From c733239f8f530872a1f80d8c45dcafbaff368737 Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Sat, 16 Mar 2024 08:35:41 +0100 Subject: bpf: Check return from set_memory_rox() arch_protect_bpf_trampoline() and alloc_new_pack() call set_memory_rox() which can fail, leading to unprotected memory. Take into account return from set_memory_rox() function and add __must_check flag to arch_protect_bpf_trampoline(). Signed-off-by: Christophe Leroy Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/fe1c163c83767fde5cab31d209a4a6be3ddb3a73.1710574353.git.christophe.leroy@csgroup.eu Signed-off-by: Martin KaFai Lau --- net/bpf/bpf_dummy_struct_ops.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c index de33dc1b0daa..25b75844891a 100644 --- a/net/bpf/bpf_dummy_struct_ops.c +++ b/net/bpf/bpf_dummy_struct_ops.c @@ -133,7 +133,9 @@ int bpf_struct_ops_test_run(struct bpf_prog *prog, const union bpf_attr *kattr, if (err < 0) goto out; - arch_protect_bpf_trampoline(image, PAGE_SIZE); + err = arch_protect_bpf_trampoline(image, PAGE_SIZE); + if (err) + goto out; prog_ret = dummy_ops_call_op(image, args); err = dummy_ops_copy_args(args); -- cgit v1.2.3 From eb166e522c77699fc19bfa705652327a1e51a117 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Fri, 15 Mar 2024 11:48:54 -0700 Subject: bpf: Allow helper bpf_get_[ns_]current_pid_tgid() for all prog types Currently bpf_get_current_pid_tgid() is allowed in tracing, cgroup and sk_msg progs while bpf_get_ns_current_pid_tgid() is only allowed in tracing progs. We have an internal use case where for an application running in a container (with pid namespace), user wants to get the pid associated with the pid namespace in a cgroup bpf program. Currently, cgroup bpf progs already allow bpf_get_current_pid_tgid(). Let us allow bpf_get_ns_current_pid_tgid() as well. With auditing the code, bpf_get_current_pid_tgid() is also used by sk_msg prog. But there are no side effect to expose these two helpers to all prog types since they do not reveal any kernel specific data. The detailed discussion is in [1]. So with this patch, both bpf_get_current_pid_tgid() and bpf_get_ns_current_pid_tgid() are put in bpf_base_func_proto(), making them available to all program types. [1] https://lore.kernel.org/bpf/20240307232659.1115872-1-yonghong.song@linux.dev/ Signed-off-by: Yonghong Song Signed-off-by: Andrii Nakryiko Acked-by: Jiri Olsa Link: https://lore.kernel.org/bpf/20240315184854.2975190-1-yonghong.song@linux.dev --- kernel/bpf/cgroup.c | 2 -- kernel/bpf/helpers.c | 4 ++++ kernel/trace/bpf_trace.c | 4 ---- net/core/filter.c | 2 -- 4 files changed, 4 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c index 82243cb6c54d..8ba73042a239 100644 --- a/kernel/bpf/cgroup.c +++ b/kernel/bpf/cgroup.c @@ -2575,8 +2575,6 @@ cgroup_current_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) switch (func_id) { case BPF_FUNC_get_current_uid_gid: return &bpf_get_current_uid_gid_proto; - case BPF_FUNC_get_current_pid_tgid: - return &bpf_get_current_pid_tgid_proto; case BPF_FUNC_get_current_comm: return &bpf_get_current_comm_proto; #ifdef CONFIG_CGROUP_NET_CLASSID diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index a89587859571..9234174ccb21 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -1730,6 +1730,10 @@ bpf_base_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) return &bpf_strtol_proto; case BPF_FUNC_strtoul: return &bpf_strtoul_proto; + case BPF_FUNC_get_current_pid_tgid: + return &bpf_get_current_pid_tgid_proto; + case BPF_FUNC_get_ns_current_pid_tgid: + return &bpf_get_ns_current_pid_tgid_proto; default: break; } diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 0a5c4efc73c3..1b041911b1d8 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -1525,8 +1525,6 @@ bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) return &bpf_ktime_get_boot_ns_proto; case BPF_FUNC_tail_call: return &bpf_tail_call_proto; - case BPF_FUNC_get_current_pid_tgid: - return &bpf_get_current_pid_tgid_proto; case BPF_FUNC_get_current_task: return &bpf_get_current_task_proto; case BPF_FUNC_get_current_task_btf: @@ -1582,8 +1580,6 @@ bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) return &bpf_send_signal_thread_proto; case BPF_FUNC_perf_event_read_value: return &bpf_perf_event_read_value_proto; - case BPF_FUNC_get_ns_current_pid_tgid: - return &bpf_get_ns_current_pid_tgid_proto; case BPF_FUNC_ringbuf_output: return &bpf_ringbuf_output_proto; case BPF_FUNC_ringbuf_reserve: diff --git a/net/core/filter.c b/net/core/filter.c index 8adf95765cdd..0c66e4a3fc5b 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -8342,8 +8342,6 @@ sk_msg_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) return &bpf_event_output_data_proto; case BPF_FUNC_get_current_uid_gid: return &bpf_get_current_uid_gid_proto; - case BPF_FUNC_get_current_pid_tgid: - return &bpf_get_current_pid_tgid_proto; case BPF_FUNC_sk_storage_get: return &bpf_sk_storage_get_proto; case BPF_FUNC_sk_storage_delete: -- cgit v1.2.3 From b919099eba0ae302231637a784aabc5a87ed0d48 Mon Sep 17 00:00:00 2001 From: Kang Yang Date: Tue, 12 Mar 2024 12:59:46 +0800 Subject: wifi: mac80211: supplement parsing of puncturing bitmap Current mac80211 won't parsing puncturing bitmap when process EHT Operation element in 6 GHz band or Bandwidth Indication element. This leads to puncturing bitmap cannot be updated in related situations, such as connecting to an EHT AP in 6 GHz band. So supplement parsing of puncturing bitmap for these elements. Signed-off-by: Kang Yang Link: https://msgid.link/20240312045947.576231-2-quic_kangyang@quicinc.com Signed-off-by: Johannes Berg --- net/mac80211/spectmgmt.c | 4 ++++ net/mac80211/util.c | 2 ++ 2 files changed, 6 insertions(+) (limited to 'net') diff --git a/net/mac80211/spectmgmt.c b/net/mac80211/spectmgmt.c index 327c74e296e2..077d7c7efc82 100644 --- a/net/mac80211/spectmgmt.c +++ b/net/mac80211/spectmgmt.c @@ -348,6 +348,10 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata, new_chandef = csa_ie->chanreq.oper; /* and update the width accordingly */ ieee80211_chandef_eht_oper(&bwi->info, &new_chandef); + + if (bwi->params & IEEE80211_BW_IND_DIS_SUBCH_PRESENT) + new_chandef.punctured = + get_unaligned_le16(bwi->info.optional); } else if (!wide_bw_chansw_ie || !wbcs_elem_to_chandef(wide_bw_chansw_ie, &new_chandef)) { if (!ieee80211_operating_class_to_chandef(new_op_class, new_chan, diff --git a/net/mac80211/util.c b/net/mac80211/util.c index a237cbcf7b49..a5d5e05688b4 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -3136,6 +3136,8 @@ bool ieee80211_chandef_he_6ghz_oper(struct ieee80211_local *local, } else { ieee80211_chandef_eht_oper((const void *)eht_oper->optional, &he_chandef); + he_chandef.punctured = + ieee80211_eht_oper_dis_subchan_bitmap(eht_oper); } if (!cfg80211_chandef_valid(&he_chandef)) -- cgit v1.2.3 From 5930a9967c78152f00f591c7d0a73c2d117e448a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 12 Mar 2024 11:20:48 +0100 Subject: wifi: mac80211: spectmgmt: simplify 6 GHz HE/EHT handling Clean up the code here a bit to have only a single call to ieee80211_chandef_he_6ghz_oper() by using a local pointer variable for the difference. Link: https://msgid.link/20240312112048.94c421d767f9.Ia7ca2f315b392c74d39b44fa9eb872a2e62e75c1@changeid Signed-off-by: Johannes Berg --- net/mac80211/spectmgmt.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/mac80211/spectmgmt.c b/net/mac80211/spectmgmt.c index 077d7c7efc82..b2de4c6fb808 100644 --- a/net/mac80211/spectmgmt.c +++ b/net/mac80211/spectmgmt.c @@ -155,6 +155,7 @@ validate_chandef_by_6ghz_he_eht_oper(struct ieee80211_sub_if_data *sdata, struct ieee80211_eht_operation _oper; struct ieee80211_eht_operation_info _oper_info; } __packed eht; + const struct ieee80211_eht_operation *eht_oper; if (conn->mode < IEEE80211_CONN_MODE_HE) { chandef->chan = NULL; @@ -203,19 +204,18 @@ validate_chandef_by_6ghz_he_eht_oper(struct ieee80211_sub_if_data *sdata, } if (conn->mode < IEEE80211_CONN_MODE_EHT) { - if (!ieee80211_chandef_he_6ghz_oper(local, &he._oper, - NULL, chandef)) - chandef->chan = NULL; + eht_oper = NULL; } else { eht._oper.params = IEEE80211_EHT_OPER_INFO_PRESENT; eht._oper_info.control = he._6ghz_oper.control; eht._oper_info.ccfs0 = he._6ghz_oper.ccfs0; eht._oper_info.ccfs1 = he._6ghz_oper.ccfs1; - - if (!ieee80211_chandef_he_6ghz_oper(local, &he._oper, - &eht._oper, chandef)) - chandef->chan = NULL; + eht_oper = &eht._oper; } + + if (!ieee80211_chandef_he_6ghz_oper(local, &he._oper, + eht_oper, chandef)) + chandef->chan = NULL; } int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata, -- cgit v1.2.3 From 56cc479188f70bbb6eacbe5ed1e43d96c91ccc6a Mon Sep 17 00:00:00 2001 From: Anjaneyulu Date: Mon, 18 Mar 2024 18:53:23 +0200 Subject: wifi: mac80211: handle indoor AFC/LPI AP on assoc success Update power_type in bss_conf based on Indoor AFC and LPI power types received in HE 6 GHz operation element on assoc success. Signed-off-by: Anjaneyulu Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.89c25dae34ff.Ifd8b2983f400623ac03dc032fc9a20025c9ca365@changeid Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 8 +++++--- net/mac80211/mlme.c | 2 ++ 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 4fd9735bb75e..7f9b829dcb1e 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -2742,9 +2742,11 @@ static inline bool ieee80211_he_capa_size_ok(const u8 *data, u8 len) #define IEEE80211_HE_OPERATION_PARTIAL_BSS_COLOR 0x40000000 #define IEEE80211_HE_OPERATION_BSS_COLOR_DISABLED 0x80000000 -#define IEEE80211_6GHZ_CTRL_REG_LPI_AP 0 -#define IEEE80211_6GHZ_CTRL_REG_SP_AP 1 -#define IEEE80211_6GHZ_CTRL_REG_VLP_AP 2 +#define IEEE80211_6GHZ_CTRL_REG_LPI_AP 0 +#define IEEE80211_6GHZ_CTRL_REG_SP_AP 1 +#define IEEE80211_6GHZ_CTRL_REG_VLP_AP 2 +#define IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP 3 +#define IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP 4 /** * struct ieee80211_he_6ghz_oper - HE 6 GHz operation Information field diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 47a2cba8313f..8cd3253c6188 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -4429,9 +4429,11 @@ static bool ieee80211_assoc_config_link(struct ieee80211_link_data *link, switch (u8_get_bits(he_6ghz_oper->control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { case IEEE80211_6GHZ_CTRL_REG_LPI_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: bss_conf->power_type = IEEE80211_REG_LPI_AP; break; case IEEE80211_6GHZ_CTRL_REG_SP_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: bss_conf->power_type = IEEE80211_REG_SP_AP; break; case IEEE80211_6GHZ_CTRL_REG_VLP_AP: -- cgit v1.2.3 From dc63b1d08335a6440aac916a20394531c8417683 Mon Sep 17 00:00:00 2001 From: Anjaneyulu Date: Mon, 18 Mar 2024 18:53:24 +0200 Subject: wifi: cfg80211: handle indoor AFC/LPI AP in probe response and beacon Mark Indoor LPI and Indoor AFC power types as valid based on channel flags. While on it, added default case. Signed-off-by: Anjaneyulu Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.091cfaaa5f45.I23cfa1104a16fd4eb9751b3d0d7b158db4ff3ecd@changeid Signed-off-by: Johannes Berg --- net/wireless/scan.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 5a5dd3ce497f..b55ffe73e1ef 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2140,11 +2140,15 @@ static bool cfg80211_6ghz_power_type_valid(const u8 *ie, size_t ielen, switch (u8_get_bits(he_6ghz_oper->control, IEEE80211_HE_6GHZ_OPER_CTRL_REG_INFO)) { case IEEE80211_6GHZ_CTRL_REG_LPI_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_LPI_AP: return true; case IEEE80211_6GHZ_CTRL_REG_SP_AP: + case IEEE80211_6GHZ_CTRL_REG_INDOOR_SP_AP: return !(flags & IEEE80211_CHAN_NO_6GHZ_AFC_CLIENT); case IEEE80211_6GHZ_CTRL_REG_VLP_AP: return !(flags & IEEE80211_CHAN_NO_6GHZ_VLP_CLIENT); + default: + return false; } } return false; -- cgit v1.2.3 From 6943e00331bb26e78edb2969165e750864a6bc6d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 18 Mar 2024 18:53:25 +0200 Subject: wifi: mac80211: add flag to disallow puncturing in 5 GHz Some devices may not be capable of handling puncturing in 5 GHz only (vs. the current flag that just removes puncturing support completely). Add a flag to support such devices: check and then downgrade the channel width if needed. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.49759510da7d.I12c5a61f0be512e0c4e574c2f794ef4b37ecaf6b@changeid Signed-off-by: Johannes Berg --- include/net/mac80211.h | 3 +++ net/mac80211/debugfs.c | 1 + net/mac80211/mlme.c | 4 ++++ 3 files changed, 8 insertions(+) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 757ec33882c2..88bb781f33f6 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -2779,6 +2779,8 @@ struct ieee80211_txq { * * @IEEE80211_HW_DISALLOW_PUNCTURING: HW requires disabling puncturing in EHT * and connecting with a lower bandwidth instead + * @IEEE80211_HW_DISALLOW_PUNCTURING_5GHZ: HW requires disabling puncturing in + * EHT in 5 GHz and connecting with a lower bandwidth instead * * @IEEE80211_HW_HANDLES_QUIET_CSA: HW/driver handles quieting for CSA, so * no need to stop queues. This really should be set by a driver that @@ -2843,6 +2845,7 @@ enum ieee80211_hw_flags { IEEE80211_HW_DETECTS_COLOR_COLLISION, IEEE80211_HW_MLO_MCAST_MULTI_LINK_TX, IEEE80211_HW_DISALLOW_PUNCTURING, + IEEE80211_HW_DISALLOW_PUNCTURING_5GHZ, IEEE80211_HW_HANDLES_QUIET_CSA, /* keep last, obviously */ diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index 2f68e92a7404..98310188f330 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -498,6 +498,7 @@ static const char *hw_flag_names[] = { FLAG(DETECTS_COLOR_COLLISION), FLAG(MLO_MCAST_MULTI_LINK_TX), FLAG(DISALLOW_PUNCTURING), + FLAG(DISALLOW_PUNCTURING_5GHZ), FLAG(HANDLES_QUIET_CSA), #undef FLAG }; diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 8cd3253c6188..20f325b6b28c 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -599,6 +599,10 @@ static bool ieee80211_chandef_usable(struct ieee80211_sub_if_data *sdata, ieee80211_hw_check(&sdata->local->hw, DISALLOW_PUNCTURING)) return false; + if (chandef->punctured && chandef->chan->band == NL80211_BAND_5GHZ && + ieee80211_hw_check(&sdata->local->hw, DISALLOW_PUNCTURING_5GHZ)) + return false; + return true; } -- cgit v1.2.3 From 0dfedd48ac539eb8d83e00830346a23a6e98f9f8 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 18 Mar 2024 18:53:27 +0200 Subject: wifi: mac80211: improve association error reporting slightly There is no reason to check the request flags for each of the links, so pull that out of the loop. Also, within the loop we can set the per-link error everywhere. Signed-off-by: Benjamin Berg Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.695faa9be279.I71b11a8d66a9cae4c27e242a47d1d92922609b03@changeid Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 20f325b6b28c..2d73f0f4dd2d 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -8206,6 +8206,14 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, if (req->ap_mld_addr) { uapsd_supported = true; + if (req->flags & (ASSOC_REQ_DISABLE_HT | + ASSOC_REQ_DISABLE_VHT | + ASSOC_REQ_DISABLE_HE | + ASSOC_REQ_DISABLE_EHT)) { + err = -EINVAL; + goto err_free; + } + for (i = 0; i < IEEE80211_MLD_MAX_NUM_LINKS; i++) { struct ieee80211_supported_band *sband; struct cfg80211_bss *link_cbss = req->links[i].bss; @@ -8218,19 +8226,13 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, if (!bss->wmm_used) { err = -EINVAL; - goto err_free; - } - - if (req->flags & (ASSOC_REQ_DISABLE_HT | - ASSOC_REQ_DISABLE_VHT | - ASSOC_REQ_DISABLE_HE | - ASSOC_REQ_DISABLE_EHT)) { - err = -EINVAL; + req->links[i].error = err; goto err_free; } if (link_cbss->channel->band == NL80211_BAND_S1GHZ) { err = -EINVAL; + req->links[i].error = err; goto err_free; } -- cgit v1.2.3 From c7378d7d8be0b44d03c990a2686e64d038c694a5 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 18 Mar 2024 18:53:28 +0200 Subject: wifi: cfg80211: check BSSID Index against MaxBSSID Add a verification that the BSSID Index does not exceed the maximum number of BSSIDs in the Multiple-BSSID set. Signed-off-by: Benjamin Berg Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.a7574d415adc.I02f40c2920a9f602898190679cc27d0c8ee2c67d@changeid Signed-off-by: Johannes Berg --- net/wireless/scan.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/scan.c b/net/wireless/scan.c index b55ffe73e1ef..8ac2d19d8ede 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2447,7 +2447,8 @@ cfg80211_parse_mbssid_data(struct wiphy *wiphy, profile, profile_len); if (!mbssid_index_ie || mbssid_index_ie[1] < 1 || mbssid_index_ie[2] == 0 || - mbssid_index_ie[2] > 46) { + mbssid_index_ie[2] > 46 || + mbssid_index_ie[2] >= (1 << elem->data[0])) { /* No valid Multiple BSSID-Index element */ continue; } -- cgit v1.2.3 From 97f8df4db4c8ef50b659d8b228c1f42fe111e7c8 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 18 Mar 2024 18:53:29 +0200 Subject: wifi: cfg80211: ignore non-TX BSSs in per-STA profile If a non-TX BSS is included in a per-STA profile, then we cannot set transmitted_bss for it. Even worse, if we do things properly we should be configuring both bssid_index and max_bssid_indicator correctly. We do not actually have both pieces of information (and, some APs currently do not include either). So, ignore any per-STA profile where the RNR says that the BSS is not transmitted. Also fix transmitted_bss to never be set for per-STA profiles. This fixes issues where mac80211 was setting the reference BSSID to an incorrect value. Fixes: 2481b5da9c6b ("wifi: cfg80211: handle BSS data contained in ML probe responses") Signed-off-by: Benjamin Berg Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.6a0babed655a.Iad447fea417c63f683da793556b97c31d07a4aab@changeid Signed-off-by: Johannes Berg --- net/wireless/scan.c | 47 ++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 38 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 8ac2d19d8ede..127853877a0a 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -2211,12 +2211,16 @@ cfg80211_inform_single_bss_data(struct wiphy *wiphy, tmp.pub.use_for = data->use_for; tmp.pub.cannot_use_reasons = data->cannot_use_reasons; - if (data->bss_source != BSS_SOURCE_DIRECT) { + switch (data->bss_source) { + case BSS_SOURCE_MBSSID: tmp.pub.transmitted_bss = data->source_bss; + fallthrough; + case BSS_SOURCE_STA_PROFILE: ts = bss_from_pub(data->source_bss)->ts; tmp.pub.bssid_index = data->bssid_index; tmp.pub.max_bssid_indicator = data->max_bssid_indicator; - } else { + break; + case BSS_SOURCE_DIRECT: ts = jiffies; if (channel->band == NL80211_BAND_60GHZ) { @@ -2231,6 +2235,7 @@ cfg80211_inform_single_bss_data(struct wiphy *wiphy, regulatory_hint_found_beacon(wiphy, channel, gfp); } + break; } /* @@ -2660,6 +2665,7 @@ struct tbtt_info_iter_data { u8 param_ch_count; u32 use_for; u8 mld_id, link_id; + bool non_tx; }; static enum cfg80211_rnr_iter_ret @@ -2670,14 +2676,20 @@ cfg802121_mld_ap_rnr_iter(void *_data, u8 type, const struct ieee80211_rnr_mld_params *mld_params; struct tbtt_info_iter_data *data = _data; u8 link_id; + bool non_tx = false; if (type == IEEE80211_TBTT_INFO_TYPE_TBTT && tbtt_info_len >= offsetofend(struct ieee80211_tbtt_info_ge_11, - mld_params)) - mld_params = (void *)(tbtt_info + - offsetof(struct ieee80211_tbtt_info_ge_11, - mld_params)); - else if (type == IEEE80211_TBTT_INFO_TYPE_MLD && + mld_params)) { + const struct ieee80211_tbtt_info_ge_11 *tbtt_info_ge_11 = + (void *)tbtt_info; + + non_tx = (tbtt_info_ge_11->bss_params & + (IEEE80211_RNR_TBTT_PARAMS_MULTI_BSSID | + IEEE80211_RNR_TBTT_PARAMS_TRANSMITTED_BSSID)) == + IEEE80211_RNR_TBTT_PARAMS_MULTI_BSSID; + mld_params = &tbtt_info_ge_11->mld_params; + } else if (type == IEEE80211_TBTT_INFO_TYPE_MLD && tbtt_info_len >= sizeof(struct ieee80211_rnr_mld_params)) mld_params = (void *)tbtt_info; else @@ -2696,6 +2708,7 @@ cfg802121_mld_ap_rnr_iter(void *_data, u8 type, data->param_ch_count = le16_get_bits(mld_params->params, IEEE80211_RNR_MLD_PARAMS_BSS_CHANGE_COUNT); + data->non_tx = non_tx; if (type == IEEE80211_TBTT_INFO_TYPE_TBTT) data->use_for = NL80211_BSS_USE_FOR_ALL; @@ -2707,7 +2720,7 @@ cfg802121_mld_ap_rnr_iter(void *_data, u8 type, static u8 cfg80211_rnr_info_for_mld_ap(const u8 *ie, size_t ielen, u8 mld_id, u8 link_id, const struct ieee80211_neighbor_ap_info **ap_info, - u8 *param_ch_count) + u8 *param_ch_count, bool *non_tx) { struct tbtt_info_iter_data data = { .mld_id = mld_id, @@ -2718,6 +2731,7 @@ cfg80211_rnr_info_for_mld_ap(const u8 *ie, size_t ielen, u8 mld_id, u8 link_id, *ap_info = data.ap_info; *param_ch_count = data.param_ch_count; + *non_tx = data.non_tx; return data.use_for; } @@ -2897,6 +2911,7 @@ cfg80211_parse_ml_elem_sta_data(struct wiphy *wiphy, ssize_t profile_len; u8 param_ch_count; u8 link_id, use_for; + bool non_tx; if (!ieee80211_mle_basic_sta_prof_size_ok((u8 *)mle->sta_prof[i], mle->sta_prof_len[i])) @@ -2942,10 +2957,24 @@ cfg80211_parse_ml_elem_sta_data(struct wiphy *wiphy, tx_data->ielen, mld_id, link_id, &ap_info, - ¶m_ch_count); + ¶m_ch_count, + &non_tx); if (!use_for) continue; + /* + * As of 802.11be_D5.0, the specification does not give us any + * way of discovering both the MaxBSSID and the Multiple-BSSID + * Index. It does seem like the Multiple-BSSID Index element + * may be provided, but section 9.4.2.45 explicitly forbids + * including a Multiple-BSSID Element (in this case without any + * subelements). + * Without both pieces of information we cannot calculate the + * reference BSSID, so simply ignore the BSS. + */ + if (non_tx) + continue; + /* We could sanity check the BSSID is included */ if (!ieee80211_operating_class_to_band(ap_info->op_class, -- cgit v1.2.3 From a17a58ad2ff24f0d201fa5f9939182f3757d1737 Mon Sep 17 00:00:00 2001 From: Ayala Beker Date: Mon, 18 Mar 2024 18:53:31 +0200 Subject: wifi: mac80211: add support for tearing down negotiated TTLM In order to activate a link that is currently inactive due to a negotiated TTLM request, need to first tear down the negotiated TTLM request. Add support for sending TTLM teardown request and update the links state accordingly. Signed-off-by: Ayala Beker Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240318184907.d480cbf46fcf.Idedad472469d2c27dd2a088cf80a13a1e1cf9b78@changeid Signed-off-by: Johannes Berg --- include/net/mac80211.h | 9 +++++++ net/mac80211/ieee80211_i.h | 2 ++ net/mac80211/mlme.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++ net/mac80211/status.c | 22 +++++++++++++++++- 4 files changed, 90 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 88bb781f33f6..f57c29de3a91 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -7591,6 +7591,15 @@ int ieee80211_set_active_links(struct ieee80211_vif *vif, u16 active_links); void ieee80211_set_active_links_async(struct ieee80211_vif *vif, u16 active_links); +/** + * ieee80211_send_teardown_neg_ttlm - tear down a negotiated TTLM request + * @vif: the interface on which the tear down request should be sent. + * + * This function can be used to tear down a previously accepted negotiated + * TTLM request. + */ +void ieee80211_send_teardown_neg_ttlm(struct ieee80211_vif *vif); + /* for older drivers - let's not document these ... */ int ieee80211_emulate_add_chanctx(struct ieee80211_hw *hw, struct ieee80211_chanctx_conf *ctx); diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index b6fead612b66..925e875f9b96 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -89,6 +89,7 @@ enum ieee80211_status_data { IEEE80211_STATUS_TYPE_MASK = 0x00f, IEEE80211_STATUS_TYPE_INVALID = 0, IEEE80211_STATUS_TYPE_SMPS = 1, + IEEE80211_STATUS_TYPE_NEG_TTLM = 2, IEEE80211_STATUS_SUBDATA_MASK = 0xff0, }; @@ -595,6 +596,7 @@ struct ieee80211_if_managed { /* TID-to-link mapping support */ struct wiphy_delayed_work ttlm_work; struct ieee80211_adv_ttlm_info ttlm_info; + struct wiphy_work teardown_ttlm_work; /* dialog token enumerator for neg TTLM request */ u8 dialog_token_alloc; diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 2d73f0f4dd2d..b653c7d5025f 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -6790,6 +6790,60 @@ void ieee80211_process_neg_ttlm_res(struct ieee80211_sub_if_data *sdata, __ieee80211_disconnect(sdata); } +static void ieee80211_teardown_ttlm_work(struct wiphy *wiphy, + struct wiphy_work *work) +{ + u16 new_dormant_links; + struct ieee80211_sub_if_data *sdata = + container_of(work, struct ieee80211_sub_if_data, + u.mgd.neg_ttlm_timeout_work.work); + + if (!sdata->vif.neg_ttlm.valid) + return; + + memset(&sdata->vif.neg_ttlm, 0, sizeof(sdata->vif.neg_ttlm)); + new_dormant_links = + sdata->vif.dormant_links & ~sdata->vif.suspended_links; + sdata->vif.suspended_links = 0; + ieee80211_vif_set_links(sdata, sdata->vif.valid_links, + new_dormant_links); + ieee80211_vif_cfg_change_notify(sdata, BSS_CHANGED_MLD_TTLM | + BSS_CHANGED_MLD_VALID_LINKS); +} + +void ieee80211_send_teardown_neg_ttlm(struct ieee80211_vif *vif) +{ + struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); + struct ieee80211_local *local = sdata->local; + struct ieee80211_mgmt *mgmt; + struct sk_buff *skb; + int frame_len = offsetofend(struct ieee80211_mgmt, + u.action.u.ttlm_tear_down); + struct ieee80211_tx_info *info; + + skb = dev_alloc_skb(local->hw.extra_tx_headroom + frame_len); + if (!skb) + return; + + skb_reserve(skb, local->hw.extra_tx_headroom); + mgmt = skb_put_zero(skb, frame_len); + mgmt->frame_control = cpu_to_le16(IEEE80211_FTYPE_MGMT | + IEEE80211_STYPE_ACTION); + memcpy(mgmt->da, sdata->vif.cfg.ap_addr, ETH_ALEN); + memcpy(mgmt->sa, sdata->vif.addr, ETH_ALEN); + memcpy(mgmt->bssid, sdata->vif.cfg.ap_addr, ETH_ALEN); + + mgmt->u.action.category = WLAN_CATEGORY_PROTECTED_EHT; + mgmt->u.action.u.ttlm_tear_down.action_code = + WLAN_PROTECTED_EHT_ACTION_TTLM_TEARDOWN; + + info = IEEE80211_SKB_CB(skb); + info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS; + info->status_data = IEEE80211_STATUS_TYPE_NEG_TTLM; + ieee80211_tx_skb(sdata, skb); +} +EXPORT_SYMBOL(ieee80211_send_teardown_neg_ttlm); + void ieee80211_sta_rx_queued_ext(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb) { @@ -7421,6 +7475,8 @@ void ieee80211_sta_setup_sdata(struct ieee80211_sub_if_data *sdata) ieee80211_tid_to_link_map_work); wiphy_delayed_work_init(&ifmgd->neg_ttlm_timeout_work, ieee80211_neg_ttlm_timeout_work); + wiphy_work_init(&ifmgd->teardown_ttlm_work, + ieee80211_teardown_ttlm_work); ifmgd->flags = 0; ifmgd->powersave = sdata->wdev.ps; @@ -8609,6 +8665,8 @@ void ieee80211_mgd_stop(struct ieee80211_sub_if_data *sdata) &ifmgd->beacon_connection_loss_work); wiphy_work_cancel(sdata->local->hw.wiphy, &ifmgd->csa_connection_drop_work); + wiphy_work_cancel(sdata->local->hw.wiphy, + &ifmgd->teardown_ttlm_work); wiphy_delayed_work_cancel(sdata->local->hw.wiphy, &ifmgd->tdls_peer_del_work); wiphy_delayed_work_cancel(sdata->local->hw.wiphy, diff --git a/net/mac80211/status.c b/net/mac80211/status.c index 1708b33cdc5e..dd8f857a1fbc 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -5,7 +5,7 @@ * Copyright 2006-2007 Jiri Benc * Copyright 2008-2010 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH - * Copyright 2021-2023 Intel Corporation + * Copyright 2021-2024 Intel Corporation */ #include @@ -696,6 +696,23 @@ static void ieee80211_handle_smps_status(struct ieee80211_sub_if_data *sdata, wiphy_work_queue(sdata->local->hw.wiphy, &link->u.mgd.recalc_smps); } +static void +ieee80211_handle_teardown_ttlm_status(struct ieee80211_sub_if_data *sdata, + bool acked) +{ + if (!sdata || !ieee80211_sdata_running(sdata)) + return; + + if (!acked) + return; + + if (sdata->vif.type != NL80211_IFTYPE_STATION) + return; + + wiphy_work_queue(sdata->local->hw.wiphy, + &sdata->u.mgd.teardown_ttlm_work); +} + static void ieee80211_report_used_skb(struct ieee80211_local *local, struct sk_buff *skb, bool dropped, ktime_t ack_hwtstamp) @@ -773,6 +790,9 @@ static void ieee80211_report_used_skb(struct ieee80211_local *local, ieee80211_handle_smps_status(sdata, acked, info->status_data); break; + case IEEE80211_STATUS_TYPE_NEG_TTLM: + ieee80211_handle_teardown_ttlm_status(sdata, acked); + break; } rcu_read_unlock(); } -- cgit v1.2.3 From 3c9ff1a1e1d5e58f0610d408e07e372f5cf15740 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Mar 2024 09:13:59 +0200 Subject: wifi: mac80211: don't enter idle during link switch When doing link switch with a disjoint set of links before and after the switch, we end up removing all channel contexts, adding new ones later. This looks like 'idle' to the code now, and we enter idle which also includes flushing queues. But we can't actually flush since we don't have a link active (bound to a channel context), and entering idle just to leave it again is also wrong. Fix this by passing through an indication that we shouldn't do any idle checks in this case. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240320091155.170328bac555.If4a522a9dd3133b91983854b909a4de13aa635da@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 34 +++++++++++++++++++--------------- net/mac80211/ieee80211_i.h | 2 ++ net/mac80211/link.c | 2 +- 3 files changed, 22 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 80e4b9784131..5a7fb0e4f89d 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -695,26 +695,29 @@ ieee80211_new_chanctx(struct ieee80211_local *local, } static void ieee80211_del_chanctx(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx) + struct ieee80211_chanctx *ctx, + bool skip_idle_recalc) { lockdep_assert_wiphy(local->hw.wiphy); drv_remove_chanctx(local, ctx); - ieee80211_recalc_idle(local); + if (!skip_idle_recalc) + ieee80211_recalc_idle(local); ieee80211_remove_wbrf(local, &ctx->conf.def); } static void ieee80211_free_chanctx(struct ieee80211_local *local, - struct ieee80211_chanctx *ctx) + struct ieee80211_chanctx *ctx, + bool skip_idle_recalc) { lockdep_assert_wiphy(local->hw.wiphy); WARN_ON_ONCE(ieee80211_chanctx_refcount(local, ctx) != 0); list_del_rcu(&ctx->list); - ieee80211_del_chanctx(local, ctx); + ieee80211_del_chanctx(local, ctx, skip_idle_recalc); kfree_rcu(ctx, rcu_head); } @@ -1002,7 +1005,7 @@ int ieee80211_link_unreserve_chanctx(struct ieee80211_link_data *link) list_del_rcu(&ctx->list); kfree_rcu(ctx, rcu_head); } else { - ieee80211_free_chanctx(sdata->local, ctx); + ieee80211_free_chanctx(sdata->local, ctx, false); } } @@ -1218,7 +1221,7 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) CHANCTX_SWMODE_REASSIGN_VIF); if (err) { if (ieee80211_chanctx_refcount(local, new_ctx) == 0) - ieee80211_free_chanctx(local, new_ctx); + ieee80211_free_chanctx(local, new_ctx, false); goto out; } @@ -1232,7 +1235,7 @@ ieee80211_link_use_reserved_reassign(struct ieee80211_link_data *link) ieee80211_check_fast_xmit_iface(sdata); if (ieee80211_chanctx_refcount(local, old_ctx) == 0) - ieee80211_free_chanctx(local, old_ctx); + ieee80211_free_chanctx(local, old_ctx, false); ieee80211_recalc_chanctx_min_def(local, new_ctx, NULL); ieee80211_recalc_smps_chanctx(local, new_ctx); @@ -1286,7 +1289,7 @@ ieee80211_link_use_reserved_assign(struct ieee80211_link_data *link) err = ieee80211_assign_link_chanctx(link, new_ctx); if (err) { if (ieee80211_chanctx_refcount(local, new_ctx) == 0) - ieee80211_free_chanctx(local, new_ctx); + ieee80211_free_chanctx(local, new_ctx, false); goto out; } @@ -1383,7 +1386,7 @@ static int ieee80211_chsw_switch_ctxs(struct ieee80211_local *local) if (!list_empty(&ctx->replace_ctx->assigned_links)) continue; - ieee80211_del_chanctx(local, ctx->replace_ctx); + ieee80211_del_chanctx(local, ctx->replace_ctx, false); err = ieee80211_add_chanctx(local, ctx); if (err) goto err; @@ -1400,7 +1403,7 @@ err: if (!list_empty(&ctx->replace_ctx->assigned_links)) continue; - ieee80211_del_chanctx(local, ctx); + ieee80211_del_chanctx(local, ctx, false); WARN_ON(ieee80211_add_chanctx(local, ctx->replace_ctx)); } @@ -1652,7 +1655,8 @@ err: return err; } -static void __ieee80211_link_release_channel(struct ieee80211_link_data *link) +void __ieee80211_link_release_channel(struct ieee80211_link_data *link, + bool skip_idle_recalc) { struct ieee80211_sub_if_data *sdata = link->sdata; struct ieee80211_bss_conf *link_conf = link->conf; @@ -1680,7 +1684,7 @@ static void __ieee80211_link_release_channel(struct ieee80211_link_data *link) ieee80211_assign_link_chanctx(link, NULL); if (ieee80211_chanctx_refcount(local, ctx) == 0) - ieee80211_free_chanctx(local, ctx); + ieee80211_free_chanctx(local, ctx, skip_idle_recalc); link->radar_required = false; @@ -1721,7 +1725,7 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, if (ret < 0) goto out; - __ieee80211_link_release_channel(link); + __ieee80211_link_release_channel(link, false); ctx = ieee80211_find_chanctx(local, chanreq, mode); if (!ctx) @@ -1737,7 +1741,7 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, if (ret) { /* if assign fails refcount stays the same */ if (ieee80211_chanctx_refcount(local, ctx) == 0) - ieee80211_free_chanctx(local, ctx); + ieee80211_free_chanctx(local, ctx, false); goto out; } @@ -1930,7 +1934,7 @@ void ieee80211_link_release_channel(struct ieee80211_link_data *link) lockdep_assert_wiphy(sdata->local->hw.wiphy); if (rcu_access_pointer(link->conf->chanctx_conf)) - __ieee80211_link_release_channel(link); + __ieee80211_link_release_channel(link, false); } void ieee80211_link_vlan_copy_chanctx(struct ieee80211_link_data *link) diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 925e875f9b96..c2212363dad3 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2567,6 +2567,8 @@ int __must_check ieee80211_link_change_chanreq(struct ieee80211_link_data *link, const struct ieee80211_chan_req *req, u64 *changed); +void __ieee80211_link_release_channel(struct ieee80211_link_data *link, + bool skip_idle_recalc); void ieee80211_link_release_channel(struct ieee80211_link_data *link); void ieee80211_link_vlan_copy_chanctx(struct ieee80211_link_data *link); void ieee80211_link_copy_chanctx_to_vlans(struct ieee80211_link_data *link, diff --git a/net/mac80211/link.c b/net/mac80211/link.c index 685ec66b4264..db49bcec5db4 100644 --- a/net/mac80211/link.c +++ b/net/mac80211/link.c @@ -358,7 +358,7 @@ static int _ieee80211_set_active_links(struct ieee80211_sub_if_data *sdata, ieee80211_teardown_tdls_peers(link); - ieee80211_link_release_channel(link); + __ieee80211_link_release_channel(link, true); } list_for_each_entry(sta, &local->sta_list, list) { -- cgit v1.2.3 From 21e29016d5d25a2dcee2e30108f5f0d87bdb398d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Mar 2024 09:14:00 +0200 Subject: wifi: mac80211: clarify IEEE80211_STATUS_SUBDATA_MASK We have 13 bits for the status_data, so restricting type to 4 and subdata to 8 bits is confusing, even if we don't need more bits now. Change subdata mask to be 9 bits instead, just to make things match up. If we actually need more types or more subdata bits we can later also reshuffle the bits between these, but we should probably keep them at 13 bits together. Signed-off-by: Johannes Berg Reviewed-by: Ayala Beker Signed-off-by: Miri Korenblit Link: https://msgid.link/20240320091155.28ac7b665039.I1abbb13e90f016cab552492e05f5cb5b52de6463@changeid Signed-off-by: Johannes Berg --- net/mac80211/ht.c | 2 +- net/mac80211/ieee80211_i.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/ht.c b/net/mac80211/ht.c index c3330aea4da3..d7e8cf8e48b7 100644 --- a/net/mac80211/ht.c +++ b/net/mac80211/ht.c @@ -580,7 +580,7 @@ int ieee80211_send_smps_action(struct ieee80211_sub_if_data *sdata, /* we'll do more on status of this frame */ info = IEEE80211_SKB_CB(skb); info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS; - /* we have 12 bits, and need 6: link_id 4, smps 2 */ + /* we have 13 bits, and need 6: link_id 4, smps 2 */ info->status_data = IEEE80211_STATUS_TYPE_SMPS | u16_encode_bits(status_link_id << 2 | smps, IEEE80211_STATUS_SUBDATA_MASK); diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index c2212363dad3..fe81ab641620 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -90,7 +90,7 @@ enum ieee80211_status_data { IEEE80211_STATUS_TYPE_INVALID = 0, IEEE80211_STATUS_TYPE_SMPS = 1, IEEE80211_STATUS_TYPE_NEG_TTLM = 2, - IEEE80211_STATUS_SUBDATA_MASK = 0xff0, + IEEE80211_STATUS_SUBDATA_MASK = 0x1ff0, }; static inline bool -- cgit v1.2.3 From 80b0aacd1ad046b46d471cf8ed6203bbd777f988 Mon Sep 17 00:00:00 2001 From: Ayala Beker Date: Wed, 20 Mar 2024 09:14:02 +0200 Subject: wifi: mac80211: don't select link ID if not provided in scan request If scan request doesn't include a link ID to be used for TSF reporting, don't select it as it might become inactive before scan is actually started by the driver. Instead, let the driver select one of the active links. Fixes: cbde0b49f276 ("wifi: mac80211: Extend support for scanning while MLO connected") Signed-off-by: Ayala Beker Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240320091155.a6b643a15755.Ic28ed9a611432387b7f85e9ca9a97a4ce34a6e0f@changeid Signed-off-by: Johannes Berg --- net/mac80211/scan.c | 16 ++++------------ 1 file changed, 4 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 0429e59ba387..977f8eb0a67b 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -707,19 +707,11 @@ static int __ieee80211_start_scan(struct ieee80211_sub_if_data *sdata, return -EBUSY; /* For an MLO connection, if a link ID was specified, validate that it - * is indeed active. If no link ID was specified, select one of the - * active links. + * is indeed active. */ - if (ieee80211_vif_is_mld(&sdata->vif)) { - if (req->tsf_report_link_id >= 0) { - if (!(sdata->vif.active_links & - BIT(req->tsf_report_link_id))) - return -EINVAL; - } else { - req->tsf_report_link_id = - __ffs(sdata->vif.active_links); - } - } + if (ieee80211_vif_is_mld(&sdata->vif) && req->tsf_report_link_id >= 0 && + !(sdata->vif.active_links & BIT(req->tsf_report_link_id))) + return -EINVAL; if (!__ieee80211_can_leave_ch(sdata)) return -EBUSY; -- cgit v1.2.3 From 2f51c87a156c2ee30bf6a66694c64856177c5303 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Mar 2024 09:14:03 +0200 Subject: wifi: mac80211: don't ask driver about no-op link changes If the links won't actually change, nothing will happen. This was previously done in the inner function (twice in some cases), but we shouldn't bother the driver with it. Clean that up. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240320091155.a8190a312a27.If4e6f5ce8228eda7afac0fc8c17dd731c5da9ed9@changeid Signed-off-by: Johannes Berg --- net/mac80211/link.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/link.c b/net/mac80211/link.c index db49bcec5db4..43f9672fc7f1 100644 --- a/net/mac80211/link.c +++ b/net/mac80211/link.c @@ -450,10 +450,13 @@ int ieee80211_set_active_links(struct ieee80211_vif *vif, u16 active_links) if (WARN_ON(!active_links)) return -EINVAL; + old_active = sdata->vif.active_links; + if (old_active == active_links) + return 0; + if (!drv_can_activate_links(local, sdata, active_links)) return -EINVAL; - old_active = sdata->vif.active_links; if (old_active & active_links) { /* * if there's at least one link that stays active across -- cgit v1.2.3 From 6e02ba7c9e0a28a87efe5a179462515e7a47008e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Mar 2024 09:14:04 +0200 Subject: wifi: mac80211: improve drop for action frame return If we use a drop we not only save the extra call to dev_kfree_skb(), but also have a better reason in tracing, so do that. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240320091155.34daf0a89eb4.I60e0639511f9de64e40e6105b640adf90f8f57f7@changeid Signed-off-by: Johannes Berg --- net/mac80211/drop.h | 3 ++- net/mac80211/rx.c | 4 ++-- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/drop.h b/net/mac80211/drop.h index 12a6f0e9eca6..59e3ec4dc960 100644 --- a/net/mac80211/drop.h +++ b/net/mac80211/drop.h @@ -2,7 +2,7 @@ /* * mac80211 drop reason list * - * Copyright (C) 2023 Intel Corporation + * Copyright (C) 2023-2024 Intel Corporation */ #ifndef MAC80211_DROP_H @@ -66,6 +66,7 @@ typedef unsigned int __bitwise ieee80211_rx_result; R(RX_DROP_U_UNEXPECTED_STA_4ADDR) \ R(RX_DROP_U_UNEXPECTED_VLAN_MCAST) \ R(RX_DROP_U_NOT_PORT_CONTROL) \ + R(RX_DROP_U_UNKNOWN_ACTION_REJECTED) \ /* this line for the trailing \ - add before this */ /* having two enums allows for checking ieee80211_rx_result use with sparse */ diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index c1f850138405..4b4cbd8bf35d 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -3958,8 +3958,8 @@ ieee80211_rx_h_action_return(struct ieee80211_rx_data *rx) __ieee80211_tx_skb_tid_band(rx->sdata, nskb, 7, -1, status->band); } - dev_kfree_skb(rx->skb); - return RX_QUEUED; + + return RX_DROP_U_UNKNOWN_ACTION_REJECTED; } static ieee80211_rx_result debug_noinline -- cgit v1.2.3 From a0b9ecffc3cb4ad7eb02e57838fdbc1ba985d183 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Mar 2024 09:14:05 +0200 Subject: wifi: mac80211: reactivate multi-link later in restart In case of restart, we currently reactivate multi-link on interfaces before reconfiguring keys etc. which means the drivers need to handle this case differently. Enable more links later to allow them to handle it the same way. Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240320091155.d0f18a56335d.Ib3338d93872a4a568f38db0d02546534d3eff810@changeid Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 2 ++ net/mac80211/util.c | 12 +++++++++--- 2 files changed, 11 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index fe81ab641620..def611e4e55f 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1161,6 +1161,8 @@ struct ieee80211_sub_if_data { struct wiphy_work activate_links_work; u16 desired_active_links; + u16 restart_active_links; + #ifdef CONFIG_MAC80211_DEBUGFS struct { struct dentry *subdir_stations; diff --git a/net/mac80211/util.c b/net/mac80211/util.c index a5d5e05688b4..cda398d8f60d 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -1932,6 +1932,8 @@ int ieee80211_reconfig(struct ieee80211_local *local) old); } + sdata->restart_active_links = active_links; + for (link_id = 0; link_id < ARRAY_SIZE(sdata->vif.link_conf); link_id++) { @@ -2059,9 +2061,6 @@ int ieee80211_reconfig(struct ieee80211_local *local) WARN_ON(1); break; } - - if (active_links) - ieee80211_set_active_links(&sdata->vif, active_links); } ieee80211_recalc_ps(local); @@ -2102,6 +2101,13 @@ int ieee80211_reconfig(struct ieee80211_local *local) list_for_each_entry(sdata, &local->interfaces, list) ieee80211_reenable_keys(sdata); + /* re-enable multi-link for client interfaces */ + list_for_each_entry(sdata, &local->interfaces, list) { + if (sdata->restart_active_links) + ieee80211_set_active_links(&sdata->vif, + sdata->restart_active_links); + } + /* Reconfigure sched scan if it was interrupted by FW restart */ sched_scan_sdata = rcu_dereference_protected(local->sched_scan_sdata, lockdep_is_held(&local->hw.wiphy->mtx)); -- cgit v1.2.3 From f40db02e8fa357196d57e002cc1e74abe8b8009d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 25 Mar 2024 15:05:03 +0100 Subject: wifi: mac80211: use kvcalloc() for codel vars This is a big array, but it's only used by software and need not be contiguous in memory. Use kvcalloc() since it's so big (order 5 allocation). Reviewed-by: Miriam Rachel Korenblit Link: https://msgid.link/20240325150509.9195643699e4.I1b94b17abc809491080d6312f31ce6b5decdd446@changeid Signed-off-by: Johannes Berg --- net/mac80211/tx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 6bf223e6cd1a..ac5ae7c05e36 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1604,8 +1604,8 @@ int ieee80211_txq_setup_flows(struct ieee80211_local *local) local->cparams.target = MS2TIME(20); local->cparams.ecn = true; - local->cvars = kcalloc(fq->flows_cnt, sizeof(local->cvars[0]), - GFP_KERNEL); + local->cvars = kvcalloc(fq->flows_cnt, sizeof(local->cvars[0]), + GFP_KERNEL); if (!local->cvars) { spin_lock_bh(&fq->lock); fq_reset(fq, fq_skb_free_func); @@ -1625,7 +1625,7 @@ void ieee80211_txq_teardown_flows(struct ieee80211_local *local) { struct fq *fq = &local->fq; - kfree(local->cvars); + kvfree(local->cvars); local->cvars = NULL; spin_lock_bh(&fq->lock); -- cgit v1.2.3 From 597d817202b5af0e3a9ff40c5d0a21e4e2e06e00 Mon Sep 17 00:00:00 2001 From: Uwe Kleine-König Date: Wed, 6 Mar 2024 19:35:38 +0100 Subject: net: rfkill: gpio: Convert to platform remove callback returning void MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by: Uwe Kleine-König Reviewed-by: Jeff Johnson Link: https://msgid.link/20240306183538.88777-2-u.kleine-koenig@pengutronix.de Signed-off-by: Johannes Berg --- net/rfkill/rfkill-gpio.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/rfkill/rfkill-gpio.c b/net/rfkill/rfkill-gpio.c index 4e32d659524e..84529886c2e6 100644 --- a/net/rfkill/rfkill-gpio.c +++ b/net/rfkill/rfkill-gpio.c @@ -156,14 +156,12 @@ err_destroy: return ret; } -static int rfkill_gpio_remove(struct platform_device *pdev) +static void rfkill_gpio_remove(struct platform_device *pdev) { struct rfkill_gpio_data *rfkill = platform_get_drvdata(pdev); rfkill_unregister(rfkill->rfkill_dev); rfkill_destroy(rfkill->rfkill_dev); - - return 0; } #ifdef CONFIG_ACPI @@ -183,7 +181,7 @@ MODULE_DEVICE_TABLE(of, rfkill_of_match); static struct platform_driver rfkill_gpio_driver = { .probe = rfkill_gpio_probe, - .remove = rfkill_gpio_remove, + .remove_new = rfkill_gpio_remove, .driver = { .name = "rfkill_gpio", .acpi_match_table = ACPI_PTR(rfkill_acpi_match), -- cgit v1.2.3 From 56364c910691f6d10ba88c964c9041b9ab777bd6 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Mon, 25 Mar 2024 08:40:28 +0100 Subject: net: Remove conditional threaded-NAPI wakeup based on task state. A NAPI thread is scheduled by first setting NAPI_STATE_SCHED bit. If successful (the bit was not yet set) then the NAPI_STATE_SCHED_THREADED is set but only if thread's state is not TASK_INTERRUPTIBLE (is TASK_RUNNING) followed by task wakeup. If the task is idle (TASK_INTERRUPTIBLE) then the NAPI_STATE_SCHED_THREADED bit is not set. The thread is no relying on the bit but always leaving the wait-loop after returning from schedule() because there must have been a wakeup. The smpboot-threads implementation for per-CPU threads requires an explicit condition and does not support "if we get out of schedule() then there must be something to do". Removing this optimisation simplifies the following integration. Set NAPI_STATE_SCHED_THREADED unconditionally on wakeup and rely on it in the wait path by removing the `woken' condition. Acked-by: Jakub Kicinski Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Paolo Abeni --- net/core/dev.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 9a67003e49db..98d7d339a8c1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4427,13 +4427,7 @@ static inline void ____napi_schedule(struct softnet_data *sd, */ thread = READ_ONCE(napi->thread); if (thread) { - /* Avoid doing set_bit() if the thread is in - * INTERRUPTIBLE state, cause napi_thread_wait() - * makes sure to proceed with napi polling - * if the thread is explicitly woken from here. - */ - if (READ_ONCE(thread->__state) != TASK_INTERRUPTIBLE) - set_bit(NAPI_STATE_SCHED_THREADED, &napi->state); + set_bit(NAPI_STATE_SCHED_THREADED, &napi->state); wake_up_process(thread); return; } @@ -6710,8 +6704,6 @@ static int napi_poll(struct napi_struct *n, struct list_head *repoll) static int napi_thread_wait(struct napi_struct *napi) { - bool woken = false; - set_current_state(TASK_INTERRUPTIBLE); while (!kthread_should_stop()) { @@ -6720,15 +6712,13 @@ static int napi_thread_wait(struct napi_struct *napi) * Testing SCHED bit is not enough because SCHED bit might be * set by some other busy poll thread or by napi_disable(). */ - if (test_bit(NAPI_STATE_SCHED_THREADED, &napi->state) || woken) { + if (test_bit(NAPI_STATE_SCHED_THREADED, &napi->state)) { WARN_ON(!list_empty(&napi->poll_list)); __set_current_state(TASK_RUNNING); return 0; } schedule(); - /* woken being true indicates this thread owns this napi. */ - woken = true; set_current_state(TASK_INTERRUPTIBLE); } __set_current_state(TASK_RUNNING); -- cgit v1.2.3 From dad6b97702639fba27a2bd3e986982ad6f0db3a7 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Mon, 25 Mar 2024 08:40:29 +0100 Subject: net: Allow to use SMP threads for backlog NAPI. Backlog NAPI is a per-CPU NAPI struct only (with no device behind it) used by drivers which don't do NAPI them self, RPS and parts of the stack which need to avoid recursive deadlocks while processing a packet. The non-NAPI driver use the CPU local backlog NAPI. If RPS is enabled then a flow for the skb is computed and based on the flow the skb can be enqueued on a remote CPU. Scheduling/ raising the softirq (for backlog's NAPI) on the remote CPU isn't trivial because the softirq is only scheduled on the local CPU and performed after the hardirq is done. In order to schedule a softirq on the remote CPU, an IPI is sent to the remote CPU which schedules the backlog-NAPI on the then local CPU. On PREEMPT_RT interrupts are force-threaded. The soft interrupts are raised within the interrupt thread and processed after the interrupt handler completed still within the context of the interrupt thread. The softirq is handled in the context where it originated. With force-threaded interrupts enabled, ksoftirqd is woken up if a softirq is raised from hardirq context. This is the case if it is raised from an IPI. Additionally there is a warning on PREEMPT_RT if the softirq is raised from the idle thread. This was done for two reasons: - With threaded interrupts the processing should happen in thread context (where it originated) and ksoftirqd is the only thread for this context if raised from hardirq. Using the currently running task instead would "punish" a random task. - Once ksoftirqd is active it consumes all further softirqs until it stops running. This changed recently and is no longer the case. Instead of keeping the backlog NAPI in ksoftirqd (in force-threaded/ PREEMPT_RT setups) I am proposing NAPI-threads for backlog. The "proper" setup with threaded-NAPI is not doable because the threads are not pinned to an individual CPU and can be modified by the user. Additionally a dummy network device would have to be assigned. Also CPU-hotplug has to be considered if additional CPUs show up. All this can be probably done/ solved but the smpboot-threads already provide this infrastructure. Sending UDP packets over loopback expects that the packet is processed within the call. Delaying it by handing it over to the thread hurts performance. It is not beneficial to the outcome if the context switch happens immediately after enqueue or after a while to process a few packets in a batch. There is no need to always use the thread if the backlog NAPI is requested on the local CPU. This restores the loopback throuput. The performance drops mostly to the same value after enabling RPS on the loopback comparing the IPI and the tread result. Create NAPI-threads for backlog if request during boot. The thread runs the inner loop from napi_threaded_poll(), the wait part is different. It checks for NAPI_STATE_SCHED (the backlog NAPI can not be disabled). The NAPI threads for backlog are optional, it has to be enabled via the boot argument "thread_backlog_napi". It is mandatory for PREEMPT_RT to avoid the wakeup of ksoftirqd from the IPI. Acked-by: Jakub Kicinski Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Paolo Abeni --- net/core/dev.c | 148 +++++++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 113 insertions(+), 35 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 98d7d339a8c1..ea574df49640 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -78,6 +78,7 @@ #include #include #include +#include #include #include #include @@ -197,6 +198,31 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex) return &net->dev_index_head[ifindex & (NETDEV_HASHENTRIES - 1)]; } +#ifndef CONFIG_PREEMPT_RT + +static DEFINE_STATIC_KEY_FALSE(use_backlog_threads_key); + +static int __init setup_backlog_napi_threads(char *arg) +{ + static_branch_enable(&use_backlog_threads_key); + return 0; +} +early_param("thread_backlog_napi", setup_backlog_napi_threads); + +static bool use_backlog_threads(void) +{ + return static_branch_unlikely(&use_backlog_threads_key); +} + +#else + +static bool use_backlog_threads(void) +{ + return true; +} + +#endif + static inline void rps_lock_irqsave(struct softnet_data *sd, unsigned long *flags) { @@ -4404,6 +4430,7 @@ EXPORT_SYMBOL(__dev_direct_xmit); /************************************************************************* * Receiver routines *************************************************************************/ +static DEFINE_PER_CPU(struct task_struct *, backlog_napi); unsigned int sysctl_skb_defer_max __read_mostly = 64; int weight_p __read_mostly = 64; /* old backlog weight */ @@ -4427,12 +4454,16 @@ static inline void ____napi_schedule(struct softnet_data *sd, */ thread = READ_ONCE(napi->thread); if (thread) { + if (use_backlog_threads() && thread == raw_cpu_read(backlog_napi)) + goto use_local_napi; + set_bit(NAPI_STATE_SCHED_THREADED, &napi->state); wake_up_process(thread); return; } } +use_local_napi: list_add_tail(&napi->poll_list, &sd->poll_list); WRITE_ONCE(napi->list_owner, smp_processor_id()); /* If not called from net_rx_action() @@ -4672,6 +4703,11 @@ static void napi_schedule_rps(struct softnet_data *sd) #ifdef CONFIG_RPS if (sd != mysd) { + if (use_backlog_threads()) { + __napi_schedule_irqoff(&sd->backlog); + return; + } + sd->rps_ipi_next = mysd->rps_ipi_list; mysd->rps_ipi_list = sd; @@ -5931,7 +5967,7 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd) #ifdef CONFIG_RPS struct softnet_data *remsd = sd->rps_ipi_list; - if (remsd) { + if (!use_backlog_threads() && remsd) { sd->rps_ipi_list = NULL; local_irq_enable(); @@ -5946,7 +5982,7 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd) static bool sd_has_rps_ipi_waiting(struct softnet_data *sd) { #ifdef CONFIG_RPS - return sd->rps_ipi_list != NULL; + return !use_backlog_threads() && sd->rps_ipi_list; #else return false; #endif @@ -5990,7 +6026,7 @@ static int process_backlog(struct napi_struct *napi, int quota) * We can use a plain write instead of clear_bit(), * and we dont need an smp_mb() memory barrier. */ - napi->state = 0; + napi->state &= NAPIF_STATE_THREADED; again = false; } else { skb_queue_splice_tail_init(&sd->input_pkt_queue, @@ -6726,43 +6762,48 @@ static int napi_thread_wait(struct napi_struct *napi) return -1; } -static int napi_threaded_poll(void *data) +static void napi_threaded_poll_loop(struct napi_struct *napi) { - struct napi_struct *napi = data; struct softnet_data *sd; - void *have; - - while (!napi_thread_wait(napi)) { - unsigned long last_qs = jiffies; + unsigned long last_qs = jiffies; - for (;;) { - bool repoll = false; + for (;;) { + bool repoll = false; + void *have; - local_bh_disable(); - sd = this_cpu_ptr(&softnet_data); - sd->in_napi_threaded_poll = true; + local_bh_disable(); + sd = this_cpu_ptr(&softnet_data); + sd->in_napi_threaded_poll = true; - have = netpoll_poll_lock(napi); - __napi_poll(napi, &repoll); - netpoll_poll_unlock(have); + have = netpoll_poll_lock(napi); + __napi_poll(napi, &repoll); + netpoll_poll_unlock(have); - sd->in_napi_threaded_poll = false; - barrier(); + sd->in_napi_threaded_poll = false; + barrier(); - if (sd_has_rps_ipi_waiting(sd)) { - local_irq_disable(); - net_rps_action_and_irq_enable(sd); - } - skb_defer_free_flush(sd); - local_bh_enable(); + if (sd_has_rps_ipi_waiting(sd)) { + local_irq_disable(); + net_rps_action_and_irq_enable(sd); + } + skb_defer_free_flush(sd); + local_bh_enable(); - if (!repoll) - break; + if (!repoll) + break; - rcu_softirq_qs_periodic(last_qs); - cond_resched(); - } + rcu_softirq_qs_periodic(last_qs); + cond_resched(); } +} + +static int napi_threaded_poll(void *data) +{ + struct napi_struct *napi = data; + + while (!napi_thread_wait(napi)) + napi_threaded_poll_loop(napi); + return 0; } @@ -11363,7 +11404,7 @@ static int dev_cpu_dead(unsigned int oldcpu) list_del_init(&napi->poll_list); if (napi->poll == process_backlog) - napi->state = 0; + napi->state &= NAPIF_STATE_THREADED; else ____napi_schedule(sd, napi); } @@ -11371,12 +11412,14 @@ static int dev_cpu_dead(unsigned int oldcpu) raise_softirq_irqoff(NET_TX_SOFTIRQ); local_irq_enable(); + if (!use_backlog_threads()) { #ifdef CONFIG_RPS - remsd = oldsd->rps_ipi_list; - oldsd->rps_ipi_list = NULL; + remsd = oldsd->rps_ipi_list; + oldsd->rps_ipi_list = NULL; #endif - /* send out pending IPI's on offline CPU */ - net_rps_send_ipi(remsd); + /* send out pending IPI's on offline CPU */ + net_rps_send_ipi(remsd); + } /* Process offline CPU's input_pkt_queue */ while ((skb = __skb_dequeue(&oldsd->process_queue))) { @@ -11715,6 +11758,38 @@ static int net_page_pool_create(int cpuid) return 0; } +static int backlog_napi_should_run(unsigned int cpu) +{ + struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu); + struct napi_struct *napi = &sd->backlog; + + return test_bit(NAPI_STATE_SCHED_THREADED, &napi->state); +} + +static void run_backlog_napi(unsigned int cpu) +{ + struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu); + + napi_threaded_poll_loop(&sd->backlog); +} + +static void backlog_napi_setup(unsigned int cpu) +{ + struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu); + struct napi_struct *napi = &sd->backlog; + + napi->thread = this_cpu_read(backlog_napi); + set_bit(NAPI_STATE_THREADED, &napi->state); +} + +static struct smp_hotplug_thread backlog_threads = { + .store = &backlog_napi, + .thread_should_run = backlog_napi_should_run, + .thread_fn = run_backlog_napi, + .thread_comm = "backlog_napi/%u", + .setup = backlog_napi_setup, +}; + /* * This is called single threaded during boot, so no need * to take the rtnl semaphore. @@ -11766,10 +11841,13 @@ static int __init net_dev_init(void) init_gro_hash(&sd->backlog); sd->backlog.poll = process_backlog; sd->backlog.weight = weight_p; + INIT_LIST_HEAD(&sd->backlog.poll_list); if (net_page_pool_create(i)) goto out; } + if (use_backlog_threads()) + smpboot_register_percpu_thread(&backlog_threads); dev_boot_phase = 0; -- cgit v1.2.3 From 80d2eefcb4c84aa9018b2a997ab3a4c567bc821a Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Mon, 25 Mar 2024 08:40:30 +0100 Subject: net: Use backlog-NAPI to clean up the defer_list. The defer_list is a per-CPU list which is used to free skbs outside of the socket lock and on the CPU on which they have been allocated. The list is processed during NAPI callbacks so ideally the list is cleaned up. Should the amount of skbs on the list exceed a certain water mark then the softirq is triggered remotely on the target CPU by invoking a remote function call. The raise of the softirqs via a remote function call leads to waking the ksoftirqd on PREEMPT_RT which is undesired. The backlog-NAPI threads already provide the infrastructure which can be utilized to perform the cleanup of the defer_list. The NAPI state is updated with the input_pkt_queue.lock acquired. It order not to break the state, it is needed to also wake the backlog-NAPI thread with the lock held. This requires to acquire the use the lock in rps_lock_irq*() if the backlog-NAPI threads are used even with RPS disabled. Move the logic of remotely starting softirqs to clean up the defer_list into kick_defer_list_purge(). Make sure a lock is held in rps_lock_irq*() if backlog-NAPI threads are used. Schedule backlog-NAPI for defer_list cleanup if backlog-NAPI is available. Acked-by: Jakub Kicinski Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Paolo Abeni --- include/linux/netdevice.h | 1 + net/core/dev.c | 25 +++++++++++++++++++++---- net/core/skbuff.c | 4 ++-- 3 files changed, 24 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index cb37817d6382..e41d30ebaca6 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3287,6 +3287,7 @@ static inline void dev_xmit_recursion_dec(void) __this_cpu_dec(softnet_data.xmit.recursion); } +void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu); void __netif_schedule(struct Qdisc *q); void netif_schedule_queue(struct netdev_queue *txq); diff --git a/net/core/dev.c b/net/core/dev.c index ea574df49640..85e11dabab48 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -226,7 +226,7 @@ static bool use_backlog_threads(void) static inline void rps_lock_irqsave(struct softnet_data *sd, unsigned long *flags) { - if (IS_ENABLED(CONFIG_RPS)) + if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_lock_irqsave(&sd->input_pkt_queue.lock, *flags); else if (!IS_ENABLED(CONFIG_PREEMPT_RT)) local_irq_save(*flags); @@ -234,7 +234,7 @@ static inline void rps_lock_irqsave(struct softnet_data *sd, static inline void rps_lock_irq_disable(struct softnet_data *sd) { - if (IS_ENABLED(CONFIG_RPS)) + if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_lock_irq(&sd->input_pkt_queue.lock); else if (!IS_ENABLED(CONFIG_PREEMPT_RT)) local_irq_disable(); @@ -243,7 +243,7 @@ static inline void rps_lock_irq_disable(struct softnet_data *sd) static inline void rps_unlock_irq_restore(struct softnet_data *sd, unsigned long *flags) { - if (IS_ENABLED(CONFIG_RPS)) + if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_unlock_irqrestore(&sd->input_pkt_queue.lock, *flags); else if (!IS_ENABLED(CONFIG_PREEMPT_RT)) local_irq_restore(*flags); @@ -251,7 +251,7 @@ static inline void rps_unlock_irq_restore(struct softnet_data *sd, static inline void rps_unlock_irq_enable(struct softnet_data *sd) { - if (IS_ENABLED(CONFIG_RPS)) + if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_unlock_irq(&sd->input_pkt_queue.lock); else if (!IS_ENABLED(CONFIG_PREEMPT_RT)) local_irq_enable(); @@ -4722,6 +4722,23 @@ static void napi_schedule_rps(struct softnet_data *sd) __napi_schedule_irqoff(&mysd->backlog); } +void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu) +{ + unsigned long flags; + + if (use_backlog_threads()) { + rps_lock_irqsave(sd, &flags); + + if (!__test_and_set_bit(NAPI_STATE_SCHED, &sd->backlog.state)) + __napi_schedule_irqoff(&sd->backlog); + + rps_unlock_irq_restore(sd, &flags); + + } else if (!cmpxchg(&sd->defer_ipi_scheduled, 0, 1)) { + smp_call_function_single_async(cpu, &sd->defer_csd); + } +} + #ifdef CONFIG_NET_FLOW_LIMIT int netdev_flow_limit_table_len __read_mostly = (1 << 12); #endif diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b99127712e67..17617c29be2d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -7039,8 +7039,8 @@ nodefer: __kfree_skb(skb); /* Make sure to trigger NET_RX_SOFTIRQ on the remote CPU * if we are unlucky enough (this seems very unlikely). */ - if (unlikely(kick) && !cmpxchg(&sd->defer_ipi_scheduled, 0, 1)) - smp_call_function_single_async(cpu, &sd->defer_csd); + if (unlikely(kick)) + kick_defer_list_purge(sd, cpu); } static void skb_splice_csum_page(struct sk_buff *skb, struct page *page, -- cgit v1.2.3 From 765b11f8f4e20b7433e4ba4a3e9106a0d59501ed Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Mon, 25 Mar 2024 08:40:31 +0100 Subject: net: Rename rps_lock to backlog_lock. The rps_lock.*() functions use the inner lock of a sk_buff_head for locking. This lock is used if RPS is enabled, otherwise the list is accessed lockless and disabling interrupts is enough for the synchronisation because it is only accessed CPU local. Not only the list is protected but also the NAPI state protected. With the addition of backlog threads, the lock is also needed because of the cross CPU access even without RPS. The clean up of the defer_list list is also done via backlog threads (if enabled). It has been suggested to rename the locking function since it is no longer just RPS. Rename the rps_lock*() functions to backlog_lock*(). Suggested-by: Jakub Kicinski Acked-by: Jakub Kicinski Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Paolo Abeni --- net/core/dev.c | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 85e11dabab48..9ce34164bcb1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -223,8 +223,8 @@ static bool use_backlog_threads(void) #endif -static inline void rps_lock_irqsave(struct softnet_data *sd, - unsigned long *flags) +static inline void backlog_lock_irq_save(struct softnet_data *sd, + unsigned long *flags) { if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_lock_irqsave(&sd->input_pkt_queue.lock, *flags); @@ -232,7 +232,7 @@ static inline void rps_lock_irqsave(struct softnet_data *sd, local_irq_save(*flags); } -static inline void rps_lock_irq_disable(struct softnet_data *sd) +static inline void backlog_lock_irq_disable(struct softnet_data *sd) { if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_lock_irq(&sd->input_pkt_queue.lock); @@ -240,8 +240,8 @@ static inline void rps_lock_irq_disable(struct softnet_data *sd) local_irq_disable(); } -static inline void rps_unlock_irq_restore(struct softnet_data *sd, - unsigned long *flags) +static inline void backlog_unlock_irq_restore(struct softnet_data *sd, + unsigned long *flags) { if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_unlock_irqrestore(&sd->input_pkt_queue.lock, *flags); @@ -249,7 +249,7 @@ static inline void rps_unlock_irq_restore(struct softnet_data *sd, local_irq_restore(*flags); } -static inline void rps_unlock_irq_enable(struct softnet_data *sd) +static inline void backlog_unlock_irq_enable(struct softnet_data *sd) { if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads()) spin_unlock_irq(&sd->input_pkt_queue.lock); @@ -4727,12 +4727,12 @@ void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu) unsigned long flags; if (use_backlog_threads()) { - rps_lock_irqsave(sd, &flags); + backlog_lock_irq_save(sd, &flags); if (!__test_and_set_bit(NAPI_STATE_SCHED, &sd->backlog.state)) __napi_schedule_irqoff(&sd->backlog); - rps_unlock_irq_restore(sd, &flags); + backlog_unlock_irq_restore(sd, &flags); } else if (!cmpxchg(&sd->defer_ipi_scheduled, 0, 1)) { smp_call_function_single_async(cpu, &sd->defer_csd); @@ -4794,7 +4794,7 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, reason = SKB_DROP_REASON_NOT_SPECIFIED; sd = &per_cpu(softnet_data, cpu); - rps_lock_irqsave(sd, &flags); + backlog_lock_irq_save(sd, &flags); if (!netif_running(skb->dev)) goto drop; qlen = skb_queue_len(&sd->input_pkt_queue); @@ -4804,7 +4804,7 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, enqueue: __skb_queue_tail(&sd->input_pkt_queue, skb); input_queue_tail_incr_save(sd, qtail); - rps_unlock_irq_restore(sd, &flags); + backlog_unlock_irq_restore(sd, &flags); return NET_RX_SUCCESS; } @@ -4819,7 +4819,7 @@ enqueue: drop: sd->dropped++; - rps_unlock_irq_restore(sd, &flags); + backlog_unlock_irq_restore(sd, &flags); dev_core_stats_rx_dropped_inc(skb->dev); kfree_skb_reason(skb, reason); @@ -5885,7 +5885,7 @@ static void flush_backlog(struct work_struct *work) local_bh_disable(); sd = this_cpu_ptr(&softnet_data); - rps_lock_irq_disable(sd); + backlog_lock_irq_disable(sd); skb_queue_walk_safe(&sd->input_pkt_queue, skb, tmp) { if (skb->dev->reg_state == NETREG_UNREGISTERING) { __skb_unlink(skb, &sd->input_pkt_queue); @@ -5893,7 +5893,7 @@ static void flush_backlog(struct work_struct *work) input_queue_head_incr(sd); } } - rps_unlock_irq_enable(sd); + backlog_unlock_irq_enable(sd); skb_queue_walk_safe(&sd->process_queue, skb, tmp) { if (skb->dev->reg_state == NETREG_UNREGISTERING) { @@ -5911,14 +5911,14 @@ static bool flush_required(int cpu) struct softnet_data *sd = &per_cpu(softnet_data, cpu); bool do_flush; - rps_lock_irq_disable(sd); + backlog_lock_irq_disable(sd); /* as insertion into process_queue happens with the rps lock held, * process_queue access may race only with dequeue */ do_flush = !skb_queue_empty(&sd->input_pkt_queue) || !skb_queue_empty_lockless(&sd->process_queue); - rps_unlock_irq_enable(sd); + backlog_unlock_irq_enable(sd); return do_flush; #endif @@ -6033,7 +6033,7 @@ static int process_backlog(struct napi_struct *napi, int quota) } - rps_lock_irq_disable(sd); + backlog_lock_irq_disable(sd); if (skb_queue_empty(&sd->input_pkt_queue)) { /* * Inline a custom version of __napi_complete(). @@ -6049,7 +6049,7 @@ static int process_backlog(struct napi_struct *napi, int quota) skb_queue_splice_tail_init(&sd->input_pkt_queue, &sd->process_queue); } - rps_unlock_irq_enable(sd); + backlog_unlock_irq_enable(sd); } return work; -- cgit v1.2.3 From 6e06312035032924fc97f2050bfe85e63ca26514 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 25 Mar 2024 13:41:55 +0000 Subject: net: remove skb_free_datagram_locked() Last user of skb_free_datagram_locked() went away in 2016 with commit 850cbaddb52d ("udp: use it's own memory accounting schema"). Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240325134155.620531-1-edumazet@google.com Signed-off-by: Paolo Abeni --- include/linux/skbuff.h | 6 ------ net/core/datagram.c | 19 ------------------- 2 files changed, 25 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 0c7c67b3a87b..b945af8a6208 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4063,12 +4063,6 @@ int skb_copy_datagram_from_iter(struct sk_buff *skb, int offset, struct iov_iter *from, int len); int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *frm); void skb_free_datagram(struct sock *sk, struct sk_buff *skb); -void __skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb, int len); -static inline void skb_free_datagram_locked(struct sock *sk, - struct sk_buff *skb) -{ - __skb_free_datagram_locked(sk, skb, 0); -} int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags); int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len); int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len); diff --git a/net/core/datagram.c b/net/core/datagram.c index a8b625abe242..e614cfd8e14a 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -324,25 +324,6 @@ void skb_free_datagram(struct sock *sk, struct sk_buff *skb) } EXPORT_SYMBOL(skb_free_datagram); -void __skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb, int len) -{ - bool slow; - - if (!skb_unref(skb)) { - sk_peek_offset_bwd(sk, len); - return; - } - - slow = lock_sock_fast(sk); - sk_peek_offset_bwd(sk, len); - skb_orphan(skb); - unlock_sock_fast(sk, slow); - - /* skb is now orphaned, can be freed outside of locked section */ - __kfree_skb(skb); -} -EXPORT_SYMBOL(__skb_free_datagram_locked); - int __sk_queue_drop_skb(struct sock *sk, struct sk_buff_head *sk_queue, struct sk_buff *skb, unsigned int flags, void (*destructor)(struct sock *sk, -- cgit v1.2.3 From 341ee1a584c8f55068cc5d7024cf1711ab81e388 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Mon, 25 Mar 2024 17:06:35 +0100 Subject: net: pin system percpu page_pools to the corresponding NUMA nodes System page_pools are percpu and one instance can be used only on one CPU. %NUMA_NO_NODE is fine for allocating pages, as the PP core always allocates local pages in this case. But for the struct &page_pool itself, this node ID means they are allocated on the boot CPU, which may belong to a different node than the target CPU. Pin system page_pools to the corresponding nodes when creating, so that all the allocated data will always be local. Use cpu_to_mem() to account memless nodes. Nodes != 0 win some Kpps when testing with xdp-trafficgen. Signed-off-by: Alexander Lobakin Link: https://lore.kernel.org/r/20240325160635.3215855-1-aleksander.lobakin@intel.com Signed-off-by: Jakub Kicinski --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 9ce34164bcb1..5d36a634f468 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -11762,7 +11762,7 @@ static int net_page_pool_create(int cpuid) struct page_pool_params page_pool_params = { .pool_size = SYSTEM_PERCPU_PAGE_POOL_SIZE, .flags = PP_FLAG_SYSTEM_POOL, - .nid = NUMA_NO_NODE, + .nid = cpu_to_mem(cpuid), }; struct page_pool *pp_ptr; -- cgit v1.2.3 From 6e9b01909a811555ff3326cf80a5847169c57806 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 Mar 2024 21:02:12 -0700 Subject: net: remove gfp_mask from napi_alloc_skb() __napi_alloc_skb() is napi_alloc_skb() with the added flexibility of choosing gfp_mask. This is a NAPI function, so GFP_ATOMIC is implied. The only practical choice the caller has is whether to set __GFP_NOWARN. But that's a false choice, too, allocation failures in atomic context will happen, and printing warnings in logs, effectively for a packet drop, is both too much and very likely non-actionable. This leads me to a conclusion that most uses of napi_alloc_skb() are simply misguided, and should use __GFP_NOWARN in the first place. We also have a "standard" way of reporting allocation failures via the queue stat API (qstats::rx-alloc-fail). The direct motivation for this patch is that one of the drivers used at Meta calls napi_alloc_skb() (so prior to this patch without __GFP_NOWARN), and the resulting OOM warning is the top networking warning in our fleet. Reviewed-by: Alexander Lobakin Reviewed-by: Simon Horman Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240327040213.3153864-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/mm/page_frags.rst | 2 +- Documentation/translations/zh_CN/mm/page_frags.rst | 2 +- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 4 +--- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 3 +-- drivers/net/ethernet/intel/iavf/iavf_txrx.c | 4 +--- drivers/net/ethernet/intel/ice/ice_txrx.c | 3 +-- drivers/net/ethernet/intel/ice/ice_xsk.c | 3 +-- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 5 ++--- drivers/net/ethernet/intel/igc/igc_main.c | 3 +-- drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 3 +-- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 5 ++--- include/linux/skbuff.h | 8 +------- net/core/skbuff.c | 9 ++++----- 13 files changed, 18 insertions(+), 36 deletions(-) (limited to 'net') diff --git a/Documentation/mm/page_frags.rst b/Documentation/mm/page_frags.rst index a81617e688a8..503ca6cdb804 100644 --- a/Documentation/mm/page_frags.rst +++ b/Documentation/mm/page_frags.rst @@ -25,7 +25,7 @@ to be disabled when executing the fragment allocation. The network stack uses two separate caches per CPU to handle fragment allocation. The netdev_alloc_cache is used by callers making use of the netdev_alloc_frag and __netdev_alloc_skb calls. The napi_alloc_cache is -used by callers of the __napi_alloc_frag and __napi_alloc_skb calls. The +used by callers of the __napi_alloc_frag and napi_alloc_skb calls. The main difference between these two calls is the context in which they may be called. The "netdev" prefixed functions are usable in any context as these functions will disable interrupts, while the "napi" prefixed functions are diff --git a/Documentation/translations/zh_CN/mm/page_frags.rst b/Documentation/translations/zh_CN/mm/page_frags.rst index 20bd3fafdc8c..a5b22486a913 100644 --- a/Documentation/translations/zh_CN/mm/page_frags.rst +++ b/Documentation/translations/zh_CN/mm/page_frags.rst @@ -25,7 +25,7 @@ sk_buff->head使用，或者用于skb_shared_info的 “frags” 部分。 网络堆栈在每个CPU使用两个独立的缓存来处理碎片分配。netdev_alloc_cache被使用 netdev_alloc_frag和__netdev_alloc_skb调用的调用者使用。napi_alloc_cache -被调用__napi_alloc_frag和__napi_alloc_skb的调用者使用。这两个调用的主要区别是 +被调用__napi_alloc_frag和napi_alloc_skb的调用者使用。这两个调用的主要区别是 它们可能被调用的环境。“netdev” 前缀的函数可以在任何上下文中使用，因为这些函数 将禁用中断，而 ”napi“ 前缀的函数只可以在softirq上下文中使用。 diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index 0d7177083708..ac2fcc5ac595 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -2144,9 +2144,7 @@ static struct sk_buff *i40e_construct_skb(struct i40e_ring *rx_ring, */ /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, - I40E_RX_HDR_SIZE, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&rx_ring->q_vector->napi, I40E_RX_HDR_SIZE); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c index 11500003af0d..a85b425794df 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c @@ -301,8 +301,7 @@ static struct sk_buff *i40e_construct_skb_zc(struct i40e_ring *rx_ring, net_prefetch(xdp->data_meta); /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, totalsize, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&rx_ring->q_vector->napi, totalsize); if (unlikely(!skb)) goto out; diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c index b71484c87a84..32bb604a1382 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c +++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c @@ -1334,9 +1334,7 @@ static struct sk_buff *iavf_construct_skb(struct iavf_ring *rx_ring, net_prefetch(va); /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, - IAVF_RX_HDR_SIZE, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&rx_ring->q_vector->napi, IAVF_RX_HDR_SIZE); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 97d41d6ebf1f..8bb743f78fcb 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1051,8 +1051,7 @@ ice_construct_skb(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) } /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&rx_ring->q_vector->napi, ICE_RX_HDR_SIZE); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 1857220d27fe..aa81d1162b81 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -555,8 +555,7 @@ ice_construct_skb_zc(struct ice_rx_ring *rx_ring, struct xdp_buff *xdp) } net_prefetch(xdp->data_meta); - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, totalsize, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&rx_ring->q_vector->napi, totalsize); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index 6dd7a66bb897..f940f650cd78 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -3005,8 +3005,7 @@ struct sk_buff *idpf_rx_construct_skb(struct idpf_queue *rxq, /* prefetch first cache line of first page */ net_prefetch(va); /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rxq->q_vector->napi, IDPF_RX_HDR_SIZE, - GFP_ATOMIC); + skb = napi_alloc_skb(&rxq->q_vector->napi, IDPF_RX_HDR_SIZE); if (unlikely(!skb)) { idpf_rx_put_page(rx_buf); @@ -3060,7 +3059,7 @@ static struct sk_buff *idpf_rx_hdr_construct_skb(struct idpf_queue *rxq, struct sk_buff *skb; /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rxq->q_vector->napi, size, GFP_ATOMIC); + skb = napi_alloc_skb(&rxq->q_vector->napi, size); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 35ad40a803cb..0cd923c8ac9b 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -2712,8 +2712,7 @@ static struct sk_buff *igc_construct_skb_zc(struct igc_ring *ring, net_prefetch(xdp->data_meta); - skb = __napi_alloc_skb(&ring->q_vector->napi, totalsize, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&ring->q_vector->napi, totalsize); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c index d34d715c59eb..397cb773fabb 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c @@ -220,8 +220,7 @@ static struct sk_buff *ixgbe_construct_skb_zc(struct ixgbe_ring *rx_ring, net_prefetch(xdp->data_meta); /* allocate a skb to store the frags */ - skb = __napi_alloc_skb(&rx_ring->q_vector->napi, totalsize, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&rx_ring->q_vector->napi, totalsize); if (unlikely(!skb)) return NULL; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 24cd80490d19..bcdde68a099a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -5109,9 +5109,8 @@ static struct sk_buff *stmmac_construct_skb_zc(struct stmmac_channel *ch, unsigned int datasize = xdp->data_end - xdp->data; struct sk_buff *skb; - skb = __napi_alloc_skb(&ch->rxtx_napi, - xdp->data_end - xdp->data_hard_start, - GFP_ATOMIC | __GFP_NOWARN); + skb = napi_alloc_skb(&ch->rxtx_napi, + xdp->data_end - xdp->data_hard_start); if (unlikely(!skb)) return NULL; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 517e546a120a..b7f1ecdaec38 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3350,13 +3350,7 @@ static inline void *napi_alloc_frag_align(unsigned int fragsz, return __napi_alloc_frag_align(fragsz, -align); } -struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, - unsigned int length, gfp_t gfp_mask); -static inline struct sk_buff *napi_alloc_skb(struct napi_struct *napi, - unsigned int length) -{ - return __napi_alloc_skb(napi, length, GFP_ATOMIC); -} +struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int length); void napi_consume_skb(struct sk_buff *skb, int budget); void napi_skb_free_stolen_head(struct sk_buff *skb); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 17617c29be2d..a1be84be5d35 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -775,10 +775,9 @@ skb_fail: EXPORT_SYMBOL(__netdev_alloc_skb); /** - * __napi_alloc_skb - allocate skbuff for rx in a specific NAPI instance + * napi_alloc_skb - allocate skbuff for rx in a specific NAPI instance * @napi: napi instance this buffer was allocated for * @len: length to allocate - * @gfp_mask: get_free_pages mask, passed to alloc_skb and alloc_pages * * Allocate a new sk_buff for use in NAPI receive. This buffer will * attempt to allocate the head from a special reserved region used @@ -787,9 +786,9 @@ EXPORT_SYMBOL(__netdev_alloc_skb); * * %NULL is returned if there is no free memory. */ -struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len, - gfp_t gfp_mask) +struct sk_buff *napi_alloc_skb(struct napi_struct *napi, unsigned int len) { + gfp_t gfp_mask = GFP_ATOMIC | __GFP_NOWARN; struct napi_alloc_cache *nc; struct sk_buff *skb; bool pfmemalloc; @@ -860,7 +859,7 @@ skb_success: skb_fail: return skb; } -EXPORT_SYMBOL(__napi_alloc_skb); +EXPORT_SYMBOL(napi_alloc_skb); void skb_add_rx_frag_netmem(struct sk_buff *skb, int i, netmem_ref netmem, int off, int size, unsigned int truesize) -- cgit v1.2.3 From 5311591fbb349fe9f5c555dcba3b13a5831aa72d Mon Sep 17 00:00:00 2001 From: Anton Protopopov Date: Tue, 26 Mar 2024 10:17:40 +0000 Subject: bpf: Add support for passing mark with bpf_fib_lookup Extend the bpf_fib_lookup() helper by making it to utilize mark if the BPF_FIB_LOOKUP_MARK flag is set. In order to pass the mark the four bytes of struct bpf_fib_lookup are used, shared with the output-only smac/dmac fields. Signed-off-by: Anton Protopopov Signed-off-by: Daniel Borkmann Reviewed-by: David Ahern Acked-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20240326101742.17421-2-aspsk@isovalent.com Signed-off-by: Alexei Starovoitov --- include/uapi/linux/bpf.h | 20 ++++++++++++++++++-- net/core/filter.c | 12 +++++++++--- tools/include/uapi/linux/bpf.h | 20 ++++++++++++++++++-- 3 files changed, 45 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 9585f5345353..96d57e483133 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -3394,6 +3394,10 @@ union bpf_attr { * for the nexthop. If the src addr cannot be derived, * **BPF_FIB_LKUP_RET_NO_SRC_ADDR** is returned. In this * case, *params*->dmac and *params*->smac are not set either. + * **BPF_FIB_LOOKUP_MARK** + * Use the mark present in *params*->mark for the fib lookup. + * This option should not be used with BPF_FIB_LOOKUP_DIRECT, + * as it only has meaning for full lookups. * * *ctx* is either **struct xdp_md** for XDP programs or * **struct sk_buff** tc cls_act programs. @@ -7120,6 +7124,7 @@ enum { BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2), BPF_FIB_LOOKUP_TBID = (1U << 3), BPF_FIB_LOOKUP_SRC = (1U << 4), + BPF_FIB_LOOKUP_MARK = (1U << 5), }; enum { @@ -7197,8 +7202,19 @@ struct bpf_fib_lookup { __u32 tbid; }; - __u8 smac[6]; /* ETH_ALEN */ - __u8 dmac[6]; /* ETH_ALEN */ + union { + /* input */ + struct { + __u32 mark; /* policy routing */ + /* 2 4-byte holes for input */ + }; + + /* output: source and dest mac */ + struct { + __u8 smac[6]; /* ETH_ALEN */ + __u8 dmac[6]; /* ETH_ALEN */ + }; + }; }; struct bpf_redir_neigh { diff --git a/net/core/filter.c b/net/core/filter.c index 0c66e4a3fc5b..1205dd777dc2 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5884,7 +5884,10 @@ static int bpf_ipv4_fib_lookup(struct net *net, struct bpf_fib_lookup *params, err = fib_table_lookup(tb, &fl4, &res, FIB_LOOKUP_NOREF); } else { - fl4.flowi4_mark = 0; + if (flags & BPF_FIB_LOOKUP_MARK) + fl4.flowi4_mark = params->mark; + else + fl4.flowi4_mark = 0; fl4.flowi4_secid = 0; fl4.flowi4_tun_key.tun_id = 0; fl4.flowi4_uid = sock_net_uid(net, NULL); @@ -6027,7 +6030,10 @@ static int bpf_ipv6_fib_lookup(struct net *net, struct bpf_fib_lookup *params, err = ipv6_stub->fib6_table_lookup(net, tb, oif, &fl6, &res, strict); } else { - fl6.flowi6_mark = 0; + if (flags & BPF_FIB_LOOKUP_MARK) + fl6.flowi6_mark = params->mark; + else + fl6.flowi6_mark = 0; fl6.flowi6_secid = 0; fl6.flowi6_tun_key.tun_id = 0; fl6.flowi6_uid = sock_net_uid(net, NULL); @@ -6105,7 +6111,7 @@ set_fwd_params: #define BPF_FIB_LOOKUP_MASK (BPF_FIB_LOOKUP_DIRECT | BPF_FIB_LOOKUP_OUTPUT | \ BPF_FIB_LOOKUP_SKIP_NEIGH | BPF_FIB_LOOKUP_TBID | \ - BPF_FIB_LOOKUP_SRC) + BPF_FIB_LOOKUP_SRC | BPF_FIB_LOOKUP_MARK) BPF_CALL_4(bpf_xdp_fib_lookup, struct xdp_buff *, ctx, struct bpf_fib_lookup *, params, int, plen, u32, flags) diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 9585f5345353..96d57e483133 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -3394,6 +3394,10 @@ union bpf_attr { * for the nexthop. If the src addr cannot be derived, * **BPF_FIB_LKUP_RET_NO_SRC_ADDR** is returned. In this * case, *params*->dmac and *params*->smac are not set either. + * **BPF_FIB_LOOKUP_MARK** + * Use the mark present in *params*->mark for the fib lookup. + * This option should not be used with BPF_FIB_LOOKUP_DIRECT, + * as it only has meaning for full lookups. * * *ctx* is either **struct xdp_md** for XDP programs or * **struct sk_buff** tc cls_act programs. @@ -7120,6 +7124,7 @@ enum { BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2), BPF_FIB_LOOKUP_TBID = (1U << 3), BPF_FIB_LOOKUP_SRC = (1U << 4), + BPF_FIB_LOOKUP_MARK = (1U << 5), }; enum { @@ -7197,8 +7202,19 @@ struct bpf_fib_lookup { __u32 tbid; }; - __u8 smac[6]; /* ETH_ALEN */ - __u8 dmac[6]; /* ETH_ALEN */ + union { + /* input */ + struct { + __u32 mark; /* policy routing */ + /* 2 4-byte holes for input */ + }; + + /* output: source and dest mac */ + struct { + __u8 smac[6]; /* ETH_ALEN */ + __u8 dmac[6]; /* ETH_ALEN */ + }; + }; }; struct bpf_redir_neigh { -- cgit v1.2.3 From 59b418c7063d30e0a3e1f592d47df096db83185c Mon Sep 17 00:00:00 2001 From: Anton Protopopov Date: Tue, 26 Mar 2024 10:17:42 +0000 Subject: bpf: Add a check for struct bpf_fib_lookup size The struct bpf_fib_lookup should not grow outside of its 64 bytes. Add a static assert to validate this. Suggested-by: David Ahern Signed-off-by: Anton Protopopov Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20240326101742.17421-4-aspsk@isovalent.com Signed-off-by: Alexei Starovoitov --- net/core/filter.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 1205dd777dc2..786d792ac816 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -87,6 +87,9 @@ #include "dev.h" +/* Keep the struct bpf_fib_lookup small so that it fits into a cacheline */ +static_assert(sizeof(struct bpf_fib_lookup) == 64, "struct bpf_fib_lookup size check"); + static const struct bpf_func_proto * bpf_sk_base_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog); -- cgit v1.2.3 From 3124591f686115aca25d772c2ccb7b1e202c3197 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 26 Mar 2024 09:21:50 -0700 Subject: bpf: add bpf_modify_return_test_tp() kfunc triggering tracepoint Add a simple bpf_modify_return_test_tp() kfunc, available to all program types, that is useful for various testing and benchmarking scenarios, as it allows to trigger most tracing BPF program types from BPF side, allowing to do complex testing and benchmarking scenarios. It is also attachable to for fmod_ret programs, making it a good and simple way to trigger fmod_ret program under test/benchmark. Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20240326162151.3981687-6-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- include/trace/events/bpf_test_run.h | 17 +++++++++++++++++ kernel/bpf/helpers.c | 1 + net/bpf/test_run.c | 8 ++++++++ 3 files changed, 26 insertions(+) (limited to 'net') diff --git a/include/trace/events/bpf_test_run.h b/include/trace/events/bpf_test_run.h index 265447e3f71a..0c924d39b7cb 100644 --- a/include/trace/events/bpf_test_run.h +++ b/include/trace/events/bpf_test_run.h @@ -7,6 +7,23 @@ #include +TRACE_EVENT(bpf_trigger_tp, + + TP_PROTO(int nonce), + + TP_ARGS(nonce), + + TP_STRUCT__entry( + __field(int, nonce) + ), + + TP_fast_assign( + __entry->nonce = nonce; + ), + + TP_printk("nonce %d", __entry->nonce) +); + DECLARE_EVENT_CLASS(bpf_test_finish, TP_PROTO(int *err), diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index f860adab3eb9..d9e7aca8ae9e 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -2625,6 +2625,7 @@ BTF_ID_FLAGS(func, bpf_dynptr_is_null) BTF_ID_FLAGS(func, bpf_dynptr_is_rdonly) BTF_ID_FLAGS(func, bpf_dynptr_size) BTF_ID_FLAGS(func, bpf_dynptr_clone) +BTF_ID_FLAGS(func, bpf_modify_return_test_tp) BTF_KFUNCS_END(common_btf_ids) static const struct btf_kfunc_id_set common_kfunc_set = { diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index 61efeadaff8d..f6aad4ed2ab2 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -575,6 +575,13 @@ __bpf_kfunc int bpf_modify_return_test2(int a, int *b, short c, int d, return a + *b + c + d + (long)e + f + g; } +__bpf_kfunc int bpf_modify_return_test_tp(int nonce) +{ + trace_bpf_trigger_tp(nonce); + + return nonce; +} + int noinline bpf_fentry_shadow_test(int a) { return a + 1; @@ -622,6 +629,7 @@ __bpf_kfunc_end_defs(); BTF_KFUNCS_START(bpf_test_modify_return_ids) BTF_ID_FLAGS(func, bpf_modify_return_test) BTF_ID_FLAGS(func, bpf_modify_return_test2) +BTF_ID_FLAGS(func, bpf_modify_return_test_tp) BTF_ID_FLAGS(func, bpf_fentry_test1, KF_SLEEPABLE) BTF_KFUNCS_END(bpf_test_modify_return_ids) -- cgit v1.2.3 From 42e4ebd390be8e0090d64d58433b6cba45d919e9 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Fri, 22 Mar 2024 12:14:32 -0700 Subject: bpf: Remove CONFIG_X86 and CONFIG_DYNAMIC_FTRACE guard from the tcp-cc kfuncs The commit 7aae231ac93b ("bpf: tcp: Limit calling some tcp cc functions to CONFIG_DYNAMIC_FTRACE") added CONFIG_DYNAMIC_FTRACE guard because pahole was only generating btf for ftrace-able functions. The ftrace filter had already been removed from pahole, so the CONFIG_DYNAMIC_FTRACE guard can be removed. The commit 569c484f9995 ("bpf: Limit static tcp-cc functions in the .BTF_ids list to x86") has added CONFIG_X86 guard because it failed the powerpc arch which prepended a "." to the local static function, so "cubictcp_init" becomes ".cubictcp_init". "__bpf_kfunc" has been added to kfunc since then and it uses the __unused compiler attribute. There is an existing "__bpf_kfunc static u32 bpf_kfunc_call_test_static_unused_arg(u32 arg, u32 unused)" test in bpf_testmod.c to cover the static kfunc case. cross compile on ppc64 with CONFIG_DYNAMIC_FTRACE disabled: > readelf -s vmlinux | grep cubictcp_ 56938: c00000000144fd00 184 FUNC LOCAL DEFAULT 2 cubictcp_cwnd_event [: 8] 56939: c00000000144fdb8 200 FUNC LOCAL DEFAULT 2 cubictcp_recalc_[...] [: 8] 56940: c00000000144fe80 296 FUNC LOCAL DEFAULT 2 cubictcp_init [: 8] 56941: c00000000144ffa8 228 FUNC LOCAL DEFAULT 2 cubictcp_state [: 8] 56942: c00000000145008c 1908 FUNC LOCAL DEFAULT 2 cubictcp_cong_avoid [: 8] 56943: c000000001450800 1644 FUNC LOCAL DEFAULT 2 cubictcp_acked [: 8] > bpftool btf dump file vmlinux | grep cubictcp_ [51540] FUNC 'cubictcp_acked' type_id=38137 linkage=static [51541] FUNC 'cubictcp_cong_avoid' type_id=38122 linkage=static [51543] FUNC 'cubictcp_cwnd_event' type_id=51542 linkage=static [51544] FUNC 'cubictcp_init' type_id=9186 linkage=static [51545] FUNC 'cubictcp_recalc_ssthresh' type_id=35021 linkage=static [51547] FUNC 'cubictcp_state' type_id=38141 linkage=static The patch removed both config guards. Cc: Jiri Olsa Signed-off-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20240322191433.4133280-1-martin.lau@linux.dev Signed-off-by: Alexei Starovoitov --- net/ipv4/tcp_bbr.c | 4 ---- net/ipv4/tcp_cubic.c | 4 ---- net/ipv4/tcp_dctcp.c | 4 ---- 3 files changed, 12 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index 05dc2d05bc7c..7e52ab24e40a 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -1156,8 +1156,6 @@ static struct tcp_congestion_ops tcp_bbr_cong_ops __read_mostly = { }; BTF_KFUNCS_START(tcp_bbr_check_kfunc_ids) -#ifdef CONFIG_X86 -#ifdef CONFIG_DYNAMIC_FTRACE BTF_ID_FLAGS(func, bbr_init) BTF_ID_FLAGS(func, bbr_main) BTF_ID_FLAGS(func, bbr_sndbuf_expand) @@ -1166,8 +1164,6 @@ BTF_ID_FLAGS(func, bbr_cwnd_event) BTF_ID_FLAGS(func, bbr_ssthresh) BTF_ID_FLAGS(func, bbr_min_tso_segs) BTF_ID_FLAGS(func, bbr_set_state) -#endif -#endif BTF_KFUNCS_END(tcp_bbr_check_kfunc_ids) static const struct btf_kfunc_id_set tcp_bbr_kfunc_set = { diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c index 44869ea089e3..5dbed91c6178 100644 --- a/net/ipv4/tcp_cubic.c +++ b/net/ipv4/tcp_cubic.c @@ -486,16 +486,12 @@ static struct tcp_congestion_ops cubictcp __read_mostly = { }; BTF_KFUNCS_START(tcp_cubic_check_kfunc_ids) -#ifdef CONFIG_X86 -#ifdef CONFIG_DYNAMIC_FTRACE BTF_ID_FLAGS(func, cubictcp_init) BTF_ID_FLAGS(func, cubictcp_recalc_ssthresh) BTF_ID_FLAGS(func, cubictcp_cong_avoid) BTF_ID_FLAGS(func, cubictcp_state) BTF_ID_FLAGS(func, cubictcp_cwnd_event) BTF_ID_FLAGS(func, cubictcp_acked) -#endif -#endif BTF_KFUNCS_END(tcp_cubic_check_kfunc_ids) static const struct btf_kfunc_id_set tcp_cubic_kfunc_set = { diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c index e33fbe4933e4..6b712a33d49f 100644 --- a/net/ipv4/tcp_dctcp.c +++ b/net/ipv4/tcp_dctcp.c @@ -261,16 +261,12 @@ static struct tcp_congestion_ops dctcp_reno __read_mostly = { }; BTF_KFUNCS_START(tcp_dctcp_check_kfunc_ids) -#ifdef CONFIG_X86 -#ifdef CONFIG_DYNAMIC_FTRACE BTF_ID_FLAGS(func, dctcp_init) BTF_ID_FLAGS(func, dctcp_update_alpha) BTF_ID_FLAGS(func, dctcp_cwnd_event) BTF_ID_FLAGS(func, dctcp_ssthresh) BTF_ID_FLAGS(func, dctcp_cwnd_undo) BTF_ID_FLAGS(func, dctcp_state) -#endif -#endif BTF_KFUNCS_END(tcp_dctcp_check_kfunc_ids) static const struct btf_kfunc_id_set tcp_dctcp_kfunc_set = { -- cgit v1.2.3 From f631ef39d81956a2ee69d25039781ceae1162f62 Mon Sep 17 00:00:00 2001 From: Asbjørn Sloth Tønnesen Date: Mon, 25 Mar 2024 20:47:34 +0000 Subject: net: sched: cls_api: add skip_sw counter MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Maintain a count of skip_sw filters. This counter is protected by the cb_lock, and is updated at the same time as offloadcnt. Signed-off-by: Asbjørn Sloth Tønnesen Reviewed-by: Jiri Pirko Reviewed-by: Simon Horman Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sch_generic.h | 1 + net/sched/cls_api.c | 4 ++++ 2 files changed, 5 insertions(+) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index cefe0c4bdae3..120a4ca6ec9b 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -471,6 +471,7 @@ struct tcf_block { struct flow_block flow_block; struct list_head owner_list; bool keep_dst; + atomic_t skipswcnt; /* Number of skip_sw filters */ atomic_t offloadcnt; /* Number of oddloaded filters */ unsigned int nooffloaddevcnt; /* Number of devs unable to do offload */ unsigned int lockeddevcnt; /* Number of devs that require rtnl lock. */ diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index ca5676b2668e..397c3d29659c 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -3483,6 +3483,8 @@ static void tcf_block_offload_inc(struct tcf_block *block, u32 *flags) if (*flags & TCA_CLS_FLAGS_IN_HW) return; *flags |= TCA_CLS_FLAGS_IN_HW; + if (tc_skip_sw(*flags)) + atomic_inc(&block->skipswcnt); atomic_inc(&block->offloadcnt); } @@ -3491,6 +3493,8 @@ static void tcf_block_offload_dec(struct tcf_block *block, u32 *flags) if (!(*flags & TCA_CLS_FLAGS_IN_HW)) return; *flags &= ~TCA_CLS_FLAGS_IN_HW; + if (tc_skip_sw(*flags)) + atomic_dec(&block->skipswcnt); atomic_dec(&block->offloadcnt); } -- cgit v1.2.3 From 2081fd3445fec6b9813c20e8b910c2abd6de31cb Mon Sep 17 00:00:00 2001 From: Asbjørn Sloth Tønnesen Date: Mon, 25 Mar 2024 20:47:35 +0000 Subject: net: sched: cls_api: add filter counter MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Maintain a count of filters per block. Counter updates are protected by cb_lock, which is also used to protect the offload counters. Signed-off-by: Asbjørn Sloth Tønnesen Reviewed-by: Simon Horman Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sch_generic.h | 2 ++ net/sched/cls_api.c | 19 +++++++++++++++++++ 2 files changed, 21 insertions(+) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 120a4ca6ec9b..eb3872c22fcd 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -422,6 +422,7 @@ struct tcf_proto { */ spinlock_t lock; bool deleting; + bool counted; refcount_t refcnt; struct rcu_head rcu; struct hlist_node destroy_ht_node; @@ -471,6 +472,7 @@ struct tcf_block { struct flow_block flow_block; struct list_head owner_list; bool keep_dst; + atomic_t filtercnt; /* Number of filters */ atomic_t skipswcnt; /* Number of skip_sw filters */ atomic_t offloadcnt; /* Number of oddloaded filters */ unsigned int nooffloaddevcnt; /* Number of devs unable to do offload */ diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 397c3d29659c..304a46ab0e0b 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -410,12 +410,30 @@ static void tcf_proto_get(struct tcf_proto *tp) refcount_inc(&tp->refcnt); } +static void tcf_block_filter_cnt_update(struct tcf_block *block, bool *counted, bool add) +{ + lockdep_assert_not_held(&block->cb_lock); + + down_write(&block->cb_lock); + if (*counted != add) { + if (add) { + atomic_inc(&block->filtercnt); + *counted = true; + } else { + atomic_dec(&block->filtercnt); + *counted = false; + } + } + up_write(&block->cb_lock); +} + static void tcf_chain_put(struct tcf_chain *chain); static void tcf_proto_destroy(struct tcf_proto *tp, bool rtnl_held, bool sig_destroy, struct netlink_ext_ack *extack) { tp->ops->destroy(tp, rtnl_held, extack); + tcf_block_filter_cnt_update(tp->chain->block, &tp->counted, false); if (sig_destroy) tcf_proto_signal_destroyed(tp->chain, tp); tcf_chain_put(tp->chain); @@ -2364,6 +2382,7 @@ replay: err = tp->ops->change(net, skb, tp, cl, t->tcm_handle, tca, &fh, flags, extack); if (err == 0) { + tcf_block_filter_cnt_update(block, &tp->counted, true); tfilter_notify(net, skb, n, tp, block, q, parent, fh, RTM_NEWTFILTER, false, rtnl_held, extack); tfilter_put(tp, fh); -- cgit v1.2.3 From 047f340b36fc550c0fc6a8947fc0a1f8e429e9ab Mon Sep 17 00:00:00 2001 From: Asbjørn Sloth Tønnesen Date: Mon, 25 Mar 2024 20:47:36 +0000 Subject: net: sched: make skip_sw actually skip software MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit TC filters come in 3 variants: - no flag (try to process in hardware, but fallback to software)) - skip_hw (do not process filter by hardware) - skip_sw (do not process filter by software) However skip_sw is implemented so that the skip_sw flag can first be checked, after it has been matched. IMHO it's common when using skip_sw, to use it on all rules. So if all filters in a block is skip_sw filters, then we can bail early, we can thus avoid having to match the filters, just to check for the skip_sw flag. This patch adds a bypass, for when only TC skip_sw rules are used. The bypass is guarded by a static key, to avoid harming other workloads. There are 3 ways that a packet from a skip_sw ruleset, can end up in the kernel path. Although the send packets to a non-existent chain way is only improved a few percents, then I believe it's worth optimizing the trap and fall-though use-cases. +----------------------------+--------+--------+--------+ | Test description | Pre- | Post- | Rel. | | | kpps | kpps | chg. | +----------------------------+--------+--------+--------+ | basic forwarding + notrack | 3589.3 | 3587.9 | 1.00x | | switch to eswitch mode | 3081.8 | 3094.7 | 1.00x | | add ingress qdisc | 3042.9 | 3063.6 | 1.01x | | tc forward in hw / skip_sw |37024.7 |37028.4 | 1.00x | | tc forward in sw / skip_hw | 3245.0 | 3245.3 | 1.00x | +----------------------------+--------+--------+--------+ | tests with only skip_sw rules below: | +----------------------------+--------+--------+--------+ | 1 non-matching rule | 2694.7 | 3058.7 | 1.14x | | 1 n-m rule, match trap | 2611.2 | 3323.1 | 1.27x | | 1 n-m rule, goto non-chain | 2886.8 | 2945.9 | 1.02x | | 5 non-matching rules | 1958.2 | 3061.3 | 1.56x | | 5 n-m rules, match trap | 1911.9 | 3327.0 | 1.74x | | 5 n-m rules, goto non-chain| 2883.1 | 2947.5 | 1.02x | | 10 non-matching rules | 1466.3 | 3062.8 | 2.09x | | 10 n-m rules, match trap | 1444.3 | 3317.9 | 2.30x | | 10 n-m rules,goto non-chain| 2883.1 | 2939.5 | 1.02x | | 25 non-matching rules | 838.5 | 3058.9 | 3.65x | | 25 n-m rules, match trap | 824.5 | 3323.0 | 4.03x | | 25 n-m rules,goto non-chain| 2875.8 | 2944.7 | 1.02x | | 50 non-matching rules | 488.1 | 3054.7 | 6.26x | | 50 n-m rules, match trap | 484.9 | 3318.5 | 6.84x | | 50 n-m rules,goto non-chain| 2884.1 | 2939.7 | 1.02x | +----------------------------+--------+--------+--------+ perf top (25 n-m skip_sw rules - pre patch): 20.39% [kernel] [k] __skb_flow_dissect 16.43% [kernel] [k] rhashtable_jhash2 10.58% [kernel] [k] fl_classify 10.23% [kernel] [k] fl_mask_lookup 4.79% [kernel] [k] memset_orig 2.58% [kernel] [k] tcf_classify 1.47% [kernel] [k] __x86_indirect_thunk_rax 1.42% [kernel] [k] __dev_queue_xmit 1.36% [kernel] [k] nft_do_chain 1.21% [kernel] [k] __rcu_read_lock perf top (25 n-m skip_sw rules - post patch): 5.12% [kernel] [k] __dev_queue_xmit 4.77% [kernel] [k] nft_do_chain 3.65% [kernel] [k] dev_gro_receive 3.41% [kernel] [k] check_preemption_disabled 3.14% [kernel] [k] mlx5e_skb_from_cqe_mpwrq_nonlinear 2.88% [kernel] [k] __netif_receive_skb_core.constprop.0 2.49% [kernel] [k] mlx5e_xmit 2.15% [kernel] [k] ip_forward 1.95% [kernel] [k] mlx5e_tc_restore_tunnel 1.92% [kernel] [k] vlan_gro_receive Test setup: DUT: Intel Xeon D-1518 (2.20GHz) w/ Nvidia/Mellanox ConnectX-6 Dx 2x100G Data rate measured on switch (Extreme X690), and DUT connected as a router on a stick, with pktgen and pktsink as VLANs. Pktgen-dpdk was in range 36.6-37.7 Mpps 64B packets across all tests. Full test data at https://files.fiberby.net/ast/2024/tc_skip_sw/v2_tests/ Signed-off-by: Asbjørn Sloth Tønnesen Reviewed-by: Simon Horman Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 9 +++++++++ include/net/sch_generic.h | 1 + net/core/dev.c | 10 ++++++++++ net/sched/cls_api.c | 18 ++++++++++++++++++ 4 files changed, 38 insertions(+) (limited to 'net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index a4ee43f493bb..41297bd38dff 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -74,6 +74,15 @@ static inline bool tcf_block_non_null_shared(struct tcf_block *block) return block && block->index; } +#ifdef CONFIG_NET_CLS_ACT +DECLARE_STATIC_KEY_FALSE(tcf_bypass_check_needed_key); + +static inline bool tcf_block_bypass_sw(struct tcf_block *block) +{ + return block && block->bypass_wanted; +} +#endif + static inline struct Qdisc *tcf_block_q(struct tcf_block *block) { WARN_ON(tcf_block_shared(block)); diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index eb3872c22fcd..76db6be16083 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -472,6 +472,7 @@ struct tcf_block { struct flow_block flow_block; struct list_head owner_list; bool keep_dst; + bool bypass_wanted; atomic_t filtercnt; /* Number of filters */ atomic_t skipswcnt; /* Number of skip_sw filters */ atomic_t offloadcnt; /* Number of oddloaded filters */ diff --git a/net/core/dev.c b/net/core/dev.c index 5d36a634f468..c136e80dea61 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2083,6 +2083,11 @@ void net_dec_egress_queue(void) EXPORT_SYMBOL_GPL(net_dec_egress_queue); #endif +#ifdef CONFIG_NET_CLS_ACT +DEFINE_STATIC_KEY_FALSE(tcf_bypass_check_needed_key); +EXPORT_SYMBOL(tcf_bypass_check_needed_key); +#endif + DEFINE_STATIC_KEY_FALSE(netstamp_needed_key); EXPORT_SYMBOL(netstamp_needed_key); #ifdef CONFIG_JUMP_LABEL @@ -3937,6 +3942,11 @@ static int tc_run(struct tcx_entry *entry, struct sk_buff *skb, if (!miniq) return ret; + if (static_branch_unlikely(&tcf_bypass_check_needed_key)) { + if (tcf_block_bypass_sw(miniq->block)) + return ret; + } + tc_skb_cb(skb)->mru = 0; tc_skb_cb(skb)->post_ct = false; tcf_set_drop_reason(skb, *drop_reason); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 304a46ab0e0b..db0653993632 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -410,6 +410,23 @@ static void tcf_proto_get(struct tcf_proto *tp) refcount_inc(&tp->refcnt); } +static void tcf_maintain_bypass(struct tcf_block *block) +{ + int filtercnt = atomic_read(&block->filtercnt); + int skipswcnt = atomic_read(&block->skipswcnt); + bool bypass_wanted = filtercnt > 0 && filtercnt == skipswcnt; + + if (bypass_wanted != block->bypass_wanted) { +#ifdef CONFIG_NET_CLS_ACT + if (bypass_wanted) + static_branch_inc(&tcf_bypass_check_needed_key); + else + static_branch_dec(&tcf_bypass_check_needed_key); +#endif + block->bypass_wanted = bypass_wanted; + } +} + static void tcf_block_filter_cnt_update(struct tcf_block *block, bool *counted, bool add) { lockdep_assert_not_held(&block->cb_lock); @@ -424,6 +441,7 @@ static void tcf_block_filter_cnt_update(struct tcf_block *block, bool *counted, *counted = false; } } + tcf_maintain_bypass(block); up_write(&block->cb_lock); } -- cgit v1.2.3 From af398bd0cb2151a581027a19bd995f72af2310ac Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Tue, 26 Mar 2024 15:29:52 +0800 Subject: net/smc: make smc_hash_sk/smc_unhash_sk static smc_hash_sk and smc_unhash_sk are only used in af_smc.c, so make them static and remove the output symbol. They can be called under the path .prot->hash()/unhash(). Signed-off-by: Zhengchao Shao Reviewed-by: Wen Gu Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- include/net/smc.h | 3 --- net/smc/af_smc.c | 6 ++---- 2 files changed, 2 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/smc.h b/include/net/smc.h index c9dcb30e3fd9..10684d0a33df 100644 --- a/include/net/smc.h +++ b/include/net/smc.h @@ -26,9 +26,6 @@ struct smc_hashinfo { struct hlist_head ht; }; -int smc_hash_sk(struct sock *sk); -void smc_unhash_sk(struct sock *sk); - /* SMCD/ISM device driver interface */ struct smcd_dmb { u64 dmb_tok; diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 4b52b3b159c0..e8dcd28a554c 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -177,7 +177,7 @@ static struct smc_hashinfo smc_v6_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v6_hashinfo.lock), }; -int smc_hash_sk(struct sock *sk) +static int smc_hash_sk(struct sock *sk) { struct smc_hashinfo *h = sk->sk_prot->h.smc_hash; struct hlist_head *head; @@ -191,9 +191,8 @@ int smc_hash_sk(struct sock *sk) return 0; } -EXPORT_SYMBOL_GPL(smc_hash_sk); -void smc_unhash_sk(struct sock *sk) +static void smc_unhash_sk(struct sock *sk) { struct smc_hashinfo *h = sk->sk_prot->h.smc_hash; @@ -202,7 +201,6 @@ void smc_unhash_sk(struct sock *sk) sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); write_unlock_bh(&h->lock); } -EXPORT_SYMBOL_GPL(smc_unhash_sk); /* This will be called before user really release sock_lock. So do the * work which we didn't do because of user hold the sock_lock in the -- cgit v1.2.3 From 50e2907ef8bb52cf80ecde9eec5c4dac07177146 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 27 Mar 2024 19:12:06 +0000 Subject: tcp/dccp: bypass empty buckets in inet_twsk_purge() TCP ehash table is often sparsely populated. inet_twsk_purge() spends too much time calling cond_resched(). This patch can reduce time spent in inet_twsk_purge() by 20x. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240327191206.508114-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/inet_timewait_sock.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index e8de45d34d56..b0cc07d9a568 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -266,12 +266,17 @@ EXPORT_SYMBOL_GPL(__inet_twsk_schedule); /* Remove all non full sockets (TIME_WAIT and NEW_SYN_RECV) for dead netns */ void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family) { + struct inet_ehash_bucket *head = &hashinfo->ehash[0]; + unsigned int ehash_mask = hashinfo->ehash_mask; struct hlist_nulls_node *node; unsigned int slot; struct sock *sk; - for (slot = 0; slot <= hashinfo->ehash_mask; slot++) { - struct inet_ehash_bucket *head = &hashinfo->ehash[slot]; + for (slot = 0; slot <= ehash_mask; slot++, head++) { + + if (hlist_nulls_empty(&head->chain)) + continue; + restart_rcu: cond_resched(); rcu_read_lock(); -- cgit v1.2.3 From 1fbfdfaa590248c1d86407f578e40e5c65136330 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:11 -0700 Subject: af_unix: Allocate struct unix_vertex for each inflight AF_UNIX fd. We will replace the garbage collection algorithm for AF_UNIX, where we will consider each inflight AF_UNIX socket as a vertex and its file descriptor as an edge in a directed graph. This patch introduces a new struct unix_vertex representing a vertex in the graph and adds its pointer to struct unix_sock. When we send a fd using the SCM_RIGHTS message, we allocate struct scm_fp_list to struct scm_cookie in scm_fp_copy(). Then, we bump each refcount of the inflight fds' struct file and save them in scm_fp_list.fp. After that, unix_attach_fds() inexplicably clones scm_fp_list of scm_cookie and sets it to skb. (We will remove this part after replacing GC.) Here, we add a new function call in unix_attach_fds() to preallocate struct unix_vertex per inflight AF_UNIX fd and link each vertex to skb's scm_fp_list.vertices. When sendmsg() succeeds later, if the socket of the inflight fd is still not inflight yet, we will set the preallocated vertex to struct unix_sock.vertex and link it to a global list unix_unvisited_vertices under spin_lock(&unix_gc_lock). If the socket is already inflight, we free the preallocated vertex. This is to avoid taking the lock unnecessarily when sendmsg() could fail later. In the following patch, we will similarly allocate another struct per edge, which will finally be linked to the inflight socket's unix_vertex.edges. And then, we will count the number of edges as unix_vertex.out_degree. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-2-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 9 +++++++++ include/net/scm.h | 3 +++ net/core/scm.c | 7 +++++++ net/unix/af_unix.c | 6 ++++++ net/unix/garbage.c | 38 ++++++++++++++++++++++++++++++++++++++ 5 files changed, 63 insertions(+) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 627ea8e2d915..c270877a5256 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -22,9 +22,17 @@ extern unsigned int unix_tot_inflight; void unix_inflight(struct user_struct *user, struct file *fp); void unix_notinflight(struct user_struct *user, struct file *fp); +int unix_prepare_fpl(struct scm_fp_list *fpl); +void unix_destroy_fpl(struct scm_fp_list *fpl); void unix_gc(void); void wait_for_unix_gc(struct scm_fp_list *fpl); +struct unix_vertex { + struct list_head edges; + struct list_head entry; + unsigned long out_degree; +}; + struct sock *unix_peer_get(struct sock *sk); #define UNIX_HASH_MOD (256 - 1) @@ -62,6 +70,7 @@ struct unix_sock { struct path path; struct mutex iolock, bindlock; struct sock *peer; + struct unix_vertex *vertex; struct list_head link; unsigned long inflight; spinlock_t lock; diff --git a/include/net/scm.h b/include/net/scm.h index 92276a2c5543..e34321b6e204 100644 --- a/include/net/scm.h +++ b/include/net/scm.h @@ -27,6 +27,9 @@ struct scm_fp_list { short count; short count_unix; short max; +#ifdef CONFIG_UNIX + struct list_head vertices; +#endif struct user_struct *user; struct file *fp[SCM_MAX_FD]; }; diff --git a/net/core/scm.c b/net/core/scm.c index 9cd4b0a01cd6..87dfec1c3378 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -89,6 +89,9 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp) fpl->count_unix = 0; fpl->max = SCM_MAX_FD; fpl->user = NULL; +#if IS_ENABLED(CONFIG_UNIX) + INIT_LIST_HEAD(&fpl->vertices); +#endif } fpp = &fpl->fp[fpl->count]; @@ -376,8 +379,12 @@ struct scm_fp_list *scm_fp_dup(struct scm_fp_list *fpl) if (new_fpl) { for (i = 0; i < fpl->count; i++) get_file(fpl->fp[i]); + new_fpl->max = new_fpl->count; new_fpl->user = get_uid(fpl->user); +#if IS_ENABLED(CONFIG_UNIX) + INIT_LIST_HEAD(&new_fpl->vertices); +#endif } return new_fpl; } diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 5b41e2321209..a3b25d311560 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -980,6 +980,7 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern, sk->sk_destruct = unix_sock_destructor; u = unix_sk(sk); u->inflight = 0; + u->vertex = NULL; u->path.dentry = NULL; u->path.mnt = NULL; spin_lock_init(&u->lock); @@ -1805,6 +1806,9 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) for (i = scm->fp->count - 1; i >= 0; i--) unix_inflight(scm->fp->user, scm->fp->fp[i]); + if (unix_prepare_fpl(UNIXCB(skb).fp)) + return -ENOMEM; + return 0; } @@ -1815,6 +1819,8 @@ static void unix_detach_fds(struct scm_cookie *scm, struct sk_buff *skb) scm->fp = UNIXCB(skb).fp; UNIXCB(skb).fp = NULL; + unix_destroy_fpl(scm->fp); + for (i = scm->fp->count - 1; i >= 0; i--) unix_notinflight(scm->fp->user, scm->fp->fp[i]); } diff --git a/net/unix/garbage.c b/net/unix/garbage.c index fa39b6265238..75bdf66b81df 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -101,6 +101,44 @@ struct unix_sock *unix_get_socket(struct file *filp) return NULL; } +static void unix_free_vertices(struct scm_fp_list *fpl) +{ + struct unix_vertex *vertex, *next_vertex; + + list_for_each_entry_safe(vertex, next_vertex, &fpl->vertices, entry) { + list_del(&vertex->entry); + kfree(vertex); + } +} + +int unix_prepare_fpl(struct scm_fp_list *fpl) +{ + struct unix_vertex *vertex; + int i; + + if (!fpl->count_unix) + return 0; + + for (i = 0; i < fpl->count_unix; i++) { + vertex = kmalloc(sizeof(*vertex), GFP_KERNEL); + if (!vertex) + goto err; + + list_add(&vertex->entry, &fpl->vertices); + } + + return 0; + +err: + unix_free_vertices(fpl); + return -ENOMEM; +} + +void unix_destroy_fpl(struct scm_fp_list *fpl) +{ + unix_free_vertices(fpl); +} + DEFINE_SPINLOCK(unix_gc_lock); unsigned int unix_tot_inflight; static LIST_HEAD(gc_candidates); -- cgit v1.2.3 From 29b64e354029cfcf1eea4d91b146c7b769305930 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:12 -0700 Subject: af_unix: Allocate struct unix_edge for each inflight AF_UNIX fd. As with the previous patch, we preallocate to skb's scm_fp_list an array of struct unix_edge in the number of inflight AF_UNIX fds. There we just preallocate memory and do not use immediately because sendmsg() could fail after this point. The actual use will be in the next patch. When we queue skb with inflight edges, we will set the inflight socket's unix_sock as unix_edge->predecessor and the receiver's unix_sock as successor, and then we will link the edge to the inflight socket's unix_vertex.edges. Note that we set NULL to cloned scm_fp_list.edges in scm_fp_dup() so that MSG_PEEK does not change the shape of the directed graph. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-3-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 6 ++++++ include/net/scm.h | 5 +++++ net/core/scm.c | 2 ++ net/unix/garbage.c | 6 ++++++ 4 files changed, 19 insertions(+) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index c270877a5256..55c4abc26a71 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -33,6 +33,12 @@ struct unix_vertex { unsigned long out_degree; }; +struct unix_edge { + struct unix_sock *predecessor; + struct unix_sock *successor; + struct list_head vertex_entry; +}; + struct sock *unix_peer_get(struct sock *sk); #define UNIX_HASH_MOD (256 - 1) diff --git a/include/net/scm.h b/include/net/scm.h index e34321b6e204..5f5154e5096d 100644 --- a/include/net/scm.h +++ b/include/net/scm.h @@ -23,12 +23,17 @@ struct scm_creds { kgid_t gid; }; +#ifdef CONFIG_UNIX +struct unix_edge; +#endif + struct scm_fp_list { short count; short count_unix; short max; #ifdef CONFIG_UNIX struct list_head vertices; + struct unix_edge *edges; #endif struct user_struct *user; struct file *fp[SCM_MAX_FD]; diff --git a/net/core/scm.c b/net/core/scm.c index 87dfec1c3378..1bcc8a2d65e3 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -90,6 +90,7 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp) fpl->max = SCM_MAX_FD; fpl->user = NULL; #if IS_ENABLED(CONFIG_UNIX) + fpl->edges = NULL; INIT_LIST_HEAD(&fpl->vertices); #endif } @@ -383,6 +384,7 @@ struct scm_fp_list *scm_fp_dup(struct scm_fp_list *fpl) new_fpl->max = new_fpl->count; new_fpl->user = get_uid(fpl->user); #if IS_ENABLED(CONFIG_UNIX) + new_fpl->edges = NULL; INIT_LIST_HEAD(&new_fpl->vertices); #endif } diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 75bdf66b81df..f31917683288 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -127,6 +127,11 @@ int unix_prepare_fpl(struct scm_fp_list *fpl) list_add(&vertex->entry, &fpl->vertices); } + fpl->edges = kvmalloc_array(fpl->count_unix, sizeof(*fpl->edges), + GFP_KERNEL_ACCOUNT); + if (!fpl->edges) + goto err; + return 0; err: @@ -136,6 +141,7 @@ err: void unix_destroy_fpl(struct scm_fp_list *fpl) { + kvfree(fpl->edges); unix_free_vertices(fpl); } -- cgit v1.2.3 From 42f298c06b30bfe0a8cbee5d38644e618699e26e Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:13 -0700 Subject: af_unix: Link struct unix_edge when queuing skb. Just before queuing skb with inflight fds, we call scm_stat_add(), which is a good place to set up the preallocated struct unix_vertex and struct unix_edge in UNIXCB(skb).fp. Then, we call unix_add_edges() and construct the directed graph as follows: 1. Set the inflight socket's unix_sock to unix_edge.predecessor. 2. Set the receiver's unix_sock to unix_edge.successor. 3. Set the preallocated vertex to inflight socket's unix_sock.vertex. 4. Link inflight socket's unix_vertex.entry to unix_unvisited_vertices. 5. Link unix_edge.vertex_entry to the inflight socket's unix_vertex.edges. Let's say we pass the fd of AF_UNIX socket A to B and the fd of B to C. The graph looks like this: +-------------------------+ | unix_unvisited_vertices | <-------------------------. +-------------------------+ | + | | +--------------+ +--------------+ | +--------------+ | | unix_sock A | <---. .---> | unix_sock B | <-|-. .---> | unix_sock C | | +--------------+ | | +--------------+ | | | +--------------+ | .-+ | vertex | | | .-+ | vertex | | | | | vertex | | | +--------------+ | | | +--------------+ | | | +--------------+ | | | | | | | | | | +--------------+ | | | +--------------+ | | | | '-> | unix_vertex | | | '-> | unix_vertex | | | | | +--------------+ | | +--------------+ | | | `---> | entry | +---------> | entry | +-' | | |--------------| | | |--------------| | | | edges | <-. | | | edges | <-. | | +--------------+ | | | +--------------+ | | | | | | | | | .----------------------' | | .----------------------' | | | | | | | | | +--------------+ | | | +--------------+ | | | | unix_edge | | | | | unix_edge | | | | +--------------+ | | | +--------------+ | | `-> | vertex_entry | | | `-> | vertex_entry | | | |--------------| | | |--------------| | | | predecessor | +---' | | predecessor | +---' | |--------------| | |--------------| | | successor | +-----' | successor | +-----' +--------------+ +--------------+ Henceforth, we denote such a graph as A -> B (-> C). Now, we can express all inflight fd graphs that do not contain embryo sockets. We will support the particular case later. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-4-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 2 ++ include/net/scm.h | 1 + net/core/scm.c | 2 ++ net/unix/af_unix.c | 8 +++-- net/unix/garbage.c | 90 ++++++++++++++++++++++++++++++++++++++++++++++++++- 5 files changed, 100 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 55c4abc26a71..f31ad1166346 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -22,6 +22,8 @@ extern unsigned int unix_tot_inflight; void unix_inflight(struct user_struct *user, struct file *fp); void unix_notinflight(struct user_struct *user, struct file *fp); +void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver); +void unix_del_edges(struct scm_fp_list *fpl); int unix_prepare_fpl(struct scm_fp_list *fpl); void unix_destroy_fpl(struct scm_fp_list *fpl); void unix_gc(void); diff --git a/include/net/scm.h b/include/net/scm.h index 5f5154e5096d..bbc5527809d1 100644 --- a/include/net/scm.h +++ b/include/net/scm.h @@ -32,6 +32,7 @@ struct scm_fp_list { short count_unix; short max; #ifdef CONFIG_UNIX + bool inflight; struct list_head vertices; struct unix_edge *edges; #endif diff --git a/net/core/scm.c b/net/core/scm.c index 1bcc8a2d65e3..5763f3320358 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -90,6 +90,7 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp) fpl->max = SCM_MAX_FD; fpl->user = NULL; #if IS_ENABLED(CONFIG_UNIX) + fpl->inflight = false; fpl->edges = NULL; INIT_LIST_HEAD(&fpl->vertices); #endif @@ -384,6 +385,7 @@ struct scm_fp_list *scm_fp_dup(struct scm_fp_list *fpl) new_fpl->max = new_fpl->count; new_fpl->user = get_uid(fpl->user); #if IS_ENABLED(CONFIG_UNIX) + new_fpl->inflight = false; new_fpl->edges = NULL; INIT_LIST_HEAD(&new_fpl->vertices); #endif diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index a3b25d311560..24adbc4d5188 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1943,8 +1943,10 @@ static void scm_stat_add(struct sock *sk, struct sk_buff *skb) struct scm_fp_list *fp = UNIXCB(skb).fp; struct unix_sock *u = unix_sk(sk); - if (unlikely(fp && fp->count)) + if (unlikely(fp && fp->count)) { atomic_add(fp->count, &u->scm_stat.nr_fds); + unix_add_edges(fp, u); + } } static void scm_stat_del(struct sock *sk, struct sk_buff *skb) @@ -1952,8 +1954,10 @@ static void scm_stat_del(struct sock *sk, struct sk_buff *skb) struct scm_fp_list *fp = UNIXCB(skb).fp; struct unix_sock *u = unix_sk(sk); - if (unlikely(fp && fp->count)) + if (unlikely(fp && fp->count)) { atomic_sub(fp->count, &u->scm_stat.nr_fds); + unix_del_edges(fp); + } } /* diff --git a/net/unix/garbage.c b/net/unix/garbage.c index f31917683288..36d665936096 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -101,6 +101,38 @@ struct unix_sock *unix_get_socket(struct file *filp) return NULL; } +static LIST_HEAD(unix_unvisited_vertices); + +static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) +{ + struct unix_vertex *vertex = edge->predecessor->vertex; + + if (!vertex) { + vertex = list_first_entry(&fpl->vertices, typeof(*vertex), entry); + vertex->out_degree = 0; + INIT_LIST_HEAD(&vertex->edges); + + list_move_tail(&vertex->entry, &unix_unvisited_vertices); + edge->predecessor->vertex = vertex; + } + + vertex->out_degree++; + list_add_tail(&edge->vertex_entry, &vertex->edges); +} + +static void unix_del_edge(struct scm_fp_list *fpl, struct unix_edge *edge) +{ + struct unix_vertex *vertex = edge->predecessor->vertex; + + list_del(&edge->vertex_entry); + vertex->out_degree--; + + if (!vertex->out_degree) { + edge->predecessor->vertex = NULL; + list_move_tail(&vertex->entry, &fpl->vertices); + } +} + static void unix_free_vertices(struct scm_fp_list *fpl) { struct unix_vertex *vertex, *next_vertex; @@ -111,6 +143,60 @@ static void unix_free_vertices(struct scm_fp_list *fpl) } } +DEFINE_SPINLOCK(unix_gc_lock); + +void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) +{ + int i = 0, j = 0; + + spin_lock(&unix_gc_lock); + + if (!fpl->count_unix) + goto out; + + do { + struct unix_sock *inflight = unix_get_socket(fpl->fp[j++]); + struct unix_edge *edge; + + if (!inflight) + continue; + + edge = fpl->edges + i++; + edge->predecessor = inflight; + edge->successor = receiver; + + unix_add_edge(fpl, edge); + } while (i < fpl->count_unix); + +out: + spin_unlock(&unix_gc_lock); + + fpl->inflight = true; + + unix_free_vertices(fpl); +} + +void unix_del_edges(struct scm_fp_list *fpl) +{ + int i = 0; + + spin_lock(&unix_gc_lock); + + if (!fpl->count_unix) + goto out; + + do { + struct unix_edge *edge = fpl->edges + i++; + + unix_del_edge(fpl, edge); + } while (i < fpl->count_unix); + +out: + spin_unlock(&unix_gc_lock); + + fpl->inflight = false; +} + int unix_prepare_fpl(struct scm_fp_list *fpl) { struct unix_vertex *vertex; @@ -141,11 +227,13 @@ err: void unix_destroy_fpl(struct scm_fp_list *fpl) { + if (fpl->inflight) + unix_del_edges(fpl); + kvfree(fpl->edges); unix_free_vertices(fpl); } -DEFINE_SPINLOCK(unix_gc_lock); unsigned int unix_tot_inflight; static LIST_HEAD(gc_candidates); static LIST_HEAD(gc_inflight_list); -- cgit v1.2.3 From 22c3c0c52d32f41cc38cd936ea0c93f22ced3315 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:14 -0700 Subject: af_unix: Bulk update unix_tot_inflight/unix_inflight when queuing skb. Currently, we track the number of inflight sockets in two variables. unix_tot_inflight is the total number of inflight AF_UNIX sockets on the host, and user->unix_inflight is the number of inflight fds per user. We update them one by one in unix_inflight(), which can be done once in batch. Also, sendmsg() could fail even after unix_inflight(), then we need to acquire unix_gc_lock only to decrement the counters. Let's bulk update the counters in unix_add_edges() and unix_del_edges(), which is called only for successfully passed fds. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-5-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 36d665936096..84c8ea524a98 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -144,6 +144,7 @@ static void unix_free_vertices(struct scm_fp_list *fpl) } DEFINE_SPINLOCK(unix_gc_lock); +unsigned int unix_tot_inflight; void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) { @@ -168,7 +169,10 @@ void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) unix_add_edge(fpl, edge); } while (i < fpl->count_unix); + WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + fpl->count_unix); out: + WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight + fpl->count); + spin_unlock(&unix_gc_lock); fpl->inflight = true; @@ -191,7 +195,10 @@ void unix_del_edges(struct scm_fp_list *fpl) unix_del_edge(fpl, edge); } while (i < fpl->count_unix); + WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - fpl->count_unix); out: + WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight - fpl->count); + spin_unlock(&unix_gc_lock); fpl->inflight = false; @@ -234,7 +241,6 @@ void unix_destroy_fpl(struct scm_fp_list *fpl) unix_free_vertices(fpl); } -unsigned int unix_tot_inflight; static LIST_HEAD(gc_candidates); static LIST_HEAD(gc_inflight_list); @@ -255,13 +261,8 @@ void unix_inflight(struct user_struct *user, struct file *filp) WARN_ON_ONCE(list_empty(&u->link)); } u->inflight++; - - /* Paired with READ_ONCE() in wait_for_unix_gc() */ - WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + 1); } - WRITE_ONCE(user->unix_inflight, user->unix_inflight + 1); - spin_unlock(&unix_gc_lock); } @@ -278,13 +279,8 @@ void unix_notinflight(struct user_struct *user, struct file *filp) u->inflight--; if (!u->inflight) list_del_init(&u->link); - - /* Paired with READ_ONCE() in wait_for_unix_gc() */ - WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - 1); } - WRITE_ONCE(user->unix_inflight, user->unix_inflight - 1); - spin_unlock(&unix_gc_lock); } -- cgit v1.2.3 From 6ba76fd2848e107594ea4f03b737230f74bc23ea Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:15 -0700 Subject: af_unix: Iterate all vertices by DFS. The new GC will use a depth first search graph algorithm to find cyclic references. The algorithm visits every vertex exactly once. Here, we implement the DFS part without recursion so that no one can abuse it. unix_walk_scc() marks every vertex unvisited by initialising index as UNIX_VERTEX_INDEX_UNVISITED and iterates inflight vertices in unix_unvisited_vertices and call __unix_walk_scc() to start DFS from an arbitrary vertex. __unix_walk_scc() iterates all edges starting from the vertex and explores the neighbour vertices with DFS using edge_stack. After visiting all neighbours, __unix_walk_scc() moves the visited vertex to unix_visited_vertices so that unix_walk_scc() will not restart DFS from the visited vertex. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-6-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 2 ++ net/unix/garbage.c | 74 +++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 76 insertions(+) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index f31ad1166346..970a91da2239 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -33,12 +33,14 @@ struct unix_vertex { struct list_head edges; struct list_head entry; unsigned long out_degree; + unsigned long index; }; struct unix_edge { struct unix_sock *predecessor; struct unix_sock *successor; struct list_head vertex_entry; + struct list_head stack_entry; }; struct sock *unix_peer_get(struct sock *sk); diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 84c8ea524a98..8b16ab9e240e 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -103,6 +103,11 @@ struct unix_sock *unix_get_socket(struct file *filp) static LIST_HEAD(unix_unvisited_vertices); +enum unix_vertex_index { + UNIX_VERTEX_INDEX_UNVISITED, + UNIX_VERTEX_INDEX_START, +}; + static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) { struct unix_vertex *vertex = edge->predecessor->vertex; @@ -241,6 +246,73 @@ void unix_destroy_fpl(struct scm_fp_list *fpl) unix_free_vertices(fpl); } +static LIST_HEAD(unix_visited_vertices); + +static void __unix_walk_scc(struct unix_vertex *vertex) +{ + unsigned long index = UNIX_VERTEX_INDEX_START; + struct unix_edge *edge; + LIST_HEAD(edge_stack); + +next_vertex: + vertex->index = index; + index++; + + /* Explore neighbour vertices (receivers of the current vertex's fd). */ + list_for_each_entry(edge, &vertex->edges, vertex_entry) { + struct unix_vertex *next_vertex = edge->successor->vertex; + + if (!next_vertex) + continue; + + if (next_vertex->index == UNIX_VERTEX_INDEX_UNVISITED) { + /* Iterative deepening depth first search + * + * 1. Push a forward edge to edge_stack and set + * the successor to vertex for the next iteration. + */ + list_add(&edge->stack_entry, &edge_stack); + + vertex = next_vertex; + goto next_vertex; + + /* 2. Pop the edge directed to the current vertex + * and restore the ancestor for backtracking. + */ +prev_vertex: + edge = list_first_entry(&edge_stack, typeof(*edge), stack_entry); + list_del_init(&edge->stack_entry); + + vertex = edge->predecessor->vertex; + } + } + + /* Don't restart DFS from this vertex in unix_walk_scc(). */ + list_move_tail(&vertex->entry, &unix_visited_vertices); + + /* Need backtracking ? */ + if (!list_empty(&edge_stack)) + goto prev_vertex; +} + +static void unix_walk_scc(void) +{ + struct unix_vertex *vertex; + + list_for_each_entry(vertex, &unix_unvisited_vertices, entry) + vertex->index = UNIX_VERTEX_INDEX_UNVISITED; + + /* Visit every vertex exactly once. + * __unix_walk_scc() moves visited vertices to unix_visited_vertices. + */ + while (!list_empty(&unix_unvisited_vertices)) { + vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); + __unix_walk_scc(vertex); + } + + list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); +} + static LIST_HEAD(gc_candidates); static LIST_HEAD(gc_inflight_list); @@ -388,6 +460,8 @@ static void __unix_gc(struct work_struct *work) spin_lock(&unix_gc_lock); + unix_walk_scc(); + /* First, select candidates for garbage collection. Only * in-flight sockets are considered, and from those only ones * which don't have any external reference. -- cgit v1.2.3 From 3484f063172dd88776b062046d721d7c2ae1af7c Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:16 -0700 Subject: af_unix: Detect Strongly Connected Components. In the new GC, we use a simple graph algorithm, Tarjan's Strongly Connected Components (SCC) algorithm, to find cyclic references. The algorithm visits every vertex exactly once using depth-first search (DFS). DFS starts by pushing an input vertex to a stack and assigning it a unique number. Two fields, index and lowlink, are initialised with the number, but lowlink could be updated later during DFS. If a vertex has an edge to an unvisited inflight vertex, we visit it and do the same processing. So, we will have vertices in the stack in the order they appear and number them consecutively in the same order. If a vertex has a back-edge to a visited vertex in the stack, we update the predecessor's lowlink with the successor's index. After iterating edges from the vertex, we check if its index equals its lowlink. If the lowlink is different from the index, it shows there was a back-edge. Then, we go backtracking and propagate the lowlink to its predecessor and resume the previous edge iteration from the next edge. If the lowlink is the same as the index, we pop vertices before and including the vertex from the stack. Then, the set of vertices is SCC, possibly forming a cycle. At the same time, we move the vertices to unix_visited_vertices. When we finish the algorithm, all vertices in each SCC will be linked via unix_vertex.scc_entry. Let's take an example. We have a graph including five inflight vertices (F is not inflight): A -> B -> C -> D -> E (-> F) ^ | `---------' Suppose that we start DFS from C. We will visit C, D, and B first and initialise their index and lowlink. Then, the stack looks like this: > B = (3, 3) (index, lowlink) D = (2, 2) C = (1, 1) When checking B's edge to C, we update B's lowlink with C's index and propagate it to D. B = (3, 1) (index, lowlink) > D = (2, 1) C = (1, 1) Next, we visit E, which has no edge to an inflight vertex. > E = (4, 4) (index, lowlink) B = (3, 1) D = (2, 1) C = (1, 1) When we leave from E, its index and lowlink are the same, so we pop E from the stack as single-vertex SCC. Next, we leave from B and D but do nothing because their lowlink are different from their index. B = (3, 1) (index, lowlink) D = (2, 1) > C = (1, 1) Then, we leave from C, whose index and lowlink are the same, so we pop B, D and C as SCC. Last, we do DFS for the rest of vertices, A, which is also a single-vertex SCC. Finally, each unix_vertex.scc_entry is linked as follows: A -. B -> C -> D E -. ^ | ^ | ^ | `--' `---------' `--' We use SCC later to decide whether we can garbage-collect the sockets. Note that we still cannot detect SCC properly if an edge points to an embryo socket. The following two patches will sort it out. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-7-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 3 +++ net/unix/garbage.c | 46 ++++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 47 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 970a91da2239..67736767b616 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -32,8 +32,11 @@ void wait_for_unix_gc(struct scm_fp_list *fpl); struct unix_vertex { struct list_head edges; struct list_head entry; + struct list_head scc_entry; unsigned long out_degree; unsigned long index; + unsigned long lowlink; + bool on_stack; }; struct unix_edge { diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 8b16ab9e240e..33aadaa35346 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -251,11 +251,19 @@ static LIST_HEAD(unix_visited_vertices); static void __unix_walk_scc(struct unix_vertex *vertex) { unsigned long index = UNIX_VERTEX_INDEX_START; + LIST_HEAD(vertex_stack); struct unix_edge *edge; LIST_HEAD(edge_stack); next_vertex: + /* Push vertex to vertex_stack. + * The vertex will be popped when finalising SCC later. + */ + vertex->on_stack = true; + list_add(&vertex->scc_entry, &vertex_stack); + vertex->index = index; + vertex->lowlink = index; index++; /* Explore neighbour vertices (receivers of the current vertex's fd). */ @@ -283,12 +291,46 @@ prev_vertex: edge = list_first_entry(&edge_stack, typeof(*edge), stack_entry); list_del_init(&edge->stack_entry); + next_vertex = vertex; vertex = edge->predecessor->vertex; + + /* If the successor has a smaller lowlink, two vertices + * are in the same SCC, so propagate the smaller lowlink + * to skip SCC finalisation. + */ + vertex->lowlink = min(vertex->lowlink, next_vertex->lowlink); + } else if (next_vertex->on_stack) { + /* Loop detected by a back/cross edge. + * + * The successor is on vertex_stack, so two vertices are + * in the same SCC. If the successor has a smaller index, + * propagate it to skip SCC finalisation. + */ + vertex->lowlink = min(vertex->lowlink, next_vertex->index); + } else { + /* The successor was already grouped as another SCC */ } } - /* Don't restart DFS from this vertex in unix_walk_scc(). */ - list_move_tail(&vertex->entry, &unix_visited_vertices); + if (vertex->index == vertex->lowlink) { + struct list_head scc; + + /* SCC finalised. + * + * If the lowlink was not updated, all the vertices above on + * vertex_stack are in the same SCC. Group them using scc_entry. + */ + __list_cut_position(&scc, &vertex_stack, &vertex->scc_entry); + + list_for_each_entry_reverse(vertex, &scc, scc_entry) { + /* Don't restart DFS from this vertex in unix_walk_scc(). */ + list_move_tail(&vertex->entry, &unix_visited_vertices); + + vertex->on_stack = false; + } + + list_del(&scc); + } /* Need backtracking ? */ if (!list_empty(&edge_stack)) -- cgit v1.2.3 From aed6ecef55d70de3762ce41c561b7f547dbaf107 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:17 -0700 Subject: af_unix: Save listener for embryo socket. This is a prep patch for the following change, where we need to fetch the listening socket from the successor embryo socket during GC. We add a new field to struct unix_sock to save a pointer to a listening socket. We set it when connect() creates a new socket, and clear it when accept() is called. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-8-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 1 + net/unix/af_unix.c | 5 ++++- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 67736767b616..dc7469191195 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -83,6 +83,7 @@ struct unix_sock { struct path path; struct mutex iolock, bindlock; struct sock *peer; + struct sock *listener; struct unix_vertex *vertex; struct list_head link; unsigned long inflight; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 24adbc4d5188..af74e7ebc35a 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -979,6 +979,7 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern, sk->sk_max_ack_backlog = net->unx.sysctl_max_dgram_qlen; sk->sk_destruct = unix_sock_destructor; u = unix_sk(sk); + u->listener = NULL; u->inflight = 0; u->vertex = NULL; u->path.dentry = NULL; @@ -1598,6 +1599,7 @@ restart: newsk->sk_type = sk->sk_type; init_peercred(newsk); newu = unix_sk(newsk); + newu->listener = other; RCU_INIT_POINTER(newsk->sk_wq, &newu->peer_wq); otheru = unix_sk(other); @@ -1693,8 +1695,8 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, bool kern) { struct sock *sk = sock->sk; - struct sock *tsk; struct sk_buff *skb; + struct sock *tsk; int err; err = -EOPNOTSUPP; @@ -1719,6 +1721,7 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, } tsk = skb->sk; + unix_sk(tsk)->listener = NULL; skb_free_datagram(sk, skb); wake_up_interruptible(&unix_sk(sk)->peer_wait); -- cgit v1.2.3 From dcf70df2048d27c5d186f013f101a4aefd63aa41 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:18 -0700 Subject: af_unix: Fix up unix_edge.successor for embryo socket. To garbage collect inflight AF_UNIX sockets, we must define the cyclic reference appropriately. This is a bit tricky if the loop consists of embryo sockets. Suppose that the fd of AF_UNIX socket A is passed to D and the fd B to C and that C and D are embryo sockets of A and B, respectively. It may appear that there are two separate graphs, A (-> D) and B (-> C), but this is not correct. A --. .-- B X C <-' `-> D Now, D holds A's refcount, and C has B's refcount, so unix_release() will never be called for A and B when we close() them. However, no one can call close() for D and C to free skbs holding refcounts of A and B because C/D is in A/B's receive queue, which should have been purged by unix_release() for A and B. So, here's another type of cyclic reference. When a fd of an AF_UNIX socket is passed to an embryo socket, the reference is indirectly held by its parent listening socket. .-> A .-> B | `- sk_receive_queue | `- sk_receive_queue | `- skb | `- skb | `- sk == C | `- sk == D | `- sk_receive_queue | `- sk_receive_queue | `- skb +---------' `- skb +-. | | `---------------------------------------------------------' Technically, the graph must be denoted as A <-> B instead of A (-> D) and B (-> C) to find such a cyclic reference without touching each socket's receive queue. .-> A --. .-- B <-. | X | == A <-> B `-- C <-' `-> D --' We apply this fixup during GC by fetching the real successor by unix_edge_successor(). When we call accept(), we clear unix_sock.listener under unix_gc_lock not to confuse GC. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-9-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 1 + net/unix/af_unix.c | 2 +- net/unix/garbage.c | 20 +++++++++++++++++++- 3 files changed, 21 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index dc7469191195..414463803b7e 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -24,6 +24,7 @@ void unix_inflight(struct user_struct *user, struct file *fp); void unix_notinflight(struct user_struct *user, struct file *fp); void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver); void unix_del_edges(struct scm_fp_list *fpl); +void unix_update_edges(struct unix_sock *receiver); int unix_prepare_fpl(struct scm_fp_list *fpl); void unix_destroy_fpl(struct scm_fp_list *fpl); void unix_gc(void); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index af74e7ebc35a..ae77e2dc0dae 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1721,7 +1721,7 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, } tsk = skb->sk; - unix_sk(tsk)->listener = NULL; + unix_update_edges(unix_sk(tsk)); skb_free_datagram(sk, skb); wake_up_interruptible(&unix_sk(sk)->peer_wait); diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 33aadaa35346..8d0912c1d01a 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -101,6 +101,17 @@ struct unix_sock *unix_get_socket(struct file *filp) return NULL; } +static struct unix_vertex *unix_edge_successor(struct unix_edge *edge) +{ + /* If an embryo socket has a fd, + * the listener indirectly holds the fd's refcnt. + */ + if (edge->successor->listener) + return unix_sk(edge->successor->listener)->vertex; + + return edge->successor->vertex; +} + static LIST_HEAD(unix_unvisited_vertices); enum unix_vertex_index { @@ -209,6 +220,13 @@ out: fpl->inflight = false; } +void unix_update_edges(struct unix_sock *receiver) +{ + spin_lock(&unix_gc_lock); + receiver->listener = NULL; + spin_unlock(&unix_gc_lock); +} + int unix_prepare_fpl(struct scm_fp_list *fpl) { struct unix_vertex *vertex; @@ -268,7 +286,7 @@ next_vertex: /* Explore neighbour vertices (receivers of the current vertex's fd). */ list_for_each_entry(edge, &vertex->edges, vertex_entry) { - struct unix_vertex *next_vertex = edge->successor->vertex; + struct unix_vertex *next_vertex = unix_edge_successor(edge); if (!next_vertex) continue; -- cgit v1.2.3 From ba31b4a4e1018f5844c6eb31734976e2184f2f9a Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:19 -0700 Subject: af_unix: Save O(n) setup of Tarjan's algo. Before starting Tarjan's algorithm, we need to mark all vertices as unvisited. We can save this O(n) setup by reserving two special indices (0, 1) and using two variables. The first time we link a vertex to unix_unvisited_vertices, we set unix_vertex_unvisited_index to index. During DFS, we can see that the index of unvisited vertices is the same as unix_vertex_unvisited_index. When we finalise SCC later, we set unix_vertex_grouped_index to each vertex's index. Then, we can know (i) that the vertex is on the stack if the index of a visited vertex is >= 2 and (ii) that it is not on the stack and belongs to a different SCC if the index is unix_vertex_grouped_index. After the whole algorithm, all indices of vertices are set as unix_vertex_grouped_index. Next time we start DFS, we know that all unvisited vertices have unix_vertex_grouped_index, and we can use unix_vertex_unvisited_index as the not-on-stack marker. To use the same variable in __unix_walk_scc(), we can swap unix_vertex_(grouped|unvisited)_index at the end of Tarjan's algorithm. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-10-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 1 - net/unix/garbage.c | 26 +++++++++++++++----------- 2 files changed, 15 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 414463803b7e..ec040caaa4b5 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -37,7 +37,6 @@ struct unix_vertex { unsigned long out_degree; unsigned long index; unsigned long lowlink; - bool on_stack; }; struct unix_edge { diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 8d0912c1d01a..cec6189126ba 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -115,16 +115,20 @@ static struct unix_vertex *unix_edge_successor(struct unix_edge *edge) static LIST_HEAD(unix_unvisited_vertices); enum unix_vertex_index { - UNIX_VERTEX_INDEX_UNVISITED, + UNIX_VERTEX_INDEX_MARK1, + UNIX_VERTEX_INDEX_MARK2, UNIX_VERTEX_INDEX_START, }; +static unsigned long unix_vertex_unvisited_index = UNIX_VERTEX_INDEX_MARK1; + static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) { struct unix_vertex *vertex = edge->predecessor->vertex; if (!vertex) { vertex = list_first_entry(&fpl->vertices, typeof(*vertex), entry); + vertex->index = unix_vertex_unvisited_index; vertex->out_degree = 0; INIT_LIST_HEAD(&vertex->edges); @@ -265,6 +269,7 @@ void unix_destroy_fpl(struct scm_fp_list *fpl) } static LIST_HEAD(unix_visited_vertices); +static unsigned long unix_vertex_grouped_index = UNIX_VERTEX_INDEX_MARK2; static void __unix_walk_scc(struct unix_vertex *vertex) { @@ -274,10 +279,10 @@ static void __unix_walk_scc(struct unix_vertex *vertex) LIST_HEAD(edge_stack); next_vertex: - /* Push vertex to vertex_stack. + /* Push vertex to vertex_stack and mark it as on-stack + * (index >= UNIX_VERTEX_INDEX_START). * The vertex will be popped when finalising SCC later. */ - vertex->on_stack = true; list_add(&vertex->scc_entry, &vertex_stack); vertex->index = index; @@ -291,7 +296,7 @@ next_vertex: if (!next_vertex) continue; - if (next_vertex->index == UNIX_VERTEX_INDEX_UNVISITED) { + if (next_vertex->index == unix_vertex_unvisited_index) { /* Iterative deepening depth first search * * 1. Push a forward edge to edge_stack and set @@ -317,7 +322,7 @@ prev_vertex: * to skip SCC finalisation. */ vertex->lowlink = min(vertex->lowlink, next_vertex->lowlink); - } else if (next_vertex->on_stack) { + } else if (next_vertex->index != unix_vertex_grouped_index) { /* Loop detected by a back/cross edge. * * The successor is on vertex_stack, so two vertices are @@ -344,7 +349,8 @@ prev_vertex: /* Don't restart DFS from this vertex in unix_walk_scc(). */ list_move_tail(&vertex->entry, &unix_visited_vertices); - vertex->on_stack = false; + /* Mark vertex as off-stack. */ + vertex->index = unix_vertex_grouped_index; } list_del(&scc); @@ -357,20 +363,18 @@ prev_vertex: static void unix_walk_scc(void) { - struct unix_vertex *vertex; - - list_for_each_entry(vertex, &unix_unvisited_vertices, entry) - vertex->index = UNIX_VERTEX_INDEX_UNVISITED; - /* Visit every vertex exactly once. * __unix_walk_scc() moves visited vertices to unix_visited_vertices. */ while (!list_empty(&unix_unvisited_vertices)) { + struct unix_vertex *vertex; + vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); __unix_walk_scc(vertex); } list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); + swap(unix_vertex_unvisited_index, unix_vertex_grouped_index); } static LIST_HEAD(gc_candidates); -- cgit v1.2.3 From 77e5593aebba823bcbcf2c4b58b07efcd63933b8 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:20 -0700 Subject: af_unix: Skip GC if no cycle exists. We do not need to run GC if there is no possible cyclic reference. We use unix_graph_maybe_cyclic to decide if we should run GC. If a fd of an AF_UNIX socket is passed to an already inflight AF_UNIX socket, they could form a cyclic reference. Then, we set true to unix_graph_maybe_cyclic and later run Tarjan's algorithm to group them into SCC. Once we run Tarjan's algorithm, we are 100% sure whether cyclic references exist or not. If there is no cycle, we set false to unix_graph_maybe_cyclic and can skip the entire garbage collection next time. When finalising SCC, we set true to unix_graph_maybe_cyclic if SCC consists of multiple vertices. Even if SCC is a single vertex, a cycle might exist as self-fd passing. Given the corner case is rare, we detect it by checking all edges of the vertex and set true to unix_graph_maybe_cyclic. With this change, __unix_gc() is just a spin_lock() dance in the normal usage. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-11-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 48 +++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 47 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index cec6189126ba..5a1fae78d6dc 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -112,6 +112,19 @@ static struct unix_vertex *unix_edge_successor(struct unix_edge *edge) return edge->successor->vertex; } +static bool unix_graph_maybe_cyclic; + +static void unix_update_graph(struct unix_vertex *vertex) +{ + /* If the receiver socket is not inflight, no cyclic + * reference could be formed. + */ + if (!vertex) + return; + + unix_graph_maybe_cyclic = true; +} + static LIST_HEAD(unix_unvisited_vertices); enum unix_vertex_index { @@ -138,12 +151,16 @@ static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) vertex->out_degree++; list_add_tail(&edge->vertex_entry, &vertex->edges); + + unix_update_graph(unix_edge_successor(edge)); } static void unix_del_edge(struct scm_fp_list *fpl, struct unix_edge *edge) { struct unix_vertex *vertex = edge->predecessor->vertex; + unix_update_graph(unix_edge_successor(edge)); + list_del(&edge->vertex_entry); vertex->out_degree--; @@ -227,6 +244,7 @@ out: void unix_update_edges(struct unix_sock *receiver) { spin_lock(&unix_gc_lock); + unix_update_graph(unix_sk(receiver->listener)->vertex); receiver->listener = NULL; spin_unlock(&unix_gc_lock); } @@ -268,6 +286,26 @@ void unix_destroy_fpl(struct scm_fp_list *fpl) unix_free_vertices(fpl); } +static bool unix_scc_cyclic(struct list_head *scc) +{ + struct unix_vertex *vertex; + struct unix_edge *edge; + + /* SCC containing multiple vertices ? */ + if (!list_is_singular(scc)) + return true; + + vertex = list_first_entry(scc, typeof(*vertex), scc_entry); + + /* Self-reference or a embryo-listener circle ? */ + list_for_each_entry(edge, &vertex->edges, vertex_entry) { + if (unix_edge_successor(edge) == vertex) + return true; + } + + return false; +} + static LIST_HEAD(unix_visited_vertices); static unsigned long unix_vertex_grouped_index = UNIX_VERTEX_INDEX_MARK2; @@ -353,6 +391,9 @@ prev_vertex: vertex->index = unix_vertex_grouped_index; } + if (!unix_graph_maybe_cyclic) + unix_graph_maybe_cyclic = unix_scc_cyclic(&scc); + list_del(&scc); } @@ -363,6 +404,8 @@ prev_vertex: static void unix_walk_scc(void) { + unix_graph_maybe_cyclic = false; + /* Visit every vertex exactly once. * __unix_walk_scc() moves visited vertices to unix_visited_vertices. */ @@ -524,6 +567,9 @@ static void __unix_gc(struct work_struct *work) spin_lock(&unix_gc_lock); + if (!unix_graph_maybe_cyclic) + goto skip_gc; + unix_walk_scc(); /* First, select candidates for garbage collection. Only @@ -617,7 +663,7 @@ static void __unix_gc(struct work_struct *work) /* All candidates should have been detached by now. */ WARN_ON_ONCE(!list_empty(&gc_candidates)); - +skip_gc: /* Paired with READ_ONCE() in wait_for_unix_gc(). */ WRITE_ONCE(gc_in_progress, false); -- cgit v1.2.3 From ad081928a8b0f57f269df999a28087fce6f2b6ce Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:21 -0700 Subject: af_unix: Avoid Tarjan's algorithm if unnecessary. Once a cyclic reference is formed, we need to run GC to check if there is dead SCC. However, we do not need to run Tarjan's algorithm if we know that the shape of the inflight graph has not been changed. If an edge is added/updated/deleted and the edge's successor is inflight, we set false to unix_graph_grouped, which means we need to re-classify SCC. Once we finalise SCC, we set true to unix_graph_grouped. While unix_graph_grouped is true, we can iterate the grouped SCC using vertex->scc_entry in unix_walk_scc_fast(). list_add() and list_for_each_entry_reverse() uses seem weird, but they are to keep the vertex order consistent and make writing test easier. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-12-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 5a1fae78d6dc..654aa8e30a8b 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -113,6 +113,7 @@ static struct unix_vertex *unix_edge_successor(struct unix_edge *edge) } static bool unix_graph_maybe_cyclic; +static bool unix_graph_grouped; static void unix_update_graph(struct unix_vertex *vertex) { @@ -123,6 +124,7 @@ static void unix_update_graph(struct unix_vertex *vertex) return; unix_graph_maybe_cyclic = true; + unix_graph_grouped = false; } static LIST_HEAD(unix_unvisited_vertices); @@ -144,6 +146,7 @@ static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) vertex->index = unix_vertex_unvisited_index; vertex->out_degree = 0; INIT_LIST_HEAD(&vertex->edges); + INIT_LIST_HEAD(&vertex->scc_entry); list_move_tail(&vertex->entry, &unix_unvisited_vertices); edge->predecessor->vertex = vertex; @@ -418,6 +421,26 @@ static void unix_walk_scc(void) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); swap(unix_vertex_unvisited_index, unix_vertex_grouped_index); + + unix_graph_grouped = true; +} + +static void unix_walk_scc_fast(void) +{ + while (!list_empty(&unix_unvisited_vertices)) { + struct unix_vertex *vertex; + struct list_head scc; + + vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); + list_add(&scc, &vertex->scc_entry); + + list_for_each_entry_reverse(vertex, &scc, scc_entry) + list_move_tail(&vertex->entry, &unix_visited_vertices); + + list_del(&scc); + } + + list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); } static LIST_HEAD(gc_candidates); @@ -570,7 +593,10 @@ static void __unix_gc(struct work_struct *work) if (!unix_graph_maybe_cyclic) goto skip_gc; - unix_walk_scc(); + if (unix_graph_grouped) + unix_walk_scc_fast(); + else + unix_walk_scc(); /* First, select candidates for garbage collection. Only * in-flight sockets are considered, and from those only ones -- cgit v1.2.3 From bfdb01283ee8f2f3089656c3ff8f62bb072dabb2 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:22 -0700 Subject: af_unix: Assign a unique index to SCC. The definition of the lowlink in Tarjan's algorithm is the smallest index of a vertex that is reachable with at most one back-edge in SCC. This is not useful for a cross-edge. If we start traversing from A in the following graph, the final lowlink of D is 3. The cross-edge here is one between D and C. A -> B -> D D = (4, 3) (index, lowlink) ^ | | C = (3, 1) | V | B = (2, 1) `--- C <--' A = (1, 1) This is because the lowlink of D is updated with the index of C. In the following patch, we detect a dead SCC by checking two conditions for each vertex. 1) vertex has no edge directed to another SCC (no bridge) 2) vertex's out_degree is the same as the refcount of its file If 1) is false, there is a receiver of all fds of the SCC and its ancestor SCC. To evaluate 1), we need to assign a unique index to each SCC and assign it to all vertices in the SCC. This patch changes the lowlink update logic for cross-edge so that in the example above, the lowlink of D is updated with the lowlink of C. A -> B -> D D = (4, 1) (index, lowlink) ^ | | C = (3, 1) | V | B = (2, 1) `--- C <--' A = (1, 1) Then, all vertices in the same SCC have the same lowlink, and we can quickly find the bridge connecting to different SCC if exists. However, it is no longer called lowlink, so we rename it to scc_index. (It's sometimes called lowpoint.) Also, we add a global variable to hold the last index used in DFS so that we do not reset the initial index in each DFS. This patch can be squashed to the SCC detection patch but is split deliberately for anyone wondering why lowlink is not used as used in the original Tarjan's algorithm and many reference implementations. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-13-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 2 +- net/unix/garbage.c | 29 +++++++++++++++-------------- 2 files changed, 16 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index ec040caaa4b5..696d997a5ac9 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -36,7 +36,7 @@ struct unix_vertex { struct list_head scc_entry; unsigned long out_degree; unsigned long index; - unsigned long lowlink; + unsigned long scc_index; }; struct unix_edge { diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 654aa8e30a8b..3f59cee3ccbc 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -312,9 +312,8 @@ static bool unix_scc_cyclic(struct list_head *scc) static LIST_HEAD(unix_visited_vertices); static unsigned long unix_vertex_grouped_index = UNIX_VERTEX_INDEX_MARK2; -static void __unix_walk_scc(struct unix_vertex *vertex) +static void __unix_walk_scc(struct unix_vertex *vertex, unsigned long *last_index) { - unsigned long index = UNIX_VERTEX_INDEX_START; LIST_HEAD(vertex_stack); struct unix_edge *edge; LIST_HEAD(edge_stack); @@ -326,9 +325,9 @@ next_vertex: */ list_add(&vertex->scc_entry, &vertex_stack); - vertex->index = index; - vertex->lowlink = index; - index++; + vertex->index = *last_index; + vertex->scc_index = *last_index; + (*last_index)++; /* Explore neighbour vertices (receivers of the current vertex's fd). */ list_for_each_entry(edge, &vertex->edges, vertex_entry) { @@ -358,30 +357,30 @@ prev_vertex: next_vertex = vertex; vertex = edge->predecessor->vertex; - /* If the successor has a smaller lowlink, two vertices - * are in the same SCC, so propagate the smaller lowlink + /* If the successor has a smaller scc_index, two vertices + * are in the same SCC, so propagate the smaller scc_index * to skip SCC finalisation. */ - vertex->lowlink = min(vertex->lowlink, next_vertex->lowlink); + vertex->scc_index = min(vertex->scc_index, next_vertex->scc_index); } else if (next_vertex->index != unix_vertex_grouped_index) { /* Loop detected by a back/cross edge. * - * The successor is on vertex_stack, so two vertices are - * in the same SCC. If the successor has a smaller index, + * The successor is on vertex_stack, so two vertices are in + * the same SCC. If the successor has a smaller *scc_index*, * propagate it to skip SCC finalisation. */ - vertex->lowlink = min(vertex->lowlink, next_vertex->index); + vertex->scc_index = min(vertex->scc_index, next_vertex->scc_index); } else { /* The successor was already grouped as another SCC */ } } - if (vertex->index == vertex->lowlink) { + if (vertex->index == vertex->scc_index) { struct list_head scc; /* SCC finalised. * - * If the lowlink was not updated, all the vertices above on + * If the scc_index was not updated, all the vertices above on * vertex_stack are in the same SCC. Group them using scc_entry. */ __list_cut_position(&scc, &vertex_stack, &vertex->scc_entry); @@ -407,6 +406,8 @@ prev_vertex: static void unix_walk_scc(void) { + unsigned long last_index = UNIX_VERTEX_INDEX_START; + unix_graph_maybe_cyclic = false; /* Visit every vertex exactly once. @@ -416,7 +417,7 @@ static void unix_walk_scc(void) struct unix_vertex *vertex; vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); - __unix_walk_scc(vertex); + __unix_walk_scc(vertex, &last_index); } list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); -- cgit v1.2.3 From a15702d8b3aad8ce5268c565bd29f0e02fd2db83 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:23 -0700 Subject: af_unix: Detect dead SCC. When iterating SCC, we call unix_vertex_dead() for each vertex to check if the vertex is close()d and has no bridge to another SCC. If both conditions are true for every vertex in SCC, we can execute garbage collection for all skb in the SCC. The actual garbage collection is done in the following patch, replacing the old implementation. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-14-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/unix/garbage.c | 44 +++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 43 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 3f59cee3ccbc..0a6b38da578c 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -289,6 +289,39 @@ void unix_destroy_fpl(struct scm_fp_list *fpl) unix_free_vertices(fpl); } +static bool unix_vertex_dead(struct unix_vertex *vertex) +{ + struct unix_edge *edge; + struct unix_sock *u; + long total_ref; + + list_for_each_entry(edge, &vertex->edges, vertex_entry) { + struct unix_vertex *next_vertex = unix_edge_successor(edge); + + /* The vertex's fd can be received by a non-inflight socket. */ + if (!next_vertex) + return false; + + /* The vertex's fd can be received by an inflight socket in + * another SCC. + */ + if (next_vertex->scc_index != vertex->scc_index) + return false; + } + + /* No receiver exists out of the same SCC. */ + + edge = list_first_entry(&vertex->edges, typeof(*edge), vertex_entry); + u = edge->predecessor; + total_ref = file_count(u->sk.sk_socket->file); + + /* If not close()d, total_ref > out_degree. */ + if (total_ref != vertex->out_degree) + return false; + + return true; +} + static bool unix_scc_cyclic(struct list_head *scc) { struct unix_vertex *vertex; @@ -377,6 +410,7 @@ prev_vertex: if (vertex->index == vertex->scc_index) { struct list_head scc; + bool scc_dead = true; /* SCC finalised. * @@ -391,6 +425,9 @@ prev_vertex: /* Mark vertex as off-stack. */ vertex->index = unix_vertex_grouped_index; + + if (scc_dead) + scc_dead = unix_vertex_dead(vertex); } if (!unix_graph_maybe_cyclic) @@ -431,13 +468,18 @@ static void unix_walk_scc_fast(void) while (!list_empty(&unix_unvisited_vertices)) { struct unix_vertex *vertex; struct list_head scc; + bool scc_dead = true; vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); list_add(&scc, &vertex->scc_entry); - list_for_each_entry_reverse(vertex, &scc, scc_entry) + list_for_each_entry_reverse(vertex, &scc, scc_entry) { list_move_tail(&vertex->entry, &unix_visited_vertices); + if (scc_dead) + scc_dead = unix_vertex_dead(vertex); + } + list_del(&scc); } -- cgit v1.2.3 From 4090fa373f0e763c43610853d2774b5979915959 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 25 Mar 2024 13:24:24 -0700 Subject: af_unix: Replace garbage collection algorithm. If we find a dead SCC during iteration, we call unix_collect_skb() to splice all skb in the SCC to the global sk_buff_head, hitlist. After iterating all SCC, we unlock unix_gc_lock and purge the queue. Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240325202425.60930-15-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 8 -- net/unix/af_unix.c | 12 -- net/unix/garbage.c | 302 +++++++++++--------------------------------------- 3 files changed, 64 insertions(+), 258 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 696d997a5ac9..226a8da2cbe3 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -19,9 +19,6 @@ static inline struct unix_sock *unix_get_socket(struct file *filp) extern spinlock_t unix_gc_lock; extern unsigned int unix_tot_inflight; - -void unix_inflight(struct user_struct *user, struct file *fp); -void unix_notinflight(struct user_struct *user, struct file *fp); void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver); void unix_del_edges(struct scm_fp_list *fpl); void unix_update_edges(struct unix_sock *receiver); @@ -85,12 +82,7 @@ struct unix_sock { struct sock *peer; struct sock *listener; struct unix_vertex *vertex; - struct list_head link; - unsigned long inflight; spinlock_t lock; - unsigned long gc_flags; -#define UNIX_GC_CANDIDATE 0 -#define UNIX_GC_MAYBE_CYCLE 1 struct socket_wq peer_wq; wait_queue_entry_t peer_wake; struct scm_stat scm_stat; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index ae77e2dc0dae..27ca50ab1cd1 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -980,12 +980,10 @@ static struct sock *unix_create1(struct net *net, struct socket *sock, int kern, sk->sk_destruct = unix_sock_destructor; u = unix_sk(sk); u->listener = NULL; - u->inflight = 0; u->vertex = NULL; u->path.dentry = NULL; u->path.mnt = NULL; spin_lock_init(&u->lock); - INIT_LIST_HEAD(&u->link); mutex_init(&u->iolock); /* single task reading lock */ mutex_init(&u->bindlock); /* single task binding lock */ init_waitqueue_head(&u->peer_wait); @@ -1793,8 +1791,6 @@ static inline bool too_many_unix_fds(struct task_struct *p) static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) { - int i; - if (too_many_unix_fds(current)) return -ETOOMANYREFS; @@ -1806,9 +1802,6 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) if (!UNIXCB(skb).fp) return -ENOMEM; - for (i = scm->fp->count - 1; i >= 0; i--) - unix_inflight(scm->fp->user, scm->fp->fp[i]); - if (unix_prepare_fpl(UNIXCB(skb).fp)) return -ENOMEM; @@ -1817,15 +1810,10 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) static void unix_detach_fds(struct scm_cookie *scm, struct sk_buff *skb) { - int i; - scm->fp = UNIXCB(skb).fp; UNIXCB(skb).fp = NULL; unix_destroy_fpl(scm->fp); - - for (i = scm->fp->count - 1; i >= 0; i--) - unix_notinflight(scm->fp->user, scm->fp->fp[i]); } static void unix_peek_fds(struct scm_cookie *scm, struct sk_buff *skb) diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 0a6b38da578c..89ea71d9297b 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -322,6 +322,52 @@ static bool unix_vertex_dead(struct unix_vertex *vertex) return true; } +enum unix_recv_queue_lock_class { + U_RECVQ_LOCK_NORMAL, + U_RECVQ_LOCK_EMBRYO, +}; + +static void unix_collect_skb(struct list_head *scc, struct sk_buff_head *hitlist) +{ + struct unix_vertex *vertex; + + list_for_each_entry_reverse(vertex, scc, scc_entry) { + struct sk_buff_head *queue; + struct unix_edge *edge; + struct unix_sock *u; + + edge = list_first_entry(&vertex->edges, typeof(*edge), vertex_entry); + u = edge->predecessor; + queue = &u->sk.sk_receive_queue; + + spin_lock(&queue->lock); + + if (u->sk.sk_state == TCP_LISTEN) { + struct sk_buff *skb; + + skb_queue_walk(queue, skb) { + struct sk_buff_head *embryo_queue = &skb->sk->sk_receive_queue; + + /* listener -> embryo order, the inversion never happens. */ + spin_lock_nested(&embryo_queue->lock, U_RECVQ_LOCK_EMBRYO); + skb_queue_splice_init(embryo_queue, hitlist); + spin_unlock(&embryo_queue->lock); + } + } else { + skb_queue_splice_init(queue, hitlist); + +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (u->oob_skb) { + kfree_skb(u->oob_skb); + u->oob_skb = NULL; + } +#endif + } + + spin_unlock(&queue->lock); + } +} + static bool unix_scc_cyclic(struct list_head *scc) { struct unix_vertex *vertex; @@ -345,7 +391,8 @@ static bool unix_scc_cyclic(struct list_head *scc) static LIST_HEAD(unix_visited_vertices); static unsigned long unix_vertex_grouped_index = UNIX_VERTEX_INDEX_MARK2; -static void __unix_walk_scc(struct unix_vertex *vertex, unsigned long *last_index) +static void __unix_walk_scc(struct unix_vertex *vertex, unsigned long *last_index, + struct sk_buff_head *hitlist) { LIST_HEAD(vertex_stack); struct unix_edge *edge; @@ -430,7 +477,9 @@ prev_vertex: scc_dead = unix_vertex_dead(vertex); } - if (!unix_graph_maybe_cyclic) + if (scc_dead) + unix_collect_skb(&scc, hitlist); + else if (!unix_graph_maybe_cyclic) unix_graph_maybe_cyclic = unix_scc_cyclic(&scc); list_del(&scc); @@ -441,7 +490,7 @@ prev_vertex: goto prev_vertex; } -static void unix_walk_scc(void) +static void unix_walk_scc(struct sk_buff_head *hitlist) { unsigned long last_index = UNIX_VERTEX_INDEX_START; @@ -454,7 +503,7 @@ static void unix_walk_scc(void) struct unix_vertex *vertex; vertex = list_first_entry(&unix_unvisited_vertices, typeof(*vertex), entry); - __unix_walk_scc(vertex, &last_index); + __unix_walk_scc(vertex, &last_index, hitlist); } list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); @@ -463,7 +512,7 @@ static void unix_walk_scc(void) unix_graph_grouped = true; } -static void unix_walk_scc_fast(void) +static void unix_walk_scc_fast(struct sk_buff_head *hitlist) { while (!list_empty(&unix_unvisited_vertices)) { struct unix_vertex *vertex; @@ -480,263 +529,40 @@ static void unix_walk_scc_fast(void) scc_dead = unix_vertex_dead(vertex); } + if (scc_dead) + unix_collect_skb(&scc, hitlist); + list_del(&scc); } list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); } -static LIST_HEAD(gc_candidates); -static LIST_HEAD(gc_inflight_list); - -/* Keep the number of times in flight count for the file - * descriptor if it is for an AF_UNIX socket. - */ -void unix_inflight(struct user_struct *user, struct file *filp) -{ - struct unix_sock *u = unix_get_socket(filp); - - spin_lock(&unix_gc_lock); - - if (u) { - if (!u->inflight) { - WARN_ON_ONCE(!list_empty(&u->link)); - list_add_tail(&u->link, &gc_inflight_list); - } else { - WARN_ON_ONCE(list_empty(&u->link)); - } - u->inflight++; - } - - spin_unlock(&unix_gc_lock); -} - -void unix_notinflight(struct user_struct *user, struct file *filp) -{ - struct unix_sock *u = unix_get_socket(filp); - - spin_lock(&unix_gc_lock); - - if (u) { - WARN_ON_ONCE(!u->inflight); - WARN_ON_ONCE(list_empty(&u->link)); - - u->inflight--; - if (!u->inflight) - list_del_init(&u->link); - } - - spin_unlock(&unix_gc_lock); -} - -static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *), - struct sk_buff_head *hitlist) -{ - struct sk_buff *skb; - struct sk_buff *next; - - spin_lock(&x->sk_receive_queue.lock); - skb_queue_walk_safe(&x->sk_receive_queue, skb, next) { - /* Do we have file descriptors ? */ - if (UNIXCB(skb).fp) { - bool hit = false; - /* Process the descriptors of this socket */ - int nfd = UNIXCB(skb).fp->count; - struct file **fp = UNIXCB(skb).fp->fp; - - while (nfd--) { - /* Get the socket the fd matches if it indeed does so */ - struct unix_sock *u = unix_get_socket(*fp++); - - /* Ignore non-candidates, they could have been added - * to the queues after starting the garbage collection - */ - if (u && test_bit(UNIX_GC_CANDIDATE, &u->gc_flags)) { - hit = true; - - func(u); - } - } - if (hit && hitlist != NULL) { - __skb_unlink(skb, &x->sk_receive_queue); - __skb_queue_tail(hitlist, skb); - } - } - } - spin_unlock(&x->sk_receive_queue.lock); -} - -static void scan_children(struct sock *x, void (*func)(struct unix_sock *), - struct sk_buff_head *hitlist) -{ - if (x->sk_state != TCP_LISTEN) { - scan_inflight(x, func, hitlist); - } else { - struct sk_buff *skb; - struct sk_buff *next; - struct unix_sock *u; - LIST_HEAD(embryos); - - /* For a listening socket collect the queued embryos - * and perform a scan on them as well. - */ - spin_lock(&x->sk_receive_queue.lock); - skb_queue_walk_safe(&x->sk_receive_queue, skb, next) { - u = unix_sk(skb->sk); - - /* An embryo cannot be in-flight, so it's safe - * to use the list link. - */ - WARN_ON_ONCE(!list_empty(&u->link)); - list_add_tail(&u->link, &embryos); - } - spin_unlock(&x->sk_receive_queue.lock); - - while (!list_empty(&embryos)) { - u = list_entry(embryos.next, struct unix_sock, link); - scan_inflight(&u->sk, func, hitlist); - list_del_init(&u->link); - } - } -} - -static void dec_inflight(struct unix_sock *usk) -{ - usk->inflight--; -} - -static void inc_inflight(struct unix_sock *usk) -{ - usk->inflight++; -} - -static void inc_inflight_move_tail(struct unix_sock *u) -{ - u->inflight++; - - /* If this still might be part of a cycle, move it to the end - * of the list, so that it's checked even if it was already - * passed over - */ - if (test_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags)) - list_move_tail(&u->link, &gc_candidates); -} - static bool gc_in_progress; static void __unix_gc(struct work_struct *work) { struct sk_buff_head hitlist; - struct unix_sock *u, *next; - LIST_HEAD(not_cycle_list); - struct list_head cursor; spin_lock(&unix_gc_lock); - if (!unix_graph_maybe_cyclic) + if (!unix_graph_maybe_cyclic) { + spin_unlock(&unix_gc_lock); goto skip_gc; - - if (unix_graph_grouped) - unix_walk_scc_fast(); - else - unix_walk_scc(); - - /* First, select candidates for garbage collection. Only - * in-flight sockets are considered, and from those only ones - * which don't have any external reference. - * - * Holding unix_gc_lock will protect these candidates from - * being detached, and hence from gaining an external - * reference. Since there are no possible receivers, all - * buffers currently on the candidates' queues stay there - * during the garbage collection. - * - * We also know that no new candidate can be added onto the - * receive queues. Other, non candidate sockets _can_ be - * added to queue, so we must make sure only to touch - * candidates. - */ - list_for_each_entry_safe(u, next, &gc_inflight_list, link) { - long total_refs; - - total_refs = file_count(u->sk.sk_socket->file); - - WARN_ON_ONCE(!u->inflight); - WARN_ON_ONCE(total_refs < u->inflight); - if (total_refs == u->inflight) { - list_move_tail(&u->link, &gc_candidates); - __set_bit(UNIX_GC_CANDIDATE, &u->gc_flags); - __set_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags); - } - } - - /* Now remove all internal in-flight reference to children of - * the candidates. - */ - list_for_each_entry(u, &gc_candidates, link) - scan_children(&u->sk, dec_inflight, NULL); - - /* Restore the references for children of all candidates, - * which have remaining references. Do this recursively, so - * only those remain, which form cyclic references. - * - * Use a "cursor" link, to make the list traversal safe, even - * though elements might be moved about. - */ - list_add(&cursor, &gc_candidates); - while (cursor.next != &gc_candidates) { - u = list_entry(cursor.next, struct unix_sock, link); - - /* Move cursor to after the current position. */ - list_move(&cursor, &u->link); - - if (u->inflight) { - list_move_tail(&u->link, ¬_cycle_list); - __clear_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags); - scan_children(&u->sk, inc_inflight_move_tail, NULL); - } } - list_del(&cursor); - /* Now gc_candidates contains only garbage. Restore original - * inflight counters for these as well, and remove the skbuffs - * which are creating the cycle(s). - */ - skb_queue_head_init(&hitlist); - list_for_each_entry(u, &gc_candidates, link) { - scan_children(&u->sk, inc_inflight, &hitlist); + __skb_queue_head_init(&hitlist); -#if IS_ENABLED(CONFIG_AF_UNIX_OOB) - if (u->oob_skb) { - kfree_skb(u->oob_skb); - u->oob_skb = NULL; - } -#endif - } - - /* not_cycle_list contains those sockets which do not make up a - * cycle. Restore these to the inflight list. - */ - while (!list_empty(¬_cycle_list)) { - u = list_entry(not_cycle_list.next, struct unix_sock, link); - __clear_bit(UNIX_GC_CANDIDATE, &u->gc_flags); - list_move_tail(&u->link, &gc_inflight_list); - } + if (unix_graph_grouped) + unix_walk_scc_fast(&hitlist); + else + unix_walk_scc(&hitlist); spin_unlock(&unix_gc_lock); - /* Here we are. Hitlist is filled. Die. */ __skb_queue_purge(&hitlist); - - spin_lock(&unix_gc_lock); - - /* All candidates should have been detached by now. */ - WARN_ON_ONCE(!list_empty(&gc_candidates)); skip_gc: - /* Paired with READ_ONCE() in wait_for_unix_gc(). */ WRITE_ONCE(gc_in_progress, false); - - spin_unlock(&unix_gc_lock); } static DECLARE_WORK(unix_gc_work, __unix_gc); -- cgit v1.2.3 From 3ae5cb7efaf1e33c2cf2e7959efea363b59fda84 Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Fri, 29 Mar 2024 20:16:23 +0100 Subject: batman-adv: Start new development cycle This version will contain all the (major or even only minor) changes for Linux 6.10. The version number isn't a semantic version number with major and minor information. It is just encoding the year of the expected publishing as Linux -rc1 and the number of published versions this year (starting at 0). Signed-off-by: Simon Wunderlich --- net/batman-adv/main.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 8ca854a75a32..3d4c36ae2e1a 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -13,7 +13,7 @@ #define BATADV_DRIVER_DEVICE "batman-adv" #ifndef BATADV_SOURCE_VERSION -#define BATADV_SOURCE_VERSION "2024.1" +#define BATADV_SOURCE_VERSION "2024.2" #endif /* B.A.T.M.A.N. parameters */ -- cgit v1.2.3 From 537db05da2ca8ccc1243c9dd1d0c148b84aa0432 Mon Sep 17 00:00:00 2001 From: Dmitry Antipov Date: Tue, 12 Mar 2024 21:16:28 +0300 Subject: batman-adv: prefer kfree_rcu() over call_rcu() with free-only callbacks Drop 'batadv_tt_local_entry_free_rcu()', 'batadv_tt_global_entry_free_rcu()' and 'batadv_tt_orig_list_entry_free_rcu()' in favor of 'kfree_rcu()' in 'batadv_tt_local_entry_release()', 'batadv_tt_global_entry_release()' and 'batadv_tt_orig_list_entry_release()', respectively. Signed-off-by: Dmitry Antipov Acked-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/translation-table.c | 47 +++----------------------------------- 1 file changed, 3 insertions(+), 44 deletions(-) (limited to 'net') diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index b95c36765d04..0555cb611489 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -208,20 +208,6 @@ batadv_tt_global_hash_find(struct batadv_priv *bat_priv, const u8 *addr, return tt_global_entry; } -/** - * batadv_tt_local_entry_free_rcu() - free the tt_local_entry - * @rcu: rcu pointer of the tt_local_entry - */ -static void batadv_tt_local_entry_free_rcu(struct rcu_head *rcu) -{ - struct batadv_tt_local_entry *tt_local_entry; - - tt_local_entry = container_of(rcu, struct batadv_tt_local_entry, - common.rcu); - - kmem_cache_free(batadv_tl_cache, tt_local_entry); -} - /** * batadv_tt_local_entry_release() - release tt_local_entry from lists and queue * for free after rcu grace period @@ -236,7 +222,7 @@ static void batadv_tt_local_entry_release(struct kref *ref) batadv_softif_vlan_put(tt_local_entry->vlan); - call_rcu(&tt_local_entry->common.rcu, batadv_tt_local_entry_free_rcu); + kfree_rcu(tt_local_entry, common.rcu); } /** @@ -254,20 +240,6 @@ batadv_tt_local_entry_put(struct batadv_tt_local_entry *tt_local_entry) batadv_tt_local_entry_release); } -/** - * batadv_tt_global_entry_free_rcu() - free the tt_global_entry - * @rcu: rcu pointer of the tt_global_entry - */ -static void batadv_tt_global_entry_free_rcu(struct rcu_head *rcu) -{ - struct batadv_tt_global_entry *tt_global_entry; - - tt_global_entry = container_of(rcu, struct batadv_tt_global_entry, - common.rcu); - - kmem_cache_free(batadv_tg_cache, tt_global_entry); -} - /** * batadv_tt_global_entry_release() - release tt_global_entry from lists and * queue for free after rcu grace period @@ -282,7 +254,7 @@ void batadv_tt_global_entry_release(struct kref *ref) batadv_tt_global_del_orig_list(tt_global_entry); - call_rcu(&tt_global_entry->common.rcu, batadv_tt_global_entry_free_rcu); + kfree_rcu(tt_global_entry, common.rcu); } /** @@ -407,19 +379,6 @@ static void batadv_tt_global_size_dec(struct batadv_orig_node *orig_node, batadv_tt_global_size_mod(orig_node, vid, -1); } -/** - * batadv_tt_orig_list_entry_free_rcu() - free the orig_entry - * @rcu: rcu pointer of the orig_entry - */ -static void batadv_tt_orig_list_entry_free_rcu(struct rcu_head *rcu) -{ - struct batadv_tt_orig_list_entry *orig_entry; - - orig_entry = container_of(rcu, struct batadv_tt_orig_list_entry, rcu); - - kmem_cache_free(batadv_tt_orig_cache, orig_entry); -} - /** * batadv_tt_orig_list_entry_release() - release tt orig entry from lists and * queue for free after rcu grace period @@ -433,7 +392,7 @@ static void batadv_tt_orig_list_entry_release(struct kref *ref) refcount); batadv_orig_node_put(orig_entry->orig_node); - call_rcu(&orig_entry->rcu, batadv_tt_orig_list_entry_free_rcu); + kfree_rcu(orig_entry, rcu); } /** -- cgit v1.2.3 From e9669a00bba79442dd4862c57761333d6a020c24 Mon Sep 17 00:00:00 2001 From: Balazs Scheidler Date: Tue, 26 Mar 2024 19:05:47 +0100 Subject: net: udp: add IP/port data to the tracepoint udp/udp_fail_queue_rcv_skb The udp_fail_queue_rcv_skb() tracepoint lacks any details on the source and destination IP/port whereas this information can be critical in case of UDP/syslog. Signed-off-by: Balazs Scheidler Reviewed-by: Jason Xing Link: https://lore.kernel.org/r/0c8b3e33dbf679e190be6f4c6736603a76988a20.1711475011.git.balazs.scheidler@axoflow.com Signed-off-by: Jakub Kicinski --- include/trace/events/udp.h | 29 ++++++++++++++++++++++++----- net/ipv4/udp.c | 2 +- net/ipv6/udp.c | 3 ++- 3 files changed, 27 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/trace/events/udp.h b/include/trace/events/udp.h index 336fe272889f..62bebe2a6ece 100644 --- a/include/trace/events/udp.h +++ b/include/trace/events/udp.h @@ -7,24 +7,43 @@ #include #include +#include TRACE_EVENT(udp_fail_queue_rcv_skb, - TP_PROTO(int rc, struct sock *sk), + TP_PROTO(int rc, struct sock *sk, struct sk_buff *skb), - TP_ARGS(rc, sk), + TP_ARGS(rc, sk, skb), TP_STRUCT__entry( __field(int, rc) - __field(__u16, lport) + + __field(__u16, sport) + __field(__u16, dport) + __field(__u16, family) + __array(__u8, saddr, sizeof(struct sockaddr_in6)) + __array(__u8, daddr, sizeof(struct sockaddr_in6)) ), TP_fast_assign( + const struct udphdr *uh = (const struct udphdr *)udp_hdr(skb); + __entry->rc = rc; - __entry->lport = inet_sk(sk)->inet_num; + + /* for filtering use */ + __entry->sport = ntohs(uh->source); + __entry->dport = ntohs(uh->dest); + __entry->family = sk->sk_family; + + memset(__entry->saddr, 0, sizeof(struct sockaddr_in6)); + memset(__entry->daddr, 0, sizeof(struct sockaddr_in6)); + + TP_STORE_ADDR_PORTS_SKB(__entry, skb, uh); ), - TP_printk("rc=%d port=%hu", __entry->rc, __entry->lport) + TP_printk("rc=%d family=%s src=%pISpc dest=%pISpc", __entry->rc, + show_family_name(__entry->family), + __entry->saddr, __entry->daddr) ); #endif /* _TRACE_UDP_H */ diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 661d0e0d273f..531882f321f2 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -2049,8 +2049,8 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) drop_reason = SKB_DROP_REASON_PROTO_MEM; } UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite); + trace_udp_fail_queue_rcv_skb(rc, sk, skb); kfree_skb_reason(skb, drop_reason); - trace_udp_fail_queue_rcv_skb(rc, sk); return -1; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 7c1e6469d091..2e4dc5e6137b 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -658,8 +659,8 @@ static int __udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) drop_reason = SKB_DROP_REASON_PROTO_MEM; } UDP6_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite); + trace_udp_fail_queue_rcv_skb(rc, sk, skb); kfree_skb_reason(skb, drop_reason); - trace_udp_fail_queue_rcv_skb(rc, sk); return -1; } -- cgit v1.2.3 From 730fffce4fd2eb7a0be2d0b6cd7e55e9194d76d5 Mon Sep 17 00:00:00 2001 From: Jian Wen Date: Wed, 27 Mar 2024 16:21:28 +0800 Subject: devlink: use kvzalloc() to allocate devlink instance resources During live migration of a virtual machine, the SR-IOV VF need to be re-registered. It may fail when the memory is badly fragmented. The related log is as follows. kernel: hv_netvsc 6045bdaa-c0d1-6045-bdaa-c0d16045bdaa eth0: VF slot 1 added ... kernel: kworker/0:0: page allocation failure: order:7, mode:0x40dc0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO), nodemask=(null),cpuset=/,mems_allowed=0 kernel: CPU: 0 PID: 24006 Comm: kworker/0:0 Tainted: G E 5.4...x86_64 #1 kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090008 12/07/2018 kernel: Workqueue: events work_for_cpu_fn kernel: Call Trace: kernel: dump_stack+0x8b/0xc8 kernel: warn_alloc+0xff/0x170 kernel: __alloc_pages_slowpath+0x92c/0xb2b kernel: ? get_page_from_freelist+0x1d4/0x1140 kernel: __alloc_pages_nodemask+0x2f9/0x320 kernel: alloc_pages_current+0x6a/0xb0 kernel: kmalloc_order+0x1e/0x70 kernel: kmalloc_order_trace+0x26/0xb0 kernel: ? __switch_to_asm+0x34/0x70 kernel: __kmalloc+0x276/0x280 kernel: ? _raw_spin_unlock_irqrestore+0x1e/0x40 kernel: devlink_alloc+0x29/0x110 kernel: mlx5_devlink_alloc+0x1a/0x20 [mlx5_core] kernel: init_one+0x1d/0x650 [mlx5_core] kernel: local_pci_probe+0x46/0x90 kernel: work_for_cpu_fn+0x1a/0x30 kernel: process_one_work+0x16d/0x390 kernel: worker_thread+0x1d3/0x3f0 kernel: kthread+0x105/0x140 kernel: ? max_active_store+0x80/0x80 kernel: ? kthread_bind+0x20/0x20 kernel: ret_from_fork+0x3a/0x50 Signed-off-by: Jian Wen Link: https://lore.kernel.org/r/20240327082128.942818-1-wenjian1@xiaomi.com Signed-off-by: Jakub Kicinski --- net/devlink/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/devlink/core.c b/net/devlink/core.c index 7f0b093208d7..f49cd83f1955 100644 --- a/net/devlink/core.c +++ b/net/devlink/core.c @@ -314,7 +314,7 @@ static void devlink_release(struct work_struct *work) mutex_destroy(&devlink->lock); lockdep_unregister_key(&devlink->lock_key); put_device(devlink->dev); - kfree(devlink); + kvfree(devlink); } void devlink_put(struct devlink *devlink) @@ -420,7 +420,7 @@ struct devlink *devlink_alloc_ns(const struct devlink_ops *ops, if (!devlink_reload_actions_valid(ops)) return NULL; - devlink = kzalloc(sizeof(*devlink) + priv_size, GFP_KERNEL); + devlink = kvzalloc(struct_size(devlink, priv, priv_size), GFP_KERNEL); if (!devlink) return NULL; @@ -455,7 +455,7 @@ struct devlink *devlink_alloc_ns(const struct devlink_ops *ops, return devlink; err_xa_alloc: - kfree(devlink); + kvfree(devlink); return NULL; } EXPORT_SYMBOL_GPL(devlink_alloc_ns); -- cgit v1.2.3 From 60557969951304dad829f2829019907dfb43ecb3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 28 Mar 2024 14:40:29 +0000 Subject: udp: annotate data-race in __udp_enqueue_schedule_skb() sk->sk_rcvbuf is read locklessly twice, while other threads could change its value. Use a READ_ONCE() to annotate the race. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240328144032.1864988-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/udp.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 531882f321f2..6a39e7fa0616 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1492,13 +1492,14 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) struct sk_buff_head *list = &sk->sk_receive_queue; int rmem, err = -ENOMEM; spinlock_t *busy = NULL; - int size; + int size, rcvbuf; - /* try to avoid the costly atomic add/sub pair when the receive - * queue is full; always allow at least a packet + /* Immediately drop when the receive queue is full. + * Always allow at least one packet. */ rmem = atomic_read(&sk->sk_rmem_alloc); - if (rmem > sk->sk_rcvbuf) + rcvbuf = READ_ONCE(sk->sk_rcvbuf); + if (rmem > rcvbuf) goto drop; /* Under mem pressure, it might be helpful to help udp_recvmsg() @@ -1507,7 +1508,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) * - Less cache line misses at copyout() time * - Less work at consume_skb() (less alien page frag freeing) */ - if (rmem > (sk->sk_rcvbuf >> 1)) { + if (rmem > (rcvbuf >> 1)) { skb_condense(skb); busy = busylock_acquire(sk); -- cgit v1.2.3 From 6a1f12dd85a8b24f871dfcf467378660af9c064d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 28 Mar 2024 14:40:30 +0000 Subject: udp: relax atomic operation on sk->sk_rmem_alloc atomic_add_return() is more expensive than atomic_add() and seems overkill in UDP rx fast path. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240328144032.1864988-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/udp.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 6a39e7fa0616..19d7db4563ac 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1516,12 +1516,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) size = skb->truesize; udp_set_dev_scratch(skb); - /* we drop only if the receive buf is full and the receive - * queue contains some other skb - */ - rmem = atomic_add_return(size, &sk->sk_rmem_alloc); - if (rmem > (size + (unsigned int)sk->sk_rcvbuf)) - goto uncharge_drop; + atomic_add(size, &sk->sk_rmem_alloc); spin_lock(&list->lock); err = udp_rmem_schedule(sk, size); -- cgit v1.2.3 From 612b1c0dec5bc7367f90fc508448b8d0d7c05414 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 28 Mar 2024 14:40:31 +0000 Subject: udp: avoid calling sock_def_readable() if possible sock_def_readable() is quite expensive (particularly when ep_poll_callback() is in the picture). We must call sk->sk_data_ready() when : - receive queue was empty, or - SO_PEEK_OFF is enabled on the socket, or - sk->sk_data_ready is not sock_def_readable. We still need to call sk_wake_async(). Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240328144032.1864988-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/udp.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 19d7db4563ac..143043cd2dcb 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1492,6 +1492,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) struct sk_buff_head *list = &sk->sk_receive_queue; int rmem, err = -ENOMEM; spinlock_t *busy = NULL; + bool becomes_readable; int size, rcvbuf; /* Immediately drop when the receive queue is full. @@ -1532,12 +1533,19 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) */ sock_skb_set_dropcount(sk, skb); + becomes_readable = skb_queue_empty(list); __skb_queue_tail(list, skb); spin_unlock(&list->lock); - if (!sock_flag(sk, SOCK_DEAD)) - INDIRECT_CALL_1(sk->sk_data_ready, sock_def_readable, sk); - + if (!sock_flag(sk, SOCK_DEAD)) { + if (becomes_readable || + sk->sk_data_ready != sock_def_readable || + READ_ONCE(sk->sk_peek_off) >= 0) + INDIRECT_CALL_1(sk->sk_data_ready, + sock_def_readable, sk); + else + sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + } busylock_release(busy); return 0; -- cgit v1.2.3 From 1abe267f173eae7ae76cf56232292e9641eb652f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 28 Mar 2024 14:40:32 +0000 Subject: net: add sk_wake_async_rcu() helper While looking at UDP receive performance, I saw sk_wake_async() was no longer inlined. This matters at least on AMD Zen1-4 platforms (see SRSO) This might be because rcu_read_lock() and rcu_read_unlock() are no longer nops in recent kernels ? Add sk_wake_async_rcu() variant, which must be called from contexts already holding rcu lock. As SOCK_FASYNC is deprecated in modern days, use unlikely() to give a hint to the compiler. sk_wake_async_rcu() is properly inlined from __udp_enqueue_schedule_skb() and sock_def_readable(). Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240328144032.1864988-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- crypto/af_alg.c | 4 ++-- include/net/sock.h | 6 ++++++ net/atm/common.c | 2 +- net/core/sock.c | 8 ++++---- net/dccp/output.c | 2 +- net/ipv4/udp.c | 2 +- net/iucv/af_iucv.c | 2 +- net/rxrpc/af_rxrpc.c | 2 +- net/sctp/socket.c | 2 +- net/smc/smc_rx.c | 4 ++-- net/unix/af_unix.c | 2 +- 11 files changed, 21 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/crypto/af_alg.c b/crypto/af_alg.c index 68cc9290cabe..5bc6d0fa7498 100644 --- a/crypto/af_alg.c +++ b/crypto/af_alg.c @@ -847,7 +847,7 @@ void af_alg_wmem_wakeup(struct sock *sk) wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN | EPOLLRDNORM | EPOLLRDBAND); - sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_IN); rcu_read_unlock(); } EXPORT_SYMBOL_GPL(af_alg_wmem_wakeup); @@ -914,7 +914,7 @@ static void af_alg_data_wakeup(struct sock *sk) wake_up_interruptible_sync_poll(&wq->wait, EPOLLOUT | EPOLLRDNORM | EPOLLRDBAND); - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); rcu_read_unlock(); } diff --git a/include/net/sock.h b/include/net/sock.h index f57bfd8a2ad2..2253eefe2848 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2513,6 +2513,12 @@ static inline void sk_wake_async(const struct sock *sk, int how, int band) } } +static inline void sk_wake_async_rcu(const struct sock *sk, int how, int band) +{ + if (unlikely(sock_flag(sk, SOCK_FASYNC))) + sock_wake_async(rcu_dereference(sk->sk_wq), how, band); +} + /* Since sk_{r,w}mem_alloc sums skb->truesize, even a small frame might * need sizeof(sk_buff) + MTU + padding, unless net driver perform copybreak. * Note: for send buffers, TCP works better if we can build two skbs at diff --git a/net/atm/common.c b/net/atm/common.c index 2a1ec014e901..9b75699992ff 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -116,7 +116,7 @@ static void vcc_write_space(struct sock *sk) if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); } rcu_read_unlock(); diff --git a/net/core/sock.c b/net/core/sock.c index 0963689a5950..5ed411231fc7 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3338,7 +3338,7 @@ static void sock_def_error_report(struct sock *sk) wq = rcu_dereference(sk->sk_wq); if (skwq_has_sleeper(wq)) wake_up_interruptible_poll(&wq->wait, EPOLLERR); - sk_wake_async(sk, SOCK_WAKE_IO, POLL_ERR); + sk_wake_async_rcu(sk, SOCK_WAKE_IO, POLL_ERR); rcu_read_unlock(); } @@ -3353,7 +3353,7 @@ void sock_def_readable(struct sock *sk) if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN | EPOLLPRI | EPOLLRDNORM | EPOLLRDBAND); - sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_IN); rcu_read_unlock(); } @@ -3373,7 +3373,7 @@ static void sock_def_write_space(struct sock *sk) EPOLLWRNORM | EPOLLWRBAND); /* Should agree with poll, otherwise some programs break */ - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); } rcu_read_unlock(); @@ -3398,7 +3398,7 @@ static void sock_def_write_space_wfree(struct sock *sk) EPOLLWRNORM | EPOLLWRBAND); /* Should agree with poll, otherwise some programs break */ - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); } } diff --git a/net/dccp/output.c b/net/dccp/output.c index fd2eb148d24d..5c2e24f3c39b 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -204,7 +204,7 @@ void dccp_write_space(struct sock *sk) wake_up_interruptible(&wq->wait); /* Should agree with poll, otherwise some programs break */ if (sock_writeable(sk)) - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); rcu_read_unlock(); } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 143043cd2dcb..11460d751e73 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1544,7 +1544,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb) INDIRECT_CALL_1(sk->sk_data_ready, sock_def_readable, sk); else - sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_IN); } busylock_release(busy); return 0; diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 7c8c3adcac6e..c951bb9cc2e0 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -184,7 +184,7 @@ static void iucv_sock_wake_msglim(struct sock *sk) wq = rcu_dereference(sk->sk_wq); if (skwq_has_sleeper(wq)) wake_up_interruptible_all(&wq->wait); - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); rcu_read_unlock(); } diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 5222bc97d192..f4844683e120 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -65,7 +65,7 @@ static void rxrpc_write_space(struct sock *sk) if (skwq_has_sleeper(wq)) wake_up_interruptible(&wq->wait); - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); } rcu_read_unlock(); } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index c67679a41044..e416b6d3d270 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -9276,7 +9276,7 @@ void sctp_data_ready(struct sock *sk) if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN | EPOLLRDNORM | EPOLLRDBAND); - sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_IN); rcu_read_unlock(); } diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c index 9a2f3638d161..f0cbe77a80b4 100644 --- a/net/smc/smc_rx.c +++ b/net/smc/smc_rx.c @@ -42,10 +42,10 @@ static void smc_rx_wake_up(struct sock *sk) if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, EPOLLIN | EPOLLPRI | EPOLLRDNORM | EPOLLRDBAND); - sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN); + sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_IN); if ((sk->sk_shutdown == SHUTDOWN_MASK) || (sk->sk_state == SMC_CLOSED)) - sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP); + sk_wake_async_rcu(sk, SOCK_WAKE_WAITD, POLL_HUP); rcu_read_unlock(); } diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 27ca50ab1cd1..533fb682c954 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -546,7 +546,7 @@ static void unix_write_space(struct sock *sk) if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, EPOLLOUT | EPOLLWRNORM | EPOLLWRBAND); - sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT); + sk_wake_async_rcu(sk, SOCK_WAKE_SPACE, POLL_OUT); } rcu_read_unlock(); } -- cgit v1.2.3 From e8058a49e67fe7bc7e4a0308851a3ca3a6d2e45d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 28 Mar 2024 20:31:45 +0100 Subject: netlink: introduce type-checking attribute iteration There are, especially with multi-attr arrays, many cases of needing to iterate all attributes of a specific type in a netlink message or a nested attribute. Add specific macros to support that case. Also convert many instances using this spatch: @@ iterator nla_for_each_attr; iterator name nla_for_each_attr_type; identifier nla; expression head, len, rem; expression ATTR; type T; identifier x; @@ -nla_for_each_attr(nla, head, len, rem) +nla_for_each_attr_type(nla, ATTR, head, len, rem) { <... T x; ...> -if (nla_type(nla) == ATTR) { ... -} } @@ identifier nla; iterator nla_for_each_nested; iterator name nla_for_each_nested_type; expression attr, rem; expression ATTR; type T; identifier x; @@ -nla_for_each_nested(nla, attr, rem) +nla_for_each_nested_type(nla, ATTR, attr, rem) { <... T x; ...> -if (nla_type(nla) == ATTR) { ... -} } @@ iterator nla_for_each_attr; iterator name nla_for_each_attr_type; identifier nla; expression head, len, rem; expression ATTR; type T; identifier x; @@ -nla_for_each_attr(nla, head, len, rem) +nla_for_each_attr_type(nla, ATTR, head, len, rem) { <... T x; ...> -if (nla_type(nla) != ATTR) continue; ... } @@ identifier nla; iterator nla_for_each_nested; iterator name nla_for_each_nested_type; expression attr, rem; expression ATTR; type T; identifier x; @@ -nla_for_each_nested(nla, attr, rem) +nla_for_each_nested_type(nla, ATTR, attr, rem) { <... T x; ...> -if (nla_type(nla) != ATTR) continue; ... } Although I had to undo one bad change this made, and I also adjusted some other code for whitespace and to use direct variable initialization now. Signed-off-by: Johannes Berg Link: https://lore.kernel.org/r/20240328203144.b5a6c895fb80.I1869b44767379f204998ff44dd239803f39c23e0@changeid Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 5 +--- drivers/net/ethernet/emulex/benet/be_main.c | 5 +--- drivers/net/ethernet/intel/i40e/i40e_main.c | 8 ++----- drivers/net/ethernet/intel/ice/ice_main.c | 7 ++---- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 11 +++------ drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 5 +--- .../net/ethernet/netronome/nfp/nfp_net_common.c | 5 +--- include/net/netlink.h | 27 ++++++++++++++++++++++ net/8021q/vlan_netlink.c | 10 ++++---- net/core/bpf_sk_storage.c | 23 ++++++++---------- net/core/rtnetlink.c | 15 ++++++------ net/devlink/dev.c | 12 ++++------ net/sched/sch_mqprio.c | 6 ++--- net/sched/sch_taprio.c | 5 +--- 14 files changed, 65 insertions(+), 79 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 388e80bf91f5..b4db4b1aaffb 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -14581,12 +14581,9 @@ static int bnxt_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { u16 mode; - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - mode = nla_get_u16(attr); if (mode == bp->br_mode) break; diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index ad862ed7888a..a8596ebcdfd6 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -4982,10 +4982,7 @@ static int be_ndo_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { mode = nla_get_u16(attr); if (BE3_chip(adapter) && mode == BRIDGE_MODE_VEPA) return -EOPNOTSUPP; diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index f86578857e8a..e427e65af205 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -13108,13 +13108,9 @@ static int i40e_ndo_bridge_setlink(struct net_device *dev, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - __u16 mode; + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { + __u16 mode = nla_get_u16(attr); - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - - mode = nla_get_u16(attr); if ((mode != BRIDGE_MODE_VEPA) && (mode != BRIDGE_MODE_VEB)) return -EINVAL; diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index f2b7d6ca8805..618570f23580 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -7993,12 +7993,9 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - __u16 mode; + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { + __u16 mode = nla_get_u16(attr); - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - mode = nla_get_u16(attr); if (mode != BRIDGE_MODE_VEPA && mode != BRIDGE_MODE_VEB) return -EINVAL; /* Continue if bridge mode is not being flipped */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index f985252c8c8d..ed05af665466 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -10061,15 +10061,10 @@ static int ixgbe_ndo_bridge_setlink(struct net_device *dev, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - int status; - __u16 mode; + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { + __u16 mode = nla_get_u16(attr); + int status = ixgbe_configure_bridge_mode(adapter, mode); - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - - mode = nla_get_u16(attr); - status = ixgbe_configure_bridge_mode(adapter, mode); if (status) return status; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 91848eae4565..81e1c1e401f9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -4950,10 +4950,7 @@ static int mlx5e_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { mode = nla_get_u16(attr); if (mode > BRIDGE_MODE_VEPA) return -EINVAL; diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index f28e769e6fda..997cc4fcffdb 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -2289,10 +2289,7 @@ static int nfp_net_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, if (!br_spec) return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { new_ctrl = nn->dp.ctrl; mode = nla_get_u16(attr); if (mode == BRIDGE_MODE_VEPA) diff --git a/include/net/netlink.h b/include/net/netlink.h index c19ff921b661..1d2bbcc50212 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -157,7 +157,11 @@ * nla_parse() parse and validate stream of attrs * nla_parse_nested() parse nested attributes * nla_for_each_attr() loop over all attributes + * nla_for_each_attr_type() loop over all attributes with the + * given type * nla_for_each_nested() loop over the nested attributes + * nla_for_each_nested_type() loop over the nested attributes with + * the given type *========================================================================= */ @@ -2070,6 +2074,18 @@ static inline int nla_total_size_64bit(int payload) nla_ok(pos, rem); \ pos = nla_next(pos, &(rem))) +/** + * nla_for_each_attr_type - iterate over a stream of attributes + * @pos: loop counter, set to current attribute + * @type: required attribute type for @pos + * @head: head of attribute stream + * @len: length of attribute stream + * @rem: initialized to len, holds bytes currently remaining in stream + */ +#define nla_for_each_attr_type(pos, type, head, len, rem) \ + nla_for_each_attr(pos, head, len, rem) \ + if (nla_type(pos) == type) + /** * nla_for_each_nested - iterate over nested attributes * @pos: loop counter, set to current attribute @@ -2079,6 +2095,17 @@ static inline int nla_total_size_64bit(int payload) #define nla_for_each_nested(pos, nla, rem) \ nla_for_each_attr(pos, nla_data(nla), nla_len(nla), rem) +/** + * nla_for_each_nested_type - iterate over nested attributes + * @pos: loop counter, set to current attribute + * @type: required attribute type for @pos + * @nla: attribute containing the nested attributes + * @rem: initialized to len, holds bytes currently remaining in stream + */ +#define nla_for_each_nested_type(pos, type, nla, rem) \ + nla_for_each_nested(pos, nla, rem) \ + if (nla_type(pos) == type) + /** * nla_is_last - Test if attribute is last in stream * @nla: attribute to test diff --git a/net/8021q/vlan_netlink.c b/net/8021q/vlan_netlink.c index a3b68243fd4b..cf5219df7903 100644 --- a/net/8021q/vlan_netlink.c +++ b/net/8021q/vlan_netlink.c @@ -117,17 +117,15 @@ static int vlan_changelink(struct net_device *dev, struct nlattr *tb[], return err; } if (data[IFLA_VLAN_INGRESS_QOS]) { - nla_for_each_nested(attr, data[IFLA_VLAN_INGRESS_QOS], rem) { - if (nla_type(attr) != IFLA_VLAN_QOS_MAPPING) - continue; + nla_for_each_nested_type(attr, IFLA_VLAN_QOS_MAPPING, + data[IFLA_VLAN_INGRESS_QOS], rem) { m = nla_data(attr); vlan_dev_set_ingress_priority(dev, m->to, m->from); } } if (data[IFLA_VLAN_EGRESS_QOS]) { - nla_for_each_nested(attr, data[IFLA_VLAN_EGRESS_QOS], rem) { - if (nla_type(attr) != IFLA_VLAN_QOS_MAPPING) - continue; + nla_for_each_nested_type(attr, IFLA_VLAN_QOS_MAPPING, + data[IFLA_VLAN_EGRESS_QOS], rem) { m = nla_data(attr); err = vlan_dev_set_egress_priority(dev, m->from, m->to); if (err) diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 6c4d90b24d46..bc01b3aa6b0f 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -496,27 +496,22 @@ bpf_sk_storage_diag_alloc(const struct nlattr *nla_stgs) if (!bpf_capable()) return ERR_PTR(-EPERM); - nla_for_each_nested(nla, nla_stgs, rem) { - if (nla_type(nla) == SK_DIAG_BPF_STORAGE_REQ_MAP_FD) { - if (nla_len(nla) != sizeof(u32)) - return ERR_PTR(-EINVAL); - nr_maps++; - } + nla_for_each_nested_type(nla, SK_DIAG_BPF_STORAGE_REQ_MAP_FD, + nla_stgs, rem) { + if (nla_len(nla) != sizeof(u32)) + return ERR_PTR(-EINVAL); + nr_maps++; } diag = kzalloc(struct_size(diag, maps, nr_maps), GFP_KERNEL); if (!diag) return ERR_PTR(-ENOMEM); - nla_for_each_nested(nla, nla_stgs, rem) { - struct bpf_map *map; - int map_fd; - - if (nla_type(nla) != SK_DIAG_BPF_STORAGE_REQ_MAP_FD) - continue; + nla_for_each_nested_type(nla, SK_DIAG_BPF_STORAGE_REQ_MAP_FD, + nla_stgs, rem) { + int map_fd = nla_get_u32(nla); + struct bpf_map *map = bpf_map_get(map_fd); - map_fd = nla_get_u32(nla); - map = bpf_map_get(map_fd); if (IS_ERR(map)) { err = PTR_ERR(map); goto err_free; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a3d7847ce69d..283e42f48af6 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -5245,15 +5245,14 @@ static int rtnl_bridge_dellink(struct sk_buff *skb, struct nlmsghdr *nlh, br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC); if (br_spec) { - nla_for_each_nested(attr, br_spec, rem) { - if (nla_type(attr) == IFLA_BRIDGE_FLAGS) { - if (nla_len(attr) < sizeof(flags)) - return -EINVAL; + nla_for_each_nested_type(attr, IFLA_BRIDGE_FLAGS, br_spec, + rem) { + if (nla_len(attr) < sizeof(flags)) + return -EINVAL; - have_flags = true; - flags = nla_get_u16(attr); - break; - } + have_flags = true; + flags = nla_get_u16(attr); + break; } } diff --git a/net/devlink/dev.c b/net/devlink/dev.c index 19dbf540748a..c609deb42e88 100644 --- a/net/devlink/dev.c +++ b/net/devlink/dev.c @@ -1202,17 +1202,13 @@ static void __devlink_compat_running_version(struct devlink *devlink, if (err) goto free_msg; - nla_for_each_attr(nlattr, (void *)msg->data, msg->len, rem) { + nla_for_each_attr_type(nlattr, DEVLINK_ATTR_INFO_VERSION_RUNNING, + (void *)msg->data, msg->len, rem) { const struct nlattr *kv; int rem_kv; - if (nla_type(nlattr) != DEVLINK_ATTR_INFO_VERSION_RUNNING) - continue; - - nla_for_each_nested(kv, nlattr, rem_kv) { - if (nla_type(kv) != DEVLINK_ATTR_INFO_VERSION_VALUE) - continue; - + nla_for_each_nested_type(kv, DEVLINK_ATTR_INFO_VERSION_VALUE, + nlattr, rem_kv) { strlcat(buf, nla_data(kv), len); strlcat(buf, " ", len); } diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 225353fbb3f1..51d4013b6121 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -215,10 +215,8 @@ static int mqprio_parse_tc_entries(struct Qdisc *sch, struct nlattr *nlattr_opt, for (tc = 0; tc < TC_QOPT_MAX_QUEUE; tc++) fp[tc] = priv->fp[tc]; - nla_for_each_attr(n, nlattr_opt, nlattr_opt_len, rem) { - if (nla_type(n) != TCA_MQPRIO_TC_ENTRY) - continue; - + nla_for_each_attr_type(n, TCA_MQPRIO_TC_ENTRY, nlattr_opt, + nlattr_opt_len, rem) { err = mqprio_parse_tc_entry(fp, n, &seen_tcs, extack); if (err) goto out; diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index a0d54b422186..1ab17e8a7260 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -1752,10 +1752,7 @@ static int taprio_parse_tc_entries(struct Qdisc *sch, fp[tc] = q->fp[tc]; } - nla_for_each_nested(n, opt, rem) { - if (nla_type(n) != TCA_TAPRIO_ATTR_TC_ENTRY) - continue; - + nla_for_each_nested_type(n, TCA_TAPRIO_ATTR_TC_ENTRY, opt, rem) { err = taprio_parse_tc_entry(sch, n, max_sdu, fp, &seen_tcs, extack); if (err) -- cgit v1.2.3 From 40dc8ab605894acae1473e434944924a22cfaaa0 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 30 Mar 2024 15:54:38 +0000 Subject: batman-adv: bypass empty buckets in batadv_purge_orig_ref() Many syzbot reports are pointing to soft lockups in batadv_purge_orig_ref() [1] Root cause is unknown, but we can avoid spending too much time there and perhaps get more interesting reports. [1] watchdog: BUG: soft lockup - CPU#0 stuck for 27s! [kworker/u4:6:621] Modules linked in: irq event stamp: 6182794 hardirqs last enabled at (6182793): [] __local_bh_enable_ip+0x224/0x44c kernel/softirq.c:386 hardirqs last disabled at (6182794): [] __el1_irq arch/arm64/kernel/entry-common.c:533 [inline] hardirqs last disabled at (6182794): [] el1_interrupt+0x24/0x68 arch/arm64/kernel/entry-common.c:551 softirqs last enabled at (6182792): [] spin_unlock_bh include/linux/spinlock.h:396 [inline] softirqs last enabled at (6182792): [] batadv_purge_orig_ref+0x114c/0x1228 net/batman-adv/originator.c:1287 softirqs last disabled at (6182790): [] spin_lock_bh include/linux/spinlock.h:356 [inline] softirqs last disabled at (6182790): [] batadv_purge_orig_ref+0x164/0x1228 net/batman-adv/originator.c:1271 CPU: 0 PID: 621 Comm: kworker/u4:6 Not tainted 6.8.0-rc7-syzkaller-g707081b61156 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/29/2024 Workqueue: bat_events batadv_purge_orig pstate: 80400005 (Nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : should_resched arch/arm64/include/asm/preempt.h:79 [inline] pc : __local_bh_enable_ip+0x228/0x44c kernel/softirq.c:388 lr : __local_bh_enable_ip+0x224/0x44c kernel/softirq.c:386 sp : ffff800099007970 x29: ffff800099007980 x28: 1fffe00018fce1bd x27: dfff800000000000 x26: ffff0000d2620008 x25: ffff0000c7e70de8 x24: 0000000000000001 x23: 1fffe00018e57781 x22: dfff800000000000 x21: ffff80008aab71c4 x20: ffff0001b40136c0 x19: ffff0000c72bbc08 x18: 1fffe0001a817bb0 x17: ffff800125414000 x16: ffff80008032116c x15: 0000000000000001 x14: 1fffe0001ee9d610 x13: 0000000000000000 x12: 0000000000000003 x11: 0000000000000000 x10: 0000000000ff0100 x9 : 0000000000000000 x8 : 00000000005e5789 x7 : ffff80008aab61dc x6 : 0000000000000000 x5 : 0000000000000000 x4 : 0000000000000001 x3 : 0000000000000000 x2 : 0000000000000006 x1 : 0000000000000080 x0 : ffff800125414000 Call trace: __daif_local_irq_enable arch/arm64/include/asm/irqflags.h:27 [inline] arch_local_irq_enable arch/arm64/include/asm/irqflags.h:49 [inline] __local_bh_enable_ip+0x228/0x44c kernel/softirq.c:386 __raw_spin_unlock_bh include/linux/spinlock_api_smp.h:167 [inline] _raw_spin_unlock_bh+0x3c/0x4c kernel/locking/spinlock.c:210 spin_unlock_bh include/linux/spinlock.h:396 [inline] batadv_purge_orig_ref+0x114c/0x1228 net/batman-adv/originator.c:1287 batadv_purge_orig+0x20/0x70 net/batman-adv/originator.c:1300 process_one_work+0x694/0x1204 kernel/workqueue.c:2633 process_scheduled_works kernel/workqueue.c:2706 [inline] worker_thread+0x938/0xef4 kernel/workqueue.c:2787 kthread+0x288/0x310 kernel/kthread.c:388 ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:860 Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 0 Comm: swapper/1 Not tainted 6.8.0-rc7-syzkaller-g707081b61156 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/29/2024 pstate: 80400005 (Nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : arch_local_irq_enable+0x8/0xc arch/arm64/include/asm/irqflags.h:51 lr : default_idle_call+0xf8/0x128 kernel/sched/idle.c:103 sp : ffff800093a17d30 x29: ffff800093a17d30 x28: dfff800000000000 x27: 1ffff00012742fb4 x26: ffff80008ec9d000 x25: 0000000000000000 x24: 0000000000000002 x23: 1ffff00011d93a74 x22: ffff80008ec9d3a0 x21: 0000000000000000 x20: ffff0000c19dbc00 x19: ffff8000802d0fd8 x18: 1fffe00036804396 x17: ffff80008ec9d000 x16: ffff8000802d089c x15: 0000000000000001 x14: 1fffe00036805f10 x13: 0000000000000000 x12: 0000000000000003 x11: 0000000000000001 x10: 0000000000000003 x9 : 0000000000000000 x8 : 00000000000ce8d1 x7 : ffff8000804609e4 x6 : 0000000000000000 x5 : 0000000000000001 x4 : 0000000000000001 x3 : ffff80008ad6aac0 x2 : 0000000000000000 x1 : ffff80008aedea60 x0 : ffff800125436000 Call trace: __daif_local_irq_enable arch/arm64/include/asm/irqflags.h:27 [inline] arch_local_irq_enable+0x8/0xc arch/arm64/include/asm/irqflags.h:49 cpuidle_idle_call kernel/sched/idle.c:170 [inline] do_idle+0x1f0/0x4e8 kernel/sched/idle.c:312 cpu_startup_entry+0x5c/0x74 kernel/sched/idle.c:410 secondary_start_kernel+0x198/0x1c0 arch/arm64/kernel/smp.c:272 __secondary_switched+0xb8/0xbc arch/arm64/kernel/head.S:404 Signed-off-by: Eric Dumazet Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/originator.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 71c143d4b6d0..ac74f6ead62d 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -1266,6 +1266,8 @@ void batadv_purge_orig_ref(struct batadv_priv *bat_priv) /* for all origins... */ for (i = 0; i < hash->size; i++) { head = &hash->table[i]; + if (hlist_empty(head)) + continue; list_lock = &hash->list_locks[i]; spin_lock_bh(list_lock); -- cgit v1.2.3 From 117aef12a7b1b797bce9f66b156c65eab850b5b5 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Wed, 27 Mar 2024 16:23:52 +0100 Subject: ip_tunnel: use a separate struct to store tunnel params in the kernel Unlike IPv6 tunnels which use purely-kernel __ip6_tnl_parm structure to store params inside the kernel, IPv4 tunnel code uses the same ip_tunnel_parm which is being used to talk with the userspace. This makes it difficult to alter or add any fields or use a different format for whatever data. Define struct ip_tunnel_parm_kern, a 1:1 copy of ip_tunnel_parm for now, and use it throughout the code. Define the pieces, where the copy user <-> kernel happens, as standalone functions, and copy the data there field-by-field, so that the kernel-side structure could be easily modified later on and the users wouldn't have to care about this. Reviewed-by: Simon Horman Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- .../net/ethernet/mellanox/mlxsw/spectrum_ipip.c | 30 ++++++----- .../net/ethernet/mellanox/mlxsw/spectrum_ipip.h | 2 +- .../net/ethernet/mellanox/mlxsw/spectrum_span.c | 4 +- include/linux/netdevice.h | 7 +-- include/net/ip_tunnels.h | 25 +++++++-- net/ipv4/ip_gre.c | 17 +++--- net/ipv4/ip_tunnel.c | 63 +++++++++++++++++----- net/ipv4/ip_tunnel_core.c | 2 +- net/ipv4/ip_vti.c | 12 ++--- net/ipv4/ipip.c | 12 ++--- net/ipv4/ipmr.c | 2 +- net/ipv6/addrconf.c | 3 +- net/ipv6/sit.c | 33 ++++++------ 13 files changed, 137 insertions(+), 75 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c index 3340b4a694c3..d67df358a52f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c @@ -8,7 +8,7 @@ #include "spectrum_ipip.h" #include "reg.h" -struct ip_tunnel_parm +struct ip_tunnel_parm_kern mlxsw_sp_ipip_netdev_parms4(const struct net_device *ol_dev) { struct ip_tunnel *tun = netdev_priv(ol_dev); @@ -24,7 +24,8 @@ mlxsw_sp_ipip_netdev_parms6(const struct net_device *ol_dev) return tun->parms; } -static bool mlxsw_sp_ipip_parms4_has_ikey(const struct ip_tunnel_parm *parms) +static bool +mlxsw_sp_ipip_parms4_has_ikey(const struct ip_tunnel_parm_kern *parms) { return !!(parms->i_flags & TUNNEL_KEY); } @@ -34,7 +35,8 @@ static bool mlxsw_sp_ipip_parms6_has_ikey(const struct __ip6_tnl_parm *parms) return !!(parms->i_flags & TUNNEL_KEY); } -static bool mlxsw_sp_ipip_parms4_has_okey(const struct ip_tunnel_parm *parms) +static bool +mlxsw_sp_ipip_parms4_has_okey(const struct ip_tunnel_parm_kern *parms) { return !!(parms->o_flags & TUNNEL_KEY); } @@ -44,7 +46,7 @@ static bool mlxsw_sp_ipip_parms6_has_okey(const struct __ip6_tnl_parm *parms) return !!(parms->o_flags & TUNNEL_KEY); } -static u32 mlxsw_sp_ipip_parms4_ikey(const struct ip_tunnel_parm *parms) +static u32 mlxsw_sp_ipip_parms4_ikey(const struct ip_tunnel_parm_kern *parms) { return mlxsw_sp_ipip_parms4_has_ikey(parms) ? be32_to_cpu(parms->i_key) : 0; @@ -56,7 +58,7 @@ static u32 mlxsw_sp_ipip_parms6_ikey(const struct __ip6_tnl_parm *parms) be32_to_cpu(parms->i_key) : 0; } -static u32 mlxsw_sp_ipip_parms4_okey(const struct ip_tunnel_parm *parms) +static u32 mlxsw_sp_ipip_parms4_okey(const struct ip_tunnel_parm_kern *parms) { return mlxsw_sp_ipip_parms4_has_okey(parms) ? be32_to_cpu(parms->o_key) : 0; @@ -69,7 +71,7 @@ static u32 mlxsw_sp_ipip_parms6_okey(const struct __ip6_tnl_parm *parms) } static union mlxsw_sp_l3addr -mlxsw_sp_ipip_parms4_saddr(const struct ip_tunnel_parm *parms) +mlxsw_sp_ipip_parms4_saddr(const struct ip_tunnel_parm_kern *parms) { return (union mlxsw_sp_l3addr) { .addr4 = parms->iph.saddr }; } @@ -81,7 +83,7 @@ mlxsw_sp_ipip_parms6_saddr(const struct __ip6_tnl_parm *parms) } static union mlxsw_sp_l3addr -mlxsw_sp_ipip_parms4_daddr(const struct ip_tunnel_parm *parms) +mlxsw_sp_ipip_parms4_daddr(const struct ip_tunnel_parm_kern *parms) { return (union mlxsw_sp_l3addr) { .addr4 = parms->iph.daddr }; } @@ -96,7 +98,7 @@ union mlxsw_sp_l3addr mlxsw_sp_ipip_netdev_saddr(enum mlxsw_sp_l3proto proto, const struct net_device *ol_dev) { - struct ip_tunnel_parm parms4; + struct ip_tunnel_parm_kern parms4; struct __ip6_tnl_parm parms6; switch (proto) { @@ -115,7 +117,9 @@ mlxsw_sp_ipip_netdev_saddr(enum mlxsw_sp_l3proto proto, static __be32 mlxsw_sp_ipip_netdev_daddr4(const struct net_device *ol_dev) { - struct ip_tunnel_parm parms4 = mlxsw_sp_ipip_netdev_parms4(ol_dev); + struct ip_tunnel_parm_kern parms4; + + parms4 = mlxsw_sp_ipip_netdev_parms4(ol_dev); return mlxsw_sp_ipip_parms4_daddr(&parms4).addr4; } @@ -124,7 +128,7 @@ static union mlxsw_sp_l3addr mlxsw_sp_ipip_netdev_daddr(enum mlxsw_sp_l3proto proto, const struct net_device *ol_dev) { - struct ip_tunnel_parm parms4; + struct ip_tunnel_parm_kern parms4; struct __ip6_tnl_parm parms6; switch (proto) { @@ -150,7 +154,7 @@ bool mlxsw_sp_l3addr_is_zero(union mlxsw_sp_l3addr addr) static struct mlxsw_sp_ipip_parms mlxsw_sp_ipip_netdev_parms_init_gre4(const struct net_device *ol_dev) { - struct ip_tunnel_parm parms = mlxsw_sp_ipip_netdev_parms4(ol_dev); + struct ip_tunnel_parm_kern parms = mlxsw_sp_ipip_netdev_parms4(ol_dev); return (struct mlxsw_sp_ipip_parms) { .proto = MLXSW_SP_L3_PROTO_IPV4, @@ -187,8 +191,8 @@ mlxsw_sp_ipip_decap_config_gre4(struct mlxsw_sp *mlxsw_sp, { u16 rif_index = mlxsw_sp_ipip_lb_rif_index(ipip_entry->ol_lb); u16 ul_rif_id = mlxsw_sp_ipip_lb_ul_rif_id(ipip_entry->ol_lb); + struct ip_tunnel_parm_kern parms; char rtdp_pl[MLXSW_REG_RTDP_LEN]; - struct ip_tunnel_parm parms; unsigned int type_check; bool has_ikey; u32 daddr4; @@ -252,7 +256,7 @@ static struct mlxsw_sp_rif_ipip_lb_config mlxsw_sp_ipip_ol_loopback_config_gre4(struct mlxsw_sp *mlxsw_sp, const struct net_device *ol_dev) { - struct ip_tunnel_parm parms = mlxsw_sp_ipip_netdev_parms4(ol_dev); + struct ip_tunnel_parm_kern parms = mlxsw_sp_ipip_netdev_parms4(ol_dev); enum mlxsw_reg_ritr_loopback_ipip_type lb_ipipt; lb_ipipt = mlxsw_sp_ipip_parms4_has_okey(&parms) ? diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.h index a35f009da561..a66173779641 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.h @@ -9,7 +9,7 @@ #include #include -struct ip_tunnel_parm +struct ip_tunnel_parm_kern mlxsw_sp_ipip_netdev_parms4(const struct net_device *ol_dev); struct __ip6_tnl_parm mlxsw_sp_ipip_netdev_parms6(const struct net_device *ol_dev); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index af50ff9e5f26..f0cbc6b9b041 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -413,8 +413,8 @@ mlxsw_sp_span_gretap4_route(const struct net_device *to_dev, __be32 *saddrp, __be32 *daddrp) { struct ip_tunnel *tun = netdev_priv(to_dev); + struct ip_tunnel_parm_kern parms; struct net_device *dev = NULL; - struct ip_tunnel_parm parms; struct rtable *rt = NULL; struct flowi4 fl4; @@ -451,7 +451,7 @@ mlxsw_sp_span_entry_gretap4_parms(struct mlxsw_sp *mlxsw_sp, const struct net_device *to_dev, struct mlxsw_sp_span_parms *sparmsp) { - struct ip_tunnel_parm tparm = mlxsw_sp_ipip_netdev_parms4(to_dev); + struct ip_tunnel_parm_kern tparm = mlxsw_sp_ipip_netdev_parms4(to_dev); union mlxsw_sp_l3addr saddr = { .addr4 = tparm.iph.saddr }; union mlxsw_sp_l3addr daddr = { .addr4 = tparm.iph.daddr }; bool inherit_tos = tparm.iph.tos & 0x1; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e41d30ebaca6..55c7cf9404a4 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -59,7 +59,7 @@ struct ethtool_ops; struct kernel_hwtstamp_config; struct phy_device; struct dsa_port; -struct ip_tunnel_parm; +struct ip_tunnel_parm_kern; struct macsec_context; struct macsec_ops; struct netdev_name_node; @@ -1327,7 +1327,7 @@ struct netdev_net_notifier { * queue id bound to an AF_XDP socket. The flags field specifies if * only RX, only Tx, or both should be woken up using the flags * XDP_WAKEUP_RX and XDP_WAKEUP_TX. - * int (*ndo_tunnel_ctl)(struct net_device *dev, struct ip_tunnel_parm *p, + * int (*ndo_tunnel_ctl)(struct net_device *dev, struct ip_tunnel_parm_kern *p, * int cmd); * Add, change, delete or get information on an IPv4 tunnel. * struct net_device *(*ndo_get_peer_dev)(struct net_device *dev); @@ -1583,7 +1583,8 @@ struct net_device_ops { int (*ndo_xsk_wakeup)(struct net_device *dev, u32 queue_id, u32 flags); int (*ndo_tunnel_ctl)(struct net_device *dev, - struct ip_tunnel_parm *p, int cmd); + struct ip_tunnel_parm_kern *p, + int cmd); struct net_device * (*ndo_get_peer_dev)(struct net_device *dev); int (*ndo_fill_forward_path)(struct net_device_path_ctx *ctx, struct net_device_path *path); diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 5cd64bb2104d..20f0319ab149 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -110,6 +110,17 @@ struct ip_tunnel_prl_entry { struct metadata_dst; +/* Kernel-side copy of ip_tunnel_parm */ +struct ip_tunnel_parm_kern { + char name[IFNAMSIZ]; + int link; + __be16 i_flags; + __be16 o_flags; + __be32 i_key; + __be32 o_key; + struct iphdr iph; +}; + struct ip_tunnel { struct ip_tunnel __rcu *next; struct hlist_node hash_node; @@ -136,7 +147,7 @@ struct ip_tunnel { struct dst_cache dst_cache; - struct ip_tunnel_parm parms; + struct ip_tunnel_parm_kern parms; int mlink; int encap_hlen; /* Encap header length (FOU,GUE) */ @@ -291,7 +302,11 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, const struct iphdr *tnl_params, const u8 protocol); void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, const u8 proto, int tunnel_hlen); -int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd); +int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, + int cmd); +bool ip_tunnel_parm_from_user(struct ip_tunnel_parm_kern *kp, + const void __user *data); +bool ip_tunnel_parm_to_user(void __user *data, struct ip_tunnel_parm_kern *kp); int ip_tunnel_siocdevprivate(struct net_device *dev, struct ifreq *ifr, void __user *data, int cmd); int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict); @@ -307,16 +322,16 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, const struct tnl_ptk_info *tpi, struct metadata_dst *tun_dst, bool log_ecn_error); int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[], - struct ip_tunnel_parm *p, __u32 fwmark); + struct ip_tunnel_parm_kern *p, __u32 fwmark); int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[], - struct ip_tunnel_parm *p, __u32 fwmark); + struct ip_tunnel_parm_kern *p, __u32 fwmark); void ip_tunnel_setup(struct net_device *dev, unsigned int net_id); bool ip_tunnel_netlink_encap_parms(struct nlattr *data[], struct ip_tunnel_encap *encap); void ip_tunnel_netlink_parms(struct nlattr *data[], - struct ip_tunnel_parm *parms); + struct ip_tunnel_parm_kern *parms); extern const struct header_ops ip_tunnel_header_ops; __be16 ip_tunnel_parse_protocol(const struct sk_buff *skb); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 7b16c211b904..8853c065a985 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -785,7 +785,8 @@ static void ipgre_link_update(struct net_device *dev, bool set_mtu) } } -static int ipgre_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, +static int ipgre_tunnel_ctl(struct net_device *dev, + struct ip_tunnel_parm_kern *p, int cmd) { int err; @@ -1131,7 +1132,7 @@ static int erspan_validate(struct nlattr *tb[], struct nlattr *data[], static int ipgre_netlink_parms(struct net_device *dev, struct nlattr *data[], struct nlattr *tb[], - struct ip_tunnel_parm *parms, + struct ip_tunnel_parm_kern *parms, __u32 *fwmark) { struct ip_tunnel *t = netdev_priv(dev); @@ -1198,7 +1199,7 @@ static int ipgre_netlink_parms(struct net_device *dev, static int erspan_netlink_parms(struct net_device *dev, struct nlattr *data[], struct nlattr *tb[], - struct ip_tunnel_parm *parms, + struct ip_tunnel_parm_kern *parms, __u32 *fwmark) { struct ip_tunnel *t = netdev_priv(dev); @@ -1357,7 +1358,7 @@ static int ipgre_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { - struct ip_tunnel_parm p; + struct ip_tunnel_parm_kern p; __u32 fwmark = 0; int err; @@ -1375,7 +1376,7 @@ static int erspan_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { - struct ip_tunnel_parm p; + struct ip_tunnel_parm_kern p; __u32 fwmark = 0; int err; @@ -1394,8 +1395,8 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); + struct ip_tunnel_parm_kern p; __u32 fwmark = t->fwmark; - struct ip_tunnel_parm p; int err; err = ipgre_newlink_encap_setup(dev, data); @@ -1423,8 +1424,8 @@ static int erspan_changelink(struct net_device *dev, struct nlattr *tb[], struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); + struct ip_tunnel_parm_kern p; __u32 fwmark = t->fwmark; - struct ip_tunnel_parm p; int err; err = ipgre_newlink_encap_setup(dev, data); @@ -1496,7 +1497,7 @@ static size_t ipgre_get_size(const struct net_device *dev) static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip_tunnel *t = netdev_priv(dev); - struct ip_tunnel_parm *p = &t->parms; + struct ip_tunnel_parm_kern *p = &t->parms; __be16 o_flags = p->o_flags; if (nla_put_u32(skb, IFLA_GRE_LINK, p->link) || diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 1b8d8ff9a237..d10f41f81463 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -56,7 +56,7 @@ static unsigned int ip_tunnel_hash(__be32 key, __be32 remote) IP_TNL_HASH_BITS); } -static bool ip_tunnel_key_match(const struct ip_tunnel_parm *p, +static bool ip_tunnel_key_match(const struct ip_tunnel_parm_kern *p, __be16 flags, __be32 key) { if (p->i_flags & TUNNEL_KEY) { @@ -171,7 +171,7 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, EXPORT_SYMBOL_GPL(ip_tunnel_lookup); static struct hlist_head *ip_bucket(struct ip_tunnel_net *itn, - struct ip_tunnel_parm *parms) + struct ip_tunnel_parm_kern *parms) { unsigned int h; __be32 remote; @@ -206,7 +206,7 @@ static void ip_tunnel_del(struct ip_tunnel_net *itn, struct ip_tunnel *t) } static struct ip_tunnel *ip_tunnel_find(struct ip_tunnel_net *itn, - struct ip_tunnel_parm *parms, + struct ip_tunnel_parm_kern *parms, int type) { __be32 remote = parms->iph.daddr; @@ -230,7 +230,7 @@ static struct ip_tunnel *ip_tunnel_find(struct ip_tunnel_net *itn, static struct net_device *__ip_tunnel_create(struct net *net, const struct rtnl_link_ops *ops, - struct ip_tunnel_parm *parms) + struct ip_tunnel_parm_kern *parms) { int err; struct ip_tunnel *tunnel; @@ -326,7 +326,7 @@ static int ip_tunnel_bind_dev(struct net_device *dev) static struct ip_tunnel *ip_tunnel_create(struct net *net, struct ip_tunnel_net *itn, - struct ip_tunnel_parm *parms) + struct ip_tunnel_parm_kern *parms) { struct ip_tunnel *nt; struct net_device *dev; @@ -871,7 +871,7 @@ EXPORT_SYMBOL_GPL(ip_tunnel_xmit); static void ip_tunnel_update(struct ip_tunnel_net *itn, struct ip_tunnel *t, struct net_device *dev, - struct ip_tunnel_parm *p, + struct ip_tunnel_parm_kern *p, bool set_mtu, __u32 fwmark) { @@ -903,7 +903,8 @@ static void ip_tunnel_update(struct ip_tunnel_net *itn, netdev_state_change(dev); } -int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) +int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, + int cmd) { int err = 0; struct ip_tunnel *t = netdev_priv(dev); @@ -1005,16 +1006,52 @@ done: } EXPORT_SYMBOL_GPL(ip_tunnel_ctl); +bool ip_tunnel_parm_from_user(struct ip_tunnel_parm_kern *kp, + const void __user *data) +{ + struct ip_tunnel_parm p; + + if (copy_from_user(&p, data, sizeof(p))) + return false; + + strscpy(kp->name, p.name); + kp->link = p.link; + kp->i_flags = p.i_flags; + kp->o_flags = p.o_flags; + kp->i_key = p.i_key; + kp->o_key = p.o_key; + memcpy(&kp->iph, &p.iph, min(sizeof(kp->iph), sizeof(p.iph))); + + return true; +} +EXPORT_SYMBOL_GPL(ip_tunnel_parm_from_user); + +bool ip_tunnel_parm_to_user(void __user *data, struct ip_tunnel_parm_kern *kp) +{ + struct ip_tunnel_parm p; + + strscpy(p.name, kp->name); + p.link = kp->link; + p.i_flags = kp->i_flags; + p.o_flags = kp->o_flags; + p.i_key = kp->i_key; + p.o_key = kp->o_key; + memcpy(&p.iph, &kp->iph, min(sizeof(p.iph), sizeof(kp->iph))); + + return !copy_to_user(data, &p, sizeof(p)); +} +EXPORT_SYMBOL_GPL(ip_tunnel_parm_to_user); + int ip_tunnel_siocdevprivate(struct net_device *dev, struct ifreq *ifr, void __user *data, int cmd) { - struct ip_tunnel_parm p; + struct ip_tunnel_parm_kern p; int err; - if (copy_from_user(&p, data, sizeof(p))) + if (!ip_tunnel_parm_from_user(&p, data)) return -EFAULT; err = dev->netdev_ops->ndo_tunnel_ctl(dev, &p, cmd); - if (!err && copy_to_user(data, &p, sizeof(p))) + if (!err && !ip_tunnel_parm_to_user(data, &p)) return -EFAULT; return err; } @@ -1093,7 +1130,7 @@ int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id, struct rtnl_link_ops *ops, char *devname) { struct ip_tunnel_net *itn = net_generic(net, ip_tnl_net_id); - struct ip_tunnel_parm parms; + struct ip_tunnel_parm_kern parms; unsigned int i; itn->rtnl_link_ops = ops; @@ -1171,7 +1208,7 @@ void ip_tunnel_delete_nets(struct list_head *net_list, unsigned int id, EXPORT_SYMBOL_GPL(ip_tunnel_delete_nets); int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[], - struct ip_tunnel_parm *p, __u32 fwmark) + struct ip_tunnel_parm_kern *p, __u32 fwmark) { struct ip_tunnel *nt; struct net *net = dev_net(dev); @@ -1225,7 +1262,7 @@ err_register_netdevice: EXPORT_SYMBOL_GPL(ip_tunnel_newlink); int ip_tunnel_changelink(struct net_device *dev, struct nlattr *tb[], - struct ip_tunnel_parm *p, __u32 fwmark) + struct ip_tunnel_parm_kern *p, __u32 fwmark) { struct ip_tunnel *t; struct ip_tunnel *tunnel = netdev_priv(dev); diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 80ccd6661aa3..98f136b07191 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -1116,7 +1116,7 @@ bool ip_tunnel_netlink_encap_parms(struct nlattr *data[], EXPORT_SYMBOL_GPL(ip_tunnel_netlink_encap_parms); void ip_tunnel_netlink_parms(struct nlattr *data[], - struct ip_tunnel_parm *parms) + struct ip_tunnel_parm_kern *parms) { if (data[IFLA_IPTUN_LINK]) parms->link = nla_get_u32(data[IFLA_IPTUN_LINK]); diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index ee587adb169f..7e595f619615 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -167,7 +167,7 @@ static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev, struct flowi *fl) { struct ip_tunnel *tunnel = netdev_priv(dev); - struct ip_tunnel_parm *parms = &tunnel->parms; + struct ip_tunnel_parm_kern *parms = &tunnel->parms; struct dst_entry *dst = skb_dst(skb); struct net_device *tdev; /* Device to other host */ int pkt_len = skb->len; @@ -373,7 +373,7 @@ static int vti4_err(struct sk_buff *skb, u32 info) } static int -vti_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) +vti_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) { int err = 0; @@ -531,7 +531,7 @@ static int vti_tunnel_validate(struct nlattr *tb[], struct nlattr *data[], } static void vti_netlink_parms(struct nlattr *data[], - struct ip_tunnel_parm *parms, + struct ip_tunnel_parm_kern *parms, __u32 *fwmark) { memset(parms, 0, sizeof(*parms)); @@ -566,7 +566,7 @@ static int vti_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { - struct ip_tunnel_parm parms; + struct ip_tunnel_parm_kern parms; __u32 fwmark = 0; vti_netlink_parms(data, &parms, &fwmark); @@ -578,8 +578,8 @@ static int vti_changelink(struct net_device *dev, struct nlattr *tb[], struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); + struct ip_tunnel_parm_kern p; __u32 fwmark = t->fwmark; - struct ip_tunnel_parm p; vti_netlink_parms(data, &p, &fwmark); return ip_tunnel_changelink(dev, tb, &p, fwmark); @@ -606,7 +606,7 @@ static size_t vti_get_size(const struct net_device *dev) static int vti_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip_tunnel *t = netdev_priv(dev); - struct ip_tunnel_parm *p = &t->parms; + struct ip_tunnel_parm_kern *p = &t->parms; if (nla_put_u32(skb, IFLA_VTI_LINK, p->link) || nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key) || diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index f2696eaadbe6..db960cd08a3b 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -330,7 +330,7 @@ static bool ipip_tunnel_ioctl_verify_protocol(u8 ipproto) } static int -ipip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) +ipip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) { if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { if (p->iph.version != 4 || @@ -405,8 +405,8 @@ static int ipip_tunnel_validate(struct nlattr *tb[], struct nlattr *data[], } static void ipip_netlink_parms(struct nlattr *data[], - struct ip_tunnel_parm *parms, bool *collect_md, - __u32 *fwmark) + struct ip_tunnel_parm_kern *parms, + bool *collect_md, __u32 *fwmark) { memset(parms, 0, sizeof(*parms)); @@ -432,8 +432,8 @@ static int ipip_newlink(struct net *src_net, struct net_device *dev, struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); - struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + struct ip_tunnel_parm_kern p; __u32 fwmark = 0; if (ip_tunnel_netlink_encap_parms(data, &ipencap)) { @@ -452,8 +452,8 @@ static int ipip_changelink(struct net_device *dev, struct nlattr *tb[], struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); - struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + struct ip_tunnel_parm_kern p; bool collect_md; __u32 fwmark = t->fwmark; @@ -510,7 +510,7 @@ static size_t ipip_get_size(const struct net_device *dev) static int ipip_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - struct ip_tunnel_parm *parm = &tunnel->parms; + struct ip_tunnel_parm_kern *parm = &tunnel->parms; if (nla_put_u32(skb, IFLA_IPTUN_LINK, parm->link) || nla_put_in_addr(skb, IFLA_IPTUN_LOCAL, parm->iph.saddr) || diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index fd5c01c8489f..6c750bd13dd8 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -441,7 +441,7 @@ static bool ipmr_init_vif_indev(const struct net_device *dev) static struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) { struct net_device *tunnel_dev, *new_dev; - struct ip_tunnel_parm p = { }; + struct ip_tunnel_parm_kern p = { }; int err; tunnel_dev = __dev_get_by_name(net, "tunl0"); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 92db9b474f2b..82ad6b8fa007 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -63,6 +63,7 @@ #include #include +#include #include #include #include @@ -2917,7 +2918,7 @@ put: static int addrconf_set_sit_dstaddr(struct net *net, struct net_device *dev, struct in6_ifreq *ireq) { - struct ip_tunnel_parm p = { }; + struct ip_tunnel_parm_kern p = { }; int err; if (!(ipv6_addr_type(&ireq->ifr6_addr) & IPV6_ADDR_COMPATv4)) diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 655c9b1a19b8..894e6a4c4be3 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -132,8 +132,8 @@ static struct ip_tunnel *ipip6_tunnel_lookup(struct net *net, return NULL; } -static struct ip_tunnel __rcu **__ipip6_bucket(struct sit_net *sitn, - struct ip_tunnel_parm *parms) +static struct ip_tunnel __rcu ** +__ipip6_bucket(struct sit_net *sitn, struct ip_tunnel_parm_kern *parms) { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; @@ -226,7 +226,8 @@ out: } static struct ip_tunnel *ipip6_tunnel_locate(struct net *net, - struct ip_tunnel_parm *parms, int create) + struct ip_tunnel_parm_kern *parms, + int create) { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; @@ -1135,7 +1136,8 @@ static void ipip6_tunnel_bind_dev(struct net_device *dev) dev->needed_headroom = t_hlen + hlen; } -static void ipip6_tunnel_update(struct ip_tunnel *t, struct ip_tunnel_parm *p, +static void ipip6_tunnel_update(struct ip_tunnel *t, + struct ip_tunnel_parm_kern *p, __u32 fwmark) { struct net *net = t->net; @@ -1196,11 +1198,11 @@ static int ipip6_tunnel_get6rd(struct net_device *dev, struct ip_tunnel_parm __user *data) { struct ip_tunnel *t = netdev_priv(dev); + struct ip_tunnel_parm_kern p; struct ip_tunnel_6rd ip6rd; - struct ip_tunnel_parm p; if (dev == dev_to_sit_net(dev)->fb_tunnel_dev) { - if (copy_from_user(&p, data, sizeof(p))) + if (!ip_tunnel_parm_from_user(&p, data)) return -EFAULT; t = ipip6_tunnel_locate(t->net, &p, 0); } @@ -1251,7 +1253,7 @@ static bool ipip6_valid_ip_proto(u8 ipproto) } static int -__ipip6_tunnel_ioctl_validate(struct net *net, struct ip_tunnel_parm *p) +__ipip6_tunnel_ioctl_validate(struct net *net, struct ip_tunnel_parm_kern *p) { if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM; @@ -1268,7 +1270,7 @@ __ipip6_tunnel_ioctl_validate(struct net *net, struct ip_tunnel_parm *p) } static int -ipip6_tunnel_get(struct net_device *dev, struct ip_tunnel_parm *p) +ipip6_tunnel_get(struct net_device *dev, struct ip_tunnel_parm_kern *p) { struct ip_tunnel *t = netdev_priv(dev); @@ -1281,7 +1283,7 @@ ipip6_tunnel_get(struct net_device *dev, struct ip_tunnel_parm *p) } static int -ipip6_tunnel_add(struct net_device *dev, struct ip_tunnel_parm *p) +ipip6_tunnel_add(struct net_device *dev, struct ip_tunnel_parm_kern *p) { struct ip_tunnel *t = netdev_priv(dev); int err; @@ -1297,7 +1299,7 @@ ipip6_tunnel_add(struct net_device *dev, struct ip_tunnel_parm *p) } static int -ipip6_tunnel_change(struct net_device *dev, struct ip_tunnel_parm *p) +ipip6_tunnel_change(struct net_device *dev, struct ip_tunnel_parm_kern *p) { struct ip_tunnel *t = netdev_priv(dev); int err; @@ -1328,7 +1330,7 @@ ipip6_tunnel_change(struct net_device *dev, struct ip_tunnel_parm *p) } static int -ipip6_tunnel_del(struct net_device *dev, struct ip_tunnel_parm *p) +ipip6_tunnel_del(struct net_device *dev, struct ip_tunnel_parm_kern *p) { struct ip_tunnel *t = netdev_priv(dev); @@ -1348,7 +1350,8 @@ ipip6_tunnel_del(struct net_device *dev, struct ip_tunnel_parm *p) } static int -ipip6_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) +ipip6_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, + int cmd) { switch (cmd) { case SIOCGETTUNNEL: @@ -1490,7 +1493,7 @@ static int ipip6_validate(struct nlattr *tb[], struct nlattr *data[], } static void ipip6_netlink_parms(struct nlattr *data[], - struct ip_tunnel_parm *parms, + struct ip_tunnel_parm_kern *parms, __u32 *fwmark) { memset(parms, 0, sizeof(*parms)); @@ -1599,8 +1602,8 @@ static int ipip6_changelink(struct net_device *dev, struct nlattr *tb[], struct netlink_ext_ack *extack) { struct ip_tunnel *t = netdev_priv(dev); - struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + struct ip_tunnel_parm_kern p; struct net *net = t->net; struct sit_net *sitn = net_generic(net, sit_net_id); #ifdef CONFIG_IPV6_SIT_6RD @@ -1687,7 +1690,7 @@ static size_t ipip6_get_size(const struct net_device *dev) static int ipip6_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - struct ip_tunnel_parm *parm = &tunnel->parms; + struct ip_tunnel_parm_kern *parm = &tunnel->parms; if (nla_put_u32(skb, IFLA_IPTUN_LINK, parm->link) || nla_put_in_addr(skb, IFLA_IPTUN_LOCAL, parm->iph.saddr) || -- cgit v1.2.3 From 5832c4a77d6931cebf9ba737129ae8f14b66ee1d Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Wed, 27 Mar 2024 16:23:53 +0100 Subject: ip_tunnel: convert __be16 tunnel flags to bitmaps Historically, tunnel flags like TUNNEL_CSUM or TUNNEL_ERSPAN_OPT have been defined as __be16. Now all of those 16 bits are occupied and there's no more free space for new flags. It can't be simply switched to a bigger container with no adjustments to the values, since it's an explicit Endian storage, and on LE systems (__be16)0x0001 equals to (__be64)0x0001000000000000. We could probably define new 64-bit flags depending on the Endianness, i.e. (__be64)0x0001 on BE and (__be64)0x00010000... on LE, but that would introduce an Endianness dependency and spawn a ton of Sparse warnings. To mitigate them, all of those places which were adjusted with this change would be touched anyway, so why not define stuff properly if there's no choice. Define IP_TUNNEL_*_BIT counterparts as a bit number instead of the value already coded and a fistful of <16 <-> bitmap> converters and helpers. The two flags which have a different bit position are SIT_ISATAP_BIT and VTI_ISVTI_BIT, as they were defined not as __cpu_to_be16(), but as (__force __be16), i.e. had different positions on LE and BE. Now they both have strongly defined places. Change all __be16 fields which were used to store those flags, to IP_TUNNEL_DECLARE_FLAGS() -> DECLARE_BITMAP(__IP_TUNNEL_FLAG_NUM) -> unsigned long[1] for now, and replace all TUNNEL_* occurrences to their bitmap counterparts. Use the converters in the places which talk to the userspace, hardware (NFP) or other hosts (GRE header). The rest must explicitly use the new flags only. This must be done at once, otherwise there will be too many conversions throughout the code in the intermediate commits. Finally, disable the old __be16 flags for use in the kernel code (except for the two 'irregular' flags mentioned above), to prevent any accidental (mis)use of them. For the userspace, nothing is changed, only additions were made. Most noticeable bloat-o-meter difference (.text): vmlinux: 307/-1 (306) gre.ko: 62/0 (62) ip_gre.ko: 941/-217 (724) [*] ip_tunnel.ko: 390/-900 (-510) [**] ip_vti.ko: 138/0 (138) ip6_gre.ko: 534/-18 (516) [*] ip6_tunnel.ko: 118/-10 (108) [*] gre_flags_to_tnl_flags() grew, but still is inlined [**] ip_tunnel_find() got uninlined, hence such decrease The average code size increase in non-extreme case is 100-200 bytes per module, mostly due to sizeof(long) > sizeof(__be16), as %__IP_TUNNEL_FLAG_NUM is less than %BITS_PER_LONG and the compilers are able to expand the majority of bitmap_*() calls here into direct operations on scalars. Reviewed-by: Simon Horman Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- drivers/net/bareudp.c | 19 ++- .../net/ethernet/mellanox/mlx5/core/en/tc_tun.h | 2 +- .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c | 6 +- .../ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c | 12 +- .../ethernet/mellanox/mlx5/core/en/tc_tun_gre.c | 8 +- .../ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c | 9 +- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 16 ++- .../net/ethernet/mellanox/mlxsw/spectrum_ipip.c | 26 +++-- .../net/ethernet/mellanox/mlxsw/spectrum_span.c | 6 +- drivers/net/ethernet/netronome/nfp/flower/action.c | 27 ++++- drivers/net/geneve.c | 44 ++++--- drivers/net/vxlan/vxlan_core.c | 14 ++- include/net/dst_metadata.h | 10 +- include/net/flow_dissector.h | 2 +- include/net/gre.h | 70 ++++++------ include/net/ip6_tunnel.h | 4 +- include/net/ip_tunnels.h | 119 ++++++++++++++++--- include/net/udp_tunnel.h | 4 +- include/uapi/linux/if_tunnel.h | 33 ++++++ net/bridge/br_vlan_tunnel.c | 9 +- net/core/filter.c | 26 +++-- net/core/flow_dissector.c | 20 +++- net/ipv4/fou_bpf.c | 2 +- net/ipv4/gre_demux.c | 2 +- net/ipv4/ip_gre.c | 127 +++++++++++++-------- net/ipv4/ip_tunnel.c | 54 +++++---- net/ipv4/ip_tunnel_core.c | 80 ++++++++----- net/ipv4/ip_vti.c | 29 +++-- net/ipv4/ipip.c | 21 +++- net/ipv4/udp_tunnel_core.c | 5 +- net/ipv6/ip6_gre.c | 85 ++++++++------ net/ipv6/ip6_tunnel.c | 14 +-- net/ipv6/sit.c | 5 +- net/netfilter/ipvs/ip_vs_core.c | 6 +- net/netfilter/ipvs/ip_vs_xmit.c | 20 ++-- net/netfilter/nft_tunnel.c | 44 ++++--- net/openvswitch/flow_netlink.c | 61 ++++++---- net/psample/psample.c | 26 +++-- net/sched/act_tunnel_key.c | 36 +++--- net/sched/cls_flower.c | 27 ++--- 40 files changed, 715 insertions(+), 415 deletions(-) (limited to 'net') diff --git a/drivers/net/bareudp.c b/drivers/net/bareudp.c index 339db6e4a1d5..d5c56ca91b77 100644 --- a/drivers/net/bareudp.c +++ b/drivers/net/bareudp.c @@ -61,6 +61,7 @@ struct bareudp_dev { static int bareudp_udp_encap_recv(struct sock *sk, struct sk_buff *skb) { struct metadata_dst *tun_dst = NULL; + IP_TUNNEL_DECLARE_FLAGS(key) = { }; struct bareudp_dev *bareudp; unsigned short family; unsigned int len; @@ -137,7 +138,10 @@ static int bareudp_udp_encap_recv(struct sock *sk, struct sk_buff *skb) bareudp->dev->stats.rx_dropped++; goto drop; } - tun_dst = udp_tun_rx_dst(skb, family, TUNNEL_KEY, 0, 0); + + __set_bit(IP_TUNNEL_KEY_BIT, key); + + tun_dst = udp_tun_rx_dst(skb, family, key, 0, 0); if (!tun_dst) { bareudp->dev->stats.rx_dropped++; goto drop; @@ -285,10 +289,10 @@ static int bareudp_xmit_skb(struct sk_buff *skb, struct net_device *dev, struct bareudp_dev *bareudp, const struct ip_tunnel_info *info) { + bool udp_sum = test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); bool xnet = !net_eq(bareudp->net, dev_net(bareudp->dev)); bool use_cache = ip_tunnel_dst_cache_usable(skb, info); struct socket *sock = rcu_dereference(bareudp->sock); - bool udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM); const struct ip_tunnel_key *key = &info->key; struct rtable *rt; __be16 sport, df; @@ -316,7 +320,8 @@ static int bareudp_xmit_skb(struct sk_buff *skb, struct net_device *dev, tos = ip_tunnel_ecn_encap(key->tos, ip_hdr(skb), skb); ttl = key->ttl; - df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; + df = test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, key->tun_flags) ? + htons(IP_DF) : 0; skb_scrub_packet(skb, xnet); err = -ENOSPC; @@ -338,7 +343,8 @@ static int bareudp_xmit_skb(struct sk_buff *skb, struct net_device *dev, udp_tunnel_xmit_skb(rt, sock->sk, skb, saddr, info->key.u.ipv4.dst, tos, ttl, df, sport, bareudp->port, !net_eq(bareudp->net, dev_net(bareudp->dev)), - !(info->key.tun_flags & TUNNEL_CSUM)); + !test_bit(IP_TUNNEL_CSUM_BIT, + info->key.tun_flags)); return 0; free_dst: @@ -350,10 +356,10 @@ static int bareudp6_xmit_skb(struct sk_buff *skb, struct net_device *dev, struct bareudp_dev *bareudp, const struct ip_tunnel_info *info) { + bool udp_sum = test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); bool xnet = !net_eq(bareudp->net, dev_net(bareudp->dev)); bool use_cache = ip_tunnel_dst_cache_usable(skb, info); struct socket *sock = rcu_dereference(bareudp->sock); - bool udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM); const struct ip_tunnel_key *key = &info->key; struct dst_entry *dst = NULL; struct in6_addr saddr, daddr; @@ -402,7 +408,8 @@ static int bareudp6_xmit_skb(struct sk_buff *skb, struct net_device *dev, udp_tunnel6_xmit_skb(dst, sock->sk, skb, dev, &saddr, &daddr, prio, ttl, info->key.label, sport, bareudp->port, - !(info->key.tun_flags & TUNNEL_CSUM)); + !test_bit(IP_TUNNEL_CSUM_BIT, + info->key.tun_flags)); return 0; free_dst: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h index 92065568bb19..6873c1201803 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h @@ -117,7 +117,7 @@ bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a, bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a, struct mlx5e_encap_key *b, - __be16 tun_flags); + u32 tun_type); #endif /* CONFIG_MLX5_ESWITCH */ #endif //__MLX5_EN_TC_TUNNEL_H__ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c index f1d1e1542e81..878cbdbf5ec8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c @@ -587,7 +587,7 @@ bool mlx5e_tc_tun_encap_info_equal_generic(struct mlx5e_encap_key *a, bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a, struct mlx5e_encap_key *b, - __be16 tun_flags) + u32 tun_type) { struct ip_tunnel_info *a_info; struct ip_tunnel_info *b_info; @@ -596,8 +596,8 @@ bool mlx5e_tc_tun_encap_info_equal_options(struct mlx5e_encap_key *a, if (!mlx5e_tc_tun_encap_info_equal_generic(a, b)) return false; - a_has_opts = !!(a->ip_tun_key->tun_flags & tun_flags); - b_has_opts = !!(b->ip_tun_key->tun_flags & tun_flags); + a_has_opts = test_bit(tun_type, a->ip_tun_key->tun_flags); + b_has_opts = test_bit(tun_type, b->ip_tun_key->tun_flags); /* keys are equal when both don't have any options attached */ if (!a_has_opts && !b_has_opts) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c index 2bcd10b6d653..bf969212cc77 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c @@ -106,12 +106,13 @@ static int mlx5e_gen_ip_tunnel_header_geneve(char buf[], memset(geneveh, 0, sizeof(*geneveh)); geneveh->ver = MLX5E_GENEVE_VER; geneveh->opt_len = tun_info->options_len / 4; - geneveh->oam = !!(tun_info->key.tun_flags & TUNNEL_OAM); - geneveh->critical = !!(tun_info->key.tun_flags & TUNNEL_CRIT_OPT); + geneveh->oam = test_bit(IP_TUNNEL_OAM_BIT, tun_info->key.tun_flags); + geneveh->critical = test_bit(IP_TUNNEL_CRIT_OPT_BIT, + tun_info->key.tun_flags); mlx5e_tunnel_id_to_vni(tun_info->key.tun_id, geneveh->vni); geneveh->proto_type = htons(ETH_P_TEB); - if (tun_info->key.tun_flags & TUNNEL_GENEVE_OPT) { + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, tun_info->key.tun_flags)) { if (!geneveh->opt_len) return -EOPNOTSUPP; ip_tunnel_info_opts_get(geneveh->options, tun_info); @@ -188,7 +189,7 @@ static int mlx5e_tc_tun_parse_geneve_options(struct mlx5e_priv *priv, /* make sure that we're talking about GENEVE options */ - if (enc_opts.key->dst_opt_type != TUNNEL_GENEVE_OPT) { + if (enc_opts.key->dst_opt_type != IP_TUNNEL_GENEVE_OPT_BIT) { NL_SET_ERR_MSG_MOD(extack, "Matching on GENEVE options: option type is not GENEVE"); netdev_warn(priv->netdev, @@ -337,7 +338,8 @@ static int mlx5e_tc_tun_parse_geneve(struct mlx5e_priv *priv, static bool mlx5e_tc_tun_encap_info_equal_geneve(struct mlx5e_encap_key *a, struct mlx5e_encap_key *b) { - return mlx5e_tc_tun_encap_info_equal_options(a, b, TUNNEL_GENEVE_OPT); + return mlx5e_tc_tun_encap_info_equal_options(a, b, + IP_TUNNEL_GENEVE_OPT_BIT); } struct mlx5e_tc_tunnel geneve_tunnel = { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c index ada14f0574dc..579eda89fc76 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_gre.c @@ -31,12 +31,16 @@ static int mlx5e_gen_ip_tunnel_header_gretap(char buf[], const struct ip_tunnel_key *tun_key = &e->tun_info->key; struct gre_base_hdr *greh = (struct gre_base_hdr *)(buf); __be32 tun_id = tunnel_id_to_key32(tun_key->tun_id); + IP_TUNNEL_DECLARE_FLAGS(unsupp) = { }; int hdr_len; *ip_proto = IPPROTO_GRE; /* the HW does not calculate GRE csum or sequences */ - if (tun_key->tun_flags & (TUNNEL_CSUM | TUNNEL_SEQ)) + __set_bit(IP_TUNNEL_CSUM_BIT, unsupp); + __set_bit(IP_TUNNEL_SEQ_BIT, unsupp); + + if (ip_tunnel_flags_intersect(tun_key->tun_flags, unsupp)) return -EOPNOTSUPP; greh->protocol = htons(ETH_P_TEB); @@ -44,7 +48,7 @@ static int mlx5e_gen_ip_tunnel_header_gretap(char buf[], /* GRE key */ hdr_len = mlx5e_tc_tun_calc_hlen_gretap(e); greh->flags = gre_tnl_flags_to_gre_flags(tun_key->tun_flags); - if (tun_key->tun_flags & TUNNEL_KEY) { + if (test_bit(IP_TUNNEL_KEY_BIT, tun_key->tun_flags)) { __be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4); *ptr = tun_id; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c index a184d739d5f8..e4e487c8431b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_vxlan.c @@ -90,7 +90,7 @@ static int mlx5e_gen_ip_tunnel_header_vxlan(char buf[], const struct vxlan_metadata *md; struct vxlanhdr *vxh; - if ((tun_key->tun_flags & TUNNEL_VXLAN_OPT) && + if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, tun_key->tun_flags) && e->tun_info->options_len != sizeof(*md)) return -EOPNOTSUPP; vxh = (struct vxlanhdr *)((char *)udp + sizeof(struct udphdr)); @@ -99,7 +99,7 @@ static int mlx5e_gen_ip_tunnel_header_vxlan(char buf[], udp->dest = tun_key->tp_dst; vxh->vx_flags = VXLAN_HF_VNI; vxh->vx_vni = vxlan_vni_field(tun_id); - if (tun_key->tun_flags & TUNNEL_VXLAN_OPT) { + if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, tun_key->tun_flags)) { md = ip_tunnel_info_opts(e->tun_info); vxlan_build_gbp_hdr(vxh, md); } @@ -125,7 +125,7 @@ static int mlx5e_tc_tun_parse_vxlan_gbp_option(struct mlx5e_priv *priv, return -EOPNOTSUPP; } - if (enc_opts.key->dst_opt_type != TUNNEL_VXLAN_OPT) { + if (enc_opts.key->dst_opt_type != IP_TUNNEL_VXLAN_OPT_BIT) { NL_SET_ERR_MSG_MOD(extack, "Wrong VxLAN option type: not GBP"); return -EOPNOTSUPP; } @@ -208,7 +208,8 @@ static int mlx5e_tc_tun_parse_vxlan(struct mlx5e_priv *priv, static bool mlx5e_tc_tun_encap_info_equal_vxlan(struct mlx5e_encap_key *a, struct mlx5e_encap_key *b) { - return mlx5e_tc_tun_encap_info_equal_options(a, b, TUNNEL_VXLAN_OPT); + return mlx5e_tc_tun_encap_info_equal_options(a, b, + IP_TUNNEL_VXLAN_OPT_BIT); } static int mlx5e_tc_tun_get_remote_ifindex(struct net_device *mirred_dev) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 31ed26cac9bf..f2f19fa37d91 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -5464,6 +5464,7 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct tunnel_match_enc_opts enc_opts = {}; struct mlx5_rep_uplink_priv *uplink_priv; + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct mlx5e_rep_priv *uplink_rpriv; struct metadata_dst *tun_dst; struct tunnel_match_key key; @@ -5471,6 +5472,8 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb struct net_device *dev; int err; + __set_bit(IP_TUNNEL_KEY_BIT, flags); + enc_opts_id = tunnel_id & ENC_OPTS_BITS_MASK; tun_id = tunnel_id >> ENC_OPTS_BITS; @@ -5503,14 +5506,14 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb case FLOW_DISSECTOR_KEY_IPV4_ADDRS: tun_dst = __ip_tun_set_dst(key.enc_ipv4.src, key.enc_ipv4.dst, key.enc_ip.tos, key.enc_ip.ttl, - key.enc_tp.dst, TUNNEL_KEY, + key.enc_tp.dst, flags, key32_to_tunnel_id(key.enc_key_id.keyid), enc_opts.key.len); break; case FLOW_DISSECTOR_KEY_IPV6_ADDRS: tun_dst = __ipv6_tun_set_dst(&key.enc_ipv6.src, &key.enc_ipv6.dst, key.enc_ip.tos, key.enc_ip.ttl, - key.enc_tp.dst, 0, TUNNEL_KEY, + key.enc_tp.dst, 0, flags, key32_to_tunnel_id(key.enc_key_id.keyid), enc_opts.key.len); break; @@ -5528,11 +5531,16 @@ static bool mlx5e_tc_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb tun_dst->u.tun_info.key.tp_src = key.enc_tp.src; - if (enc_opts.key.len) + if (enc_opts.key.len) { + ip_tunnel_flags_zero(flags); + if (enc_opts.key.dst_opt_type) + __set_bit(enc_opts.key.dst_opt_type, flags); + ip_tunnel_info_opts_set(&tun_dst->u.tun_info, enc_opts.key.data, enc_opts.key.len, - enc_opts.key.dst_opt_type); + flags); + } skb_dst_set(skb, (struct dst_entry *)tun_dst); dev = dev_get_by_index(&init_net, key.filter_ifindex); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c index d67df358a52f..d761a1235994 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ipip.c @@ -27,23 +27,23 @@ mlxsw_sp_ipip_netdev_parms6(const struct net_device *ol_dev) static bool mlxsw_sp_ipip_parms4_has_ikey(const struct ip_tunnel_parm_kern *parms) { - return !!(parms->i_flags & TUNNEL_KEY); + return test_bit(IP_TUNNEL_KEY_BIT, parms->i_flags); } static bool mlxsw_sp_ipip_parms6_has_ikey(const struct __ip6_tnl_parm *parms) { - return !!(parms->i_flags & TUNNEL_KEY); + return test_bit(IP_TUNNEL_KEY_BIT, parms->i_flags); } static bool mlxsw_sp_ipip_parms4_has_okey(const struct ip_tunnel_parm_kern *parms) { - return !!(parms->o_flags & TUNNEL_KEY); + return test_bit(IP_TUNNEL_KEY_BIT, parms->o_flags); } static bool mlxsw_sp_ipip_parms6_has_okey(const struct __ip6_tnl_parm *parms) { - return !!(parms->o_flags & TUNNEL_KEY); + return test_bit(IP_TUNNEL_KEY_BIT, parms->o_flags); } static u32 mlxsw_sp_ipip_parms4_ikey(const struct ip_tunnel_parm_kern *parms) @@ -242,12 +242,15 @@ static bool mlxsw_sp_ipip_can_offload_gre4(const struct mlxsw_sp *mlxsw_sp, const struct net_device *ol_dev) { struct ip_tunnel *tunnel = netdev_priv(ol_dev); - __be16 okflags = TUNNEL_KEY; /* We can't offload any other features. */ bool inherit_ttl = tunnel->parms.iph.ttl == 0; bool inherit_tos = tunnel->parms.iph.tos & 0x1; + IP_TUNNEL_DECLARE_FLAGS(okflags) = { }; - return (tunnel->parms.i_flags & ~okflags) == 0 && - (tunnel->parms.o_flags & ~okflags) == 0 && + /* We can't offload any other features. */ + __set_bit(IP_TUNNEL_KEY_BIT, okflags); + + return ip_tunnel_flags_subset(tunnel->parms.i_flags, okflags) && + ip_tunnel_flags_subset(tunnel->parms.o_flags, okflags) && inherit_ttl && inherit_tos && mlxsw_sp_ipip_tunnel_complete(MLXSW_SP_L3_PROTO_IPV4, ol_dev); } @@ -443,10 +446,13 @@ static bool mlxsw_sp_ipip_can_offload_gre6(const struct mlxsw_sp *mlxsw_sp, struct __ip6_tnl_parm tparm = mlxsw_sp_ipip_netdev_parms6(ol_dev); bool inherit_tos = tparm.flags & IP6_TNL_F_USE_ORIG_TCLASS; bool inherit_ttl = tparm.hop_limit == 0; - __be16 okflags = TUNNEL_KEY; /* We can't offload any other features. */ + IP_TUNNEL_DECLARE_FLAGS(okflags) = { }; + + /* We can't offload any other features. */ + __set_bit(IP_TUNNEL_KEY_BIT, okflags); - return (tparm.i_flags & ~okflags) == 0 && - (tparm.o_flags & ~okflags) == 0 && + return ip_tunnel_flags_subset(tparm.i_flags, okflags) && + ip_tunnel_flags_subset(tparm.o_flags, okflags) && inherit_ttl && inherit_tos && mlxsw_sp_ipip_tunnel_complete(MLXSW_SP_L3_PROTO_IPV6, ol_dev); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index f0cbc6b9b041..3de69b2eb5c4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -461,7 +461,8 @@ mlxsw_sp_span_entry_gretap4_parms(struct mlxsw_sp *mlxsw_sp, if (!(to_dev->flags & IFF_UP) || /* Reject tunnels with GRE keys, checksums, etc. */ - tparm.i_flags || tparm.o_flags || + !ip_tunnel_flags_empty(tparm.i_flags) || + !ip_tunnel_flags_empty(tparm.o_flags) || /* Require a fixed TTL and a TOS copied from the mirrored packet. */ inherit_ttl || !inherit_tos || /* A destination address may not be "any". */ @@ -565,7 +566,8 @@ mlxsw_sp_span_entry_gretap6_parms(struct mlxsw_sp *mlxsw_sp, if (!(to_dev->flags & IFF_UP) || /* Reject tunnels with GRE keys, checksums, etc. */ - tparm.i_flags || tparm.o_flags || + !ip_tunnel_flags_empty(tparm.i_flags) || + !ip_tunnel_flags_empty(tparm.o_flags) || /* Require a fixed TTL and a TOS copied from the mirrored packet. */ inherit_ttl || !inherit_tos || /* A destination address may not be "any". */ diff --git a/drivers/net/ethernet/netronome/nfp/flower/action.c b/drivers/net/ethernet/netronome/nfp/flower/action.c index 2c3f62907958..aca2a7417af3 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/action.c +++ b/drivers/net/ethernet/netronome/nfp/flower/action.c @@ -396,6 +396,17 @@ nfp_fl_push_geneve_options(struct nfp_fl_payload *nfp_fl, int *list_len, return 0; } +#define NFP_FL_CHECK(flag) ({ \ + IP_TUNNEL_DECLARE_FLAGS(__check) = { }; \ + __be16 __res; \ + \ + __set_bit(IP_TUNNEL_##flag##_BIT, __check); \ + __res = ip_tunnel_flags_to_be16(__check); \ + \ + BUILD_BUG_ON(__builtin_constant_p(__res) && \ + NFP_FL_TUNNEL_##flag != __res); \ +}) + static int nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun, const struct flow_action_entry *act, @@ -410,6 +421,7 @@ nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun, u32 tmp_set_ip_tun_type_index = 0; /* Currently support one pre-tunnel so index is always 0. */ int pretun_idx = 0; + __be16 tun_flags; if (!IS_ENABLED(CONFIG_IPV6) && ipv6) return -EOPNOTSUPP; @@ -417,9 +429,10 @@ nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun, if (ipv6 && !(priv->flower_ext_feats & NFP_FL_FEATS_IPV6_TUN)) return -EOPNOTSUPP; - BUILD_BUG_ON(NFP_FL_TUNNEL_CSUM != TUNNEL_CSUM || - NFP_FL_TUNNEL_KEY != TUNNEL_KEY || - NFP_FL_TUNNEL_GENEVE_OPT != TUNNEL_GENEVE_OPT); + NFP_FL_CHECK(CSUM); + NFP_FL_CHECK(KEY); + NFP_FL_CHECK(GENEVE_OPT); + if (ip_tun->options_len && (tun_type != NFP_FL_TUNNEL_GENEVE || !(priv->flower_ext_feats & NFP_FL_FEATS_GENEVE_OPT))) { @@ -427,7 +440,9 @@ nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun, return -EOPNOTSUPP; } - if (ip_tun->key.tun_flags & ~NFP_FL_SUPPORTED_UDP_TUN_FLAGS) { + tun_flags = ip_tunnel_flags_to_be16(ip_tun->key.tun_flags); + if (!ip_tunnel_flags_is_be16_compat(ip_tun->key.tun_flags) || + (tun_flags & ~NFP_FL_SUPPORTED_UDP_TUN_FLAGS)) { NL_SET_ERR_MSG_MOD(extack, "unsupported offload: loaded firmware does not support tunnel flag offload"); return -EOPNOTSUPP; @@ -442,7 +457,7 @@ nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun, FIELD_PREP(NFP_FL_PRE_TUN_INDEX, pretun_idx); set_tun->tun_type_index = cpu_to_be32(tmp_set_ip_tun_type_index); - if (ip_tun->key.tun_flags & NFP_FL_TUNNEL_KEY) + if (tun_flags & NFP_FL_TUNNEL_KEY) set_tun->tun_id = ip_tun->key.tun_id; if (ip_tun->key.ttl) { @@ -486,7 +501,7 @@ nfp_fl_set_tun(struct nfp_app *app, struct nfp_fl_set_tun *set_tun, } set_tun->tos = ip_tun->key.tos; - set_tun->tun_flags = ip_tun->key.tun_flags; + set_tun->tun_flags = tun_flags; if (tun_type == NFP_FL_TUNNEL_GENEVE) { set_tun->tun_proto = htons(ETH_P_TEB); diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 2f6739fe78af..163f94a5a58f 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -225,10 +225,11 @@ static void geneve_rx(struct geneve_dev *geneve, struct geneve_sock *gs, void *oiph; if (ip_tunnel_collect_metadata() || gs->collect_md) { - __be16 flags; + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; - flags = TUNNEL_KEY | (gnvh->oam ? TUNNEL_OAM : 0) | - (gnvh->critical ? TUNNEL_CRIT_OPT : 0); + __set_bit(IP_TUNNEL_KEY_BIT, flags); + __assign_bit(IP_TUNNEL_OAM_BIT, flags, gnvh->oam); + __assign_bit(IP_TUNNEL_CRIT_OPT_BIT, flags, gnvh->critical); tun_dst = udp_tun_rx_dst(skb, geneve_get_sk_family(gs), flags, vni_to_tunnel_id(gnvh->vni), @@ -238,9 +239,11 @@ static void geneve_rx(struct geneve_dev *geneve, struct geneve_sock *gs, goto drop; } /* Update tunnel dst according to Geneve options. */ + ip_tunnel_flags_zero(flags); + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, flags); ip_tunnel_info_opts_set(&tun_dst->u.tun_info, gnvh->options, gnvh->opt_len * 4, - TUNNEL_GENEVE_OPT); + flags); } else { /* Drop packets w/ critical options, * since we don't support any... @@ -745,14 +748,15 @@ static void geneve_build_header(struct genevehdr *geneveh, { geneveh->ver = GENEVE_VER; geneveh->opt_len = info->options_len / 4; - geneveh->oam = !!(info->key.tun_flags & TUNNEL_OAM); - geneveh->critical = !!(info->key.tun_flags & TUNNEL_CRIT_OPT); + geneveh->oam = test_bit(IP_TUNNEL_OAM_BIT, info->key.tun_flags); + geneveh->critical = test_bit(IP_TUNNEL_CRIT_OPT_BIT, + info->key.tun_flags); geneveh->rsvd1 = 0; tunnel_id_to_vni(info->key.tun_id, geneveh->vni); geneveh->proto_type = inner_proto; geneveh->rsvd2 = 0; - if (info->key.tun_flags & TUNNEL_GENEVE_OPT) + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, info->key.tun_flags)) ip_tunnel_info_opts_get(geneveh->options, info); } @@ -761,7 +765,7 @@ static int geneve_build_skb(struct dst_entry *dst, struct sk_buff *skb, bool xnet, int ip_hdr_len, bool inner_proto_inherit) { - bool udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM); + bool udp_sum = test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); struct genevehdr *gnvh; __be16 inner_proto; int min_headroom; @@ -878,7 +882,8 @@ static int geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, if (geneve->cfg.collect_md) { ttl = key->ttl; - df = key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; + df = test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, key->tun_flags) ? + htons(IP_DF) : 0; } else { if (geneve->cfg.ttl_inherit) ttl = ip_tunnel_get_ttl(ip_hdr(skb), skb); @@ -910,7 +915,8 @@ static int geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, udp_tunnel_xmit_skb(rt, gs4->sock->sk, skb, saddr, info->key.u.ipv4.dst, tos, ttl, df, sport, geneve->cfg.info.key.tp_dst, !net_eq(geneve->net, dev_net(geneve->dev)), - !(info->key.tun_flags & TUNNEL_CSUM)); + !test_bit(IP_TUNNEL_CSUM_BIT, + info->key.tun_flags)); return 0; } @@ -998,7 +1004,8 @@ static int geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev, udp_tunnel6_xmit_skb(dst, gs6->sock->sk, skb, dev, &saddr, &key->u.ipv6.dst, prio, ttl, info->key.label, sport, geneve->cfg.info.key.tp_dst, - !(info->key.tun_flags & TUNNEL_CSUM)); + !test_bit(IP_TUNNEL_CSUM_BIT, + info->key.tun_flags)); return 0; } #endif @@ -1297,7 +1304,8 @@ static struct geneve_dev *geneve_find_dev(struct geneve_net *gn, static bool is_tnl_info_zero(const struct ip_tunnel_info *info) { - return !(info->key.tun_id || info->key.tun_flags || info->key.tos || + return !(info->key.tun_id || info->key.tos || + !ip_tunnel_flags_empty(info->key.tun_flags) || info->key.ttl || info->key.label || info->key.tp_src || memchr_inv(&info->key.u, 0, sizeof(info->key.u))); } @@ -1435,7 +1443,7 @@ static int geneve_nl2info(struct nlattr *tb[], struct nlattr *data[], "Remote IPv6 address cannot be Multicast"); return -EINVAL; } - info->key.tun_flags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); cfg->use_udp6_rx_checksums = true; #else NL_SET_ERR_MSG_ATTR(extack, data[IFLA_GENEVE_REMOTE6], @@ -1510,7 +1518,7 @@ static int geneve_nl2info(struct nlattr *tb[], struct nlattr *data[], goto change_notsup; } if (nla_get_u8(data[IFLA_GENEVE_UDP_CSUM])) - info->key.tun_flags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); } if (data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX]) { @@ -1520,7 +1528,7 @@ static int geneve_nl2info(struct nlattr *tb[], struct nlattr *data[], goto change_notsup; } if (nla_get_u8(data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX])) - info->key.tun_flags &= ~TUNNEL_CSUM; + __clear_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); #else NL_SET_ERR_MSG_ATTR(extack, data[IFLA_GENEVE_UDP_ZERO_CSUM6_TX], "IPv6 support not enabled in the kernel"); @@ -1753,7 +1761,8 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev) info->key.u.ipv4.dst)) goto nla_put_failure; if (nla_put_u8(skb, IFLA_GENEVE_UDP_CSUM, - !!(info->key.tun_flags & TUNNEL_CSUM))) + test_bit(IP_TUNNEL_CSUM_BIT, + info->key.tun_flags))) goto nla_put_failure; #if IS_ENABLED(CONFIG_IPV6) @@ -1762,7 +1771,8 @@ static int geneve_fill_info(struct sk_buff *skb, const struct net_device *dev) &info->key.u.ipv6.dst)) goto nla_put_failure; if (nla_put_u8(skb, IFLA_GENEVE_UDP_ZERO_CSUM6_TX, - !(info->key.tun_flags & TUNNEL_CSUM))) + !test_bit(IP_TUNNEL_CSUM_BIT, + info->key.tun_flags))) goto nla_put_failure; #endif } diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index 3495591a5c29..72ecf6cf809c 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -1584,7 +1584,8 @@ static void vxlan_parse_gbp_hdr(struct vxlanhdr *unparsed, tun_dst = (struct metadata_dst *)skb_dst(skb); if (tun_dst) { - tun_dst->u.tun_info.key.tun_flags |= TUNNEL_VXLAN_OPT; + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, + tun_dst->u.tun_info.key.tun_flags); tun_dst->u.tun_info.options_len = sizeof(*md); } if (gbp->dont_learn) @@ -1716,9 +1717,11 @@ static int vxlan_rcv(struct sock *sk, struct sk_buff *skb) goto drop; if (vxlan_collect_metadata(vs)) { + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct metadata_dst *tun_dst; - tun_dst = udp_tun_rx_dst(skb, vxlan_get_sk_family(vs), TUNNEL_KEY, + __set_bit(IP_TUNNEL_KEY_BIT, flags); + tun_dst = udp_tun_rx_dst(skb, vxlan_get_sk_family(vs), flags, key32_to_tunnel_id(vni), sizeof(*md)); if (!tun_dst) @@ -2403,14 +2406,14 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, vni = tunnel_id_to_key32(info->key.tun_id); ifindex = 0; dst_cache = &info->dst_cache; - if (info->key.tun_flags & TUNNEL_VXLAN_OPT) { + if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, info->key.tun_flags)) { if (info->options_len < sizeof(*md)) goto drop; md = ip_tunnel_info_opts(info); } ttl = info->key.ttl; tos = info->key.tos; - udp_sum = !!(info->key.tun_flags & TUNNEL_CSUM); + udp_sum = test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); } src_port = udp_flow_src_port(dev_net(dev), skb, vxlan->cfg.port_min, vxlan->cfg.port_max, true); @@ -2451,7 +2454,8 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, old_iph->frag_off & htons(IP_DF))) df = htons(IP_DF); } - } else if (info->key.tun_flags & TUNNEL_DONT_FRAGMENT) { + } else if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, + info->key.tun_flags)) { df = htons(IP_DF); } diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h index 1b7fae4c6b24..4160731dcb6e 100644 --- a/include/net/dst_metadata.h +++ b/include/net/dst_metadata.h @@ -198,7 +198,7 @@ static inline struct metadata_dst *__ip_tun_set_dst(__be32 saddr, __be32 daddr, __u8 tos, __u8 ttl, __be16 tp_dst, - __be16 flags, + const unsigned long *flags, __be64 tunnel_id, int md_size) { @@ -215,7 +215,7 @@ static inline struct metadata_dst *__ip_tun_set_dst(__be32 saddr, } static inline struct metadata_dst *ip_tun_rx_dst(struct sk_buff *skb, - __be16 flags, + const unsigned long *flags, __be64 tunnel_id, int md_size) { @@ -230,7 +230,7 @@ static inline struct metadata_dst *__ipv6_tun_set_dst(const struct in6_addr *sad __u8 tos, __u8 ttl, __be16 tp_dst, __be32 label, - __be16 flags, + const unsigned long *flags, __be64 tunnel_id, int md_size) { @@ -243,7 +243,7 @@ static inline struct metadata_dst *__ipv6_tun_set_dst(const struct in6_addr *sad info = &tun_dst->u.tun_info; info->mode = IP_TUNNEL_INFO_IPV6; - info->key.tun_flags = flags; + ip_tunnel_flags_copy(info->key.tun_flags, flags); info->key.tun_id = tunnel_id; info->key.tp_src = 0; info->key.tp_dst = tp_dst; @@ -259,7 +259,7 @@ static inline struct metadata_dst *__ipv6_tun_set_dst(const struct in6_addr *sad } static inline struct metadata_dst *ipv6_tun_rx_dst(struct sk_buff *skb, - __be16 flags, + const unsigned long *flags, __be64 tunnel_id, int md_size) { diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h index 1a7131d6cb0e..9ab376d1a677 100644 --- a/include/net/flow_dissector.h +++ b/include/net/flow_dissector.h @@ -97,7 +97,7 @@ struct flow_dissector_key_enc_opts { * here but seems difficult to #include */ u8 len; - __be16 dst_opt_type; + u32 dst_opt_type; }; struct flow_dissector_key_keyid { diff --git a/include/net/gre.h b/include/net/gre.h index 4e209708b754..ccd293203284 100644 --- a/include/net/gre.h +++ b/include/net/gre.h @@ -49,67 +49,61 @@ static inline bool netif_is_ip6gretap(const struct net_device *dev) !strcmp(dev->rtnl_link_ops->kind, "ip6gretap"); } -static inline int gre_calc_hlen(__be16 o_flags) +static inline int gre_calc_hlen(const unsigned long *o_flags) { int addend = 4; - if (o_flags & TUNNEL_CSUM) + if (test_bit(IP_TUNNEL_CSUM_BIT, o_flags)) addend += 4; - if (o_flags & TUNNEL_KEY) + if (test_bit(IP_TUNNEL_KEY_BIT, o_flags)) addend += 4; - if (o_flags & TUNNEL_SEQ) + if (test_bit(IP_TUNNEL_SEQ_BIT, o_flags)) addend += 4; return addend; } -static inline __be16 gre_flags_to_tnl_flags(__be16 flags) +static inline void gre_flags_to_tnl_flags(unsigned long *dst, __be16 flags) { - __be16 tflags = 0; - - if (flags & GRE_CSUM) - tflags |= TUNNEL_CSUM; - if (flags & GRE_ROUTING) - tflags |= TUNNEL_ROUTING; - if (flags & GRE_KEY) - tflags |= TUNNEL_KEY; - if (flags & GRE_SEQ) - tflags |= TUNNEL_SEQ; - if (flags & GRE_STRICT) - tflags |= TUNNEL_STRICT; - if (flags & GRE_REC) - tflags |= TUNNEL_REC; - if (flags & GRE_VERSION) - tflags |= TUNNEL_VERSION; - - return tflags; + IP_TUNNEL_DECLARE_FLAGS(res) = { }; + + __assign_bit(IP_TUNNEL_CSUM_BIT, res, flags & GRE_CSUM); + __assign_bit(IP_TUNNEL_ROUTING_BIT, res, flags & GRE_ROUTING); + __assign_bit(IP_TUNNEL_KEY_BIT, res, flags & GRE_KEY); + __assign_bit(IP_TUNNEL_SEQ_BIT, res, flags & GRE_SEQ); + __assign_bit(IP_TUNNEL_STRICT_BIT, res, flags & GRE_STRICT); + __assign_bit(IP_TUNNEL_REC_BIT, res, flags & GRE_REC); + __assign_bit(IP_TUNNEL_VERSION_BIT, res, flags & GRE_VERSION); + + ip_tunnel_flags_copy(dst, res); } -static inline __be16 gre_tnl_flags_to_gre_flags(__be16 tflags) +static inline __be16 gre_tnl_flags_to_gre_flags(const unsigned long *tflags) { __be16 flags = 0; - if (tflags & TUNNEL_CSUM) + if (test_bit(IP_TUNNEL_CSUM_BIT, tflags)) flags |= GRE_CSUM; - if (tflags & TUNNEL_ROUTING) + if (test_bit(IP_TUNNEL_ROUTING_BIT, tflags)) flags |= GRE_ROUTING; - if (tflags & TUNNEL_KEY) + if (test_bit(IP_TUNNEL_KEY_BIT, tflags)) flags |= GRE_KEY; - if (tflags & TUNNEL_SEQ) + if (test_bit(IP_TUNNEL_SEQ_BIT, tflags)) flags |= GRE_SEQ; - if (tflags & TUNNEL_STRICT) + if (test_bit(IP_TUNNEL_STRICT_BIT, tflags)) flags |= GRE_STRICT; - if (tflags & TUNNEL_REC) + if (test_bit(IP_TUNNEL_REC_BIT, tflags)) flags |= GRE_REC; - if (tflags & TUNNEL_VERSION) + if (test_bit(IP_TUNNEL_VERSION_BIT, tflags)) flags |= GRE_VERSION; return flags; } static inline void gre_build_header(struct sk_buff *skb, int hdr_len, - __be16 flags, __be16 proto, + const unsigned long *flags, __be16 proto, __be32 key, __be32 seq) { + IP_TUNNEL_DECLARE_FLAGS(cond) = { }; struct gre_base_hdr *greh; skb_push(skb, hdr_len); @@ -120,18 +114,22 @@ static inline void gre_build_header(struct sk_buff *skb, int hdr_len, greh->flags = gre_tnl_flags_to_gre_flags(flags); greh->protocol = proto; - if (flags & (TUNNEL_KEY | TUNNEL_CSUM | TUNNEL_SEQ)) { + __set_bit(IP_TUNNEL_KEY_BIT, cond); + __set_bit(IP_TUNNEL_CSUM_BIT, cond); + __set_bit(IP_TUNNEL_SEQ_BIT, cond); + + if (ip_tunnel_flags_intersect(flags, cond)) { __be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4); - if (flags & TUNNEL_SEQ) { + if (test_bit(IP_TUNNEL_SEQ_BIT, flags)) { *ptr = seq; ptr--; } - if (flags & TUNNEL_KEY) { + if (test_bit(IP_TUNNEL_KEY_BIT, flags)) { *ptr = key; ptr--; } - if (flags & TUNNEL_CSUM && + if (test_bit(IP_TUNNEL_CSUM_BIT, flags) && !(skb_shinfo(skb)->gso_type & (SKB_GSO_GRE | SKB_GSO_GRE_CSUM))) { *ptr = 0; diff --git a/include/net/ip6_tunnel.h b/include/net/ip6_tunnel.h index 74b369bddf49..399592405c72 100644 --- a/include/net/ip6_tunnel.h +++ b/include/net/ip6_tunnel.h @@ -30,8 +30,8 @@ struct __ip6_tnl_parm { struct in6_addr laddr; /* local tunnel end-point address */ struct in6_addr raddr; /* remote tunnel end-point address */ - __be16 i_flags; - __be16 o_flags; + IP_TUNNEL_DECLARE_FLAGS(i_flags); + IP_TUNNEL_DECLARE_FLAGS(o_flags); __be32 i_key; __be32 o_key; diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 20f0319ab149..ed8e48cc9054 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -36,6 +36,24 @@ (sizeof_field(struct ip_tunnel_key, u) - \ sizeof_field(struct ip_tunnel_key, u.ipv4)) +#define __ipt_flag_op(op, ...) \ + op(__VA_ARGS__, __IP_TUNNEL_FLAG_NUM) + +#define IP_TUNNEL_DECLARE_FLAGS(...) \ + __ipt_flag_op(DECLARE_BITMAP, __VA_ARGS__) + +#define ip_tunnel_flags_zero(...) __ipt_flag_op(bitmap_zero, __VA_ARGS__) +#define ip_tunnel_flags_copy(...) __ipt_flag_op(bitmap_copy, __VA_ARGS__) +#define ip_tunnel_flags_and(...) __ipt_flag_op(bitmap_and, __VA_ARGS__) +#define ip_tunnel_flags_or(...) __ipt_flag_op(bitmap_or, __VA_ARGS__) + +#define ip_tunnel_flags_empty(...) \ + __ipt_flag_op(bitmap_empty, __VA_ARGS__) +#define ip_tunnel_flags_intersect(...) \ + __ipt_flag_op(bitmap_intersects, __VA_ARGS__) +#define ip_tunnel_flags_subset(...) \ + __ipt_flag_op(bitmap_subset, __VA_ARGS__) + struct ip_tunnel_key { __be64 tun_id; union { @@ -48,11 +66,11 @@ struct ip_tunnel_key { struct in6_addr dst; } ipv6; } u; - __be16 tun_flags; - u8 tos; /* TOS for IPv4, TC for IPv6 */ - u8 ttl; /* TTL for IPv4, HL for IPv6 */ + IP_TUNNEL_DECLARE_FLAGS(tun_flags); __be32 label; /* Flow Label for IPv6 */ u32 nhid; + u8 tos; /* TOS for IPv4, TC for IPv6 */ + u8 ttl; /* TTL for IPv4, HL for IPv6 */ __be16 tp_src; __be16 tp_dst; __u8 flow_flags; @@ -110,14 +128,14 @@ struct ip_tunnel_prl_entry { struct metadata_dst; -/* Kernel-side copy of ip_tunnel_parm */ +/* Kernel-side variant of ip_tunnel_parm */ struct ip_tunnel_parm_kern { char name[IFNAMSIZ]; - int link; - __be16 i_flags; - __be16 o_flags; + IP_TUNNEL_DECLARE_FLAGS(i_flags); + IP_TUNNEL_DECLARE_FLAGS(o_flags); __be32 i_key; __be32 o_key; + int link; struct iphdr iph; }; @@ -168,7 +186,7 @@ struct ip_tunnel { }; struct tnl_ptk_info { - __be16 flags; + IP_TUNNEL_DECLARE_FLAGS(flags); __be16 proto; __be32 key; __be32 seq; @@ -190,11 +208,77 @@ struct ip_tunnel_net { int type; }; +static inline void ip_tunnel_set_options_present(unsigned long *flags) +{ + IP_TUNNEL_DECLARE_FLAGS(present) = { }; + + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present); + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present); + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present); + __set_bit(IP_TUNNEL_GTP_OPT_BIT, present); + + ip_tunnel_flags_or(flags, flags, present); +} + +static inline void ip_tunnel_clear_options_present(unsigned long *flags) +{ + IP_TUNNEL_DECLARE_FLAGS(present) = { }; + + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present); + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present); + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present); + __set_bit(IP_TUNNEL_GTP_OPT_BIT, present); + + __ipt_flag_op(bitmap_andnot, flags, flags, present); +} + +static inline bool ip_tunnel_is_options_present(const unsigned long *flags) +{ + IP_TUNNEL_DECLARE_FLAGS(present) = { }; + + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, present); + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present); + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present); + __set_bit(IP_TUNNEL_GTP_OPT_BIT, present); + + return ip_tunnel_flags_intersect(flags, present); +} + +static inline bool ip_tunnel_flags_is_be16_compat(const unsigned long *flags) +{ + IP_TUNNEL_DECLARE_FLAGS(supp) = { }; + + bitmap_set(supp, 0, BITS_PER_TYPE(__be16)); + __set_bit(IP_TUNNEL_VTI_BIT, supp); + + return ip_tunnel_flags_subset(flags, supp); +} + +static inline void ip_tunnel_flags_from_be16(unsigned long *dst, __be16 flags) +{ + ip_tunnel_flags_zero(dst); + + bitmap_write(dst, be16_to_cpu(flags), 0, BITS_PER_TYPE(__be16)); + __assign_bit(IP_TUNNEL_VTI_BIT, dst, flags & VTI_ISVTI); +} + +static inline __be16 ip_tunnel_flags_to_be16(const unsigned long *flags) +{ + __be16 ret; + + ret = cpu_to_be16(bitmap_read(flags, 0, BITS_PER_TYPE(__be16))); + if (test_bit(IP_TUNNEL_VTI_BIT, flags)) + ret |= VTI_ISVTI; + + return ret; +} + static inline void ip_tunnel_key_init(struct ip_tunnel_key *key, __be32 saddr, __be32 daddr, u8 tos, u8 ttl, __be32 label, __be16 tp_src, __be16 tp_dst, - __be64 tun_id, __be16 tun_flags) + __be64 tun_id, + const unsigned long *tun_flags) { key->tun_id = tun_id; key->u.ipv4.src = saddr; @@ -204,7 +288,7 @@ static inline void ip_tunnel_key_init(struct ip_tunnel_key *key, key->tos = tos; key->ttl = ttl; key->label = label; - key->tun_flags = tun_flags; + ip_tunnel_flags_copy(key->tun_flags, tun_flags); /* For the tunnel types on the top of IPsec, the tp_src and tp_dst of * the upper tunnel are used. @@ -225,12 +309,8 @@ ip_tunnel_dst_cache_usable(const struct sk_buff *skb, { if (skb->mark) return false; - if (!info) - return true; - if (info->key.tun_flags & TUNNEL_NOCACHE) - return false; - return true; + return !info || !test_bit(IP_TUNNEL_NOCACHE_BIT, info->key.tun_flags); } static inline unsigned short ip_tunnel_info_af(const struct ip_tunnel_info @@ -313,7 +393,7 @@ int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict); int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu); struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, - int link, __be16 flags, + int link, const unsigned long *flags, __be32 remote, __be32 local, __be32 key); @@ -529,12 +609,13 @@ static inline void ip_tunnel_info_opts_get(void *to, static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info, const void *from, int len, - __be16 flags) + const unsigned long *flags) { info->options_len = len; if (len > 0) { memcpy(ip_tunnel_info_opts(info), from, len); - info->key.tun_flags |= flags; + ip_tunnel_flags_or(info->key.tun_flags, info->key.tun_flags, + flags); } } @@ -578,7 +659,7 @@ static inline void ip_tunnel_info_opts_get(void *to, static inline void ip_tunnel_info_opts_set(struct ip_tunnel_info *info, const void *from, int len, - __be16 flags) + const unsigned long *flags) { info->options_len = 0; } diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index d716214fe03d..a93dc51f6323 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -179,8 +179,8 @@ struct dst_entry *udp_tunnel6_dst_lookup(struct sk_buff *skb, struct dst_cache *dst_cache); struct metadata_dst *udp_tun_rx_dst(struct sk_buff *skb, unsigned short family, - __be16 flags, __be64 tunnel_id, - int md_size); + const unsigned long *flags, + __be64 tunnel_id, int md_size); #ifdef CONFIG_INET static inline int udp_tunnel_handle_offloads(struct sk_buff *skb, bool udp_csum) diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index 102119628ff5..838927dd73a4 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -161,6 +161,14 @@ enum { #define IFLA_VTI_MAX (__IFLA_VTI_MAX - 1) +#ifndef __KERNEL__ +/* Historically, tunnel flags have been defined as __be16 and now there are + * no free bits left. It is strongly advised to switch the already existing + * userspace code to the new *_BIT definitions from down below, as __be16 + * can't be simply cast to a wider type on LE systems. All new flags and + * code must use *_BIT only. + */ + #define TUNNEL_CSUM __cpu_to_be16(0x01) #define TUNNEL_ROUTING __cpu_to_be16(0x02) #define TUNNEL_KEY __cpu_to_be16(0x04) @@ -181,5 +189,30 @@ enum { #define TUNNEL_OPTIONS_PRESENT \ (TUNNEL_GENEVE_OPT | TUNNEL_VXLAN_OPT | TUNNEL_ERSPAN_OPT | \ TUNNEL_GTP_OPT) +#endif + +enum { + IP_TUNNEL_CSUM_BIT = 0U, + IP_TUNNEL_ROUTING_BIT, + IP_TUNNEL_KEY_BIT, + IP_TUNNEL_SEQ_BIT, + IP_TUNNEL_STRICT_BIT, + IP_TUNNEL_REC_BIT, + IP_TUNNEL_VERSION_BIT, + IP_TUNNEL_NO_KEY_BIT, + IP_TUNNEL_DONT_FRAGMENT_BIT, + IP_TUNNEL_OAM_BIT, + IP_TUNNEL_CRIT_OPT_BIT, + IP_TUNNEL_GENEVE_OPT_BIT, /* OPTIONS_PRESENT */ + IP_TUNNEL_VXLAN_OPT_BIT, /* OPTIONS_PRESENT */ + IP_TUNNEL_NOCACHE_BIT, + IP_TUNNEL_ERSPAN_OPT_BIT, /* OPTIONS_PRESENT */ + IP_TUNNEL_GTP_OPT_BIT, /* OPTIONS_PRESENT */ + + IP_TUNNEL_VTI_BIT, + IP_TUNNEL_SIT_ISATAP_BIT = IP_TUNNEL_VTI_BIT, + + __IP_TUNNEL_FLAG_NUM, +}; #endif /* _UAPI_IF_TUNNEL_H_ */ diff --git a/net/bridge/br_vlan_tunnel.c b/net/bridge/br_vlan_tunnel.c index 81833ca7a2c7..a966a6ec8263 100644 --- a/net/bridge/br_vlan_tunnel.c +++ b/net/bridge/br_vlan_tunnel.c @@ -65,13 +65,14 @@ static int __vlan_tunnel_info_add(struct net_bridge_vlan_group *vg, { struct metadata_dst *metadata = rtnl_dereference(vlan->tinfo.tunnel_dst); __be64 key = key32_to_tunnel_id(cpu_to_be32(tun_id)); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; int err; if (metadata) return -EEXIST; - metadata = __ip_tun_set_dst(0, 0, 0, 0, 0, TUNNEL_KEY, - key, 0); + __set_bit(IP_TUNNEL_KEY_BIT, flags); + metadata = __ip_tun_set_dst(0, 0, 0, 0, 0, flags, key, 0); if (!metadata) return -EINVAL; @@ -185,6 +186,7 @@ void br_handle_ingress_vlan_tunnel(struct sk_buff *skb, int br_handle_egress_vlan_tunnel(struct sk_buff *skb, struct net_bridge_vlan *vlan) { + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct metadata_dst *tunnel_dst; __be64 tunnel_id; int err; @@ -202,7 +204,8 @@ int br_handle_egress_vlan_tunnel(struct sk_buff *skb, return err; if (BR_INPUT_SKB_CB(skb)->backup_nhid) { - tunnel_dst = __ip_tun_set_dst(0, 0, 0, 0, 0, TUNNEL_KEY, + __set_bit(IP_TUNNEL_KEY_BIT, flags); + tunnel_dst = __ip_tun_set_dst(0, 0, 0, 0, 0, flags, tunnel_id, 0); if (!tunnel_dst) return -ENOMEM; diff --git a/net/core/filter.c b/net/core/filter.c index 0c66e4a3fc5b..294670d3850d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4662,7 +4662,7 @@ set_compat: to->tunnel_tos = info->key.tos; to->tunnel_ttl = info->key.ttl; if (flags & BPF_F_TUNINFO_FLAGS) - to->tunnel_flags = info->key.tun_flags; + to->tunnel_flags = ip_tunnel_flags_to_be16(info->key.tun_flags); else to->tunnel_ext = 0; @@ -4705,7 +4705,7 @@ BPF_CALL_3(bpf_skb_get_tunnel_opt, struct sk_buff *, skb, u8 *, to, u32, size) int err; if (unlikely(!info || - !(info->key.tun_flags & TUNNEL_OPTIONS_PRESENT))) { + !ip_tunnel_is_options_present(info->key.tun_flags))) { err = -ENOENT; goto err_clear; } @@ -4775,15 +4775,15 @@ BPF_CALL_4(bpf_skb_set_tunnel_key, struct sk_buff *, skb, memset(info, 0, sizeof(*info)); info->mode = IP_TUNNEL_INFO_TX; - info->key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM | TUNNEL_NOCACHE; - if (flags & BPF_F_DONT_FRAGMENT) - info->key.tun_flags |= TUNNEL_DONT_FRAGMENT; - if (flags & BPF_F_ZERO_CSUM_TX) - info->key.tun_flags &= ~TUNNEL_CSUM; - if (flags & BPF_F_SEQ_NUMBER) - info->key.tun_flags |= TUNNEL_SEQ; - if (flags & BPF_F_NO_TUNNEL_KEY) - info->key.tun_flags &= ~TUNNEL_KEY; + __set_bit(IP_TUNNEL_NOCACHE_BIT, info->key.tun_flags); + __assign_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, info->key.tun_flags, + flags & BPF_F_DONT_FRAGMENT); + __assign_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags, + !(flags & BPF_F_ZERO_CSUM_TX)); + __assign_bit(IP_TUNNEL_SEQ_BIT, info->key.tun_flags, + flags & BPF_F_SEQ_NUMBER); + __assign_bit(IP_TUNNEL_KEY_BIT, info->key.tun_flags, + !(flags & BPF_F_NO_TUNNEL_KEY)); info->key.tun_id = cpu_to_be64(from->tunnel_id); info->key.tos = from->tunnel_tos; @@ -4821,13 +4821,15 @@ BPF_CALL_3(bpf_skb_set_tunnel_opt, struct sk_buff *, skb, { struct ip_tunnel_info *info = skb_tunnel_info(skb); const struct metadata_dst *md = this_cpu_ptr(md_dst); + IP_TUNNEL_DECLARE_FLAGS(present) = { }; if (unlikely(info != &md->u.tun_info || (size & (sizeof(u32) - 1)))) return -EINVAL; if (unlikely(size > IP_TUNNEL_OPTS_MAX)) return -ENOMEM; - ip_tunnel_info_opts_set(info, from, size, TUNNEL_OPTIONS_PRESENT); + ip_tunnel_set_options_present(present); + ip_tunnel_info_opts_set(info, from, size, present); return 0; } diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 272f09251343..f82e9a7d3b37 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -455,17 +455,25 @@ skb_flow_dissect_tunnel_info(const struct sk_buff *skb, if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_OPTS)) { struct flow_dissector_key_enc_opts *enc_opt; + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; + u32 val; enc_opt = skb_flow_dissector_target(flow_dissector, FLOW_DISSECTOR_KEY_ENC_OPTS, target_container); - if (info->options_len) { - enc_opt->len = info->options_len; - ip_tunnel_info_opts_get(enc_opt->data, info); - enc_opt->dst_opt_type = info->key.tun_flags & - TUNNEL_OPTIONS_PRESENT; - } + if (!info->options_len) + return; + + enc_opt->len = info->options_len; + ip_tunnel_info_opts_get(enc_opt->data, info); + + ip_tunnel_set_options_present(flags); + ip_tunnel_flags_and(flags, info->key.tun_flags, flags); + + val = find_next_bit(flags, __IP_TUNNEL_FLAG_NUM, + IP_TUNNEL_GENEVE_OPT_BIT); + enc_opt->dst_opt_type = val < __IP_TUNNEL_FLAG_NUM ? val : 0; } } EXPORT_SYMBOL(skb_flow_dissect_tunnel_info); diff --git a/net/ipv4/fou_bpf.c b/net/ipv4/fou_bpf.c index 06e5572f296f..54984f3170a8 100644 --- a/net/ipv4/fou_bpf.c +++ b/net/ipv4/fou_bpf.c @@ -64,7 +64,7 @@ __bpf_kfunc int bpf_skb_set_fou_encap(struct __sk_buff *skb_ctx, info->encap.type = TUNNEL_ENCAP_NONE; } - if (info->key.tun_flags & TUNNEL_CSUM) + if (test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags)) info->encap.flags |= TUNNEL_ENCAP_FLAG_CSUM; info->encap.sport = encap->sport; diff --git a/net/ipv4/gre_demux.c b/net/ipv4/gre_demux.c index 3757fd93523f..6701a98d9a9f 100644 --- a/net/ipv4/gre_demux.c +++ b/net/ipv4/gre_demux.c @@ -73,7 +73,7 @@ int gre_parse_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, if (unlikely(greh->flags & (GRE_VERSION | GRE_ROUTING))) return -EINVAL; - tpi->flags = gre_flags_to_tnl_flags(greh->flags); + gre_flags_to_tnl_flags(tpi->flags, greh->flags); hdr_len = gre_calc_hlen(tpi->flags); if (!pskb_may_pull(skb, nhs + hdr_len)) diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 8853c065a985..3c46a7fef8db 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -265,6 +265,7 @@ static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, struct net *net = dev_net(skb->dev); struct metadata_dst *tun_dst = NULL; struct erspan_base_hdr *ershdr; + IP_TUNNEL_DECLARE_FLAGS(flags); struct ip_tunnel_net *itn; struct ip_tunnel *tunnel; const struct iphdr *iph; @@ -272,18 +273,20 @@ static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, int ver; int len; + ip_tunnel_flags_copy(flags, tpi->flags); + itn = net_generic(net, erspan_net_id); iph = ip_hdr(skb); if (is_erspan_type1(gre_hdr_len)) { ver = 0; - tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, - tpi->flags | TUNNEL_NO_KEY, + __set_bit(IP_TUNNEL_NO_KEY_BIT, flags); + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, flags, iph->saddr, iph->daddr, 0); } else { ershdr = (struct erspan_base_hdr *)(skb->data + gre_hdr_len); ver = ershdr->ver; - tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, - tpi->flags | TUNNEL_KEY, + __set_bit(IP_TUNNEL_KEY_BIT, flags); + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, flags, iph->saddr, iph->daddr, tpi->key); } @@ -307,10 +310,9 @@ static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, struct ip_tunnel_info *info; unsigned char *gh; __be64 tun_id; - __be16 flags; - tpi->flags |= TUNNEL_KEY; - flags = tpi->flags; + __set_bit(IP_TUNNEL_KEY_BIT, tpi->flags); + ip_tunnel_flags_copy(flags, tpi->flags); tun_id = key32_to_tunnel_id(tpi->key); tun_dst = ip_tun_rx_dst(skb, flags, @@ -333,7 +335,8 @@ static int erspan_rcv(struct sk_buff *skb, struct tnl_ptk_info *tpi, ERSPAN_V2_MDSIZE); info = &tun_dst->u.tun_info; - info->key.tun_flags |= TUNNEL_ERSPAN_OPT; + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, + info->key.tun_flags); info->options_len = sizeof(*md); } @@ -376,10 +379,13 @@ static int __ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi, tnl_params = &tunnel->parms.iph; if (tunnel->collect_md || tnl_params->daddr == 0) { - __be16 flags; + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; __be64 tun_id; - flags = tpi->flags & (TUNNEL_CSUM | TUNNEL_KEY); + __set_bit(IP_TUNNEL_CSUM_BIT, flags); + __set_bit(IP_TUNNEL_KEY_BIT, flags); + ip_tunnel_flags_and(flags, tpi->flags, flags); + tun_id = key32_to_tunnel_id(tpi->key); tun_dst = ip_tun_rx_dst(skb, flags, tun_id, 0); if (!tun_dst) @@ -459,12 +465,15 @@ static void __gre_xmit(struct sk_buff *skb, struct net_device *dev, __be16 proto) { struct ip_tunnel *tunnel = netdev_priv(dev); - __be16 flags = tunnel->parms.o_flags; + IP_TUNNEL_DECLARE_FLAGS(flags); + + ip_tunnel_flags_copy(flags, tunnel->parms.o_flags); /* Push GRE header. */ gre_build_header(skb, tunnel->tun_hlen, flags, proto, tunnel->parms.o_key, - (flags & TUNNEL_SEQ) ? htonl(atomic_fetch_inc(&tunnel->o_seqno)) : 0); + test_bit(IP_TUNNEL_SEQ_BIT, flags) ? + htonl(atomic_fetch_inc(&tunnel->o_seqno)) : 0); ip_tunnel_xmit(skb, dev, tnl_params, tnl_params->protocol); } @@ -478,10 +487,10 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev, __be16 proto) { struct ip_tunnel *tunnel = netdev_priv(dev); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct ip_tunnel_info *tun_info; const struct ip_tunnel_key *key; int tunnel_hlen; - __be16 flags; tun_info = skb_tunnel_info(skb); if (unlikely(!tun_info || !(tun_info->mode & IP_TUNNEL_INFO_TX) || @@ -495,14 +504,19 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev, goto err_free_skb; /* Push Tunnel header. */ - if (gre_handle_offloads(skb, !!(tun_info->key.tun_flags & TUNNEL_CSUM))) + if (gre_handle_offloads(skb, test_bit(IP_TUNNEL_CSUM_BIT, + tunnel->parms.o_flags))) goto err_free_skb; - flags = tun_info->key.tun_flags & - (TUNNEL_CSUM | TUNNEL_KEY | TUNNEL_SEQ); + __set_bit(IP_TUNNEL_CSUM_BIT, flags); + __set_bit(IP_TUNNEL_KEY_BIT, flags); + __set_bit(IP_TUNNEL_SEQ_BIT, flags); + ip_tunnel_flags_and(flags, tun_info->key.tun_flags, flags); + gre_build_header(skb, tunnel_hlen, flags, proto, tunnel_id_to_key32(tun_info->key.tun_id), - (flags & TUNNEL_SEQ) ? htonl(atomic_fetch_inc(&tunnel->o_seqno)) : 0); + test_bit(IP_TUNNEL_SEQ_BIT, flags) ? + htonl(atomic_fetch_inc(&tunnel->o_seqno)) : 0); ip_md_tunnel_xmit(skb, dev, IPPROTO_GRE, tunnel_hlen); @@ -516,6 +530,7 @@ err_free_skb: static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct ip_tunnel_info *tun_info; const struct ip_tunnel_key *key; struct erspan_metadata *md; @@ -531,7 +546,7 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) goto err_free_skb; key = &tun_info->key; - if (!(tun_info->key.tun_flags & TUNNEL_ERSPAN_OPT)) + if (!test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, tun_info->key.tun_flags)) goto err_free_skb; if (tun_info->options_len < sizeof(*md)) goto err_free_skb; @@ -584,8 +599,9 @@ static void erspan_fb_xmit(struct sk_buff *skb, struct net_device *dev) goto err_free_skb; } - gre_build_header(skb, 8, TUNNEL_SEQ, - proto, 0, htonl(atomic_fetch_inc(&tunnel->o_seqno))); + __set_bit(IP_TUNNEL_SEQ_BIT, flags); + gre_build_header(skb, 8, flags, proto, 0, + htonl(atomic_fetch_inc(&tunnel->o_seqno))); ip_md_tunnel_xmit(skb, dev, IPPROTO_GRE, tunnel_hlen); @@ -659,7 +675,8 @@ static netdev_tx_t ipgre_xmit(struct sk_buff *skb, tnl_params = &tunnel->parms.iph; } - if (gre_handle_offloads(skb, !!(tunnel->parms.o_flags & TUNNEL_CSUM))) + if (gre_handle_offloads(skb, test_bit(IP_TUNNEL_CSUM_BIT, + tunnel->parms.o_flags))) goto free_skb; __gre_xmit(skb, dev, tnl_params, skb->protocol); @@ -701,7 +718,7 @@ static netdev_tx_t erspan_xmit(struct sk_buff *skb, /* Push ERSPAN header */ if (tunnel->erspan_ver == 0) { proto = htons(ETH_P_ERSPAN); - tunnel->parms.o_flags &= ~TUNNEL_SEQ; + __clear_bit(IP_TUNNEL_SEQ_BIT, tunnel->parms.o_flags); } else if (tunnel->erspan_ver == 1) { erspan_build_header(skb, ntohl(tunnel->parms.o_key), tunnel->index, @@ -716,7 +733,7 @@ static netdev_tx_t erspan_xmit(struct sk_buff *skb, goto free_skb; } - tunnel->parms.o_flags &= ~TUNNEL_KEY; + __clear_bit(IP_TUNNEL_KEY_BIT, tunnel->parms.o_flags); __gre_xmit(skb, dev, &tunnel->parms.iph, proto); return NETDEV_TX_OK; @@ -739,7 +756,8 @@ static netdev_tx_t gre_tap_xmit(struct sk_buff *skb, return NETDEV_TX_OK; } - if (gre_handle_offloads(skb, !!(tunnel->parms.o_flags & TUNNEL_CSUM))) + if (gre_handle_offloads(skb, test_bit(IP_TUNNEL_CSUM_BIT, + tunnel->parms.o_flags))) goto free_skb; if (skb_cow_head(skb, dev->needed_headroom)) @@ -757,7 +775,6 @@ free_skb: static void ipgre_link_update(struct net_device *dev, bool set_mtu) { struct ip_tunnel *tunnel = netdev_priv(dev); - __be16 flags; int len; len = tunnel->tun_hlen; @@ -773,10 +790,9 @@ static void ipgre_link_update(struct net_device *dev, bool set_mtu) if (set_mtu) dev->mtu = max_t(int, dev->mtu - len, 68); - flags = tunnel->parms.o_flags; - - if (flags & TUNNEL_SEQ || - (flags & TUNNEL_CSUM && tunnel->encap.type != TUNNEL_ENCAP_NONE)) { + if (test_bit(IP_TUNNEL_SEQ_BIT, tunnel->parms.o_flags) || + (test_bit(IP_TUNNEL_CSUM_BIT, tunnel->parms.o_flags) && + tunnel->encap.type != TUNNEL_ENCAP_NONE)) { dev->features &= ~NETIF_F_GSO_SOFTWARE; dev->hw_features &= ~NETIF_F_GSO_SOFTWARE; } else { @@ -789,17 +805,25 @@ static int ipgre_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) { + __be16 i_flags, o_flags; int err; + if (!ip_tunnel_flags_is_be16_compat(p->i_flags) || + !ip_tunnel_flags_is_be16_compat(p->o_flags)) + return -EOVERFLOW; + + i_flags = ip_tunnel_flags_to_be16(p->i_flags); + o_flags = ip_tunnel_flags_to_be16(p->o_flags); + if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { if (p->iph.version != 4 || p->iph.protocol != IPPROTO_GRE || p->iph.ihl != 5 || (p->iph.frag_off & htons(~IP_DF)) || - ((p->i_flags | p->o_flags) & (GRE_VERSION | GRE_ROUTING))) + ((i_flags | o_flags) & (GRE_VERSION | GRE_ROUTING))) return -EINVAL; } - p->i_flags = gre_flags_to_tnl_flags(p->i_flags); - p->o_flags = gre_flags_to_tnl_flags(p->o_flags); + gre_flags_to_tnl_flags(p->i_flags, i_flags); + gre_flags_to_tnl_flags(p->o_flags, o_flags); err = ip_tunnel_ctl(dev, p, cmd); if (err) @@ -808,15 +832,18 @@ static int ipgre_tunnel_ctl(struct net_device *dev, if (cmd == SIOCCHGTUNNEL) { struct ip_tunnel *t = netdev_priv(dev); - t->parms.i_flags = p->i_flags; - t->parms.o_flags = p->o_flags; + ip_tunnel_flags_copy(t->parms.i_flags, p->i_flags); + ip_tunnel_flags_copy(t->parms.o_flags, p->o_flags); if (strcmp(dev->rtnl_link_ops->kind, "erspan")) ipgre_link_update(dev, true); } - p->i_flags = gre_tnl_flags_to_gre_flags(p->i_flags); - p->o_flags = gre_tnl_flags_to_gre_flags(p->o_flags); + i_flags = gre_tnl_flags_to_gre_flags(p->i_flags); + ip_tunnel_flags_from_be16(p->i_flags, i_flags); + o_flags = gre_tnl_flags_to_gre_flags(p->o_flags); + ip_tunnel_flags_from_be16(p->o_flags, o_flags); + return 0; } @@ -956,7 +983,6 @@ static void ipgre_tunnel_setup(struct net_device *dev) static void __gre_tunnel_init(struct net_device *dev) { struct ip_tunnel *tunnel; - __be16 flags; tunnel = netdev_priv(dev); tunnel->tun_hlen = gre_calc_hlen(tunnel->parms.o_flags); @@ -968,14 +994,13 @@ static void __gre_tunnel_init(struct net_device *dev) dev->features |= GRE_FEATURES | NETIF_F_LLTX; dev->hw_features |= GRE_FEATURES; - flags = tunnel->parms.o_flags; - /* TCP offload with GRE SEQ is not supported, nor can we support 2 * levels of outer headers requiring an update. */ - if (flags & TUNNEL_SEQ) + if (test_bit(IP_TUNNEL_SEQ_BIT, tunnel->parms.o_flags)) return; - if (flags & TUNNEL_CSUM && tunnel->encap.type != TUNNEL_ENCAP_NONE) + if (test_bit(IP_TUNNEL_CSUM_BIT, tunnel->parms.o_flags) && + tunnel->encap.type != TUNNEL_ENCAP_NONE) return; dev->features |= NETIF_F_GSO_SOFTWARE; @@ -1148,10 +1173,12 @@ static int ipgre_netlink_parms(struct net_device *dev, parms->link = nla_get_u32(data[IFLA_GRE_LINK]); if (data[IFLA_GRE_IFLAGS]) - parms->i_flags = gre_flags_to_tnl_flags(nla_get_be16(data[IFLA_GRE_IFLAGS])); + gre_flags_to_tnl_flags(parms->i_flags, + nla_get_be16(data[IFLA_GRE_IFLAGS])); if (data[IFLA_GRE_OFLAGS]) - parms->o_flags = gre_flags_to_tnl_flags(nla_get_be16(data[IFLA_GRE_OFLAGS])); + gre_flags_to_tnl_flags(parms->o_flags, + nla_get_be16(data[IFLA_GRE_OFLAGS])); if (data[IFLA_GRE_IKEY]) parms->i_key = nla_get_be32(data[IFLA_GRE_IKEY]); @@ -1411,8 +1438,8 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], if (err < 0) return err; - t->parms.i_flags = p.i_flags; - t->parms.o_flags = p.o_flags; + ip_tunnel_flags_copy(t->parms.i_flags, p.i_flags); + ip_tunnel_flags_copy(t->parms.o_flags, p.o_flags); ipgre_link_update(dev, !tb[IFLA_MTU]); @@ -1440,8 +1467,8 @@ static int erspan_changelink(struct net_device *dev, struct nlattr *tb[], if (err < 0) return err; - t->parms.i_flags = p.i_flags; - t->parms.o_flags = p.o_flags; + ip_tunnel_flags_copy(t->parms.i_flags, p.i_flags); + ip_tunnel_flags_copy(t->parms.o_flags, p.o_flags); return 0; } @@ -1498,7 +1525,9 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip_tunnel *t = netdev_priv(dev); struct ip_tunnel_parm_kern *p = &t->parms; - __be16 o_flags = p->o_flags; + IP_TUNNEL_DECLARE_FLAGS(o_flags); + + ip_tunnel_flags_copy(o_flags, p->o_flags); if (nla_put_u32(skb, IFLA_GRE_LINK, p->link) || nla_put_be16(skb, IFLA_GRE_IFLAGS, @@ -1546,7 +1575,7 @@ static int erspan_fill_info(struct sk_buff *skb, const struct net_device *dev) if (t->erspan_ver <= 2) { if (t->erspan_ver != 0 && !t->collect_md) - t->parms.o_flags |= TUNNEL_KEY; + __set_bit(IP_TUNNEL_KEY_BIT, t->parms.o_flags); if (nla_put_u8(skb, IFLA_GRE_ERSPAN_VER, t->erspan_ver)) goto nla_put_failure; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index d10f41f81463..f65170a28106 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -57,16 +57,12 @@ static unsigned int ip_tunnel_hash(__be32 key, __be32 remote) } static bool ip_tunnel_key_match(const struct ip_tunnel_parm_kern *p, - __be16 flags, __be32 key) + const unsigned long *flags, __be32 key) { - if (p->i_flags & TUNNEL_KEY) { - if (flags & TUNNEL_KEY) - return key == p->i_key; - else - /* key expected, none present */ - return false; - } else - return !(flags & TUNNEL_KEY); + if (!test_bit(IP_TUNNEL_KEY_BIT, flags)) + return !test_bit(IP_TUNNEL_KEY_BIT, p->i_flags); + + return test_bit(IP_TUNNEL_KEY_BIT, p->i_flags) && p->i_key == key; } /* Fallback tunnel: no source, no destination, no key, no options @@ -81,7 +77,7 @@ static bool ip_tunnel_key_match(const struct ip_tunnel_parm_kern *p, Given src, dst and key, find appropriate for input tunnel. */ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, - int link, __be16 flags, + int link, const unsigned long *flags, __be32 remote, __be32 local, __be32 key) { @@ -143,7 +139,8 @@ struct ip_tunnel *ip_tunnel_lookup(struct ip_tunnel_net *itn, } hlist_for_each_entry_rcu(t, head, hash_node) { - if ((!(flags & TUNNEL_NO_KEY) && t->parms.i_key != key) || + if ((!test_bit(IP_TUNNEL_NO_KEY_BIT, flags) && + t->parms.i_key != key) || t->parms.iph.saddr != 0 || t->parms.iph.daddr != 0 || !(t->dev->flags & IFF_UP)) @@ -182,7 +179,8 @@ static struct hlist_head *ip_bucket(struct ip_tunnel_net *itn, else remote = 0; - if (!(parms->i_flags & TUNNEL_KEY) && (parms->i_flags & VTI_ISVTI)) + if (!test_bit(IP_TUNNEL_KEY_BIT, parms->i_flags) && + test_bit(IP_TUNNEL_VTI_BIT, parms->i_flags)) i_key = 0; h = ip_tunnel_hash(i_key, remote); @@ -211,12 +209,14 @@ static struct ip_tunnel *ip_tunnel_find(struct ip_tunnel_net *itn, { __be32 remote = parms->iph.daddr; __be32 local = parms->iph.saddr; + IP_TUNNEL_DECLARE_FLAGS(flags); __be32 key = parms->i_key; - __be16 flags = parms->i_flags; int link = parms->link; struct ip_tunnel *t = NULL; struct hlist_head *head = ip_bucket(itn, parms); + ip_tunnel_flags_copy(flags, parms->i_flags); + hlist_for_each_entry_rcu(t, head, hash_node) { if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr && @@ -386,15 +386,15 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, } #endif - if ((!(tpi->flags&TUNNEL_CSUM) && (tunnel->parms.i_flags&TUNNEL_CSUM)) || - ((tpi->flags&TUNNEL_CSUM) && !(tunnel->parms.i_flags&TUNNEL_CSUM))) { + if (test_bit(IP_TUNNEL_CSUM_BIT, tunnel->parms.i_flags) != + test_bit(IP_TUNNEL_CSUM_BIT, tpi->flags)) { DEV_STATS_INC(tunnel->dev, rx_crc_errors); DEV_STATS_INC(tunnel->dev, rx_errors); goto drop; } - if (tunnel->parms.i_flags&TUNNEL_SEQ) { - if (!(tpi->flags&TUNNEL_SEQ) || + if (test_bit(IP_TUNNEL_SEQ_BIT, tunnel->parms.i_flags)) { + if (!test_bit(IP_TUNNEL_SEQ_BIT, tpi->flags) || (tunnel->i_seqno && (s32)(ntohl(tpi->seq) - tunnel->i_seqno) < 0)) { DEV_STATS_INC(tunnel->dev, rx_fifo_errors); DEV_STATS_INC(tunnel->dev, rx_errors); @@ -638,7 +638,7 @@ void ip_md_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, goto tx_error; } - if (key->tun_flags & TUNNEL_DONT_FRAGMENT) + if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, key->tun_flags)) df = htons(IP_DF); if (tnl_update_pmtu(dev, skb, rt, df, inner_iph, tunnel_hlen, key->u.ipv4.dst, true)) { @@ -928,10 +928,10 @@ int ip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, goto done; if (p->iph.ttl) p->iph.frag_off |= htons(IP_DF); - if (!(p->i_flags & VTI_ISVTI)) { - if (!(p->i_flags & TUNNEL_KEY)) + if (!test_bit(IP_TUNNEL_VTI_BIT, p->i_flags)) { + if (!test_bit(IP_TUNNEL_KEY_BIT, p->i_flags)) p->i_key = 0; - if (!(p->o_flags & TUNNEL_KEY)) + if (!test_bit(IP_TUNNEL_KEY_BIT, p->o_flags)) p->o_key = 0; } @@ -1016,8 +1016,8 @@ bool ip_tunnel_parm_from_user(struct ip_tunnel_parm_kern *kp, strscpy(kp->name, p.name); kp->link = p.link; - kp->i_flags = p.i_flags; - kp->o_flags = p.o_flags; + ip_tunnel_flags_from_be16(kp->i_flags, p.i_flags); + ip_tunnel_flags_from_be16(kp->o_flags, p.o_flags); kp->i_key = p.i_key; kp->o_key = p.o_key; memcpy(&kp->iph, &p.iph, min(sizeof(kp->iph), sizeof(p.iph))); @@ -1030,10 +1030,14 @@ bool ip_tunnel_parm_to_user(void __user *data, struct ip_tunnel_parm_kern *kp) { struct ip_tunnel_parm p; + if (!ip_tunnel_flags_is_be16_compat(kp->i_flags) || + !ip_tunnel_flags_is_be16_compat(kp->o_flags)) + return false; + strscpy(p.name, kp->name); p.link = kp->link; - p.i_flags = kp->i_flags; - p.o_flags = kp->o_flags; + p.i_flags = ip_tunnel_flags_to_be16(kp->i_flags); + p.o_flags = ip_tunnel_flags_to_be16(kp->o_flags); p.i_key = kp->i_key; p.o_key = kp->o_key; memcpy(&p.iph, &kp->iph, min(sizeof(p.iph), sizeof(kp->iph))); diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 98f136b07191..a3676155be78 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -125,6 +125,7 @@ EXPORT_SYMBOL_GPL(__iptunnel_pull_header); struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, gfp_t flags) { + IP_TUNNEL_DECLARE_FLAGS(tun_flags) = { }; struct metadata_dst *res; struct ip_tunnel_info *dst, *src; @@ -144,10 +145,10 @@ struct metadata_dst *iptunnel_metadata_reply(struct metadata_dst *md, sizeof(struct in6_addr)); else dst->key.u.ipv4.dst = src->key.u.ipv4.src; - dst->key.tun_flags = src->key.tun_flags; + ip_tunnel_flags_copy(dst->key.tun_flags, src->key.tun_flags); dst->mode = src->mode | IP_TUNNEL_INFO_TX; ip_tunnel_info_opts_set(dst, ip_tunnel_info_opts(src), - src->options_len, 0); + src->options_len, tun_flags); return res; } @@ -497,7 +498,7 @@ static int ip_tun_parse_opts_geneve(struct nlattr *attr, opt->opt_class = nla_get_be16(attr); attr = tb[LWTUNNEL_IP_OPT_GENEVE_TYPE]; opt->type = nla_get_u8(attr); - info->key.tun_flags |= TUNNEL_GENEVE_OPT; + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, info->key.tun_flags); } return sizeof(struct geneve_opt) + data_len; @@ -525,7 +526,7 @@ static int ip_tun_parse_opts_vxlan(struct nlattr *attr, attr = tb[LWTUNNEL_IP_OPT_VXLAN_GBP]; md->gbp = nla_get_u32(attr); md->gbp &= VXLAN_GBP_MASK; - info->key.tun_flags |= TUNNEL_VXLAN_OPT; + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, info->key.tun_flags); } return sizeof(struct vxlan_metadata); @@ -574,7 +575,7 @@ static int ip_tun_parse_opts_erspan(struct nlattr *attr, set_hwid(&md->u.md2, nla_get_u8(attr)); } - info->key.tun_flags |= TUNNEL_ERSPAN_OPT; + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, info->key.tun_flags); } return sizeof(struct erspan_metadata); @@ -585,7 +586,7 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, { int err, rem, opt_len, opts_len = 0; struct nlattr *nla; - __be16 type = 0; + u32 type = 0; if (!attr) return 0; @@ -598,7 +599,7 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, nla_for_each_attr(nla, nla_data(attr), nla_len(attr), rem) { switch (nla_type(nla)) { case LWTUNNEL_IP_OPTS_GENEVE: - if (type && type != TUNNEL_GENEVE_OPT) + if (type && type != IP_TUNNEL_GENEVE_OPT_BIT) return -EINVAL; opt_len = ip_tun_parse_opts_geneve(nla, info, opts_len, extack); @@ -607,7 +608,7 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, opts_len += opt_len; if (opts_len > IP_TUNNEL_OPTS_MAX) return -EINVAL; - type = TUNNEL_GENEVE_OPT; + type = IP_TUNNEL_GENEVE_OPT_BIT; break; case LWTUNNEL_IP_OPTS_VXLAN: if (type) @@ -617,7 +618,7 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, if (opt_len < 0) return opt_len; opts_len += opt_len; - type = TUNNEL_VXLAN_OPT; + type = IP_TUNNEL_VXLAN_OPT_BIT; break; case LWTUNNEL_IP_OPTS_ERSPAN: if (type) @@ -627,7 +628,7 @@ static int ip_tun_parse_opts(struct nlattr *attr, struct ip_tunnel_info *info, if (opt_len < 0) return opt_len; opts_len += opt_len; - type = TUNNEL_ERSPAN_OPT; + type = IP_TUNNEL_ERSPAN_OPT_BIT; break; default: return -EINVAL; @@ -705,10 +706,16 @@ static int ip_tun_build_state(struct net *net, struct nlattr *attr, if (tb[LWTUNNEL_IP_TOS]) tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]); - if (tb[LWTUNNEL_IP_FLAGS]) - tun_info->key.tun_flags |= - (nla_get_be16(tb[LWTUNNEL_IP_FLAGS]) & - ~TUNNEL_OPTIONS_PRESENT); + if (tb[LWTUNNEL_IP_FLAGS]) { + IP_TUNNEL_DECLARE_FLAGS(flags); + + ip_tunnel_flags_from_be16(flags, + nla_get_be16(tb[LWTUNNEL_IP_FLAGS])); + ip_tunnel_clear_options_present(flags); + + ip_tunnel_flags_or(tun_info->key.tun_flags, + tun_info->key.tun_flags, flags); + } tun_info->mode = IP_TUNNEL_INFO_TX; tun_info->options_len = opt_len; @@ -812,18 +819,18 @@ static int ip_tun_fill_encap_opts(struct sk_buff *skb, int type, struct nlattr *nest; int err = 0; - if (!(tun_info->key.tun_flags & TUNNEL_OPTIONS_PRESENT)) + if (!ip_tunnel_is_options_present(tun_info->key.tun_flags)) return 0; nest = nla_nest_start_noflag(skb, type); if (!nest) return -ENOMEM; - if (tun_info->key.tun_flags & TUNNEL_GENEVE_OPT) + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, tun_info->key.tun_flags)) err = ip_tun_fill_encap_opts_geneve(skb, tun_info); - else if (tun_info->key.tun_flags & TUNNEL_VXLAN_OPT) + else if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, tun_info->key.tun_flags)) err = ip_tun_fill_encap_opts_vxlan(skb, tun_info); - else if (tun_info->key.tun_flags & TUNNEL_ERSPAN_OPT) + else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, tun_info->key.tun_flags)) err = ip_tun_fill_encap_opts_erspan(skb, tun_info); if (err) { @@ -846,7 +853,8 @@ static int ip_tun_fill_encap_info(struct sk_buff *skb, nla_put_in_addr(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) || nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) || nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) || - nla_put_be16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags) || + nla_put_be16(skb, LWTUNNEL_IP_FLAGS, + ip_tunnel_flags_to_be16(tun_info->key.tun_flags)) || ip_tun_fill_encap_opts(skb, LWTUNNEL_IP_OPTS, tun_info)) return -ENOMEM; @@ -857,11 +865,11 @@ static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) { int opt_len; - if (!(info->key.tun_flags & TUNNEL_OPTIONS_PRESENT)) + if (!ip_tunnel_is_options_present(info->key.tun_flags)) return 0; opt_len = nla_total_size(0); /* LWTUNNEL_IP_OPTS */ - if (info->key.tun_flags & TUNNEL_GENEVE_OPT) { + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, info->key.tun_flags)) { struct geneve_opt *opt; int offset = 0; @@ -874,10 +882,10 @@ static int ip_tun_opts_nlsize(struct ip_tunnel_info *info) /* OPT_GENEVE_DATA */ offset += sizeof(*opt) + opt->length * 4; } - } else if (info->key.tun_flags & TUNNEL_VXLAN_OPT) { + } else if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, info->key.tun_flags)) { opt_len += nla_total_size(0) /* LWTUNNEL_IP_OPTS_VXLAN */ + nla_total_size(4); /* OPT_VXLAN_GBP */ - } else if (info->key.tun_flags & TUNNEL_ERSPAN_OPT) { + } else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, info->key.tun_flags)) { struct erspan_metadata *md = ip_tunnel_info_opts(info); opt_len += nla_total_size(0) /* LWTUNNEL_IP_OPTS_ERSPAN */ @@ -984,10 +992,17 @@ static int ip6_tun_build_state(struct net *net, struct nlattr *attr, if (tb[LWTUNNEL_IP6_TC]) tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP6_TC]); - if (tb[LWTUNNEL_IP6_FLAGS]) - tun_info->key.tun_flags |= - (nla_get_be16(tb[LWTUNNEL_IP6_FLAGS]) & - ~TUNNEL_OPTIONS_PRESENT); + if (tb[LWTUNNEL_IP6_FLAGS]) { + IP_TUNNEL_DECLARE_FLAGS(flags); + __be16 data; + + data = nla_get_be16(tb[LWTUNNEL_IP6_FLAGS]); + ip_tunnel_flags_from_be16(flags, data); + ip_tunnel_clear_options_present(flags); + + ip_tunnel_flags_or(tun_info->key.tun_flags, + tun_info->key.tun_flags, flags); + } tun_info->mode = IP_TUNNEL_INFO_TX | IP_TUNNEL_INFO_IPV6; tun_info->options_len = opt_len; @@ -1008,7 +1023,8 @@ static int ip6_tun_fill_encap_info(struct sk_buff *skb, nla_put_in6_addr(skb, LWTUNNEL_IP6_SRC, &tun_info->key.u.ipv6.src) || nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.tos) || nla_put_u8(skb, LWTUNNEL_IP6_HOPLIMIT, tun_info->key.ttl) || - nla_put_be16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags) || + nla_put_be16(skb, LWTUNNEL_IP6_FLAGS, + ip_tunnel_flags_to_be16(tun_info->key.tun_flags)) || ip_tun_fill_encap_opts(skb, LWTUNNEL_IP6_OPTS, tun_info)) return -ENOMEM; @@ -1139,8 +1155,12 @@ void ip_tunnel_netlink_parms(struct nlattr *data[], if (!data[IFLA_IPTUN_PMTUDISC] || nla_get_u8(data[IFLA_IPTUN_PMTUDISC])) parms->iph.frag_off = htons(IP_DF); - if (data[IFLA_IPTUN_FLAGS]) - parms->i_flags = nla_get_be16(data[IFLA_IPTUN_FLAGS]); + if (data[IFLA_IPTUN_FLAGS]) { + __be16 flags; + + flags = nla_get_be16(data[IFLA_IPTUN_FLAGS]); + ip_tunnel_flags_from_be16(parms->i_flags, flags); + } if (data[IFLA_IPTUN_PROTO]) parms->iph.protocol = nla_get_u8(data[IFLA_IPTUN_PROTO]); diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 7e595f619615..14536da9f5dc 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -51,8 +51,11 @@ static int vti_input(struct sk_buff *skb, int nexthdr, __be32 spi, const struct iphdr *iph = ip_hdr(skb); struct net *net = dev_net(skb->dev); struct ip_tunnel_net *itn = net_generic(net, vti_net_id); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; - tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY, + __set_bit(IP_TUNNEL_NO_KEY_BIT, flags); + + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, flags, iph->saddr, iph->daddr, 0); if (tunnel) { if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) @@ -322,8 +325,11 @@ static int vti4_err(struct sk_buff *skb, u32 info) const struct iphdr *iph = (const struct iphdr *)skb->data; int protocol = iph->protocol; struct ip_tunnel_net *itn = net_generic(net, vti_net_id); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; + + __set_bit(IP_TUNNEL_NO_KEY_BIT, flags); - tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY, + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, flags, iph->daddr, iph->saddr, 0); if (!tunnel) return -1; @@ -375,6 +381,7 @@ static int vti4_err(struct sk_buff *skb, u32 info) static int vti_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) { + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; int err = 0; if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { @@ -383,20 +390,26 @@ vti_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) return -EINVAL; } - if (!(p->i_flags & GRE_KEY)) + if (!ip_tunnel_flags_is_be16_compat(p->i_flags) || + !ip_tunnel_flags_is_be16_compat(p->o_flags)) + return -EOVERFLOW; + + if (!(ip_tunnel_flags_to_be16(p->i_flags) & GRE_KEY)) p->i_key = 0; - if (!(p->o_flags & GRE_KEY)) + if (!(ip_tunnel_flags_to_be16(p->o_flags) & GRE_KEY)) p->o_key = 0; - p->i_flags = VTI_ISVTI; + __set_bit(IP_TUNNEL_VTI_BIT, flags); + ip_tunnel_flags_copy(p->i_flags, flags); err = ip_tunnel_ctl(dev, p, cmd); if (err) return err; if (cmd != SIOCDELTUNNEL) { - p->i_flags |= GRE_KEY; - p->o_flags |= GRE_KEY; + ip_tunnel_flags_from_be16(flags, GRE_KEY); + ip_tunnel_flags_or(p->i_flags, p->i_flags, flags); + ip_tunnel_flags_or(p->o_flags, p->o_flags, flags); } return 0; } @@ -541,7 +554,7 @@ static void vti_netlink_parms(struct nlattr *data[], if (!data) return; - parms->i_flags = VTI_ISVTI; + __set_bit(IP_TUNNEL_VTI_BIT, parms->i_flags); if (data[IFLA_VTI_LINK]) parms->link = nla_get_u32(data[IFLA_VTI_LINK]); diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index db960cd08a3b..923a2ef68c2f 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -130,13 +130,16 @@ static int ipip_err(struct sk_buff *skb, u32 info) struct net *net = dev_net(skb->dev); struct ip_tunnel_net *itn = net_generic(net, ipip_net_id); const struct iphdr *iph = (const struct iphdr *)skb->data; + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; struct ip_tunnel *t; int err = 0; - t = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY, - iph->daddr, iph->saddr, 0); + __set_bit(IP_TUNNEL_NO_KEY_BIT, flags); + + t = ip_tunnel_lookup(itn, skb->dev->ifindex, flags, iph->daddr, + iph->saddr, 0); if (!t) { err = -ENOENT; goto out; @@ -213,13 +216,16 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto) { struct net *net = dev_net(skb->dev); struct ip_tunnel_net *itn = net_generic(net, ipip_net_id); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct metadata_dst *tun_dst = NULL; struct ip_tunnel *tunnel; const struct iphdr *iph; + __set_bit(IP_TUNNEL_NO_KEY_BIT, flags); + iph = ip_hdr(skb); - tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY, - iph->saddr, iph->daddr, 0); + tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, flags, iph->saddr, + iph->daddr, 0); if (tunnel) { const struct tnl_ptk_info *tpi; @@ -238,7 +244,9 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto) if (iptunnel_pull_header(skb, 0, tpi->proto, false)) goto drop; if (tunnel->collect_md) { - tun_dst = ip_tun_rx_dst(skb, 0, 0, 0); + ip_tunnel_flags_zero(flags); + + tun_dst = ip_tun_rx_dst(skb, flags, 0, 0); if (!tun_dst) return 0; ip_tunnel_md_udp_encap(skb, &tun_dst->u.tun_info); @@ -340,7 +348,8 @@ ipip_tunnel_ctl(struct net_device *dev, struct ip_tunnel_parm_kern *p, int cmd) } p->i_key = p->o_key = 0; - p->i_flags = p->o_flags = 0; + ip_tunnel_flags_zero(p->i_flags); + ip_tunnel_flags_zero(p->o_flags); return ip_tunnel_ctl(dev, p, cmd); } diff --git a/net/ipv4/udp_tunnel_core.c b/net/ipv4/udp_tunnel_core.c index 860aff5f8599..e4e0fa869fa4 100644 --- a/net/ipv4/udp_tunnel_core.c +++ b/net/ipv4/udp_tunnel_core.c @@ -183,7 +183,8 @@ void udp_tunnel_sock_release(struct socket *sock) EXPORT_SYMBOL_GPL(udp_tunnel_sock_release); struct metadata_dst *udp_tun_rx_dst(struct sk_buff *skb, unsigned short family, - __be16 flags, __be64 tunnel_id, int md_size) + const unsigned long *flags, + __be64 tunnel_id, int md_size) { struct metadata_dst *tun_dst; struct ip_tunnel_info *info; @@ -199,7 +200,7 @@ struct metadata_dst *udp_tun_rx_dst(struct sk_buff *skb, unsigned short family, info->key.tp_src = udp_hdr(skb)->source; info->key.tp_dst = udp_hdr(skb)->dest; if (udp_hdr(skb)->check) - info->key.tun_flags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags); return tun_dst; } EXPORT_SYMBOL_GPL(udp_tun_rx_dst); diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index ca7e77e84283..2146cf69d8e2 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -496,11 +496,11 @@ static int ip6gre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi) tpi->proto); if (tunnel) { if (tunnel->parms.collect_md) { + IP_TUNNEL_DECLARE_FLAGS(flags); struct metadata_dst *tun_dst; __be64 tun_id; - __be16 flags; - flags = tpi->flags; + ip_tunnel_flags_copy(flags, tpi->flags); tun_id = key32_to_tunnel_id(tpi->key); tun_dst = ipv6_tun_rx_dst(skb, flags, tun_id, 0); @@ -548,14 +548,14 @@ static int ip6erspan_rcv(struct sk_buff *skb, if (tunnel->parms.collect_md) { struct erspan_metadata *pkt_md, *md; + IP_TUNNEL_DECLARE_FLAGS(flags); struct metadata_dst *tun_dst; struct ip_tunnel_info *info; unsigned char *gh; __be64 tun_id; - __be16 flags; - tpi->flags |= TUNNEL_KEY; - flags = tpi->flags; + __set_bit(IP_TUNNEL_KEY_BIT, tpi->flags); + ip_tunnel_flags_copy(flags, tpi->flags); tun_id = key32_to_tunnel_id(tpi->key); tun_dst = ipv6_tun_rx_dst(skb, flags, tun_id, @@ -577,7 +577,8 @@ static int ip6erspan_rcv(struct sk_buff *skb, md2 = &md->u.md2; memcpy(md2, pkt_md, ver == 1 ? ERSPAN_V1_MDSIZE : ERSPAN_V2_MDSIZE); - info->key.tun_flags |= TUNNEL_ERSPAN_OPT; + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, + info->key.tun_flags); info->options_len = sizeof(*md); ip6_tnl_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error); @@ -745,8 +746,8 @@ static netdev_tx_t __gre6_xmit(struct sk_buff *skb, __u32 *pmtu, __be16 proto) { struct ip6_tnl *tunnel = netdev_priv(dev); + IP_TUNNEL_DECLARE_FLAGS(flags); __be16 protocol; - __be16 flags; if (dev->type == ARPHRD_ETHER) IPCB(skb)->flags = 0; @@ -778,8 +779,11 @@ static netdev_tx_t __gre6_xmit(struct sk_buff *skb, fl6->fl6_gre_key = tunnel_id_to_key32(key->tun_id); dsfield = key->tos; - flags = key->tun_flags & - (TUNNEL_CSUM | TUNNEL_KEY | TUNNEL_SEQ); + ip_tunnel_flags_zero(flags); + __set_bit(IP_TUNNEL_CSUM_BIT, flags); + __set_bit(IP_TUNNEL_KEY_BIT, flags); + __set_bit(IP_TUNNEL_SEQ_BIT, flags); + ip_tunnel_flags_and(flags, flags, key->tun_flags); tun_hlen = gre_calc_hlen(flags); if (skb_cow_head(skb, dev->needed_headroom ?: tun_hlen + tunnel->encap_hlen)) @@ -788,19 +792,21 @@ static netdev_tx_t __gre6_xmit(struct sk_buff *skb, gre_build_header(skb, tun_hlen, flags, protocol, tunnel_id_to_key32(tun_info->key.tun_id), - (flags & TUNNEL_SEQ) ? htonl(atomic_fetch_inc(&tunnel->o_seqno)) - : 0); + test_bit(IP_TUNNEL_SEQ_BIT, flags) ? + htonl(atomic_fetch_inc(&tunnel->o_seqno)) : + 0); } else { if (skb_cow_head(skb, dev->needed_headroom ?: tunnel->hlen)) return -ENOMEM; - flags = tunnel->parms.o_flags; + ip_tunnel_flags_copy(flags, tunnel->parms.o_flags); gre_build_header(skb, tunnel->tun_hlen, flags, protocol, tunnel->parms.o_key, - (flags & TUNNEL_SEQ) ? htonl(atomic_fetch_inc(&tunnel->o_seqno)) - : 0); + test_bit(IP_TUNNEL_SEQ_BIT, flags) ? + htonl(atomic_fetch_inc(&tunnel->o_seqno)) : + 0); } return ip6_tnl_xmit(skb, dev, dsfield, fl6, encap_limit, pmtu, @@ -822,7 +828,8 @@ static inline int ip6gre_xmit_ipv4(struct sk_buff *skb, struct net_device *dev) prepare_ip6gre_xmit_ipv4(skb, dev, &fl6, &dsfield, &encap_limit); - err = gre_handle_offloads(skb, !!(t->parms.o_flags & TUNNEL_CSUM)); + err = gre_handle_offloads(skb, test_bit(IP_TUNNEL_CSUM_BIT, + t->parms.o_flags)); if (err) return -1; @@ -856,7 +863,8 @@ static inline int ip6gre_xmit_ipv6(struct sk_buff *skb, struct net_device *dev) prepare_ip6gre_xmit_ipv6(skb, dev, &fl6, &dsfield, &encap_limit)) return -1; - if (gre_handle_offloads(skb, !!(t->parms.o_flags & TUNNEL_CSUM))) + if (gre_handle_offloads(skb, test_bit(IP_TUNNEL_CSUM_BIT, + t->parms.o_flags))) return -1; err = __gre6_xmit(skb, dev, dsfield, &fl6, encap_limit, @@ -883,7 +891,8 @@ static int ip6gre_xmit_other(struct sk_buff *skb, struct net_device *dev) prepare_ip6gre_xmit_other(skb, dev, &fl6, &dsfield, &encap_limit)) return -1; - err = gre_handle_offloads(skb, !!(t->parms.o_flags & TUNNEL_CSUM)); + err = gre_handle_offloads(skb, test_bit(IP_TUNNEL_CSUM_BIT, + t->parms.o_flags)); if (err) return err; err = __gre6_xmit(skb, dev, dsfield, &fl6, encap_limit, &mtu, skb->protocol); @@ -936,6 +945,7 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, struct ip_tunnel_info *tun_info = NULL; struct ip6_tnl *t = netdev_priv(dev); struct dst_entry *dst = skb_dst(skb); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; bool truncate = false; int encap_limit = -1; __u8 dsfield = false; @@ -979,7 +989,7 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, if (skb_cow_head(skb, dev->needed_headroom ?: t->hlen)) goto tx_err; - t->parms.o_flags &= ~TUNNEL_KEY; + __clear_bit(IP_TUNNEL_KEY_BIT, t->parms.o_flags); IPCB(skb)->flags = 0; /* For collect_md mode, derive fl6 from the tunnel key, @@ -1004,7 +1014,8 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, fl6.fl6_gre_key = tunnel_id_to_key32(key->tun_id); dsfield = key->tos; - if (!(tun_info->key.tun_flags & TUNNEL_ERSPAN_OPT)) + if (!test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, + tun_info->key.tun_flags)) goto tx_err; if (tun_info->options_len < sizeof(*md)) goto tx_err; @@ -1065,7 +1076,9 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb, } /* Push GRE header. */ - gre_build_header(skb, 8, TUNNEL_SEQ, proto, 0, htonl(atomic_fetch_inc(&t->o_seqno))); + __set_bit(IP_TUNNEL_SEQ_BIT, flags); + gre_build_header(skb, 8, flags, proto, 0, + htonl(atomic_fetch_inc(&t->o_seqno))); /* TooBig packet may have updated dst->dev's mtu */ if (!t->parms.collect_md && dst && dst_mtu(dst) > dst->dev->mtu) @@ -1208,8 +1221,8 @@ static void ip6gre_tnl_copy_tnl_parm(struct ip6_tnl *t, t->parms.proto = p->proto; t->parms.i_key = p->i_key; t->parms.o_key = p->o_key; - t->parms.i_flags = p->i_flags; - t->parms.o_flags = p->o_flags; + ip_tunnel_flags_copy(t->parms.i_flags, p->i_flags); + ip_tunnel_flags_copy(t->parms.o_flags, p->o_flags); t->parms.fwmark = p->fwmark; t->parms.erspan_ver = p->erspan_ver; t->parms.index = p->index; @@ -1238,8 +1251,8 @@ static void ip6gre_tnl_parm_from_user(struct __ip6_tnl_parm *p, p->link = u->link; p->i_key = u->i_key; p->o_key = u->o_key; - p->i_flags = gre_flags_to_tnl_flags(u->i_flags); - p->o_flags = gre_flags_to_tnl_flags(u->o_flags); + gre_flags_to_tnl_flags(p->i_flags, u->i_flags); + gre_flags_to_tnl_flags(p->o_flags, u->o_flags); memcpy(p->name, u->name, sizeof(u->name)); } @@ -1391,7 +1404,7 @@ static int ip6gre_header(struct sk_buff *skb, struct net_device *dev, ipv6h->daddr = t->parms.raddr; p = (__be16 *)(ipv6h + 1); - p[0] = t->parms.o_flags; + p[0] = ip_tunnel_flags_to_be16(t->parms.o_flags); p[1] = htons(type); /* @@ -1455,19 +1468,17 @@ static void ip6gre_tunnel_setup(struct net_device *dev) static void ip6gre_tnl_init_features(struct net_device *dev) { struct ip6_tnl *nt = netdev_priv(dev); - __be16 flags; dev->features |= GRE6_FEATURES | NETIF_F_LLTX; dev->hw_features |= GRE6_FEATURES; - flags = nt->parms.o_flags; - /* TCP offload with GRE SEQ is not supported, nor can we support 2 * levels of outer headers requiring an update. */ - if (flags & TUNNEL_SEQ) + if (test_bit(IP_TUNNEL_SEQ_BIT, nt->parms.o_flags)) return; - if (flags & TUNNEL_CSUM && nt->encap.type != TUNNEL_ENCAP_NONE) + if (test_bit(IP_TUNNEL_CSUM_BIT, nt->parms.o_flags) && + nt->encap.type != TUNNEL_ENCAP_NONE) return; dev->features |= NETIF_F_GSO_SOFTWARE; @@ -1792,12 +1803,12 @@ static void ip6gre_netlink_parms(struct nlattr *data[], parms->link = nla_get_u32(data[IFLA_GRE_LINK]); if (data[IFLA_GRE_IFLAGS]) - parms->i_flags = gre_flags_to_tnl_flags( - nla_get_be16(data[IFLA_GRE_IFLAGS])); + gre_flags_to_tnl_flags(parms->i_flags, + nla_get_be16(data[IFLA_GRE_IFLAGS])); if (data[IFLA_GRE_OFLAGS]) - parms->o_flags = gre_flags_to_tnl_flags( - nla_get_be16(data[IFLA_GRE_OFLAGS])); + gre_flags_to_tnl_flags(parms->o_flags, + nla_get_be16(data[IFLA_GRE_OFLAGS])); if (data[IFLA_GRE_IKEY]) parms->i_key = nla_get_be32(data[IFLA_GRE_IKEY]); @@ -2144,11 +2155,13 @@ static int ip6gre_fill_info(struct sk_buff *skb, const struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); struct __ip6_tnl_parm *p = &t->parms; - __be16 o_flags = p->o_flags; + IP_TUNNEL_DECLARE_FLAGS(o_flags); + + ip_tunnel_flags_copy(o_flags, p->o_flags); if (p->erspan_ver == 1 || p->erspan_ver == 2) { if (!p->collect_md) - o_flags |= TUNNEL_KEY; + __set_bit(IP_TUNNEL_KEY_BIT, o_flags); if (nla_put_u8(skb, IFLA_GRE_ERSPAN_VER, p->erspan_ver)) goto nla_put_failure; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index e9cc315832cb..57bb3b3ea0c5 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -798,17 +798,15 @@ static int __ip6_tnl_rcv(struct ip6_tnl *tunnel, struct sk_buff *skb, const struct ipv6hdr *ipv6h; int nh, err; - if ((!(tpi->flags & TUNNEL_CSUM) && - (tunnel->parms.i_flags & TUNNEL_CSUM)) || - ((tpi->flags & TUNNEL_CSUM) && - !(tunnel->parms.i_flags & TUNNEL_CSUM))) { + if (test_bit(IP_TUNNEL_CSUM_BIT, tunnel->parms.i_flags) != + test_bit(IP_TUNNEL_CSUM_BIT, tpi->flags)) { DEV_STATS_INC(tunnel->dev, rx_crc_errors); DEV_STATS_INC(tunnel->dev, rx_errors); goto drop; } - if (tunnel->parms.i_flags & TUNNEL_SEQ) { - if (!(tpi->flags & TUNNEL_SEQ) || + if (test_bit(IP_TUNNEL_SEQ_BIT, tunnel->parms.i_flags)) { + if (!test_bit(IP_TUNNEL_SEQ_BIT, tpi->flags) || (tunnel->i_seqno && (s32)(ntohl(tpi->seq) - tunnel->i_seqno) < 0)) { DEV_STATS_INC(tunnel->dev, rx_fifo_errors); @@ -946,7 +944,9 @@ static int ipxip6_rcv(struct sk_buff *skb, u8 ipproto, if (iptunnel_pull_header(skb, 0, tpi->proto, false)) goto drop; if (t->parms.collect_md) { - tun_dst = ipv6_tun_rx_dst(skb, 0, 0, 0); + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; + + tun_dst = ipv6_tun_rx_dst(skb, flags, 0, 0); if (!tun_dst) goto drop; } diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 894e6a4c4be3..83b195f09561 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -207,7 +207,7 @@ static int ipip6_tunnel_create(struct net_device *dev) __dev_addr_set(dev, &t->parms.iph.saddr, 4); memcpy(dev->broadcast, &t->parms.iph.daddr, 4); - if ((__force u16)t->parms.i_flags & SIT_ISATAP) + if (test_bit(IP_TUNNEL_SIT_ISATAP_BIT, t->parms.i_flags)) dev->priv_flags |= IFF_ISATAP; dev->rtnl_link_ops = &sit_link_ops; @@ -1700,7 +1700,8 @@ static int ipip6_fill_info(struct sk_buff *skb, const struct net_device *dev) nla_put_u8(skb, IFLA_IPTUN_PMTUDISC, !!(parm->iph.frag_off & htons(IP_DF))) || nla_put_u8(skb, IFLA_IPTUN_PROTO, parm->iph.protocol) || - nla_put_be16(skb, IFLA_IPTUN_FLAGS, parm->i_flags) || + nla_put_be16(skb, IFLA_IPTUN_FLAGS, + ip_tunnel_flags_to_be16(parm->i_flags)) || nla_put_u32(skb, IFLA_IPTUN_FWMARK, tunnel->fwmark)) goto nla_put_failure; diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index a2c16b501087..c7a8a08b7308 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1550,6 +1550,7 @@ static int ipvs_gre_decap(struct netns_ipvs *ipvs, struct sk_buff *skb, if (!dest) goto unk; if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GRE) { + IP_TUNNEL_DECLARE_FLAGS(flags); __be16 type; /* Only support version 0 and C (csum) */ @@ -1560,7 +1561,10 @@ static int ipvs_gre_decap(struct netns_ipvs *ipvs, struct sk_buff *skb, if (type != htons(ETH_P_IP)) goto unk; *proto = IPPROTO_IPIP; - return gre_calc_hlen(gre_flags_to_tnl_flags(greh->flags)); + + gre_flags_to_tnl_flags(flags, greh->flags); + + return gre_calc_hlen(flags); } unk: diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 65e0259178da..39b5fd6bbf65 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -390,10 +390,10 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, skb->ip_summed == CHECKSUM_PARTIAL) mtu -= GUE_PLEN_REMCSUM + GUE_LEN_PRIV; } else if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GRE) { - __be16 tflags = 0; + IP_TUNNEL_DECLARE_FLAGS(tflags) = { }; if (dest->tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) - tflags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, tflags); mtu -= gre_calc_hlen(tflags); } if (mtu < 68) { @@ -553,10 +553,10 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, skb->ip_summed == CHECKSUM_PARTIAL) mtu -= GUE_PLEN_REMCSUM + GUE_LEN_PRIV; } else if (dest->tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GRE) { - __be16 tflags = 0; + IP_TUNNEL_DECLARE_FLAGS(tflags) = { }; if (dest->tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) - tflags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, tflags); mtu -= gre_calc_hlen(tflags); } if (mtu < IPV6_MIN_MTU) { @@ -1082,11 +1082,11 @@ ipvs_gre_encap(struct net *net, struct sk_buff *skb, { __be16 proto = *next_protocol == IPPROTO_IPIP ? htons(ETH_P_IP) : htons(ETH_P_IPV6); - __be16 tflags = 0; + IP_TUNNEL_DECLARE_FLAGS(tflags) = { }; size_t hdrlen; if (cp->dest->tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) - tflags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, tflags); hdrlen = gre_calc_hlen(tflags); gre_build_header(skb, hdrlen, tflags, proto, 0, 0); @@ -1165,11 +1165,11 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, max_headroom += sizeof(struct udphdr) + gue_hdrlen; } else if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GRE) { + IP_TUNNEL_DECLARE_FLAGS(tflags) = { }; size_t gre_hdrlen; - __be16 tflags = 0; if (tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) - tflags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, tflags); gre_hdrlen = gre_calc_hlen(tflags); max_headroom += gre_hdrlen; @@ -1310,11 +1310,11 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, max_headroom += sizeof(struct udphdr) + gue_hdrlen; } else if (tun_type == IP_VS_CONN_F_TUNNEL_TYPE_GRE) { + IP_TUNNEL_DECLARE_FLAGS(tflags) = { }; size_t gre_hdrlen; - __be16 tflags = 0; if (tun_flags & IP_VS_TUNNEL_ENCAP_FLAG_CSUM) - tflags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, tflags); gre_hdrlen = gre_calc_hlen(tflags); max_headroom += gre_hdrlen; diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c index f735d79d8be5..60a76e6e348e 100644 --- a/net/netfilter/nft_tunnel.c +++ b/net/netfilter/nft_tunnel.c @@ -174,8 +174,8 @@ struct nft_tunnel_opts { struct erspan_metadata erspan; u8 data[IP_TUNNEL_OPTS_MAX]; } u; + IP_TUNNEL_DECLARE_FLAGS(flags); u32 len; - __be16 flags; }; struct nft_tunnel_obj { @@ -271,7 +271,8 @@ static int nft_tunnel_obj_vxlan_init(const struct nlattr *attr, opts->u.vxlan.gbp = ntohl(nla_get_be32(tb[NFTA_TUNNEL_KEY_VXLAN_GBP])); opts->len = sizeof(struct vxlan_metadata); - opts->flags = TUNNEL_VXLAN_OPT; + ip_tunnel_flags_zero(opts->flags); + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, opts->flags); return 0; } @@ -325,7 +326,8 @@ static int nft_tunnel_obj_erspan_init(const struct nlattr *attr, opts->u.erspan.version = version; opts->len = sizeof(struct erspan_metadata); - opts->flags = TUNNEL_ERSPAN_OPT; + ip_tunnel_flags_zero(opts->flags); + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, opts->flags); return 0; } @@ -366,7 +368,8 @@ static int nft_tunnel_obj_geneve_init(const struct nlattr *attr, opt->length = data_len / 4; opt->opt_class = nla_get_be16(tb[NFTA_TUNNEL_KEY_GENEVE_CLASS]); opt->type = nla_get_u8(tb[NFTA_TUNNEL_KEY_GENEVE_TYPE]); - opts->flags = TUNNEL_GENEVE_OPT; + ip_tunnel_flags_zero(opts->flags); + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, opts->flags); return 0; } @@ -385,8 +388,8 @@ static int nft_tunnel_obj_opts_init(const struct nft_ctx *ctx, struct nft_tunnel_opts *opts) { struct nlattr *nla; - __be16 type = 0; int err, rem; + u32 type = 0; err = nla_validate_nested_deprecated(attr, NFTA_TUNNEL_KEY_OPTS_MAX, nft_tunnel_opts_policy, NULL); @@ -401,7 +404,7 @@ static int nft_tunnel_obj_opts_init(const struct nft_ctx *ctx, err = nft_tunnel_obj_vxlan_init(nla, opts); if (err) return err; - type = TUNNEL_VXLAN_OPT; + type = IP_TUNNEL_VXLAN_OPT_BIT; break; case NFTA_TUNNEL_KEY_OPTS_ERSPAN: if (type) @@ -409,15 +412,15 @@ static int nft_tunnel_obj_opts_init(const struct nft_ctx *ctx, err = nft_tunnel_obj_erspan_init(nla, opts); if (err) return err; - type = TUNNEL_ERSPAN_OPT; + type = IP_TUNNEL_ERSPAN_OPT_BIT; break; case NFTA_TUNNEL_KEY_OPTS_GENEVE: - if (type && type != TUNNEL_GENEVE_OPT) + if (type && type != IP_TUNNEL_GENEVE_OPT_BIT) return -EINVAL; err = nft_tunnel_obj_geneve_init(nla, opts); if (err) return err; - type = TUNNEL_GENEVE_OPT; + type = IP_TUNNEL_GENEVE_OPT_BIT; break; default: return -EOPNOTSUPP; @@ -454,7 +457,9 @@ static int nft_tunnel_obj_init(const struct nft_ctx *ctx, memset(&info, 0, sizeof(info)); info.mode = IP_TUNNEL_INFO_TX; info.key.tun_id = key32_to_tunnel_id(nla_get_be32(tb[NFTA_TUNNEL_KEY_ID])); - info.key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM | TUNNEL_NOCACHE; + __set_bit(IP_TUNNEL_KEY_BIT, info.key.tun_flags); + __set_bit(IP_TUNNEL_CSUM_BIT, info.key.tun_flags); + __set_bit(IP_TUNNEL_NOCACHE_BIT, info.key.tun_flags); if (tb[NFTA_TUNNEL_KEY_IP]) { err = nft_tunnel_obj_ip_init(ctx, tb[NFTA_TUNNEL_KEY_IP], &info); @@ -483,11 +488,12 @@ static int nft_tunnel_obj_init(const struct nft_ctx *ctx, return -EOPNOTSUPP; if (tun_flags & NFT_TUNNEL_F_ZERO_CSUM_TX) - info.key.tun_flags &= ~TUNNEL_CSUM; + __clear_bit(IP_TUNNEL_CSUM_BIT, info.key.tun_flags); if (tun_flags & NFT_TUNNEL_F_DONT_FRAGMENT) - info.key.tun_flags |= TUNNEL_DONT_FRAGMENT; + __set_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, + info.key.tun_flags); if (tun_flags & NFT_TUNNEL_F_SEQ_NUMBER) - info.key.tun_flags |= TUNNEL_SEQ; + __set_bit(IP_TUNNEL_SEQ_BIT, info.key.tun_flags); } if (tb[NFTA_TUNNEL_KEY_TOS]) info.key.tos = nla_get_u8(tb[NFTA_TUNNEL_KEY_TOS]); @@ -583,7 +589,7 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb, if (!nest) return -1; - if (opts->flags & TUNNEL_VXLAN_OPT) { + if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, opts->flags)) { inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_VXLAN); if (!inner) goto failure; @@ -591,7 +597,7 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb, htonl(opts->u.vxlan.gbp))) goto inner_failure; nla_nest_end(skb, inner); - } else if (opts->flags & TUNNEL_ERSPAN_OPT) { + } else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, opts->flags)) { inner = nla_nest_start_noflag(skb, NFTA_TUNNEL_KEY_OPTS_ERSPAN); if (!inner) goto failure; @@ -613,7 +619,7 @@ static int nft_tunnel_opts_dump(struct sk_buff *skb, break; } nla_nest_end(skb, inner); - } else if (opts->flags & TUNNEL_GENEVE_OPT) { + } else if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, opts->flags)) { struct geneve_opt *opt; int offset = 0; @@ -658,11 +664,11 @@ static int nft_tunnel_flags_dump(struct sk_buff *skb, { u32 flags = 0; - if (info->key.tun_flags & TUNNEL_DONT_FRAGMENT) + if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, info->key.tun_flags)) flags |= NFT_TUNNEL_F_DONT_FRAGMENT; - if (!(info->key.tun_flags & TUNNEL_CSUM)) + if (!test_bit(IP_TUNNEL_CSUM_BIT, info->key.tun_flags)) flags |= NFT_TUNNEL_F_ZERO_CSUM_TX; - if (info->key.tun_flags & TUNNEL_SEQ) + if (test_bit(IP_TUNNEL_SEQ_BIT, info->key.tun_flags)) flags |= NFT_TUNNEL_F_SEQ_NUMBER; if (nla_put_be32(skb, NFTA_TUNNEL_KEY_FLAGS, htonl(flags)) < 0) diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index ebc5728aab4e..f224d9bcea5e 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -152,6 +152,13 @@ static void update_range(struct sw_flow_match *match, sizeof((match)->key->field)); \ } while (0) +#define SW_FLOW_KEY_BITMAP_COPY(match, field, value_p, nbits, is_mask) ({ \ + update_range(match, offsetof(struct sw_flow_key, field), \ + bitmap_size(nbits), is_mask); \ + bitmap_copy(is_mask ? (match)->mask->key.field : (match)->key->field, \ + value_p, nbits); \ +}) + static bool match_validate(const struct sw_flow_match *match, u64 key_attrs, u64 mask_attrs, bool log) { @@ -670,8 +677,8 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, bool log) { bool ttl = false, ipv4 = false, ipv6 = false; + IP_TUNNEL_DECLARE_FLAGS(tun_flags) = { }; bool info_bridge_mode = false; - __be16 tun_flags = 0; int opts_type = 0; struct nlattr *a; int rem; @@ -697,7 +704,7 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, case OVS_TUNNEL_KEY_ATTR_ID: SW_FLOW_KEY_PUT(match, tun_key.tun_id, nla_get_be64(a), is_mask); - tun_flags |= TUNNEL_KEY; + __set_bit(IP_TUNNEL_KEY_BIT, tun_flags); break; case OVS_TUNNEL_KEY_ATTR_IPV4_SRC: SW_FLOW_KEY_PUT(match, tun_key.u.ipv4.src, @@ -729,10 +736,10 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, ttl = true; break; case OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT: - tun_flags |= TUNNEL_DONT_FRAGMENT; + __set_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, tun_flags); break; case OVS_TUNNEL_KEY_ATTR_CSUM: - tun_flags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, tun_flags); break; case OVS_TUNNEL_KEY_ATTR_TP_SRC: SW_FLOW_KEY_PUT(match, tun_key.tp_src, @@ -743,7 +750,7 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, nla_get_be16(a), is_mask); break; case OVS_TUNNEL_KEY_ATTR_OAM: - tun_flags |= TUNNEL_OAM; + __set_bit(IP_TUNNEL_OAM_BIT, tun_flags); break; case OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS: if (opts_type) { @@ -755,7 +762,7 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, if (err) return err; - tun_flags |= TUNNEL_GENEVE_OPT; + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, tun_flags); opts_type = type; break; case OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS: @@ -768,7 +775,7 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, if (err) return err; - tun_flags |= TUNNEL_VXLAN_OPT; + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, tun_flags); opts_type = type; break; case OVS_TUNNEL_KEY_ATTR_PAD: @@ -784,7 +791,7 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, if (err) return err; - tun_flags |= TUNNEL_ERSPAN_OPT; + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, tun_flags); opts_type = type; break; case OVS_TUNNEL_KEY_ATTR_IPV4_INFO_BRIDGE: @@ -798,7 +805,8 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, } } - SW_FLOW_KEY_PUT(match, tun_key.tun_flags, tun_flags, is_mask); + SW_FLOW_KEY_BITMAP_COPY(match, tun_key.tun_flags, tun_flags, + __IP_TUNNEL_FLAG_NUM, is_mask); if (is_mask) SW_FLOW_KEY_MEMSET_FIELD(match, tun_proto, 0xff, true); else @@ -823,13 +831,15 @@ static int ip_tun_from_nlattr(const struct nlattr *attr, } if (ipv4) { if (info_bridge_mode) { + __clear_bit(IP_TUNNEL_KEY_BIT, tun_flags); + if (match->key->tun_key.u.ipv4.src || match->key->tun_key.u.ipv4.dst || match->key->tun_key.tp_src || match->key->tun_key.tp_dst || match->key->tun_key.ttl || match->key->tun_key.tos || - tun_flags & ~TUNNEL_KEY) { + !ip_tunnel_flags_empty(tun_flags)) { OVS_NLERR(log, "IPv4 tun info is not correct"); return -EINVAL; } @@ -874,7 +884,7 @@ static int __ip_tun_to_nlattr(struct sk_buff *skb, const void *tun_opts, int swkey_tun_opts_len, unsigned short tun_proto, u8 mode) { - if (output->tun_flags & TUNNEL_KEY && + if (test_bit(IP_TUNNEL_KEY_BIT, output->tun_flags) && nla_put_be64(skb, OVS_TUNNEL_KEY_ATTR_ID, output->tun_id, OVS_TUNNEL_KEY_ATTR_PAD)) return -EMSGSIZE; @@ -910,10 +920,10 @@ static int __ip_tun_to_nlattr(struct sk_buff *skb, return -EMSGSIZE; if (nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TTL, output->ttl)) return -EMSGSIZE; - if ((output->tun_flags & TUNNEL_DONT_FRAGMENT) && + if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, output->tun_flags) && nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT)) return -EMSGSIZE; - if ((output->tun_flags & TUNNEL_CSUM) && + if (test_bit(IP_TUNNEL_CSUM_BIT, output->tun_flags) && nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_CSUM)) return -EMSGSIZE; if (output->tp_src && @@ -922,18 +932,20 @@ static int __ip_tun_to_nlattr(struct sk_buff *skb, if (output->tp_dst && nla_put_be16(skb, OVS_TUNNEL_KEY_ATTR_TP_DST, output->tp_dst)) return -EMSGSIZE; - if ((output->tun_flags & TUNNEL_OAM) && + if (test_bit(IP_TUNNEL_OAM_BIT, output->tun_flags) && nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_OAM)) return -EMSGSIZE; if (swkey_tun_opts_len) { - if (output->tun_flags & TUNNEL_GENEVE_OPT && + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, output->tun_flags) && nla_put(skb, OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS, swkey_tun_opts_len, tun_opts)) return -EMSGSIZE; - else if (output->tun_flags & TUNNEL_VXLAN_OPT && + else if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, + output->tun_flags) && vxlan_opt_to_nlattr(skb, tun_opts, swkey_tun_opts_len)) return -EMSGSIZE; - else if (output->tun_flags & TUNNEL_ERSPAN_OPT && + else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, + output->tun_flags) && nla_put(skb, OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS, swkey_tun_opts_len, tun_opts)) return -EMSGSIZE; @@ -2029,7 +2041,7 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey, if ((swkey->tun_proto || is_mask)) { const void *opts = NULL; - if (output->tun_key.tun_flags & TUNNEL_OPTIONS_PRESENT) + if (ip_tunnel_is_options_present(output->tun_key.tun_flags)) opts = TUN_METADATA_OPTS(output, swkey->tun_opts_len); if (ip_tun_to_nlattr(skb, &output->tun_key, opts, @@ -2752,7 +2764,8 @@ static int validate_geneve_opts(struct sw_flow_key *key) opts_len -= len; } - key->tun_key.tun_flags |= crit_opt ? TUNNEL_CRIT_OPT : 0; + if (crit_opt) + __set_bit(IP_TUNNEL_CRIT_OPT_BIT, key->tun_key.tun_flags); return 0; } @@ -2760,6 +2773,7 @@ static int validate_geneve_opts(struct sw_flow_key *key) static int validate_and_copy_set_tun(const struct nlattr *attr, struct sw_flow_actions **sfa, bool log) { + IP_TUNNEL_DECLARE_FLAGS(dst_opt_type) = { }; struct sw_flow_match match; struct sw_flow_key key; struct metadata_dst *tun_dst; @@ -2767,9 +2781,7 @@ static int validate_and_copy_set_tun(const struct nlattr *attr, struct ovs_tunnel_info *ovs_tun; struct nlattr *a; int err = 0, start, opts_type; - __be16 dst_opt_type; - dst_opt_type = 0; ovs_match_init(&match, &key, true, NULL); opts_type = ip_tun_from_nlattr(nla_data(attr), &match, false, log); if (opts_type < 0) @@ -2781,13 +2793,14 @@ static int validate_and_copy_set_tun(const struct nlattr *attr, err = validate_geneve_opts(&key); if (err < 0) return err; - dst_opt_type = TUNNEL_GENEVE_OPT; + + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, dst_opt_type); break; case OVS_TUNNEL_KEY_ATTR_VXLAN_OPTS: - dst_opt_type = TUNNEL_VXLAN_OPT; + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, dst_opt_type); break; case OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS: - dst_opt_type = TUNNEL_ERSPAN_OPT; + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, dst_opt_type); break; } } diff --git a/net/psample/psample.c b/net/psample/psample.c index ddd211a151d0..a5d9b8446f77 100644 --- a/net/psample/psample.c +++ b/net/psample/psample.c @@ -221,7 +221,7 @@ static int __psample_ip_tun_to_nlattr(struct sk_buff *skb, const struct ip_tunnel_key *tun_key = &tun_info->key; int tun_opts_len = tun_info->options_len; - if (tun_key->tun_flags & TUNNEL_KEY && + if (test_bit(IP_TUNNEL_KEY_BIT, tun_key->tun_flags) && nla_put_be64(skb, PSAMPLE_TUNNEL_KEY_ATTR_ID, tun_key->tun_id, PSAMPLE_TUNNEL_KEY_ATTR_PAD)) return -EMSGSIZE; @@ -257,10 +257,10 @@ static int __psample_ip_tun_to_nlattr(struct sk_buff *skb, return -EMSGSIZE; if (nla_put_u8(skb, PSAMPLE_TUNNEL_KEY_ATTR_TTL, tun_key->ttl)) return -EMSGSIZE; - if ((tun_key->tun_flags & TUNNEL_DONT_FRAGMENT) && + if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, tun_key->tun_flags) && nla_put_flag(skb, PSAMPLE_TUNNEL_KEY_ATTR_DONT_FRAGMENT)) return -EMSGSIZE; - if ((tun_key->tun_flags & TUNNEL_CSUM) && + if (test_bit(IP_TUNNEL_CSUM_BIT, tun_key->tun_flags) && nla_put_flag(skb, PSAMPLE_TUNNEL_KEY_ATTR_CSUM)) return -EMSGSIZE; if (tun_key->tp_src && @@ -269,15 +269,16 @@ static int __psample_ip_tun_to_nlattr(struct sk_buff *skb, if (tun_key->tp_dst && nla_put_be16(skb, PSAMPLE_TUNNEL_KEY_ATTR_TP_DST, tun_key->tp_dst)) return -EMSGSIZE; - if ((tun_key->tun_flags & TUNNEL_OAM) && + if (test_bit(IP_TUNNEL_OAM_BIT, tun_key->tun_flags) && nla_put_flag(skb, PSAMPLE_TUNNEL_KEY_ATTR_OAM)) return -EMSGSIZE; if (tun_opts_len) { - if (tun_key->tun_flags & TUNNEL_GENEVE_OPT && + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, tun_key->tun_flags) && nla_put(skb, PSAMPLE_TUNNEL_KEY_ATTR_GENEVE_OPTS, tun_opts_len, tun_opts)) return -EMSGSIZE; - else if (tun_key->tun_flags & TUNNEL_ERSPAN_OPT && + else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, + tun_key->tun_flags) && nla_put(skb, PSAMPLE_TUNNEL_KEY_ATTR_ERSPAN_OPTS, tun_opts_len, tun_opts)) return -EMSGSIZE; @@ -314,7 +315,7 @@ static int psample_tunnel_meta_len(struct ip_tunnel_info *tun_info) int tun_opts_len = tun_info->options_len; int sum = nla_total_size(0); /* PSAMPLE_ATTR_TUNNEL */ - if (tun_key->tun_flags & TUNNEL_KEY) + if (test_bit(IP_TUNNEL_KEY_BIT, tun_key->tun_flags)) sum += nla_total_size_64bit(sizeof(u64)); if (tun_info->mode & IP_TUNNEL_INFO_BRIDGE) @@ -337,20 +338,21 @@ static int psample_tunnel_meta_len(struct ip_tunnel_info *tun_info) if (tun_key->tos) sum += nla_total_size(sizeof(u8)); sum += nla_total_size(sizeof(u8)); /* TTL */ - if (tun_key->tun_flags & TUNNEL_DONT_FRAGMENT) + if (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, tun_key->tun_flags)) sum += nla_total_size(0); - if (tun_key->tun_flags & TUNNEL_CSUM) + if (test_bit(IP_TUNNEL_CSUM_BIT, tun_key->tun_flags)) sum += nla_total_size(0); if (tun_key->tp_src) sum += nla_total_size(sizeof(u16)); if (tun_key->tp_dst) sum += nla_total_size(sizeof(u16)); - if (tun_key->tun_flags & TUNNEL_OAM) + if (test_bit(IP_TUNNEL_OAM_BIT, tun_key->tun_flags)) sum += nla_total_size(0); if (tun_opts_len) { - if (tun_key->tun_flags & TUNNEL_GENEVE_OPT) + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, tun_key->tun_flags)) sum += nla_total_size(tun_opts_len); - else if (tun_key->tun_flags & TUNNEL_ERSPAN_OPT) + else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, + tun_key->tun_flags)) sum += nla_total_size(tun_opts_len); } diff --git a/net/sched/act_tunnel_key.c b/net/sched/act_tunnel_key.c index 1536f8b16f1b..af7c99845948 100644 --- a/net/sched/act_tunnel_key.c +++ b/net/sched/act_tunnel_key.c @@ -230,7 +230,7 @@ static int tunnel_key_copy_opts(const struct nlattr *nla, u8 *dst, nla_for_each_attr(attr, head, len, rem) { switch (nla_type(attr)) { case TCA_TUNNEL_KEY_ENC_OPTS_GENEVE: - if (type && type != TUNNEL_GENEVE_OPT) { + if (type && type != IP_TUNNEL_GENEVE_OPT_BIT) { NL_SET_ERR_MSG(extack, "Duplicate type for geneve options"); return -EINVAL; } @@ -247,7 +247,7 @@ static int tunnel_key_copy_opts(const struct nlattr *nla, u8 *dst, dst_len -= opt_len; dst += opt_len; } - type = TUNNEL_GENEVE_OPT; + type = IP_TUNNEL_GENEVE_OPT_BIT; break; case TCA_TUNNEL_KEY_ENC_OPTS_VXLAN: if (type) { @@ -259,7 +259,7 @@ static int tunnel_key_copy_opts(const struct nlattr *nla, u8 *dst, if (opt_len < 0) return opt_len; opts_len += opt_len; - type = TUNNEL_VXLAN_OPT; + type = IP_TUNNEL_VXLAN_OPT_BIT; break; case TCA_TUNNEL_KEY_ENC_OPTS_ERSPAN: if (type) { @@ -271,7 +271,7 @@ static int tunnel_key_copy_opts(const struct nlattr *nla, u8 *dst, if (opt_len < 0) return opt_len; opts_len += opt_len; - type = TUNNEL_ERSPAN_OPT; + type = IP_TUNNEL_ERSPAN_OPT_BIT; break; } } @@ -302,7 +302,7 @@ static int tunnel_key_opts_set(struct nlattr *nla, struct ip_tunnel_info *info, switch (nla_type(nla_data(nla))) { case TCA_TUNNEL_KEY_ENC_OPTS_GENEVE: #if IS_ENABLED(CONFIG_INET) - info->key.tun_flags |= TUNNEL_GENEVE_OPT; + __set_bit(IP_TUNNEL_GENEVE_OPT_BIT, info->key.tun_flags); return tunnel_key_copy_opts(nla, ip_tunnel_info_opts(info), opts_len, extack); #else @@ -310,7 +310,7 @@ static int tunnel_key_opts_set(struct nlattr *nla, struct ip_tunnel_info *info, #endif case TCA_TUNNEL_KEY_ENC_OPTS_VXLAN: #if IS_ENABLED(CONFIG_INET) - info->key.tun_flags |= TUNNEL_VXLAN_OPT; + __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, info->key.tun_flags); return tunnel_key_copy_opts(nla, ip_tunnel_info_opts(info), opts_len, extack); #else @@ -318,7 +318,7 @@ static int tunnel_key_opts_set(struct nlattr *nla, struct ip_tunnel_info *info, #endif case TCA_TUNNEL_KEY_ENC_OPTS_ERSPAN: #if IS_ENABLED(CONFIG_INET) - info->key.tun_flags |= TUNNEL_ERSPAN_OPT; + __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, info->key.tun_flags); return tunnel_key_copy_opts(nla, ip_tunnel_info_opts(info), opts_len, extack); #else @@ -363,6 +363,7 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, bool bind = act_flags & TCA_ACT_FLAGS_BIND; struct nlattr *tb[TCA_TUNNEL_KEY_MAX + 1]; struct tcf_tunnel_key_params *params_new; + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; struct metadata_dst *metadata = NULL; struct tcf_chain *goto_ch = NULL; struct tc_tunnel_key *parm; @@ -371,7 +372,6 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, __be16 dst_port = 0; __be64 key_id = 0; int opts_len = 0; - __be16 flags = 0; u8 tos, ttl; int ret = 0; u32 index; @@ -412,16 +412,16 @@ static int tunnel_key_init(struct net *net, struct nlattr *nla, key32 = nla_get_be32(tb[TCA_TUNNEL_KEY_ENC_KEY_ID]); key_id = key32_to_tunnel_id(key32); - flags = TUNNEL_KEY; + __set_bit(IP_TUNNEL_KEY_BIT, flags); } - flags |= TUNNEL_CSUM; + __set_bit(IP_TUNNEL_CSUM_BIT, flags); if (tb[TCA_TUNNEL_KEY_NO_CSUM] && nla_get_u8(tb[TCA_TUNNEL_KEY_NO_CSUM])) - flags &= ~TUNNEL_CSUM; + __clear_bit(IP_TUNNEL_CSUM_BIT, flags); if (nla_get_flag(tb[TCA_TUNNEL_KEY_NO_FRAG])) - flags |= TUNNEL_DONT_FRAGMENT; + __set_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, flags); if (tb[TCA_TUNNEL_KEY_ENC_DST_PORT]) dst_port = nla_get_be16(tb[TCA_TUNNEL_KEY_ENC_DST_PORT]); @@ -663,15 +663,15 @@ static int tunnel_key_opts_dump(struct sk_buff *skb, if (!start) return -EMSGSIZE; - if (info->key.tun_flags & TUNNEL_GENEVE_OPT) { + if (test_bit(IP_TUNNEL_GENEVE_OPT_BIT, info->key.tun_flags)) { err = tunnel_key_geneve_opts_dump(skb, info); if (err) goto err_out; - } else if (info->key.tun_flags & TUNNEL_VXLAN_OPT) { + } else if (test_bit(IP_TUNNEL_VXLAN_OPT_BIT, info->key.tun_flags)) { err = tunnel_key_vxlan_opts_dump(skb, info); if (err) goto err_out; - } else if (info->key.tun_flags & TUNNEL_ERSPAN_OPT) { + } else if (test_bit(IP_TUNNEL_ERSPAN_OPT_BIT, info->key.tun_flags)) { err = tunnel_key_erspan_opts_dump(skb, info); if (err) goto err_out; @@ -741,7 +741,7 @@ static int tunnel_key_dump(struct sk_buff *skb, struct tc_action *a, struct ip_tunnel_key *key = &info->key; __be32 key_id = tunnel_id_to_key32(key->tun_id); - if (((key->tun_flags & TUNNEL_KEY) && + if ((test_bit(IP_TUNNEL_KEY_BIT, key->tun_flags) && nla_put_be32(skb, TCA_TUNNEL_KEY_ENC_KEY_ID, key_id)) || tunnel_key_dump_addresses(skb, ¶ms->tcft_enc_metadata->u.tun_info) || @@ -749,8 +749,8 @@ static int tunnel_key_dump(struct sk_buff *skb, struct tc_action *a, nla_put_be16(skb, TCA_TUNNEL_KEY_ENC_DST_PORT, key->tp_dst)) || nla_put_u8(skb, TCA_TUNNEL_KEY_NO_CSUM, - !(key->tun_flags & TUNNEL_CSUM)) || - ((key->tun_flags & TUNNEL_DONT_FRAGMENT) && + !test_bit(IP_TUNNEL_CSUM_BIT, key->tun_flags)) || + (test_bit(IP_TUNNEL_DONT_FRAGMENT_BIT, key->tun_flags) && nla_put_flag(skb, TCA_TUNNEL_KEY_NO_FRAG)) || tunnel_key_opts_dump(skb, info)) goto nla_put_failure; diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index e1314674b4a9..ae14d649140f 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -1454,12 +1454,13 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, switch (nla_type(nla_opt_key)) { case TCA_FLOWER_KEY_ENC_OPTS_GENEVE: if (key->enc_opts.dst_opt_type && - key->enc_opts.dst_opt_type != TUNNEL_GENEVE_OPT) { + key->enc_opts.dst_opt_type != + IP_TUNNEL_GENEVE_OPT_BIT) { NL_SET_ERR_MSG(extack, "Duplicate type for geneve options"); return -EINVAL; } option_len = 0; - key->enc_opts.dst_opt_type = TUNNEL_GENEVE_OPT; + key->enc_opts.dst_opt_type = IP_TUNNEL_GENEVE_OPT_BIT; option_len = fl_set_geneve_opt(nla_opt_key, key, key_depth, option_len, extack); @@ -1470,7 +1471,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, /* At the same time we need to parse through the mask * in order to verify exact and mask attribute lengths. */ - mask->enc_opts.dst_opt_type = TUNNEL_GENEVE_OPT; + mask->enc_opts.dst_opt_type = IP_TUNNEL_GENEVE_OPT_BIT; option_len = fl_set_geneve_opt(nla_opt_msk, mask, msk_depth, option_len, extack); @@ -1489,7 +1490,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, return -EINVAL; } option_len = 0; - key->enc_opts.dst_opt_type = TUNNEL_VXLAN_OPT; + key->enc_opts.dst_opt_type = IP_TUNNEL_VXLAN_OPT_BIT; option_len = fl_set_vxlan_opt(nla_opt_key, key, key_depth, option_len, extack); @@ -1500,7 +1501,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, /* At the same time we need to parse through the mask * in order to verify exact and mask attribute lengths. */ - mask->enc_opts.dst_opt_type = TUNNEL_VXLAN_OPT; + mask->enc_opts.dst_opt_type = IP_TUNNEL_VXLAN_OPT_BIT; option_len = fl_set_vxlan_opt(nla_opt_msk, mask, msk_depth, option_len, extack); @@ -1519,7 +1520,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, return -EINVAL; } option_len = 0; - key->enc_opts.dst_opt_type = TUNNEL_ERSPAN_OPT; + key->enc_opts.dst_opt_type = IP_TUNNEL_ERSPAN_OPT_BIT; option_len = fl_set_erspan_opt(nla_opt_key, key, key_depth, option_len, extack); @@ -1530,7 +1531,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, /* At the same time we need to parse through the mask * in order to verify exact and mask attribute lengths. */ - mask->enc_opts.dst_opt_type = TUNNEL_ERSPAN_OPT; + mask->enc_opts.dst_opt_type = IP_TUNNEL_ERSPAN_OPT_BIT; option_len = fl_set_erspan_opt(nla_opt_msk, mask, msk_depth, option_len, extack); @@ -1550,7 +1551,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, return -EINVAL; } option_len = 0; - key->enc_opts.dst_opt_type = TUNNEL_GTP_OPT; + key->enc_opts.dst_opt_type = IP_TUNNEL_GTP_OPT_BIT; option_len = fl_set_gtp_opt(nla_opt_key, key, key_depth, option_len, extack); @@ -1561,7 +1562,7 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, /* At the same time we need to parse through the mask * in order to verify exact and mask attribute lengths. */ - mask->enc_opts.dst_opt_type = TUNNEL_GTP_OPT; + mask->enc_opts.dst_opt_type = IP_TUNNEL_GTP_OPT_BIT; option_len = fl_set_gtp_opt(nla_opt_msk, mask, msk_depth, option_len, extack); @@ -3202,22 +3203,22 @@ static int fl_dump_key_options(struct sk_buff *skb, int enc_opt_type, goto nla_put_failure; switch (enc_opts->dst_opt_type) { - case TUNNEL_GENEVE_OPT: + case IP_TUNNEL_GENEVE_OPT_BIT: err = fl_dump_key_geneve_opt(skb, enc_opts); if (err) goto nla_put_failure; break; - case TUNNEL_VXLAN_OPT: + case IP_TUNNEL_VXLAN_OPT_BIT: err = fl_dump_key_vxlan_opt(skb, enc_opts); if (err) goto nla_put_failure; break; - case TUNNEL_ERSPAN_OPT: + case IP_TUNNEL_ERSPAN_OPT_BIT: err = fl_dump_key_erspan_opt(skb, enc_opts); if (err) goto nla_put_failure; break; - case TUNNEL_GTP_OPT: + case IP_TUNNEL_GTP_OPT_BIT: err = fl_dump_key_gtp_opt(skb, enc_opts); if (err) goto nla_put_failure; -- cgit v1.2.3 From 5b2be2ab76d11e20dda502712858dd0c2d4dfcd0 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Wed, 27 Mar 2024 16:23:54 +0100 Subject: net: net_test: add tests for IP tunnel flags conversion helpers Now that there are helpers for converting IP tunnel flags between the old __be16 format and the bitmap format, make sure they work as expected by adding a couple of tests to the networking testing suite. The helpers are all inline, so no dependencies on the related CONFIG_* (or a standalone module) are needed. Cover three possible cases: 1. No bits past BIT(15) are set, VTI/SIT bits are not set. This conversion is almost a direct assignment. 2. No bits past BIT(15) are set, but VTI/SIT bit is set. During the conversion, it must be transformed into BIT(16) in the bitmap, but still compatible with the __be16 format. 3. The bitmap has bits past BIT(15) set (not the VTI/SIT one). The result will be truncated. Note that currently __IP_TUNNEL_FLAG_NUM is 17 (incl. special), which means that the result of this case is currently semi-false-positive. When BIT(17) is finally here, it will be adjusted accordingly. Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- net/core/Makefile | 2 +- net/core/gso_test.c | 274 ------------------------------------ net/core/net_test.c | 390 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 391 insertions(+), 275 deletions(-) delete mode 100644 net/core/gso_test.c create mode 100644 net/core/net_test.c (limited to 'net') diff --git a/net/core/Makefile b/net/core/Makefile index 6e6548011fae..21d6fbc7e884 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -41,4 +41,4 @@ obj-$(CONFIG_NET_SOCK_MSG) += skmsg.o obj-$(CONFIG_BPF_SYSCALL) += sock_map.o obj-$(CONFIG_BPF_SYSCALL) += bpf_sk_storage.o obj-$(CONFIG_OF) += of_net.o -obj-$(CONFIG_NET_TEST) += gso_test.o +obj-$(CONFIG_NET_TEST) += net_test.o diff --git a/net/core/gso_test.c b/net/core/gso_test.c deleted file mode 100644 index 358c44680d91..000000000000 --- a/net/core/gso_test.c +++ /dev/null @@ -1,274 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later - -#include -#include - -static const char hdr[] = "abcdefgh"; -#define GSO_TEST_SIZE 1000 - -static void __init_skb(struct sk_buff *skb) -{ - skb_reset_mac_header(skb); - memcpy(skb_mac_header(skb), hdr, sizeof(hdr)); - - /* skb_segment expects skb->data at start of payload */ - skb_pull(skb, sizeof(hdr)); - skb_reset_network_header(skb); - skb_reset_transport_header(skb); - - /* proto is arbitrary, as long as not ETH_P_TEB or vlan */ - skb->protocol = htons(ETH_P_ATALK); - skb_shinfo(skb)->gso_size = GSO_TEST_SIZE; -} - -enum gso_test_nr { - GSO_TEST_LINEAR, - GSO_TEST_NO_GSO, - GSO_TEST_FRAGS, - GSO_TEST_FRAGS_PURE, - GSO_TEST_GSO_PARTIAL, - GSO_TEST_FRAG_LIST, - GSO_TEST_FRAG_LIST_PURE, - GSO_TEST_FRAG_LIST_NON_UNIFORM, - GSO_TEST_GSO_BY_FRAGS, -}; - -struct gso_test_case { - enum gso_test_nr id; - const char *name; - - /* input */ - unsigned int linear_len; - unsigned int nr_frags; - const unsigned int *frags; - unsigned int nr_frag_skbs; - const unsigned int *frag_skbs; - - /* output as expected */ - unsigned int nr_segs; - const unsigned int *segs; -}; - -static struct gso_test_case cases[] = { - { - .id = GSO_TEST_NO_GSO, - .name = "no_gso", - .linear_len = GSO_TEST_SIZE, - .nr_segs = 1, - .segs = (const unsigned int[]) { GSO_TEST_SIZE }, - }, - { - .id = GSO_TEST_LINEAR, - .name = "linear", - .linear_len = GSO_TEST_SIZE + GSO_TEST_SIZE + 1, - .nr_segs = 3, - .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 1 }, - }, - { - .id = GSO_TEST_FRAGS, - .name = "frags", - .linear_len = GSO_TEST_SIZE, - .nr_frags = 2, - .frags = (const unsigned int[]) { GSO_TEST_SIZE, 1 }, - .nr_segs = 3, - .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 1 }, - }, - { - .id = GSO_TEST_FRAGS_PURE, - .name = "frags_pure", - .nr_frags = 3, - .frags = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 2 }, - .nr_segs = 3, - .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 2 }, - }, - { - .id = GSO_TEST_GSO_PARTIAL, - .name = "gso_partial", - .linear_len = GSO_TEST_SIZE, - .nr_frags = 2, - .frags = (const unsigned int[]) { GSO_TEST_SIZE, 3 }, - .nr_segs = 2, - .segs = (const unsigned int[]) { 2 * GSO_TEST_SIZE, 3 }, - }, - { - /* commit 89319d3801d1: frag_list on mss boundaries */ - .id = GSO_TEST_FRAG_LIST, - .name = "frag_list", - .linear_len = GSO_TEST_SIZE, - .nr_frag_skbs = 2, - .frag_skbs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE }, - .nr_segs = 3, - .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, GSO_TEST_SIZE }, - }, - { - .id = GSO_TEST_FRAG_LIST_PURE, - .name = "frag_list_pure", - .nr_frag_skbs = 2, - .frag_skbs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE }, - .nr_segs = 2, - .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE }, - }, - { - /* commit 43170c4e0ba7: GRO of frag_list trains */ - .id = GSO_TEST_FRAG_LIST_NON_UNIFORM, - .name = "frag_list_non_uniform", - .linear_len = GSO_TEST_SIZE, - .nr_frag_skbs = 4, - .frag_skbs = (const unsigned int[]) { GSO_TEST_SIZE, 1, GSO_TEST_SIZE, 2 }, - .nr_segs = 4, - .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, GSO_TEST_SIZE, 3 }, - }, - { - /* commit 3953c46c3ac7 ("sk_buff: allow segmenting based on frag sizes") and - * commit 90017accff61 ("sctp: Add GSO support") - * - * "there will be a cover skb with protocol headers and - * children ones containing the actual segments" - */ - .id = GSO_TEST_GSO_BY_FRAGS, - .name = "gso_by_frags", - .nr_frag_skbs = 4, - .frag_skbs = (const unsigned int[]) { 100, 200, 300, 400 }, - .nr_segs = 4, - .segs = (const unsigned int[]) { 100, 200, 300, 400 }, - }, -}; - -static void gso_test_case_to_desc(struct gso_test_case *t, char *desc) -{ - sprintf(desc, "%s", t->name); -} - -KUNIT_ARRAY_PARAM(gso_test, cases, gso_test_case_to_desc); - -static void gso_test_func(struct kunit *test) -{ - const int shinfo_size = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); - struct sk_buff *skb, *segs, *cur, *next, *last; - const struct gso_test_case *tcase; - netdev_features_t features; - struct page *page; - int i; - - tcase = test->param_value; - - page = alloc_page(GFP_KERNEL); - KUNIT_ASSERT_NOT_NULL(test, page); - skb = build_skb(page_address(page), sizeof(hdr) + tcase->linear_len + shinfo_size); - KUNIT_ASSERT_NOT_NULL(test, skb); - __skb_put(skb, sizeof(hdr) + tcase->linear_len); - - __init_skb(skb); - - if (tcase->nr_frags) { - unsigned int pg_off = 0; - - page = alloc_page(GFP_KERNEL); - KUNIT_ASSERT_NOT_NULL(test, page); - page_ref_add(page, tcase->nr_frags - 1); - - for (i = 0; i < tcase->nr_frags; i++) { - skb_fill_page_desc(skb, i, page, pg_off, tcase->frags[i]); - pg_off += tcase->frags[i]; - } - - KUNIT_ASSERT_LE(test, pg_off, PAGE_SIZE); - - skb->data_len = pg_off; - skb->len += skb->data_len; - skb->truesize += skb->data_len; - } - - if (tcase->frag_skbs) { - unsigned int total_size = 0, total_true_size = 0; - struct sk_buff *frag_skb, *prev = NULL; - - for (i = 0; i < tcase->nr_frag_skbs; i++) { - unsigned int frag_size; - - page = alloc_page(GFP_KERNEL); - KUNIT_ASSERT_NOT_NULL(test, page); - - frag_size = tcase->frag_skbs[i]; - frag_skb = build_skb(page_address(page), - frag_size + shinfo_size); - KUNIT_ASSERT_NOT_NULL(test, frag_skb); - __skb_put(frag_skb, frag_size); - - if (prev) - prev->next = frag_skb; - else - skb_shinfo(skb)->frag_list = frag_skb; - prev = frag_skb; - - total_size += frag_size; - total_true_size += frag_skb->truesize; - } - - skb->len += total_size; - skb->data_len += total_size; - skb->truesize += total_true_size; - - if (tcase->id == GSO_TEST_GSO_BY_FRAGS) - skb_shinfo(skb)->gso_size = GSO_BY_FRAGS; - } - - features = NETIF_F_SG | NETIF_F_HW_CSUM; - if (tcase->id == GSO_TEST_GSO_PARTIAL) - features |= NETIF_F_GSO_PARTIAL; - - /* TODO: this should also work with SG, - * rather than hit BUG_ON(i >= nfrags) - */ - if (tcase->id == GSO_TEST_FRAG_LIST_NON_UNIFORM) - features &= ~NETIF_F_SG; - - segs = skb_segment(skb, features); - if (IS_ERR(segs)) { - KUNIT_FAIL(test, "segs error %pe", segs); - goto free_gso_skb; - } else if (!segs) { - KUNIT_FAIL(test, "no segments"); - goto free_gso_skb; - } - - last = segs->prev; - for (cur = segs, i = 0; cur; cur = next, i++) { - next = cur->next; - - KUNIT_ASSERT_EQ(test, cur->len, sizeof(hdr) + tcase->segs[i]); - - /* segs have skb->data pointing to the mac header */ - KUNIT_ASSERT_PTR_EQ(test, skb_mac_header(cur), cur->data); - KUNIT_ASSERT_PTR_EQ(test, skb_network_header(cur), cur->data + sizeof(hdr)); - - /* header was copied to all segs */ - KUNIT_ASSERT_EQ(test, memcmp(skb_mac_header(cur), hdr, sizeof(hdr)), 0); - - /* last seg can be found through segs->prev pointer */ - if (!next) - KUNIT_ASSERT_PTR_EQ(test, cur, last); - - consume_skb(cur); - } - - KUNIT_ASSERT_EQ(test, i, tcase->nr_segs); - -free_gso_skb: - consume_skb(skb); -} - -static struct kunit_case gso_test_cases[] = { - KUNIT_CASE_PARAM(gso_test_func, gso_test_gen_params), - {} -}; - -static struct kunit_suite gso_test_suite = { - .name = "net_core_gso", - .test_cases = gso_test_cases, -}; - -kunit_test_suite(gso_test_suite); - -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("KUnit tests for segmentation offload"); diff --git a/net/core/net_test.c b/net/core/net_test.c new file mode 100644 index 000000000000..30062b9621e4 --- /dev/null +++ b/net/core/net_test.c @@ -0,0 +1,390 @@ +// SPDX-License-Identifier: GPL-2.0-or-later + +#include + +/* GSO */ + +#include + +static const char hdr[] = "abcdefgh"; +#define GSO_TEST_SIZE 1000 + +static void __init_skb(struct sk_buff *skb) +{ + skb_reset_mac_header(skb); + memcpy(skb_mac_header(skb), hdr, sizeof(hdr)); + + /* skb_segment expects skb->data at start of payload */ + skb_pull(skb, sizeof(hdr)); + skb_reset_network_header(skb); + skb_reset_transport_header(skb); + + /* proto is arbitrary, as long as not ETH_P_TEB or vlan */ + skb->protocol = htons(ETH_P_ATALK); + skb_shinfo(skb)->gso_size = GSO_TEST_SIZE; +} + +enum gso_test_nr { + GSO_TEST_LINEAR, + GSO_TEST_NO_GSO, + GSO_TEST_FRAGS, + GSO_TEST_FRAGS_PURE, + GSO_TEST_GSO_PARTIAL, + GSO_TEST_FRAG_LIST, + GSO_TEST_FRAG_LIST_PURE, + GSO_TEST_FRAG_LIST_NON_UNIFORM, + GSO_TEST_GSO_BY_FRAGS, +}; + +struct gso_test_case { + enum gso_test_nr id; + const char *name; + + /* input */ + unsigned int linear_len; + unsigned int nr_frags; + const unsigned int *frags; + unsigned int nr_frag_skbs; + const unsigned int *frag_skbs; + + /* output as expected */ + unsigned int nr_segs; + const unsigned int *segs; +}; + +static struct gso_test_case cases[] = { + { + .id = GSO_TEST_NO_GSO, + .name = "no_gso", + .linear_len = GSO_TEST_SIZE, + .nr_segs = 1, + .segs = (const unsigned int[]) { GSO_TEST_SIZE }, + }, + { + .id = GSO_TEST_LINEAR, + .name = "linear", + .linear_len = GSO_TEST_SIZE + GSO_TEST_SIZE + 1, + .nr_segs = 3, + .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 1 }, + }, + { + .id = GSO_TEST_FRAGS, + .name = "frags", + .linear_len = GSO_TEST_SIZE, + .nr_frags = 2, + .frags = (const unsigned int[]) { GSO_TEST_SIZE, 1 }, + .nr_segs = 3, + .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 1 }, + }, + { + .id = GSO_TEST_FRAGS_PURE, + .name = "frags_pure", + .nr_frags = 3, + .frags = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 2 }, + .nr_segs = 3, + .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, 2 }, + }, + { + .id = GSO_TEST_GSO_PARTIAL, + .name = "gso_partial", + .linear_len = GSO_TEST_SIZE, + .nr_frags = 2, + .frags = (const unsigned int[]) { GSO_TEST_SIZE, 3 }, + .nr_segs = 2, + .segs = (const unsigned int[]) { 2 * GSO_TEST_SIZE, 3 }, + }, + { + /* commit 89319d3801d1: frag_list on mss boundaries */ + .id = GSO_TEST_FRAG_LIST, + .name = "frag_list", + .linear_len = GSO_TEST_SIZE, + .nr_frag_skbs = 2, + .frag_skbs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE }, + .nr_segs = 3, + .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, GSO_TEST_SIZE }, + }, + { + .id = GSO_TEST_FRAG_LIST_PURE, + .name = "frag_list_pure", + .nr_frag_skbs = 2, + .frag_skbs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE }, + .nr_segs = 2, + .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE }, + }, + { + /* commit 43170c4e0ba7: GRO of frag_list trains */ + .id = GSO_TEST_FRAG_LIST_NON_UNIFORM, + .name = "frag_list_non_uniform", + .linear_len = GSO_TEST_SIZE, + .nr_frag_skbs = 4, + .frag_skbs = (const unsigned int[]) { GSO_TEST_SIZE, 1, GSO_TEST_SIZE, 2 }, + .nr_segs = 4, + .segs = (const unsigned int[]) { GSO_TEST_SIZE, GSO_TEST_SIZE, GSO_TEST_SIZE, 3 }, + }, + { + /* commit 3953c46c3ac7 ("sk_buff: allow segmenting based on frag sizes") and + * commit 90017accff61 ("sctp: Add GSO support") + * + * "there will be a cover skb with protocol headers and + * children ones containing the actual segments" + */ + .id = GSO_TEST_GSO_BY_FRAGS, + .name = "gso_by_frags", + .nr_frag_skbs = 4, + .frag_skbs = (const unsigned int[]) { 100, 200, 300, 400 }, + .nr_segs = 4, + .segs = (const unsigned int[]) { 100, 200, 300, 400 }, + }, +}; + +static void gso_test_case_to_desc(struct gso_test_case *t, char *desc) +{ + sprintf(desc, "%s", t->name); +} + +KUNIT_ARRAY_PARAM(gso_test, cases, gso_test_case_to_desc); + +static void gso_test_func(struct kunit *test) +{ + const int shinfo_size = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); + struct sk_buff *skb, *segs, *cur, *next, *last; + const struct gso_test_case *tcase; + netdev_features_t features; + struct page *page; + int i; + + tcase = test->param_value; + + page = alloc_page(GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, page); + skb = build_skb(page_address(page), sizeof(hdr) + tcase->linear_len + shinfo_size); + KUNIT_ASSERT_NOT_NULL(test, skb); + __skb_put(skb, sizeof(hdr) + tcase->linear_len); + + __init_skb(skb); + + if (tcase->nr_frags) { + unsigned int pg_off = 0; + + page = alloc_page(GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, page); + page_ref_add(page, tcase->nr_frags - 1); + + for (i = 0; i < tcase->nr_frags; i++) { + skb_fill_page_desc(skb, i, page, pg_off, tcase->frags[i]); + pg_off += tcase->frags[i]; + } + + KUNIT_ASSERT_LE(test, pg_off, PAGE_SIZE); + + skb->data_len = pg_off; + skb->len += skb->data_len; + skb->truesize += skb->data_len; + } + + if (tcase->frag_skbs) { + unsigned int total_size = 0, total_true_size = 0; + struct sk_buff *frag_skb, *prev = NULL; + + for (i = 0; i < tcase->nr_frag_skbs; i++) { + unsigned int frag_size; + + page = alloc_page(GFP_KERNEL); + KUNIT_ASSERT_NOT_NULL(test, page); + + frag_size = tcase->frag_skbs[i]; + frag_skb = build_skb(page_address(page), + frag_size + shinfo_size); + KUNIT_ASSERT_NOT_NULL(test, frag_skb); + __skb_put(frag_skb, frag_size); + + if (prev) + prev->next = frag_skb; + else + skb_shinfo(skb)->frag_list = frag_skb; + prev = frag_skb; + + total_size += frag_size; + total_true_size += frag_skb->truesize; + } + + skb->len += total_size; + skb->data_len += total_size; + skb->truesize += total_true_size; + + if (tcase->id == GSO_TEST_GSO_BY_FRAGS) + skb_shinfo(skb)->gso_size = GSO_BY_FRAGS; + } + + features = NETIF_F_SG | NETIF_F_HW_CSUM; + if (tcase->id == GSO_TEST_GSO_PARTIAL) + features |= NETIF_F_GSO_PARTIAL; + + /* TODO: this should also work with SG, + * rather than hit BUG_ON(i >= nfrags) + */ + if (tcase->id == GSO_TEST_FRAG_LIST_NON_UNIFORM) + features &= ~NETIF_F_SG; + + segs = skb_segment(skb, features); + if (IS_ERR(segs)) { + KUNIT_FAIL(test, "segs error %pe", segs); + goto free_gso_skb; + } else if (!segs) { + KUNIT_FAIL(test, "no segments"); + goto free_gso_skb; + } + + last = segs->prev; + for (cur = segs, i = 0; cur; cur = next, i++) { + next = cur->next; + + KUNIT_ASSERT_EQ(test, cur->len, sizeof(hdr) + tcase->segs[i]); + + /* segs have skb->data pointing to the mac header */ + KUNIT_ASSERT_PTR_EQ(test, skb_mac_header(cur), cur->data); + KUNIT_ASSERT_PTR_EQ(test, skb_network_header(cur), cur->data + sizeof(hdr)); + + /* header was copied to all segs */ + KUNIT_ASSERT_EQ(test, memcmp(skb_mac_header(cur), hdr, sizeof(hdr)), 0); + + /* last seg can be found through segs->prev pointer */ + if (!next) + KUNIT_ASSERT_PTR_EQ(test, cur, last); + + consume_skb(cur); + } + + KUNIT_ASSERT_EQ(test, i, tcase->nr_segs); + +free_gso_skb: + consume_skb(skb); +} + +/* IP tunnel flags */ + +#include + +struct ip_tunnel_flags_test { + const char *name; + + const u16 *src_bits; + const u16 *exp_bits; + u8 src_num; + u8 exp_num; + + __be16 exp_val; + bool exp_comp; +}; + +#define IP_TUNNEL_FLAGS_TEST(n, src, comp, eval, exp) { \ + .name = (n), \ + .src_bits = (src), \ + .src_num = ARRAY_SIZE(src), \ + .exp_comp = (comp), \ + .exp_val = (eval), \ + .exp_bits = (exp), \ + .exp_num = ARRAY_SIZE(exp), \ +} + +/* These are __be16-compatible and can be compared as is */ +static const u16 ip_tunnel_flags_1[] = { + IP_TUNNEL_KEY_BIT, + IP_TUNNEL_STRICT_BIT, + IP_TUNNEL_ERSPAN_OPT_BIT, +}; + +/* Due to the previous flags design limitation, setting either + * ``IP_TUNNEL_CSUM_BIT`` (on Big Endian) or ``IP_TUNNEL_DONT_FRAGMENT_BIT`` + * (on Little) also sets VTI/ISATAP bit. In the bitmap implementation, they + * correspond to ``BIT(16)``, which is bigger than ``U16_MAX``, but still is + * backward-compatible. + */ +#ifdef __LITTLE_ENDIAN +#define IP_TUNNEL_CONFLICT_BIT IP_TUNNEL_DONT_FRAGMENT_BIT +#else +#define IP_TUNNEL_CONFLICT_BIT IP_TUNNEL_CSUM_BIT +#endif + +static const u16 ip_tunnel_flags_2_src[] = { + IP_TUNNEL_CONFLICT_BIT, +}; + +static const u16 ip_tunnel_flags_2_exp[] = { + IP_TUNNEL_CONFLICT_BIT, + IP_TUNNEL_SIT_ISATAP_BIT, +}; + +/* Bits 17 and higher are not compatible with __be16 flags */ +static const u16 ip_tunnel_flags_3_src[] = { + IP_TUNNEL_VXLAN_OPT_BIT, + 17, + 18, + 20, +}; + +static const u16 ip_tunnel_flags_3_exp[] = { + IP_TUNNEL_VXLAN_OPT_BIT, +}; + +static const struct ip_tunnel_flags_test ip_tunnel_flags_test[] = { + IP_TUNNEL_FLAGS_TEST("compat", ip_tunnel_flags_1, true, + cpu_to_be16(BIT(IP_TUNNEL_KEY_BIT) | + BIT(IP_TUNNEL_STRICT_BIT) | + BIT(IP_TUNNEL_ERSPAN_OPT_BIT)), + ip_tunnel_flags_1), + IP_TUNNEL_FLAGS_TEST("conflict", ip_tunnel_flags_2_src, true, + VTI_ISVTI, ip_tunnel_flags_2_exp), + IP_TUNNEL_FLAGS_TEST("new", ip_tunnel_flags_3_src, + /* This must be set to ``false`` once + * ``__IP_TUNNEL_FLAG_NUM`` goes above 17. + */ + true, cpu_to_be16(BIT(IP_TUNNEL_VXLAN_OPT_BIT)), + ip_tunnel_flags_3_exp), +}; + +static void +ip_tunnel_flags_test_case_to_desc(const struct ip_tunnel_flags_test *t, + char *desc) +{ + strscpy(desc, t->name, KUNIT_PARAM_DESC_SIZE); +} +KUNIT_ARRAY_PARAM(ip_tunnel_flags_test, ip_tunnel_flags_test, + ip_tunnel_flags_test_case_to_desc); + +static void ip_tunnel_flags_test_run(struct kunit *test) +{ + const struct ip_tunnel_flags_test *t = test->param_value; + IP_TUNNEL_DECLARE_FLAGS(src) = { }; + IP_TUNNEL_DECLARE_FLAGS(exp) = { }; + IP_TUNNEL_DECLARE_FLAGS(out); + + for (u32 j = 0; j < t->src_num; j++) + __set_bit(t->src_bits[j], src); + for (u32 j = 0; j < t->exp_num; j++) + __set_bit(t->exp_bits[j], exp); + + KUNIT_ASSERT_EQ(test, t->exp_comp, + ip_tunnel_flags_is_be16_compat(src)); + KUNIT_ASSERT_EQ(test, (__force u16)t->exp_val, + (__force u16)ip_tunnel_flags_to_be16(src)); + + ip_tunnel_flags_from_be16(out, t->exp_val); + KUNIT_ASSERT_TRUE(test, __ipt_flag_op(bitmap_equal, exp, out)); +} + +static struct kunit_case net_test_cases[] = { + KUNIT_CASE_PARAM(gso_test_func, gso_test_gen_params), + KUNIT_CASE_PARAM(ip_tunnel_flags_test_run, + ip_tunnel_flags_test_gen_params), + { }, +}; + +static struct kunit_suite net_test_suite = { + .name = "net_core", + .test_cases = net_test_cases, +}; +kunit_test_suite(net_test_suite); + +MODULE_DESCRIPTION("KUnit tests for networking core"); +MODULE_LICENSE("GPL"); -- cgit v1.2.3 From 6dd514f48110ebb4bf36875b9e7e02d07b589caa Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Wed, 27 Mar 2024 16:23:56 +0100 Subject: pfcp: always set pfcp metadata In PFCP receive path set metadata needed by flower code to do correct classification based on this metadata. Signed-off-by: Michal Swiatkowski Signed-off-by: Marcin Szycik Reviewed-by: Simon Horman Signed-off-by: Alexander Lobakin Signed-off-by: David S. Miller --- drivers/net/pfcp.c | 81 ++++++++++++++++++++++++++++++- include/net/ip_tunnels.h | 3 ++ include/net/pfcp.h | 73 ++++++++++++++++++++++++++++ include/uapi/linux/if_tunnel.h | 3 ++ include/uapi/linux/pkt_cls.h | 14 ++++++ net/core/net_test.c | 7 +-- net/sched/cls_flower.c | 107 +++++++++++++++++++++++++++++++++++++++++ 7 files changed, 282 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/drivers/net/pfcp.c b/drivers/net/pfcp.c index 3f1ee0ae7111..cc5b28c5f99f 100644 --- a/drivers/net/pfcp.c +++ b/drivers/net/pfcp.c @@ -21,6 +21,8 @@ struct pfcp_dev { struct socket *sock; struct net_device *dev; struct net *net; + + struct gro_cells gro_cells; }; static unsigned int pfcp_net_id __read_mostly; @@ -29,6 +31,78 @@ struct pfcp_net { struct list_head pfcp_dev_list; }; +static void +pfcp_session_recv(struct pfcp_dev *pfcp, struct sk_buff *skb, + struct pfcp_metadata *md) +{ + struct pfcphdr_session *unparsed = pfcp_hdr_session(skb); + + md->seid = unparsed->seid; + md->type = PFCP_TYPE_SESSION; +} + +static void +pfcp_node_recv(struct pfcp_dev *pfcp, struct sk_buff *skb, + struct pfcp_metadata *md) +{ + md->type = PFCP_TYPE_NODE; +} + +static int pfcp_encap_recv(struct sock *sk, struct sk_buff *skb) +{ + IP_TUNNEL_DECLARE_FLAGS(flags) = { }; + struct metadata_dst *tun_dst; + struct pfcp_metadata *md; + struct pfcphdr *unparsed; + struct pfcp_dev *pfcp; + + if (unlikely(!pskb_may_pull(skb, PFCP_HLEN))) + goto drop; + + pfcp = rcu_dereference_sk_user_data(sk); + if (unlikely(!pfcp)) + goto drop; + + unparsed = pfcp_hdr(skb); + + ip_tunnel_flags_zero(flags); + tun_dst = udp_tun_rx_dst(skb, sk->sk_family, flags, 0, + sizeof(*md)); + if (unlikely(!tun_dst)) + goto drop; + + md = ip_tunnel_info_opts(&tun_dst->u.tun_info); + if (unlikely(!md)) + goto drop; + + if (unparsed->flags & PFCP_SEID_FLAG) + pfcp_session_recv(pfcp, skb, md); + else + pfcp_node_recv(pfcp, skb, md); + + __set_bit(IP_TUNNEL_PFCP_OPT_BIT, flags); + ip_tunnel_info_opts_set(&tun_dst->u.tun_info, md, sizeof(*md), + flags); + + if (unlikely(iptunnel_pull_header(skb, PFCP_HLEN, skb->protocol, + !net_eq(sock_net(sk), + dev_net(pfcp->dev))))) + goto drop; + + skb_dst_set(skb, (struct dst_entry *)tun_dst); + + skb_reset_network_header(skb); + skb_reset_mac_header(skb); + skb->dev = pfcp->dev; + + gro_cells_receive(&pfcp->gro_cells, skb); + + return 0; +drop: + kfree_skb(skb); + return 0; +} + static void pfcp_del_sock(struct pfcp_dev *pfcp) { udp_tunnel_sock_release(pfcp->sock); @@ -39,6 +113,7 @@ static void pfcp_dev_uninit(struct net_device *dev) { struct pfcp_dev *pfcp = netdev_priv(dev); + gro_cells_destroy(&pfcp->gro_cells); pfcp_del_sock(pfcp); } @@ -48,7 +123,7 @@ static int pfcp_dev_init(struct net_device *dev) pfcp->dev = dev; - return 0; + return gro_cells_init(&pfcp->gro_cells, dev); } static const struct net_device_ops pfcp_netdev_ops = { @@ -94,6 +169,10 @@ static struct socket *pfcp_create_sock(struct pfcp_dev *pfcp) if (err) return ERR_PTR(err); + tuncfg.sk_user_data = pfcp; + tuncfg.encap_rcv = pfcp_encap_recv; + tuncfg.encap_type = 1; + setup_udp_tunnel_sock(net, sock, &tuncfg); return sock; diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index ed8e48cc9054..d8f574fbb11e 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -216,6 +216,7 @@ static inline void ip_tunnel_set_options_present(unsigned long *flags) __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present); __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present); __set_bit(IP_TUNNEL_GTP_OPT_BIT, present); + __set_bit(IP_TUNNEL_PFCP_OPT_BIT, present); ip_tunnel_flags_or(flags, flags, present); } @@ -228,6 +229,7 @@ static inline void ip_tunnel_clear_options_present(unsigned long *flags) __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present); __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present); __set_bit(IP_TUNNEL_GTP_OPT_BIT, present); + __set_bit(IP_TUNNEL_PFCP_OPT_BIT, present); __ipt_flag_op(bitmap_andnot, flags, flags, present); } @@ -240,6 +242,7 @@ static inline bool ip_tunnel_is_options_present(const unsigned long *flags) __set_bit(IP_TUNNEL_VXLAN_OPT_BIT, present); __set_bit(IP_TUNNEL_ERSPAN_OPT_BIT, present); __set_bit(IP_TUNNEL_GTP_OPT_BIT, present); + __set_bit(IP_TUNNEL_PFCP_OPT_BIT, present); return ip_tunnel_flags_intersect(flags, present); } diff --git a/include/net/pfcp.h b/include/net/pfcp.h index 3f9ebf27a8ff..af14f970b80e 100644 --- a/include/net/pfcp.h +++ b/include/net/pfcp.h @@ -2,12 +2,85 @@ #ifndef _PFCP_H_ #define _PFCP_H_ +#include +#include #include +#include +#include +#include +#include #include #include +#include #define PFCP_PORT 8805 +/* PFCP protocol header */ +struct pfcphdr { + u8 flags; + u8 message_type; + __be16 message_length; +}; + +/* PFCP header flags */ +#define PFCP_SEID_FLAG BIT(0) +#define PFCP_MP_FLAG BIT(1) + +#define PFCP_VERSION_MASK GENMASK(4, 0) + +#define PFCP_HLEN (sizeof(struct udphdr) + sizeof(struct pfcphdr)) + +/* PFCP node related messages */ +struct pfcphdr_node { + u8 seq_number[3]; + u8 reserved; +}; + +/* PFCP session related messages */ +struct pfcphdr_session { + __be64 seid; + u8 seq_number[3]; +#ifdef __LITTLE_ENDIAN_BITFIELD + u8 message_priority:4, + reserved:4; +#elif defined(__BIG_ENDIAN_BITFIELD) + u8 reserved:4, + message_priprity:4; +#else +#error "Please fix " +#endif +}; + +struct pfcp_metadata { + u8 type; + __be64 seid; +} __packed; + +enum { + PFCP_TYPE_NODE = 0, + PFCP_TYPE_SESSION = 1, +}; + +#define PFCP_HEADROOM (sizeof(struct iphdr) + sizeof(struct udphdr) + \ + sizeof(struct pfcphdr) + sizeof(struct ethhdr)) +#define PFCP6_HEADROOM (sizeof(struct ipv6hdr) + sizeof(struct udphdr) + \ + sizeof(struct pfcphdr) + sizeof(struct ethhdr)) + +static inline struct pfcphdr *pfcp_hdr(struct sk_buff *skb) +{ + return (struct pfcphdr *)(udp_hdr(skb) + 1); +} + +static inline struct pfcphdr_node *pfcp_hdr_node(struct sk_buff *skb) +{ + return (struct pfcphdr_node *)(pfcp_hdr(skb) + 1); +} + +static inline struct pfcphdr_session *pfcp_hdr_session(struct sk_buff *skb) +{ + return (struct pfcphdr_session *)(pfcp_hdr(skb) + 1); +} + static inline bool netif_is_pfcp(const struct net_device *dev) { return dev->rtnl_link_ops && diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index 838927dd73a4..e1a246dd8c62 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -212,6 +212,9 @@ enum { IP_TUNNEL_VTI_BIT, IP_TUNNEL_SIT_ISATAP_BIT = IP_TUNNEL_VTI_BIT, + /* Flags starting from here are not available via the old UAPI */ + IP_TUNNEL_PFCP_OPT_BIT, /* OPTIONS_PRESENT */ + __IP_TUNNEL_FLAG_NUM, }; diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index ea277039f89d..229fc925ec3a 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -587,6 +587,10 @@ enum { * TCA_FLOWER_KEY_ENC_OPT_GTP_ * attributes */ + TCA_FLOWER_KEY_ENC_OPTS_PFCP, /* Nested + * TCA_FLOWER_KEY_ENC_IPT_PFCP + * attributes + */ __TCA_FLOWER_KEY_ENC_OPTS_MAX, }; @@ -636,6 +640,16 @@ enum { #define TCA_FLOWER_KEY_ENC_OPT_GTP_MAX \ (__TCA_FLOWER_KEY_ENC_OPT_GTP_MAX - 1) +enum { + TCA_FLOWER_KEY_ENC_OPT_PFCP_UNSPEC, + TCA_FLOWER_KEY_ENC_OPT_PFCP_TYPE, /* u8 */ + TCA_FLOWER_KEY_ENC_OPT_PFCP_SEID, /* be64 */ + __TCA_FLOWER_KEY_ENC_OPT_PFCP_MAX, +}; + +#define TCA_FLOWER_KEY_ENC_OPT_PFCP_MAX \ + (__TCA_FLOWER_KEY_ENC_OPT_PFCP_MAX - 1) + enum { TCA_FLOWER_KEY_MPLS_OPTS_UNSPEC, TCA_FLOWER_KEY_MPLS_OPTS_LSE, diff --git a/net/core/net_test.c b/net/core/net_test.c index 30062b9621e4..9c3a590865d2 100644 --- a/net/core/net_test.c +++ b/net/core/net_test.c @@ -335,11 +335,8 @@ static const struct ip_tunnel_flags_test ip_tunnel_flags_test[] = { ip_tunnel_flags_1), IP_TUNNEL_FLAGS_TEST("conflict", ip_tunnel_flags_2_src, true, VTI_ISVTI, ip_tunnel_flags_2_exp), - IP_TUNNEL_FLAGS_TEST("new", ip_tunnel_flags_3_src, - /* This must be set to ``false`` once - * ``__IP_TUNNEL_FLAG_NUM`` goes above 17. - */ - true, cpu_to_be16(BIT(IP_TUNNEL_VXLAN_OPT_BIT)), + IP_TUNNEL_FLAGS_TEST("new", ip_tunnel_flags_3_src, false, + cpu_to_be16(BIT(IP_TUNNEL_VXLAN_OPT_BIT)), ip_tunnel_flags_3_exp), }; diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index ae14d649140f..fd9a6f20b60b 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -741,6 +742,7 @@ enc_opts_policy[TCA_FLOWER_KEY_ENC_OPTS_MAX + 1] = { [TCA_FLOWER_KEY_ENC_OPTS_VXLAN] = { .type = NLA_NESTED }, [TCA_FLOWER_KEY_ENC_OPTS_ERSPAN] = { .type = NLA_NESTED }, [TCA_FLOWER_KEY_ENC_OPTS_GTP] = { .type = NLA_NESTED }, + [TCA_FLOWER_KEY_ENC_OPTS_PFCP] = { .type = NLA_NESTED }, }; static const struct nla_policy @@ -770,6 +772,12 @@ gtp_opt_policy[TCA_FLOWER_KEY_ENC_OPT_GTP_MAX + 1] = { [TCA_FLOWER_KEY_ENC_OPT_GTP_QFI] = { .type = NLA_U8 }, }; +static const struct nla_policy +pfcp_opt_policy[TCA_FLOWER_KEY_ENC_OPT_PFCP_MAX + 1] = { + [TCA_FLOWER_KEY_ENC_OPT_PFCP_TYPE] = { .type = NLA_U8 }, + [TCA_FLOWER_KEY_ENC_OPT_PFCP_SEID] = { .type = NLA_U64 }, +}; + static const struct nla_policy mpls_stack_entry_policy[TCA_FLOWER_KEY_MPLS_OPT_LSE_MAX + 1] = { [TCA_FLOWER_KEY_MPLS_OPT_LSE_DEPTH] = { .type = NLA_U8 }, @@ -1419,6 +1427,44 @@ static int fl_set_gtp_opt(const struct nlattr *nla, struct fl_flow_key *key, return sizeof(*sinfo); } +static int fl_set_pfcp_opt(const struct nlattr *nla, struct fl_flow_key *key, + int depth, int option_len, + struct netlink_ext_ack *extack) +{ + struct nlattr *tb[TCA_FLOWER_KEY_ENC_OPT_PFCP_MAX + 1]; + struct pfcp_metadata *md; + int err; + + md = (struct pfcp_metadata *)&key->enc_opts.data[key->enc_opts.len]; + memset(md, 0xff, sizeof(*md)); + + if (!depth) + return sizeof(*md); + + if (nla_type(nla) != TCA_FLOWER_KEY_ENC_OPTS_PFCP) { + NL_SET_ERR_MSG_MOD(extack, "Non-pfcp option type for mask"); + return -EINVAL; + } + + err = nla_parse_nested(tb, TCA_FLOWER_KEY_ENC_OPT_PFCP_MAX, nla, + pfcp_opt_policy, extack); + if (err < 0) + return err; + + if (!option_len && !tb[TCA_FLOWER_KEY_ENC_OPT_PFCP_TYPE]) { + NL_SET_ERR_MSG_MOD(extack, "Missing tunnel key pfcp option type"); + return -EINVAL; + } + + if (tb[TCA_FLOWER_KEY_ENC_OPT_PFCP_TYPE]) + md->type = nla_get_u8(tb[TCA_FLOWER_KEY_ENC_OPT_PFCP_TYPE]); + + if (tb[TCA_FLOWER_KEY_ENC_OPT_PFCP_SEID]) + md->seid = nla_get_be64(tb[TCA_FLOWER_KEY_ENC_OPT_PFCP_SEID]); + + return sizeof(*md); +} + static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, struct fl_flow_key *mask, struct netlink_ext_ack *extack) @@ -1576,6 +1622,36 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key, return -EINVAL; } break; + case TCA_FLOWER_KEY_ENC_OPTS_PFCP: + if (key->enc_opts.dst_opt_type) { + NL_SET_ERR_MSG_MOD(extack, "Duplicate type for pfcp options"); + return -EINVAL; + } + option_len = 0; + key->enc_opts.dst_opt_type = IP_TUNNEL_PFCP_OPT_BIT; + option_len = fl_set_pfcp_opt(nla_opt_key, key, + key_depth, option_len, + extack); + if (option_len < 0) + return option_len; + + key->enc_opts.len += option_len; + /* At the same time we need to parse through the mask + * in order to verify exact and mask attribute lengths. + */ + mask->enc_opts.dst_opt_type = IP_TUNNEL_PFCP_OPT_BIT; + option_len = fl_set_pfcp_opt(nla_opt_msk, mask, + msk_depth, option_len, + extack); + if (option_len < 0) + return option_len; + + mask->enc_opts.len += option_len; + if (key->enc_opts.len != mask->enc_opts.len) { + NL_SET_ERR_MSG_MOD(extack, "Key and mask miss aligned"); + return -EINVAL; + } + break; default: NL_SET_ERR_MSG(extack, "Unknown tunnel option type"); return -EINVAL; @@ -3118,6 +3194,32 @@ nla_put_failure: return -EMSGSIZE; } +static int fl_dump_key_pfcp_opt(struct sk_buff *skb, + struct flow_dissector_key_enc_opts *enc_opts) +{ + struct pfcp_metadata *md; + struct nlattr *nest; + + nest = nla_nest_start_noflag(skb, TCA_FLOWER_KEY_ENC_OPTS_PFCP); + if (!nest) + goto nla_put_failure; + + md = (struct pfcp_metadata *)&enc_opts->data[0]; + if (nla_put_u8(skb, TCA_FLOWER_KEY_ENC_OPT_PFCP_TYPE, md->type)) + goto nla_put_failure; + + if (nla_put_be64(skb, TCA_FLOWER_KEY_ENC_OPT_PFCP_SEID, + md->seid, 0)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + static int fl_dump_key_ct(struct sk_buff *skb, struct flow_dissector_key_ct *key, struct flow_dissector_key_ct *mask) @@ -3223,6 +3325,11 @@ static int fl_dump_key_options(struct sk_buff *skb, int enc_opt_type, if (err) goto nla_put_failure; break; + case IP_TUNNEL_PFCP_OPT_BIT: + err = fl_dump_key_pfcp_opt(skb, enc_opts); + if (err) + goto nla_put_failure; + break; default: goto nla_put_failure; } -- cgit v1.2.3 From b9495b564d91a0afe4125db64d2bc25c310bda9c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:18 +0000 Subject: net: move kick_defer_list_purge() to net/core/dev.h kick_defer_list_purge() is defined in net/core/dev.c and used from net/core/skubff.c Because we need softnet_data, include from net/core/dev.h Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 - net/core/dev.h | 6 +++--- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 55c7cf9404a4..15e70527b703 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3288,7 +3288,6 @@ static inline void dev_xmit_recursion_dec(void) __this_cpu_dec(softnet_data.xmit.recursion); } -void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu); void __netif_schedule(struct Qdisc *q); void netif_schedule_queue(struct netdev_queue *txq); diff --git a/net/core/dev.h b/net/core/dev.h index 2bcaf8eee50c..9d0f8b4587f8 100644 --- a/net/core/dev.h +++ b/net/core/dev.h @@ -4,11 +4,9 @@ #include #include +#include struct net; -struct net_device; -struct netdev_bpf; -struct netdev_phys_item_id; struct netlink_ext_ack; struct cpumask; @@ -150,4 +148,6 @@ static inline void xdp_do_check_flushed(struct napi_struct *napi) { } #endif struct napi_struct *napi_by_id(unsigned int napi_id); +void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu); + #endif -- cgit v1.2.3 From 2fe50a4d7225cf10748775e290361896637091a9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:19 +0000 Subject: net: move dev_xmit_recursion() helpers to net/core/dev.h Move dev_xmit_recursion() and friends to net/core/dev.h They are only used from net/core/dev.c and net/core/filter.c. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 17 ----------------- net/core/dev.h | 17 +++++++++++++++++ 2 files changed, 17 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 15e70527b703..0cd9ee83f554 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3271,23 +3271,6 @@ static inline int dev_recursion_level(void) return this_cpu_read(softnet_data.xmit.recursion); } -#define XMIT_RECURSION_LIMIT 8 -static inline bool dev_xmit_recursion(void) -{ - return unlikely(__this_cpu_read(softnet_data.xmit.recursion) > - XMIT_RECURSION_LIMIT); -} - -static inline void dev_xmit_recursion_inc(void) -{ - __this_cpu_inc(softnet_data.xmit.recursion); -} - -static inline void dev_xmit_recursion_dec(void) -{ - __this_cpu_dec(softnet_data.xmit.recursion); -} - void __netif_schedule(struct Qdisc *q); void netif_schedule_queue(struct netdev_queue *txq); diff --git a/net/core/dev.h b/net/core/dev.h index 9d0f8b4587f8..8572d2c8dc4a 100644 --- a/net/core/dev.h +++ b/net/core/dev.h @@ -150,4 +150,21 @@ static inline void xdp_do_check_flushed(struct napi_struct *napi) { } struct napi_struct *napi_by_id(unsigned int napi_id); void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu); +#define XMIT_RECURSION_LIMIT 8 +static inline bool dev_xmit_recursion(void) +{ + return unlikely(__this_cpu_read(softnet_data.xmit.recursion) > + XMIT_RECURSION_LIMIT); +} + +static inline void dev_xmit_recursion_inc(void) +{ + __this_cpu_inc(softnet_data.xmit.recursion); +} + +static inline void dev_xmit_recursion_dec(void) +{ + __this_cpu_dec(softnet_data.xmit.recursion); +} + #endif -- cgit v1.2.3 From 95e48d862ada73188be6d91a33c49d1712815bd2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:20 +0000 Subject: net: enqueue_to_backlog() change vs not running device If the device attached to the packet given to enqueue_to_backlog() is not running, we drop the packet. But we accidentally increase sd->dropped, giving false signals to admins: sd->dropped should be reserved to cpu backlog pressure, not to temporary glitches at device dismantles. While we are at it, perform the netif_running() test before we get the rps lock, and use REASON_DEV_READY drop reason instead of NOT_SPECIFIED. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index c136e80dea61..4ad7836365e6 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4801,12 +4801,13 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, unsigned long flags; unsigned int qlen; - reason = SKB_DROP_REASON_NOT_SPECIFIED; + reason = SKB_DROP_REASON_DEV_READY; + if (!netif_running(skb->dev)) + goto bad_dev; + sd = &per_cpu(softnet_data, cpu); backlog_lock_irq_save(sd, &flags); - if (!netif_running(skb->dev)) - goto drop; qlen = skb_queue_len(&sd->input_pkt_queue); if (qlen <= READ_ONCE(net_hotdata.max_backlog) && !skb_flow_limit(skb, qlen)) { @@ -4827,10 +4828,10 @@ enqueue: } reason = SKB_DROP_REASON_CPU_BACKLOG; -drop: sd->dropped++; backlog_unlock_irq_restore(sd, &flags); +bad_dev: dev_core_stats_rx_dropped_inc(skb->dev); kfree_skb_reason(skb, reason); return NET_RX_DROP; -- cgit v1.2.3 From a7ae7b0b2ea014ff3ed4be812c3efa1b1d86e153 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:21 +0000 Subject: net: make softnet_data.dropped an atomic_t If under extreme cpu backlog pressure enqueue_to_backlog() has to drop a packet, it could do this without dirtying a cache line and potentially slowing down the target cpu. Move sd->dropped into a separate cache line, and make it atomic. In non pressure mode, this field is not touched, no need to consume valuable space in a hot cache line. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 ++- net/core/dev.c | 13 +++++++++---- net/core/net-procfs.c | 3 ++- 3 files changed, 13 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 0cd9ee83f554..be9f071a340d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3237,10 +3237,11 @@ struct softnet_data { unsigned int input_queue_tail; #endif unsigned int received_rps; - unsigned int dropped; struct sk_buff_head input_pkt_queue; struct napi_struct backlog; + atomic_t dropped ____cacheline_aligned_in_smp; + /* Another possibly contended cache line */ spinlock_t defer_lock ____cacheline_aligned_in_smp; int defer_count; diff --git a/net/core/dev.c b/net/core/dev.c index 4ad7836365e6..02c98f115243 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4800,17 +4800,22 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, struct softnet_data *sd; unsigned long flags; unsigned int qlen; + int max_backlog; reason = SKB_DROP_REASON_DEV_READY; if (!netif_running(skb->dev)) goto bad_dev; + reason = SKB_DROP_REASON_CPU_BACKLOG; sd = &per_cpu(softnet_data, cpu); + qlen = skb_queue_len_lockless(&sd->input_pkt_queue); + max_backlog = READ_ONCE(net_hotdata.max_backlog); + if (unlikely(qlen > max_backlog)) + goto cpu_backlog_drop; backlog_lock_irq_save(sd, &flags); qlen = skb_queue_len(&sd->input_pkt_queue); - if (qlen <= READ_ONCE(net_hotdata.max_backlog) && - !skb_flow_limit(skb, qlen)) { + if (qlen <= max_backlog && !skb_flow_limit(skb, qlen)) { if (qlen) { enqueue: __skb_queue_tail(&sd->input_pkt_queue, skb); @@ -4826,11 +4831,11 @@ enqueue: napi_schedule_rps(sd); goto enqueue; } - reason = SKB_DROP_REASON_CPU_BACKLOG; - sd->dropped++; backlog_unlock_irq_restore(sd, &flags); +cpu_backlog_drop: + atomic_inc(&sd->dropped); bad_dev: dev_core_stats_rx_dropped_inc(skb->dev); kfree_skb_reason(skb, reason); diff --git a/net/core/net-procfs.c b/net/core/net-procfs.c index a97eceb84e61..fa6d3969734a 100644 --- a/net/core/net-procfs.c +++ b/net/core/net-procfs.c @@ -144,7 +144,8 @@ static int softnet_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08x " "%08x %08x\n", - sd->processed, sd->dropped, sd->time_squeeze, 0, + sd->processed, atomic_read(&sd->dropped), + sd->time_squeeze, 0, 0, 0, 0, 0, /* was fastroute */ 0, /* was cpu_collision */ sd->received_rps, flow_limit_count, -- cgit v1.2.3 From f7efd01fe21431a98677e5505e5de46649121ca7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:22 +0000 Subject: net: enqueue_to_backlog() cleanup We can remove a goto and a label by reversing a condition. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 02c98f115243..0a8ccb0451c3 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4816,20 +4816,18 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, backlog_lock_irq_save(sd, &flags); qlen = skb_queue_len(&sd->input_pkt_queue); if (qlen <= max_backlog && !skb_flow_limit(skb, qlen)) { - if (qlen) { -enqueue: - __skb_queue_tail(&sd->input_pkt_queue, skb); - input_queue_tail_incr_save(sd, qtail); - backlog_unlock_irq_restore(sd, &flags); - return NET_RX_SUCCESS; + if (!qlen) { + /* Schedule NAPI for backlog device. We can use + * non atomic operation as we own the queue lock. + */ + if (!__test_and_set_bit(NAPI_STATE_SCHED, + &sd->backlog.state)) + napi_schedule_rps(sd); } - - /* Schedule NAPI for backlog device - * We can use non atomic operation since we own the queue lock - */ - if (!__test_and_set_bit(NAPI_STATE_SCHED, &sd->backlog.state)) - napi_schedule_rps(sd); - goto enqueue; + __skb_queue_tail(&sd->input_pkt_queue, skb); + input_queue_tail_incr_save(sd, qtail); + backlog_unlock_irq_restore(sd, &flags); + return NET_RX_SUCCESS; } backlog_unlock_irq_restore(sd, &flags); -- cgit v1.2.3 From 36b83ffcf209a2e6099dae1070df6a2001dfab27 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:23 +0000 Subject: net: rps: change input_queue_tail_incr_save() input_queue_tail_incr_save() is incrementing the sd queue_tail and save it in the flow last_qtail. Two issues here : - no lock protects the write on last_qtail, we should use appropriate annotations. - We can perform this write after releasing the per-cpu backlog lock, to decrease this lock hold duration (move away the cache line miss) Also move input_queue_head_incr() and rps helpers to include/net/rps.h, while adding rps_ prefix to better reflect their role. v2: Fixed a build issue (Jakub and kernel build bots) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/netdevice.h | 15 --------------- include/net/rps.h | 23 +++++++++++++++++++++++ net/core/dev.c | 20 ++++++++++++-------- 3 files changed, 35 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index be9f071a340d..2c11c295cf64 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3250,21 +3250,6 @@ struct softnet_data { call_single_data_t defer_csd; }; -static inline void input_queue_head_incr(struct softnet_data *sd) -{ -#ifdef CONFIG_RPS - sd->input_queue_head++; -#endif -} - -static inline void input_queue_tail_incr_save(struct softnet_data *sd, - unsigned int *qtail) -{ -#ifdef CONFIG_RPS - *qtail = ++sd->input_queue_tail; -#endif -} - DECLARE_PER_CPU_ALIGNED(struct softnet_data, softnet_data); static inline int dev_recursion_level(void) diff --git a/include/net/rps.h b/include/net/rps.h index 7660243e905b..10ca25731c1e 100644 --- a/include/net/rps.h +++ b/include/net/rps.h @@ -122,4 +122,27 @@ static inline void sock_rps_record_flow(const struct sock *sk) #endif } +static inline u32 rps_input_queue_tail_incr(struct softnet_data *sd) +{ +#ifdef CONFIG_RPS + return ++sd->input_queue_tail; +#else + return 0; +#endif +} + +static inline void rps_input_queue_tail_save(u32 *dest, u32 tail) +{ +#ifdef CONFIG_RPS + WRITE_ONCE(*dest, tail); +#endif +} + +static inline void rps_input_queue_head_incr(struct softnet_data *sd) +{ +#ifdef CONFIG_RPS + sd->input_queue_head++; +#endif +} + #endif /* _NET_RPS_H */ diff --git a/net/core/dev.c b/net/core/dev.c index 0a8ccb0451c3..79073bbc9a64 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4611,7 +4611,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, if (unlikely(tcpu != next_cpu) && (tcpu >= nr_cpu_ids || !cpu_online(tcpu) || ((int)(per_cpu(softnet_data, tcpu).input_queue_head - - rflow->last_qtail)) >= 0)) { + READ_ONCE(rflow->last_qtail))) >= 0)) { tcpu = next_cpu; rflow = set_rps_cpu(dev, skb, rflow, next_cpu); } @@ -4666,7 +4666,7 @@ bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, cpu = READ_ONCE(rflow->cpu); if (rflow->filter == filter_id && cpu < nr_cpu_ids && ((int)(per_cpu(softnet_data, cpu).input_queue_head - - rflow->last_qtail) < + READ_ONCE(rflow->last_qtail)) < (int)(10 * flow_table->mask))) expire = false; } @@ -4801,6 +4801,7 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, unsigned long flags; unsigned int qlen; int max_backlog; + u32 tail; reason = SKB_DROP_REASON_DEV_READY; if (!netif_running(skb->dev)) @@ -4825,8 +4826,11 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, napi_schedule_rps(sd); } __skb_queue_tail(&sd->input_pkt_queue, skb); - input_queue_tail_incr_save(sd, qtail); + tail = rps_input_queue_tail_incr(sd); backlog_unlock_irq_restore(sd, &flags); + + /* save the tail outside of the critical section */ + rps_input_queue_tail_save(qtail, tail); return NET_RX_SUCCESS; } @@ -5904,7 +5908,7 @@ static void flush_backlog(struct work_struct *work) if (skb->dev->reg_state == NETREG_UNREGISTERING) { __skb_unlink(skb, &sd->input_pkt_queue); dev_kfree_skb_irq(skb); - input_queue_head_incr(sd); + rps_input_queue_head_incr(sd); } } backlog_unlock_irq_enable(sd); @@ -5913,7 +5917,7 @@ static void flush_backlog(struct work_struct *work) if (skb->dev->reg_state == NETREG_UNREGISTERING) { __skb_unlink(skb, &sd->process_queue); kfree_skb(skb); - input_queue_head_incr(sd); + rps_input_queue_head_incr(sd); } } local_bh_enable(); @@ -6041,7 +6045,7 @@ static int process_backlog(struct napi_struct *napi, int quota) rcu_read_lock(); __netif_receive_skb(skb); rcu_read_unlock(); - input_queue_head_incr(sd); + rps_input_queue_head_incr(sd); if (++work >= quota) return work; @@ -11455,11 +11459,11 @@ static int dev_cpu_dead(unsigned int oldcpu) /* Process offline CPU's input_pkt_queue */ while ((skb = __skb_dequeue(&oldsd->process_queue))) { netif_rx(skb); - input_queue_head_incr(oldsd); + rps_input_queue_head_incr(oldsd); } while ((skb = skb_dequeue(&oldsd->input_pkt_queue))) { netif_rx(skb); - input_queue_head_incr(oldsd); + rps_input_queue_head_incr(oldsd); } return 0; -- cgit v1.2.3 From c62fdf5b11ef12b89ac2450c25c12bffc3d924c5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:42:24 +0000 Subject: net: rps: add rps_input_queue_head_add() helper process_backlog() can batch increments of sd->input_queue_head, saving some memory bandwidth. Also add READ_ONCE()/WRITE_ONCE() annotations around sd->input_queue_head accesses. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/rps.h | 9 +++++++-- net/core/dev.c | 13 ++++++++----- 2 files changed, 15 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/rps.h b/include/net/rps.h index 10ca25731c1e..a93401d23d66 100644 --- a/include/net/rps.h +++ b/include/net/rps.h @@ -138,11 +138,16 @@ static inline void rps_input_queue_tail_save(u32 *dest, u32 tail) #endif } -static inline void rps_input_queue_head_incr(struct softnet_data *sd) +static inline void rps_input_queue_head_add(struct softnet_data *sd, int val) { #ifdef CONFIG_RPS - sd->input_queue_head++; + WRITE_ONCE(sd->input_queue_head, sd->input_queue_head + val); #endif } +static inline void rps_input_queue_head_incr(struct softnet_data *sd) +{ + rps_input_queue_head_add(sd, 1); +} + #endif /* _NET_RPS_H */ diff --git a/net/core/dev.c b/net/core/dev.c index 79073bbc9a64..818699dea9d7 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4528,7 +4528,7 @@ set_rps_cpu(struct net_device *dev, struct sk_buff *skb, out: #endif rflow->last_qtail = - per_cpu(softnet_data, next_cpu).input_queue_head; + READ_ONCE(per_cpu(softnet_data, next_cpu).input_queue_head); } rflow->cpu = next_cpu; @@ -4610,7 +4610,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, */ if (unlikely(tcpu != next_cpu) && (tcpu >= nr_cpu_ids || !cpu_online(tcpu) || - ((int)(per_cpu(softnet_data, tcpu).input_queue_head - + ((int)(READ_ONCE(per_cpu(softnet_data, tcpu).input_queue_head) - READ_ONCE(rflow->last_qtail))) >= 0)) { tcpu = next_cpu; rflow = set_rps_cpu(dev, skb, rflow, next_cpu); @@ -4665,7 +4665,7 @@ bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, rflow = &flow_table->flows[flow_id]; cpu = READ_ONCE(rflow->cpu); if (rflow->filter == filter_id && cpu < nr_cpu_ids && - ((int)(per_cpu(softnet_data, cpu).input_queue_head - + ((int)(READ_ONCE(per_cpu(softnet_data, cpu).input_queue_head) - READ_ONCE(rflow->last_qtail)) < (int)(10 * flow_table->mask))) expire = false; @@ -6045,9 +6045,10 @@ static int process_backlog(struct napi_struct *napi, int quota) rcu_read_lock(); __netif_receive_skb(skb); rcu_read_unlock(); - rps_input_queue_head_incr(sd); - if (++work >= quota) + if (++work >= quota) { + rps_input_queue_head_add(sd, work); return work; + } } @@ -6070,6 +6071,8 @@ static int process_backlog(struct napi_struct *napi, int quota) backlog_unlock_irq_enable(sd); } + if (work) + rps_input_queue_head_add(sd, work); return work; } -- cgit v1.2.3 From 58169ec9c40309541509181e068177eab72e6caa Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 14:49:31 +0000 Subject: inet: preserve const qualifier in inet_csk() We can change inet_csk() to propagate its argument const qualifier, thanks to container_of_const(). We have to fix few places that had mistakes, like tcp_bound_rto(). Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240329144931.295800-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/espintcp.h | 2 +- include/net/inet_connection_sock.h | 5 +---- include/net/tcp.h | 2 +- include/net/tls.h | 2 +- net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_timer.c | 4 ++-- net/mptcp/protocol.h | 2 +- 7 files changed, 8 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/net/espintcp.h b/include/net/espintcp.h index 0335bbd76552..c70efd704b6d 100644 --- a/include/net/espintcp.h +++ b/include/net/espintcp.h @@ -32,7 +32,7 @@ struct espintcp_ctx { static inline struct espintcp_ctx *espintcp_getctx(const struct sock *sk) { - struct inet_connection_sock *icsk = inet_csk(sk); + const struct inet_connection_sock *icsk = inet_csk(sk); /* RCU is only needed for diag */ return (__force void *)icsk->icsk_ulp_data; diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index ccf171f7eb60..d96c871b0d58 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -147,10 +147,7 @@ struct inet_connection_sock { #define ICSK_TIME_LOSS_PROBE 5 /* Tail loss probe timer */ #define ICSK_TIME_REO_TIMEOUT 6 /* Reordering timer */ -static inline struct inet_connection_sock *inet_csk(const struct sock *sk) -{ - return (struct inet_connection_sock *)sk; -} +#define inet_csk(ptr) container_of_const(ptr, struct inet_connection_sock, icsk_inet.sk) static inline void *inet_csk_ca(const struct sock *sk) { diff --git a/include/net/tcp.h b/include/net/tcp.h index 6ae35199d3b3..6c27195619ab 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -742,7 +742,7 @@ int tcp_mtu_to_mss(struct sock *sk, int pmtu); int tcp_mss_to_mtu(struct sock *sk, int mss); void tcp_mtup_init(struct sock *sk); -static inline void tcp_bound_rto(const struct sock *sk) +static inline void tcp_bound_rto(struct sock *sk) { if (inet_csk(sk)->icsk_rto > TCP_RTO_MAX) inet_csk(sk)->icsk_rto = TCP_RTO_MAX; diff --git a/include/net/tls.h b/include/net/tls.h index 340ad43971e4..400f4857dc11 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -361,7 +361,7 @@ static inline bool tls_is_skb_tx_device_offloaded(const struct sk_buff *skb) static inline struct tls_context *tls_get_ctx(const struct sock *sk) { - struct inet_connection_sock *icsk = inet_csk(sk); + const struct inet_connection_sock *icsk = inet_csk(sk); /* Use RCU on icsk_ulp_data only for sock diag code, * TLS data path doesn't need rcu_dereference(). diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5d874817a78d..1b6cd3840012 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6999,7 +6999,7 @@ EXPORT_SYMBOL(inet_reqsk_alloc); /* * Return true if a syncookie should be sent */ -static bool tcp_syn_flood_action(const struct sock *sk, const char *proto) +static bool tcp_syn_flood_action(struct sock *sk, const char *proto) { struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue; const char *msg = "Dropping request"; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index d1ad20ce1c8c..976db57b95d4 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -25,7 +25,7 @@ static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk) { - struct inet_connection_sock *icsk = inet_csk(sk); + const struct inet_connection_sock *icsk = inet_csk(sk); const struct tcp_sock *tp = tcp_sk(sk); u32 elapsed, user_timeout; s32 remaining; @@ -47,7 +47,7 @@ static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk) u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when) { - struct inet_connection_sock *icsk = inet_csk(sk); + const struct inet_connection_sock *icsk = inet_csk(sk); u32 remaining, user_timeout; s32 elapsed; diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index a10ebf3ee10a..46f4655b7123 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -558,7 +558,7 @@ struct mptcp_subflow_context { static inline struct mptcp_subflow_context * mptcp_subflow_ctx(const struct sock *sk) { - struct inet_connection_sock *icsk = inet_csk(sk); + const struct inet_connection_sock *icsk = inet_csk(sk); /* Use RCU on icsk_ulp_data only for sock diag code */ return (__force struct mptcp_subflow_context *)icsk->icsk_ulp_data; -- cgit v1.2.3 From 1eeb5043573981f3a1278876515851b7f6b1df1b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 15:32:03 +0000 Subject: tcp/dccp: do not care about families in inet_twsk_purge() We lost ability to unload ipv6 module a long time ago. Instead of calling expensive inet_twsk_purge() twice, we can handle all families in one round. Also remove an extra line added in my prior patch, per Kuniyuki Iwashima feedback. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/netdev/20240327192934.6843-1-kuniyu@amazon.com/ Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240329153203.345203-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/inet_timewait_sock.h | 2 +- include/net/tcp.h | 2 +- net/dccp/ipv4.c | 2 +- net/dccp/ipv6.c | 6 ------ net/ipv4/inet_timewait_sock.c | 9 +++------ net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/tcp_minisocks.c | 6 +++--- net/ipv6/tcp_ipv6.c | 6 ------ 8 files changed, 10 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index f28da08a37b4..2a536eea9424 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -111,7 +111,7 @@ static inline void inet_twsk_reschedule(struct inet_timewait_sock *tw, int timeo void inet_twsk_deschedule_put(struct inet_timewait_sock *tw); -void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family); +void inet_twsk_purge(struct inet_hashinfo *hashinfo); static inline struct net *twsk_net(const struct inet_timewait_sock *twsk) diff --git a/include/net/tcp.h b/include/net/tcp.h index 6c27195619ab..9ab5b37e9d53 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -353,7 +353,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb); void tcp_rcv_space_adjust(struct sock *sk); int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp); void tcp_twsk_destructor(struct sock *sk); -void tcp_twsk_purge(struct list_head *net_exit_list, int family); +void tcp_twsk_purge(struct list_head *net_exit_list); ssize_t tcp_splice_read(struct socket *sk, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 44b033fe1ef6..9fc9cea4c251 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1039,7 +1039,7 @@ static void __net_exit dccp_v4_exit_net(struct net *net) static void __net_exit dccp_v4_exit_batch(struct list_head *net_exit_list) { - inet_twsk_purge(&dccp_hashinfo, AF_INET); + inet_twsk_purge(&dccp_hashinfo); } static struct pernet_operations dccp_v4_ops = { diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index ded07e09f813..c8ca703dc331 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -1119,15 +1119,9 @@ static void __net_exit dccp_v6_exit_net(struct net *net) inet_ctl_sock_destroy(pn->v6_ctl_sk); } -static void __net_exit dccp_v6_exit_batch(struct list_head *net_exit_list) -{ - inet_twsk_purge(&dccp_hashinfo, AF_INET6); -} - static struct pernet_operations dccp_v6_ops = { .init = dccp_v6_init_net, .exit = dccp_v6_exit_net, - .exit_batch = dccp_v6_exit_batch, .id = &dccp_v6_pernet_id, .size = sizeof(struct dccp_v6_pernet), }; diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index b0cc07d9a568..e28075f0006e 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -264,7 +264,7 @@ void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm) EXPORT_SYMBOL_GPL(__inet_twsk_schedule); /* Remove all non full sockets (TIME_WAIT and NEW_SYN_RECV) for dead netns */ -void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family) +void inet_twsk_purge(struct inet_hashinfo *hashinfo) { struct inet_ehash_bucket *head = &hashinfo->ehash[0]; unsigned int ehash_mask = hashinfo->ehash_mask; @@ -273,7 +273,6 @@ void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family) struct sock *sk; for (slot = 0; slot <= ehash_mask; slot++, head++) { - if (hlist_nulls_empty(&head->chain)) continue; @@ -288,15 +287,13 @@ restart: TCPF_NEW_SYN_RECV)) continue; - if (sk->sk_family != family || - refcount_read(&sock_net(sk)->ns.count)) + if (refcount_read(&sock_net(sk)->ns.count)) continue; if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt))) continue; - if (unlikely(sk->sk_family != family || - refcount_read(&sock_net(sk)->ns.count))) { + if (refcount_read(&sock_net(sk)->ns.count)) { sock_gen_put(sk); goto restart; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a22ee5838751..1e0a9762f92e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -3501,7 +3501,7 @@ static void __net_exit tcp_sk_exit_batch(struct list_head *net_exit_list) { struct net *net; - tcp_twsk_purge(net_exit_list, AF_INET); + tcp_twsk_purge(net_exit_list); list_for_each_entry(net, net_exit_list, exit_list) { inet_pernet_hashinfo_free(net->ipv4.tcp_death_row.hashinfo); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index f0761f060a83..5b21a07ddf9a 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -388,7 +388,7 @@ void tcp_twsk_destructor(struct sock *sk) } EXPORT_SYMBOL_GPL(tcp_twsk_destructor); -void tcp_twsk_purge(struct list_head *net_exit_list, int family) +void tcp_twsk_purge(struct list_head *net_exit_list) { bool purged_once = false; struct net *net; @@ -396,9 +396,9 @@ void tcp_twsk_purge(struct list_head *net_exit_list, int family) list_for_each_entry(net, net_exit_list, exit_list) { if (net->ipv4.tcp_death_row.hashinfo->pernet) { /* Even if tw_refcount == 1, we must clean up kernel reqsk */ - inet_twsk_purge(net->ipv4.tcp_death_row.hashinfo, family); + inet_twsk_purge(net->ipv4.tcp_death_row.hashinfo); } else if (!purged_once) { - inet_twsk_purge(&tcp_hashinfo, family); + inet_twsk_purge(&tcp_hashinfo); purged_once = true; } } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 3f4cba49e9ee..5ae74f661d25 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2389,15 +2389,9 @@ static void __net_exit tcpv6_net_exit(struct net *net) inet_ctl_sock_destroy(net->ipv6.tcp_sk); } -static void __net_exit tcpv6_net_exit_batch(struct list_head *net_exit_list) -{ - tcp_twsk_purge(net_exit_list, AF_INET6); -} - static struct pernet_operations tcpv6_net_ops = { .init = tcpv6_net_init, .exit = tcpv6_net_exit, - .exit_batch = tcpv6_net_exit_batch, }; int __init tcpv6_init(void) -- cgit v1.2.3 From 5bc63d3a6f466add504f283d9f743f20ca9ec334 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 29 Mar 2024 10:57:08 -0700 Subject: netlink: create a new header for internal genetlink symbols There are things in linux/genetlink.h which are only used under net/netlink/. Move them to a new local header. A new header with just 2 externs isn't great, but alternative would be to include af_netlink.h in genetlink.c which feels even worse. Link: https://lore.kernel.org/r/20240329175710.291749-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- include/linux/genetlink.h | 5 ----- net/netlink/af_netlink.c | 2 +- net/netlink/genetlink.c | 2 ++ net/netlink/genetlink.h | 11 +++++++++++ 4 files changed, 14 insertions(+), 6 deletions(-) create mode 100644 net/netlink/genetlink.h (limited to 'net') diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h index c285968e437a..9dbd7ba9b858 100644 --- a/include/linux/genetlink.h +++ b/include/linux/genetlink.h @@ -4,15 +4,10 @@ #include - /* All generic netlink requests are serialized by a global lock. */ extern void genl_lock(void); extern void genl_unlock(void); -/* for synchronisation between af_netlink and genetlink */ -extern atomic_t genl_sk_destructing_cnt; -extern wait_queue_head_t genl_sk_destructing_waitq; - #define MODULE_ALIAS_GENL_FAMILY(family)\ MODULE_ALIAS_NET_PF_PROTO_NAME(PF_NETLINK, NETLINK_GENERIC, "-family-" family) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 7554803218a2..dc8c3c01d51b 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -59,7 +59,6 @@ #include #include #include -#include #include #include #include @@ -73,6 +72,7 @@ #include #include "af_netlink.h" +#include "genetlink.h" struct listeners { struct rcu_head rcu; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 3b7666944b11..feb54c63a116 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -22,6 +22,8 @@ #include #include +#include "genetlink.h" + static DEFINE_MUTEX(genl_mutex); /* serialization of message processing */ static DECLARE_RWSEM(cb_lock); diff --git a/net/netlink/genetlink.h b/net/netlink/genetlink.h new file mode 100644 index 000000000000..89bd9d2631c3 --- /dev/null +++ b/net/netlink/genetlink.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __NET_GENETLINK_H +#define __NET_GENETLINK_H + +#include + +/* for synchronisation between af_netlink and genetlink */ +extern atomic_t genl_sk_destructing_cnt; +extern wait_queue_head_t genl_sk_destructing_waitq; + +#endif /* __LINUX_GENERIC_NETLINK_H */ -- cgit v1.2.3 From f97c9b533a1dc60a77ff329e0117acc5ae17def5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 29 Mar 2024 10:57:09 -0700 Subject: net: openvswitch: remove unnecessary linux/genetlink.h include The only legit reason I could think of for net/genetlink.h and linux/genetlink.h to be separate would be if one was included by other headers and we wanted to keep it lightweight. That is not the case, net/openvswitch/meter.h includes linux/genetlink.h but for no apparent reason (for struct genl_family perhaps? it's not necessary, types of externs do not need to be known). Link: https://lore.kernel.org/r/20240329175710.291749-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/openvswitch/meter.h | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/openvswitch/meter.h b/net/openvswitch/meter.h index ed11cd12b512..8bbf983cd244 100644 --- a/net/openvswitch/meter.h +++ b/net/openvswitch/meter.h @@ -11,7 +11,6 @@ #include #include #include -#include #include #include -- cgit v1.2.3 From cd7209628cdb2a7edd7656c126d2455e7102e949 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 29 Mar 2024 10:57:10 -0700 Subject: genetlink: remove linux/genetlink.h genetlink.h is a shell of what used to be a combined uAPI and kernel header over a decade ago. It has fewer than 10 lines of code. Merge it into net/genetlink.h. In some ways it'd be better to keep the combined header under linux/ but it would make looking through git history harder. Acked-by: Sven Eckelmann Link: https://lore.kernel.org/r/20240329175710.291749-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/wireguard/main.c | 2 +- include/linux/genetlink.h | 14 -------------- include/linux/genl_magic_struct.h | 2 +- include/net/genetlink.h | 10 +++++++++- net/batman-adv/main.c | 2 +- net/batman-adv/netlink.c | 1 - net/openvswitch/datapath.c | 1 - 7 files changed, 12 insertions(+), 20 deletions(-) delete mode 100644 include/linux/genetlink.h (limited to 'net') diff --git a/drivers/net/wireguard/main.c b/drivers/net/wireguard/main.c index ee4da9ab8013..a00671b58701 100644 --- a/drivers/net/wireguard/main.c +++ b/drivers/net/wireguard/main.c @@ -14,7 +14,7 @@ #include #include -#include +#include #include static int __init wg_mod_init(void) diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h deleted file mode 100644 index 9dbd7ba9b858..000000000000 --- a/include/linux/genetlink.h +++ /dev/null @@ -1,14 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __LINUX_GENERIC_NETLINK_H -#define __LINUX_GENERIC_NETLINK_H - -#include - -/* All generic netlink requests are serialized by a global lock. */ -extern void genl_lock(void); -extern void genl_unlock(void); - -#define MODULE_ALIAS_GENL_FAMILY(family)\ - MODULE_ALIAS_NET_PF_PROTO_NAME(PF_NETLINK, NETLINK_GENERIC, "-family-" family) - -#endif /* __LINUX_GENERIC_NETLINK_H */ diff --git a/include/linux/genl_magic_struct.h b/include/linux/genl_magic_struct.h index a419d93789ff..621b87a87d74 100644 --- a/include/linux/genl_magic_struct.h +++ b/include/linux/genl_magic_struct.h @@ -15,8 +15,8 @@ #endif #include -#include #include +#include extern int CONCATENATE(GENL_MAGIC_FAMILY, _genl_register)(void); extern void CONCATENATE(GENL_MAGIC_FAMILY, _genl_unregister)(void); diff --git a/include/net/genetlink.h b/include/net/genetlink.h index 9ece6e5a3ea8..9ab49bfeae78 100644 --- a/include/net/genetlink.h +++ b/include/net/genetlink.h @@ -2,12 +2,20 @@ #ifndef __NET_GENERIC_NETLINK_H #define __NET_GENERIC_NETLINK_H -#include +#include #include #include +#include #define GENLMSG_DEFAULT_SIZE (NLMSG_DEFAULT_SIZE - GENL_HDRLEN) +/* Non-parallel generic netlink requests are serialized by a global lock. */ +void genl_lock(void); +void genl_unlock(void); + +#define MODULE_ALIAS_GENL_FAMILY(family) \ + MODULE_ALIAS_NET_PF_PROTO_NAME(PF_NETLINK, NETLINK_GENERIC, "-family-" family) + /* Binding to multicast group requires %CAP_NET_ADMIN */ #define GENL_MCAST_CAP_NET_ADMIN BIT(0) /* Binding to multicast group requires %CAP_SYS_ADMIN */ diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 75119f1ffccc..8e0f44c71696 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include @@ -38,6 +37,7 @@ #include #include #include +#include #include #include #include diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index 0954757f0b8b..9362cd9d6f3d 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include #include diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 11c69415c605..99d72543abd3 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include #include -- cgit v1.2.3 From 5fc68320c1fb3c7d456ddcae0b4757326a043e6f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Mar 2024 18:30:53 +0000 Subject: ipv6: remove RTNL protection from inet6_dump_fib() No longer hold RTNL while calling inet6_dump_fib(). Also change return value for a completed dump, so that NLMSG_DONE can be appended to current skb, saving one recvmsg() system call. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240329183053.644630-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_fib.c | 51 ++++++++++++++++++++++++++------------------------- 1 file changed, 26 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 5c558dc1c683..284002a2a890 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -623,23 +623,22 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) struct rt6_rtnl_dump_arg arg = { .filter.dump_exceptions = true, .filter.dump_routes = true, - .filter.rtnl_held = true, + .filter.rtnl_held = false, }; const struct nlmsghdr *nlh = cb->nlh; struct net *net = sock_net(skb->sk); - unsigned int h, s_h; unsigned int e = 0, s_e; + struct hlist_head *head; struct fib6_walker *w; struct fib6_table *tb; - struct hlist_head *head; - int res = 0; + unsigned int h, s_h; + int err = 0; + rcu_read_lock(); if (cb->strict_check) { - int err; - err = ip_valid_fib_dump_req(net, nlh, &arg.filter, cb); if (err < 0) - return err; + goto unlock; } else if (nlmsg_len(nlh) >= sizeof(struct rtmsg)) { struct rtmsg *rtm = nlmsg_data(nlh); @@ -660,8 +659,10 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) * 2. allocate and initialize walker. */ w = kzalloc(sizeof(*w), GFP_ATOMIC); - if (!w) - return -ENOMEM; + if (!w) { + err = -ENOMEM; + goto unlock; + } w->func = fib6_dump_node; cb->args[2] = (long)w; } @@ -675,46 +676,46 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) tb = fib6_get_table(net, arg.filter.table_id); if (!tb) { if (rtnl_msg_family(cb->nlh) != PF_INET6) - goto out; + goto unlock; NL_SET_ERR_MSG_MOD(cb->extack, "FIB table does not exist"); - return -ENOENT; + err = -ENOENT; + goto unlock; } if (!cb->args[0]) { - res = fib6_dump_table(tb, skb, cb); - if (!res) + err = fib6_dump_table(tb, skb, cb); + if (!err) cb->args[0] = 1; } - goto out; + goto unlock; } s_h = cb->args[0]; s_e = cb->args[1]; - rcu_read_lock(); for (h = s_h; h < FIB6_TABLE_HASHSZ; h++, s_e = 0) { e = 0; head = &net->ipv6.fib_table_hash[h]; hlist_for_each_entry_rcu(tb, head, tb6_hlist) { if (e < s_e) goto next; - res = fib6_dump_table(tb, skb, cb); - if (res != 0) - goto out_unlock; + err = fib6_dump_table(tb, skb, cb); + if (err != 0) + goto out; next: e++; } } -out_unlock: - rcu_read_unlock(); +out: cb->args[1] = e; cb->args[0] = h; -out: - res = res < 0 ? res : skb->len; - if (res <= 0) + +unlock: + rcu_read_unlock(); + if (err <= 0) fib6_dump_end(cb); - return res; + return err; } void fib6_metric_set(struct fib6_info *f6i, int metric, u32 val) @@ -2506,7 +2507,7 @@ int __init fib6_init(void) goto out_kmem_cache_create; ret = rtnl_register_module(THIS_MODULE, PF_INET6, RTM_GETROUTE, NULL, - inet6_dump_fib, 0); + inet6_dump_fib, RTNL_FLAG_DUMP_UNLOCKED); if (ret) goto out_unregister_subsys; -- cgit v1.2.3 From 4a96a4e807c390a9d91b450ebe04eeb2e0ecc076 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Fri, 29 Mar 2024 17:55:06 +0100 Subject: page_pool: check for PP direct cache locality later Since we have pool->p.napi (Jakub) and pool->cpuid (Lorenzo) to check whether it's safe to use direct recycling, we can use both globally for each page instead of relying solely on @allow_direct argument. Let's assume that @allow_direct means "I'm sure it's local, don't waste time rechecking this" and when it's false, try the mentioned params to still recycle the page directly. If neither is true, we'll lose some CPU cycles, but then it surely won't be hotpath. On the other hand, paths where it's possible to use direct cache, but not possible to safely set @allow_direct, will benefit from this move. The whole propagation of @napi_safe through a dozen of skb freeing functions can now go away, which saves us some stack space. Signed-off-by: Alexander Lobakin Link: https://lore.kernel.org/r/20240329165507.3240110-2-aleksander.lobakin@intel.com Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 12 ++++----- net/core/page_pool.c | 31 +++++++++++++++++++--- net/core/skbuff.c | 70 ++++++++++++++++---------------------------------- net/ipv4/esp4.c | 2 +- net/ipv6/esp6.c | 2 +- 5 files changed, 58 insertions(+), 59 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index b7f1ecdaec38..03ea36a82cdd 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3510,25 +3510,25 @@ int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb, unsigned int headroom); int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb, struct bpf_prog *prog); -bool napi_pp_put_page(struct page *page, bool napi_safe); +bool napi_pp_put_page(struct page *page); static inline void -skb_page_unref(const struct sk_buff *skb, struct page *page, bool napi_safe) +skb_page_unref(const struct sk_buff *skb, struct page *page) { #ifdef CONFIG_PAGE_POOL - if (skb->pp_recycle && napi_pp_put_page(page, napi_safe)) + if (skb->pp_recycle && napi_pp_put_page(page)) return; #endif put_page(page); } static inline void -napi_frag_unref(skb_frag_t *frag, bool recycle, bool napi_safe) +napi_frag_unref(skb_frag_t *frag, bool recycle) { struct page *page = skb_frag_page(frag); #ifdef CONFIG_PAGE_POOL - if (recycle && napi_pp_put_page(page, napi_safe)) + if (recycle && napi_pp_put_page(page)) return; #endif put_page(page); @@ -3544,7 +3544,7 @@ napi_frag_unref(skb_frag_t *frag, bool recycle, bool napi_safe) */ static inline void __skb_frag_unref(skb_frag_t *frag, bool recycle) { - napi_frag_unref(frag, recycle, false); + napi_frag_unref(frag, recycle); } /** diff --git a/net/core/page_pool.c b/net/core/page_pool.c index dd364d738c00..9d56257e444b 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -690,8 +690,7 @@ __page_pool_put_page(struct page_pool *pool, struct page *page, page_pool_dma_sync_for_device(pool, page, dma_sync_size); - if (allow_direct && in_softirq() && - page_pool_recycle_in_cache(page, pool)) + if (allow_direct && page_pool_recycle_in_cache(page, pool)) return NULL; /* Page found as candidate for recycling */ @@ -716,9 +715,35 @@ __page_pool_put_page(struct page_pool *pool, struct page *page, return NULL; } +static bool page_pool_napi_local(const struct page_pool *pool) +{ + const struct napi_struct *napi; + u32 cpuid; + + if (unlikely(!in_softirq())) + return false; + + /* Allow direct recycle if we have reasons to believe that we are + * in the same context as the consumer would run, so there's + * no possible race. + * __page_pool_put_page() makes sure we're not in hardirq context + * and interrupts are enabled prior to accessing the cache. + */ + cpuid = smp_processor_id(); + if (READ_ONCE(pool->cpuid) == cpuid) + return true; + + napi = READ_ONCE(pool->p.napi); + + return napi && READ_ONCE(napi->list_owner) == cpuid; +} + void page_pool_put_unrefed_page(struct page_pool *pool, struct page *page, unsigned int dma_sync_size, bool allow_direct) { + if (!allow_direct) + allow_direct = page_pool_napi_local(pool); + page = __page_pool_put_page(pool, page, dma_sync_size, allow_direct); if (page && !page_pool_recycle_in_ring(pool, page)) { /* Cache full, fallback to free pages */ @@ -969,7 +994,7 @@ void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *), static void page_pool_disable_direct_recycling(struct page_pool *pool) { /* Disable direct recycling based on pool->cpuid. - * Paired with READ_ONCE() in napi_pp_put_page(). + * Paired with READ_ONCE() in page_pool_napi_local(). */ WRITE_ONCE(pool->cpuid, -1); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index a1be84be5d35..2a5ce6667bbb 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1004,11 +1004,8 @@ int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb, EXPORT_SYMBOL(skb_cow_data_for_xdp); #if IS_ENABLED(CONFIG_PAGE_POOL) -bool napi_pp_put_page(struct page *page, bool napi_safe) +bool napi_pp_put_page(struct page *page) { - bool allow_direct = false; - struct page_pool *pp; - page = compound_head(page); /* page->pp_magic is OR'ed with PP_SIGNATURE after the allocation @@ -1021,39 +1018,18 @@ bool napi_pp_put_page(struct page *page, bool napi_safe) if (unlikely(!is_pp_page(page))) return false; - pp = page->pp; - - /* Allow direct recycle if we have reasons to believe that we are - * in the same context as the consumer would run, so there's - * no possible race. - * __page_pool_put_page() makes sure we're not in hardirq context - * and interrupts are enabled prior to accessing the cache. - */ - if (napi_safe || in_softirq()) { - const struct napi_struct *napi = READ_ONCE(pp->p.napi); - unsigned int cpuid = smp_processor_id(); - - allow_direct = napi && READ_ONCE(napi->list_owner) == cpuid; - allow_direct |= READ_ONCE(pp->cpuid) == cpuid; - } - - /* Driver set this to memory recycling info. Reset it on recycle. - * This will *not* work for NIC using a split-page memory model. - * The page will be returned to the pool here regardless of the - * 'flipped' fragment being in use or not. - */ - page_pool_put_full_page(pp, page, allow_direct); + page_pool_put_full_page(page->pp, page, false); return true; } EXPORT_SYMBOL(napi_pp_put_page); #endif -static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe) +static bool skb_pp_recycle(struct sk_buff *skb, void *data) { if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle) return false; - return napi_pp_put_page(virt_to_page(data), napi_safe); + return napi_pp_put_page(virt_to_page(data)); } /** @@ -1095,12 +1071,12 @@ static void skb_kfree_head(void *head, unsigned int end_offset) kfree(head); } -static void skb_free_head(struct sk_buff *skb, bool napi_safe) +static void skb_free_head(struct sk_buff *skb) { unsigned char *head = skb->head; if (skb->head_frag) { - if (skb_pp_recycle(skb, head, napi_safe)) + if (skb_pp_recycle(skb, head)) return; skb_free_frag(head); } else { @@ -1108,8 +1084,7 @@ static void skb_free_head(struct sk_buff *skb, bool napi_safe) } } -static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason, - bool napi_safe) +static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason) { struct skb_shared_info *shinfo = skb_shinfo(skb); int i; @@ -1126,13 +1101,13 @@ static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason, } for (i = 0; i < shinfo->nr_frags; i++) - napi_frag_unref(&shinfo->frags[i], skb->pp_recycle, napi_safe); + napi_frag_unref(&shinfo->frags[i], skb->pp_recycle); free_head: if (shinfo->frag_list) kfree_skb_list_reason(shinfo->frag_list, reason); - skb_free_head(skb, napi_safe); + skb_free_head(skb); exit: /* When we clone an SKB we copy the reycling bit. The pp_recycle * bit is only set on the head though, so in order to avoid races @@ -1193,12 +1168,11 @@ void skb_release_head_state(struct sk_buff *skb) } /* Free everything but the sk_buff shell. */ -static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason, - bool napi_safe) +static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason) { skb_release_head_state(skb); if (likely(skb->head)) - skb_release_data(skb, reason, napi_safe); + skb_release_data(skb, reason); } /** @@ -1212,7 +1186,7 @@ static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason, void __kfree_skb(struct sk_buff *skb) { - skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, false); + skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED); kfree_skbmem(skb); } EXPORT_SYMBOL(__kfree_skb); @@ -1269,7 +1243,7 @@ static void kfree_skb_add_bulk(struct sk_buff *skb, return; } - skb_release_all(skb, reason, false); + skb_release_all(skb, reason); sa->skb_array[sa->skb_count++] = skb; if (unlikely(sa->skb_count == KFREE_SKB_BULK_SIZE)) { @@ -1443,7 +1417,7 @@ EXPORT_SYMBOL(consume_skb); void __consume_stateless_skb(struct sk_buff *skb) { trace_consume_skb(skb, __builtin_return_address(0)); - skb_release_data(skb, SKB_CONSUMED, false); + skb_release_data(skb, SKB_CONSUMED); kfree_skbmem(skb); } @@ -1470,7 +1444,7 @@ static void napi_skb_cache_put(struct sk_buff *skb) void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason) { - skb_release_all(skb, reason, true); + skb_release_all(skb, reason); napi_skb_cache_put(skb); } @@ -1508,7 +1482,7 @@ void napi_consume_skb(struct sk_buff *skb, int budget) return; } - skb_release_all(skb, SKB_CONSUMED, !!budget); + skb_release_all(skb, SKB_CONSUMED); napi_skb_cache_put(skb); } EXPORT_SYMBOL(napi_consume_skb); @@ -1639,7 +1613,7 @@ EXPORT_SYMBOL_GPL(alloc_skb_for_msg); */ struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src) { - skb_release_all(dst, SKB_CONSUMED, false); + skb_release_all(dst, SKB_CONSUMED); return __skb_clone(dst, src); } EXPORT_SYMBOL_GPL(skb_morph); @@ -2271,9 +2245,9 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail, if (skb_has_frag_list(skb)) skb_clone_fraglist(skb); - skb_release_data(skb, SKB_CONSUMED, false); + skb_release_data(skb, SKB_CONSUMED); } else { - skb_free_head(skb, false); + skb_free_head(skb); } off = (data + nhead) - skb->head; @@ -6574,12 +6548,12 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off, skb_frag_ref(skb, i); if (skb_has_frag_list(skb)) skb_clone_fraglist(skb); - skb_release_data(skb, SKB_CONSUMED, false); + skb_release_data(skb, SKB_CONSUMED); } else { /* we can reuse existing recount- all we did was * relocate values */ - skb_free_head(skb, false); + skb_free_head(skb); } skb->head = data; @@ -6714,7 +6688,7 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off, skb_kfree_head(data, size); return -ENOMEM; } - skb_release_data(skb, SKB_CONSUMED, false); + skb_release_data(skb, SKB_CONSUMED); skb->head = data; skb->head_frag = 0; diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index d33d12421814..3d647c9a7a21 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -114,7 +114,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) */ if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) - skb_page_unref(skb, sg_page(sg), false); + skb_page_unref(skb, sg_page(sg)); } #ifdef CONFIG_INET_ESPINTCP diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 7371886d4f9f..fe8d53f5a5ee 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -131,7 +131,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) */ if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) - skb_page_unref(skb, sg_page(sg), false); + skb_page_unref(skb, sg_page(sg)); } #ifdef CONFIG_INET6_ESPINTCP -- cgit v1.2.3 From 39806b96c89ae5d52092c8f86393ecbfaae26697 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Fri, 29 Mar 2024 17:55:07 +0100 Subject: page_pool: try direct bulk recycling Now that the checks for direct recycling possibility live inside the Page Pool core, reuse them when performing bulk recycling. page_pool_put_page_bulk() can be called from process context as well, page_pool_napi_local() takes care of this at the very beginning. Under high .ndo_xdp_xmit() traffic load, the win is 2-3% Pps assuming the sending driver uses xdp_return_frame_bulk() on Tx completion. Signed-off-by: Alexander Lobakin Link: https://lore.kernel.org/r/20240329165507.3240110-3-aleksander.lobakin@intel.com Signed-off-by: Jakub Kicinski --- net/core/page_pool.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 9d56257e444b..4c175091fc0a 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -772,8 +772,11 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data, int count) { int i, bulk_len = 0; + bool allow_direct; bool in_softirq; + allow_direct = page_pool_napi_local(pool); + for (i = 0; i < count; i++) { struct page *page = virt_to_head_page(data[i]); @@ -781,13 +784,13 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data, if (!page_pool_is_last_ref(page)) continue; - page = __page_pool_put_page(pool, page, -1, false); + page = __page_pool_put_page(pool, page, -1, allow_direct); /* Approved for bulk recycling in ptr_ring cache */ if (page) data[bulk_len++] = page; } - if (unlikely(!bulk_len)) + if (!bulk_len) return; /* Bulk producer into ptr_ring page_pool cache */ -- cgit v1.2.3 From be4e1304419c99a164b4c0e101c7c2a756b635b9 Mon Sep 17 00:00:00 2001 From: Dawei Li Date: Sun, 31 Mar 2024 13:34:40 +0800 Subject: net/iucv: Avoid explicit cpumask var allocation on stack For CONFIG_CPUMASK_OFFSTACK=y kernel, explicit allocation of cpumask variable on stack is not recommended since it can cause potential stack overflow. Instead, kernel code should always use *cpumask_var API(s) to allocate cpumask var in config-neutral way, leaving allocation strategy to CONFIG_CPUMASK_OFFSTACK. Use *cpumask_var API(s) to address it. Signed-off-by: Dawei Li Reviewed-by: Alexandra Winter Link: https://lore.kernel.org/r/20240331053441.1276826-2-dawei.li@shingroup.cn Signed-off-by: Jakub Kicinski --- net/iucv/iucv.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index a4ab615ca3e3..5e37a8ceebcb 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -520,7 +520,7 @@ static void iucv_setmask_mp(void) */ static void iucv_setmask_up(void) { - cpumask_t cpumask; + static cpumask_t cpumask; int cpu; /* Disable all cpu but the first in cpu_irq_cpumask. */ @@ -628,23 +628,33 @@ static int iucv_cpu_online(unsigned int cpu) static int iucv_cpu_down_prep(unsigned int cpu) { - cpumask_t cpumask; + cpumask_var_t cpumask; + int ret = 0; if (!iucv_path_table) return 0; - cpumask_copy(&cpumask, &iucv_buffer_cpumask); - cpumask_clear_cpu(cpu, &cpumask); - if (cpumask_empty(&cpumask)) + if (!alloc_cpumask_var(&cpumask, GFP_KERNEL)) + return -ENOMEM; + + cpumask_copy(cpumask, &iucv_buffer_cpumask); + cpumask_clear_cpu(cpu, cpumask); + if (cpumask_empty(cpumask)) { /* Can't offline last IUCV enabled cpu. */ - return -EINVAL; + ret = -EINVAL; + goto __free_cpumask; + } iucv_retrieve_cpu(NULL); if (!cpumask_empty(&iucv_irq_cpumask)) - return 0; + goto __free_cpumask; + smp_call_function_single(cpumask_first(&iucv_buffer_cpumask), iucv_allow_cpu, NULL, 1); - return 0; + +__free_cpumask: + free_cpumask_var(cpumask); + return ret; } /** -- cgit v1.2.3 From f9a4506438a1068d9ff613319bc64dcaa65f681d Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 30 Mar 2024 09:32:12 +0100 Subject: caif: Use UTILITY_NAME_LENGTH instead of hard-coding 16 UTILITY_NAME_LENGTH is 16. So better use the former when defining the 'utility_name' array. This makes the intent clearer when it is used around line 260. While at it, declare variable in reverse xmas tree style. Signed-off-by: Christophe JAILLET Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/8c1160501f69b64bb2d45ce9f26f746eec80ac77.1711787352.git.christophe.jaillet@wanadoo.fr Signed-off-by: Jakub Kicinski --- net/caif/cfctrl.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/caif/cfctrl.c b/net/caif/cfctrl.c index 8480684f2762..20139fa1be1f 100644 --- a/net/caif/cfctrl.c +++ b/net/caif/cfctrl.c @@ -201,14 +201,14 @@ int cfctrl_linkup_request(struct cflayer *layer, struct cflayer *user_layer) { struct cfctrl *cfctrl = container_obj(layer); + struct cflayer *dn = cfctrl->serv.layer.dn; + char utility_name[UTILITY_NAME_LENGTH]; + struct cfctrl_request_info *req; + struct cfpkt *pkt; u32 tmp32; u16 tmp16; u8 tmp8; - struct cfctrl_request_info *req; int ret; - char utility_name[16]; - struct cfpkt *pkt; - struct cflayer *dn = cfctrl->serv.layer.dn; if (!dn) { pr_debug("not able to send linkup request\n"); -- cgit v1.2.3 From 9748dbc9f26541d35c4231348c64499bf7a19735 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Thu, 28 Mar 2024 15:52:49 -0600 Subject: net/smc: Avoid -Wflex-array-member-not-at-end warnings -Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. There are currently a couple of objects in `struct smc_clc_msg_proposal_area` that contain a couple of flexible structures: struct smc_clc_msg_proposal_area { ... struct smc_clc_v2_extension pclc_v2_ext; ... struct smc_clc_smcd_v2_extension pclc_smcd_v2_ext; ... }; So, in order to avoid ending up with a couple of flexible-array members in the middle of a struct, we use the `struct_group_tagged()` helper to separate the flexible array from the rest of the members in the flexible structure: struct smc_clc_smcd_v2_extension { struct_group_tagged(smc_clc_smcd_v2_extension_fixed, fixed, u8 system_eid[SMC_MAX_EID_LEN]; u8 reserved[16]; ); struct smc_clc_smcd_gid_chid gidchid[]; }; With the change described above, we now declare objects of the type of the tagged struct without embedding flexible arrays in the middle of another struct: struct smc_clc_msg_proposal_area { ... struct smc_clc_v2_extension_fixed pclc_v2_ext; ... struct smc_clc_smcd_v2_extension_fixed pclc_smcd_v2_ext; ... }; We also use `container_of()` when we need to retrieve a pointer to the flexible structures. So, with these changes, fix the following warnings: In file included from net/smc/af_smc.c:42: net/smc/smc_clc.h:186:49: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] 186 | struct smc_clc_v2_extension pclc_v2_ext; | ^~~~~~~~~~~ net/smc/smc_clc.h:188:49: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] 188 | struct smc_clc_smcd_v2_extension pclc_smcd_v2_ext; | ^~~~~~~~~~~~~~~~ Reviewed-by: Kees Cook Signed-off-by: Gustavo A. R. Silva Reviewed-by: Wen Gu Signed-off-by: David S. Miller --- net/smc/smc_clc.c | 6 ++++-- net/smc/smc_clc.h | 26 ++++++++++++++++---------- 2 files changed, 20 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/smc/smc_clc.c b/net/smc/smc_clc.c index e55026c7529c..33fa787c28eb 100644 --- a/net/smc/smc_clc.c +++ b/net/smc/smc_clc.c @@ -853,8 +853,10 @@ int smc_clc_send_proposal(struct smc_sock *smc, struct smc_init_info *ini) pclc_smcd = &pclc->pclc_smcd; pclc_prfx = &pclc->pclc_prfx; ipv6_prfx = pclc->pclc_prfx_ipv6; - v2_ext = &pclc->pclc_v2_ext; - smcd_v2_ext = &pclc->pclc_smcd_v2_ext; + v2_ext = container_of(&pclc->pclc_v2_ext, + struct smc_clc_v2_extension, fixed); + smcd_v2_ext = container_of(&pclc->pclc_smcd_v2_ext, + struct smc_clc_smcd_v2_extension, fixed); gidchids = pclc->pclc_gidchids; trl = &pclc->pclc_trl; diff --git a/net/smc/smc_clc.h b/net/smc/smc_clc.h index 7cc7070b9772..467effb50cd6 100644 --- a/net/smc/smc_clc.h +++ b/net/smc/smc_clc.h @@ -134,12 +134,15 @@ struct smc_clc_smcd_gid_chid { */ struct smc_clc_v2_extension { - struct smc_clnt_opts_area_hdr hdr; - u8 roce[16]; /* RoCEv2 GID */ - u8 max_conns; - u8 max_links; - __be16 feature_mask; - u8 reserved[12]; + /* New members must be added within the struct_group() macro below. */ + struct_group_tagged(smc_clc_v2_extension_fixed, fixed, + struct smc_clnt_opts_area_hdr hdr; + u8 roce[16]; /* RoCEv2 GID */ + u8 max_conns; + u8 max_links; + __be16 feature_mask; + u8 reserved[12]; + ); u8 user_eids[][SMC_MAX_EID_LEN]; }; @@ -159,8 +162,11 @@ struct smc_clc_msg_smcd { /* SMC-D GID information */ }; struct smc_clc_smcd_v2_extension { - u8 system_eid[SMC_MAX_EID_LEN]; - u8 reserved[16]; + /* New members must be added within the struct_group() macro below. */ + struct_group_tagged(smc_clc_smcd_v2_extension_fixed, fixed, + u8 system_eid[SMC_MAX_EID_LEN]; + u8 reserved[16]; + ); struct smc_clc_smcd_gid_chid gidchid[]; }; @@ -183,9 +189,9 @@ struct smc_clc_msg_proposal_area { struct smc_clc_msg_smcd pclc_smcd; struct smc_clc_msg_proposal_prefix pclc_prfx; struct smc_clc_ipv6_prefix pclc_prfx_ipv6[SMC_CLC_MAX_V6_PREFIX]; - struct smc_clc_v2_extension pclc_v2_ext; + struct smc_clc_v2_extension_fixed pclc_v2_ext; u8 user_eids[SMC_CLC_MAX_UEID][SMC_MAX_EID_LEN]; - struct smc_clc_smcd_v2_extension pclc_smcd_v2_ext; + struct smc_clc_smcd_v2_extension_fixed pclc_smcd_v2_ext; struct smc_clc_smcd_gid_chid pclc_gidchids[SMCD_CLC_MAX_V2_GID_ENTRIES]; struct smc_clc_msg_trail pclc_trl; -- cgit v1.2.3 From 2f3c7195a702178629575b637f3f9a91eabdab73 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Mon, 1 Apr 2024 12:36:40 -0300 Subject: net: enable timestamp static key if CPU For systems that use CPU isolation (via nohz_full), creating or destroying a socket with SO_TIMESTAMP, SO_TIMESTAMPNS or SO_TIMESTAMPING with flag SOF_TIMESTAMPING_RX_SOFTWARE will cause a static key to be enabled/disabled. This in turn causes undesired IPIs to isolated CPUs. So enable the static key unconditionally, if CPU isolation is enabled, thus avoiding the IPIs. Signed-off-by: Marcelo Tosatti Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/ZgrUiLLtbEUf9SFn@tpad Signed-off-by: Jakub Kicinski --- net/core/dev.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 818699dea9d7..9b821d96eff3 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -77,6 +77,7 @@ #include #include #include +#include #include #include #include @@ -11911,6 +11912,10 @@ static int __init net_dev_init(void) NULL, dev_cpu_dead); WARN_ON(rc < 0); rc = 0; + + /* avoid static key IPIs to isolated CPUs */ + if (housekeeping_enabled(HK_TYPE_MISC)) + net_enable_timestamp(); out: if (rc < 0) { for_each_possible_cpu(i) { -- cgit v1.2.3 From 19822a980e1956a6572998887a7df5a0607a32f6 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Mon, 1 Apr 2024 15:36:05 +0800 Subject: trace: tcp: fully support trace_tcp_send_reset Prior to this patch, what we can see by enabling trace_tcp_send is only happening under two circumstances: 1) active rst mode 2) non-active rst mode and based on the full socket That means the inconsistency occurs if we use tcpdump and trace simultaneously to see how rst happens. It's necessary that we should take into other cases into considerations, say: 1) time-wait socket 2) no socket ... By parsing the incoming skb and reversing its 4-tuple can we know the exact 'flow' which might not exist. Samples after applied this patch: 1. tcp_send_reset: skbaddr=XXX skaddr=XXX src=ip:port dest=ip:port state=TCP_ESTABLISHED 2. tcp_send_reset: skbaddr=000...000 skaddr=XXX src=ip:port dest=ip:port state=UNKNOWN Note: 1) UNKNOWN means we cannot extract the right information from skb. 2) skbaddr/skaddr could be 0 Signed-off-by: Jason Xing Link: https://lore.kernel.org/r/20240401073605.37335-3-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- include/trace/events/tcp.h | 40 ++++++++++++++++++++++++++++++++++++++-- net/ipv4/tcp_ipv4.c | 7 +++---- net/ipv6/tcp_ipv6.c | 3 ++- 3 files changed, 43 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h index cf14b6fcbeed..5c04a61a11c2 100644 --- a/include/trace/events/tcp.h +++ b/include/trace/events/tcp.h @@ -78,11 +78,47 @@ DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb, * skb of trace_tcp_send_reset is the skb that caused RST. In case of * active reset, skb should be NULL */ -DEFINE_EVENT(tcp_event_sk_skb, tcp_send_reset, +TRACE_EVENT(tcp_send_reset, TP_PROTO(const struct sock *sk, const struct sk_buff *skb), - TP_ARGS(sk, skb) + TP_ARGS(sk, skb), + + TP_STRUCT__entry( + __field(const void *, skbaddr) + __field(const void *, skaddr) + __field(int, state) + __array(__u8, saddr, sizeof(struct sockaddr_in6)) + __array(__u8, daddr, sizeof(struct sockaddr_in6)) + ), + + TP_fast_assign( + __entry->skbaddr = skb; + __entry->skaddr = sk; + /* Zero means unknown state. */ + __entry->state = sk ? sk->sk_state : 0; + + memset(__entry->saddr, 0, sizeof(struct sockaddr_in6)); + memset(__entry->daddr, 0, sizeof(struct sockaddr_in6)); + + if (sk && sk_fullsock(sk)) { + const struct inet_sock *inet = inet_sk(sk); + + TP_STORE_ADDR_PORTS(__entry, inet, sk); + } else if (skb) { + const struct tcphdr *th = (const struct tcphdr *)skb->data; + /* + * We should reverse the 4-tuple of skb, so later + * it can print the right flow direction of rst. + */ + TP_STORE_ADDR_PORTS_SKB(skb, th, entry->daddr, entry->saddr); + } + ), + + TP_printk("skbaddr=%p skaddr=%p src=%pISpc dest=%pISpc state=%s", + __entry->skbaddr, __entry->skaddr, + __entry->saddr, __entry->daddr, + __entry->state ? show_tcp_state_name(__entry->state) : "UNKNOWN") ); /* diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1e0a9762f92e..56b75efcfd12 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -866,11 +866,10 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) * routing might fail in this case. No choice here, if we choose to force * input interface, we will misroute in case of asymmetric route. */ - if (sk) { + if (sk) arg.bound_dev_if = sk->sk_bound_dev_if; - if (sk_fullsock(sk)) - trace_tcp_send_reset(sk, skb); - } + + trace_tcp_send_reset(sk, skb); BUILD_BUG_ON(offsetof(struct sock, sk_bound_dev_if) != offsetof(struct inet_timewait_sock, tw_bound_dev_if)); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 5ae74f661d25..cffebaec66f1 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1113,7 +1113,6 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb) if (sk) { oif = sk->sk_bound_dev_if; if (sk_fullsock(sk)) { - trace_tcp_send_reset(sk, skb); if (inet6_test_bit(REPFLOW, sk)) label = ip6_flowlabel(ipv6h); priority = READ_ONCE(sk->sk_priority); @@ -1129,6 +1128,8 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb) label = ip6_flowlabel(ipv6h); } + trace_tcp_send_reset(sk, skb); + tcp_v6_send_response(sk, skb, seq, ack_seq, 0, 0, 0, oif, 1, ipv6_get_dsfield(ipv6h), label, priority, txhash, &key); -- cgit v1.2.3 From 7c349ed090318b1c88a3e5dff3b24f732296edce Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 1 Apr 2024 10:31:24 -0700 Subject: af_unix: Remove scm_fp_dup() in unix_attach_fds(). When we passed fds, we used to bump each file's refcount twice in scm_fp_copy() and scm_fp_dup() before linking the socket to gc_inflight_list. This is because we incremented the inflight count of the socket and linked it to the list in advance before passing skb to the destination socket. Otherwise, the inflight socket could have been garbage-collected in a small race window between linking the socket to the list and queuing skb: CPU 1 : sendmsg(X) w/ A's fd CPU 2 : close(A) ----- ----- /* Here A's refcount is 1, and inflight count is 0 */ bump A's refcount to 2 in scm_fp_copy() bump A's inflight count to 1 link A to gc_inflight_list decrement A's refcount to 1 /* A's refcount == inflight count, thus A could be GC candidate */ start GC mark A as candidate purge A's receive queue queue skb w/ A's fd to X /* A is queued, but all data has been lost */ After commit 4090fa373f0e ("af_unix: Replace garbage collection algorithm."), we increment the inflight count and link the socket to the global list only when queuing the skb. The race no longer exists, so let's not clone the fd nor bump the count in unix_attach_fds(). Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240401173125.92184-2-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 533fb682c954..78be8b520cef 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1794,13 +1794,8 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb) if (too_many_unix_fds(current)) return -ETOOMANYREFS; - /* Need to duplicate file references for the sake of garbage - * collection. Otherwise a socket in the fps might become a - * candidate for GC while the skb is not yet queued. - */ - UNIXCB(skb).fp = scm_fp_dup(scm->fp); - if (!UNIXCB(skb).fp) - return -ENOMEM; + UNIXCB(skb).fp = scm->fp; + scm->fp = NULL; if (unix_prepare_fpl(UNIXCB(skb).fp)) return -ENOMEM; -- cgit v1.2.3 From 118f457da9ed58a79e24b73c2ef0aa1987241f0e Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 1 Apr 2024 10:31:25 -0700 Subject: af_unix: Remove lock dance in unix_peek_fds(). In the previous GC implementation, the shape of the inflight socket graph was not expected to change while GC was in progress. MSG_PEEK was tricky because it could install inflight fd silently and transform the graph. Let's say we peeked a fd, which was a listening socket, and accept()ed some embryo sockets from it. The garbage collection algorithm would have been confused because the set of sockets visited in scan_inflight() would change within the same GC invocation. That's why we placed spin_lock(&unix_gc_lock) and spin_unlock() in unix_peek_fds() with a fat comment. In the new GC implementation, we no longer garbage-collect the socket if it exists in another queue, that is, if it has a bridge to another SCC. Also, accept() will require the lock if it has edges. Thus, we need not do the complicated lock dance. Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240401173125.92184-3-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/af_unix.h | 1 - net/unix/af_unix.c | 42 ------------------------------------------ net/unix/garbage.c | 2 +- 3 files changed, 1 insertion(+), 44 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 226a8da2cbe3..7311b77edfc7 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -17,7 +17,6 @@ static inline struct unix_sock *unix_get_socket(struct file *filp) } #endif -extern spinlock_t unix_gc_lock; extern unsigned int unix_tot_inflight; void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver); void unix_del_edges(struct scm_fp_list *fpl); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 78be8b520cef..61ecfa9c9c6b 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1814,48 +1814,6 @@ static void unix_detach_fds(struct scm_cookie *scm, struct sk_buff *skb) static void unix_peek_fds(struct scm_cookie *scm, struct sk_buff *skb) { scm->fp = scm_fp_dup(UNIXCB(skb).fp); - - /* - * Garbage collection of unix sockets starts by selecting a set of - * candidate sockets which have reference only from being in flight - * (total_refs == inflight_refs). This condition is checked once during - * the candidate collection phase, and candidates are marked as such, so - * that non-candidates can later be ignored. While inflight_refs is - * protected by unix_gc_lock, total_refs (file count) is not, hence this - * is an instantaneous decision. - * - * Once a candidate, however, the socket must not be reinstalled into a - * file descriptor while the garbage collection is in progress. - * - * If the above conditions are met, then the directed graph of - * candidates (*) does not change while unix_gc_lock is held. - * - * Any operations that changes the file count through file descriptors - * (dup, close, sendmsg) does not change the graph since candidates are - * not installed in fds. - * - * Dequeing a candidate via recvmsg would install it into an fd, but - * that takes unix_gc_lock to decrement the inflight count, so it's - * serialized with garbage collection. - * - * MSG_PEEK is special in that it does not change the inflight count, - * yet does install the socket into an fd. The following lock/unlock - * pair is to ensure serialization with garbage collection. It must be - * done between incrementing the file count and installing the file into - * an fd. - * - * If garbage collection starts after the barrier provided by the - * lock/unlock, then it will see the elevated refcount and not mark this - * as a candidate. If a garbage collection is already in progress - * before the file count was incremented, then the lock/unlock pair will - * ensure that garbage collection is finished before progressing to - * installing the fd. - * - * (*) A -> B where B is on the queue of A or B is on the queue of C - * which is on the queue of listening socket A. - */ - spin_lock(&unix_gc_lock); - spin_unlock(&unix_gc_lock); } static void unix_destruct_scm(struct sk_buff *skb) diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 89ea71d9297b..12a4ec27e0d4 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -183,7 +183,7 @@ static void unix_free_vertices(struct scm_fp_list *fpl) } } -DEFINE_SPINLOCK(unix_gc_lock); +static DEFINE_SPINLOCK(unix_gc_lock); unsigned int unix_tot_inflight; void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) -- cgit v1.2.3 From 8e088a20dbe33919695a8082c0b32deb62d23b4a Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Tue, 2 Apr 2024 14:41:38 -0400 Subject: SUNRPC: add a missing rpc_stat for TCP TLS Commit 1548036ef120 ("nfs: make the rpc_stat per net namespace") added functionality to specify rpc_stats function but missed adding it to the TCP TLS functionality. As the result, mounting with xprtsec=tls lead to the following kernel oops. [ 128.984192] Unable to handle kernel NULL pointer dereference at virtual address 000000000000001c [ 128.985058] Mem abort info: [ 128.985372] ESR = 0x0000000096000004 [ 128.985709] EC = 0x25: DABT (current EL), IL = 32 bits [ 128.986176] SET = 0, FnV = 0 [ 128.986521] EA = 0, S1PTW = 0 [ 128.986804] FSC = 0x04: level 0 translation fault [ 128.987229] Data abort info: [ 128.987597] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000 [ 128.988169] CM = 0, WnR = 0, TnD = 0, TagAccess = 0 [ 128.988811] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 [ 128.989302] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000106c84000 [ 128.990048] [000000000000001c] pgd=0000000000000000, p4d=0000000000000000 [ 128.990736] Internal error: Oops: 0000000096000004 [#1] SMP [ 128.991168] Modules linked in: nfs_layout_nfsv41_files rpcsec_gss_krb5 auth_rpcgss nfsv4 dns_resolver nfs lockd grace netfs uinput dm_mod nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 rfkill ip_set nf_tables nfnetlink qrtr vsock_loopback vmw_vsock_virtio_transport_common vmw_vsock_vmci_transport vsock sunrpc vfat fat uvcvideo videobuf2_vmalloc videobuf2_memops uvc videobuf2_v4l2 videodev videobuf2_common mc vmw_vmci xfs libcrc32c e1000e crct10dif_ce ghash_ce sha2_ce vmwgfx nvme sha256_arm64 nvme_core sr_mod cdrom sha1_ce drm_ttm_helper ttm drm_kms_helper drm sg fuse [ 128.996466] CPU: 0 PID: 179 Comm: kworker/u4:26 Kdump: loaded Not tainted 6.8.0-rc6+ #12 [ 128.997226] Hardware name: VMware, Inc. VMware20,1/VBSA, BIOS VMW201.00V.21805430.BA64.2305221830 05/22/2023 [ 128.998084] Workqueue: xprtiod xs_tcp_tls_setup_socket [sunrpc] [ 128.998701] pstate: 81400005 (Nzcv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--) [ 128.999384] pc : call_start+0x74/0x138 [sunrpc] [ 128.999809] lr : __rpc_execute+0xb8/0x3e0 [sunrpc] [ 129.000244] sp : ffff8000832b3a00 [ 129.000508] x29: ffff8000832b3a00 x28: ffff800081ac79c0 x27: ffff800081ac7000 [ 129.001111] x26: 0000000004248060 x25: 0000000000000000 x24: ffff800081596008 [ 129.001757] x23: ffff80007b087240 x22: ffff00009a509d30 x21: 0000000000000000 [ 129.002345] x20: ffff000090075600 x19: ffff00009a509d00 x18: ffffffffffffffff [ 129.002912] x17: 733d4d4554535953 x16: 42555300312d746e x15: ffff8000832b3a88 [ 129.003464] x14: ffffffffffffffff x13: ffff8000832b3a7d x12: 0000000000000008 [ 129.004021] x11: 0101010101010101 x10: ffff8000150cb560 x9 : ffff80007b087c00 [ 129.004577] x8 : ffff00009a509de0 x7 : 0000000000000000 x6 : 00000000be8c4ee3 [ 129.005026] x5 : 0000000000000000 x4 : 0000000000000000 x3 : ffff000094d56680 [ 129.005425] x2 : ffff80007b0637f8 x1 : ffff000090075600 x0 : ffff00009a509d00 [ 129.005824] Call trace: [ 129.005967] call_start+0x74/0x138 [sunrpc] [ 129.006233] __rpc_execute+0xb8/0x3e0 [sunrpc] [ 129.006506] rpc_execute+0x160/0x1d8 [sunrpc] [ 129.006778] rpc_run_task+0x148/0x1f8 [sunrpc] [ 129.007204] tls_probe+0x80/0xd0 [sunrpc] [ 129.007460] rpc_ping+0x28/0x80 [sunrpc] [ 129.007715] rpc_create_xprt+0x134/0x1a0 [sunrpc] [ 129.007999] rpc_create+0x128/0x2a0 [sunrpc] [ 129.008264] xs_tcp_tls_setup_socket+0xdc/0x508 [sunrpc] [ 129.008583] process_one_work+0x174/0x3c8 [ 129.008813] worker_thread+0x2c8/0x3e0 [ 129.009033] kthread+0x100/0x110 [ 129.009225] ret_from_fork+0x10/0x20 [ 129.009432] Code: f0ffffc2 911fe042 aa1403e1 aa1303e0 (b9401c83) Fixes: 1548036ef120 ("nfs: make the rpc_stat per net namespace") Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index bb81050c870e..ea7bf1b4172d 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2664,6 +2664,7 @@ static void xs_tcp_tls_setup_socket(struct work_struct *work) .xprtsec = { .policy = RPC_XPRTSEC_NONE, }, + .stats = upper_clnt->cl_stats, }; unsigned int pflags = current->flags; struct rpc_clnt *lower_clnt; -- cgit v1.2.3 From 0e9c127729be19adcdf2f5cc1f3450a4322fdf3a Mon Sep 17 00:00:00 2001 From: Rahul Rameshbabu Date: Wed, 3 Apr 2024 14:28:39 -0700 Subject: ethtool: add interface to read Tx hardware timestamping statistics Multiple network devices that support hardware timestamping appear to have common behavior with regards to timestamp handling. Implement common Tx hardware timestamping statistics in a tx_stats struct_group. Common Rx hardware timestamping statistics can subsequently be implemented in a rx_stats struct_group for ethtool_ts_stats. Signed-off-by: Rahul Rameshbabu Reviewed-by: Dragos Tatulea Link: https://lore.kernel.org/r/20240403212931.128541-2-rrameshbabu@nvidia.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/ethtool.yaml | 17 +++++++++ Documentation/networking/ethtool-netlink.rst | 9 +++++ include/linux/ethtool.h | 27 ++++++++++++++- include/uapi/linux/ethtool_netlink.h | 14 ++++++++ net/ethtool/tsinfo.c | 52 +++++++++++++++++++++++++++- 5 files changed, 117 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml index d0e4a47e0f21..b76916394ec9 100644 --- a/Documentation/netlink/specs/ethtool.yaml +++ b/Documentation/netlink/specs/ethtool.yaml @@ -565,6 +565,18 @@ attribute-sets: - name: tx-lpi-timer type: u32 + - + name: ts-stat + attributes: + - + name: tx-pkts + type: uint + - + name: tx-lost + type: uint + - + name: tx-err + type: uint - name: tsinfo attributes: @@ -587,6 +599,10 @@ attribute-sets: - name: phc-index type: u32 + - + name: stats + type: nest + nested-attributes: ts-stat - name: cable-result attributes: @@ -1394,6 +1410,7 @@ operations: - tx-types - rx-filters - phc-index + - stats dump: *tsinfo-get-op - name: cable-test-act diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index d583d9abf2f8..08d330b0f50f 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -1237,12 +1237,21 @@ Kernel response contents: ``ETHTOOL_A_TSINFO_TX_TYPES`` bitset supported Tx types ``ETHTOOL_A_TSINFO_RX_FILTERS`` bitset supported Rx filters ``ETHTOOL_A_TSINFO_PHC_INDEX`` u32 PTP hw clock index + ``ETHTOOL_A_TSINFO_STATS`` nested HW timestamping statistics ===================================== ====== ========================== ``ETHTOOL_A_TSINFO_PHC_INDEX`` is absent if there is no associated PHC (there is no special value for this case). The bitset attributes are omitted if they would be empty (no bit set). +Additional hardware timestamping statistics response contents: + + ===================================== ====== =================================== + ``ETHTOOL_A_TS_STAT_TX_PKTS`` u64 Packets with Tx HW timestamps + ``ETHTOOL_A_TS_STAT_TX_LOST`` u64 Tx HW timestamp not arrived count + ``ETHTOOL_A_TS_STAT_TX_ERR`` u64 HW error request Tx timestamp count + ===================================== ====== =================================== + CABLE_TEST ========== diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 9901e563f706..6fd9107d3cc0 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -480,6 +480,26 @@ struct ethtool_rmon_stats { ); }; +/** + * struct ethtool_ts_stats - HW timestamping statistics + * @pkts: Number of packets successfully timestamped by the hardware. + * @lost: Number of hardware timestamping requests where the timestamping + * information from the hardware never arrived for submission with + * the skb. + * @err: Number of arbitrary timestamp generation error events that the + * hardware encountered, exclusive of @lost statistics. Cases such + * as resource exhaustion, unavailability, firmware errors, and + * detected illogical timestamp values not submitted with the skb + * are inclusive to this counter. + */ +struct ethtool_ts_stats { + struct_group(tx_stats, + u64 pkts; + u64 lost; + u64 err; + ); +}; + #define ETH_MODULE_EEPROM_PAGE_LEN 128 #define ETH_MODULE_MAX_I2C_ADDRESS 0x7f @@ -755,7 +775,10 @@ struct ethtool_rxfh_param { * @get_ts_info: Get the time stamping and PTP hardware clock capabilities. * It may be called with RCU, or rtnl or reference on the device. * Drivers supporting transmit time stamps in software should set this to - * ethtool_op_get_ts_info(). + * ethtool_op_get_ts_info(). Drivers must not zero statistics which they + * don't report. The stats structure is initialized to ETHTOOL_STAT_NOT_SET + * indicating driver does not report statistics. + * @get_ts_stats: Query the device hardware timestamping statistics. * @get_module_info: Get the size and type of the eeprom contained within * a plug-in module. * @get_module_eeprom: Get the eeprom information from the plug-in module @@ -898,6 +921,8 @@ struct ethtool_ops { struct ethtool_dump *, void *); int (*set_dump)(struct net_device *, struct ethtool_dump *); int (*get_ts_info)(struct net_device *, struct ethtool_ts_info *); + void (*get_ts_stats)(struct net_device *dev, + struct ethtool_ts_stats *ts_stats); int (*get_module_info)(struct net_device *, struct ethtool_modinfo *); int (*get_module_eeprom)(struct net_device *, diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h index accbb1a231df..708272026d80 100644 --- a/include/uapi/linux/ethtool_netlink.h +++ b/include/uapi/linux/ethtool_netlink.h @@ -478,12 +478,26 @@ enum { ETHTOOL_A_TSINFO_TX_TYPES, /* bitset */ ETHTOOL_A_TSINFO_RX_FILTERS, /* bitset */ ETHTOOL_A_TSINFO_PHC_INDEX, /* u32 */ + ETHTOOL_A_TSINFO_STATS, /* nest - _A_TSINFO_STAT */ /* add new constants above here */ __ETHTOOL_A_TSINFO_CNT, ETHTOOL_A_TSINFO_MAX = (__ETHTOOL_A_TSINFO_CNT - 1) }; +enum { + ETHTOOL_A_TS_STAT_UNSPEC, + + ETHTOOL_A_TS_STAT_TX_PKTS, /* u64 */ + ETHTOOL_A_TS_STAT_TX_LOST, /* u64 */ + ETHTOOL_A_TS_STAT_TX_ERR, /* u64 */ + + /* add new constants above here */ + __ETHTOOL_A_TS_STAT_CNT, + ETHTOOL_A_TS_STAT_MAX = (__ETHTOOL_A_TS_STAT_CNT - 1) + +}; + /* PHC VCLOCKS */ enum { diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c index 9daed0aab162..be2755c8d8fd 100644 --- a/net/ethtool/tsinfo.c +++ b/net/ethtool/tsinfo.c @@ -13,14 +13,18 @@ struct tsinfo_req_info { struct tsinfo_reply_data { struct ethnl_reply_data base; struct ethtool_ts_info ts_info; + struct ethtool_ts_stats stats; }; #define TSINFO_REPDATA(__reply_base) \ container_of(__reply_base, struct tsinfo_reply_data, base) +#define ETHTOOL_TS_STAT_CNT \ + (__ETHTOOL_A_TS_STAT_CNT - (ETHTOOL_A_TS_STAT_UNSPEC + 1)) + const struct nla_policy ethnl_tsinfo_get_policy[] = { [ETHTOOL_A_TSINFO_HEADER] = - NLA_POLICY_NESTED(ethnl_header_policy), + NLA_POLICY_NESTED(ethnl_header_policy_stats), }; static int tsinfo_prepare_data(const struct ethnl_req_info *req_base, @@ -34,6 +38,12 @@ static int tsinfo_prepare_data(const struct ethnl_req_info *req_base, ret = ethnl_ops_begin(dev); if (ret < 0) return ret; + if (req_base->flags & ETHTOOL_FLAG_STATS && + dev->ethtool_ops->get_ts_stats) { + ethtool_stats_init((u64 *)&data->stats, + sizeof(data->stats) / sizeof(u64)); + dev->ethtool_ops->get_ts_stats(dev, &data->stats); + } ret = __ethtool_get_ts_info(dev, &data->ts_info); ethnl_ops_complete(dev); @@ -79,10 +89,47 @@ static int tsinfo_reply_size(const struct ethnl_req_info *req_base, } if (ts_info->phc_index >= 0) len += nla_total_size(sizeof(u32)); /* _TSINFO_PHC_INDEX */ + if (req_base->flags & ETHTOOL_FLAG_STATS) + len += nla_total_size(0) + /* _TSINFO_STATS */ + nla_total_size_64bit(sizeof(u64)) * ETHTOOL_TS_STAT_CNT; return len; } +static int tsinfo_put_stat(struct sk_buff *skb, u64 val, u16 attrtype) +{ + if (val == ETHTOOL_STAT_NOT_SET) + return 0; + if (nla_put_uint(skb, attrtype, val)) + return -EMSGSIZE; + return 0; +} + +static int tsinfo_put_stats(struct sk_buff *skb, + const struct ethtool_ts_stats *stats) +{ + struct nlattr *nest; + + nest = nla_nest_start(skb, ETHTOOL_A_TSINFO_STATS); + if (!nest) + return -EMSGSIZE; + + if (tsinfo_put_stat(skb, stats->tx_stats.pkts, + ETHTOOL_A_TS_STAT_TX_PKTS) || + tsinfo_put_stat(skb, stats->tx_stats.lost, + ETHTOOL_A_TS_STAT_TX_LOST) || + tsinfo_put_stat(skb, stats->tx_stats.err, + ETHTOOL_A_TS_STAT_TX_ERR)) + goto err_cancel; + + nla_nest_end(skb, nest); + return 0; + +err_cancel: + nla_nest_cancel(skb, nest); + return -EMSGSIZE; +} + static int tsinfo_fill_reply(struct sk_buff *skb, const struct ethnl_req_info *req_base, const struct ethnl_reply_data *reply_base) @@ -119,6 +166,9 @@ static int tsinfo_fill_reply(struct sk_buff *skb, if (ts_info->phc_index >= 0 && nla_put_u32(skb, ETHTOOL_A_TSINFO_PHC_INDEX, ts_info->phc_index)) return -EMSGSIZE; + if (req_base->flags & ETHTOOL_FLAG_STATS && + tsinfo_put_stats(skb, &data->stats)) + return -EMSGSIZE; return 0; } -- cgit v1.2.3 From f410cbea9f3d2675b4c8e52af1d1985b11b387d1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 4 Apr 2024 11:42:31 +0000 Subject: tcp: annotate data-races around tp->window_clamp tp->window_clamp can be read locklessly, add READ_ONCE() and WRITE_ONCE() annotations. Signed-off-by: Eric Dumazet Reviewed-by: Jason Xing Link: https://lore.kernel.org/r/20240404114231.2195171-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/syncookies.c | 3 ++- net/ipv4/tcp.c | 8 ++++---- net/ipv4/tcp_input.c | 17 ++++++++++------- net/ipv4/tcp_output.c | 18 ++++++++++-------- net/ipv6/syncookies.c | 2 +- net/mptcp/protocol.c | 2 +- net/mptcp/sockopt.c | 2 +- 7 files changed, 29 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 500f665f98cb..b61d36810fe3 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -462,7 +462,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) } /* Try to redo what tcp_v4_send_synack did. */ - req->rsk_window_clamp = tp->window_clamp ? :dst_metric(&rt->dst, RTAX_WINDOW); + req->rsk_window_clamp = READ_ONCE(tp->window_clamp) ? : + dst_metric(&rt->dst, RTAX_WINDOW); /* limit the window selection if the user enforce a smaller rx buffer */ full_space = tcp_full_space(sk); if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index e767721b3a58..92ee60492314 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1721,7 +1721,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val) space = tcp_space_from_win(sk, val); if (space > sk->sk_rcvbuf) { WRITE_ONCE(sk->sk_rcvbuf, space); - tcp_sk(sk)->window_clamp = val; + WRITE_ONCE(tcp_sk(sk)->window_clamp, val); } return 0; } @@ -3379,7 +3379,7 @@ int tcp_set_window_clamp(struct sock *sk, int val) if (!val) { if (sk->sk_state != TCP_CLOSE) return -EINVAL; - tp->window_clamp = 0; + WRITE_ONCE(tp->window_clamp, 0); } else { u32 new_rcv_ssthresh, old_window_clamp = tp->window_clamp; u32 new_window_clamp = val < SOCK_MIN_RCVBUF / 2 ? @@ -3388,7 +3388,7 @@ int tcp_set_window_clamp(struct sock *sk, int val) if (new_window_clamp == old_window_clamp) return 0; - tp->window_clamp = new_window_clamp; + WRITE_ONCE(tp->window_clamp, new_window_clamp); if (new_window_clamp < old_window_clamp) { /* need to apply the reserved mem provisioning only * when shrinking the window clamp @@ -4057,7 +4057,7 @@ int do_tcp_getsockopt(struct sock *sk, int level, TCP_RTO_MAX / HZ); break; case TCP_WINDOW_CLAMP: - val = tp->window_clamp; + val = READ_ONCE(tp->window_clamp); break; case TCP_INFO: { struct tcp_info info; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 1b6cd3840012..8d44ab5671ea 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -563,19 +563,20 @@ static void tcp_init_buffer_space(struct sock *sk) maxwin = tcp_full_space(sk); if (tp->window_clamp >= maxwin) { - tp->window_clamp = maxwin; + WRITE_ONCE(tp->window_clamp, maxwin); if (tcp_app_win && maxwin > 4 * tp->advmss) - tp->window_clamp = max(maxwin - - (maxwin >> tcp_app_win), - 4 * tp->advmss); + WRITE_ONCE(tp->window_clamp, + max(maxwin - (maxwin >> tcp_app_win), + 4 * tp->advmss)); } /* Force reservation of one segment. */ if (tcp_app_win && tp->window_clamp > 2 * tp->advmss && tp->window_clamp + tp->advmss > maxwin) - tp->window_clamp = max(2 * tp->advmss, maxwin - tp->advmss); + WRITE_ONCE(tp->window_clamp, + max(2 * tp->advmss, maxwin - tp->advmss)); tp->rcv_ssthresh = min(tp->rcv_ssthresh, tp->window_clamp); tp->snd_cwnd_stamp = tcp_jiffies32; @@ -773,7 +774,8 @@ void tcp_rcv_space_adjust(struct sock *sk) WRITE_ONCE(sk->sk_rcvbuf, rcvbuf); /* Make the window clamp follow along. */ - tp->window_clamp = tcp_win_from_space(sk, rcvbuf); + WRITE_ONCE(tp->window_clamp, + tcp_win_from_space(sk, rcvbuf)); } } tp->rcvq_space.space = copied; @@ -6426,7 +6428,8 @@ consume: if (!tp->rx_opt.wscale_ok) { tp->rx_opt.snd_wscale = tp->rx_opt.rcv_wscale = 0; - tp->window_clamp = min(tp->window_clamp, 65535U); + WRITE_ONCE(tp->window_clamp, + min(tp->window_clamp, 65535U)); } if (tp->rx_opt.saw_tstamp) { diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index e3167ad96567..9282fafc0e61 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -203,16 +203,17 @@ static inline void tcp_event_ack_sent(struct sock *sk, u32 rcv_nxt) * This MUST be enforced by all callers. */ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, - __u32 *rcv_wnd, __u32 *window_clamp, + __u32 *rcv_wnd, __u32 *__window_clamp, int wscale_ok, __u8 *rcv_wscale, __u32 init_rcv_wnd) { unsigned int space = (__space < 0 ? 0 : __space); + u32 window_clamp = READ_ONCE(*__window_clamp); /* If no clamp set the clamp to the max possible scaled window */ - if (*window_clamp == 0) - (*window_clamp) = (U16_MAX << TCP_MAX_WSCALE); - space = min(*window_clamp, space); + if (window_clamp == 0) + window_clamp = (U16_MAX << TCP_MAX_WSCALE); + space = min(window_clamp, space); /* Quantize space offering to a multiple of mss if possible. */ if (space > mss) @@ -239,12 +240,13 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, /* Set window scaling on max possible window */ space = max_t(u32, space, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])); space = max_t(u32, space, READ_ONCE(sysctl_rmem_max)); - space = min_t(u32, space, *window_clamp); + space = min_t(u32, space, window_clamp); *rcv_wscale = clamp_t(int, ilog2(space) - 15, 0, TCP_MAX_WSCALE); } /* Set the clamp no higher than max representable value */ - (*window_clamp) = min_t(__u32, U16_MAX << (*rcv_wscale), *window_clamp); + WRITE_ONCE(*__window_clamp, + min_t(__u32, U16_MAX << (*rcv_wscale), window_clamp)); } EXPORT_SYMBOL(tcp_select_initial_window); @@ -3855,7 +3857,7 @@ static void tcp_connect_init(struct sock *sk) tcp_ca_dst_init(sk, dst); if (!tp->window_clamp) - tp->window_clamp = dst_metric(dst, RTAX_WINDOW); + WRITE_ONCE(tp->window_clamp, dst_metric(dst, RTAX_WINDOW)); tp->advmss = tcp_mss_clamp(tp, dst_metric_advmss(dst)); tcp_initialize_rcv_mss(sk); @@ -3863,7 +3865,7 @@ static void tcp_connect_init(struct sock *sk) /* limit the window selection if the user enforce a smaller rx buffer */ if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && (tp->window_clamp > tcp_full_space(sk) || tp->window_clamp == 0)) - tp->window_clamp = tcp_full_space(sk); + WRITE_ONCE(tp->window_clamp, tcp_full_space(sk)); rcv_wnd = tcp_rwnd_init_bpf(sk); if (rcv_wnd == 0) diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index 6d8286c299c9..bfad1e89b6a6 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -246,7 +246,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) } } - req->rsk_window_clamp = tp->window_clamp ? :dst_metric(dst, RTAX_WINDOW); + req->rsk_window_clamp = READ_ONCE(tp->window_clamp) ? :dst_metric(dst, RTAX_WINDOW); /* limit the window selection if the user enforce a smaller rx buffer */ full_space = tcp_full_space(sk); if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 7e74b812e366..995b53cd021c 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2056,7 +2056,7 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied) ssk = mptcp_subflow_tcp_sock(subflow); slow = lock_sock_fast(ssk); WRITE_ONCE(ssk->sk_rcvbuf, rcvbuf); - tcp_sk(ssk)->window_clamp = window_clamp; + WRITE_ONCE(tcp_sk(ssk)->window_clamp, window_clamp); tcp_cleanup_rbuf(ssk, 1); unlock_sock_fast(ssk, slow); } diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 73fdf423de44..9d5d42a77bcc 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -1523,7 +1523,7 @@ int mptcp_set_rcvlowat(struct sock *sk, int val) slow = lock_sock_fast(ssk); WRITE_ONCE(ssk->sk_rcvbuf, space); - tcp_sk(ssk)->window_clamp = val; + WRITE_ONCE(tcp_sk(ssk)->window_clamp, val); unlock_sock_fast(ssk, slow); } return 0; -- cgit v1.2.3 From 9f06f87fef689d28588cde8c7ebb00a67da34026 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 3 Apr 2024 13:21:39 -0700 Subject: net: skbuff: generalize the skb->decrypted bit The ->decrypted bit can be reused for other crypto protocols. Remove the direct dependency on TLS, add helpers to clean up the ifdefs leaking out everywhere. Signed-off-by: Jakub Kicinski Reviewed-by: David Ahern Signed-off-by: David S. Miller --- include/linux/skbuff.h | 15 ++++++++++++--- include/net/sock.h | 4 +--- net/Kconfig | 3 +++ net/core/sock.c | 5 ++--- net/ipv4/tcp_input.c | 12 +++--------- net/ipv4/tcp_ipv4.c | 4 +--- net/ipv4/tcp_offload.c | 4 +--- net/tls/Kconfig | 1 + 8 files changed, 24 insertions(+), 24 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 03ea36a82cdd..7dfb906d92f7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -992,7 +992,7 @@ struct sk_buff { #ifdef CONFIG_NETFILTER_SKIP_EGRESS __u8 nf_skip_egress:1; #endif -#ifdef CONFIG_TLS_DEVICE +#ifdef CONFIG_SKB_DECRYPTED __u8 decrypted:1; #endif __u8 slow_gro:1; @@ -1615,17 +1615,26 @@ static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from) static inline int skb_cmp_decrypted(const struct sk_buff *skb1, const struct sk_buff *skb2) { -#ifdef CONFIG_TLS_DEVICE +#ifdef CONFIG_SKB_DECRYPTED return skb2->decrypted - skb1->decrypted; #else return 0; #endif } +static inline bool skb_is_decrypted(const struct sk_buff *skb) +{ +#ifdef CONFIG_SKB_DECRYPTED + return skb->decrypted; +#else + return false; +#endif +} + static inline void skb_copy_decrypted(struct sk_buff *to, const struct sk_buff *from) { -#ifdef CONFIG_TLS_DEVICE +#ifdef CONFIG_SKB_DECRYPTED to->decrypted = from->decrypted; #endif } diff --git a/include/net/sock.h b/include/net/sock.h index 2253eefe2848..a495330c5c49 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2835,12 +2835,10 @@ static inline struct sk_buff *sk_validate_xmit_skb(struct sk_buff *skb, if (sk && sk_fullsock(sk) && sk->sk_validate_xmit_skb) { skb = sk->sk_validate_xmit_skb(sk, dev, skb); -#ifdef CONFIG_TLS_DEVICE - } else if (unlikely(skb->decrypted)) { + } else if (unlikely(skb_is_decrypted(skb))) { pr_warn_ratelimited("unencrypted skb with no associated socket - dropping\n"); kfree_skb(skb); skb = NULL; -#endif } #endif diff --git a/net/Kconfig b/net/Kconfig index 3e57ccf0da27..d5ab791f7afa 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -60,6 +60,9 @@ config NET_XGRESS config NET_REDIRECT bool +config SKB_DECRYPTED + bool + config SKB_EXTENSIONS bool diff --git a/net/core/sock.c b/net/core/sock.c index 5ed411231fc7..fe9195186c13 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2526,13 +2526,12 @@ EXPORT_SYMBOL(skb_set_owner_w); static bool can_skb_orphan_partial(const struct sk_buff *skb) { -#ifdef CONFIG_TLS_DEVICE /* Drivers depend on in-order delivery for crypto offload, * partial orphan breaks out-of-order-OK logic. */ - if (skb->decrypted) + if (skb_is_decrypted(skb)) return false; -#endif + return (skb->destructor == sock_wfree || (IS_ENABLED(CONFIG_INET) && skb->destructor == tcp_wfree)); } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 8d44ab5671ea..1f28a2561795 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4805,10 +4805,8 @@ static bool tcp_try_coalesce(struct sock *sk, if (!mptcp_skb_can_collapse(to, from)) return false; -#ifdef CONFIG_TLS_DEVICE - if (from->decrypted != to->decrypted) + if (skb_cmp_decrypted(from, to)) return false; -#endif if (!skb_try_coalesce(to, from, fragstolen, &delta)) return false; @@ -5377,9 +5375,7 @@ restart: break; memcpy(nskb->cb, skb->cb, sizeof(skb->cb)); -#ifdef CONFIG_TLS_DEVICE - nskb->decrypted = skb->decrypted; -#endif + skb_copy_decrypted(nskb, skb); TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start; if (list) __skb_queue_before(list, skb, nskb); @@ -5409,10 +5405,8 @@ restart: !mptcp_skb_can_collapse(nskb, skb) || (TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN))) goto end; -#ifdef CONFIG_TLS_DEVICE - if (skb->decrypted != nskb->decrypted) + if (skb_cmp_decrypted(skb, nskb)) goto end; -#endif } } } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 56b75efcfd12..52963c3bb8ca 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2044,10 +2044,8 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb, TCP_SKB_CB(skb)->tcp_flags) & TCPHDR_ACK) || ((TCP_SKB_CB(tail)->tcp_flags ^ TCP_SKB_CB(skb)->tcp_flags) & (TCPHDR_ECE | TCPHDR_CWR)) || -#ifdef CONFIG_TLS_DEVICE - tail->decrypted != skb->decrypted || -#endif !mptcp_skb_can_collapse(tail, skb) || + skb_cmp_decrypted(tail, skb) || thtail->doff != th->doff || memcmp(thtail + 1, th + 1, hdrlen - sizeof(*th))) goto no_coalesce; diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index ebe4722bb020..fab0973f995b 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -265,9 +265,7 @@ found: flush |= (len - 1) >= mss; flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq); -#ifdef CONFIG_TLS_DEVICE - flush |= p->decrypted ^ skb->decrypted; -#endif + flush |= skb_cmp_decrypted(p, skb); if (flush || skb_gro_receive(p, skb)) { mss = 1; diff --git a/net/tls/Kconfig b/net/tls/Kconfig index 0cdc1f7b6b08..ce8d56a19187 100644 --- a/net/tls/Kconfig +++ b/net/tls/Kconfig @@ -20,6 +20,7 @@ config TLS config TLS_DEVICE bool "Transport Layer Security HW offload" depends on TLS + select SKB_DECRYPTED select SOCK_VALIDATE_XMIT select SOCK_RX_QUEUE_MAPPING default n -- cgit v1.2.3 From 8e69b3459ca1ed4f6f7bd0b0a11962ddb3e7d34a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 3 Apr 2024 13:22:59 -0700 Subject: netlink: add nlmsg_consume() and use it in devlink compat devlink_compat_running_version() sticks out when running netdevsim tests and watching dropped skbs. Add nlmsg_consume() for cases were we want to free a netlink skb but it is expected, rather than a drop. af_netlink code uses consume_skb() directly, which is fine, but some may prefer the symmetry of nlmsg_new() / nlmsg_consume(). Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- include/net/netlink.h | 14 ++++++++++++-- net/devlink/dev.c | 2 +- 2 files changed, 13 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/netlink.h b/include/net/netlink.h index 1d2bbcc50212..61cef3bd2d31 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -41,7 +41,8 @@ * nlmsg_get_pos() return current position in message * nlmsg_trim() trim part of message * nlmsg_cancel() cancel message construction - * nlmsg_free() free a netlink message + * nlmsg_consume() free a netlink message (expected) + * nlmsg_free() free a netlink message (drop) * * Message Sending: * nlmsg_multicast() multicast message to several groups @@ -1082,7 +1083,7 @@ static inline void nlmsg_cancel(struct sk_buff *skb, struct nlmsghdr *nlh) } /** - * nlmsg_free - free a netlink message + * nlmsg_free - drop a netlink message * @skb: socket buffer of netlink message */ static inline void nlmsg_free(struct sk_buff *skb) @@ -1090,6 +1091,15 @@ static inline void nlmsg_free(struct sk_buff *skb) kfree_skb(skb); } +/** + * nlmsg_consume - free a netlink message + * @skb: socket buffer of netlink message + */ +static inline void nlmsg_consume(struct sk_buff *skb) +{ + consume_skb(skb); +} + /** * nlmsg_multicast_filtered - multicast a netlink message with filter function * @sk: netlink socket to spread messages to diff --git a/net/devlink/dev.c b/net/devlink/dev.c index c609deb42e88..13c73f50da3d 100644 --- a/net/devlink/dev.c +++ b/net/devlink/dev.c @@ -1214,7 +1214,7 @@ static void __devlink_compat_running_version(struct devlink *devlink, } } free_msg: - nlmsg_free(msg); + nlmsg_consume(msg); } void devlink_compat_running_version(struct devlink *devlink, -- cgit v1.2.3 From 6916e461e7933d3d003441291c543938f2ccb371 Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Thu, 4 Apr 2024 11:29:51 +0200 Subject: net: phy: Introduce ethernet link topology representation Link topologies containing multiple network PHYs attached to the same net_device can be found when using a PHY as a media converter for use with an SFP connector, on which an SFP transceiver containing a PHY can be used. With the current model, the transceiver's PHY can't be used for operations such as cable testing, timestamping, macsec offload, etc. The reason being that most of the logic for these configuration, coming from either ethtool netlink or ioctls tend to use netdev->phydev, which in multi-phy systems will reference the PHY closest to the MAC. Introduce a numbering scheme allowing to enumerate PHY devices that belong to any netdev, which can in turn allow userspace to take more precise decisions with regard to each PHY's configuration. The numbering is maintained per-netdev, in a phy_device_list. The numbering works similarly to a netdevice's ifindex, with identifiers that are only recycled once INT_MAX has been reached. This prevents races that could occur between PHY listing and SFP transceiver removal/insertion. The identifiers are assigned at phy_attach time, as the numbering depends on the netdevice the phy is attached to. The PHY index can be re-used for PHYs that are persistent. Signed-off-by: Maxime Chevallier Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- MAINTAINERS | 2 + drivers/net/phy/Makefile | 2 +- drivers/net/phy/phy_device.c | 7 +++ drivers/net/phy/phy_link_topology.c | 105 +++++++++++++++++++++++++++++++++ include/linux/netdevice.h | 4 +- include/linux/phy.h | 4 ++ include/linux/phy_link_topology.h | 72 ++++++++++++++++++++++ include/linux/phy_link_topology_core.h | 25 ++++++++ include/uapi/linux/ethtool.h | 16 +++++ net/core/dev.c | 9 +++ 10 files changed, 244 insertions(+), 2 deletions(-) create mode 100644 drivers/net/phy/phy_link_topology.c create mode 100644 include/linux/phy_link_topology.h create mode 100644 include/linux/phy_link_topology_core.h (limited to 'net') diff --git a/MAINTAINERS b/MAINTAINERS index 046598b471b7..4745ea94d463 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8015,6 +8015,8 @@ F: include/linux/mii.h F: include/linux/of_net.h F: include/linux/phy.h F: include/linux/phy_fixed.h +F: include/linux/phy_link_topology.h +F: include/linux/phy_link_topology_core.h F: include/linux/phylib_stubs.h F: include/linux/platform_data/mdio-bcm-unimac.h F: include/linux/platform_data/mdio-gpio.h diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index 202ed7f450da..1d8be374915f 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -2,7 +2,7 @@ # Makefile for Linux PHY drivers libphy-y := phy.o phy-c45.o phy-core.o phy_device.o \ - linkmode.o + linkmode.o phy_link_topology.o mdio-bus-y += mdio_bus.o mdio_device.o ifdef CONFIG_MDIO_DEVICE diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 6c6ec9475709..452fc8b3406d 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include @@ -1511,6 +1512,11 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, if (phydev->sfp_bus_attached) dev->sfp_bus = phydev->sfp_bus; + + err = phy_link_topo_add_phy(dev->link_topo, phydev, + PHY_UPSTREAM_MAC, dev); + if (err) + goto error; } /* Some Ethernet drivers try to connect to a PHY device before @@ -1938,6 +1944,7 @@ void phy_detach(struct phy_device *phydev) if (dev) { phydev->attached_dev->phydev = NULL; phydev->attached_dev = NULL; + phy_link_topo_del_phy(dev->link_topo, phydev); } phydev->phylink = NULL; diff --git a/drivers/net/phy/phy_link_topology.c b/drivers/net/phy/phy_link_topology.c new file mode 100644 index 000000000000..985941c5c558 --- /dev/null +++ b/drivers/net/phy/phy_link_topology.c @@ -0,0 +1,105 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * Infrastructure to handle all PHY devices connected to a given netdev, + * either directly or indirectly attached. + * + * Copyright (c) 2023 Maxime Chevallier + */ + +#include +#include +#include +#include +#include + +struct phy_link_topology *phy_link_topo_create(struct net_device *dev) +{ + struct phy_link_topology *topo; + + topo = kzalloc(sizeof(*topo), GFP_KERNEL); + if (!topo) + return ERR_PTR(-ENOMEM); + + xa_init_flags(&topo->phys, XA_FLAGS_ALLOC1); + topo->next_phy_index = 1; + + return topo; +} + +void phy_link_topo_destroy(struct phy_link_topology *topo) +{ + if (!topo) + return; + + xa_destroy(&topo->phys); + kfree(topo); +} + +int phy_link_topo_add_phy(struct phy_link_topology *topo, + struct phy_device *phy, + enum phy_upstream upt, void *upstream) +{ + struct phy_device_node *pdn; + int ret; + + pdn = kzalloc(sizeof(*pdn), GFP_KERNEL); + if (!pdn) + return -ENOMEM; + + pdn->phy = phy; + switch (upt) { + case PHY_UPSTREAM_MAC: + pdn->upstream.netdev = (struct net_device *)upstream; + if (phy_on_sfp(phy)) + pdn->parent_sfp_bus = pdn->upstream.netdev->sfp_bus; + break; + case PHY_UPSTREAM_PHY: + pdn->upstream.phydev = (struct phy_device *)upstream; + if (phy_on_sfp(phy)) + pdn->parent_sfp_bus = pdn->upstream.phydev->sfp_bus; + break; + default: + ret = -EINVAL; + goto err; + } + pdn->upstream_type = upt; + + /* Attempt to re-use a previously allocated phy_index */ + if (phy->phyindex) { + ret = xa_insert(&topo->phys, phy->phyindex, pdn, GFP_KERNEL); + + /* Errors could be either -ENOMEM or -EBUSY. If the phy has an + * index, and there's another entry at the same index, this is + * unexpected and we still error-out + */ + if (ret) + goto err; + return 0; + } + + ret = xa_alloc_cyclic(&topo->phys, &phy->phyindex, pdn, xa_limit_32b, + &topo->next_phy_index, GFP_KERNEL); + if (ret) + goto err; + + return 0; + +err: + kfree(pdn); + return ret; +} +EXPORT_SYMBOL_GPL(phy_link_topo_add_phy); + +void phy_link_topo_del_phy(struct phy_link_topology *topo, + struct phy_device *phy) +{ + struct phy_device_node *pdn = xa_erase(&topo->phys, phy->phyindex); + + /* We delete the PHY from the topology, however we don't re-set the + * phy->phyindex field. If the PHY isn't gone, we can re-assign it the + * same index next time it's added back to the topology + */ + + kfree(pdn); +} +EXPORT_SYMBOL_GPL(phy_link_topo_del_phy); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 0c198620ac93..d45f330d083d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -40,7 +40,6 @@ #include #endif #include - #include #include #include @@ -52,6 +51,7 @@ #include #include #include +#include struct netpoll_info; struct device; @@ -1974,6 +1974,7 @@ enum netdev_reg_state { * @fcoe_ddp_xid: Max exchange id for FCoE LRO by ddp * * @priomap: XXX: need comments on this one + * @link_topo: Physical link topology tracking attached PHYs * @phydev: Physical device may attach itself * for hardware timestamping * @sfp_bus: attached &struct sfp_bus structure. @@ -2364,6 +2365,7 @@ struct net_device { #if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) struct netprio_map __rcu *priomap; #endif + struct phy_link_topology *link_topo; struct phy_device *phydev; struct sfp_bus *sfp_bus; struct lock_class_key *qdisc_tx_busylock; diff --git a/include/linux/phy.h b/include/linux/phy.h index e6e83304558e..8c848c79b1fd 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -550,6 +550,9 @@ struct macsec_ops; * @drv: Pointer to the driver for this PHY instance * @devlink: Create a link between phy dev and mac dev, if the external phy * used by current mac interface is managed by another mac interface. + * @phyindex: Unique id across the phy's parent tree of phys to address the PHY + * from userspace, similar to ifindex. A zero index means the PHY + * wasn't assigned an id yet. * @phy_id: UID for this device found during discovery * @c45_ids: 802.3-c45 Device Identifiers if is_c45. * @is_c45: Set to true if this PHY uses clause 45 addressing. @@ -650,6 +653,7 @@ struct phy_device { struct device_link *devlink; + u32 phyindex; u32 phy_id; struct phy_c45_device_ids c45_ids; diff --git a/include/linux/phy_link_topology.h b/include/linux/phy_link_topology.h new file mode 100644 index 000000000000..6b79feb607e7 --- /dev/null +++ b/include/linux/phy_link_topology.h @@ -0,0 +1,72 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * PHY device list allow maintaining a list of PHY devices that are + * part of a netdevice's link topology. PHYs can for example be chained, + * as is the case when using a PHY that exposes an SFP module, on which an + * SFP transceiver that embeds a PHY is connected. + * + * This list can then be used by userspace to leverage individual PHY + * capabilities. + */ +#ifndef __PHY_LINK_TOPOLOGY_H +#define __PHY_LINK_TOPOLOGY_H + +#include +#include + +struct xarray; +struct phy_device; +struct net_device; +struct sfp_bus; + +struct phy_device_node { + enum phy_upstream upstream_type; + + union { + struct net_device *netdev; + struct phy_device *phydev; + } upstream; + + struct sfp_bus *parent_sfp_bus; + + struct phy_device *phy; +}; + +struct phy_link_topology { + struct xarray phys; + u32 next_phy_index; +}; + +static inline struct phy_device * +phy_link_topo_get_phy(struct phy_link_topology *topo, u32 phyindex) +{ + struct phy_device_node *pdn = xa_load(&topo->phys, phyindex); + + if (pdn) + return pdn->phy; + + return NULL; +} + +#if IS_REACHABLE(CONFIG_PHYLIB) +int phy_link_topo_add_phy(struct phy_link_topology *topo, + struct phy_device *phy, + enum phy_upstream upt, void *upstream); + +void phy_link_topo_del_phy(struct phy_link_topology *lt, struct phy_device *phy); + +#else +static inline int phy_link_topo_add_phy(struct phy_link_topology *topo, + struct phy_device *phy, + enum phy_upstream upt, void *upstream) +{ + return 0; +} + +static inline void phy_link_topo_del_phy(struct phy_link_topology *topo, + struct phy_device *phy) +{ +} +#endif + +#endif /* __PHY_LINK_TOPOLOGY_H */ diff --git a/include/linux/phy_link_topology_core.h b/include/linux/phy_link_topology_core.h new file mode 100644 index 000000000000..0a6479055745 --- /dev/null +++ b/include/linux/phy_link_topology_core.h @@ -0,0 +1,25 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __PHY_LINK_TOPOLOGY_CORE_H +#define __PHY_LINK_TOPOLOGY_CORE_H + +struct phy_link_topology; + +#if IS_REACHABLE(CONFIG_PHYLIB) + +struct phy_link_topology *phy_link_topo_create(struct net_device *dev); +void phy_link_topo_destroy(struct phy_link_topology *topo); + +#else + +static inline struct phy_link_topology *phy_link_topo_create(struct net_device *dev) +{ + return NULL; +} + +static inline void phy_link_topo_destroy(struct phy_link_topology *topo) +{ +} + +#endif + +#endif /* __PHY_LINK_TOPOLOGY_CORE_H */ diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index 11fc18988bc2..95c2f09f0d0a 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -2268,4 +2268,20 @@ struct ethtool_link_settings { * __u32 map_lp_advertising[link_mode_masks_nwords]; */ }; + +/** + * enum phy_upstream - Represents the upstream component a given PHY device + * is connected to, as in what is on the other end of the MII bus. Most PHYs + * will be attached to an Ethernet MAC controller, but in some cases, there's + * an intermediate PHY used as a media-converter, which will driver another + * MII interface as its output. + * @PHY_UPSTREAM_MAC: Upstream component is a MAC (a switch port, + * or ethernet controller) + * @PHY_UPSTREAM_PHY: Upstream component is a PHY (likely a media converter) + */ +enum phy_upstream { + PHY_UPSTREAM_MAC, + PHY_UPSTREAM_PHY, +}; + #endif /* _UAPI_LINUX_ETHTOOL_H */ diff --git a/net/core/dev.c b/net/core/dev.c index 92f5bddbc2de..854a3a28a8d8 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -158,6 +158,7 @@ #include #include #include +#include #include "dev.h" #include "net-sysfs.h" @@ -10962,6 +10963,12 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, #ifdef CONFIG_NET_SCHED hash_init(dev->qdisc_hash); #endif + dev->link_topo = phy_link_topo_create(dev); + if (IS_ERR(dev->link_topo)) { + dev->link_topo = NULL; + goto free_all; + } + dev->priv_flags = IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM; setup(dev); @@ -11050,6 +11057,8 @@ void free_netdev(struct net_device *dev) free_percpu(dev->xdp_bulkq); dev->xdp_bulkq = NULL; + phy_link_topo_destroy(dev->link_topo); + /* Compatibility with error handling in drivers */ if (dev->reg_state == NETREG_UNINITIALIZED) { netdev_freemem(dev); -- cgit v1.2.3 From 841942bc62122c59f654f799c2d00ce3fda93efa Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Thu, 4 Apr 2024 11:29:55 +0200 Subject: net: ethtool: Allow passing a phy index for some commands Some netlink commands are target towards ethernet PHYs, to control some of their features. As there's several such commands, add the ability to pass a PHY index in the ethnl request, which will populate the generic ethnl_req_info with the relevant phydev when the command targets a PHY. Signed-off-by: Maxime Chevallier Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- Documentation/networking/ethtool-netlink.rst | 7 ++++ include/uapi/linux/ethtool_netlink.h | 1 + net/ethtool/netlink.c | 48 ++++++++++++++++++++++++++-- net/ethtool/netlink.h | 5 +++ 4 files changed, 59 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index 08d330b0f50f..5dc42f7ce429 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -57,6 +57,7 @@ Structure of this header is ``ETHTOOL_A_HEADER_DEV_INDEX`` u32 device ifindex ``ETHTOOL_A_HEADER_DEV_NAME`` string device name ``ETHTOOL_A_HEADER_FLAGS`` u32 flags common for all requests + ``ETHTOOL_A_HEADER_PHY_INDEX`` u32 phy device index ============================== ====== ============================= ``ETHTOOL_A_HEADER_DEV_INDEX`` and ``ETHTOOL_A_HEADER_DEV_NAME`` identify the @@ -81,6 +82,12 @@ the behaviour is backward compatible, i.e. requests from old clients not aware of the flag should be interpreted the way the client expects. A client must not set flags it does not understand. +``ETHTOOL_A_HEADER_PHY_INDEX`` identifies the Ethernet PHY the message relates to. +As there are numerous commands that are related to PHY configuration, and because +there may be more than one PHY on the link, the PHY index can be passed in the +request for the commands that needs it. It is, however, not mandatory, and if it +is not passed for commands that target a PHY, the net_device.phydev pointer +is used. Bit sets ======== diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h index e06a34cded78..23e225f00fb0 100644 --- a/include/uapi/linux/ethtool_netlink.h +++ b/include/uapi/linux/ethtool_netlink.h @@ -132,6 +132,7 @@ enum { ETHTOOL_A_HEADER_DEV_INDEX, /* u32 */ ETHTOOL_A_HEADER_DEV_NAME, /* string */ ETHTOOL_A_HEADER_FLAGS, /* u32 - ETHTOOL_FLAG_* */ + ETHTOOL_A_HEADER_PHY_INDEX, /* u32 */ /* add new constants above here */ __ETHTOOL_A_HEADER_CNT, diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c index bd04f28d5cf4..563e94e0cbd8 100644 --- a/net/ethtool/netlink.c +++ b/net/ethtool/netlink.c @@ -4,6 +4,7 @@ #include #include #include "netlink.h" +#include static struct genl_family ethtool_genl_family; @@ -30,6 +31,24 @@ const struct nla_policy ethnl_header_policy_stats[] = { ETHTOOL_FLAGS_STATS), }; +const struct nla_policy ethnl_header_policy_phy[] = { + [ETHTOOL_A_HEADER_DEV_INDEX] = { .type = NLA_U32 }, + [ETHTOOL_A_HEADER_DEV_NAME] = { .type = NLA_NUL_STRING, + .len = ALTIFNAMSIZ - 1 }, + [ETHTOOL_A_HEADER_FLAGS] = NLA_POLICY_MASK(NLA_U32, + ETHTOOL_FLAGS_BASIC), + [ETHTOOL_A_HEADER_PHY_INDEX] = NLA_POLICY_MIN(NLA_U32, 1), +}; + +const struct nla_policy ethnl_header_policy_phy_stats[] = { + [ETHTOOL_A_HEADER_DEV_INDEX] = { .type = NLA_U32 }, + [ETHTOOL_A_HEADER_DEV_NAME] = { .type = NLA_NUL_STRING, + .len = ALTIFNAMSIZ - 1 }, + [ETHTOOL_A_HEADER_FLAGS] = NLA_POLICY_MASK(NLA_U32, + ETHTOOL_FLAGS_STATS), + [ETHTOOL_A_HEADER_PHY_INDEX] = NLA_POLICY_MIN(NLA_U32, 1), +}; + int ethnl_ops_begin(struct net_device *dev) { int ret; @@ -89,8 +108,9 @@ int ethnl_parse_header_dev_get(struct ethnl_req_info *req_info, const struct nlattr *header, struct net *net, struct netlink_ext_ack *extack, bool require_dev) { - struct nlattr *tb[ARRAY_SIZE(ethnl_header_policy)]; + struct nlattr *tb[ARRAY_SIZE(ethnl_header_policy_phy)]; const struct nlattr *devname_attr; + struct phy_device *phydev = NULL; struct net_device *dev = NULL; u32 flags = 0; int ret; @@ -104,7 +124,7 @@ int ethnl_parse_header_dev_get(struct ethnl_req_info *req_info, /* No validation here, command policy should have a nested policy set * for the header, therefore validation should have already been done. */ - ret = nla_parse_nested(tb, ARRAY_SIZE(ethnl_header_policy) - 1, header, + ret = nla_parse_nested(tb, ARRAY_SIZE(ethnl_header_policy_phy) - 1, header, NULL, extack); if (ret < 0) return ret; @@ -145,6 +165,30 @@ int ethnl_parse_header_dev_get(struct ethnl_req_info *req_info, return -EINVAL; } + if (dev) { + if (tb[ETHTOOL_A_HEADER_PHY_INDEX]) { + struct nlattr *phy_id; + + phy_id = tb[ETHTOOL_A_HEADER_PHY_INDEX]; + phydev = phy_link_topo_get_phy(dev->link_topo, + nla_get_u32(phy_id)); + if (!phydev) { + NL_SET_BAD_ATTR(extack, phy_id); + return -ENODEV; + } + } else { + /* If we need a PHY but no phy index is specified, fallback + * to dev->phydev + */ + phydev = dev->phydev; + } + } else if (tb[ETHTOOL_A_HEADER_PHY_INDEX]) { + NL_SET_ERR_MSG_ATTR(extack, header, + "can't target a PHY without a netdev"); + return -EINVAL; + } + + req_info->phydev = phydev; req_info->dev = dev; req_info->flags = flags; return 0; diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h index 9a333a8d04c1..d57a890b5d9e 100644 --- a/net/ethtool/netlink.h +++ b/net/ethtool/netlink.h @@ -250,6 +250,7 @@ static inline unsigned int ethnl_reply_header_size(void) * @dev: network device the request is for (may be null) * @dev_tracker: refcount tracker for @dev reference * @flags: request flags common for all request types + * @phydev: phy_device connected to @dev this request is for (may be null) * * This is a common base for request specific structures holding data from * parsed userspace request. These always embed struct ethnl_req_info at @@ -259,6 +260,7 @@ struct ethnl_req_info { struct net_device *dev; netdevice_tracker dev_tracker; u32 flags; + struct phy_device *phydev; }; static inline void ethnl_parse_header_dev_put(struct ethnl_req_info *req_info) @@ -395,9 +397,12 @@ extern const struct ethnl_request_ops ethnl_rss_request_ops; extern const struct ethnl_request_ops ethnl_plca_cfg_request_ops; extern const struct ethnl_request_ops ethnl_plca_status_request_ops; extern const struct ethnl_request_ops ethnl_mm_request_ops; +extern const struct ethnl_request_ops ethnl_phy_request_ops; extern const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_FLAGS + 1]; extern const struct nla_policy ethnl_header_policy_stats[ETHTOOL_A_HEADER_FLAGS + 1]; +extern const struct nla_policy ethnl_header_policy_phy[ETHTOOL_A_HEADER_PHY_INDEX + 1]; +extern const struct nla_policy ethnl_header_policy_phy_stats[ETHTOOL_A_HEADER_PHY_INDEX + 1]; extern const struct nla_policy ethnl_strset_get_policy[ETHTOOL_A_STRSET_COUNTS_ONLY + 1]; extern const struct nla_policy ethnl_linkinfo_get_policy[ETHTOOL_A_LINKINFO_HEADER + 1]; extern const struct nla_policy ethnl_linkinfo_set_policy[ETHTOOL_A_LINKINFO_TP_MDIX_CTRL + 1]; -- cgit v1.2.3 From 6648e613226e18897231ab5e42ffc29e63fa3365 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 4 Apr 2024 10:10:01 +0800 Subject: bpf, skmsg: Fix NULL pointer dereference in sk_psock_skb_ingress_enqueue Fix NULL pointer data-races in sk_psock_skb_ingress_enqueue() which syzbot reported [1]. [1] BUG: KCSAN: data-race in sk_psock_drop / sk_psock_skb_ingress_enqueue write to 0xffff88814b3278b8 of 8 bytes by task 10724 on cpu 1: sk_psock_stop_verdict net/core/skmsg.c:1257 [inline] sk_psock_drop+0x13e/0x1f0 net/core/skmsg.c:843 sk_psock_put include/linux/skmsg.h:459 [inline] sock_map_close+0x1a7/0x260 net/core/sock_map.c:1648 unix_release+0x4b/0x80 net/unix/af_unix.c:1048 __sock_release net/socket.c:659 [inline] sock_close+0x68/0x150 net/socket.c:1421 __fput+0x2c1/0x660 fs/file_table.c:422 __fput_sync+0x44/0x60 fs/file_table.c:507 __do_sys_close fs/open.c:1556 [inline] __se_sys_close+0x101/0x1b0 fs/open.c:1541 __x64_sys_close+0x1f/0x30 fs/open.c:1541 do_syscall_64+0xd3/0x1d0 entry_SYSCALL_64_after_hwframe+0x6d/0x75 read to 0xffff88814b3278b8 of 8 bytes by task 10713 on cpu 0: sk_psock_data_ready include/linux/skmsg.h:464 [inline] sk_psock_skb_ingress_enqueue+0x32d/0x390 net/core/skmsg.c:555 sk_psock_skb_ingress_self+0x185/0x1e0 net/core/skmsg.c:606 sk_psock_verdict_apply net/core/skmsg.c:1008 [inline] sk_psock_verdict_recv+0x3e4/0x4a0 net/core/skmsg.c:1202 unix_read_skb net/unix/af_unix.c:2546 [inline] unix_stream_read_skb+0x9e/0xf0 net/unix/af_unix.c:2682 sk_psock_verdict_data_ready+0x77/0x220 net/core/skmsg.c:1223 unix_stream_sendmsg+0x527/0x860 net/unix/af_unix.c:2339 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x140/0x180 net/socket.c:745 ____sys_sendmsg+0x312/0x410 net/socket.c:2584 ___sys_sendmsg net/socket.c:2638 [inline] __sys_sendmsg+0x1e9/0x280 net/socket.c:2667 __do_sys_sendmsg net/socket.c:2676 [inline] __se_sys_sendmsg net/socket.c:2674 [inline] __x64_sys_sendmsg+0x46/0x50 net/socket.c:2674 do_syscall_64+0xd3/0x1d0 entry_SYSCALL_64_after_hwframe+0x6d/0x75 value changed: 0xffffffff83d7feb0 -> 0x0000000000000000 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 10713 Comm: syz-executor.4 Tainted: G W 6.8.0-syzkaller-08951-gfe46a7dd189e #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/29/2024 Prior to this, commit 4cd12c6065df ("bpf, sockmap: Fix NULL pointer dereference in sk_psock_verdict_data_ready()") fixed one NULL pointer similarly due to no protection of saved_data_ready. Here is another different caller causing the same issue because of the same reason. So we should protect it with sk_callback_lock read lock because the writer side in the sk_psock_drop() uses "write_lock_bh(&sk->sk_callback_lock);". To avoid errors that could happen in future, I move those two pairs of lock into the sk_psock_data_ready(), which is suggested by John Fastabend. Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface") Reported-by: syzbot+aa8c8ec2538929f18f2d@syzkaller.appspotmail.com Signed-off-by: Jason Xing Signed-off-by: Daniel Borkmann Reviewed-by: John Fastabend Closes: https://syzkaller.appspot.com/bug?extid=aa8c8ec2538929f18f2d Link: https://lore.kernel.org/all/20240329134037.92124-1-kerneljasonxing@gmail.com Link: https://lore.kernel.org/bpf/20240404021001.94815-1-kerneljasonxing@gmail.com --- include/linux/skmsg.h | 2 ++ net/core/skmsg.c | 5 +---- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index e65ec3fd2799..a509caf823d6 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -461,10 +461,12 @@ static inline void sk_psock_put(struct sock *sk, struct sk_psock *psock) static inline void sk_psock_data_ready(struct sock *sk, struct sk_psock *psock) { + read_lock_bh(&sk->sk_callback_lock); if (psock->saved_data_ready) psock->saved_data_ready(sk); else sk->sk_data_ready(sk); + read_unlock_bh(&sk->sk_callback_lock); } static inline void psock_set_prog(struct bpf_prog **pprog, diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 4d75ef9d24bf..fd20aae30be2 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -1226,11 +1226,8 @@ static void sk_psock_verdict_data_ready(struct sock *sk) rcu_read_lock(); psock = sk_psock(sk); - if (psock) { - read_lock_bh(&sk->sk_callback_lock); + if (psock) sk_psock_data_ready(sk, psock); - read_unlock_bh(&sk->sk_callback_lock); - } rcu_read_unlock(); } } -- cgit v1.2.3 From a9b2d55a8f1e9652b800c84c3986442ca6d1fb50 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 4 Apr 2024 05:52:51 -0700 Subject: ip6_vti: Do not use custom stat allocator With commit 34d21de99cea9 ("net: Move {l,t,d}stats allocation to core and convert veth & vrf"), stats allocation could be done on net core instead of in this driver. With this new approach, the driver doesn't have to bother with error handling (allocation failure checking, making sure free happens in the right spot, etc). This is core responsibility now. Remove the allocation in the ip6_vti and leverage the network core allocation instead. Signed-off-by: Breno Leitao Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/ip6_vti.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 7f4f976aa24a..71c749bb4b3e 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -174,11 +174,6 @@ vti6_tnl_unlink(struct vti6_net *ip6n, struct ip6_tnl *t) } } -static void vti6_dev_free(struct net_device *dev) -{ - free_percpu(dev->tstats); -} - static int vti6_tnl_create2(struct net_device *dev) { struct ip6_tnl *t = netdev_priv(dev); @@ -907,9 +902,8 @@ static void vti6_dev_setup(struct net_device *dev) { dev->netdev_ops = &vti6_netdev_ops; dev->header_ops = &ip_tunnel_header_ops; - dev->needs_free_netdev = true; - dev->priv_destructor = vti6_dev_free; + dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS; dev->type = ARPHRD_TUNNEL6; dev->min_mtu = IPV4_MIN_MTU; dev->max_mtu = IP_MAX_MTU - sizeof(struct ipv6hdr); @@ -931,9 +925,6 @@ static inline int vti6_dev_init_gen(struct net_device *dev) t->dev = dev; t->net = dev_net(dev); - dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; netdev_hold(dev, &t->dev_tracker, GFP_KERNEL); netdev_lockdep_set_classes(dev); return 0; -- cgit v1.2.3 From b2c919c108abda11ebf16c4e31cf687a4db95a6f Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 4 Apr 2024 05:52:52 -0700 Subject: ip6_vti: Remove generic .ndo_get_stats64 Commit 3e2f544dd8a33 ("net: get stats64 if device if driver is configured") moved the callback to dev_get_tstats64() to net core, so, unless the driver is doing some custom stats collection, it does not need to set .ndo_get_stats64. Since this driver is now relying in NETDEV_PCPU_STAT_TSTATS, then, it doesn't need to set the dev_get_tstats64() generic .ndo_get_stats64 function pointer. Signed-off-by: Breno Leitao Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/ip6_vti.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 71c749bb4b3e..4d68a0777b0c 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -887,7 +887,6 @@ static const struct net_device_ops vti6_netdev_ops = { .ndo_uninit = vti6_dev_uninit, .ndo_start_xmit = vti6_tnl_xmit, .ndo_siocdevprivate = vti6_siocdevprivate, - .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip6_tnl_get_iflink, }; -- cgit v1.2.3 From 802e12ff9cbff3bafb0b6f0fe79cff04dce9de16 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 4 Apr 2024 13:07:51 +0000 Subject: inet: frags: delay fqdir_free_fn() fqdir_free_fn() is using very expensive rcu_barrier() When one netns is dismantled, we often call fqdir_exit() multiple times, typically lauching fqdir_free_fn() twice. Delaying by one second fqdir_free_fn() helps to reduce the number of rcu_barrier() calls, and lock contention on rcu_state.barrier_mutex. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/inet_fragment.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index c88c9034d630..faaec92a46ac 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -175,7 +175,7 @@ static void fqdir_free_fn(struct work_struct *work) } } -static DECLARE_WORK(fqdir_free_work, fqdir_free_fn); +static DECLARE_DELAYED_WORK(fqdir_free_work, fqdir_free_fn); static void fqdir_work_fn(struct work_struct *work) { @@ -184,7 +184,7 @@ static void fqdir_work_fn(struct work_struct *work) rhashtable_free_and_destroy(&fqdir->rhashtable, inet_frags_free_cb, NULL); if (llist_add(&fqdir->free_list, &fqdir_free_list)) - queue_work(system_wq, &fqdir_free_work); + queue_delayed_work(system_wq, &fqdir_free_work, HZ); } int fqdir_init(struct fqdir **fqdirp, struct inet_frags *f, struct net *net) -- cgit v1.2.3 From eec53cc38c4aba41bc3fabd0a75bc30c2c8ea332 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 4 Apr 2024 13:24:13 +0000 Subject: ipv6: remove RTNL protection from ip6addrlbl_dump() No longer hold RTNL while calling ip6addrlbl_dump() ("ip addrlabel show") ip6addrlbl_dump() was already mostly relying on RCU anyway. Add READ_ONCE()/WRITE_ONCE() annotations around net->ipv6.ip6addrlbl_table.seq Note that ifal_seq value is currently ignored in iproute2, and a bit weak. We might user later cb->seq / nl_dump_check_consistent() protocol if needed. Also change return value for a completed dump, so that NLMSG_DONE can be appended to current skb, saving one recvmsg() system call. v2: read net->ipv6.ip6addrlbl_table.seq once, (David Ahern) Signed-off-by: Eric Dumazet Link:https://lore.kernel.org/netdev/67f5cb70-14a4-4455-8372-f039da2f15c2@kernel.org/ Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/addrlabel.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index 17ac45aa7194..acd70b5992a7 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -234,7 +234,8 @@ static int __ip6addrlbl_add(struct net *net, struct ip6addrlbl_entry *newp, hlist_add_head_rcu(&newp->list, &net->ipv6.ip6addrlbl_table.head); out: if (!ret) - net->ipv6.ip6addrlbl_table.seq++; + WRITE_ONCE(net->ipv6.ip6addrlbl_table.seq, + net->ipv6.ip6addrlbl_table.seq + 1); return ret; } @@ -445,7 +446,7 @@ static void ip6addrlbl_putmsg(struct nlmsghdr *nlh, }; static int ip6addrlbl_fill(struct sk_buff *skb, - struct ip6addrlbl_entry *p, + const struct ip6addrlbl_entry *p, u32 lseq, u32 portid, u32 seq, int event, unsigned int flags) @@ -498,7 +499,8 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) struct net *net = sock_net(skb->sk); struct ip6addrlbl_entry *p; int idx = 0, s_idx = cb->args[0]; - int err; + int err = 0; + u32 lseq; if (cb->strict_check) { err = ip6addrlbl_valid_dump_req(nlh, cb->extack); @@ -507,10 +509,11 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) } rcu_read_lock(); + lseq = READ_ONCE(net->ipv6.ip6addrlbl_table.seq); hlist_for_each_entry_rcu(p, &net->ipv6.ip6addrlbl_table.head, list) { if (idx >= s_idx) { err = ip6addrlbl_fill(skb, p, - net->ipv6.ip6addrlbl_table.seq, + lseq, NETLINK_CB(cb->skb).portid, nlh->nlmsg_seq, RTM_NEWADDRLABEL, @@ -522,7 +525,7 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) } rcu_read_unlock(); cb->args[0] = idx; - return skb->len; + return err; } static inline int ip6addrlbl_msgsize(void) @@ -614,7 +617,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr *nlh, rcu_read_lock(); p = __ipv6_addr_label(net, addr, ipv6_addr_type(addr), ifal->ifal_index); - lseq = net->ipv6.ip6addrlbl_table.seq; + lseq = READ_ONCE(net->ipv6.ip6addrlbl_table.seq); if (p) err = ip6addrlbl_fill(skb, p, lseq, NETLINK_CB(in_skb).portid, @@ -647,6 +650,7 @@ int __init ipv6_addr_label_rtnl_register(void) return ret; ret = rtnl_register_module(THIS_MODULE, PF_INET6, RTM_GETADDRLABEL, ip6addrlbl_get, - ip6addrlbl_dump, RTNL_FLAG_DOIT_UNLOCKED); + ip6addrlbl_dump, RTNL_FLAG_DOIT_UNLOCKED | + RTNL_FLAG_DUMP_UNLOCKED); return ret; } -- cgit v1.2.3 From 5a66cda52d7daccc88aa358872816c35093759f9 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Thu, 4 Apr 2024 18:03:02 +0200 Subject: ip_tunnel: harden copying IP tunnel params to userspace Structures which are about to be copied to userspace shouldn't have uninitialized fields or paddings. memset() the whole &ip_tunnel_parm in ip_tunnel_parm_to_user() before filling it with the kernel data. The compilers will hopefully combine writes to it. Fixes: 117aef12a7b1 ("ip_tunnel: use a separate struct to store tunnel params in the kernel") Reported-by: Dan Carpenter Closes: https://lore.kernel.org/netdev/5f63dd25-de94-4ca3-84e6-14095953db13@moroto.mountain Signed-off-by: Alexander Lobakin Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/ipv4/ip_tunnel.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index f65170a28106..177f40c3a8e8 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -1034,6 +1034,8 @@ bool ip_tunnel_parm_to_user(void __user *data, struct ip_tunnel_parm_kern *kp) !ip_tunnel_flags_is_be16_compat(kp->o_flags)) return false; + memset(&p, 0, sizeof(p)); + strscpy(p.name, kp->name); p.link = kp->link; p.i_flags = ip_tunnel_flags_to_be16(kp->i_flags); -- cgit v1.2.3 From db77cdc69684890b37ff2bb9fb9e0be2ba4d6efe Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 4 Apr 2024 16:46:04 +0000 Subject: net: dqs: use sysfs_emit() in favor of sprintf() Commit 6025b9135f7a ("net: dqs: add NIC stall detector based on BQL") added three sysfs files. Use the recommended sysfs_emit() helper. Signed-off-by: Eric Dumazet Cc: Jakub Kicinski Cc: Breno Leitao Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index e3d7a8cfa20b..ff3ee45be64a 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -1419,7 +1419,7 @@ static ssize_t bql_show_stall_thrs(struct netdev_queue *queue, char *buf) { struct dql *dql = &queue->dql; - return sprintf(buf, "%u\n", jiffies_to_msecs(dql->stall_thrs)); + return sysfs_emit(buf, "%u\n", jiffies_to_msecs(dql->stall_thrs)); } static ssize_t bql_set_stall_thrs(struct netdev_queue *queue, @@ -1451,7 +1451,7 @@ static struct netdev_queue_attribute bql_stall_thrs_attribute __ro_after_init = static ssize_t bql_show_stall_max(struct netdev_queue *queue, char *buf) { - return sprintf(buf, "%u\n", READ_ONCE(queue->dql.stall_max)); + return sysfs_emit(buf, "%u\n", READ_ONCE(queue->dql.stall_max)); } static ssize_t bql_set_stall_max(struct netdev_queue *queue, @@ -1468,7 +1468,7 @@ static ssize_t bql_show_stall_cnt(struct netdev_queue *queue, char *buf) { struct dql *dql = &queue->dql; - return sprintf(buf, "%lu\n", dql->stall_cnt); + return sysfs_emit(buf, "%lu\n", dql->stall_cnt); } static struct netdev_queue_attribute bql_stall_cnt_attribute __ro_after_init = -- cgit v1.2.3 From d2c3a7eb1afada8cfb5fde41489913ea5733a319 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 5 Apr 2024 10:29:26 +0000 Subject: tcp: more struct tcp_sock adjustments tp->recvmsg_inq is used from tcp recvmsg() thus should be in tcp_sock_read_rx group. tp->tcp_clock_cache and tp->tcp_mstamp are written both in rx and tx paths, thus are better placed in tcp_sock_write_txrx group. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/tcp.h | 6 +++--- net/ipv4/tcp.c | 8 ++++---- 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 55399ee2a57e..6a5e08b937b3 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -244,6 +244,7 @@ struct tcp_sock { /* OOO segments go in this rbtree. Socket lock must be held. */ struct rb_root out_of_order_queue; u32 snd_ssthresh; /* Slow start size threshold */ + u8 recvmsg_inq : 1;/* Indicate # of bytes in queue upon recvmsg */ __cacheline_group_end(tcp_sock_read_rx); /* TX read-write hotpath cache lines */ @@ -266,8 +267,6 @@ struct tcp_sock { u32 mdev_us; /* medium deviation */ u32 rtt_seq; /* sequence number to update rttvar */ u64 tcp_wstamp_ns; /* departure time for next sent data packet */ - u64 tcp_clock_cache; /* cache last tcp_clock_ns() (see tcp_mstamp_refresh()) */ - u64 tcp_mstamp; /* most recent packet received/sent */ struct list_head tsorted_sent_queue; /* time-sorted sent but un-SACKed skbs */ struct sk_buff *highest_sack; /* skb just after the highest * skb with SACKed bit set @@ -284,6 +283,8 @@ struct tcp_sock { * 0x5?10 << 16 + snd_wnd in net byte order */ __be32 pred_flags; + u64 tcp_clock_cache; /* cache last tcp_clock_ns() (see tcp_mstamp_refresh()) */ + u64 tcp_mstamp; /* most recent packet received/sent */ u32 rcv_nxt; /* What we want to receive next */ u32 snd_nxt; /* Next sequence we send */ u32 snd_una; /* First byte we want an ack for */ @@ -370,7 +371,6 @@ struct tcp_sock { tlp_retrans:1, /* TLP is a retransmission */ unused:5; u8 thin_lto : 1,/* Use linear timeouts for thin streams */ - recvmsg_inq : 1,/* Indicate # of bytes in queue upon recvmsg */ fastopen_connect:1, /* FASTOPEN_CONNECT sockopt */ fastopen_no_cookie:1, /* Allow send/recv SYN+data without a cookie */ fastopen_client_fail:2, /* reason why fastopen failed */ diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 92ee60492314..664c8ecb076b 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -4648,16 +4648,16 @@ static void __init tcp_struct_check(void) CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, lsndtime); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, mdev_us); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, tcp_wstamp_ns); - CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, tcp_clock_cache); - CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, tcp_mstamp); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, rtt_seq); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, tsorted_sent_queue); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, highest_sack); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, ecn_flags); - CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_tx, 105); + CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_tx, 89); /* TXRX read-write hotpath cache lines */ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, pred_flags); + CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, tcp_clock_cache); + CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, tcp_mstamp); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rcv_nxt); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, snd_nxt); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, snd_una); @@ -4670,7 +4670,7 @@ static void __init tcp_struct_check(void) CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, app_limited); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rcv_wnd); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rx_opt); - CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 76); + CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 92); /* RX read-write hotpath cache lines */ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_received); -- cgit v1.2.3 From a29689e60ed3e65463d6462390caad669d08a6b7 Mon Sep 17 00:00:00 2001 From: Niklas Schnelle Date: Fri, 5 Apr 2024 13:18:31 +0200 Subject: net: handle HAS_IOPORT dependencies In a future patch HAS_IOPORT=n will disable inb()/outb() and friends at compile time. We thus need to add HAS_IOPORT as dependency for those drivers requiring them. For the DEFXX driver the use of I/O ports is optional and we only need to fence specific code paths. It also turns out that with HAS_IOPORT handled explicitly HAMRADIO does not need the !S390 dependency and successfully builds the bpqether driver. Acked-by: Marc Kleine-Budde Acked-by: Jakub Kicinski Acked-by: Maciej W. Rozycki Co-developed-by: Arnd Bergmann Signed-off-by: Arnd Bergmann Signed-off-by: Niklas Schnelle Signed-off-by: David S. Miller --- drivers/net/Kconfig | 2 +- drivers/net/arcnet/Kconfig | 2 +- drivers/net/can/cc770/Kconfig | 1 + drivers/net/can/sja1000/Kconfig | 1 + drivers/net/ethernet/3com/Kconfig | 4 ++-- drivers/net/ethernet/8390/Kconfig | 6 +++--- drivers/net/ethernet/amd/Kconfig | 4 ++-- drivers/net/ethernet/fujitsu/Kconfig | 2 +- drivers/net/ethernet/intel/Kconfig | 2 +- drivers/net/ethernet/sis/Kconfig | 4 ++-- drivers/net/ethernet/smsc/Kconfig | 2 +- drivers/net/ethernet/ti/Kconfig | 2 +- drivers/net/ethernet/via/Kconfig | 1 + drivers/net/ethernet/xircom/Kconfig | 2 +- drivers/net/fddi/defxx.c | 2 +- drivers/net/hamradio/Kconfig | 6 +++--- drivers/net/wan/Kconfig | 2 +- net/ax25/Kconfig | 2 +- 18 files changed, 25 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 172d84e39129..f0663cf1f755 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -520,7 +520,7 @@ source "drivers/net/ipa/Kconfig" config NET_SB1000 tristate "General Instruments Surfboard 1000" - depends on PNP + depends on ISA && PNP help This is a driver for the General Instrument (also known as NextLevel) SURFboard 1000 internal diff --git a/drivers/net/arcnet/Kconfig b/drivers/net/arcnet/Kconfig index a51b9dab6d3a..d1d07a1d4fbc 100644 --- a/drivers/net/arcnet/Kconfig +++ b/drivers/net/arcnet/Kconfig @@ -4,7 +4,7 @@ # menuconfig ARCNET - depends on NETDEVICES && (ISA || PCI || PCMCIA) + depends on NETDEVICES && (ISA || PCI || PCMCIA) && HAS_IOPORT tristate "ARCnet support" help If you have a network card of this type, say Y and check out the diff --git a/drivers/net/can/cc770/Kconfig b/drivers/net/can/cc770/Kconfig index 9ef1359319f0..467ef19de1c1 100644 --- a/drivers/net/can/cc770/Kconfig +++ b/drivers/net/can/cc770/Kconfig @@ -7,6 +7,7 @@ if CAN_CC770 config CAN_CC770_ISA tristate "ISA Bus based legacy CC770 driver" + depends on ISA help This driver adds legacy support for CC770 and AN82527 chips connected to the ISA bus using I/O port, memory mapped or diff --git a/drivers/net/can/sja1000/Kconfig b/drivers/net/can/sja1000/Kconfig index 4b2f9cb17fc3..01168db4c106 100644 --- a/drivers/net/can/sja1000/Kconfig +++ b/drivers/net/can/sja1000/Kconfig @@ -87,6 +87,7 @@ config CAN_PLX_PCI config CAN_SJA1000_ISA tristate "ISA Bus based legacy SJA1000 driver" + depends on ISA help This driver adds legacy support for SJA1000 chips connected to the ISA bus using I/O port, memory mapped or indirect access. diff --git a/drivers/net/ethernet/3com/Kconfig b/drivers/net/ethernet/3com/Kconfig index 706bd59bf645..1fbab79e2be4 100644 --- a/drivers/net/ethernet/3com/Kconfig +++ b/drivers/net/ethernet/3com/Kconfig @@ -44,7 +44,7 @@ config 3C515 config PCMCIA_3C574 tristate "3Com 3c574 PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT help Say Y here if you intend to attach a 3Com 3c574 or compatible PCMCIA (PC-card) Fast Ethernet card to your computer. @@ -54,7 +54,7 @@ config PCMCIA_3C574 config PCMCIA_3C589 tristate "3Com 3c589 PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT help Say Y here if you intend to attach a 3Com 3c589 or compatible PCMCIA (PC-card) Ethernet card to your computer. diff --git a/drivers/net/ethernet/8390/Kconfig b/drivers/net/ethernet/8390/Kconfig index a4130e643342..345f250781c6 100644 --- a/drivers/net/ethernet/8390/Kconfig +++ b/drivers/net/ethernet/8390/Kconfig @@ -19,7 +19,7 @@ if NET_VENDOR_8390 config PCMCIA_AXNET tristate "Asix AX88190 PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT help Say Y here if you intend to attach an Asix AX88190-based PCMCIA (PC-card) Fast Ethernet card to your computer. These cards are @@ -117,7 +117,7 @@ config NE2000 config NE2K_PCI tristate "PCI NE2000 and clones support (see help)" - depends on PCI + depends on PCI && HAS_IOPORT select CRC32 help This driver is for NE2000 compatible PCI cards. It will not work @@ -146,7 +146,7 @@ config APNE config PCMCIA_PCNET tristate "NE2000 compatible PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT select CRC32 help Say Y here if you intend to attach an NE2000 compatible PCMCIA diff --git a/drivers/net/ethernet/amd/Kconfig b/drivers/net/ethernet/amd/Kconfig index f8cc8925161c..b39c6f3e1eda 100644 --- a/drivers/net/ethernet/amd/Kconfig +++ b/drivers/net/ethernet/amd/Kconfig @@ -56,7 +56,7 @@ config LANCE config PCNET32 tristate "AMD PCnet32 PCI support" - depends on PCI + depends on PCI && HAS_IOPORT select CRC32 select MII help @@ -122,7 +122,7 @@ config MVME147_NET config PCMCIA_NMCLAN tristate "New Media PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT help Say Y here if you intend to attach a New Media Ethernet or LiveWire PCMCIA (PC-card) Ethernet card to your computer. diff --git a/drivers/net/ethernet/fujitsu/Kconfig b/drivers/net/ethernet/fujitsu/Kconfig index 0a1400cb410a..06a28bce5d27 100644 --- a/drivers/net/ethernet/fujitsu/Kconfig +++ b/drivers/net/ethernet/fujitsu/Kconfig @@ -18,7 +18,7 @@ if NET_VENDOR_FUJITSU config PCMCIA_FMVJ18X tristate "Fujitsu FMV-J18x PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT select CRC32 help Say Y here if you intend to attach a Fujitsu FMV-J18x or compatible diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 639fbb12bd35..6e7901e12699 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -41,7 +41,7 @@ config E100 config E1000 tristate "Intel(R) PRO/1000 Gigabit Ethernet support" - depends on PCI + depends on PCI && HAS_IOPORT help This driver supports Intel(R) PRO/1000 gigabit ethernet family of adapters. For more information on how to identify your adapter, go diff --git a/drivers/net/ethernet/sis/Kconfig b/drivers/net/ethernet/sis/Kconfig index 775d76d9890e..7e498bdbca73 100644 --- a/drivers/net/ethernet/sis/Kconfig +++ b/drivers/net/ethernet/sis/Kconfig @@ -19,7 +19,7 @@ if NET_VENDOR_SIS config SIS900 tristate "SiS 900/7016 PCI Fast Ethernet Adapter support" - depends on PCI + depends on PCI && HAS_IOPORT select CRC32 select MII help @@ -35,7 +35,7 @@ config SIS900 config SIS190 tristate "SiS190/SiS191 gigabit ethernet support" - depends on PCI + depends on PCI && HAS_IOPORT select CRC32 select MII help diff --git a/drivers/net/ethernet/smsc/Kconfig b/drivers/net/ethernet/smsc/Kconfig index 5f22a8a4d27b..13ce9086a9ca 100644 --- a/drivers/net/ethernet/smsc/Kconfig +++ b/drivers/net/ethernet/smsc/Kconfig @@ -54,7 +54,7 @@ config SMC91X config PCMCIA_SMC91C92 tristate "SMC 91Cxx PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT select CRC32 select MII help diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig index 1530d13984d4..cfe9bd6b1d62 100644 --- a/drivers/net/ethernet/ti/Kconfig +++ b/drivers/net/ethernet/ti/Kconfig @@ -167,7 +167,7 @@ config TI_KEYSTONE_NETCP_ETHSS config TLAN tristate "TI ThunderLAN support" - depends on (PCI || EISA) + depends on (PCI || EISA) && HAS_IOPORT help If you have a PCI Ethernet network card based on the ThunderLAN chip which is supported by this driver, say Y here. diff --git a/drivers/net/ethernet/via/Kconfig b/drivers/net/ethernet/via/Kconfig index da287ef65be7..00773f5e4d7e 100644 --- a/drivers/net/ethernet/via/Kconfig +++ b/drivers/net/ethernet/via/Kconfig @@ -20,6 +20,7 @@ config VIA_RHINE tristate "VIA Rhine support" depends on PCI || (OF_IRQ && GENERIC_PCI_IOMAP) depends on PCI || ARCH_VT8500 || COMPILE_TEST + depends on HAS_IOPORT depends on HAS_DMA select CRC32 select MII diff --git a/drivers/net/ethernet/xircom/Kconfig b/drivers/net/ethernet/xircom/Kconfig index 7497b9bea511..bfbdcf758afb 100644 --- a/drivers/net/ethernet/xircom/Kconfig +++ b/drivers/net/ethernet/xircom/Kconfig @@ -19,7 +19,7 @@ if NET_VENDOR_XIRCOM config PCMCIA_XIRC2PS tristate "Xircom 16-bit PCMCIA support" - depends on PCMCIA + depends on PCMCIA && HAS_IOPORT help Say Y here if you intend to attach a Xircom 16-bit PCMCIA (PC-card) Ethernet or Fast Ethernet card to your computer. diff --git a/drivers/net/fddi/defxx.c b/drivers/net/fddi/defxx.c index 1fef8a9b1a0f..0fbbb7286008 100644 --- a/drivers/net/fddi/defxx.c +++ b/drivers/net/fddi/defxx.c @@ -254,7 +254,7 @@ static const char version[] = #define DFX_BUS_TC(dev) 0 #endif -#if defined(CONFIG_EISA) || defined(CONFIG_PCI) +#ifdef CONFIG_HAS_IOPORT #define dfx_use_mmio bp->mmio #else #define dfx_use_mmio true diff --git a/drivers/net/hamradio/Kconfig b/drivers/net/hamradio/Kconfig index 25b1f929c422..36a9aade9f33 100644 --- a/drivers/net/hamradio/Kconfig +++ b/drivers/net/hamradio/Kconfig @@ -83,7 +83,7 @@ config SCC_TRXECHO config BAYCOM_SER_FDX tristate "BAYCOM ser12 fullduplex driver for AX.25" - depends on AX25 && !S390 + depends on AX25 && HAS_IOPORT select CRC_CCITT help This is one of two drivers for Baycom style simple amateur radio @@ -103,7 +103,7 @@ config BAYCOM_SER_FDX config BAYCOM_SER_HDX tristate "BAYCOM ser12 halfduplex driver for AX.25" - depends on AX25 && !S390 + depends on AX25 && HAS_IOPORT select CRC_CCITT help This is one of two drivers for Baycom style simple amateur radio @@ -150,7 +150,7 @@ config BAYCOM_EPP config YAM tristate "YAM driver for AX.25" - depends on AX25 && !S390 + depends on AX25 && HAS_IOPORT help The YAM is a modem for packet radio which connects to the serial port and includes some of the functions of a Terminal Node diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig index 31ab2136cdf1..67be9857c86c 100644 --- a/drivers/net/wan/Kconfig +++ b/drivers/net/wan/Kconfig @@ -180,7 +180,7 @@ config C101 config FARSYNC tristate "FarSync T-Series support" - depends on HDLC && PCI + depends on HDLC && PCI && HAS_IOPORT help Support for the FarSync T-Series X.21 (and V.35/V.24) cards by FarSite Communications Ltd. diff --git a/net/ax25/Kconfig b/net/ax25/Kconfig index fdb666607f10..e23a3dc14b93 100644 --- a/net/ax25/Kconfig +++ b/net/ax25/Kconfig @@ -4,7 +4,7 @@ # menuconfig HAMRADIO - depends on NET && !S390 + depends on NET bool "Amateur Radio support" help If you want to connect your Linux box to an amateur radio, answer Y -- cgit v1.2.3 From 86d43e2bf93ccac88ef71cee36a23282ebd9e427 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 5 Apr 2024 11:49:39 +0000 Subject: af_packet: avoid a false positive warning in packet_setsockopt() Although the code is correct, the following line copy_from_sockptr(&req_u.req, optval, len)); triggers this warning : memcpy: detected field-spanning write (size 28) of single field "dst" at include/linux/sockptr.h:49 (size 16) Refactor the code to be more explicit. Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: Kees Cook Cc: Willem de Bruijn Reviewed-by: Kees Cook Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/packet/af_packet.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 18f616f487ea..8c6d3fbb4ed8 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -3800,28 +3800,30 @@ packet_setsockopt(struct socket *sock, int level, int optname, sockptr_t optval, case PACKET_TX_RING: { union tpacket_req_u req_u; - int len; + ret = -EINVAL; lock_sock(sk); switch (po->tp_version) { case TPACKET_V1: case TPACKET_V2: - len = sizeof(req_u.req); + if (optlen < sizeof(req_u.req)) + break; + ret = copy_from_sockptr(&req_u.req, optval, + sizeof(req_u.req)) ? + -EINVAL : 0; break; case TPACKET_V3: default: - len = sizeof(req_u.req3); + if (optlen < sizeof(req_u.req3)) + break; + ret = copy_from_sockptr(&req_u.req3, optval, + sizeof(req_u.req3)) ? + -EINVAL : 0; break; } - if (optlen < len) { - ret = -EINVAL; - } else { - if (copy_from_sockptr(&req_u.req, optval, len)) - ret = -EFAULT; - else - ret = packet_set_ring(sk, &req_u, 0, - optname == PACKET_TX_RING); - } + if (!ret) + ret = packet_set_ring(sk, &req_u, 0, + optname == PACKET_TX_RING); release_sock(sk); return ret; } -- cgit v1.2.3 From ec20b283009346adc66d5a460b1f8fb5adafbcfe Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Fri, 5 Apr 2024 22:05:00 +0200 Subject: ipv4: Set scope explicitly in ip_route_output(). Add a "scope" parameter to ip_route_output() so that callers don't have to override the tos parameter with the RTO_ONLINK flag if they want a local scope. This will allow converting flowi4_tos to dscp_t in the future, thus allowing static analysers to flag invalid interactions between "tos" (the DSCP bits) and ECN. Only three users ask for local scope (bonding, arp and atm). The others continue to use RT_SCOPE_UNIVERSE. While there, add a comment to warn users about the limitations of ip_route_output(). Signed-off-by: Guillaume Nault Acked-by: Leon Romanovsky # infiniband Signed-off-by: David S. Miller --- drivers/infiniband/hw/irdma/cm.c | 3 ++- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 3 ++- drivers/net/bonding/bond_main.c | 4 ++-- drivers/net/ethernet/broadcom/cnic.c | 3 ++- include/net/route.h | 9 ++++++++- net/atm/clip.c | 2 +- net/bridge/br_netfilter_hooks.c | 3 ++- net/ipv4/arp.c | 9 ++++++--- net/ipv4/igmp.c | 3 ++- net/mpls/af_mpls.c | 2 +- 10 files changed, 28 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/drivers/infiniband/hw/irdma/cm.c b/drivers/infiniband/hw/irdma/cm.c index 1ee7a4e0d8d8..36bb7e5ce638 100644 --- a/drivers/infiniband/hw/irdma/cm.c +++ b/drivers/infiniband/hw/irdma/cm.c @@ -1985,7 +1985,8 @@ static int irdma_addr_resolve_neigh(struct irdma_device *iwdev, u32 src_ip, __be32 dst_ipaddr = htonl(dst_ip); __be32 src_ipaddr = htonl(src_ip); - rt = ip_route_output(&init_net, dst_ipaddr, src_ipaddr, 0, 0); + rt = ip_route_output(&init_net, dst_ipaddr, src_ipaddr, 0, 0, + RT_SCOPE_UNIVERSE); if (IS_ERR(rt)) { ibdev_dbg(&iwdev->ibdev, "CM: ip_route_output fail\n"); return -EINVAL; diff --git a/drivers/infiniband/hw/qedr/qedr_iw_cm.c b/drivers/infiniband/hw/qedr/qedr_iw_cm.c index a51fc6854984..259303b9907c 100644 --- a/drivers/infiniband/hw/qedr/qedr_iw_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_iw_cm.c @@ -447,7 +447,8 @@ qedr_addr4_resolve(struct qedr_dev *dev, struct rtable *rt = NULL; int rc = 0; - rt = ip_route_output(&init_net, dst_ip, src_ip, 0, 0); + rt = ip_route_output(&init_net, dst_ip, src_ip, 0, 0, + RT_SCOPE_UNIVERSE); if (IS_ERR(rt)) { DP_ERR(dev, "ip_route_output returned error\n"); return -EINVAL; diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 2c5ed0a7cb18..c9f0415f780a 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3014,8 +3014,8 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) tags = NULL; /* Find out through which dev should the packet go */ - rt = ip_route_output(dev_net(bond->dev), targets[i], 0, - RTO_ONLINK, 0); + rt = ip_route_output(dev_net(bond->dev), targets[i], 0, 0, 0, + RT_SCOPE_LINK); if (IS_ERR(rt)) { /* there's no route to target - try to send arp * probe to generate any traffic (arp_validate=0) diff --git a/drivers/net/ethernet/broadcom/cnic.c b/drivers/net/ethernet/broadcom/cnic.c index 3d63177e7e52..c2b4188a1ef1 100644 --- a/drivers/net/ethernet/broadcom/cnic.c +++ b/drivers/net/ethernet/broadcom/cnic.c @@ -3682,7 +3682,8 @@ static int cnic_get_v4_route(struct sockaddr_in *dst_addr, #if defined(CONFIG_INET) struct rtable *rt; - rt = ip_route_output(&init_net, dst_addr->sin_addr.s_addr, 0, 0, 0); + rt = ip_route_output(&init_net, dst_addr->sin_addr.s_addr, 0, 0, 0, + RT_SCOPE_UNIVERSE); if (!IS_ERR(rt)) { *dst = &rt->dst; return 0; diff --git a/include/net/route.h b/include/net/route.h index d4a0147942f1..315a8acee6c6 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -141,15 +141,22 @@ static inline struct rtable *ip_route_output_key(struct net *net, struct flowi4 return ip_route_output_flow(net, flp, NULL); } +/* Simplistic IPv4 route lookup function. + * This is only suitable for some particular use cases: since the flowi4 + * structure is only partially set, it may bypass some fib-rules. + */ static inline struct rtable *ip_route_output(struct net *net, __be32 daddr, - __be32 saddr, u8 tos, int oif) + __be32 saddr, u8 tos, int oif, + __u8 scope) { struct flowi4 fl4 = { .flowi4_oif = oif, .flowi4_tos = tos, + .flowi4_scope = scope, .daddr = daddr, .saddr = saddr, }; + return ip_route_output_key(net, &fl4); } diff --git a/net/atm/clip.c b/net/atm/clip.c index 294cb9efe3d3..362e8d25a79e 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -463,7 +463,7 @@ static int clip_setentry(struct atm_vcc *vcc, __be32 ip) unlink_clip_vcc(clip_vcc); return 0; } - rt = ip_route_output(&init_net, ip, 0, 1, 0); + rt = ip_route_output(&init_net, ip, 0, 0, 0, RT_SCOPE_LINK); if (IS_ERR(rt)) return PTR_ERR(rt); neigh = __neigh_lookup(&arp_tbl, &ip, rt->dst.dev, 1); diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index 35e10c5a766d..4242447be322 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -399,7 +399,8 @@ static int br_nf_pre_routing_finish(struct net *net, struct sock *sk, struct sk_ goto free_skb; rt = ip_route_output(net, iph->daddr, 0, - RT_TOS(iph->tos), 0); + RT_TOS(iph->tos), 0, + RT_SCOPE_UNIVERSE); if (!IS_ERR(rt)) { /* - Bridged-and-DNAT'ed traffic doesn't * require ip_forwarding. */ diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 0d0d725b46ad..ab82ca104496 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -456,7 +456,8 @@ static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev) /*unsigned long now; */ struct net *net = dev_net(dev); - rt = ip_route_output(net, sip, tip, 0, l3mdev_master_ifindex_rcu(dev)); + rt = ip_route_output(net, sip, tip, 0, l3mdev_master_ifindex_rcu(dev), + RT_SCOPE_UNIVERSE); if (IS_ERR(rt)) return 1; if (rt->dst.dev != dev) { @@ -1056,7 +1057,8 @@ static int arp_req_set(struct net *net, struct arpreq *r, if (r->arp_flags & ATF_PERM) r->arp_flags |= ATF_COM; if (!dev) { - struct rtable *rt = ip_route_output(net, ip, 0, RTO_ONLINK, 0); + struct rtable *rt = ip_route_output(net, ip, 0, 0, 0, + RT_SCOPE_LINK); if (IS_ERR(rt)) return PTR_ERR(rt); @@ -1188,7 +1190,8 @@ static int arp_req_delete(struct net *net, struct arpreq *r, ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; if (!dev) { - struct rtable *rt = ip_route_output(net, ip, 0, RTO_ONLINK, 0); + struct rtable *rt = ip_route_output(net, ip, 0, 0, 0, + RT_SCOPE_LINK); if (IS_ERR(rt)) return PTR_ERR(rt); dev = rt->dst.dev; diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 717e97a389a8..9bf09de6a2e7 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1842,7 +1842,8 @@ static struct in_device *ip_mc_find_dev(struct net *net, struct ip_mreqn *imr) if (!dev) { struct rtable *rt = ip_route_output(net, imr->imr_multiaddr.s_addr, - 0, 0, 0); + 0, 0, 0, + RT_SCOPE_UNIVERSE); if (!IS_ERR(rt)) { dev = rt->dst.dev; ip_rt_put(rt); diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 6dab883a08dd..1303acb9cdd2 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -594,7 +594,7 @@ static struct net_device *inet_fib_lookup_dev(struct net *net, struct in_addr daddr; memcpy(&daddr, addr, sizeof(struct in_addr)); - rt = ip_route_output(net, daddr.s_addr, 0, 0, 0); + rt = ip_route_output(net, daddr.s_addr, 0, 0, 0, RT_SCOPE_UNIVERSE); if (IS_ERR(rt)) return ERR_CAST(rt); -- cgit v1.2.3 From 382c60019ee769a8f3743e8340c063e4aa71231e Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 6 Apr 2024 09:48:48 +0800 Subject: mptcp: add reset reason options in some places The reason codes are handled in two ways nowadays (quoting Mat Martineau): 1. Sending in the MPTCP option on RST packets when there is no subflow context available (these use subflow_add_reset_reason() and directly call a TCP-level send_reset function) 2. The "normal" way via subflow->reset_reason. This will propagate to both the outgoing reset packet and to a local path manager process via netlink in mptcp_event_sub_closed() RFC 8684 defines the skb reset reason behaviour which is not required even though in some places: A host sends a TCP RST in order to close a subflow or reject an attempt to open a subflow (MP_JOIN). In order to let the receiving host know why a subflow is being closed or rejected, the TCP RST packet MAY include the MP_TCPRST option (Figure 15). The host MAY use this information to decide, for example, whether it tries to re-establish the subflow immediately, later, or never. Since the commit dc87efdb1a5cd ("mptcp: add mptcp reset option support") introduced this feature about three years ago, we can fully use it. There remains some places where we could insert reason into skb as we can see in this patch. Many thanks to Mat and Paolo for help:) Signed-off-by: Jason Xing Acked-by: Paolo Abeni Signed-off-by: David S. Miller --- net/mptcp/subflow.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 6042a47da61b..162b218d9858 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -150,8 +150,10 @@ static int subflow_check_req(struct request_sock *req, /* no MPTCP if MD5SIG is enabled on this socket or we may run out of * TCP option space. */ - if (rcu_access_pointer(tcp_sk(sk_listener)->md5sig_info)) + if (rcu_access_pointer(tcp_sk(sk_listener)->md5sig_info)) { + subflow_add_reset_reason(skb, MPTCP_RST_EMPTCP); return -EINVAL; + } #endif mptcp_get_options(skb, &mp_opt); @@ -219,6 +221,7 @@ again: ntohs(inet_sk((struct sock *)subflow_req->msk)->inet_sport)); if (!mptcp_pm_sport_in_anno_list(subflow_req->msk, sk_listener)) { SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MISMATCHPORTSYNRX); + subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT); return -EPERM; } SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINPORTSYNRX); @@ -227,10 +230,12 @@ again: subflow_req_create_thmac(subflow_req); if (unlikely(req->syncookie)) { - if (mptcp_can_accept_new_subflow(subflow_req->msk)) - subflow_init_req_cookie_join_save(subflow_req, skb); - else + if (!mptcp_can_accept_new_subflow(subflow_req->msk)) { + subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT); return -EPERM; + } + + subflow_init_req_cookie_join_save(subflow_req, skb); } pr_debug("token=%u, remote_nonce=%u msk=%p", subflow_req->token, @@ -873,13 +878,18 @@ create_child: ntohs(inet_sk((struct sock *)owner)->inet_sport)); if (!mptcp_pm_sport_in_anno_list(owner, sk)) { SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MISMATCHPORTACKRX); + subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT); goto dispose_child; } SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINPORTACKRX); } - if (!mptcp_finish_join(child)) + if (!mptcp_finish_join(child)) { + struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(child); + + subflow_add_reset_reason(skb, subflow->reset_reason); goto dispose_child; + } SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKRX); tcp_rsk(req)->drop_req = true; -- cgit v1.2.3 From 4308811ba90118ae1b71a95fee79ab7dada6400c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 7 Apr 2024 08:06:06 +0000 Subject: net: display more skb fields in skb_dump() Print these additional fields in skb_dump() to ease debugging. - mac_len - csum_start (in v2, at Willem suggestion) - csum_offset (in v2, at Willem suggestion) - priority - mark - alloc_cpu - vlan_all - encapsulation - inner_protocol - inner_mac_header - inner_network_header - inner_transport_header Signed-off-by: Eric Dumazet Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/core/skbuff.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 2a5ce6667bbb..21cd01641f4c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1304,22 +1304,28 @@ void skb_dump(const char *level, const struct sk_buff *skb, bool full_pkt) has_trans = skb_transport_header_was_set(skb); printk("%sskb len=%u headroom=%u headlen=%u tailroom=%u\n" - "mac=(%d,%d) net=(%d,%d) trans=%d\n" + "mac=(%d,%d) mac_len=%u net=(%d,%d) trans=%d\n" "shinfo(txflags=%u nr_frags=%u gso(size=%hu type=%u segs=%hu))\n" - "csum(0x%x ip_summed=%u complete_sw=%u valid=%u level=%u)\n" - "hash(0x%x sw=%u l4=%u) proto=0x%04x pkttype=%u iif=%d\n", + "csum(0x%x start=%u offset=%u ip_summed=%u complete_sw=%u valid=%u level=%u)\n" + "hash(0x%x sw=%u l4=%u) proto=0x%04x pkttype=%u iif=%d\n" + "priority=0x%x mark=0x%x alloc_cpu=%u vlan_all=0x%x\n" + "encapsulation=%d inner(proto=0x%04x, mac=%u, net=%u, trans=%u)\n", level, skb->len, headroom, skb_headlen(skb), tailroom, has_mac ? skb->mac_header : -1, has_mac ? skb_mac_header_len(skb) : -1, + skb->mac_len, skb->network_header, has_trans ? skb_network_header_len(skb) : -1, has_trans ? skb->transport_header : -1, sh->tx_flags, sh->nr_frags, sh->gso_size, sh->gso_type, sh->gso_segs, - skb->csum, skb->ip_summed, skb->csum_complete_sw, - skb->csum_valid, skb->csum_level, + skb->csum, skb->csum_start, skb->csum_offset, skb->ip_summed, + skb->csum_complete_sw, skb->csum_valid, skb->csum_level, skb->hash, skb->sw_hash, skb->l4_hash, - ntohs(skb->protocol), skb->pkt_type, skb->skb_iif); + ntohs(skb->protocol), skb->pkt_type, skb->skb_iif, + skb->priority, skb->mark, skb->alloc_cpu, skb->vlan_all, + skb->encapsulation, skb->inner_protocol, skb->inner_mac_header, + skb->inner_network_header, skb->inner_transport_header); if (dev) printk("%sdev name=%s feat=%pNF\n", -- cgit v1.2.3 From 5af3e3876d567fb79a355bec1cb48e432d69b4fb Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Sat, 6 Apr 2024 04:05:37 +0300 Subject: devlink: Support setting max_io_eqs Many devices send event notifications for the IO queues, such as tx and rx queues, through event queues. Enable a privileged owner, such as a hypervisor PF, to set the number of IO event queues for the VF and SF during the provisioning stage. example: Get maximum IO event queues of the VF device:: $ devlink port show pci/0000:06:00.0/2 pci/0000:06:00.0/2: type eth netdev enp6s0pf0vf1 flavour pcivf pfnum 0 vfnum 1 function: hw_addr 00:00:00:00:00:00 ipsec_packet disabled max_io_eqs 10 Set maximum IO event queues of the VF device:: $ devlink port function set pci/0000:06:00.0/2 max_io_eqs 32 $ devlink port show pci/0000:06:00.0/2 pci/0000:06:00.0/2: type eth netdev enp6s0pf0vf1 flavour pcivf pfnum 0 vfnum 1 function: hw_addr 00:00:00:00:00:00 ipsec_packet disabled max_io_eqs 32 Reviewed-by: Jiri Pirko Reviewed-by: Shay Drory Signed-off-by: Parav Pandit Signed-off-by: David S. Miller --- Documentation/networking/devlink/devlink-port.rst | 33 ++++++++++++++ include/net/devlink.h | 14 ++++++ include/uapi/linux/devlink.h | 1 + net/devlink/port.c | 53 +++++++++++++++++++++++ 4 files changed, 101 insertions(+) (limited to 'net') diff --git a/Documentation/networking/devlink/devlink-port.rst b/Documentation/networking/devlink/devlink-port.rst index 562f46b41274..9d22d41a7cd1 100644 --- a/Documentation/networking/devlink/devlink-port.rst +++ b/Documentation/networking/devlink/devlink-port.rst @@ -134,6 +134,9 @@ Users may also set the IPsec crypto capability of the function using Users may also set the IPsec packet capability of the function using `devlink port function set ipsec_packet` command. +Users may also set the maximum IO event queues of the function +using `devlink port function set max_io_eqs` command. + Function attributes =================== @@ -295,6 +298,36 @@ policy is processed in software by the kernel. function: hw_addr 00:00:00:00:00:00 ipsec_packet enabled +Maximum IO events queues setup +------------------------------ +When user sets maximum number of IO event queues for a SF or +a VF, such function driver is limited to consume only enforced +number of IO event queues. + +IO event queues deliver events related to IO queues, including network +device transmit and receive queues (txq and rxq) and RDMA Queue Pairs (QPs). +For example, the number of netdevice channels and RDMA device completion +vectors are derived from the function's IO event queues. Usually, the number +of interrupt vectors consumed by the driver is limited by the number of IO +event queues per device, as each of the IO event queues is connected to an +interrupt vector. + +- Get maximum IO event queues of the VF device:: + + $ devlink port show pci/0000:06:00.0/2 + pci/0000:06:00.0/2: type eth netdev enp6s0pf0vf1 flavour pcivf pfnum 0 vfnum 1 + function: + hw_addr 00:00:00:00:00:00 ipsec_packet disabled max_io_eqs 10 + +- Set maximum IO event queues of the VF device:: + + $ devlink port function set pci/0000:06:00.0/2 max_io_eqs 32 + + $ devlink port show pci/0000:06:00.0/2 + pci/0000:06:00.0/2: type eth netdev enp6s0pf0vf1 flavour pcivf pfnum 0 vfnum 1 + function: + hw_addr 00:00:00:00:00:00 ipsec_packet disabled max_io_eqs 32 + Subfunction ============ diff --git a/include/net/devlink.h b/include/net/devlink.h index 9ac394bdfbe4..bb1af599d101 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -1602,6 +1602,14 @@ void devlink_free(struct devlink *devlink); * capability. Should be used by device drivers to * enable/disable ipsec_packet capability of a * function managed by the devlink port. + * @port_fn_max_io_eqs_get: Callback used to get port function's maximum number + * of event queues. Should be used by device drivers to + * report the maximum event queues of a function + * managed by the devlink port. + * @port_fn_max_io_eqs_set: Callback used to set port function's maximum number + * of event queues. Should be used by device drivers to + * configure maximum number of event queues + * of a function managed by the devlink port. * * Note: Driver should return -EOPNOTSUPP if it doesn't support * port function (@port_fn_*) handling for a particular port. @@ -1651,6 +1659,12 @@ struct devlink_port_ops { int (*port_fn_ipsec_packet_set)(struct devlink_port *devlink_port, bool enable, struct netlink_ext_ack *extack); + int (*port_fn_max_io_eqs_get)(struct devlink_port *devlink_port, + u32 *max_eqs, + struct netlink_ext_ack *extack); + int (*port_fn_max_io_eqs_set)(struct devlink_port *devlink_port, + u32 max_eqs, + struct netlink_ext_ack *extack); }; void devlink_port_init(struct devlink *devlink, diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h index 2da0c7eb6710..9401aa343673 100644 --- a/include/uapi/linux/devlink.h +++ b/include/uapi/linux/devlink.h @@ -686,6 +686,7 @@ enum devlink_port_function_attr { DEVLINK_PORT_FN_ATTR_OPSTATE, /* u8 */ DEVLINK_PORT_FN_ATTR_CAPS, /* bitfield32 */ DEVLINK_PORT_FN_ATTR_DEVLINK, /* nested */ + DEVLINK_PORT_FN_ATTR_MAX_IO_EQS, /* u32 */ __DEVLINK_PORT_FUNCTION_ATTR_MAX, DEVLINK_PORT_FUNCTION_ATTR_MAX = __DEVLINK_PORT_FUNCTION_ATTR_MAX - 1 diff --git a/net/devlink/port.c b/net/devlink/port.c index 118d130d2afd..be9158b4453c 100644 --- a/net/devlink/port.c +++ b/net/devlink/port.c @@ -16,6 +16,7 @@ static const struct nla_policy devlink_function_nl_policy[DEVLINK_PORT_FUNCTION_ DEVLINK_PORT_FN_STATE_ACTIVE), [DEVLINK_PORT_FN_ATTR_CAPS] = NLA_POLICY_BITFIELD32(DEVLINK_PORT_FN_CAPS_VALID_MASK), + [DEVLINK_PORT_FN_ATTR_MAX_IO_EQS] = { .type = NLA_U32 }, }; #define ASSERT_DEVLINK_PORT_REGISTERED(devlink_port) \ @@ -182,6 +183,30 @@ static int devlink_port_fn_caps_fill(struct devlink_port *devlink_port, return 0; } +static int devlink_port_fn_max_io_eqs_fill(struct devlink_port *port, + struct sk_buff *msg, + struct netlink_ext_ack *extack, + bool *msg_updated) +{ + u32 max_io_eqs; + int err; + + if (!port->ops->port_fn_max_io_eqs_get) + return 0; + + err = port->ops->port_fn_max_io_eqs_get(port, &max_io_eqs, extack); + if (err) { + if (err == -EOPNOTSUPP) + return 0; + return err; + } + err = nla_put_u32(msg, DEVLINK_PORT_FN_ATTR_MAX_IO_EQS, max_io_eqs); + if (err) + return err; + *msg_updated = true; + return 0; +} + int devlink_nl_port_handle_fill(struct sk_buff *msg, struct devlink_port *devlink_port) { if (devlink_nl_put_handle(msg, devlink_port->devlink)) @@ -409,6 +434,18 @@ static int devlink_port_fn_caps_set(struct devlink_port *devlink_port, return 0; } +static int +devlink_port_fn_max_io_eqs_set(struct devlink_port *devlink_port, + const struct nlattr *attr, + struct netlink_ext_ack *extack) +{ + u32 max_io_eqs; + + max_io_eqs = nla_get_u32(attr); + return devlink_port->ops->port_fn_max_io_eqs_set(devlink_port, + max_io_eqs, extack); +} + static int devlink_nl_port_function_attrs_put(struct sk_buff *msg, struct devlink_port *port, struct netlink_ext_ack *extack) @@ -428,6 +465,9 @@ devlink_nl_port_function_attrs_put(struct sk_buff *msg, struct devlink_port *por if (err) goto out; err = devlink_port_fn_state_fill(port, msg, extack, &msg_updated); + if (err) + goto out; + err = devlink_port_fn_max_io_eqs_fill(port, msg, extack, &msg_updated); if (err) goto out; err = devlink_rel_devlink_handle_put(msg, port->devlink, @@ -726,6 +766,12 @@ static int devlink_port_function_validate(struct devlink_port *devlink_port, } } } + if (tb[DEVLINK_PORT_FN_ATTR_MAX_IO_EQS] && + !ops->port_fn_max_io_eqs_set) { + NL_SET_ERR_MSG_ATTR(extack, tb[DEVLINK_PORT_FN_ATTR_MAX_IO_EQS], + "Function does not support max_io_eqs setting"); + return -EOPNOTSUPP; + } return 0; } @@ -761,6 +807,13 @@ static int devlink_port_function_set(struct devlink_port *port, return err; } + attr = tb[DEVLINK_PORT_FN_ATTR_MAX_IO_EQS]; + if (attr) { + err = devlink_port_fn_max_io_eqs_set(port, attr, extack); + if (err) + return err; + } + /* Keep this as the last function attribute set, so that when * multiple port function attributes are set along with state, * Those can be applied first before activating the state. -- cgit v1.2.3 From 499921d3dcabd48cb17f4f4ca97c0e17f05cd7b6 Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Thu, 28 Mar 2024 22:44:13 -0700 Subject: wifi: mac80211: correctly document struct mesh_table Currently kernel-doc -Wall reports: net/mac80211/ieee80211_i.h:687: warning: missing initial short description on line: * struct mesh_table So add a short description. Reported-by: Randy Dunlap Link: https://lore.kernel.org/linux-wireless/a009a21a-56d7-4a1a-aaf9-feefa5acc561@infradead.org/ Signed-off-by: Jeff Johnson Reviewed-by: Randy Dunlap Link: https://msgid.link/20240328-mesh_table-kerneldoc-v1-1-174c4df341b1@quicinc.com Signed-off-by: Johannes Berg --- net/mac80211/ieee80211_i.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 458f63ce9a32..81a9c5b7642d 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -686,7 +686,7 @@ struct mesh_csa_settings { }; /** - * struct mesh_table + * struct mesh_table - mesh hash table * * @known_gates: list of known mesh gates and their mpaths by the station. The * gate's mpath may or may not be resolved and active. -- cgit v1.2.3 From a35b36e6ee5dcf323d846ec3881159e4af90c9b4 Mon Sep 17 00:00:00 2001 From: Michael-CY Lee Date: Tue, 26 Mar 2024 08:30:36 +0800 Subject: wifi: mac80211: extend IEEE80211_KEY_FLAG_GENERATE_MMIE to other ciphers Extend the flag IEEE80211_KEY_FLAG_GENERATE_MMIE to BIP-CMAC-256, BIP-GMAC-128 and BIP-GMAC-256 for the same reason and in the same way that the flag was added originally in commit a0b4496a4368 ("mac80211: add IEEE80211_KEY_FLAG_GENERATE_MMIE to ieee80211_key_flags"). Signed-off-by: Michael-CY Lee Link: https://msgid.link/20240326003036.15215-1-michael-cy.lee@mediatek.com Signed-off-by: Johannes Berg --- include/net/mac80211.h | 4 ++-- net/mac80211/wpa.c | 12 ++++++++++-- 2 files changed, 12 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 460387916c2e..41f135a793ab 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -2124,8 +2124,8 @@ static inline bool lockdep_vif_wiphy_mutex_held(struct ieee80211_vif *vif) * @IEEE80211_KEY_FLAG_GENERATE_MMIC on the same key. * @IEEE80211_KEY_FLAG_NO_AUTO_TX: Key needs explicit Tx activation. * @IEEE80211_KEY_FLAG_GENERATE_MMIE: This flag should be set by the driver - * for a AES_CMAC key to indicate that it requires sequence number - * generation only + * for a AES_CMAC or a AES_GMAC key to indicate that it requires sequence + * number generation only * @IEEE80211_KEY_FLAG_SPP_AMSDU: SPP A-MSDUs can be used with this key * (set by mac80211 from the sta->spp_amsdu flag) */ diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index e40529b8c5c9..047a33797020 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -895,7 +895,8 @@ ieee80211_crypto_aes_cmac_256_encrypt(struct ieee80211_tx_data *tx) info = IEEE80211_SKB_CB(skb); - if (info->control.hw_key) + if (info->control.hw_key && + !(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIE)) return TX_CONTINUE; if (WARN_ON(skb_tailroom(skb) < sizeof(*mmie))) @@ -911,6 +912,9 @@ ieee80211_crypto_aes_cmac_256_encrypt(struct ieee80211_tx_data *tx) bip_ipn_set64(mmie->sequence_number, pn64); + if (info->control.hw_key) + return TX_CONTINUE; + bip_aad(skb, aad); /* MIC = AES-256-CMAC(IGTK, AAD || Management Frame Body || MMIE, 128) @@ -1040,7 +1044,8 @@ ieee80211_crypto_aes_gmac_encrypt(struct ieee80211_tx_data *tx) info = IEEE80211_SKB_CB(skb); - if (info->control.hw_key) + if (info->control.hw_key && + !(key->conf.flags & IEEE80211_KEY_FLAG_GENERATE_MMIE)) return TX_CONTINUE; if (WARN_ON(skb_tailroom(skb) < sizeof(*mmie))) @@ -1056,6 +1061,9 @@ ieee80211_crypto_aes_gmac_encrypt(struct ieee80211_tx_data *tx) bip_ipn_set64(mmie->sequence_number, pn64); + if (info->control.hw_key) + return TX_CONTINUE; + bip_aad(skb, aad); hdr = (struct ieee80211_hdr *)skb->data; -- cgit v1.2.3 From b9e810405880c99baafd550ada7043e86465396e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 7 Apr 2024 09:33:21 +0000 Subject: tcp: propagate tcp_tw_isn via an extra parameter to ->route_req() tcp_v6_init_req() reads TCP_SKB_CB(skb)->tcp_tw_isn to find out if the request socket is created by a SYN hitting a TIMEWAIT socket. This has been buggy for a decade, lets directly pass the information from tcp_conn_request(). This is a preparatory patch to make the following one easier to review. Signed-off-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/net/tcp.h | 3 ++- net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_ipv4.c | 3 ++- net/ipv6/tcp_ipv6.c | 10 ++++++---- net/mptcp/subflow.c | 10 ++++++---- 5 files changed, 17 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 9ab5b37e9d53..fa0ab77acee2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2284,7 +2284,8 @@ struct tcp_request_sock_ops { struct dst_entry *(*route_req)(const struct sock *sk, struct sk_buff *skb, struct flowi *fl, - struct request_sock *req); + struct request_sock *req, + u32 tw_isn); u32 (*init_seq)(const struct sk_buff *skb); u32 (*init_ts_off)(const struct net *net, const struct sk_buff *skb); int (*send_synack)(const struct sock *sk, struct dst_entry *dst, diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 1f28a2561795..48c275e6ef02 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -7160,7 +7160,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, /* Note: tcp_v6_init_req() might override ir_iif for link locals */ inet_rsk(req)->ir_iif = inet_request_bound_dev_if(sk, skb); - dst = af_ops->route_req(sk, skb, &fl, req); + dst = af_ops->route_req(sk, skb, &fl, req, isn); if (!dst) goto drop_and_free; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 52963c3bb8ca..81e2f05c244d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1666,7 +1666,8 @@ static void tcp_v4_init_req(struct request_sock *req, static struct dst_entry *tcp_v4_route_req(const struct sock *sk, struct sk_buff *skb, struct flowi *fl, - struct request_sock *req) + struct request_sock *req, + u32 tw_isn) { tcp_v4_init_req(req, sk, skb); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index cffebaec66f1..5141f7033abd 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -793,7 +793,8 @@ clear_hash_nostart: static void tcp_v6_init_req(struct request_sock *req, const struct sock *sk_listener, - struct sk_buff *skb) + struct sk_buff *skb, + u32 tw_isn) { bool l3_slave = ipv6_l3mdev_skb(TCP_SKB_CB(skb)->header.h6.flags); struct inet_request_sock *ireq = inet_rsk(req); @@ -807,7 +808,7 @@ static void tcp_v6_init_req(struct request_sock *req, ipv6_addr_type(&ireq->ir_v6_rmt_addr) & IPV6_ADDR_LINKLOCAL) ireq->ir_iif = tcp_v6_iif(skb); - if (!TCP_SKB_CB(skb)->tcp_tw_isn && + if (!tw_isn && (ipv6_opt_accepted(sk_listener, skb, &TCP_SKB_CB(skb)->header.h6) || np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo || np->rxopt.bits.rxhlim || @@ -820,9 +821,10 @@ static void tcp_v6_init_req(struct request_sock *req, static struct dst_entry *tcp_v6_route_req(const struct sock *sk, struct sk_buff *skb, struct flowi *fl, - struct request_sock *req) + struct request_sock *req, + u32 tw_isn) { - tcp_v6_init_req(req, sk, skb); + tcp_v6_init_req(req, sk, skb, tw_isn); if (security_inet_conn_request(sk, skb, req)) return NULL; diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 162b218d9858..b94d1dca1094 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -289,7 +289,8 @@ EXPORT_SYMBOL_GPL(mptcp_subflow_init_cookie_req); static struct dst_entry *subflow_v4_route_req(const struct sock *sk, struct sk_buff *skb, struct flowi *fl, - struct request_sock *req) + struct request_sock *req, + u32 tw_isn) { struct dst_entry *dst; int err; @@ -297,7 +298,7 @@ static struct dst_entry *subflow_v4_route_req(const struct sock *sk, tcp_rsk(req)->is_mptcp = 1; subflow_init_req(req, sk); - dst = tcp_request_sock_ipv4_ops.route_req(sk, skb, fl, req); + dst = tcp_request_sock_ipv4_ops.route_req(sk, skb, fl, req, tw_isn); if (!dst) return NULL; @@ -356,7 +357,8 @@ static int subflow_v6_send_synack(const struct sock *sk, struct dst_entry *dst, static struct dst_entry *subflow_v6_route_req(const struct sock *sk, struct sk_buff *skb, struct flowi *fl, - struct request_sock *req) + struct request_sock *req, + u32 tw_isn) { struct dst_entry *dst; int err; @@ -364,7 +366,7 @@ static struct dst_entry *subflow_v6_route_req(const struct sock *sk, tcp_rsk(req)->is_mptcp = 1; subflow_init_req(req, sk); - dst = tcp_request_sock_ipv6_ops.route_req(sk, skb, fl, req); + dst = tcp_request_sock_ipv6_ops.route_req(sk, skb, fl, req, tw_isn); if (!dst) return NULL; -- cgit v1.2.3 From 41eecbd712b73f0d5dcf1152b9a1c27b1f238028 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 7 Apr 2024 09:33:22 +0000 Subject: tcp: replace TCP_SKB_CB(skb)->tcp_tw_isn with a per-cpu field TCP can transform a TIMEWAIT socket into a SYN_RECV one from a SYN packet, and the ISN of the SYNACK packet is normally generated using TIMEWAIT tw_snd_nxt : tcp_timewait_state_process() ... u32 isn = tcptw->tw_snd_nxt + 65535 + 2; if (isn == 0) isn++; TCP_SKB_CB(skb)->tcp_tw_isn = isn; return TCP_TW_SYN; This SYN packet also bypasses normal checks against listen queue being full or not. tcp_conn_request() ... __u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn; ... /* TW buckets are converted to open requests without * limitations, they conserve resources and peer is * evidently real one. */ if ((syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) && !isn) { want_cookie = tcp_syn_flood_action(sk, rsk_ops->slab_name); if (!want_cookie) goto drop; } This was using TCP_SKB_CB(skb)->tcp_tw_isn field in skb. Unfortunately this field has been accidentally cleared after the call to tcp_timewait_state_process() returning TCP_TW_SYN. Using a field in TCP_SKB_CB(skb) for a temporary state is overkill. Switch instead to a per-cpu variable. As a bonus, we do not have to clear tcp_tw_isn in TCP receive fast path. It is temporarily set then cleared only in the TCP_TW_SYN dance. Fixes: 4ad19de8774e ("net: tcp6: fix double call of tcp_v6_fill_cb()") Fixes: eeea10b83a13 ("tcp: add tcp_v4_fill_cb()/tcp_v4_restore_cb()") Signed-off-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/net/tcp.h | 10 +++++----- net/ipv4/tcp.c | 3 +++ net/ipv4/tcp_input.c | 26 ++++++++++++++++---------- net/ipv4/tcp_ipv4.c | 5 +++-- net/ipv4/tcp_minisocks.c | 4 ++-- net/ipv6/tcp_ipv6.c | 5 +++-- 6 files changed, 32 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index fa0ab77acee2..ba6c5ae86e22 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -52,6 +52,8 @@ extern struct inet_hashinfo tcp_hashinfo; DECLARE_PER_CPU(unsigned int, tcp_orphan_count); int tcp_orphan_count_sum(void); +DECLARE_PER_CPU(u32, tcp_tw_isn); + void tcp_time_wait(struct sock *sk, int state, int timeo); #define MAX_TCP_HEADER L1_CACHE_ALIGN(128 + MAX_HEADER) @@ -392,7 +394,8 @@ enum tcp_tw_status { enum tcp_tw_status tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb, - const struct tcphdr *th); + const struct tcphdr *th, + u32 *tw_isn); struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, struct request_sock *req, bool fastopen, bool *lost_race); @@ -935,13 +938,10 @@ struct tcp_skb_cb { __u32 seq; /* Starting sequence number */ __u32 end_seq; /* SEQ + FIN + SYN + datalen */ union { - /* Note : tcp_tw_isn is used in input path only - * (isn chosen by tcp_timewait_state_process()) - * + /* Note : * tcp_gso_segs/size are used in write queue only, * cf tcp_skb_pcount()/tcp_skb_mss() */ - __u32 tcp_tw_isn; struct { u16 tcp_gso_segs; u16 tcp_gso_size; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 664c8ecb076b..b07aa71b24ec 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -290,6 +290,9 @@ enum { DEFINE_PER_CPU(unsigned int, tcp_orphan_count); EXPORT_PER_CPU_SYMBOL_GPL(tcp_orphan_count); +DEFINE_PER_CPU(u32, tcp_tw_isn); +EXPORT_PER_CPU_SYMBOL_GPL(tcp_tw_isn); + long sysctl_tcp_mem[3] __read_mostly; EXPORT_SYMBOL(sysctl_tcp_mem); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 48c275e6ef02..5a45a0923a1f 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -7097,7 +7097,6 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, struct sock *sk, struct sk_buff *skb) { struct tcp_fastopen_cookie foc = { .len = -1 }; - __u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn; struct tcp_options_received tmp_opt; struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); @@ -7107,21 +7106,28 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, struct dst_entry *dst; struct flowi fl; u8 syncookies; + u32 isn; #ifdef CONFIG_TCP_AO const struct tcp_ao_hdr *aoh; #endif - syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies); + isn = __this_cpu_read(tcp_tw_isn); + if (isn) { + /* TW buckets are converted to open requests without + * limitations, they conserve resources and peer is + * evidently real one. + */ + __this_cpu_write(tcp_tw_isn, 0); + } else { + syncookies = READ_ONCE(net->ipv4.sysctl_tcp_syncookies); - /* TW buckets are converted to open requests without - * limitations, they conserve resources and peer is - * evidently real one. - */ - if ((syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) && !isn) { - want_cookie = tcp_syn_flood_action(sk, rsk_ops->slab_name); - if (!want_cookie) - goto drop; + if (syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) { + want_cookie = tcp_syn_flood_action(sk, + rsk_ops->slab_name); + if (!want_cookie) + goto drop; + } } if (sk_acceptq_is_full(sk)) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 81e2f05c244d..1e650ec71d2f 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2146,7 +2146,6 @@ static void tcp_v4_fill_cb(struct sk_buff *skb, const struct iphdr *iph, skb->len - th->doff * 4); TCP_SKB_CB(skb)->ack_seq = ntohl(th->ack_seq); TCP_SKB_CB(skb)->tcp_flags = tcp_flag_byte(th); - TCP_SKB_CB(skb)->tcp_tw_isn = 0; TCP_SKB_CB(skb)->ip_dsfield = ipv4_get_dsfield(iph); TCP_SKB_CB(skb)->sacked = 0; TCP_SKB_CB(skb)->has_rxtstamp = @@ -2168,6 +2167,7 @@ int tcp_v4_rcv(struct sk_buff *skb) bool refcounted; struct sock *sk; int ret; + u32 isn; drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; if (skb->pkt_type != PACKET_HOST) @@ -2383,7 +2383,7 @@ do_time_wait: inet_twsk_put(inet_twsk(sk)); goto csum_error; } - switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) { + switch (tcp_timewait_state_process(inet_twsk(sk), skb, th, &isn)) { case TCP_TW_SYN: { struct sock *sk2 = inet_lookup_listener(net, net->ipv4.tcp_death_row.hashinfo, @@ -2397,6 +2397,7 @@ do_time_wait: sk = sk2; tcp_v4_restore_cb(skb); refcounted = false; + __this_cpu_write(tcp_tw_isn, isn); goto process; } } diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 5b21a07ddf9a..f53c7ada2ace 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -95,7 +95,7 @@ static void twsk_rcv_nxt_update(struct tcp_timewait_sock *tcptw, u32 seq) */ enum tcp_tw_status tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb, - const struct tcphdr *th) + const struct tcphdr *th, u32 *tw_isn) { struct tcp_options_received tmp_opt; struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw); @@ -228,7 +228,7 @@ kill: u32 isn = tcptw->tw_snd_nxt + 65535 + 2; if (isn == 0) isn++; - TCP_SKB_CB(skb)->tcp_tw_isn = isn; + *tw_isn = isn; return TCP_TW_SYN; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 5141f7033abd..3aa9da5c9a66 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1741,7 +1741,6 @@ static void tcp_v6_fill_cb(struct sk_buff *skb, const struct ipv6hdr *hdr, skb->len - th->doff*4); TCP_SKB_CB(skb)->ack_seq = ntohl(th->ack_seq); TCP_SKB_CB(skb)->tcp_flags = tcp_flag_byte(th); - TCP_SKB_CB(skb)->tcp_tw_isn = 0; TCP_SKB_CB(skb)->ip_dsfield = ipv6_get_dsfield(hdr); TCP_SKB_CB(skb)->sacked = 0; TCP_SKB_CB(skb)->has_rxtstamp = @@ -1758,6 +1757,7 @@ INDIRECT_CALLABLE_SCOPE int tcp_v6_rcv(struct sk_buff *skb) bool refcounted; struct sock *sk; int ret; + u32 isn; struct net *net = dev_net(skb->dev); drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; @@ -1967,7 +1967,7 @@ do_time_wait: goto csum_error; } - switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) { + switch (tcp_timewait_state_process(inet_twsk(sk), skb, th, &isn)) { case TCP_TW_SYN: { struct sock *sk2; @@ -1985,6 +1985,7 @@ do_time_wait: sk = sk2; tcp_v6_restore_cb(skb); refcounted = false; + __this_cpu_write(tcp_tw_isn, isn); goto process; } } -- cgit v1.2.3 From e0cd1185900e638d41d9cccb4c259051e05f69e9 Mon Sep 17 00:00:00 2001 From: Baochen Qiang Date: Mon, 8 Apr 2024 17:41:50 +0300 Subject: net: qrtr: support suspend/hibernation MHI devices may not be destroyed during suspend/hibernation, so need to unprepare/prepare MHI channels throughout the transition, this is done by adding suspend/resume callbacks. The suspend callback is called in the late suspend stage, this means MHI channels are still alive at suspend stage, and that makes it possible for an MHI controller driver to communicate with others over those channels at suspend stage. While the resume callback is called in the early resume stage, for a similar reason. Also note that we won't do unprepare/prepare when MHI device is in suspend state because it's pointless if MHI is only meant to go through a suspend/resume transition, instead of a complete power cycle. Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30 Signed-off-by: Baochen Qiang Reviewed-by: Manivannan Sadhasivam Reviewed-by: Jeff Johnson Signed-off-by: Kalle Valo Link: https://msgid.link/20240305021320.3367-3-quic_bqiang@quicinc.com --- net/qrtr/mhi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) (limited to 'net') diff --git a/net/qrtr/mhi.c b/net/qrtr/mhi.c index 9ced13c0627a..69f53625a049 100644 --- a/net/qrtr/mhi.c +++ b/net/qrtr/mhi.c @@ -118,6 +118,51 @@ static const struct mhi_device_id qcom_mhi_qrtr_id_table[] = { }; MODULE_DEVICE_TABLE(mhi, qcom_mhi_qrtr_id_table); +static int __maybe_unused qcom_mhi_qrtr_pm_suspend_late(struct device *dev) +{ + struct mhi_device *mhi_dev = container_of(dev, struct mhi_device, dev); + enum mhi_state state; + + state = mhi_get_mhi_state(mhi_dev->mhi_cntrl); + /* + * If the device is in suspend state, then no need for the + * client driver to unprepare the channels. + */ + if (state == MHI_STATE_M3) + return 0; + + mhi_unprepare_from_transfer(mhi_dev); + + return 0; +} + +static int __maybe_unused qcom_mhi_qrtr_pm_resume_early(struct device *dev) +{ + struct mhi_device *mhi_dev = container_of(dev, struct mhi_device, dev); + enum mhi_state state; + int rc; + + state = mhi_get_mhi_state(mhi_dev->mhi_cntrl); + /* + * If the device is in suspend state, we won't unprepare channels + * in suspend callback, therefore no need to prepare channels when + * resume. + */ + if (state == MHI_STATE_M3) + return 0; + + rc = mhi_prepare_for_transfer_autoqueue(mhi_dev); + if (rc) + dev_err(dev, "failed to prepare for autoqueue transfer %d\n", rc); + + return rc; +} + +static const struct dev_pm_ops qcom_mhi_qrtr_pm_ops = { + SET_LATE_SYSTEM_SLEEP_PM_OPS(qcom_mhi_qrtr_pm_suspend_late, + qcom_mhi_qrtr_pm_resume_early) +}; + static struct mhi_driver qcom_mhi_qrtr_driver = { .probe = qcom_mhi_qrtr_probe, .remove = qcom_mhi_qrtr_remove, @@ -126,6 +171,7 @@ static struct mhi_driver qcom_mhi_qrtr_driver = { .id_table = qcom_mhi_qrtr_id_table, .driver = { .name = "qcom_mhi_qrtr", + .pm = &qcom_mhi_qrtr_pm_ops, }, }; -- cgit v1.2.3 From d034d02de882d0ac600806724f9ffeb52de11e2e Mon Sep 17 00:00:00 2001 From: Kuan-Wei Chiu Date: Tue, 9 Apr 2024 01:47:16 +0800 Subject: net: sched: cake: Optimize the number of function calls and branches in heap construction MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When constructing a heap, heapify operations are required on all non-leaf nodes. Thus, determining the index of the first non-leaf node is crucial. In a heap, the left child's index of node i is 2 * i + 1 and the right child's index is 2 * i + 2. Node CAKE_MAX_TINS * CAKE_QUEUES / 2 has its left and right children at indexes CAKE_MAX_TINS * CAKE_QUEUES + 1 and CAKE_MAX_TINS * CAKE_QUEUES + 2, respectively, which are beyond the heap's range, indicating it as a leaf node. Conversely, node CAKE_MAX_TINS * CAKE_QUEUES / 2 - 1 has a left child at index CAKE_MAX_TINS * CAKE_QUEUES - 1, confirming its non-leaf status. The loop should start from it since it's not a leaf node. By starting the loop from CAKE_MAX_TINS * CAKE_QUEUES / 2 - 1, we minimize function calls and branch condition evaluations. This adjustment theoretically reduces two function calls (one for cake_heapify() and another for cake_heap_get_backlog()) and five branch evaluations (one for iterating all non-leaf nodes, one within cake_heapify()'s while loop, and three more within the while loop with if conditions). Signed-off-by: Kuan-Wei Chiu Acked-by: Toke Høiland-Jørgensen Link: https://lore.kernel.org/r/20240408174716.751069-1-visitorckw@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_cake.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index edee926ccde8..2eabc4dc5b79 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -1512,7 +1512,7 @@ static unsigned int cake_drop(struct Qdisc *sch, struct sk_buff **to_free) if (!q->overflow_timeout) { int i; /* Build fresh max-heap */ - for (i = CAKE_MAX_TINS * CAKE_QUEUES / 2; i >= 0; i--) + for (i = CAKE_MAX_TINS * CAKE_QUEUES / 2 - 1; i >= 0; i--) cake_heapify(q, i); } q->overflow_timeout = 65535; -- cgit v1.2.3 From 959fa5c188bf095558c417554e4772ac1fda3531 Mon Sep 17 00:00:00 2001 From: Mina Almasry Date: Mon, 8 Apr 2024 08:29:56 -0700 Subject: net: make napi_frag_unref reuse skb_page_unref The implementations of these 2 functions are almost identical. Remove the implementation of napi_frag_unref, and make it a call into skb_page_unref so we don't duplicate the implementation. Signed-off-by: Mina Almasry Reviewed-by: Eric Dumazet Reviewed-by: Jacob Keller Link: https://lore.kernel.org/r/20240408153000.2152844-2-almasrymina@google.com Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 12 +++--------- net/ipv4/esp4.c | 2 +- net/ipv6/esp6.c | 2 +- 3 files changed, 5 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7dfb906d92f7..c0ff85bb087a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3522,10 +3522,10 @@ int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb, bool napi_pp_put_page(struct page *page); static inline void -skb_page_unref(const struct sk_buff *skb, struct page *page) +skb_page_unref(struct page *page, bool recycle) { #ifdef CONFIG_PAGE_POOL - if (skb->pp_recycle && napi_pp_put_page(page)) + if (recycle && napi_pp_put_page(page)) return; #endif put_page(page); @@ -3534,13 +3534,7 @@ skb_page_unref(const struct sk_buff *skb, struct page *page) static inline void napi_frag_unref(skb_frag_t *frag, bool recycle) { - struct page *page = skb_frag_page(frag); - -#ifdef CONFIG_PAGE_POOL - if (recycle && napi_pp_put_page(page)) - return; -#endif - put_page(page); + skb_page_unref(skb_frag_page(frag), recycle); } /** diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 3d647c9a7a21..40330253f076 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -114,7 +114,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) */ if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) - skb_page_unref(skb, sg_page(sg)); + skb_page_unref(sg_page(sg), skb->pp_recycle); } #ifdef CONFIG_INET_ESPINTCP diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index fe8d53f5a5ee..fb431d0a3475 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -131,7 +131,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) */ if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) - skb_page_unref(skb, sg_page(sg)); + skb_page_unref(sg_page(sg), skb->pp_recycle); } #ifdef CONFIG_INET6_ESPINTCP -- cgit v1.2.3 From f58f3c9563409e618e591d0d540316286cb0665f Mon Sep 17 00:00:00 2001 From: Mina Almasry Date: Mon, 8 Apr 2024 08:29:58 -0700 Subject: net: remove napi_frag_unref With the changes in the last patches, napi_frag_unref() is now reduandant. Remove it and use skb_page_unref directly. Signed-off-by: Mina Almasry Reviewed-by: Dragos Tatulea Reviewed-by: Eric Dumazet Reviewed-by: Jacob Keller Link: https://lore.kernel.org/r/20240408153000.2152844-4-almasrymina@google.com Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 8 +------- net/core/skbuff.c | 2 +- 2 files changed, 2 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index c0ff85bb087a..7135a3e94afd 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3531,12 +3531,6 @@ skb_page_unref(struct page *page, bool recycle) put_page(page); } -static inline void -napi_frag_unref(skb_frag_t *frag, bool recycle) -{ - skb_page_unref(skb_frag_page(frag), recycle); -} - /** * __skb_frag_unref - release a reference on a paged fragment. * @frag: the paged fragment @@ -3547,7 +3541,7 @@ napi_frag_unref(skb_frag_t *frag, bool recycle) */ static inline void __skb_frag_unref(skb_frag_t *frag, bool recycle) { - napi_frag_unref(frag, recycle); + skb_page_unref(skb_frag_page(frag), recycle); } /** diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 21cd01641f4c..888874ef8566 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1101,7 +1101,7 @@ static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason) } for (i = 0; i < shinfo->nr_frags; i++) - napi_frag_unref(&shinfo->frags[i], skb->pp_recycle); + __skb_frag_unref(&shinfo->frags[i], skb->pp_recycle); free_head: if (shinfo->frag_list) -- cgit v1.2.3 From 2ecd487b670fcbb1ad4893fff1af4aafdecb6023 Mon Sep 17 00:00:00 2001 From: Jianbo Liu Date: Mon, 8 Apr 2024 16:48:17 +0300 Subject: net: sched: cls_api: fix slab-use-after-free in fl_dump_key MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The filter counter is updated under the protection of cb_lock in the cited commit. While waiting for the lock, it's possible the filter is being deleted by other thread, and thus causes UAF when dump it. Fix this issue by moving tcf_block_filter_cnt_update() after tfilter_put(). ================================================================== BUG: KASAN: slab-use-after-free in fl_dump_key+0x1d3e/0x20d0 [cls_flower] Read of size 4 at addr ffff88814f864000 by task tc/2973 CPU: 7 PID: 2973 Comm: tc Not tainted 6.9.0-rc2_for_upstream_debug_2024_04_02_12_41 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: dump_stack_lvl+0x7e/0xc0 print_report+0xc1/0x600 ? __virt_addr_valid+0x1cf/0x390 ? fl_dump_key+0x1d3e/0x20d0 [cls_flower] ? fl_dump_key+0x1d3e/0x20d0 [cls_flower] kasan_report+0xb9/0xf0 ? fl_dump_key+0x1d3e/0x20d0 [cls_flower] fl_dump_key+0x1d3e/0x20d0 [cls_flower] ? lock_acquire+0x1c2/0x530 ? fl_dump+0x172/0x5c0 [cls_flower] ? lockdep_hardirqs_on_prepare+0x400/0x400 ? fl_dump_key_options.part.0+0x10f0/0x10f0 [cls_flower] ? do_raw_spin_lock+0x12d/0x270 ? spin_bug+0x1d0/0x1d0 fl_dump+0x21d/0x5c0 [cls_flower] ? fl_tmplt_dump+0x1f0/0x1f0 [cls_flower] ? nla_put+0x15f/0x1c0 tcf_fill_node+0x51b/0x9a0 ? tc_skb_ext_tc_enable+0x150/0x150 ? __alloc_skb+0x17b/0x310 ? __build_skb_around+0x340/0x340 ? down_write+0x1b0/0x1e0 tfilter_notify+0x1a5/0x390 ? fl_terse_dump+0x400/0x400 [cls_flower] tc_new_tfilter+0x963/0x2170 ? tc_del_tfilter+0x1490/0x1490 ? print_usage_bug.part.0+0x670/0x670 ? lock_downgrade+0x680/0x680 ? security_capable+0x51/0x90 ? tc_del_tfilter+0x1490/0x1490 rtnetlink_rcv_msg+0x75e/0xac0 ? if_nlmsg_stats_size+0x4c0/0x4c0 ? lockdep_set_lock_cmp_fn+0x190/0x190 ? __netlink_lookup+0x35e/0x6e0 netlink_rcv_skb+0x12c/0x360 ? if_nlmsg_stats_size+0x4c0/0x4c0 ? netlink_ack+0x15e0/0x15e0 ? lockdep_hardirqs_on_prepare+0x400/0x400 ? netlink_deliver_tap+0xcd/0xa60 ? netlink_deliver_tap+0xcd/0xa60 ? netlink_deliver_tap+0x1c9/0xa60 netlink_unicast+0x43e/0x700 ? netlink_attachskb+0x750/0x750 ? lock_acquire+0x1c2/0x530 ? __might_fault+0xbb/0x170 netlink_sendmsg+0x749/0xc10 ? netlink_unicast+0x700/0x700 ? __might_fault+0xbb/0x170 ? netlink_unicast+0x700/0x700 __sock_sendmsg+0xc5/0x190 ____sys_sendmsg+0x534/0x6b0 ? import_iovec+0x7/0x10 ? kernel_sendmsg+0x30/0x30 ? __copy_msghdr+0x3c0/0x3c0 ? entry_SYSCALL_64_after_hwframe+0x46/0x4e ? lock_acquire+0x1c2/0x530 ? __virt_addr_valid+0x116/0x390 ___sys_sendmsg+0xeb/0x170 ? __virt_addr_valid+0x1ca/0x390 ? copy_msghdr_from_user+0x110/0x110 ? __delete_object+0xb8/0x100 ? __virt_addr_valid+0x1cf/0x390 ? do_sys_openat2+0x102/0x150 ? lockdep_hardirqs_on_prepare+0x284/0x400 ? do_sys_openat2+0x102/0x150 ? __fget_light+0x53/0x1d0 ? sockfd_lookup_light+0x1a/0x150 __sys_sendmsg+0xb5/0x140 ? __sys_sendmsg_sock+0x20/0x20 ? lock_downgrade+0x680/0x680 do_syscall_64+0x70/0x140 entry_SYSCALL_64_after_hwframe+0x46/0x4e RIP: 0033:0x7f98e3713367 Code: 0e 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 RSP: 002b:00007ffc74a64608 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 000000000047eae0 RCX: 00007f98e3713367 RDX: 0000000000000000 RSI: 00007ffc74a64670 RDI: 0000000000000003 RBP: 0000000000000008 R08: 0000000000000000 R09: 0000000000000000 R10: 00007f98e360c5e8 R11: 0000000000000246 R12: 00007ffc74a6a508 R13: 00000000660d518d R14: 0000000000484a80 R15: 00007ffc74a6a50b Allocated by task 2973: kasan_save_stack+0x20/0x40 kasan_save_track+0x10/0x30 __kasan_kmalloc+0x77/0x90 fl_change+0x27a6/0x4540 [cls_flower] tc_new_tfilter+0x879/0x2170 rtnetlink_rcv_msg+0x75e/0xac0 netlink_rcv_skb+0x12c/0x360 netlink_unicast+0x43e/0x700 netlink_sendmsg+0x749/0xc10 __sock_sendmsg+0xc5/0x190 ____sys_sendmsg+0x534/0x6b0 ___sys_sendmsg+0xeb/0x170 __sys_sendmsg+0xb5/0x140 do_syscall_64+0x70/0x140 entry_SYSCALL_64_after_hwframe+0x46/0x4e Freed by task 283: kasan_save_stack+0x20/0x40 kasan_save_track+0x10/0x30 kasan_save_free_info+0x37/0x50 poison_slab_object+0x105/0x190 __kasan_slab_free+0x11/0x30 kfree+0x111/0x340 process_one_work+0x787/0x1490 worker_thread+0x586/0xd30 kthread+0x2df/0x3b0 ret_from_fork+0x2d/0x70 ret_from_fork_asm+0x11/0x20 Last potentially related work creation: kasan_save_stack+0x20/0x40 __kasan_record_aux_stack+0x9b/0xb0 insert_work+0x25/0x1b0 __queue_work+0x640/0xc90 rcu_work_rcufn+0x42/0x70 rcu_core+0x6a9/0x1850 __do_softirq+0x264/0x88f Second to last potentially related work creation: kasan_save_stack+0x20/0x40 __kasan_record_aux_stack+0x9b/0xb0 __call_rcu_common.constprop.0+0x6f/0xac0 queue_rcu_work+0x56/0x70 fl_mask_put+0x20d/0x270 [cls_flower] __fl_delete+0x352/0x6b0 [cls_flower] fl_delete+0x97/0x160 [cls_flower] tc_del_tfilter+0x7d1/0x1490 rtnetlink_rcv_msg+0x75e/0xac0 netlink_rcv_skb+0x12c/0x360 netlink_unicast+0x43e/0x700 netlink_sendmsg+0x749/0xc10 __sock_sendmsg+0xc5/0x190 ____sys_sendmsg+0x534/0x6b0 ___sys_sendmsg+0xeb/0x170 __sys_sendmsg+0xb5/0x140 do_syscall_64+0x70/0x140 entry_SYSCALL_64_after_hwframe+0x46/0x4e Fixes: 2081fd3445fe ("net: sched: cls_api: add filter counter") Signed-off-by: Jianbo Liu Reviewed-by: Cosmin Ratiu Tested-by: Asbjørn Sloth Tønnesen Signed-off-by: David S. Miller --- net/sched/cls_api.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index db0653993632..17d97bbe890f 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -2400,10 +2400,10 @@ replay: err = tp->ops->change(net, skb, tp, cl, t->tcm_handle, tca, &fh, flags, extack); if (err == 0) { - tcf_block_filter_cnt_update(block, &tp->counted, true); tfilter_notify(net, skb, n, tp, block, q, parent, fh, RTM_NEWTFILTER, false, rtnl_held, extack); tfilter_put(tp, fh); + tcf_block_filter_cnt_update(block, &tp->counted, true); /* q pointer is NULL for shared blocks */ if (q) q->flags &= ~TCQ_F_CAN_BYPASS; -- cgit v1.2.3 From 9b9fd45869e744bea7d32a94793736e3d9fd7d26 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 9 Apr 2024 14:09:14 +0000 Subject: tcp: tweak tcp_sock_write_txrx size assertion I forgot 32bit arches might have 64bit alignment for u64 fields. tcp_sock_write_txrx group does not contain pointers, but two u64 fields. It is possible that on 32bit kernel, a 32bit hole is before tp->tcp_clock_cache. I will try to remember a group can be bigger on 32bit kernels in the future. With help from Vladimir Oltean. Fixes: d2c3a7eb1afa ("tcp: more struct tcp_sock adjustments") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202404082207.HCEdQhUO-lkp@intel.com/ Signed-off-by: Eric Dumazet Tested-by: Vladimir Oltean Link: https://lore.kernel.org/r/20240409140914.4105429-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b07aa71b24ec..e1bf468e0d22 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -4673,7 +4673,11 @@ static void __init tcp_struct_check(void) CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, app_limited); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rcv_wnd); CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rx_opt); - CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 92); + + /* 32bit arches with 8byte alignment on u64 fields might need padding + * before tcp_clock_cache. + */ + CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 92 + 4); /* RX read-write hotpath cache lines */ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_received); -- cgit v1.2.3 From 7cb31c46b9cc37a5e564667fe46daf9a35dbafdd Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 10 Apr 2024 02:28:09 +0100 Subject: net: cache for same cpu skb_attempt_defer_free Optimise skb_attempt_defer_free() when run by the same CPU the skb was allocated on. Instead of __kfree_skb() -> kmem_cache_free() we can disable softirqs and put the buffer into cpu local caches. CPU bound TCP ping pong style benchmarking (i.e. netbench) showed a 1% throughput increase (392.2 -> 396.4 Krps). Cross checking with profiles, the total CPU share of skb_attempt_defer_free() dropped by 0.6%. Note, I'd expect the win doubled with rx only benchmarks, as the optimisation is for the receive path, but the test spends >55% of CPU doing writes. Signed-off-by: Pavel Begunkov Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/a887463fb219d973ec5ad275e31194812571f1f5.1712711977.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 888874ef8566..18612f255ef5 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6974,6 +6974,19 @@ free_now: EXPORT_SYMBOL(__skb_ext_put); #endif /* CONFIG_SKB_EXTENSIONS */ +static void kfree_skb_napi_cache(struct sk_buff *skb) +{ + /* if SKB is a clone, don't handle this case */ + if (skb->fclone != SKB_FCLONE_UNAVAILABLE) { + __kfree_skb(skb); + return; + } + + local_bh_disable(); + __napi_kfree_skb(skb, SKB_DROP_REASON_NOT_SPECIFIED); + local_bh_enable(); +} + /** * skb_attempt_defer_free - queue skb for remote freeing * @skb: buffer @@ -6992,7 +7005,7 @@ void skb_attempt_defer_free(struct sk_buff *skb) if (WARN_ON_ONCE(cpu >= nr_cpu_ids) || !cpu_online(cpu) || cpu == raw_smp_processor_id()) { -nodefer: __kfree_skb(skb); +nodefer: kfree_skb_napi_cache(skb); return; } -- cgit v1.2.3 From d8415a165c43905ccf9ad7a4c7eaf6006b91a00b Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Wed, 10 Apr 2024 02:28:10 +0100 Subject: net: use SKB_CONSUMED in skb_attempt_defer_free() skb_attempt_defer_free() is used to free already processed skbs, so pass SKB_CONSUMED as the reason in kfree_skb_napi_cache(). Suggested-by: Jason Xing Suggested-by: Eric Dumazet Signed-off-by: Pavel Begunkov Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/bcf5dbdda79688b074ab7ae2238535840a6d3fc2.1712711977.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 18612f255ef5..ab970ded8a7b 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6983,7 +6983,7 @@ static void kfree_skb_napi_cache(struct sk_buff *skb) } local_bh_disable(); - __napi_kfree_skb(skb, SKB_DROP_REASON_NOT_SPECIFIED); + __napi_kfree_skb(skb, SKB_CONSUMED); local_bh_enable(); } -- cgit v1.2.3 From 699c23f02c65cbfc3e638f14ce0d70c23a2e1f02 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Tue, 9 Apr 2024 21:35:27 -0700 Subject: bpf: Add bpf_link support for sk_msg and sk_skb progs Add bpf_link support for sk_msg and sk_skb programs. We have an internal request to support bpf_link for sk_msg programs so user space can have a uniform handling with bpf_link based libbpf APIs. Using bpf_link based libbpf API also has a benefit which makes system robust by decoupling prog life cycle and attachment life cycle. Reviewed-by: John Fastabend Signed-off-by: Yonghong Song Link: https://lore.kernel.org/r/20240410043527.3737160-1-yonghong.song@linux.dev Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 6 + include/linux/skmsg.h | 4 + include/uapi/linux/bpf.h | 5 + kernel/bpf/syscall.c | 4 + net/core/sock_map.c | 263 ++++++++++++++++++++++++++++++++++++++--- tools/include/uapi/linux/bpf.h | 5 + 6 files changed, 271 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 62762390c93d..5034c1b4ded7 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -2996,6 +2996,7 @@ int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype); int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags); int sock_map_bpf_prog_query(const union bpf_attr *attr, union bpf_attr __user *uattr); +int sock_map_link_create(const union bpf_attr *attr, struct bpf_prog *prog); void sock_map_unhash(struct sock *sk); void sock_map_destroy(struct sock *sk); @@ -3094,6 +3095,11 @@ static inline int sock_map_bpf_prog_query(const union bpf_attr *attr, { return -EINVAL; } + +static inline int sock_map_link_create(const union bpf_attr *attr, struct bpf_prog *prog) +{ + return -EOPNOTSUPP; +} #endif /* CONFIG_BPF_SYSCALL */ #endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */ diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index e65ec3fd2799..9c8dd4c01412 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -58,6 +58,10 @@ struct sk_psock_progs { struct bpf_prog *stream_parser; struct bpf_prog *stream_verdict; struct bpf_prog *skb_verdict; + struct bpf_link *msg_parser_link; + struct bpf_link *stream_parser_link; + struct bpf_link *stream_verdict_link; + struct bpf_link *skb_verdict_link; }; enum sk_psock_state_bits { diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 6fe9f11c8abe..cee0a7915c08 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -1135,6 +1135,7 @@ enum bpf_link_type { BPF_LINK_TYPE_TCX = 11, BPF_LINK_TYPE_UPROBE_MULTI = 12, BPF_LINK_TYPE_NETKIT = 13, + BPF_LINK_TYPE_SOCKMAP = 14, __MAX_BPF_LINK_TYPE, }; @@ -6724,6 +6725,10 @@ struct bpf_link_info { __u32 ifindex; __u32 attach_type; } netkit; + struct { + __u32 map_id; + __u32 attach_type; + } sockmap; }; } __attribute__((aligned(8))); diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index e44c276e8617..7d392ec83655 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -5213,6 +5213,10 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr) case BPF_PROG_TYPE_SK_LOOKUP: ret = netns_bpf_link_create(attr, prog); break; + case BPF_PROG_TYPE_SK_MSG: + case BPF_PROG_TYPE_SK_SKB: + ret = sock_map_link_create(attr, prog); + break; #ifdef CONFIG_NET case BPF_PROG_TYPE_XDP: ret = bpf_xdp_link_attach(attr, prog); diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 27d733c0f65e..63c016b4c169 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -24,8 +24,16 @@ struct bpf_stab { #define SOCK_CREATE_FLAG_MASK \ (BPF_F_NUMA_NODE | BPF_F_RDONLY | BPF_F_WRONLY) +/* This mutex is used to + * - protect race between prog/link attach/detach and link prog update, and + * - protect race between releasing and accessing map in bpf_link. + * A single global mutex lock is used since it is expected contention is low. + */ +static DEFINE_MUTEX(sockmap_mutex); + static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog, - struct bpf_prog *old, u32 which); + struct bpf_prog *old, struct bpf_link *link, + u32 which); static struct sk_psock_progs *sock_map_progs(struct bpf_map *map); static struct bpf_map *sock_map_alloc(union bpf_attr *attr) @@ -71,7 +79,9 @@ int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog) map = __bpf_map_get(f); if (IS_ERR(map)) return PTR_ERR(map); - ret = sock_map_prog_update(map, prog, NULL, attr->attach_type); + mutex_lock(&sockmap_mutex); + ret = sock_map_prog_update(map, prog, NULL, NULL, attr->attach_type); + mutex_unlock(&sockmap_mutex); fdput(f); return ret; } @@ -103,7 +113,9 @@ int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype) goto put_prog; } - ret = sock_map_prog_update(map, NULL, prog, attr->attach_type); + mutex_lock(&sockmap_mutex); + ret = sock_map_prog_update(map, NULL, prog, NULL, attr->attach_type); + mutex_unlock(&sockmap_mutex); put_prog: bpf_prog_put(prog); put_map: @@ -1454,55 +1466,84 @@ static struct sk_psock_progs *sock_map_progs(struct bpf_map *map) return NULL; } -static int sock_map_prog_lookup(struct bpf_map *map, struct bpf_prog ***pprog, - u32 which) +static int sock_map_prog_link_lookup(struct bpf_map *map, struct bpf_prog ***pprog, + struct bpf_link ***plink, u32 which) { struct sk_psock_progs *progs = sock_map_progs(map); + struct bpf_prog **cur_pprog; + struct bpf_link **cur_plink; if (!progs) return -EOPNOTSUPP; switch (which) { case BPF_SK_MSG_VERDICT: - *pprog = &progs->msg_parser; + cur_pprog = &progs->msg_parser; + cur_plink = &progs->msg_parser_link; break; #if IS_ENABLED(CONFIG_BPF_STREAM_PARSER) case BPF_SK_SKB_STREAM_PARSER: - *pprog = &progs->stream_parser; + cur_pprog = &progs->stream_parser; + cur_plink = &progs->stream_parser_link; break; #endif case BPF_SK_SKB_STREAM_VERDICT: if (progs->skb_verdict) return -EBUSY; - *pprog = &progs->stream_verdict; + cur_pprog = &progs->stream_verdict; + cur_plink = &progs->stream_verdict_link; break; case BPF_SK_SKB_VERDICT: if (progs->stream_verdict) return -EBUSY; - *pprog = &progs->skb_verdict; + cur_pprog = &progs->skb_verdict; + cur_plink = &progs->skb_verdict_link; break; default: return -EOPNOTSUPP; } + *pprog = cur_pprog; + if (plink) + *plink = cur_plink; return 0; } +/* Handle the following four cases: + * prog_attach: prog != NULL, old == NULL, link == NULL + * prog_detach: prog == NULL, old != NULL, link == NULL + * link_attach: prog != NULL, old == NULL, link != NULL + * link_detach: prog == NULL, old != NULL, link != NULL + */ static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog, - struct bpf_prog *old, u32 which) + struct bpf_prog *old, struct bpf_link *link, + u32 which) { struct bpf_prog **pprog; + struct bpf_link **plink; int ret; - ret = sock_map_prog_lookup(map, &pprog, which); + ret = sock_map_prog_link_lookup(map, &pprog, &plink, which); if (ret) return ret; - if (old) - return psock_replace_prog(pprog, prog, old); + /* for prog_attach/prog_detach/link_attach, return error if a bpf_link + * exists for that prog. + */ + if ((!link || prog) && *plink) + return -EBUSY; - psock_set_prog(pprog, prog); - return 0; + if (old) { + ret = psock_replace_prog(pprog, prog, old); + if (!ret) + *plink = NULL; + } else { + psock_set_prog(pprog, prog); + if (link) + *plink = link; + } + + return ret; } int sock_map_bpf_prog_query(const union bpf_attr *attr, @@ -1527,7 +1568,7 @@ int sock_map_bpf_prog_query(const union bpf_attr *attr, rcu_read_lock(); - ret = sock_map_prog_lookup(map, &pprog, attr->query.attach_type); + ret = sock_map_prog_link_lookup(map, &pprog, NULL, attr->query.attach_type); if (ret) goto end; @@ -1657,6 +1698,196 @@ void sock_map_close(struct sock *sk, long timeout) } EXPORT_SYMBOL_GPL(sock_map_close); +struct sockmap_link { + struct bpf_link link; + struct bpf_map *map; + enum bpf_attach_type attach_type; +}; + +static void sock_map_link_release(struct bpf_link *link) +{ + struct sockmap_link *sockmap_link = container_of(link, struct sockmap_link, link); + + mutex_lock(&sockmap_mutex); + if (!sockmap_link->map) + goto out; + + WARN_ON_ONCE(sock_map_prog_update(sockmap_link->map, NULL, link->prog, link, + sockmap_link->attach_type)); + + bpf_map_put_with_uref(sockmap_link->map); + sockmap_link->map = NULL; +out: + mutex_unlock(&sockmap_mutex); +} + +static int sock_map_link_detach(struct bpf_link *link) +{ + sock_map_link_release(link); + return 0; +} + +static void sock_map_link_dealloc(struct bpf_link *link) +{ + kfree(link); +} + +/* Handle the following two cases: + * case 1: link != NULL, prog != NULL, old != NULL + * case 2: link != NULL, prog != NULL, old == NULL + */ +static int sock_map_link_update_prog(struct bpf_link *link, + struct bpf_prog *prog, + struct bpf_prog *old) +{ + const struct sockmap_link *sockmap_link = container_of(link, struct sockmap_link, link); + struct bpf_prog **pprog, *old_link_prog; + struct bpf_link **plink; + int ret = 0; + + mutex_lock(&sockmap_mutex); + + /* If old prog is not NULL, ensure old prog is the same as link->prog. */ + if (old && link->prog != old) { + ret = -EPERM; + goto out; + } + /* Ensure link->prog has the same type/attach_type as the new prog. */ + if (link->prog->type != prog->type || + link->prog->expected_attach_type != prog->expected_attach_type) { + ret = -EINVAL; + goto out; + } + + ret = sock_map_prog_link_lookup(sockmap_link->map, &pprog, &plink, + sockmap_link->attach_type); + if (ret) + goto out; + + /* return error if the stored bpf_link does not match the incoming bpf_link. */ + if (link != *plink) { + ret = -EBUSY; + goto out; + } + + if (old) { + ret = psock_replace_prog(pprog, prog, old); + if (ret) + goto out; + } else { + psock_set_prog(pprog, prog); + } + + bpf_prog_inc(prog); + old_link_prog = xchg(&link->prog, prog); + bpf_prog_put(old_link_prog); + +out: + mutex_unlock(&sockmap_mutex); + return ret; +} + +static u32 sock_map_link_get_map_id(const struct sockmap_link *sockmap_link) +{ + u32 map_id = 0; + + mutex_lock(&sockmap_mutex); + if (sockmap_link->map) + map_id = sockmap_link->map->id; + mutex_unlock(&sockmap_mutex); + return map_id; +} + +static int sock_map_link_fill_info(const struct bpf_link *link, + struct bpf_link_info *info) +{ + const struct sockmap_link *sockmap_link = container_of(link, struct sockmap_link, link); + u32 map_id = sock_map_link_get_map_id(sockmap_link); + + info->sockmap.map_id = map_id; + info->sockmap.attach_type = sockmap_link->attach_type; + return 0; +} + +static void sock_map_link_show_fdinfo(const struct bpf_link *link, + struct seq_file *seq) +{ + const struct sockmap_link *sockmap_link = container_of(link, struct sockmap_link, link); + u32 map_id = sock_map_link_get_map_id(sockmap_link); + + seq_printf(seq, "map_id:\t%u\n", map_id); + seq_printf(seq, "attach_type:\t%u\n", sockmap_link->attach_type); +} + +static const struct bpf_link_ops sock_map_link_ops = { + .release = sock_map_link_release, + .dealloc = sock_map_link_dealloc, + .detach = sock_map_link_detach, + .update_prog = sock_map_link_update_prog, + .fill_link_info = sock_map_link_fill_info, + .show_fdinfo = sock_map_link_show_fdinfo, +}; + +int sock_map_link_create(const union bpf_attr *attr, struct bpf_prog *prog) +{ + struct bpf_link_primer link_primer; + struct sockmap_link *sockmap_link; + enum bpf_attach_type attach_type; + struct bpf_map *map; + int ret; + + if (attr->link_create.flags) + return -EINVAL; + + map = bpf_map_get_with_uref(attr->link_create.target_fd); + if (IS_ERR(map)) + return PTR_ERR(map); + if (map->map_type != BPF_MAP_TYPE_SOCKMAP && map->map_type != BPF_MAP_TYPE_SOCKHASH) { + ret = -EINVAL; + goto out; + } + + sockmap_link = kzalloc(sizeof(*sockmap_link), GFP_USER); + if (!sockmap_link) { + ret = -ENOMEM; + goto out; + } + + attach_type = attr->link_create.attach_type; + bpf_link_init(&sockmap_link->link, BPF_LINK_TYPE_SOCKMAP, &sock_map_link_ops, prog); + sockmap_link->map = map; + sockmap_link->attach_type = attach_type; + + ret = bpf_link_prime(&sockmap_link->link, &link_primer); + if (ret) { + kfree(sockmap_link); + goto out; + } + + mutex_lock(&sockmap_mutex); + ret = sock_map_prog_update(map, prog, NULL, &sockmap_link->link, attach_type); + mutex_unlock(&sockmap_mutex); + if (ret) { + bpf_link_cleanup(&link_primer); + goto out; + } + + /* Increase refcnt for the prog since when old prog is replaced with + * psock_replace_prog() and psock_set_prog() its refcnt will be decreased. + * + * Actually, we do not need to increase refcnt for the prog since bpf_link + * will hold a reference. But in order to have less complexity w.r.t. + * replacing/setting prog, let us increase the refcnt to make things simpler. + */ + bpf_prog_inc(prog); + + return bpf_link_settle(&link_primer); + +out: + bpf_map_put_with_uref(map); + return ret; +} + static int sock_map_iter_attach_target(struct bpf_prog *prog, union bpf_iter_link_info *linfo, struct bpf_iter_aux_info *aux) diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 6fe9f11c8abe..cee0a7915c08 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -1135,6 +1135,7 @@ enum bpf_link_type { BPF_LINK_TYPE_TCX = 11, BPF_LINK_TYPE_UPROBE_MULTI = 12, BPF_LINK_TYPE_NETKIT = 13, + BPF_LINK_TYPE_SOCKMAP = 14, __MAX_BPF_LINK_TYPE, }; @@ -6724,6 +6725,10 @@ struct bpf_link_info { __u32 ifindex; __u32 attach_type; } netkit; + struct { + __u32 map_id; + __u32 attach_type; + } sockmap; }; } __attribute__((aligned(8))); -- cgit v1.2.3 From 8b06a24bb625728ac709f2c69405eb01025687e1 Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Wed, 10 Apr 2024 19:27:12 +0200 Subject: xfrm: fix possible derferencing in error path Fix derferencing pointer when xfrm_policy_lookup_bytype returns an error. Fixes: 63b21caba17e ("xfrm: introduce forwarding of ICMP Error messages") Reported-by: Dan Carpenter Closes: https://lore.kernel.org/kernel-janitors/f6ef0d0d-96de-4e01-9dc3-c1b3a6338653@moroto.mountain/ Signed-off-by: Antony Antony Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 6affe5cd85d8..53d8fabfa685 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -3593,6 +3593,8 @@ xfrm_policy *xfrm_in_fwd_icmp(struct sk_buff *skb, return pol; pol = xfrm_policy_lookup(net, &fl1, family, XFRM_POLICY_FWD, if_id); + if (IS_ERR(pol)) + pol = NULL; } return pol; -- cgit v1.2.3 From 18d82cde743237def9f80171e32be1153f5cadac Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Wed, 10 Apr 2024 11:48:24 +0200 Subject: mptcp: add last time fields in mptcp_info This patch adds "last time" fields last_data_sent, last_data_recv and last_ack_recv in struct mptcp_sock to record the last time data_sent, data_recv and ack_recv happened. They all are initialized as tcp_jiffies32 in __mptcp_init_sock(), and updated as tcp_jiffies32 too when data is sent in __subflow_push_pending(), data is received in __mptcp_move_skbs_from_subflow(), and ack is received in ack_update_msk(). Similar to tcpi_last_data_sent, tcpi_last_data_recv and tcpi_last_ack_recv exposed with TCP, this patch exposes the last time "an action happened" for MPTCP in mptcp_info, named mptcpi_last_data_sent, mptcpi_last_data_recv and mptcpi_last_ack_recv, calculated in mptcp_diag_fill_info() as the time deltas between now and the newly added last time fields in mptcp_sock. Since msk->last_ack_recv needs to be protected by mptcp_data_lock/unlock, and lock_sock_fast can sleep and be quite slow, move the entire mptcp_data_lock/unlock block after the lock/unlock_sock_fast block. Then mptcpi_last_data_sent and mptcpi_last_data_recv are set in lock/unlock_sock_fast block, while mptcpi_last_ack_recv is set in mptcp_data_lock/unlock block, which is protected by a spinlock and should not block for too long. Also add three reserved bytes in struct mptcp_info not to have holes in this structure exposed to userspace. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/446 Signed-off-by: Geliang Tang Reviewed-by: Mat Martineau Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240410-upstream-net-next-20240405-mptcp-last-time-info-v2-1-f95bd6b33e51@kernel.org Signed-off-by: Jakub Kicinski --- include/uapi/linux/mptcp.h | 4 ++++ net/mptcp/options.c | 1 + net/mptcp/protocol.c | 7 +++++++ net/mptcp/protocol.h | 3 +++ net/mptcp/sockopt.c | 16 +++++++++++----- 5 files changed, 26 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/mptcp.h b/include/uapi/linux/mptcp.h index 74cfe496891e..67d015df8893 100644 --- a/include/uapi/linux/mptcp.h +++ b/include/uapi/linux/mptcp.h @@ -58,6 +58,10 @@ struct mptcp_info { __u64 mptcpi_bytes_received; __u64 mptcpi_bytes_acked; __u8 mptcpi_subflows_total; + __u8 reserved[3]; + __u32 mptcpi_last_data_sent; + __u32 mptcpi_last_data_recv; + __u32 mptcpi_last_ack_recv; }; /* MPTCP Reset reason codes, rfc8684 */ diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 27ca42c77b02..8e8dcfbc2993 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -1068,6 +1068,7 @@ static void ack_update_msk(struct mptcp_sock *msk, __mptcp_snd_una_update(msk, new_snd_una); __mptcp_data_acked(sk); } + msk->last_ack_recv = tcp_jiffies32; mptcp_data_unlock(sk); trace_ack_update_msk(mp_opt->data_ack, diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 995b53cd021c..f8bc34f0d973 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -706,6 +706,8 @@ static bool __mptcp_move_skbs_from_subflow(struct mptcp_sock *msk, } } while (more_data_avail); + if (moved > 0) + msk->last_data_recv = tcp_jiffies32; *bytes += moved; return done; } @@ -1556,6 +1558,8 @@ static int __subflow_push_pending(struct sock *sk, struct sock *ssk, err = copied; out: + if (err > 0) + msk->last_data_sent = tcp_jiffies32; return err; } @@ -2793,6 +2797,9 @@ static void __mptcp_init_sock(struct sock *sk) WRITE_ONCE(msk->allow_infinite_fallback, true); msk->recovery = false; msk->subflow_id = 1; + msk->last_data_sent = tcp_jiffies32; + msk->last_data_recv = tcp_jiffies32; + msk->last_ack_recv = tcp_jiffies32; mptcp_pm_data_init(msk); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 46f4655b7123..fdfa843e2d88 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -282,6 +282,9 @@ struct mptcp_sock { u64 bytes_acked; u64 snd_una; u64 wnd_end; + u32 last_data_sent; + u32 last_data_recv; + u32 last_ack_recv; unsigned long timer_ival; u32 token; int rmem_released; diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 9d5d42a77bcc..1fea43f5b6f3 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -898,6 +898,7 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) struct sock *sk = (struct sock *)msk; u32 flags = 0; bool slow; + u32 now; memset(info, 0, sizeof(*info)); @@ -926,11 +927,6 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) if (READ_ONCE(msk->can_ack)) flags |= MPTCP_INFO_FLAG_REMOTE_KEY_RECEIVED; info->mptcpi_flags = flags; - mptcp_data_lock(sk); - info->mptcpi_snd_una = msk->snd_una; - info->mptcpi_rcv_nxt = msk->ack_seq; - info->mptcpi_bytes_acked = msk->bytes_acked; - mptcp_data_unlock(sk); slow = lock_sock_fast(sk); info->mptcpi_csum_enabled = READ_ONCE(msk->csum_enabled); @@ -942,7 +938,17 @@ void mptcp_diag_fill_info(struct mptcp_sock *msk, struct mptcp_info *info) info->mptcpi_bytes_retrans = msk->bytes_retrans; info->mptcpi_subflows_total = info->mptcpi_subflows + __mptcp_has_initial_subflow(msk); + now = tcp_jiffies32; + info->mptcpi_last_data_sent = jiffies_to_msecs(now - msk->last_data_sent); + info->mptcpi_last_data_recv = jiffies_to_msecs(now - msk->last_data_recv); unlock_sock_fast(sk, slow); + + mptcp_data_lock(sk); + info->mptcpi_last_ack_recv = jiffies_to_msecs(now - msk->last_ack_recv); + info->mptcpi_snd_una = msk->snd_una; + info->mptcpi_rcv_nxt = msk->ack_seq; + info->mptcpi_bytes_acked = msk->bytes_acked; + mptcp_data_unlock(sk); } EXPORT_SYMBOL_GPL(mptcp_diag_fill_info); -- cgit v1.2.3 From f6d827b180bda01f8805bf5e85307419b0d6f890 Mon Sep 17 00:00:00 2001 From: Mina Almasry Date: Wed, 10 Apr 2024 12:05:01 -0700 Subject: net: move skb ref helpers to new header Add a new header, linux/skbuff_ref.h, which contains all the skb_*_ref() helpers. Many of the consumers of skbuff.h do not actually use any of the skb ref helpers, and we can speed up compilation a bit by minimizing this header file. Additionally in the later patch in the series we add page_pool support to skb_frag_ref(), which requires some page_pool dependencies. We can now add these dependencies to skbuff_ref.h instead of a very ubiquitous skbuff.h Signed-off-by: Mina Almasry Link: https://lore.kernel.org/r/20240410190505.1225848-2-almasrymina@google.com Signed-off-by: Jakub Kicinski --- .../chelsio/inline_crypto/ch_ktls/chcr_ktls.c | 1 + drivers/net/ethernet/marvell/sky2.c | 1 + drivers/net/ethernet/mellanox/mlx4/en_rx.c | 1 + drivers/net/ethernet/sun/cassini.c | 1 + drivers/net/veth.c | 1 + drivers/net/xen-netback/netback.c | 1 + include/linux/skbuff.h | 63 ------------------ include/linux/skbuff_ref.h | 75 ++++++++++++++++++++++ net/core/gro.c | 1 + net/core/skbuff.c | 1 + net/ipv4/esp4.c | 1 + net/ipv4/tcp_output.c | 1 + net/ipv6/esp6.c | 1 + net/tls/tls_device.c | 1 + net/tls/tls_device_fallback.c | 1 + net/tls/tls_strp.c | 1 + 16 files changed, 89 insertions(+), 63 deletions(-) create mode 100644 include/linux/skbuff_ref.h (limited to 'net') diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c index 6482728794dd..e8e460a92e0e 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c @@ -10,6 +10,7 @@ #include #include #include +#include #include "chcr_ktls.h" static LIST_HEAD(uld_ctx_list); diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c index 07720841a8d7..f3f7f4cc27b3 100644 --- a/drivers/net/ethernet/marvell/sky2.c +++ b/drivers/net/ethernet/marvell/sky2.c @@ -34,6 +34,7 @@ #include #include #include +#include #include diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index eac49657bd07..8328df8645d5 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -42,6 +42,7 @@ #include #include #include +#include #include #if IS_ENABLED(CONFIG_IPV6) diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index bfb903506367..8f1f43dbb76d 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -73,6 +73,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/net/veth.c b/drivers/net/veth.c index bcdfbf61eb66..426e68a95067 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #define DRV_NAME "veth" diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index ef76850d9bcd..48254fc07d64 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -38,6 +38,7 @@ #include #include #include +#include #include diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7135a3e94afd..4072a7ee3859 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3492,73 +3492,10 @@ static inline struct page *skb_frag_page(const skb_frag_t *frag) return netmem_to_page(frag->netmem); } -/** - * __skb_frag_ref - take an addition reference on a paged fragment. - * @frag: the paged fragment - * - * Takes an additional reference on the paged fragment @frag. - */ -static inline void __skb_frag_ref(skb_frag_t *frag) -{ - get_page(skb_frag_page(frag)); -} - -/** - * skb_frag_ref - take an addition reference on a paged fragment of an skb. - * @skb: the buffer - * @f: the fragment offset. - * - * Takes an additional reference on the @f'th paged fragment of @skb. - */ -static inline void skb_frag_ref(struct sk_buff *skb, int f) -{ - __skb_frag_ref(&skb_shinfo(skb)->frags[f]); -} - int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb, unsigned int headroom); int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb, struct bpf_prog *prog); -bool napi_pp_put_page(struct page *page); - -static inline void -skb_page_unref(struct page *page, bool recycle) -{ -#ifdef CONFIG_PAGE_POOL - if (recycle && napi_pp_put_page(page)) - return; -#endif - put_page(page); -} - -/** - * __skb_frag_unref - release a reference on a paged fragment. - * @frag: the paged fragment - * @recycle: recycle the page if allocated via page_pool - * - * Releases a reference on the paged fragment @frag - * or recycles the page via the page_pool API. - */ -static inline void __skb_frag_unref(skb_frag_t *frag, bool recycle) -{ - skb_page_unref(skb_frag_page(frag), recycle); -} - -/** - * skb_frag_unref - release a reference on a paged fragment of an skb. - * @skb: the buffer - * @f: the fragment offset - * - * Releases a reference on the @f'th paged fragment of @skb. - */ -static inline void skb_frag_unref(struct sk_buff *skb, int f) -{ - struct skb_shared_info *shinfo = skb_shinfo(skb); - - if (!skb_zcopy_managed(skb)) - __skb_frag_unref(&shinfo->frags[f], skb->pp_recycle); -} - /** * skb_frag_address - gets the address of the data contained in a paged fragment * @frag: the paged fragment buffer diff --git a/include/linux/skbuff_ref.h b/include/linux/skbuff_ref.h new file mode 100644 index 000000000000..11f0a4063403 --- /dev/null +++ b/include/linux/skbuff_ref.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Skb ref helpers. + * + */ + +#ifndef _LINUX_SKBUFF_REF_H +#define _LINUX_SKBUFF_REF_H + +#include + +/** + * __skb_frag_ref - take an addition reference on a paged fragment. + * @frag: the paged fragment + * + * Takes an additional reference on the paged fragment @frag. + */ +static inline void __skb_frag_ref(skb_frag_t *frag) +{ + get_page(skb_frag_page(frag)); +} + +/** + * skb_frag_ref - take an addition reference on a paged fragment of an skb. + * @skb: the buffer + * @f: the fragment offset. + * + * Takes an additional reference on the @f'th paged fragment of @skb. + */ +static inline void skb_frag_ref(struct sk_buff *skb, int f) +{ + __skb_frag_ref(&skb_shinfo(skb)->frags[f]); +} + +bool napi_pp_put_page(struct page *page); + +static inline void +skb_page_unref(struct page *page, bool recycle) +{ +#ifdef CONFIG_PAGE_POOL + if (recycle && napi_pp_put_page(page)) + return; +#endif + put_page(page); +} + +/** + * __skb_frag_unref - release a reference on a paged fragment. + * @frag: the paged fragment + * @recycle: recycle the page if allocated via page_pool + * + * Releases a reference on the paged fragment @frag + * or recycles the page via the page_pool API. + */ +static inline void __skb_frag_unref(skb_frag_t *frag, bool recycle) +{ + skb_page_unref(skb_frag_page(frag), recycle); +} + +/** + * skb_frag_unref - release a reference on a paged fragment of an skb. + * @skb: the buffer + * @f: the fragment offset + * + * Releases a reference on the @f'th paged fragment of @skb. + */ +static inline void skb_frag_unref(struct sk_buff *skb, int f) +{ + struct skb_shared_info *shinfo = skb_shinfo(skb); + + if (!skb_zcopy_managed(skb)) + __skb_frag_unref(&shinfo->frags[f], skb->pp_recycle); +} + +#endif /* _LINUX_SKBUFF_REF_H */ diff --git a/net/core/gro.c b/net/core/gro.c index 83f35d99a682..2459ab697f7f 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -3,6 +3,7 @@ #include #include #include +#include #define MAX_GRO_SKBS 8 diff --git a/net/core/skbuff.c b/net/core/skbuff.c index ab970ded8a7b..2554a6f5f386 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -51,6 +51,7 @@ #endif #include #include +#include #include #include #include diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 40330253f076..dff04580318f 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -20,6 +20,7 @@ #include #include #include +#include #include diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 9282fafc0e61..61119d42b0fd 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -44,6 +44,7 @@ #include #include #include +#include #include diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index fb431d0a3475..6bc0a84c8d05 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -36,6 +36,7 @@ #include #include #include +#include #include diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index bf8ed36b1ad6..ab6e694f7bc2 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -37,6 +37,7 @@ #include #include #include +#include #include "tls.h" #include "trace.h" diff --git a/net/tls/tls_device_fallback.c b/net/tls/tls_device_fallback.c index 4e7228f275fa..f9e3d3d90dcf 100644 --- a/net/tls/tls_device_fallback.c +++ b/net/tls/tls_device_fallback.c @@ -33,6 +33,7 @@ #include #include #include +#include #include "tls.h" diff --git a/net/tls/tls_strp.c b/net/tls/tls_strp.c index ca1e0e198ceb..58c4b06f4f0c 100644 --- a/net/tls/tls_strp.c +++ b/net/tls/tls_strp.c @@ -2,6 +2,7 @@ /* Copyright (c) 2016 Tom Herbert */ #include +#include #include #include #include -- cgit v1.2.3 From a580ea994fd37f4105028f5a85c38ff6508a2b25 Mon Sep 17 00:00:00 2001 From: Mina Almasry Date: Wed, 10 Apr 2024 12:05:02 -0700 Subject: net: mirror skb frag ref/unref helpers Refactor some of the skb frag ref/unref helpers for improved clarity. Implement napi_pp_get_page() to be the mirror counterpart of napi_pp_put_page(). Implement skb_page_ref() to be the mirror of skb_page_unref(). Improve __skb_frag_ref() to become a mirror counterpart of __skb_frag_unref(). Previously unref could handle pp & non-pp pages, while the ref could only handle non-pp pages. Now both the ref & unref helpers can correctly handle both pp & non-pp pages. Now that __skb_frag_ref() can handle both pp & non-pp pages, remove skb_pp_frag_ref(), and use __skb_frag_ref() instead. This lets us remove pp specific handling from skb_try_coalesce. Additionally, since __skb_frag_ref() can now handle both pp & non-pp pages, a latent issue in skb_shift() should now be fixed. Previously this function would do a non-pp ref & pp unref on potential pp frags (fragfrom). After this patch, skb_shift() should correctly do a pp ref/unref on pp frags. Signed-off-by: Mina Almasry Reviewed-by: Dragos Tatulea Reviewed-by: Jacob Keller Link: https://lore.kernel.org/r/20240410190505.1225848-3-almasrymina@google.com Signed-off-by: Jakub Kicinski --- .../chelsio/inline_crypto/ch_ktls/chcr_ktls.c | 2 +- drivers/net/ethernet/sun/cassini.c | 4 +- drivers/net/veth.c | 2 +- include/linux/skbuff_ref.h | 39 ++++++++++++++++-- net/core/skbuff.c | 46 ++-------------------- net/tls/tls_device_fallback.c | 2 +- 6 files changed, 44 insertions(+), 51 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c index e8e460a92e0e..3832c2e8ea5a 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c @@ -1659,7 +1659,7 @@ static void chcr_ktls_copy_record_in_skb(struct sk_buff *nskb, for (i = 0; i < record->num_frags; i++) { skb_shinfo(nskb)->frags[i] = record->frags[i]; /* increase the frag ref count */ - __skb_frag_ref(&skb_shinfo(nskb)->frags[i]); + __skb_frag_ref(&skb_shinfo(nskb)->frags[i], nskb->pp_recycle); } skb_shinfo(nskb)->nr_frags = record->num_frags; diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index 8f1f43dbb76d..f058e154a3bc 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -2000,7 +2000,7 @@ static int cas_rx_process_pkt(struct cas *cp, struct cas_rx_comp *rxc, skb->len += hlen - swivel; skb_frag_fill_page_desc(frag, page->buffer, off, hlen - swivel); - __skb_frag_ref(frag); + __skb_frag_ref(frag, skb->pp_recycle); /* any more data? */ if ((words[0] & RX_COMP1_SPLIT_PKT) && ((dlen -= hlen) > 0)) { @@ -2024,7 +2024,7 @@ static int cas_rx_process_pkt(struct cas *cp, struct cas_rx_comp *rxc, frag++; skb_frag_fill_page_desc(frag, page->buffer, 0, hlen); - __skb_frag_ref(frag); + __skb_frag_ref(frag, skb->pp_recycle); RX_USED_ADD(page, hlen + cp->crc_size); } diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 426e68a95067..0b0293629329 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -717,7 +717,7 @@ static void veth_xdp_get(struct xdp_buff *xdp) return; for (i = 0; i < sinfo->nr_frags; i++) - __skb_frag_ref(&sinfo->frags[i]); + __skb_frag_ref(&sinfo->frags[i], false); } static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq, diff --git a/include/linux/skbuff_ref.h b/include/linux/skbuff_ref.h index 11f0a4063403..4dcdbe9fbc5f 100644 --- a/include/linux/skbuff_ref.h +++ b/include/linux/skbuff_ref.h @@ -8,16 +8,47 @@ #define _LINUX_SKBUFF_REF_H #include +#include + +#ifdef CONFIG_PAGE_POOL +static inline bool is_pp_page(struct page *page) +{ + return (page->pp_magic & ~0x3UL) == PP_SIGNATURE; +} + +static inline bool napi_pp_get_page(struct page *page) +{ + page = compound_head(page); + + if (!is_pp_page(page)) + return false; + + page_pool_ref_page(page); + return true; +} +#endif + +static inline void skb_page_ref(struct page *page, bool recycle) +{ +#ifdef CONFIG_PAGE_POOL + if (recycle && napi_pp_get_page(page)) + return; +#endif + get_page(page); +} /** * __skb_frag_ref - take an addition reference on a paged fragment. * @frag: the paged fragment + * @recycle: skb->pp_recycle param of the parent skb. False if no parent skb. * - * Takes an additional reference on the paged fragment @frag. + * Takes an additional reference on the paged fragment @frag. Obtains the + * correct reference count depending on whether skb->pp_recycle is set and + * whether the frag is a page pool frag. */ -static inline void __skb_frag_ref(skb_frag_t *frag) +static inline void __skb_frag_ref(skb_frag_t *frag, bool recycle) { - get_page(skb_frag_page(frag)); + skb_page_ref(skb_frag_page(frag), recycle); } /** @@ -29,7 +60,7 @@ static inline void __skb_frag_ref(skb_frag_t *frag) */ static inline void skb_frag_ref(struct sk_buff *skb, int f) { - __skb_frag_ref(&skb_shinfo(skb)->frags[f]); + __skb_frag_ref(&skb_shinfo(skb)->frags[f], skb->pp_recycle); } bool napi_pp_put_page(struct page *page); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 2554a6f5f386..ea052fa710d8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -907,11 +907,6 @@ static void skb_clone_fraglist(struct sk_buff *skb) skb_get(list); } -static bool is_pp_page(struct page *page) -{ - return (page->pp_magic & ~0x3UL) == PP_SIGNATURE; -} - int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb, unsigned int headroom) { @@ -1033,37 +1028,6 @@ static bool skb_pp_recycle(struct sk_buff *skb, void *data) return napi_pp_put_page(virt_to_page(data)); } -/** - * skb_pp_frag_ref() - Increase fragment references of a page pool aware skb - * @skb: page pool aware skb - * - * Increase the fragment reference count (pp_ref_count) of a skb. This is - * intended to gain fragment references only for page pool aware skbs, - * i.e. when skb->pp_recycle is true, and not for fragments in a - * non-pp-recycling skb. It has a fallback to increase references on normal - * pages, as page pool aware skbs may also have normal page fragments. - */ -static int skb_pp_frag_ref(struct sk_buff *skb) -{ - struct skb_shared_info *shinfo; - struct page *head_page; - int i; - - if (!skb->pp_recycle) - return -EINVAL; - - shinfo = skb_shinfo(skb); - - for (i = 0; i < shinfo->nr_frags; i++) { - head_page = compound_head(skb_frag_page(&shinfo->frags[i])); - if (likely(is_pp_page(head_page))) - page_pool_ref_page(head_page); - else - page_ref_inc(head_page); - } - return 0; -} - static void skb_kfree_head(void *head, unsigned int end_offset) { if (end_offset == SKB_SMALL_HEAD_HEADROOM) @@ -4176,7 +4140,7 @@ int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen) to++; } else { - __skb_frag_ref(fragfrom); + __skb_frag_ref(fragfrom, skb->pp_recycle); skb_frag_page_copy(fragto, fragfrom); skb_frag_off_copy(fragto, fragfrom); skb_frag_size_set(fragto, todo); @@ -4826,7 +4790,7 @@ normal: } *nskb_frag = (i < 0) ? skb_head_frag_to_page_desc(frag_skb) : *frag; - __skb_frag_ref(nskb_frag); + __skb_frag_ref(nskb_frag, nskb->pp_recycle); size = skb_frag_size(nskb_frag); if (pos < offset) { @@ -5957,10 +5921,8 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, /* if the skb is not cloned this does nothing * since we set nr_frags to 0. */ - if (skb_pp_frag_ref(from)) { - for (i = 0; i < from_shinfo->nr_frags; i++) - __skb_frag_ref(&from_shinfo->frags[i]); - } + for (i = 0; i < from_shinfo->nr_frags; i++) + __skb_frag_ref(&from_shinfo->frags[i], from->pp_recycle); to->truesize += delta; to->len += len; diff --git a/net/tls/tls_device_fallback.c b/net/tls/tls_device_fallback.c index f9e3d3d90dcf..9237dded4467 100644 --- a/net/tls/tls_device_fallback.c +++ b/net/tls/tls_device_fallback.c @@ -278,7 +278,7 @@ static int fill_sg_in(struct scatterlist *sg_in, for (i = 0; remaining > 0; i++) { skb_frag_t *frag = &record->frags[i]; - __skb_frag_ref(frag); + __skb_frag_ref(frag, false); sg_set_page(sg_in + i, skb_frag_page(frag), skb_frag_size(frag), skb_frag_off(frag)); -- cgit v1.2.3 From e0f89d2864b062b027196925ea19f94b2ce50d6a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 10 Apr 2024 11:19:50 +0000 Subject: mpls: no longer hold RTNL in mpls_netconf_dump_devconf() - Use for_each_netdev_dump() to no longer rely on net->dev_index_head hash table. - No longer care of net->dev_base_seq - Fix return value at the end of a dump, so that NLMSG_DONE can be appended to current skb, saving one recvmsg() system call. - No longer grab RTNL, RCU protection is enough, afer adding one READ_ONCE(mdev->input_enabled) in mpls_netconf_fill_devconf() Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240410111951.2673193-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/mpls/af_mpls.c | 59 ++++++++++++++++++++---------------------------------- 1 file changed, 22 insertions(+), 37 deletions(-) (limited to 'net') diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 1303acb9cdd2..0315b8deed3f 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1154,7 +1154,7 @@ static int mpls_netconf_fill_devconf(struct sk_buff *skb, struct mpls_dev *mdev, if ((all || type == NETCONFA_INPUT) && nla_put_s32(skb, NETCONFA_INPUT, - mdev->input_enabled) < 0) + READ_ONCE(mdev->input_enabled)) < 0) goto nla_put_failure; nlmsg_end(skb, nlh); @@ -1303,11 +1303,12 @@ static int mpls_netconf_dump_devconf(struct sk_buff *skb, { const struct nlmsghdr *nlh = cb->nlh; struct net *net = sock_net(skb->sk); - struct hlist_head *head; + struct { + unsigned long ifindex; + } *ctx = (void *)cb->ctx; struct net_device *dev; struct mpls_dev *mdev; - int idx, s_idx; - int h, s_h; + int err = 0; if (cb->strict_check) { struct netlink_ext_ack *extack = cb->extack; @@ -1324,40 +1325,23 @@ static int mpls_netconf_dump_devconf(struct sk_buff *skb, } } - s_h = cb->args[0]; - s_idx = idx = cb->args[1]; - - for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) { - idx = 0; - head = &net->dev_index_head[h]; - rcu_read_lock(); - cb->seq = net->dev_base_seq; - hlist_for_each_entry_rcu(dev, head, index_hlist) { - if (idx < s_idx) - goto cont; - mdev = mpls_dev_get(dev); - if (!mdev) - goto cont; - if (mpls_netconf_fill_devconf(skb, mdev, - NETLINK_CB(cb->skb).portid, - nlh->nlmsg_seq, - RTM_NEWNETCONF, - NLM_F_MULTI, - NETCONFA_ALL) < 0) { - rcu_read_unlock(); - goto done; - } - nl_dump_check_consistent(cb, nlmsg_hdr(skb)); -cont: - idx++; - } - rcu_read_unlock(); + rcu_read_lock(); + for_each_netdev_dump(net, dev, ctx->ifindex) { + mdev = mpls_dev_get(dev); + if (!mdev) + continue; + err = mpls_netconf_fill_devconf(skb, mdev, + NETLINK_CB(cb->skb).portid, + nlh->nlmsg_seq, + RTM_NEWNETCONF, + NLM_F_MULTI, + NETCONFA_ALL); + if (err < 0) + break; } -done: - cb->args[0] = h; - cb->args[1] = idx; + rcu_read_unlock(); - return skb->len; + return err; } #define MPLS_PERDEV_SYSCTL_OFFSET(field) \ @@ -2773,7 +2757,8 @@ static int __init mpls_init(void) mpls_getroute, mpls_dump_routes, 0); rtnl_register_module(THIS_MODULE, PF_MPLS, RTM_GETNETCONF, mpls_netconf_get_devconf, - mpls_netconf_dump_devconf, 0); + mpls_netconf_dump_devconf, + RTNL_FLAG_DUMP_UNLOCKED); err = ipgre_tunnel_encap_add_mpls_ops(); if (err) pr_err("Can't add mpls over gre tunnel ops\n"); -- cgit v1.2.3 From 05ea491641d338422edbd87af3b311c16ea762f5 Mon Sep 17 00:00:00 2001 From: Jon Maloy Date: Tue, 9 Apr 2024 11:28:05 -0400 Subject: tcp: add support for SO_PEEK_OFF socket option When reading received messages from a socket with MSG_PEEK, we may want to read the contents with an offset, like we can do with pread/preadv() when reading files. Currently, it is not possible to do that. In this commit, we add support for the SO_PEEK_OFF socket option for TCP, in a similar way it is done for Unix Domain sockets. In the iperf3 log examples shown below, we can observe a throughput improvement of 15-20 % in the direction host->namespace when using the protocol splicer 'pasta' (https://passt.top). This is a consistent result. pasta(1) and passt(1) implement user-mode networking for network namespaces (containers) and virtual machines by means of a translation layer between Layer-2 network interface and native Layer-4 sockets (TCP, UDP, ICMP/ICMPv6 echo). Received, pending TCP data to the container/guest is kept in kernel buffers until acknowledged, so the tool routinely needs to fetch new data from socket, skipping data that was already sent. At the moment this is implemented using a dummy buffer passed to recvmsg(). With this change, we don't need a dummy buffer and the related buffer copy (copy_to_user()) anymore. passt and pasta are supported in KubeVirt and libvirt/qemu. jmaloy@freyr:~/passt$ perf record -g ./pasta --config-net -f SO_PEEK_OFF not supported by kernel. jmaloy@freyr:~/passt# iperf3 -s ----------------------------------------------------------- Server listening on 5201 (test #1) ----------------------------------------------------------- Accepted connection from 192.168.122.1, port 44822 [ 5] local 192.168.122.180 port 5201 connected to 192.168.122.1 port 44832 [ ID] Interval Transfer Bitrate [ 5] 0.00-1.00 sec 1.02 GBytes 8.78 Gbits/sec [ 5] 1.00-2.00 sec 1.06 GBytes 9.08 Gbits/sec [ 5] 2.00-3.00 sec 1.07 GBytes 9.15 Gbits/sec [ 5] 3.00-4.00 sec 1.10 GBytes 9.46 Gbits/sec [ 5] 4.00-5.00 sec 1.03 GBytes 8.85 Gbits/sec [ 5] 5.00-6.00 sec 1.10 GBytes 9.44 Gbits/sec [ 5] 6.00-7.00 sec 1.11 GBytes 9.56 Gbits/sec [ 5] 7.00-8.00 sec 1.07 GBytes 9.20 Gbits/sec [ 5] 8.00-9.00 sec 667 MBytes 5.59 Gbits/sec [ 5] 9.00-10.00 sec 1.03 GBytes 8.83 Gbits/sec [ 5] 10.00-10.04 sec 30.1 MBytes 6.36 Gbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate [ 5] 0.00-10.04 sec 10.3 GBytes 8.78 Gbits/sec receiver ----------------------------------------------------------- Server listening on 5201 (test #2) ----------------------------------------------------------- ^Ciperf3: interrupt - the server has terminated jmaloy@freyr:~/passt# logout [ perf record: Woken up 23 times to write data ] [ perf record: Captured and wrote 5.696 MB perf.data (35580 samples) ] jmaloy@freyr:~/passt$ jmaloy@freyr:~/passt$ perf record -g ./pasta --config-net -f SO_PEEK_OFF supported by kernel. jmaloy@freyr:~/passt# iperf3 -s ----------------------------------------------------------- Server listening on 5201 (test #1) ----------------------------------------------------------- Accepted connection from 192.168.122.1, port 52084 [ 5] local 192.168.122.180 port 5201 connected to 192.168.122.1 port 52098 [ ID] Interval Transfer Bitrate [ 5] 0.00-1.00 sec 1.32 GBytes 11.3 Gbits/sec [ 5] 1.00-2.00 sec 1.19 GBytes 10.2 Gbits/sec [ 5] 2.00-3.00 sec 1.26 GBytes 10.8 Gbits/sec [ 5] 3.00-4.00 sec 1.36 GBytes 11.7 Gbits/sec [ 5] 4.00-5.00 sec 1.33 GBytes 11.4 Gbits/sec [ 5] 5.00-6.00 sec 1.21 GBytes 10.4 Gbits/sec [ 5] 6.00-7.00 sec 1.31 GBytes 11.2 Gbits/sec [ 5] 7.00-8.00 sec 1.25 GBytes 10.7 Gbits/sec [ 5] 8.00-9.00 sec 1.33 GBytes 11.5 Gbits/sec [ 5] 9.00-10.00 sec 1.24 GBytes 10.7 Gbits/sec [ 5] 10.00-10.04 sec 56.0 MBytes 12.1 Gbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - [ ID] Interval Transfer Bitrate [ 5] 0.00-10.04 sec 12.9 GBytes 11.0 Gbits/sec receiver ----------------------------------------------------------- Server listening on 5201 (test #2) ----------------------------------------------------------- ^Ciperf3: interrupt - the server has terminated logout [ perf record: Woken up 20 times to write data ] [ perf record: Captured and wrote 5.040 MB perf.data (33411 samples) ] jmaloy@freyr:~/passt$ The perf record confirms this result. Below, we can observe that the CPU spends significantly less time in the function ____sys_recvmsg() when we have offset support. Without offset support: ---------------------- jmaloy@freyr:~/passt$ perf report -q --symbol-filter=do_syscall_64 \ -p ____sys_recvmsg -x --stdio -i perf.data | head -1 46.32% 0.00% passt.avx2 [kernel.vmlinux] [k] do_syscall_64 ____sys_recvmsg With offset support: ---------------------- jmaloy@freyr:~/passt$ perf report -q --symbol-filter=do_syscall_64 \ -p ____sys_recvmsg -x --stdio -i perf.data | head -1 28.12% 0.00% passt.avx2 [kernel.vmlinux] [k] do_syscall_64 ____sys_recvmsg Suggested-by: Paolo Abeni Reviewed-by: Stefano Brivio Signed-off-by: Jon Maloy Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240409152805.913891-1-jmaloy@redhat.com Signed-off-by: Jakub Kicinski --- net/ipv4/af_inet.c | 1 + net/ipv4/tcp.c | 16 ++++++++++------ 2 files changed, 11 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 55bd72997b31..a7cfeda28bb2 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1072,6 +1072,7 @@ const struct proto_ops inet_stream_ops = { #endif .splice_eof = inet_splice_eof, .splice_read = tcp_splice_read, + .set_peek_off = sk_set_peek_off, .read_sock = tcp_read_sock, .read_skb = tcp_read_skb, .sendmsg_locked = tcp_sendmsg_locked, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index e1bf468e0d22..f23b97777ea5 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1419,8 +1419,6 @@ static int tcp_peek_sndq(struct sock *sk, struct msghdr *msg, int len) struct sk_buff *skb; int copied = 0, err = 0; - /* XXX -- need to support SO_PEEK_OFF */ - skb_rbtree_walk(skb, &sk->tcp_rtx_queue) { err = skb_copy_datagram_msg(skb, 0, msg, skb->len); if (err) @@ -2331,6 +2329,7 @@ static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, int target; /* Read at least this many bytes */ long timeo; struct sk_buff *skb, *last; + u32 peek_offset = 0; u32 urg_hole = 0; err = -ENOTCONN; @@ -2364,7 +2363,8 @@ static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, seq = &tp->copied_seq; if (flags & MSG_PEEK) { - peek_seq = tp->copied_seq; + peek_offset = max(sk_peek_offset(sk, flags), 0); + peek_seq = tp->copied_seq + peek_offset; seq = &peek_seq; } @@ -2467,11 +2467,11 @@ static int tcp_recvmsg_locked(struct sock *sk, struct msghdr *msg, size_t len, } if ((flags & MSG_PEEK) && - (peek_seq - copied - urg_hole != tp->copied_seq)) { + (peek_seq - peek_offset - copied - urg_hole != tp->copied_seq)) { net_dbg_ratelimited("TCP(%s:%d): Application bug, race in MSG_PEEK\n", current->comm, task_pid_nr(current)); - peek_seq = tp->copied_seq; + peek_seq = tp->copied_seq + peek_offset; } continue; @@ -2512,7 +2512,10 @@ found_ok_skb: WRITE_ONCE(*seq, *seq + used); copied += used; len -= used; - + if (flags & MSG_PEEK) + sk_peek_offset_fwd(sk, used); + else + sk_peek_offset_bwd(sk, used); tcp_rcv_space_adjust(sk); skip_copy: @@ -3013,6 +3016,7 @@ int tcp_disconnect(struct sock *sk, int flags) __skb_queue_purge(&sk->sk_receive_queue); WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); WRITE_ONCE(tp->urg_data, 0); + sk_set_peek_off(sk, -1); tcp_write_queue_purge(sk); tcp_fastopen_active_disable_ofo_check(sk); skb_rbtree_purge(&tp->out_of_order_queue); -- cgit v1.2.3 From 5618603f5d063c8d769d2e64cd440f787567fd13 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Wed, 10 Apr 2024 15:14:29 +0200 Subject: ipv4: Remove RTO_ONLINK. RTO_ONLINK was a flag used in ->flowi4_tos that allowed to alter the scope of an IPv4 route lookup. Setting this flag was equivalent to specifying RT_SCOPE_LINK in ->flowi4_scope. With commit ec20b2830093 ("ipv4: Set scope explicitly in ip_route_output()."), the last users of RTO_ONLINK have been removed. Therefore, we can now drop the code that checked this bit and stop modifying ->flowi4_scope in ip_route_output_key_hash(). Signed-off-by: Guillaume Nault Reviewed-by: Przemek Kitszel Reviewed-by: David Ahern Link: https://lore.kernel.org/r/57de760565cab55df7b129f523530ac6475865b2.1712754146.git.gnault@redhat.com Signed-off-by: Jakub Kicinski --- include/net/route.h | 2 -- net/ipv4/route.c | 14 +------------- 2 files changed, 1 insertion(+), 15 deletions(-) (limited to 'net') diff --git a/include/net/route.h b/include/net/route.h index 315a8acee6c6..630d1ef6868a 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -35,8 +35,6 @@ #include #include -#define RTO_ONLINK 0x01 - static inline __u8 ip_sock_rt_scope(const struct sock *sk) { if (sock_flag(sk, SOCK_LOCALROUTE)) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index d36ace160d42..6133f1fd848e 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -106,9 +106,6 @@ #include "fib_lookup.h" -#define RT_FL_TOS(oldflp4) \ - ((oldflp4)->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK)) - #define RT_GC_TIMEOUT (300*HZ) #define DEFAULT_MIN_PMTU (512 + 20 + 20) @@ -498,15 +495,6 @@ void __ip_select_ident(struct net *net, struct iphdr *iph, int segs) } EXPORT_SYMBOL(__ip_select_ident); -static void ip_rt_fix_tos(struct flowi4 *fl4) -{ - __u8 tos = RT_FL_TOS(fl4); - - fl4->flowi4_tos = tos & IPTOS_RT_MASK; - if (tos & RTO_ONLINK) - fl4->flowi4_scope = RT_SCOPE_LINK; -} - static void __build_flow_key(const struct net *net, struct flowi4 *fl4, const struct sock *sk, const struct iphdr *iph, int oif, __u8 tos, u8 prot, u32 mark, @@ -2636,7 +2624,7 @@ struct rtable *ip_route_output_key_hash(struct net *net, struct flowi4 *fl4, struct rtable *rth; fl4->flowi4_iif = LOOPBACK_IFINDEX; - ip_rt_fix_tos(fl4); + fl4->flowi4_tos &= IPTOS_RT_MASK; rcu_read_lock(); rth = ip_route_output_key_hash_rcu(net, fl4, &res, skb); -- cgit v1.2.3 From f7ac8fbd3215e72824c08b53cd2ff272d42eb823 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 10 Apr 2024 15:41:36 +0100 Subject: tls: remove redundant assignment to variable decrypted The variable decrypted is being assigned a value that is never read, the control of flow after the assignment is via an return path and decrypted is not referenced in this path. The assignment is redundant and can be removed. Cleans up clang scan warning: net/tls/tls_sw.c:2150:4: warning: Value stored to 'decrypted' is never read [deadcode.DeadStores] Signed-off-by: Colin Ian King Link: https://lore.kernel.org/r/20240410144136.289030-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski --- net/tls/tls_sw.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index b783231668c6..305a412785f5 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -2147,7 +2147,6 @@ recv_end: if (ret) { if (err >= 0 || err == -EINPROGRESS) err = ret; - decrypted = 0; goto end; } -- cgit v1.2.3 From dd0c9855b41310470086500c9963bbb64bb90dd0 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 10 Apr 2024 20:42:38 +0100 Subject: net: dsa: introduce dsa_phylink_to_port() We convert from a phylink_config struct to a dsa_port struct in many places, let's provide a helper for this. Reviewed-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: Russell King (Oracle) Reviewed-by: Vladimir Oltean Link: https://lore.kernel.org/r/E1rudqA-006K9B-85@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 6 ++++++ net/dsa/port.c | 12 ++++++------ 2 files changed, 12 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 7c0da9effe4e..f228b479a5fd 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -327,6 +327,12 @@ struct dsa_port { }; }; +static inline struct dsa_port * +dsa_phylink_to_port(struct phylink_config *config) +{ + return container_of(config, struct dsa_port, pl_config); +} + /* TODO: ideally DSA ports would have a single dp->link_dp member, * and no dst->rtable nor this struct dsa_link would be needed, * but this would require some more complex tree walking, diff --git a/net/dsa/port.c b/net/dsa/port.c index c42dac87671b..02bf1c306bdc 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -1558,7 +1558,7 @@ static struct phylink_pcs * dsa_port_phylink_mac_select_pcs(struct phylink_config *config, phy_interface_t interface) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct phylink_pcs *pcs = ERR_PTR(-EOPNOTSUPP); struct dsa_switch *ds = dp->ds; @@ -1572,7 +1572,7 @@ static int dsa_port_phylink_mac_prepare(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct dsa_switch *ds = dp->ds; int err = 0; @@ -1587,7 +1587,7 @@ static void dsa_port_phylink_mac_config(struct phylink_config *config, unsigned int mode, const struct phylink_link_state *state) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct dsa_switch *ds = dp->ds; if (!ds->ops->phylink_mac_config) @@ -1600,7 +1600,7 @@ static int dsa_port_phylink_mac_finish(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct dsa_switch *ds = dp->ds; int err = 0; @@ -1615,7 +1615,7 @@ static void dsa_port_phylink_mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct phy_device *phydev = NULL; struct dsa_switch *ds = dp->ds; @@ -1638,7 +1638,7 @@ static void dsa_port_phylink_mac_link_up(struct phylink_config *config, int speed, int duplex, bool tx_pause, bool rx_pause) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct dsa_switch *ds = dp->ds; if (!ds->ops->phylink_mac_link_up) { -- cgit v1.2.3 From cae425cb43feddd9fd62fc1b25567f9463da4915 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 10 Apr 2024 20:42:43 +0100 Subject: net: dsa: allow DSA switch drivers to provide their own phylink mac ops Rather than having a shim for each and every phylink MAC operation, allow DSA switch drivers to provide their own ops structure. When a DSA driver provides the phylink MAC operations, the shimmed ops must not be provided, so fail an attempt to register a switch with both the phylink_mac_ops in struct dsa_switch and the phylink_mac_* operations populated in dsa_switch_ops populated. Signed-off-by: Russell King (Oracle) Reviewed-by: Vladimir Oltean Reviewed-by: Florian Fainelli Link: https://lore.kernel.org/r/E1rudqF-006K9H-Cc@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 5 +++++ net/dsa/dsa.c | 11 +++++++++++ net/dsa/port.c | 26 ++++++++++++++++++++------ 3 files changed, 36 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/net/dsa.h b/include/net/dsa.h index f228b479a5fd..7edfd8de8882 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -457,6 +457,11 @@ struct dsa_switch { */ const struct dsa_switch_ops *ops; + /* + * Allow a DSA switch driver to override the phylink MAC ops + */ + const struct phylink_mac_ops *phylink_mac_ops; + /* * User mii_bus and devices for the individual ports. */ diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 09d2f5d4b3dd..2f347cd37316 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -1505,6 +1505,17 @@ static int dsa_switch_probe(struct dsa_switch *ds) if (!ds->num_ports) return -EINVAL; + if (ds->phylink_mac_ops) { + if (ds->ops->phylink_mac_select_pcs || + ds->ops->phylink_mac_prepare || + ds->ops->phylink_mac_config || + ds->ops->phylink_mac_finish || + ds->ops->phylink_mac_link_down || + ds->ops->phylink_mac_link_up || + ds->ops->adjust_link) + return -EINVAL; + } + if (np) { err = dsa_switch_parse_of(ds, np); if (err) diff --git a/net/dsa/port.c b/net/dsa/port.c index 02bf1c306bdc..c6febc3d96d9 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -1662,6 +1662,7 @@ static const struct phylink_mac_ops dsa_port_phylink_mac_ops = { int dsa_port_phylink_create(struct dsa_port *dp) { + const struct phylink_mac_ops *mac_ops; struct dsa_switch *ds = dp->ds; phy_interface_t mode; struct phylink *pl; @@ -1685,8 +1686,12 @@ int dsa_port_phylink_create(struct dsa_port *dp) } } - pl = phylink_create(&dp->pl_config, of_fwnode_handle(dp->dn), - mode, &dsa_port_phylink_mac_ops); + mac_ops = &dsa_port_phylink_mac_ops; + if (ds->phylink_mac_ops) + mac_ops = ds->phylink_mac_ops; + + pl = phylink_create(&dp->pl_config, of_fwnode_handle(dp->dn), mode, + mac_ops); if (IS_ERR(pl)) { pr_err("error creating PHYLINK: %ld\n", PTR_ERR(pl)); return PTR_ERR(pl); @@ -1952,12 +1957,23 @@ static void dsa_shared_port_validate_of(struct dsa_port *dp, dn, dsa_port_is_cpu(dp) ? "CPU" : "DSA", dp->index); } +static void dsa_shared_port_link_down(struct dsa_port *dp) +{ + struct dsa_switch *ds = dp->ds; + + if (ds->phylink_mac_ops && ds->phylink_mac_ops->mac_link_down) + ds->phylink_mac_ops->mac_link_down(&dp->pl_config, MLO_AN_FIXED, + PHY_INTERFACE_MODE_NA); + else if (ds->ops->phylink_mac_link_down) + ds->ops->phylink_mac_link_down(ds, dp->index, MLO_AN_FIXED, + PHY_INTERFACE_MODE_NA); +} + int dsa_shared_port_link_register_of(struct dsa_port *dp) { struct dsa_switch *ds = dp->ds; bool missing_link_description; bool missing_phy_mode; - int port = dp->index; dsa_shared_port_validate_of(dp, &missing_phy_mode, &missing_link_description); @@ -1973,9 +1989,7 @@ int dsa_shared_port_link_register_of(struct dsa_port *dp) "Skipping phylink registration for %s port %d\n", dsa_port_is_cpu(dp) ? "CPU" : "DSA", dp->index); } else { - if (ds->ops->phylink_mac_link_down) - ds->ops->phylink_mac_link_down(ds, port, - MLO_AN_FIXED, PHY_INTERFACE_MODE_NA); + dsa_shared_port_link_down(dp); return dsa_shared_port_phylink_register(dp); } -- cgit v1.2.3 From a799de0e598583c18c03ce438648783b92c653de Mon Sep 17 00:00:00 2001 From: Lin Ma Date: Wed, 10 Apr 2024 11:48:46 +0800 Subject: net: nfc: remove inappropriate attrs check Revert "NFC: fix attrs checks in netlink interface" This reverts commit 18917d51472fe3b126a3a8f756c6b18085eb8130. Our checks found weird attrs present check in function nfc_genl_dep_link_down() and nfc_genl_llc_get_params(), which are introduced by commit 18917d51472f ("NFC: fix attrs checks in netlink interface"). According to its message, it should add checks for functions nfc_genl_deactivate_target() and nfc_genl_fw_download(). However, it didn't do that. In fact, the expected checks are added by (1) commit 385097a36757 ("nfc: Ensure presence of required attributes in the deactivate_target handler") and (2) commit 280e3ebdafb8 ("nfc: Ensure presence of NFC_ATTR_FIRMWARE_NAME attribute in nfc_genl_fw_download()"). Perhaps something went wrong. Anyway, the attr NFC_ATTR_TARGET_INDEX is never accessed in callback nfc_genl_dep_link_down() and same for NFC_ATTR_FIRMWARE_NAME and nfc_genl_llc_get_params(). Thus, remove those checks. Signed-off-by: Lin Ma Link: https://lore.kernel.org/r/20240410034846.167421-1-linma@zju.edu.cn Signed-off-by: Jakub Kicinski --- net/nfc/netlink.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index aa1dbf654c3e..dd2ce73a24fb 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -969,8 +969,7 @@ static int nfc_genl_dep_link_down(struct sk_buff *skb, struct genl_info *info) int rc; u32 idx; - if (!info->attrs[NFC_ATTR_DEVICE_INDEX] || - !info->attrs[NFC_ATTR_TARGET_INDEX]) + if (!info->attrs[NFC_ATTR_DEVICE_INDEX]) return -EINVAL; idx = nla_get_u32(info->attrs[NFC_ATTR_DEVICE_INDEX]); @@ -1018,8 +1017,7 @@ static int nfc_genl_llc_get_params(struct sk_buff *skb, struct genl_info *info) struct sk_buff *msg = NULL; u32 idx; - if (!info->attrs[NFC_ATTR_DEVICE_INDEX] || - !info->attrs[NFC_ATTR_FIRMWARE_NAME]) + if (!info->attrs[NFC_ATTR_DEVICE_INDEX]) return -EINVAL; idx = nla_get_u32(info->attrs[NFC_ATTR_DEVICE_INDEX]); -- cgit v1.2.3 From d13b05962369bade0fb8e2d55af56492f866cd36 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 11 Apr 2024 08:25:29 +0000 Subject: tcp: small optimization when TCP_TW_SYN is processed When TCP_TW_SYN is processed, we perform a lookup to find a listener and jump back in tcp_v6_rcv() and tcp_v4_rcv() Paolo suggested that we do not have to check if the found socket is a TIME_WAIT or NEW_SYN_RECV one. Suggested-by: Paolo Abeni Link: https://lore.kernel.org/netdev/68085c8a84538cacaac991415e4ccc72f45e76c2.camel@redhat.com/ Signed-off-by: Eric Dumazet Cc: Neal Cardwell Link: https://lore.kernel.org/r/20240411082530.907113-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1e650ec71d2f..88c83ac42129 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2205,7 +2205,6 @@ lookup: if (!sk) goto no_tcp_socket; -process: if (sk->sk_state == TCP_TIME_WAIT) goto do_time_wait; @@ -2285,6 +2284,7 @@ process: } } +process: if (static_branch_unlikely(&ip4_min_ttl)) { /* min_ttl can be changed concurrently from do_ip_setsockopt() */ if (unlikely(iph->ttl < READ_ONCE(inet_sk(sk)->min_ttl))) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 3aa9da5c9a66..bb7c3caf4f85 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1794,7 +1794,6 @@ lookup: if (!sk) goto no_tcp_socket; -process: if (sk->sk_state == TCP_TIME_WAIT) goto do_time_wait; @@ -1871,6 +1870,7 @@ process: } } +process: if (static_branch_unlikely(&ip6_min_hopcount)) { /* min_hopcount can be changed concurrently from do_ipv6_setsockopt() */ if (unlikely(hdr->hop_limit < READ_ONCE(tcp_inet6_sk(sk)->min_hopcount))) { -- cgit v1.2.3 From 195b7fc53c6f45feb0f649c7c68af70900928253 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 11 Apr 2024 10:17:04 +0100 Subject: tipc: remove redundant assignment to ret, simplify code Variable err is being assigned a zero value and it is never read afterwards in either the break path or continue path, the assignment is redundant and can be removed. With it removed, the if statement can also be simplified. Cleans up clang scan warning: net/tipc/socket.c:3570:5: warning: Value stored to 'err' is never read [deadcode.DeadStores] Signed-off-by: Colin Ian King Link: https://lore.kernel.org/r/20240411091704.306752-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski --- net/tipc/socket.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'net') diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 7e4135db5816..798397b6811e 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -3565,11 +3565,8 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb, rhashtable_walk_start(iter); while ((tsk = rhashtable_walk_next(iter)) != NULL) { if (IS_ERR(tsk)) { - err = PTR_ERR(tsk); - if (err == -EAGAIN) { - err = 0; + if (PTR_ERR(tsk) == -EAGAIN) continue; - } break; } -- cgit v1.2.3 From 32affa5578f0e6b9abef3623d3976395afbd265c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 11 Apr 2024 13:33:40 +0000 Subject: fib: rules: no longer hold RTNL in fib_nl_dumprule() - fib rules are already RCU protected, RTNL is not needed to get them. - Fix return value at the end of a dump, so that NLMSG_DONE can be appended to current skb, saving one recvmsg() system call. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240411133340.1332796-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/fib_rules.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 3f933ffcefc3..6ebffbc63236 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -1142,10 +1142,10 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb) const struct nlmsghdr *nlh = cb->nlh; struct net *net = sock_net(skb->sk); struct fib_rules_ops *ops; - int idx = 0, family; + int err, idx = 0, family; if (cb->strict_check) { - int err = fib_valid_dumprule_req(nlh, cb->extack); + err = fib_valid_dumprule_req(nlh, cb->extack); if (err < 0) return err; @@ -1158,17 +1158,17 @@ static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb) if (ops == NULL) return -EAFNOSUPPORT; - dump_rules(skb, cb, ops); - - return skb->len; + return dump_rules(skb, cb, ops); } + err = 0; rcu_read_lock(); list_for_each_entry_rcu(ops, &net->rules_ops, list) { if (idx < cb->args[0] || !try_module_get(ops->owner)) goto skip; - if (dump_rules(skb, cb, ops) < 0) + err = dump_rules(skb, cb, ops); + if (err < 0) break; cb->args[1] = 0; @@ -1178,7 +1178,7 @@ skip: rcu_read_unlock(); cb->args[0] = idx; - return skb->len; + return err; } static void notify_rule_change(int event, struct fib_rule *rule, @@ -1293,7 +1293,8 @@ static int __init fib_rules_init(void) int err; rtnl_register(PF_UNSPEC, RTM_NEWRULE, fib_nl_newrule, NULL, 0); rtnl_register(PF_UNSPEC, RTM_DELRULE, fib_nl_delrule, NULL, 0); - rtnl_register(PF_UNSPEC, RTM_GETRULE, NULL, fib_nl_dumprule, 0); + rtnl_register(PF_UNSPEC, RTM_GETRULE, NULL, fib_nl_dumprule, + RTNL_FLAG_DUMP_UNLOCKED); err = register_pernet_subsys(&fib_rules_net_ops); if (err < 0) -- cgit v1.2.3 From f1e197a665c2148ebc25fe09c53689e60afea195 Mon Sep 17 00:00:00 2001 From: Wander Lairson Costa Date: Thu, 11 Apr 2024 11:13:46 -0300 Subject: drop_monitor: replace spin_lock by raw_spin_lock trace_drop_common() is called with preemption disabled, and it acquires a spin_lock. This is problematic for RT kernels because spin_locks are sleeping locks in this configuration, which causes the following splat: BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48 in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 449, name: rcuc/47 preempt_count: 1, expected: 0 RCU nest depth: 2, expected: 2 5 locks held by rcuc/47/449: #0: ff1100086ec30a60 ((softirq_ctrl.lock)){+.+.}-{2:2}, at: __local_bh_disable_ip+0x105/0x210 #1: ffffffffb394a280 (rcu_read_lock){....}-{1:2}, at: rt_spin_lock+0xbf/0x130 #2: ffffffffb394a280 (rcu_read_lock){....}-{1:2}, at: __local_bh_disable_ip+0x11c/0x210 #3: ffffffffb394a160 (rcu_callback){....}-{0:0}, at: rcu_do_batch+0x360/0xc70 #4: ff1100086ee07520 (&data->lock){+.+.}-{2:2}, at: trace_drop_common.constprop.0+0xb5/0x290 irq event stamp: 139909 hardirqs last enabled at (139908): [] _raw_spin_unlock_irqrestore+0x63/0x80 hardirqs last disabled at (139909): [] trace_drop_common.constprop.0+0x26d/0x290 softirqs last enabled at (139892): [] __local_bh_enable_ip+0x103/0x170 softirqs last disabled at (139898): [] rcu_cpu_kthread+0x93/0x1f0 Preemption disabled at: [] rt_mutex_slowunlock+0xab/0x2e0 CPU: 47 PID: 449 Comm: rcuc/47 Not tainted 6.9.0-rc2-rt1+ #7 Hardware name: Dell Inc. PowerEdge R650/0Y2G81, BIOS 1.6.5 04/15/2022 Call Trace: dump_stack_lvl+0x8c/0xd0 dump_stack+0x14/0x20 __might_resched+0x21e/0x2f0 rt_spin_lock+0x5e/0x130 ? trace_drop_common.constprop.0+0xb5/0x290 ? skb_queue_purge_reason.part.0+0x1bf/0x230 trace_drop_common.constprop.0+0xb5/0x290 ? preempt_count_sub+0x1c/0xd0 ? _raw_spin_unlock_irqrestore+0x4a/0x80 ? __pfx_trace_drop_common.constprop.0+0x10/0x10 ? rt_mutex_slowunlock+0x26a/0x2e0 ? skb_queue_purge_reason.part.0+0x1bf/0x230 ? __pfx_rt_mutex_slowunlock+0x10/0x10 ? skb_queue_purge_reason.part.0+0x1bf/0x230 trace_kfree_skb_hit+0x15/0x20 trace_kfree_skb+0xe9/0x150 kfree_skb_reason+0x7b/0x110 skb_queue_purge_reason.part.0+0x1bf/0x230 ? __pfx_skb_queue_purge_reason.part.0+0x10/0x10 ? mark_lock.part.0+0x8a/0x520 ... trace_drop_common() also disables interrupts, but this is a minor issue because we could easily replace it with a local_lock. Replace the spin_lock with raw_spin_lock to avoid sleeping in atomic context. Signed-off-by: Wander Lairson Costa Reported-by: Hu Chunyu Signed-off-by: David S. Miller --- net/core/drop_monitor.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c index b0f221d658be..430ed18f8584 100644 --- a/net/core/drop_monitor.c +++ b/net/core/drop_monitor.c @@ -74,7 +74,7 @@ struct net_dm_hw_entries { }; struct per_cpu_dm_data { - spinlock_t lock; /* Protects 'skb', 'hw_entries' and + raw_spinlock_t lock; /* Protects 'skb', 'hw_entries' and * 'send_timer' */ union { @@ -168,9 +168,9 @@ static struct sk_buff *reset_per_cpu_data(struct per_cpu_dm_data *data) err: mod_timer(&data->send_timer, jiffies + HZ / 10); out: - spin_lock_irqsave(&data->lock, flags); + raw_spin_lock_irqsave(&data->lock, flags); swap(data->skb, skb); - spin_unlock_irqrestore(&data->lock, flags); + raw_spin_unlock_irqrestore(&data->lock, flags); if (skb) { struct nlmsghdr *nlh = (struct nlmsghdr *)skb->data; @@ -225,7 +225,7 @@ static void trace_drop_common(struct sk_buff *skb, void *location) local_irq_save(flags); data = this_cpu_ptr(&dm_cpu_data); - spin_lock(&data->lock); + raw_spin_lock(&data->lock); dskb = data->skb; if (!dskb) @@ -259,7 +259,7 @@ static void trace_drop_common(struct sk_buff *skb, void *location) } out: - spin_unlock_irqrestore(&data->lock, flags); + raw_spin_unlock_irqrestore(&data->lock, flags); } static void trace_kfree_skb_hit(void *ignore, struct sk_buff *skb, @@ -314,9 +314,9 @@ net_dm_hw_reset_per_cpu_data(struct per_cpu_dm_data *hw_data) mod_timer(&hw_data->send_timer, jiffies + HZ / 10); } - spin_lock_irqsave(&hw_data->lock, flags); + raw_spin_lock_irqsave(&hw_data->lock, flags); swap(hw_data->hw_entries, hw_entries); - spin_unlock_irqrestore(&hw_data->lock, flags); + raw_spin_unlock_irqrestore(&hw_data->lock, flags); return hw_entries; } @@ -448,7 +448,7 @@ net_dm_hw_trap_summary_probe(void *ignore, const struct devlink *devlink, return; hw_data = this_cpu_ptr(&dm_hw_cpu_data); - spin_lock_irqsave(&hw_data->lock, flags); + raw_spin_lock_irqsave(&hw_data->lock, flags); hw_entries = hw_data->hw_entries; if (!hw_entries) @@ -477,7 +477,7 @@ net_dm_hw_trap_summary_probe(void *ignore, const struct devlink *devlink, } out: - spin_unlock_irqrestore(&hw_data->lock, flags); + raw_spin_unlock_irqrestore(&hw_data->lock, flags); } static const struct net_dm_alert_ops net_dm_alert_summary_ops = { @@ -1673,7 +1673,7 @@ static struct notifier_block dropmon_net_notifier = { static void __net_dm_cpu_data_init(struct per_cpu_dm_data *data) { - spin_lock_init(&data->lock); + raw_spin_lock_init(&data->lock); skb_queue_head_init(&data->drop_queue); u64_stats_init(&data->stats.syncp); } -- cgit v1.2.3 From 3db3b62955cd6d73afde05a17d7e8e106695c3b9 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 11 Apr 2024 11:32:22 -0700 Subject: net: dev_addr_lists: move locking out of init/exit in kunit We lock and unlock rtnl in init/exit for convenience, but it started causing problems if the exit is handled by a different thread. To avoid having to futz with disabling locking assertions move the locking into the test cases. We don't use ASSERTs so it should be safe. ============= dev-addr-list-test (6 subtests) ============== [PASSED] dev_addr_test_basic [PASSED] dev_addr_test_sync_one [PASSED] dev_addr_test_add_del [PASSED] dev_addr_test_del_main [PASSED] dev_addr_test_add_set [PASSED] dev_addr_test_add_excl =============== [PASSED] dev-addr-list-test ================ Link: https://lore.kernel.org/all/20240403131936.787234-7-linux@roeck-us.net Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- net/core/dev_addr_lists_test.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/dev_addr_lists_test.c b/net/core/dev_addr_lists_test.c index 4dbd0dc6aea2..8e1dba825e94 100644 --- a/net/core/dev_addr_lists_test.c +++ b/net/core/dev_addr_lists_test.c @@ -49,7 +49,6 @@ static int dev_addr_test_init(struct kunit *test) KUNIT_FAIL(test, "Can't register netdev %d", err); } - rtnl_lock(); return 0; } @@ -57,7 +56,6 @@ static void dev_addr_test_exit(struct kunit *test) { struct net_device *netdev = test->priv; - rtnl_unlock(); unregister_netdev(netdev); free_netdev(netdev); } @@ -67,6 +65,7 @@ static void dev_addr_test_basic(struct kunit *test) struct net_device *netdev = test->priv; u8 addr[ETH_ALEN]; + rtnl_lock(); KUNIT_EXPECT_TRUE(test, !!netdev->dev_addr); memset(addr, 2, sizeof(addr)); @@ -76,6 +75,7 @@ static void dev_addr_test_basic(struct kunit *test) memset(addr, 3, sizeof(addr)); dev_addr_set(netdev, addr); KUNIT_EXPECT_MEMEQ(test, netdev->dev_addr, addr, sizeof(addr)); + rtnl_unlock(); } static void dev_addr_test_sync_one(struct kunit *test) @@ -86,6 +86,7 @@ static void dev_addr_test_sync_one(struct kunit *test) datp = netdev_priv(netdev); + rtnl_lock(); memset(addr, 1, sizeof(addr)); eth_hw_addr_set(netdev, addr); @@ -103,6 +104,7 @@ static void dev_addr_test_sync_one(struct kunit *test) * considered synced and we overwrite in place. */ KUNIT_EXPECT_EQ(test, 0, datp->addr_seen); + rtnl_unlock(); } static void dev_addr_test_add_del(struct kunit *test) @@ -114,6 +116,7 @@ static void dev_addr_test_add_del(struct kunit *test) datp = netdev_priv(netdev); + rtnl_lock(); for (i = 1; i < 4; i++) { memset(addr, i, sizeof(addr)); KUNIT_EXPECT_EQ(test, 0, dev_addr_add(netdev, addr, @@ -143,6 +146,7 @@ static void dev_addr_test_add_del(struct kunit *test) __hw_addr_sync_dev(&netdev->dev_addrs, netdev, dev_addr_test_sync, dev_addr_test_unsync); KUNIT_EXPECT_EQ(test, 1, datp->addr_seen); + rtnl_unlock(); } static void dev_addr_test_del_main(struct kunit *test) @@ -150,6 +154,7 @@ static void dev_addr_test_del_main(struct kunit *test) struct net_device *netdev = test->priv; u8 addr[ETH_ALEN]; + rtnl_lock(); memset(addr, 1, sizeof(addr)); eth_hw_addr_set(netdev, addr); @@ -161,6 +166,7 @@ static void dev_addr_test_del_main(struct kunit *test) NETDEV_HW_ADDR_T_LAN)); KUNIT_EXPECT_EQ(test, -ENOENT, dev_addr_del(netdev, addr, NETDEV_HW_ADDR_T_LAN)); + rtnl_unlock(); } static void dev_addr_test_add_set(struct kunit *test) @@ -172,6 +178,7 @@ static void dev_addr_test_add_set(struct kunit *test) datp = netdev_priv(netdev); + rtnl_lock(); /* There is no external API like dev_addr_add_excl(), * so shuffle the tree a little bit and exploit aliasing. */ @@ -191,6 +198,7 @@ static void dev_addr_test_add_set(struct kunit *test) __hw_addr_sync_dev(&netdev->dev_addrs, netdev, dev_addr_test_sync, dev_addr_test_unsync); KUNIT_EXPECT_EQ(test, 0xffff, datp->addr_seen); + rtnl_unlock(); } static void dev_addr_test_add_excl(struct kunit *test) @@ -199,6 +207,7 @@ static void dev_addr_test_add_excl(struct kunit *test) u8 addr[ETH_ALEN]; int i; + rtnl_lock(); for (i = 0; i < 10; i++) { memset(addr, i, sizeof(addr)); KUNIT_EXPECT_EQ(test, 0, dev_uc_add_excl(netdev, addr)); @@ -213,6 +222,7 @@ static void dev_addr_test_add_excl(struct kunit *test) memset(addr, i, sizeof(addr)); KUNIT_EXPECT_EQ(test, -EEXIST, dev_uc_add_excl(netdev, addr)); } + rtnl_unlock(); } static struct kunit_case dev_addr_test_cases[] = { -- cgit v1.2.3 From 4d0470b9ad73e965f5a1e52f1deb0edbb6d03c89 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Fri, 12 Apr 2024 11:07:18 +0800 Subject: net: save some cycles when doing skb_attempt_defer_free() Normally, we don't face these two exceptions very often meanwhile we have some chance to meet the condition where the current cpu id is the same as skb->alloc_cpu. One simple test that can help us see the frequency of this statement 'cpu == raw_smp_processor_id()': 1. running iperf -s and iperf -c [ip] -P [MAX CPU] 2. using BPF to capture skb_attempt_defer_free() I can see around 4% chance that happens to satisfy the statement. So moving this statement at the beginning can save some cycles in most cases. Signed-off-by: Jason Xing Reviewed-by: Alexander Lobakin Signed-off-by: David S. Miller --- net/core/skbuff.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index ea052fa710d8..37c858dc11a6 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6965,9 +6965,9 @@ void skb_attempt_defer_free(struct sk_buff *skb) unsigned int defer_max; bool kick; - if (WARN_ON_ONCE(cpu >= nr_cpu_ids) || - !cpu_online(cpu) || - cpu == raw_smp_processor_id()) { + if (cpu == raw_smp_processor_id() || + WARN_ON_ONCE(cpu >= nr_cpu_ids) || + !cpu_online(cpu)) { nodefer: kfree_skb_napi_cache(skb); return; } -- cgit v1.2.3 From 9382b4f338d261494aad7eeffef0b13ff663b542 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Fri, 12 Apr 2024 12:17:57 +0200 Subject: net: constify net_class AFAICS all users of net_class take a const struct class * argument. Therefore fully constify net_class. Signed-off-by: Heiner Kallweit Acked-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/core/net-sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index ff3ee45be64a..1f7f09e56771 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -2046,7 +2046,7 @@ static void net_get_ownership(const struct device *d, kuid_t *uid, kgid_t *gid) net_ns_get_ownership(net, uid, gid); } -static struct class net_class __ro_after_init = { +static const struct class net_class = { .name = "net", .dev_release = netdev_release, .dev_groups = net_class_groups, -- cgit v1.2.3 From a788fafff56f940dbb1753c5bbbff387f9b97619 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Fri, 12 Apr 2024 16:15:03 +0100 Subject: net: dsa: convert dsa_user_phylink_fixed_state() to use dsa_phylink_to_port() Convert dsa_user_phylink_fixed_state() to use the newly introduced dsa_phylink_to_port() helper. Suggested-by: Vladimir Oltean Signed-off-by: Russell King (Oracle) Signed-off-by: David S. Miller --- net/dsa/user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/dsa/user.c b/net/dsa/user.c index 16d395bb1a1f..c94b868855aa 100644 --- a/net/dsa/user.c +++ b/net/dsa/user.c @@ -2445,7 +2445,7 @@ EXPORT_SYMBOL_GPL(dsa_port_phylink_mac_change); static void dsa_user_phylink_fixed_state(struct phylink_config *config, struct phylink_link_state *state) { - struct dsa_port *dp = container_of(config, struct dsa_port, pl_config); + struct dsa_port *dp = dsa_phylink_to_port(config); struct dsa_switch *ds = dp->ds; /* No need to check that this operation is valid, the callback would -- cgit v1.2.3 From 8622f90a371b65a8e454ca3d1fe95aeed1bf1da4 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 12 Apr 2024 08:19:25 -0700 Subject: net: ipv6_gre: Do not use custom stat allocator With commit 34d21de99cea9 ("net: Move {l,t,d}stats allocation to core and convert veth & vrf"), stats allocation could be done on net core instead of in this driver. With this new approach, the driver doesn't have to bother with error handling (allocation failure checking, making sure free happens in the right spot, etc). This is core responsibility now. Remove the allocation in the ip6_gre and leverage the network core allocation instead. Signed-off-by: Breno Leitao Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 22 +++++----------------- 1 file changed, 5 insertions(+), 17 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 95fbdaf85711..b5b417902c0a 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1444,7 +1444,6 @@ static void ip6gre_dev_free(struct net_device *dev) gro_cells_destroy(&t->gro_cells); dst_cache_destroy(&t->dst_cache); - free_percpu(dev->tstats); } static void ip6gre_tunnel_setup(struct net_device *dev) @@ -1453,6 +1452,7 @@ static void ip6gre_tunnel_setup(struct net_device *dev) dev->needs_free_netdev = true; dev->priv_destructor = ip6gre_dev_free; + dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS; dev->type = ARPHRD_IP6GRE; dev->flags |= IFF_NOARP; @@ -1500,13 +1500,9 @@ static int ip6gre_tunnel_init_common(struct net_device *dev) tunnel->net = dev_net(dev); strcpy(tunnel->parms.name, dev->name); - dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; - ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL); if (ret) - goto cleanup_alloc_pcpu_stats; + return ret; ret = gro_cells_init(&tunnel->gro_cells, dev); if (ret) @@ -1530,9 +1526,6 @@ static int ip6gre_tunnel_init_common(struct net_device *dev) cleanup_dst_cache_init: dst_cache_destroy(&tunnel->dst_cache); -cleanup_alloc_pcpu_stats: - free_percpu(dev->tstats); - dev->tstats = NULL; return ret; } @@ -1893,13 +1886,9 @@ static int ip6erspan_tap_init(struct net_device *dev) tunnel->net = dev_net(dev); strcpy(tunnel->parms.name, dev->name); - dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; - ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL); if (ret) - goto cleanup_alloc_pcpu_stats; + return ret; ret = gro_cells_init(&tunnel->gro_cells, dev); if (ret) @@ -1921,9 +1910,6 @@ static int ip6erspan_tap_init(struct net_device *dev) cleanup_dst_cache_init: dst_cache_destroy(&tunnel->dst_cache); -cleanup_alloc_pcpu_stats: - free_percpu(dev->tstats); - dev->tstats = NULL; return ret; } @@ -1948,6 +1934,7 @@ static void ip6gre_tap_setup(struct net_device *dev) dev->needs_free_netdev = true; dev->priv_destructor = ip6gre_dev_free; + dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS; dev->priv_flags &= ~IFF_TX_SKB_SHARING; dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; netif_keep_dst(dev); @@ -2250,6 +2237,7 @@ static void ip6erspan_tap_setup(struct net_device *dev) dev->needs_free_netdev = true; dev->priv_destructor = ip6gre_dev_free; + dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS; dev->priv_flags &= ~IFF_TX_SKB_SHARING; dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; netif_keep_dst(dev); -- cgit v1.2.3 From 05d604a57773a499b158e5fe84108301228392ec Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Fri, 12 Apr 2024 08:19:26 -0700 Subject: net: ip6_gre: Remove generic .ndo_get_stats64 Commit 3e2f544dd8a33 ("net: get stats64 if device if driver is configured") moved the callback to dev_get_tstats64() to net core, so, unless the driver is doing some custom stats collection, it does not need to set .ndo_get_stats64. Since this driver is now relying in NETDEV_PCPU_STAT_TSTATS, then, it doesn't need to set the dev_get_tstats64() generic .ndo_get_stats64 function pointer. Signed-off-by: Breno Leitao Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/ip6_gre.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index b5b417902c0a..3942bd2ade78 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1434,7 +1434,6 @@ static const struct net_device_ops ip6gre_netdev_ops = { .ndo_start_xmit = ip6gre_tunnel_xmit, .ndo_siocdevprivate = ip6gre_tunnel_siocdevprivate, .ndo_change_mtu = ip6_tnl_change_mtu, - .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip6_tnl_get_iflink, }; @@ -1857,7 +1856,6 @@ static const struct net_device_ops ip6gre_tap_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = ip6_tnl_change_mtu, - .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip6_tnl_get_iflink, }; @@ -1920,7 +1918,6 @@ static const struct net_device_ops ip6erspan_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = ip6_tnl_change_mtu, - .ndo_get_stats64 = dev_get_tstats64, .ndo_get_iflink = ip6_tnl_get_iflink, }; -- cgit v1.2.3 From 50aee97d15113b95a68848db1f0cb2a6c09f753a Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Fri, 12 Apr 2024 17:20:04 -0400 Subject: udp: Avoid call to compute_score on multiple sites We've observed a 7-12% performance regression in iperf3 UDP ipv4 and ipv6 tests with multiple sockets on Zen3 cpus, which we traced back to commit f0ea27e7bfe1 ("udp: re-score reuseport groups when connected sockets are present"). The failing tests were those that would spawn UDP sockets per-cpu on systems that have a high number of cpus. Unsurprisingly, it is not caused by the extra re-scoring of the reused socket, but due to the compiler no longer inlining compute_score, once it has the extra call site in udp4_lib_lookup2. This is augmented by the "Safe RET" mitigation for SRSO, needed in our Zen3 cpus. We could just explicitly inline it, but compute_score() is quite a large function, around 300b. Inlining in two sites would almost double udp4_lib_lookup2, which is a silly thing to do just to workaround a mitigation. Instead, this patch shuffles the code a bit to avoid the multiple calls to compute_score. Since it is a static function used in one spot, the compiler can safely fold it in, as it did before, without increasing the text size. With this patch applied I ran my original iperf3 testcases. The failing cases all looked like this (ipv4): iperf3 -c 127.0.0.1 --udp -4 -f K -b $R -l 8920 -t 30 -i 5 -P 64 -O 2 where $R is either 1G/10G/0 (max, unlimited). I ran 3 times each. baseline is v6.9-rc3. harmean == harmonic mean; CV == coefficient of variation. ipv4: 1G 10G MAX HARMEAN (CV) HARMEAN (CV) HARMEAN (CV) baseline 1743852.66(0.0208) 1725933.02(0.0167) 1705203.78(0.0386) patched 1968727.61(0.0035) 1962283.22(0.0195) 1923853.50(0.0256) ipv6: 1G 10G MAX HARMEAN (CV) HARMEAN (CV) HARMEAN (CV) baseline 1729020.03(0.0028) 1691704.49(0.0243) 1692251.34(0.0083) patched 1900422.19(0.0067) 1900968.01(0.0067) 1568532.72(0.1519) This restores the performance we had before the change above with this benchmark. We obviously don't expect any real impact when mitigations are disabled, but just to be sure it also doesn't regresses: mitigations=off ipv4: 1G 10G MAX HARMEAN (CV) HARMEAN (CV) HARMEAN (CV) baseline 3230279.97(0.0066) 3229320.91(0.0060) 2605693.19(0.0697) patched 3242802.36(0.0073) 3239310.71(0.0035) 2502427.19(0.0882) Cc: Lorenz Bauer Fixes: f0ea27e7bfe1 ("udp: re-score reuseport groups when connected sockets are present") Signed-off-by: Gabriel Krisman Bertazi Reviewed-by: Kuniyuki Iwashima Reviewed-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/ipv4/udp.c | 21 ++++++++++++++++----- net/ipv6/udp.c | 20 ++++++++++++++++---- 2 files changed, 32 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7613daa339b0..944922172c98 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -427,15 +427,21 @@ static struct sock *udp4_lib_lookup2(struct net *net, { struct sock *sk, *result; int score, badness; + bool need_rescore; result = NULL; badness = 0; udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { - score = compute_score(sk, net, saddr, sport, - daddr, hnum, dif, sdif); + need_rescore = false; +rescore: + score = compute_score(need_rescore ? result : sk, net, saddr, + sport, daddr, hnum, dif, sdif); if (score > badness) { badness = score; + if (need_rescore) + continue; + if (sk->sk_state == TCP_ESTABLISHED) { result = sk; continue; @@ -456,9 +462,14 @@ static struct sock *udp4_lib_lookup2(struct net *net, if (IS_ERR(result)) continue; - badness = compute_score(result, net, saddr, sport, - daddr, hnum, dif, sdif); - + /* compute_score is too long of a function to be + * inlined, and calling it again here yields + * measureable overhead for some + * workloads. Work around it by jumping + * backwards to rescore 'result'. + */ + need_rescore = true; + goto rescore; } } return result; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index a9466f881f1b..9914e73f4785 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -169,15 +169,21 @@ static struct sock *udp6_lib_lookup2(struct net *net, { struct sock *sk, *result; int score, badness; + bool need_rescore; result = NULL; badness = -1; udp_portaddr_for_each_entry_rcu(sk, &hslot2->head) { - score = compute_score(sk, net, saddr, sport, - daddr, hnum, dif, sdif); + need_rescore = false; +rescore: + score = compute_score(need_rescore ? result : sk, net, saddr, + sport, daddr, hnum, dif, sdif); if (score > badness) { badness = score; + if (need_rescore) + continue; + if (sk->sk_state == TCP_ESTABLISHED) { result = sk; continue; @@ -198,8 +204,14 @@ static struct sock *udp6_lib_lookup2(struct net *net, if (IS_ERR(result)) continue; - badness = compute_score(sk, net, saddr, sport, - daddr, hnum, dif, sdif); + /* compute_score is too long of a function to be + * inlined, and calling it again here yields + * measureable overhead for some + * workloads. Work around it by jumping + * backwards to rescore 'result'. + */ + need_rescore = true; + goto rescore; } } return result; -- cgit v1.2.3 From 8c9a6f549e65912825e31dc1e0e3f7995984649d Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Tue, 9 Apr 2024 14:05:53 -0700 Subject: io_uring: separate header for exported net bits We're exporting some io_uring bits to networking, e.g. for implementing a net callback for io_uring cmds, but we don't want to expose more than needed. Add a separate header for networking. Signed-off-by: Pavel Begunkov Signed-off-by: David Wei Link: https://lore.kernel.org/r/20240409210554.1878789-1-dw@davidwei.uk Signed-off-by: Jens Axboe --- include/linux/io_uring.h | 6 ------ include/linux/io_uring/net.h | 18 ++++++++++++++++++ io_uring/uring_cmd.c | 1 + net/socket.c | 2 +- 4 files changed, 20 insertions(+), 7 deletions(-) create mode 100644 include/linux/io_uring/net.h (limited to 'net') diff --git a/include/linux/io_uring.h b/include/linux/io_uring.h index 68ed6697fece..e123d5e17b52 100644 --- a/include/linux/io_uring.h +++ b/include/linux/io_uring.h @@ -11,7 +11,6 @@ void __io_uring_cancel(bool cancel_all); void __io_uring_free(struct task_struct *tsk); void io_uring_unreg_ringfd(void); const char *io_uring_get_opcode(u8 opcode); -int io_uring_cmd_sock(struct io_uring_cmd *cmd, unsigned int issue_flags); bool io_is_uring_fops(struct file *file); static inline void io_uring_files_cancel(void) @@ -45,11 +44,6 @@ static inline const char *io_uring_get_opcode(u8 opcode) { return ""; } -static inline int io_uring_cmd_sock(struct io_uring_cmd *cmd, - unsigned int issue_flags) -{ - return -EOPNOTSUPP; -} static inline bool io_is_uring_fops(struct file *file) { return false; diff --git a/include/linux/io_uring/net.h b/include/linux/io_uring/net.h new file mode 100644 index 000000000000..b58f39fed4d5 --- /dev/null +++ b/include/linux/io_uring/net.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef _LINUX_IO_URING_NET_H +#define _LINUX_IO_URING_NET_H + +struct io_uring_cmd; + +#if defined(CONFIG_IO_URING) +int io_uring_cmd_sock(struct io_uring_cmd *cmd, unsigned int issue_flags); + +#else +static inline int io_uring_cmd_sock(struct io_uring_cmd *cmd, + unsigned int issue_flags) +{ + return -EOPNOTSUPP; +} +#endif + +#endif diff --git a/io_uring/uring_cmd.c b/io_uring/uring_cmd.c index 334d31dd6628..21ac5fb2d5f0 100644 --- a/io_uring/uring_cmd.c +++ b/io_uring/uring_cmd.c @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include diff --git a/net/socket.c b/net/socket.c index e5f3af49a8b6..01a71ae10c35 100644 --- a/net/socket.c +++ b/net/socket.c @@ -88,7 +88,7 @@ #include #include #include -#include +#include #include #include -- cgit v1.2.3 From fd86344823b521149bb31d91eba900ba3525efa6 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 12 Apr 2024 19:19:28 -0700 Subject: af_unix: Try not to hold unix_gc_lock during accept(). Commit dcf70df2048d ("af_unix: Fix up unix_edge.successor for embryo socket.") added spin_lock(&unix_gc_lock) in accept() path, and it caused regression in a stress test as reported by kernel test robot. If the embryo socket is not part of the inflight graph, we need not hold the lock. To decide that in O(1) time and avoid the regression in the normal use case, 1. add a new stat unix_sk(sk)->scm_stat.nr_unix_fds 2. count the number of inflight AF_UNIX sockets in the receive queue under unix_state_lock() 3. move unix_update_edges() call under unix_state_lock() 4. avoid locking if nr_unix_fds is 0 in unix_update_edges() Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202404101427.92a08551-oliver.sang@intel.com Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240413021928.20946-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni --- include/net/af_unix.h | 1 + net/unix/af_unix.c | 2 +- net/unix/garbage.c | 20 ++++++++++++++++---- 3 files changed, 18 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 7311b77edfc7..872ff2a50372 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -67,6 +67,7 @@ struct unix_skb_parms { struct scm_stat { atomic_t nr_fds; + unsigned long nr_unix_fds; }; #define UNIXCB(skb) (*(struct unix_skb_parms *)&((skb)->cb)) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 142d210b5b03..ed16d5f66df8 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1719,12 +1719,12 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, } tsk = skb->sk; - unix_update_edges(unix_sk(tsk)); skb_free_datagram(sk, skb); wake_up_interruptible(&unix_sk(sk)->peer_wait); /* attach accepted sock to socket */ unix_state_lock(tsk); + unix_update_edges(unix_sk(tsk)); newsock->state = SS_CONNECTED; unix_sock_inherit_flags(sock, newsock); sock_graft(tsk, newsock); diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 12a4ec27e0d4..95240a59808f 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -209,6 +209,7 @@ void unix_add_edges(struct scm_fp_list *fpl, struct unix_sock *receiver) unix_add_edge(fpl, edge); } while (i < fpl->count_unix); + receiver->scm_stat.nr_unix_fds += fpl->count_unix; WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + fpl->count_unix); out: WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight + fpl->count); @@ -222,6 +223,7 @@ out: void unix_del_edges(struct scm_fp_list *fpl) { + struct unix_sock *receiver; int i = 0; spin_lock(&unix_gc_lock); @@ -235,6 +237,8 @@ void unix_del_edges(struct scm_fp_list *fpl) unix_del_edge(fpl, edge); } while (i < fpl->count_unix); + receiver = fpl->edges[0].successor; + receiver->scm_stat.nr_unix_fds -= fpl->count_unix; WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - fpl->count_unix); out: WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight - fpl->count); @@ -246,10 +250,18 @@ out: void unix_update_edges(struct unix_sock *receiver) { - spin_lock(&unix_gc_lock); - unix_update_graph(unix_sk(receiver->listener)->vertex); - receiver->listener = NULL; - spin_unlock(&unix_gc_lock); + /* nr_unix_fds is only updated under unix_state_lock(). + * If it's 0 here, the embryo socket is not part of the + * inflight graph, and GC will not see it, so no lock needed. + */ + if (!receiver->scm_stat.nr_unix_fds) { + receiver->listener = NULL; + } else { + spin_lock(&unix_gc_lock); + unix_update_graph(unix_sk(receiver->listener)->vertex); + receiver->listener = NULL; + spin_unlock(&unix_gc_lock); + } } int unix_prepare_fpl(struct scm_fp_list *fpl) -- cgit v1.2.3 From c2b640529ec7595b4f82f1630da1ac9697a0c927 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Mon, 15 Apr 2024 11:07:13 +0100 Subject: net/handshake: remove redundant assignment to variable ret The variable is being assigned an value and then is being re-assigned a new value in the next statement. The assignment is redundant and can be removed. Cleans up clang scan build warning: net/handshake/tlshd.c:216:2: warning: Value stored to 'ret' is never read [deadcode.DeadStores] Signed-off-by: Colin Ian King Reviewed-by: Jason Xing Acked-by: Chuck Lever Link: https://lore.kernel.org/r/20240415100713.483399-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski --- net/handshake/tlshd.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index d697f68c598c..d6f52839827e 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -213,7 +213,6 @@ static int tls_handshake_accept(struct handshake_req *req, if (!hdr) goto out_cancel; - ret = -EMSGSIZE; ret = nla_put_s32(msg, HANDSHAKE_A_ACCEPT_SOCKFD, fd); if (ret < 0) goto out_cancel; -- cgit v1.2.3 From 86600ea11dc18ecd10110c13148d0fb04a80ceea Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 15 Apr 2024 14:23:44 +0200 Subject: ip6_vti: fix memleak on netns dismantle kmemleak reports net_device resources are no longer released, restore needs_free_netdev toggle. Sample backtrace: unreferenced object 0xffff88810874f000 (size 4096): [..] [<00000000a2b8af8b>] __kmalloc_node+0x209/0x290 [<0000000040b0a1a9>] alloc_netdev_mqs+0x58/0x470 [<00000000b4be1e78>] vti6_init_net+0x94/0x230 [<000000008830c1ea>] ops_init+0x32/0xc0 [<000000006a26fa8f>] setup_net+0x134/0x2e0 [..] Cc: Breno Leitao Fixes: a9b2d55a8f1e ("ip6_vti: Do not use custom stat allocator") Signed-off-by: Florian Westphal Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240415122346.26503-1-fw@strlen.de Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_vti.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 4d68a0777b0c..78344cf3867e 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -901,6 +901,7 @@ static void vti6_dev_setup(struct net_device *dev) { dev->netdev_ops = &vti6_netdev_ops; dev->header_ops = &ip_tunnel_header_ops; + dev->needs_free_netdev = true; dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS; dev->type = ARPHRD_TUNNEL6; -- cgit v1.2.3 From b010bf72da5ada39f4e722fa28d4275df8673229 Mon Sep 17 00:00:00 2001 From: "Kory Maincent (Dent Project)" Date: Sun, 14 Apr 2024 16:21:52 +0200 Subject: net: pse-pd: Rectify and adapt the naming of admin_cotrol member of struct pse_control_config In commit 18ff0bcda6d1 ("ethtool: add interface to interact with Ethernet Power Equipment"), the 'pse_control_config' structure was introduced, housing a single member labeled 'admin_cotrol' responsible for maintaining the operational state of the PoDL PSE functions. A noticeable typographical error exists in the naming of this field ('cotrol' should be corrected to 'control'), which this commit aims to rectify. Furthermore, with upcoming extensions of this structure to encompass PoE functionalities, the field is being renamed to 'podl_admin_state' to distinctly indicate that this state is tailored specifically for PoDL." Reviewed-by: Oleksij Rempel Reviewed-by: Andrew Lunn Signed-off-by: Kory Maincent Link: https://lore.kernel.org/r/20240414-feature_poe-v8-3-e4bf1e860da5@bootlin.com Signed-off-by: Jakub Kicinski --- drivers/net/pse-pd/pse_regulator.c | 8 ++++---- include/linux/pse-pd/pse.h | 4 ++-- net/ethtool/pse-pd.c | 2 +- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/drivers/net/pse-pd/pse_regulator.c b/drivers/net/pse-pd/pse_regulator.c index e2bf8306ca90..1dedf4de296e 100644 --- a/drivers/net/pse-pd/pse_regulator.c +++ b/drivers/net/pse-pd/pse_regulator.c @@ -31,10 +31,10 @@ pse_reg_ethtool_set_config(struct pse_controller_dev *pcdev, unsigned long id, struct pse_reg_priv *priv = to_pse_reg(pcdev); int ret; - if (priv->admin_state == config->admin_cotrol) + if (priv->admin_state == config->podl_admin_control) return 0; - switch (config->admin_cotrol) { + switch (config->podl_admin_control) { case ETHTOOL_PODL_PSE_ADMIN_STATE_ENABLED: ret = regulator_enable(priv->ps); break; @@ -43,14 +43,14 @@ pse_reg_ethtool_set_config(struct pse_controller_dev *pcdev, unsigned long id, break; default: dev_err(pcdev->dev, "Unknown admin state %i\n", - config->admin_cotrol); + config->podl_admin_control); ret = -ENOTSUPP; } if (ret) return ret; - priv->admin_state = config->admin_cotrol; + priv->admin_state = config->podl_admin_control; return 0; } diff --git a/include/linux/pse-pd/pse.h b/include/linux/pse-pd/pse.h index fb724c65c77b..199cf4ae3cf2 100644 --- a/include/linux/pse-pd/pse.h +++ b/include/linux/pse-pd/pse.h @@ -15,11 +15,11 @@ struct pse_controller_dev; /** * struct pse_control_config - PSE control/channel configuration. * - * @admin_cotrol: set PoDL PSE admin control as described in + * @podl_admin_control: set PoDL PSE admin control as described in * IEEE 802.3-2018 30.15.1.2.1 acPoDLPSEAdminControl */ struct pse_control_config { - enum ethtool_podl_pse_admin_state admin_cotrol; + enum ethtool_podl_pse_admin_state podl_admin_control; }; /** diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c index cc478af77111..aef57a058f0d 100644 --- a/net/ethtool/pse-pd.c +++ b/net/ethtool/pse-pd.c @@ -130,7 +130,7 @@ ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info) struct phy_device *phydev; /* this values are already validated by the ethnl_pse_set_policy */ - config.admin_cotrol = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]); + config.podl_admin_control = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]); phydev = dev->phydev; if (!phydev) { -- cgit v1.2.3 From 2bd99aef1b19e6da09eff692bc0a09d61d785782 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 16 Apr 2024 09:50:54 +0000 Subject: tcp: accept bare FIN packets under memory pressure Andrew Oates reported that some macOS hosts could repeatedly send FIN packets even if the remote peer drops them and send back DUP ACK RWIN 0 packets. 20:27:16.968254 gif0 In IP macos > victim: Flags [SEW], seq 1950399762, win 65535, options [mss 1460,nop,wscale 6,nop,nop,TS val 501897188 ecr 0,sackOK,eol], length 0 20:27:16.968339 gif0 Out IP victim > macos: Flags [S.E], seq 2995489058, ack 1950399763, win 1448, options [mss 1460,sackOK,TS val 3829877593 ecr 501897188,nop,wscale 0], length 0 20:27:16.968833 gif0 In IP macos > victim: Flags [.], ack 1, win 2058, options [nop,nop,TS val 501897188 ecr 3829877593], length 0 20:27:16.968885 gif0 In IP macos > victim: Flags [P.], seq 1:1449, ack 1, win 2058, options [nop,nop,TS val 501897188 ecr 3829877593], length 1448 20:27:16.968896 gif0 Out IP victim > macos: Flags [.], ack 1449, win 0, options [nop,nop,TS val 3829877593 ecr 501897188], length 0 20:27:19.454593 gif0 In IP macos > victim: Flags [F.], seq 1449, ack 1, win 2058, options [nop,nop,TS val 501899674 ecr 3829877593], length 0 20:27:19.454675 gif0 Out IP victim > macos: Flags [.], ack 1449, win 0, options [nop,nop,TS val 3829880079 ecr 501899674], length 0 20:27:19.455116 gif0 In IP macos > victim: Flags [F.], seq 1449, ack 1, win 2058, options [nop,nop,TS val 501899674 ecr 3829880079], length 0 The retransmits/dup-ACKs then repeat in a tight loop. RFC 9293 3.4. Sequence Numbers states : Note that when the receive window is zero no segments should be acceptable except ACK segments. Thus, it is be possible for a TCP to maintain a zero receive window while transmitting data and receiving ACKs. However, even when the receive window is zero, a TCP must process the RST and URG fields of all incoming segments. Even if we could consider a bare FIN.ACK packet to be an ACK in RFC terms, the retransmits should use exponential backoff. Accepting the FIN in linux does not add extra memory costs, because the FIN flag will simply be merged to the tail skb in the receive queue, and incoming packet is freed. Reported-by: Andrew Oates Signed-off-by: Eric Dumazet Cc: Neal Cardwell Cc: Christoph Paasch Cc: Vidhi Goel Acked-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5a45a0923a1f..384fa5e2f065 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5174,6 +5174,16 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) */ if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) { if (tcp_receive_window(tp) == 0) { + /* Some stacks are known to send bare FIN packets + * in a loop even if we send RWIN 0 in our ACK. + * Accepting this FIN does not hurt memory pressure + * because the FIN flag will simply be merged to the + * receive queue tail skb in most cases. + */ + if (!skb->len && + (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)) + goto queue_and_out; + reason = SKB_DROP_REASON_TCP_ZEROWINDOW; NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPZEROWINDOWDROP); goto out_of_window; @@ -5188,7 +5198,7 @@ queue_and_out: inet_csk_schedule_ack(sk); sk->sk_data_ready(sk); - if (skb_queue_len(&sk->sk_receive_queue)) { + if (skb_queue_len(&sk->sk_receive_queue) && skb->len) { reason = SKB_DROP_REASON_PROTO_MEM; NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRCVQDROP); goto drop; -- cgit v1.2.3 From 1514b06aff1681a418089060f0ebe936c27481bc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 16 Apr 2024 14:07:39 +0000 Subject: netns: no longer hold RTNL in rtnl_net_dumpid() - rtnl_net_dumpid() is already fully RCU protected, RTNL is not needed there. - Fix return value at the end of a dump, so that NLMSG_DONE can be appended to current skb, saving one recvmsg() system call. Signed-off-by: Eric Dumazet Reviewed-by: Guillaume Nault Link: https://lore.kernel.org/r/20240416140739.967941-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/net_namespace.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index f0540c557515..2f5190aa2f15 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -1090,7 +1090,7 @@ static int rtnl_net_dumpid(struct sk_buff *skb, struct netlink_callback *cb) end: if (net_cb.fillargs.add_ref) put_net(net_cb.tgt_net); - return err < 0 ? err : skb->len; + return err; } static void rtnl_net_notifyid(struct net *net, int cmd, int id, u32 portid, @@ -1205,7 +1205,8 @@ void __init net_ns_init(void) rtnl_register(PF_UNSPEC, RTM_NEWNSID, rtnl_net_newid, NULL, RTNL_FLAG_DOIT_UNLOCKED); rtnl_register(PF_UNSPEC, RTM_GETNSID, rtnl_net_getid, rtnl_net_dumpid, - RTNL_FLAG_DOIT_UNLOCKED); + RTNL_FLAG_DOIT_UNLOCKED | + RTNL_FLAG_DUMP_UNLOCKED); } static void free_exit_list(struct pernet_operations *ops, struct list_head *net_exit_list) -- cgit v1.2.3 From ade1c9cc404a15057b62888352e57fd5f54c5331 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 16 Apr 2024 16:11:12 +0000 Subject: tcp_metrics: fix tcp_metrics_nl_dump() return value Change tcp_metrics_nl_dump() to return 0 at the end of a dump so that NLMSG_DONE can be appended to the current skb, saving one recvmsg() system call. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240416161112.1199265-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_metrics.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index c2a925538542..301881eb23f3 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -766,6 +766,7 @@ static int tcp_metrics_nl_dump(struct sk_buff *skb, unsigned int max_rows = 1U << tcp_metrics_hash_log; unsigned int row, s_row = cb->args[0]; int s_col = cb->args[1], col = s_col; + int res = 0; for (row = s_row; row < max_rows; row++, s_col = 0) { struct tcp_metrics_block *tm; @@ -778,7 +779,8 @@ static int tcp_metrics_nl_dump(struct sk_buff *skb, continue; if (col < s_col) continue; - if (tcp_metrics_dump_info(skb, cb, tm) < 0) { + res = tcp_metrics_dump_info(skb, cb, tm); + if (res < 0) { rcu_read_unlock(); goto done; } @@ -789,7 +791,7 @@ static int tcp_metrics_nl_dump(struct sk_buff *skb, done: cb->args[0] = row; cb->args[1] = col; - return skb->len; + return res; } static int __parse_nl_addr(struct genl_info *info, struct inetpeer_addr *addr, -- cgit v1.2.3 From ba3de6d8035ef1efdd27e9a5047412f7cf3fb3a6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 16 Apr 2024 16:20:25 +0000 Subject: tcp_metrics: use parallel_ops for tcp_metrics_nl_family TCP_METRICS_CMD_GET and TCP_METRICS_CMD_DEL use their own locking (tcp_metrics_lock and RCU), they do not need genl_mutex protection. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240416162025.1251547-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_metrics.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index 301881eb23f3..e93df98de3f4 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -988,6 +988,7 @@ static struct genl_family tcp_metrics_nl_family __ro_after_init = { .maxattr = TCP_METRICS_ATTR_MAX, .policy = tcp_metrics_nl_policy, .netnsok = true, + .parallel_ops = true, .module = THIS_MODULE, .small_ops = tcp_metrics_nl_ops, .n_small_ops = ARRAY_SIZE(tcp_metrics_nl_ops), -- cgit v1.2.3 From aeb48a428d7dbe636203ae892e981bcc3e2ac042 Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Fri, 12 Apr 2024 13:50:44 +0200 Subject: udpencap: Remove Obsolete UDP_ENCAP_ESPINUDP_NON_IKE Support The UDP_ENCAP_ESPINUDP_NON_IKE mode, introduced into the Linux kernel in 2004 [2], has remained inactive and obsolete for an extended period. This mode was originally defined in an early version of an IETF draft [1] from 2001. By the time it was integrated into the kernel in 2004 [2], it had already been replaced by UDP_ENCAP_ESPINUDP [3] in later versions of draft-ietf-ipsec-udp-encaps, particularly in version 06. Over time, UDP_ENCAP_ESPINUDP_NON_IKE has lost its relevance, with no known use cases. With this commit, we remove support for UDP_ENCAP_ESPINUDP_NON_IKE, simplifying the codebase and eliminating unnecessary complexity. Kernel will return an error -ENOPROTOOPT if the userspace tries to set this option. References: [1] https://datatracker.ietf.org/doc/html/draft-ietf-ipsec-udp-encaps-00.txt [2] Commit that added UDP_ENCAP_ESPINUDP_NON_IKE to the Linux historic repository. Author: Andreas Gruenbacher Date: Fri Apr 9 01:47:47 2004 -0700 [IPSEC]: Support draft-ietf-ipsec-udp-encaps-00/01, some ipec impls need it. [3] Commit that added UDP_ENCAP_ESPINUDP to the Linux historic repository. Author: Derek Atkins Date: Wed Apr 2 13:21:02 2003 -0800 [IPSEC]: Implement UDP Encapsulation framework. Signed-off-by: Antony Antony Signed-off-by: Steffen Klassert --- include/uapi/linux/udp.h | 2 +- net/ipv4/esp4.c | 12 ------------ net/ipv4/udp.c | 2 -- net/ipv4/xfrm4_input.c | 13 ------------- net/ipv6/esp6.c | 12 ------------ net/ipv6/xfrm6_input.c | 13 ------------- 6 files changed, 1 insertion(+), 53 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/udp.h b/include/uapi/linux/udp.h index 4828794efcf8..1a0fe8b151fb 100644 --- a/include/uapi/linux/udp.h +++ b/include/uapi/linux/udp.h @@ -36,7 +36,7 @@ struct udphdr { #define UDP_GRO 104 /* This socket can receive UDP GRO packets */ /* UDP encapsulation types */ -#define UDP_ENCAP_ESPINUDP_NON_IKE 1 /* draft-ietf-ipsec-nat-t-ike-00/01 */ +#define UDP_ENCAP_ESPINUDP_NON_IKE 1 /* unused draft-ietf-ipsec-nat-t-ike-00/01 */ #define UDP_ENCAP_ESPINUDP 2 /* draft-ietf-ipsec-udp-encaps-06 */ #define UDP_ENCAP_L2TPINUDP 3 /* rfc2661 */ #define UDP_ENCAP_GTP0 4 /* GSM TS 09.60 */ diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 3d647c9a7a21..7d38ddd64115 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -347,7 +347,6 @@ static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb, __be16 dport) { struct udphdr *uh; - __be32 *udpdata32; unsigned int len; len = skb->len + esp->tailen - skb_transport_offset(skb); @@ -362,12 +361,6 @@ static struct ip_esp_hdr *esp_output_udp_encap(struct sk_buff *skb, *skb_mac_header(skb) = IPPROTO_UDP; - if (encap_type == UDP_ENCAP_ESPINUDP_NON_IKE) { - udpdata32 = (__be32 *)(uh + 1); - udpdata32[0] = udpdata32[1] = 0; - return (struct ip_esp_hdr *)(udpdata32 + 2); - } - return (struct ip_esp_hdr *)(uh + 1); } @@ -423,7 +416,6 @@ static int esp_output_encap(struct xfrm_state *x, struct sk_buff *skb, switch (encap_type) { default: case UDP_ENCAP_ESPINUDP: - case UDP_ENCAP_ESPINUDP_NON_IKE: esph = esp_output_udp_encap(skb, encap_type, esp, sport, dport); break; case TCP_ENCAP_ESPINTCP: @@ -775,7 +767,6 @@ int esp_input_done2(struct sk_buff *skb, int err) source = th->source; break; case UDP_ENCAP_ESPINUDP: - case UDP_ENCAP_ESPINUDP_NON_IKE: source = uh->source; break; default: @@ -1179,9 +1170,6 @@ static int esp_init_state(struct xfrm_state *x, struct netlink_ext_ack *extack) case UDP_ENCAP_ESPINUDP: x->props.header_len += sizeof(struct udphdr); break; - case UDP_ENCAP_ESPINUDP_NON_IKE: - x->props.header_len += sizeof(struct udphdr) + 2 * sizeof(u32); - break; #ifdef CONFIG_INET_ESPINTCP case TCP_ENCAP_ESPINTCP: /* only the length field, TCP encap is done by diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7613daa339b0..4ca781065a07 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -2699,8 +2699,6 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, #ifdef CONFIG_XFRM case UDP_ENCAP_ESPINUDP: set_xfrm_gro_udp_encap_rcv(val, sk->sk_family, sk); - fallthrough; - case UDP_ENCAP_ESPINUDP_NON_IKE: #if IS_ENABLED(CONFIG_IPV6) if (sk->sk_family == AF_INET6) WRITE_ONCE(up->encap_rcv, diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c index dae35101d189..0918b0682174 100644 --- a/net/ipv4/xfrm4_input.c +++ b/net/ipv4/xfrm4_input.c @@ -113,19 +113,6 @@ static int __xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb, bool pull /* Must be an IKE packet.. pass it through */ return 1; break; - case UDP_ENCAP_ESPINUDP_NON_IKE: - /* Check if this is a keepalive packet. If so, eat it. */ - if (len == 1 && udpdata[0] == 0xff) { - return -EINVAL; - } else if (len > 2 * sizeof(u32) + sizeof(struct ip_esp_hdr) && - udpdata32[0] == 0 && udpdata32[1] == 0) { - - /* ESP Packet with Non-IKE marker */ - len = sizeof(struct udphdr) + 2 * sizeof(u32); - } else - /* Must be an IKE packet.. pass it through */ - return 1; - break; } /* At this point we are sure that this is an ESPinUDP packet, diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index fe8d53f5a5ee..27df148530a6 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -383,7 +383,6 @@ static struct ip_esp_hdr *esp6_output_udp_encap(struct sk_buff *skb, __be16 dport) { struct udphdr *uh; - __be32 *udpdata32; unsigned int len; len = skb->len + esp->tailen - skb_transport_offset(skb); @@ -398,12 +397,6 @@ static struct ip_esp_hdr *esp6_output_udp_encap(struct sk_buff *skb, *skb_mac_header(skb) = IPPROTO_UDP; - if (encap_type == UDP_ENCAP_ESPINUDP_NON_IKE) { - udpdata32 = (__be32 *)(uh + 1); - udpdata32[0] = udpdata32[1] = 0; - return (struct ip_esp_hdr *)(udpdata32 + 2); - } - return (struct ip_esp_hdr *)(uh + 1); } @@ -459,7 +452,6 @@ static int esp6_output_encap(struct xfrm_state *x, struct sk_buff *skb, switch (encap_type) { default: case UDP_ENCAP_ESPINUDP: - case UDP_ENCAP_ESPINUDP_NON_IKE: esph = esp6_output_udp_encap(skb, encap_type, esp, sport, dport); break; case TCP_ENCAP_ESPINTCP: @@ -822,7 +814,6 @@ int esp6_input_done2(struct sk_buff *skb, int err) source = th->source; break; case UDP_ENCAP_ESPINUDP: - case UDP_ENCAP_ESPINUDP_NON_IKE: source = uh->source; break; default: @@ -1232,9 +1223,6 @@ static int esp6_init_state(struct xfrm_state *x, struct netlink_ext_ack *extack) case UDP_ENCAP_ESPINUDP: x->props.header_len += sizeof(struct udphdr); break; - case UDP_ENCAP_ESPINUDP_NON_IKE: - x->props.header_len += sizeof(struct udphdr) + 2 * sizeof(u32); - break; #ifdef CONFIG_INET6_ESPINTCP case TCP_ENCAP_ESPINTCP: /* only the length field, TCP encap is done by diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index a17d783dc7c0..2c6aeb090b7a 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -109,19 +109,6 @@ static int __xfrm6_udp_encap_rcv(struct sock *sk, struct sk_buff *skb, bool pull /* Must be an IKE packet.. pass it through */ return 1; break; - case UDP_ENCAP_ESPINUDP_NON_IKE: - /* Check if this is a keepalive packet. If so, eat it. */ - if (len == 1 && udpdata[0] == 0xff) { - return -EINVAL; - } else if (len > 2 * sizeof(u32) + sizeof(struct ip_esp_hdr) && - udpdata32[0] == 0 && udpdata32[1] == 0) { - - /* ESP Packet with Non-IKE marker */ - len = sizeof(struct udphdr) + 2 * sizeof(u32); - } else - /* Must be an IKE packet.. pass it through */ - return 1; - break; } /* At this point we are sure that this is an ESPinUDP packet, -- cgit v1.2.3 From eabf425bc6ad32fa49cfb35c7bc59db07dfdd36e Mon Sep 17 00:00:00 2001 From: Zheng Li Date: Tue, 16 Apr 2024 17:53:43 +0800 Subject: neighbour: guarantee the localhost connections be established successfully even the ARP table is full Inter-process communication on localhost should be established successfully even the ARP table is full, many processes on server machine use the localhost to communicate such as command-line interface (CLI), servers hope all CLI commands can be executed successfully even the arp table is full. Right now CLI commands got timeout when the arp table is full. Set the parameter of exempt_from_gc to be true for LOOPBACK net device to keep localhost neigh in arp table, not removed by gc. the steps of reproduced: server with "gc_thresh3 = 1024" setting, ping server from more than 1024 same netmask Lan IPv4 addresses, run "ssh localhost" on console interface, then the command will get timeout. Signed-off-by: Zheng Li Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240416095343.540-1-lizheng043@gmail.com Signed-off-by: Paolo Abeni --- net/core/neighbour.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 552719c3bbc3..47d07b122f7a 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -734,7 +734,9 @@ out_neigh_release: struct neighbour *__neigh_create(struct neigh_table *tbl, const void *pkey, struct net_device *dev, bool want_ref) { - return ___neigh_create(tbl, pkey, dev, 0, false, want_ref); + bool exempt_from_gc = !!(dev->flags & IFF_LOOPBACK); + + return ___neigh_create(tbl, pkey, dev, 0, exempt_from_gc, want_ref); } EXPORT_SYMBOL(__neigh_create); -- cgit v1.2.3 From 4d18e3ddf427d93a2895a1c9d61477e1f1726cad Mon Sep 17 00:00:00 2001 From: "Kory Maincent (Dent Project)" Date: Wed, 17 Apr 2024 16:39:51 +0200 Subject: net: ethtool: pse-pd: Expand pse commands with the PSE PoE interface Add PSE PoE interface support in the ethtool pse command. Reviewed-by: Andrew Lunn Signed-off-by: Kory Maincent Link: https://lore.kernel.org/r/20240417-feature_poe-v9-3-242293fd1900@bootlin.com Signed-off-by: Jakub Kicinski --- Documentation/networking/ethtool-netlink.rst | 20 ++++++++++ net/ethtool/pse-pd.c | 60 +++++++++++++++++++++++----- 2 files changed, 70 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index 4e63d3708ed9..8bc71f249448 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -1733,6 +1733,10 @@ Kernel response contents: PSE functions ``ETHTOOL_A_PODL_PSE_PW_D_STATUS`` u32 power detection status of the PoDL PSE. + ``ETHTOOL_A_C33_PSE_ADMIN_STATE`` u32 Operational state of the PoE + PSE functions. + ``ETHTOOL_A_C33_PSE_PW_D_STATUS`` u32 power detection status of the + PoE PSE. ====================================== ====== ============================= When set, the optional ``ETHTOOL_A_PODL_PSE_ADMIN_STATE`` attribute identifies @@ -1744,6 +1748,12 @@ aPoDLPSEAdminState. Possible values are: .. kernel-doc:: include/uapi/linux/ethtool.h :identifiers: ethtool_podl_pse_admin_state +The same goes for ``ETHTOOL_A_C33_PSE_ADMIN_STATE`` implementing +``IEEE 802.3-2022`` 30.9.1.1.2 aPSEAdminState. + +.. kernel-doc:: include/uapi/linux/ethtool.h + :identifiers: ethtool_c33_pse_admin_state + When set, the optional ``ETHTOOL_A_PODL_PSE_PW_D_STATUS`` attribute identifies the power detection status of the PoDL PSE. The status depend on internal PSE state machine and automatic PD classification support. This option is @@ -1753,6 +1763,12 @@ Possible values are: .. kernel-doc:: include/uapi/linux/ethtool.h :identifiers: ethtool_podl_pse_pw_d_status +The same goes for ``ETHTOOL_A_C33_PSE_ADMIN_PW_D_STATUS`` implementing +``IEEE 802.3-2022`` 30.9.1.1.5 aPSEPowerDetectionStatus. + +.. kernel-doc:: include/uapi/linux/ethtool.h + :identifiers: ethtool_c33_pse_pw_d_status + PSE_SET ======= @@ -1763,6 +1779,7 @@ Request contents: ====================================== ====== ============================= ``ETHTOOL_A_PSE_HEADER`` nested request header ``ETHTOOL_A_PODL_PSE_ADMIN_CONTROL`` u32 Control PoDL PSE Admin state + ``ETHTOOL_A_C33_PSE_ADMIN_CONTROL`` u32 Control PSE Admin state ====================================== ====== ============================= When set, the optional ``ETHTOOL_A_PODL_PSE_ADMIN_CONTROL`` attribute is used @@ -1770,6 +1787,9 @@ to control PoDL PSE Admin functions. This option is implementing ``IEEE 802.3-2018`` 30.15.1.2.1 acPoDLPSEAdminControl. See ``ETHTOOL_A_PODL_PSE_ADMIN_STATE`` for supported values. +The same goes for ``ETHTOOL_A_C33_PSE_ADMIN_CONTROL`` implementing +``IEEE 802.3-2022`` 30.9.1.2.1 acPSEAdminControl. + RSS_GET ======= diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c index aef57a058f0d..2c981d443f27 100644 --- a/net/ethtool/pse-pd.c +++ b/net/ethtool/pse-pd.c @@ -82,6 +82,10 @@ static int pse_reply_size(const struct ethnl_req_info *req_base, len += nla_total_size(sizeof(u32)); /* _PODL_PSE_ADMIN_STATE */ if (st->podl_pw_status > 0) len += nla_total_size(sizeof(u32)); /* _PODL_PSE_PW_D_STATUS */ + if (st->c33_admin_state > 0) + len += nla_total_size(sizeof(u32)); /* _C33_PSE_ADMIN_STATE */ + if (st->c33_pw_status > 0) + len += nla_total_size(sizeof(u32)); /* _C33_PSE_PW_D_STATUS */ return len; } @@ -103,6 +107,16 @@ static int pse_fill_reply(struct sk_buff *skb, st->podl_pw_status)) return -EMSGSIZE; + if (st->c33_admin_state > 0 && + nla_put_u32(skb, ETHTOOL_A_C33_PSE_ADMIN_STATE, + st->c33_admin_state)) + return -EMSGSIZE; + + if (st->c33_pw_status > 0 && + nla_put_u32(skb, ETHTOOL_A_C33_PSE_PW_D_STATUS, + st->c33_pw_status)) + return -EMSGSIZE; + return 0; } @@ -113,25 +127,18 @@ const struct nla_policy ethnl_pse_set_policy[ETHTOOL_A_PSE_MAX + 1] = { [ETHTOOL_A_PODL_PSE_ADMIN_CONTROL] = NLA_POLICY_RANGE(NLA_U32, ETHTOOL_PODL_PSE_ADMIN_STATE_DISABLED, ETHTOOL_PODL_PSE_ADMIN_STATE_ENABLED), + [ETHTOOL_A_C33_PSE_ADMIN_CONTROL] = + NLA_POLICY_RANGE(NLA_U32, ETHTOOL_C33_PSE_ADMIN_STATE_DISABLED, + ETHTOOL_C33_PSE_ADMIN_STATE_ENABLED), }; static int ethnl_set_pse_validate(struct ethnl_req_info *req_info, struct genl_info *info) -{ - return !!info->attrs[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]; -} - -static int -ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info) { struct net_device *dev = req_info->dev; - struct pse_control_config config = {}; struct nlattr **tb = info->attrs; struct phy_device *phydev; - /* this values are already validated by the ethnl_pse_set_policy */ - config.podl_admin_control = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]); - phydev = dev->phydev; if (!phydev) { NL_SET_ERR_MSG(info->extack, "No PHY is attached"); @@ -143,6 +150,39 @@ ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info) return -EOPNOTSUPP; } + if (tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL] && + !pse_has_podl(phydev->psec)) { + NL_SET_ERR_MSG_ATTR(info->extack, + tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL], + "setting PoDL PSE admin control not supported"); + return -EOPNOTSUPP; + } + if (tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL] && + !pse_has_c33(phydev->psec)) { + NL_SET_ERR_MSG_ATTR(info->extack, + tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL], + "setting C33 PSE admin control not supported"); + return -EOPNOTSUPP; + } + + return 1; +} + +static int +ethnl_set_pse(struct ethnl_req_info *req_info, struct genl_info *info) +{ + struct net_device *dev = req_info->dev; + struct pse_control_config config = {}; + struct nlattr **tb = info->attrs; + struct phy_device *phydev; + + phydev = dev->phydev; + /* These values are already validated by the ethnl_pse_set_policy */ + if (pse_has_podl(phydev->psec)) + config.podl_admin_control = nla_get_u32(tb[ETHTOOL_A_PODL_PSE_ADMIN_CONTROL]); + if (pse_has_c33(phydev->psec)) + config.c33_admin_control = nla_get_u32(tb[ETHTOOL_A_C33_PSE_ADMIN_CONTROL]); + /* Return errno directly - PSE has no notification */ return pse_ethtool_set_config(phydev->psec, info->extack, &config); } -- cgit v1.2.3 From 570944a094c24ee3a09b2cb5e580063cfde64d7a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 15 Apr 2024 11:27:12 +0300 Subject: wifi: mac80211: transmit deauth only if link is available There's an issue in that when we disconnect from an AP due to the AP switching to an unsupported channel, we might not tell the driver about this before we try to send the deauth. If the underlying implementation has detected the quiet CSA, this may cause issues if this is the only active link. Avoid this by transmitting (and flushing) the deauth only when there's an active link available that's not affected by quiet CSA. Since this introduces link->u.mgd.csa_blocked_tx and we no longer check sdata->csa_blocked_tx for the TX itself also rename the latter to csa_blocked_queues. Fixes: 6f0107d195a8 ("wifi: mac80211: introduce a feature flag for quiet in CSA") Signed-off-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://msgid.link/20240415112355.1d91db5e95aa.Iad3a5df3367f305dff48cd61776abfd6cf0fd4ab@changeid Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 12 +++++------ net/mac80211/ieee80211_i.h | 3 ++- net/mac80211/iface.c | 4 ++-- net/mac80211/mlme.c | 53 +++++++++++++++++++++++++++++++--------------- 4 files changed, 46 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index f67c1d021812..07abaf7820c5 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1607,10 +1607,10 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev, /* abort any running channel switch or color change */ link_conf->csa_active = false; link_conf->color_change_active = false; - if (sdata->csa_blocked_tx) { + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } ieee80211_free_next_beacon(link); @@ -3648,7 +3648,7 @@ void ieee80211_channel_switch_disconnect(struct ieee80211_vif *vif, bool block_t struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; struct ieee80211_local *local = sdata->local; - sdata->csa_blocked_tx = block_tx; + sdata->csa_blocked_queues = block_tx; sdata_info(sdata, "channel switch failed, disconnecting\n"); wiphy_work_queue(local->hw.wiphy, &ifmgd->csa_connection_drop_work); } @@ -3734,10 +3734,10 @@ static int __ieee80211_csa_finalize(struct ieee80211_link_data *link_data) ieee80211_link_info_change_notify(sdata, link_data, changed); - if (sdata->csa_blocked_tx) { + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } err = drv_post_channel_switch(link_data); @@ -4019,7 +4019,7 @@ __ieee80211_channel_switch(struct wiphy *wiphy, struct net_device *dev, !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA)) { ieee80211_stop_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = true; + sdata->csa_blocked_queues = true; } cfg80211_ch_switch_started_notify(sdata->dev, diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 81a9c5b7642d..5c95ec9f499f 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -976,6 +976,7 @@ struct ieee80211_link_data_managed { bool csa_waiting_bcn; bool csa_ignored_same_chan; + bool csa_blocked_tx; struct wiphy_delayed_work chswitch_work; struct wiphy_work request_smps_work; @@ -1094,7 +1095,7 @@ struct ieee80211_sub_if_data { unsigned long state; - bool csa_blocked_tx; + bool csa_blocked_queues; char name[IFNAMSIZ]; diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 395de62d9cb2..ef6b0fc82d02 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -544,10 +544,10 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_do sdata->vif.bss_conf.csa_active = false; if (sdata->vif.type == NL80211_IFTYPE_STATION) sdata->deflink.u.mgd.csa_waiting_bcn = false; - if (sdata->csa_blocked_tx) { + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } wiphy_work_cancel(local->hw.wiphy, &sdata->deflink.csa_finalize_work); diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index cd4b54e02653..742cf2561bda 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -1930,13 +1930,14 @@ static void ieee80211_chswitch_post_beacon(struct ieee80211_link_data *link) WARN_ON(!link->conf->csa_active); - if (sdata->csa_blocked_tx) { + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } link->conf->csa_active = false; + link->u.mgd.csa_blocked_tx = false; link->u.mgd.csa_waiting_bcn = false; ret = drv_post_channel_switch(link); @@ -1996,13 +1997,14 @@ ieee80211_sta_abort_chanswitch(struct ieee80211_link_data *link) ieee80211_link_unreserve_chanctx(link); - if (sdata->csa_blocked_tx) { + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } link->conf->csa_active = false; + link->u.mgd.csa_blocked_tx = false; drv_abort_channel_switch(link); } @@ -2162,12 +2164,13 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link, link->csa_chanreq = csa_ie.chanreq; link->u.mgd.csa_ignored_same_chan = false; link->u.mgd.beacon_crc_valid = false; + link->u.mgd.csa_blocked_tx = csa_ie.mode; if (csa_ie.mode && !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA)) { ieee80211_stop_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = true; + sdata->csa_blocked_queues = true; } cfg80211_ch_switch_started_notify(sdata->dev, &csa_ie.chanreq.oper, @@ -2196,7 +2199,8 @@ ieee80211_sta_process_chanswitch(struct ieee80211_link_data *link, * reset when the disconnection worker runs. */ link->conf->csa_active = true; - sdata->csa_blocked_tx = + link->u.mgd.csa_blocked_tx = csa_ie.mode; + sdata->csa_blocked_queues = csa_ie.mode && !ieee80211_hw_check(&local->hw, HANDLES_QUIET_CSA); wiphy_work_queue(sdata->local->hw.wiphy, @@ -3249,12 +3253,13 @@ static void ieee80211_set_disassoc(struct ieee80211_sub_if_data *sdata, } sdata->vif.bss_conf.csa_active = false; + sdata->deflink.u.mgd.csa_blocked_tx = false; sdata->deflink.u.mgd.csa_waiting_bcn = false; sdata->deflink.u.mgd.csa_ignored_same_chan = false; - if (sdata->csa_blocked_tx) { + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } /* existing TX TSPEC sessions no longer exist */ @@ -3560,19 +3565,32 @@ static void __ieee80211_disconnect(struct ieee80211_sub_if_data *sdata) struct ieee80211_local *local = sdata->local; struct ieee80211_if_managed *ifmgd = &sdata->u.mgd; u8 frame_buf[IEEE80211_DEAUTH_FRAME_LEN]; - bool tx; + bool tx = false; lockdep_assert_wiphy(local->hw.wiphy); if (!ifmgd->associated) return; - /* - * MLO drivers should have HANDLES_QUIET_CSA, so that csa_blocked_tx - * is always false; if they don't then this may try to transmit the - * frame but queues will be stopped. - */ - tx = !sdata->csa_blocked_tx; + /* only transmit if we have a link that makes that worthwhile */ + for (unsigned int link_id = 0; + link_id < ARRAY_SIZE(sdata->link); + link_id++) { + struct ieee80211_link_data *link; + + if (!ieee80211_vif_link_active(&sdata->vif, link_id)) + continue; + + link = sdata_dereference(sdata->link[link_id], sdata); + if (WARN_ON_ONCE(!link)) + continue; + + if (link->u.mgd.csa_blocked_tx) + continue; + + tx = true; + break; + } if (!ifmgd->driver_disconnect) { unsigned int link_id; @@ -3605,10 +3623,11 @@ static void __ieee80211_disconnect(struct ieee80211_sub_if_data *sdata) /* the other links will be destroyed */ sdata->vif.bss_conf.csa_active = false; sdata->deflink.u.mgd.csa_waiting_bcn = false; - if (sdata->csa_blocked_tx) { + sdata->deflink.u.mgd.csa_blocked_tx = false; + if (sdata->csa_blocked_queues) { ieee80211_wake_vif_queues(local, sdata, IEEE80211_QUEUE_STOP_REASON_CSA); - sdata->csa_blocked_tx = false; + sdata->csa_blocked_queues = false; } ieee80211_report_disconnect(sdata, frame_buf, sizeof(frame_buf), tx, -- cgit v1.2.3 From c6d075be966ddc1f08f843f78a82d092e5e4362e Mon Sep 17 00:00:00 2001 From: Miri Korenblit Date: Mon, 15 Apr 2024 11:27:13 +0300 Subject: wifi: mac80211: defer link switch work in reconfig If a link switch work was queued, and then a restart happened, the worker might be executed before the reconfig, and obviously it will fail (the HW might not respond to updates etc.) So, don't perform the switch if we are in reconfig, instead - do it at the end of the reconfig. Signed-off-by: Miri Korenblit Link: https://msgid.link/20240415112355.1ef1008e3a0a.I19add3f2152dcfd55a759de97b1d09265c1cde98@changeid Signed-off-by: Johannes Berg --- net/mac80211/iface.c | 5 +++++ net/mac80211/util.c | 7 +++++++ 2 files changed, 12 insertions(+) (limited to 'net') diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index ef6b0fc82d02..dc42902e2693 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -1699,8 +1699,13 @@ static void ieee80211_activate_links_work(struct wiphy *wiphy, struct ieee80211_sub_if_data *sdata = container_of(work, struct ieee80211_sub_if_data, activate_links_work); + struct ieee80211_local *local = wiphy_priv(wiphy); + + if (local->in_reconfig) + return; ieee80211_set_active_links(&sdata->vif, sdata->desired_active_links); + sdata->desired_active_links = 0; } /* diff --git a/net/mac80211/util.c b/net/mac80211/util.c index cda398d8f60d..0b893e958959 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -2106,6 +2106,13 @@ int ieee80211_reconfig(struct ieee80211_local *local) if (sdata->restart_active_links) ieee80211_set_active_links(&sdata->vif, sdata->restart_active_links); + /* + * If a link switch was scheduled before the restart, and ran + * before reconfig, it will do nothing, so re-schedule. + */ + if (sdata->desired_active_links) + wiphy_work_queue(sdata->local->hw.wiphy, + &sdata->activate_links_work); } /* Reconfigure sched scan if it was interrupted by FW restart */ -- cgit v1.2.3 From 5e0c422d12b535cb3c7281d017f1a5b1b16098a5 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 18 Apr 2024 11:52:19 +0200 Subject: wifi: mac80211: reserve chanctx during find When searching for a chanctx for re-use, it's later adjusted and assigned. It may also be that another one is already assigned to the link in question, so unassign can also happen. In short, the driver is called multiple times. During these callbacks, it may thus change active links (on another interface), which then can in turn cause the found chanctx (that's going to be reused) to get removed and freed. To avoid this, temporarily assign it to the reserved chanctx and track the link that wants to use it in the reserved_links list. This causes the ieee80211_chanctx_refcount() to be increased by one during these operations, thus avoiding the free. Reviewed-by: Miriam Rachel Korenblit Signed-off-by: Johannes Berg Link: https://msgid.link/20240418115219.94ea84c8ee1e.I0b247dbc0cd937ae6367bc0fc7e8d156b5d5f9b1@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 31 +++++++++++++++++++++++++++++-- 1 file changed, 29 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 5a7fb0e4f89d..6d0148dea126 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -547,8 +547,10 @@ static void ieee80211_change_chanctx(struct ieee80211_local *local, _ieee80211_change_chanctx(local, ctx, old_ctx, chanreq, NULL); } +/* Note: if successful, the returned chanctx is reserved for the link */ static struct ieee80211_chanctx * ieee80211_find_chanctx(struct ieee80211_local *local, + struct ieee80211_link_data *link, const struct ieee80211_chan_req *chanreq, enum ieee80211_chanctx_mode mode) { @@ -560,6 +562,9 @@ ieee80211_find_chanctx(struct ieee80211_local *local, if (mode == IEEE80211_CHANCTX_EXCLUSIVE) return NULL; + if (WARN_ON(link->reserved_chanctx)) + return NULL; + list_for_each_entry(ctx, &local->chanctx_list, list) { const struct ieee80211_chan_req *compat; @@ -578,6 +583,16 @@ ieee80211_find_chanctx(struct ieee80211_local *local, if (!compat) continue; + /* + * Reserve the chanctx temporarily, as the driver might change + * active links during callbacks we make into it below and/or + * later during assignment, which could (otherwise) cause the + * context to actually be removed. + */ + link->reserved_chanctx = ctx; + list_add(&link->reserved_chanctx_list, + &ctx->reserved_links); + ieee80211_change_chanctx(local, ctx, ctx, compat); return ctx; @@ -1701,6 +1716,7 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, struct ieee80211_local *local = sdata->local; struct ieee80211_chanctx *ctx; u8 radar_detect_width = 0; + bool reserved = false; int ret; lockdep_assert_wiphy(local->hw.wiphy); @@ -1727,8 +1743,11 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, __ieee80211_link_release_channel(link, false); - ctx = ieee80211_find_chanctx(local, chanreq, mode); - if (!ctx) + ctx = ieee80211_find_chanctx(local, link, chanreq, mode); + /* Note: context is now reserved */ + if (ctx) + reserved = true; + else ctx = ieee80211_new_chanctx(local, chanreq, mode); if (IS_ERR(ctx)) { ret = PTR_ERR(ctx); @@ -1738,6 +1757,14 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, ieee80211_link_update_chanreq(link, chanreq); ret = ieee80211_assign_link_chanctx(link, ctx); + + if (reserved) { + /* remove reservation */ + WARN_ON(link->reserved_chanctx != ctx); + link->reserved_chanctx = NULL; + list_del(&link->reserved_chanctx_list); + } + if (ret) { /* if assign fails refcount stays the same */ if (ieee80211_chanctx_refcount(local, ctx) == 0) -- cgit v1.2.3 From 87f5500285fbd27eb36efbcd02ddcee1ad27ec91 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 18 Apr 2024 11:52:20 +0200 Subject: wifi: mac80211: simplify ieee80211_assign_link_chanctx() There's no need for a label/goto here, the only thing is that drv_assign_vif_chanctx() must succeed to set 'conf' and add the new context to the list, the remaining code is (and must be) the same regardless. Reviewed-by: Miriam Rachel Korenblit Signed-off-by: Johannes Berg Link: https://msgid.link/20240418115219.a94852030d33.I9d647178ab25636372ed79e5312c68a06e0bf60c@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 6d0148dea126..2c338014ded5 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -815,7 +815,7 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, struct ieee80211_local *local = sdata->local; struct ieee80211_chanctx_conf *conf; struct ieee80211_chanctx *curr_ctx = NULL; - int ret = 0; + int ret; if (WARN_ON(sdata->vif.type == NL80211_IFTYPE_NAN)) return -EOPNOTSUPP; @@ -836,15 +836,16 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, ieee80211_recalc_chanctx_min_def(local, new_ctx, link); ret = drv_assign_vif_chanctx(local, sdata, link->conf, new_ctx); - if (ret) - goto out; - - conf = &new_ctx->conf; - list_add(&link->assigned_chanctx_list, - &new_ctx->assigned_links); + if (!ret) { + /* succeeded, so commit it to the data structures */ + conf = &new_ctx->conf; + list_add(&link->assigned_chanctx_list, + &new_ctx->assigned_links); + } + } else { + ret = 0; } -out: rcu_assign_pointer(link->conf->chanctx_conf, conf); sdata->vif.cfg.idle = !conf; -- cgit v1.2.3 From cc3ea42cfaac955b485d23507c6bac84fb21a7da Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Thu, 18 Apr 2024 11:52:21 +0200 Subject: wifi: mac80211: keep mac80211 consistent on link activation failure In the unlikely event that link_use_channel fails while activating a link, mac80211 would go into a bad state. Unfortunately, we cannot completely avoid failures from drivers in this case. However, what we can do is to just continue internally anyway and assume the driver is going to trigger a recovery flow from its side. Doing that means that we at least have a consistent state in mac80211 allowing such a recovery flow to succeed. Reviewed-by: Miriam Rachel Korenblit Signed-off-by: Benjamin Berg Signed-off-by: Johannes Berg Link: https://msgid.link/20240418115219.1129e89f4b55.I6299678353e50e88b55c99b0bce15c64b52c2804@changeid Signed-off-by: Johannes Berg --- net/mac80211/chan.c | 35 +++++++++++++++++++++++------------ net/mac80211/ieee80211_i.h | 12 +++++++++++- net/mac80211/link.c | 21 ++++++++++++++++++--- 3 files changed, 52 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/net/mac80211/chan.c b/net/mac80211/chan.c index 2c338014ded5..dedf11eeb43c 100644 --- a/net/mac80211/chan.c +++ b/net/mac80211/chan.c @@ -688,7 +688,8 @@ static int ieee80211_add_chanctx(struct ieee80211_local *local, static struct ieee80211_chanctx * ieee80211_new_chanctx(struct ieee80211_local *local, const struct ieee80211_chan_req *chanreq, - enum ieee80211_chanctx_mode mode) + enum ieee80211_chanctx_mode mode, + bool assign_on_failure) { struct ieee80211_chanctx *ctx; int err; @@ -700,10 +701,12 @@ ieee80211_new_chanctx(struct ieee80211_local *local, return ERR_PTR(-ENOMEM); err = ieee80211_add_chanctx(local, ctx); - if (err) { + if (!assign_on_failure && err) { kfree(ctx); return ERR_PTR(err); } + /* We ignored a driver error, see _ieee80211_set_active_links */ + WARN_ON_ONCE(err && !local->in_reconfig); list_add_rcu(&ctx->list, &local->chanctx_list); return ctx; @@ -809,7 +812,8 @@ static void ieee80211_recalc_radar_chanctx(struct ieee80211_local *local, } static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, - struct ieee80211_chanctx *new_ctx) + struct ieee80211_chanctx *new_ctx, + bool assign_on_failure) { struct ieee80211_sub_if_data *sdata = link->sdata; struct ieee80211_local *local = sdata->local; @@ -836,7 +840,11 @@ static int ieee80211_assign_link_chanctx(struct ieee80211_link_data *link, ieee80211_recalc_chanctx_min_def(local, new_ctx, link); ret = drv_assign_vif_chanctx(local, sdata, link->conf, new_ctx); - if (!ret) { + if (assign_on_failure || !ret) { + /* Need to continue, see _ieee80211_set_active_links */ + WARN_ON_ONCE(ret && !local->in_reconfig); + ret = 0; + /* succeeded, so commit it to the data structures */ conf = &new_ctx->conf; list_add(&link->assigned_chanctx_list, @@ -1046,7 +1054,8 @@ int ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link, new_ctx = ieee80211_find_reservation_chanctx(local, chanreq, mode); if (!new_ctx) { if (ieee80211_can_create_new_chanctx(local)) { - new_ctx = ieee80211_new_chanctx(local, chanreq, mode); + new_ctx = ieee80211_new_chanctx(local, chanreq, mode, + false); if (IS_ERR(new_ctx)) return PTR_ERR(new_ctx); } else { @@ -1302,7 +1311,7 @@ ieee80211_link_use_reserved_assign(struct ieee80211_link_data *link) list_del(&link->reserved_chanctx_list); link->reserved_chanctx = NULL; - err = ieee80211_assign_link_chanctx(link, new_ctx); + err = ieee80211_assign_link_chanctx(link, new_ctx, false); if (err) { if (ieee80211_chanctx_refcount(local, new_ctx) == 0) ieee80211_free_chanctx(local, new_ctx, false); @@ -1698,7 +1707,7 @@ void __ieee80211_link_release_channel(struct ieee80211_link_data *link, ieee80211_link_unreserve_chanctx(link); } - ieee80211_assign_link_chanctx(link, NULL); + ieee80211_assign_link_chanctx(link, NULL, false); if (ieee80211_chanctx_refcount(local, ctx) == 0) ieee80211_free_chanctx(local, ctx, skip_idle_recalc); @@ -1709,9 +1718,10 @@ void __ieee80211_link_release_channel(struct ieee80211_link_data *link, ieee80211_vif_use_reserved_switch(local); } -int ieee80211_link_use_channel(struct ieee80211_link_data *link, - const struct ieee80211_chan_req *chanreq, - enum ieee80211_chanctx_mode mode) +int _ieee80211_link_use_channel(struct ieee80211_link_data *link, + const struct ieee80211_chan_req *chanreq, + enum ieee80211_chanctx_mode mode, + bool assign_on_failure) { struct ieee80211_sub_if_data *sdata = link->sdata; struct ieee80211_local *local = sdata->local; @@ -1749,7 +1759,8 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, if (ctx) reserved = true; else - ctx = ieee80211_new_chanctx(local, chanreq, mode); + ctx = ieee80211_new_chanctx(local, chanreq, mode, + assign_on_failure); if (IS_ERR(ctx)) { ret = PTR_ERR(ctx); goto out; @@ -1757,7 +1768,7 @@ int ieee80211_link_use_channel(struct ieee80211_link_data *link, ieee80211_link_update_chanreq(link, chanreq); - ret = ieee80211_assign_link_chanctx(link, ctx); + ret = ieee80211_assign_link_chanctx(link, ctx, assign_on_failure); if (reserved) { /* remove reservation */ diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 5c95ec9f499f..eb62b7d4b4f7 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -2554,9 +2554,19 @@ bool ieee80211_chanreq_identical(const struct ieee80211_chan_req *a, const struct ieee80211_chan_req *b); int __must_check +_ieee80211_link_use_channel(struct ieee80211_link_data *link, + const struct ieee80211_chan_req *req, + enum ieee80211_chanctx_mode mode, + bool assign_on_failure); + +static inline int __must_check ieee80211_link_use_channel(struct ieee80211_link_data *link, const struct ieee80211_chan_req *req, - enum ieee80211_chanctx_mode mode); + enum ieee80211_chanctx_mode mode) +{ + return _ieee80211_link_use_channel(link, req, mode, false); +} + int __must_check ieee80211_link_reserve_chanctx(struct ieee80211_link_data *link, const struct ieee80211_chan_req *req, diff --git a/net/mac80211/link.c b/net/mac80211/link.c index 43f9672fc7f1..af0321408a97 100644 --- a/net/mac80211/link.c +++ b/net/mac80211/link.c @@ -404,9 +404,24 @@ static int _ieee80211_set_active_links(struct ieee80211_sub_if_data *sdata, link = sdata_dereference(sdata->link[link_id], sdata); - ret = ieee80211_link_use_channel(link, - &link->conf->chanreq, - IEEE80211_CHANCTX_SHARED); + /* + * This call really should not fail. Unfortunately, it appears + * that this may happen occasionally with some drivers. Should + * it happen, we are stuck in a bad place as going backwards is + * not really feasible. + * + * So lets just tell link_use_channel that it must not fail to + * assign the channel context (from mac80211's perspective) and + * assume the driver is going to trigger a recovery flow if it + * had a failure. + * That really is not great nor guaranteed to work. But at least + * the internal mac80211 state remains consistent and there is + * a chance that we can recover. + */ + ret = _ieee80211_link_use_channel(link, + &link->conf->chanreq, + IEEE80211_CHANCTX_SHARED, + true); WARN_ON_ONCE(ret); ieee80211_mgd_set_link_qos_params(link); -- cgit v1.2.3 From 630009e2440d31683d734165a06365f49acc37d4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 19 Apr 2024 10:21:27 +0200 Subject: wifi: mac80211: add return docs for sta_info_flush() Use the Return: annotation instead of spelling out "Returns" in the documentation, for both sta_info_flush()/__sta_info_flush(). Signed-off-by: Johannes Berg --- net/mac80211/sta_info.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index a52fb76386d0..bd5e2f7146f6 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -886,7 +886,7 @@ void sta_info_stop(struct ieee80211_local *local); /** * __sta_info_flush - flush matching STA entries from the STA table * - * Returns the number of removed STA entries. + * Return: the number of removed STA entries. * * @sdata: sdata to remove all stations from * @vlans: if the given interface is an AP interface, also flush VLANs @@ -900,7 +900,7 @@ int __sta_info_flush(struct ieee80211_sub_if_data *sdata, bool vlans, /** * sta_info_flush - flush matching STA entries from the STA table * - * Returns the number of removed STA entries. + * Return: the number of removed STA entries. * * @sdata: sdata to remove all stations from * @link_id: if given (>=0), all those STA entries using @link_id only -- cgit v1.2.3 From dbda949b7ff005cec4c58bf93c5b4058148f0c59 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 19 Apr 2024 10:25:48 +0200 Subject: wifi: cfg80211: make some regulatory functions void The return value of regulatory_hint_indoor() is always 0 for success, and the return value of regulatory_hint_found_beacon() is always ignored. Make them both have void return. Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 3 ++- net/wireless/reg.c | 18 +++++++----------- net/wireless/reg.h | 8 ++++---- 3 files changed, 13 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index b4edba6b0b7b..e5d8ea4211ea 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -8116,7 +8116,8 @@ static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info) is_indoor = true; } - return regulatory_hint_indoor(is_indoor, owner_nlportid); + regulatory_hint_indoor(is_indoor, owner_nlportid); + return 0; default: return -EINVAL; } diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 753f8e9aa4b1..3cef0021a3db 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -3284,7 +3284,7 @@ int regulatory_hint_user(const char *alpha2, return 0; } -int regulatory_hint_indoor(bool is_indoor, u32 portid) +void regulatory_hint_indoor(bool is_indoor, u32 portid) { spin_lock(®_indoor_lock); @@ -3307,8 +3307,6 @@ int regulatory_hint_indoor(bool is_indoor, u32 portid) if (!is_indoor) reg_check_channels(); - - return 0; } void regulatory_netlink_notify(u32 portid) @@ -3666,9 +3664,9 @@ static bool pending_reg_beacon(struct ieee80211_channel *beacon_chan) return false; } -int regulatory_hint_found_beacon(struct wiphy *wiphy, - struct ieee80211_channel *beacon_chan, - gfp_t gfp) +void regulatory_hint_found_beacon(struct wiphy *wiphy, + struct ieee80211_channel *beacon_chan, + gfp_t gfp) { struct reg_beacon *reg_beacon; bool processing; @@ -3677,18 +3675,18 @@ int regulatory_hint_found_beacon(struct wiphy *wiphy, beacon_chan->flags & IEEE80211_CHAN_RADAR || (beacon_chan->band == NL80211_BAND_2GHZ && !freq_is_chan_12_13_14(beacon_chan->center_freq))) - return 0; + return; spin_lock_bh(®_pending_beacons_lock); processing = pending_reg_beacon(beacon_chan); spin_unlock_bh(®_pending_beacons_lock); if (processing) - return 0; + return; reg_beacon = kzalloc(sizeof(struct reg_beacon), gfp); if (!reg_beacon) - return -ENOMEM; + return; pr_debug("Found new beacon on frequency: %d.%03d MHz (Ch %d) on %s\n", beacon_chan->center_freq, beacon_chan->freq_offset, @@ -3708,8 +3706,6 @@ int regulatory_hint_found_beacon(struct wiphy *wiphy, spin_unlock_bh(®_pending_beacons_lock); schedule_work(®_work); - - return 0; } static void print_rd_rules(const struct ieee80211_regdomain *rd) diff --git a/net/wireless/reg.h b/net/wireless/reg.h index a02ef5609f52..0f839c790a4f 100644 --- a/net/wireless/reg.h +++ b/net/wireless/reg.h @@ -42,7 +42,7 @@ int regulatory_hint_user(const char *alpha2, * device is operating in an indoor environment. * @portid: the netlink port ID on which the hint was given. */ -int regulatory_hint_indoor(bool is_indoor, u32 portid); +void regulatory_hint_indoor(bool is_indoor, u32 portid); /** * regulatory_netlink_notify - notify on released netlink socket @@ -82,9 +82,9 @@ bool reg_last_request_cell_base(void); * on a newly found BSS. If you cannot make use of this feature you can * set the wiphy->disable_beacon_hints to true. */ -int regulatory_hint_found_beacon(struct wiphy *wiphy, - struct ieee80211_channel *beacon_chan, - gfp_t gfp); +void regulatory_hint_found_beacon(struct wiphy *wiphy, + struct ieee80211_channel *beacon_chan, + gfp_t gfp); /** * regulatory_hint_country_ie - hints a country IE as a regulatory domain -- cgit v1.2.3 From 2bf78f0f107300d2128451a239e94ccc7284090d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 19 Apr 2024 10:26:18 +0200 Subject: wifi: cfg80211: add return docs for regulatory functions Add return value documentation for regulatory functions that are missing it. Signed-off-by: Johannes Berg --- net/wireless/reg.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/wireless/reg.h b/net/wireless/reg.h index 0f839c790a4f..e1b211c4f75c 100644 --- a/net/wireless/reg.h +++ b/net/wireless/reg.h @@ -137,13 +137,14 @@ void regulatory_hint_disconnect(void); * Get a value specifying the U-NII band frequency belongs to. * U-NII bands are defined by the FCC in C.F.R 47 part 15. * - * Returns -EINVAL if freq is invalid, 0 for UNII-1, 1 for UNII-2A, + * Return: -EINVAL if freq is invalid, 0 for UNII-1, 1 for UNII-2A, * 2 for UNII-2B, 3 for UNII-2C and 4 for UNII-3. */ int cfg80211_get_unii(int freq); /** * regulatory_indoor_allowed - is indoor operation allowed + * Return: %true if indoor operation is allowed, %false otherwise */ bool regulatory_indoor_allowed(void); @@ -173,11 +174,13 @@ void regulatory_propagate_dfs_state(struct wiphy *wiphy, * reg_dfs_domain_same - Checks if both wiphy have same DFS domain configured * @wiphy1: wiphy it's dfs_region to be checked against that of wiphy2 * @wiphy2: wiphy it's dfs_region to be checked against that of wiphy1 + * Return: %true if both wiphys have the same DFS domain, %false otherwise */ bool reg_dfs_domain_same(struct wiphy *wiphy1, struct wiphy *wiphy2); /** * reg_reload_regdb - reload the regulatory.db firmware file + * Return: 0 for success, an error code otherwise */ int reg_reload_regdb(void); -- cgit v1.2.3 From 6d4ed5b3562c43394db01906e2a94904caaec054 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Tue, 9 Apr 2024 15:10:17 +0530 Subject: wifi: mac80211: handle sdata->u.ap.active flag with MLO Currently whenever link AP beacon is assigned, sdata->u.ap.active flag is set and whenever it is brought down, the flag is reset. However, with MLO, all the links of the same MLD would use the same sdata. Hence there is no need to set/reset for each link up/down. Also, resetting it when only one of the links went down is not desirable. Add changes to set the active flag only when first link is assigned beacon. Similarly, add changes to reset that flag only when last link is brought down. Signed-off-by: Aditya Kumar Singh Link: https://msgid.link/20240409094017.3165560-1-quic_adisi@quicinc.com [remove unnecessary check before constant true assignment] Signed-off-by: Johannes Berg --- net/mac80211/cfg.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 07abaf7820c5..44971a62d798 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -1486,7 +1486,10 @@ static int ieee80211_start_ap(struct wiphy *wiphy, struct net_device *dev, if (old) kfree_rcu(old, rcu_head); RCU_INIT_POINTER(link->u.ap.beacon, NULL); - sdata->u.ap.active = false; + + if (ieee80211_num_beaconing_links(sdata) == 0) + sdata->u.ap.active = false; + goto error; } @@ -1619,11 +1622,12 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev, list_for_each_entry(vlan, &sdata->u.ap.vlans, u.vlan.list) netif_carrier_off(vlan->dev); - if (ieee80211_num_beaconing_links(sdata) <= 1) + if (ieee80211_num_beaconing_links(sdata) <= 1) { netif_carrier_off(dev); + sdata->u.ap.active = false; + } /* remove beacon and probe response */ - sdata->u.ap.active = false; RCU_INIT_POINTER(link->u.ap.beacon, NULL); RCU_INIT_POINTER(link->u.ap.probe_resp, NULL); RCU_INIT_POINTER(link->u.ap.fils_discovery, NULL); -- cgit v1.2.3 From b0d2d8f996c1ce3d4c8a1c8cb7d634536e7c2118 Mon Sep 17 00:00:00 2001 From: Sriram R Date: Wed, 10 Apr 2024 10:57:05 +0530 Subject: wifi: mac80211: handle link ID during management Tx During non-STA management Tx, when source address is same as one of the link addresses and even when userspace requested Tx on a specific link, the link ID is not set in the TX control information. Now if the MLD address is also the same as that of the link address, then mac80211 fills link as "unspecified", since it looks like MLD TX. This is unexpected, however, since non-STA TX must specify which link to use. In hwsim, this will (after warnings) result in dropping such frames as well. Use and set the link id if the link bss is matching the address and requested channel. Signed-off-by: Sriram R Signed-off-by: Aditya Kumar Singh Link: https://msgid.link/20240410052705.169865-1-quic_adisi@quicinc.com Link: https://lore.kernel.org/r/0496fb7e-53cc-476f-8052-985d82fd8d01@quicinc.com [reword commit message, should spell out hwsim etc.] Signed-off-by: Johannes Berg --- net/mac80211/offchannel.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mac80211/offchannel.c b/net/mac80211/offchannel.c index 221695d841fd..65e1e9e971fd 100644 --- a/net/mac80211/offchannel.c +++ b/net/mac80211/offchannel.c @@ -897,8 +897,18 @@ int ieee80211_mgmt_tx(struct wiphy *wiphy, struct wireless_dev *wdev, break; } - if (ether_addr_equal(conf->addr, mgmt->sa)) + if (ether_addr_equal(conf->addr, mgmt->sa)) { + /* If userspace requested Tx on a specific link + * use the same link id if the link bss is matching + * the requested chan. + */ + if (sdata->vif.valid_links && + params->link_id >= 0 && params->link_id == i && + params->chan == chanctx_conf->def.chan) + link_id = i; + break; + } chanctx_conf = NULL; } -- cgit v1.2.3 From 24bcc307679031079986fa0d30382cfc959b5593 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:35 +0000 Subject: net_sched: sch_fq: implement lockless fq_dump() Instead of relying on RTNL, fq_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() in fq_change() v2: Addressed Simon feedback in V1: https://lore.kernel.org/netdev/20240416181915.GT2320920@kernel.org/ Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_fq.c | 108 ++++++++++++++++++++++++++++++++++------------------- 1 file changed, 69 insertions(+), 39 deletions(-) (limited to 'net') diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index cdf23ff16f40..238974725679 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -106,6 +106,8 @@ struct fq_perband_flows { int quantum; /* based on band nr : 576KB, 192KB, 64KB */ }; +#define FQ_PRIO2BAND_CRUMB_SIZE ((TC_PRIO_MAX + 1) >> 2) + struct fq_sched_data { /* Read mostly cache line */ @@ -122,7 +124,7 @@ struct fq_sched_data { u8 rate_enable; u8 fq_trees_log; u8 horizon_drop; - u8 prio2band[(TC_PRIO_MAX + 1) >> 2]; + u8 prio2band[FQ_PRIO2BAND_CRUMB_SIZE]; u32 timer_slack; /* hrtimer slack in ns */ /* Read/Write fields. */ @@ -159,7 +161,7 @@ struct fq_sched_data { /* return the i-th 2-bit value ("crumb") */ static u8 fq_prio2band(const u8 *prio2band, unsigned int prio) { - return (prio2band[prio / 4] >> (2 * (prio & 0x3))) & 0x3; + return (READ_ONCE(prio2band[prio / 4]) >> (2 * (prio & 0x3))) & 0x3; } /* @@ -888,7 +890,7 @@ static int fq_resize(struct Qdisc *sch, u32 log) fq_rehash(q, old_fq_root, q->fq_trees_log, array, log); q->fq_root = array; - q->fq_trees_log = log; + WRITE_ONCE(q->fq_trees_log, log); sch_tree_unlock(sch); @@ -927,11 +929,15 @@ static const struct nla_policy fq_policy[TCA_FQ_MAX + 1] = { static void fq_prio2band_compress_crumb(const u8 *in, u8 *out) { const int num_elems = TC_PRIO_MAX + 1; + u8 tmp[FQ_PRIO2BAND_CRUMB_SIZE]; int i; - memset(out, 0, num_elems / 4); + memset(tmp, 0, sizeof(tmp)); for (i = 0; i < num_elems; i++) - out[i / 4] |= in[i] << (2 * (i & 0x3)); + tmp[i / 4] |= in[i] << (2 * (i & 0x3)); + + for (i = 0; i < FQ_PRIO2BAND_CRUMB_SIZE; i++) + WRITE_ONCE(out[i], tmp[i]); } static void fq_prio2band_decompress_crumb(const u8 *in, u8 *out) @@ -958,7 +964,7 @@ static int fq_load_weights(struct fq_sched_data *q, } } for (i = 0; i < FQ_BANDS; i++) - q->band_flows[i].quantum = weights[i]; + WRITE_ONCE(q->band_flows[i].quantum, weights[i]); return 0; } @@ -1011,16 +1017,18 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt, err = -EINVAL; } if (tb[TCA_FQ_PLIMIT]) - sch->limit = nla_get_u32(tb[TCA_FQ_PLIMIT]); + WRITE_ONCE(sch->limit, + nla_get_u32(tb[TCA_FQ_PLIMIT])); if (tb[TCA_FQ_FLOW_PLIMIT]) - q->flow_plimit = nla_get_u32(tb[TCA_FQ_FLOW_PLIMIT]); + WRITE_ONCE(q->flow_plimit, + nla_get_u32(tb[TCA_FQ_FLOW_PLIMIT])); if (tb[TCA_FQ_QUANTUM]) { u32 quantum = nla_get_u32(tb[TCA_FQ_QUANTUM]); if (quantum > 0 && quantum <= (1 << 20)) { - q->quantum = quantum; + WRITE_ONCE(q->quantum, quantum); } else { NL_SET_ERR_MSG_MOD(extack, "invalid quantum"); err = -EINVAL; @@ -1028,7 +1036,8 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt, } if (tb[TCA_FQ_INITIAL_QUANTUM]) - q->initial_quantum = nla_get_u32(tb[TCA_FQ_INITIAL_QUANTUM]); + WRITE_ONCE(q->initial_quantum, + nla_get_u32(tb[TCA_FQ_INITIAL_QUANTUM])); if (tb[TCA_FQ_FLOW_DEFAULT_RATE]) pr_warn_ratelimited("sch_fq: defrate %u ignored.\n", @@ -1037,17 +1046,19 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt, if (tb[TCA_FQ_FLOW_MAX_RATE]) { u32 rate = nla_get_u32(tb[TCA_FQ_FLOW_MAX_RATE]); - q->flow_max_rate = (rate == ~0U) ? ~0UL : rate; + WRITE_ONCE(q->flow_max_rate, + (rate == ~0U) ? ~0UL : rate); } if (tb[TCA_FQ_LOW_RATE_THRESHOLD]) - q->low_rate_threshold = - nla_get_u32(tb[TCA_FQ_LOW_RATE_THRESHOLD]); + WRITE_ONCE(q->low_rate_threshold, + nla_get_u32(tb[TCA_FQ_LOW_RATE_THRESHOLD])); if (tb[TCA_FQ_RATE_ENABLE]) { u32 enable = nla_get_u32(tb[TCA_FQ_RATE_ENABLE]); if (enable <= 1) - q->rate_enable = enable; + WRITE_ONCE(q->rate_enable, + enable); else err = -EINVAL; } @@ -1055,7 +1066,8 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt, if (tb[TCA_FQ_FLOW_REFILL_DELAY]) { u32 usecs_delay = nla_get_u32(tb[TCA_FQ_FLOW_REFILL_DELAY]) ; - q->flow_refill_delay = usecs_to_jiffies(usecs_delay); + WRITE_ONCE(q->flow_refill_delay, + usecs_to_jiffies(usecs_delay)); } if (!err && tb[TCA_FQ_PRIOMAP]) @@ -1065,21 +1077,26 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt, err = fq_load_weights(q, tb[TCA_FQ_WEIGHTS], extack); if (tb[TCA_FQ_ORPHAN_MASK]) - q->orphan_mask = nla_get_u32(tb[TCA_FQ_ORPHAN_MASK]); + WRITE_ONCE(q->orphan_mask, + nla_get_u32(tb[TCA_FQ_ORPHAN_MASK])); if (tb[TCA_FQ_CE_THRESHOLD]) - q->ce_threshold = (u64)NSEC_PER_USEC * - nla_get_u32(tb[TCA_FQ_CE_THRESHOLD]); + WRITE_ONCE(q->ce_threshold, + (u64)NSEC_PER_USEC * + nla_get_u32(tb[TCA_FQ_CE_THRESHOLD])); if (tb[TCA_FQ_TIMER_SLACK]) - q->timer_slack = nla_get_u32(tb[TCA_FQ_TIMER_SLACK]); + WRITE_ONCE(q->timer_slack, + nla_get_u32(tb[TCA_FQ_TIMER_SLACK])); if (tb[TCA_FQ_HORIZON]) - q->horizon = (u64)NSEC_PER_USEC * - nla_get_u32(tb[TCA_FQ_HORIZON]); + WRITE_ONCE(q->horizon, + (u64)NSEC_PER_USEC * + nla_get_u32(tb[TCA_FQ_HORIZON])); if (tb[TCA_FQ_HORIZON_DROP]) - q->horizon_drop = nla_get_u8(tb[TCA_FQ_HORIZON_DROP]); + WRITE_ONCE(q->horizon_drop, + nla_get_u8(tb[TCA_FQ_HORIZON_DROP])); if (!err) { @@ -1160,13 +1177,13 @@ static int fq_init(struct Qdisc *sch, struct nlattr *opt, static int fq_dump(struct Qdisc *sch, struct sk_buff *skb) { struct fq_sched_data *q = qdisc_priv(sch); - u64 ce_threshold = q->ce_threshold; struct tc_prio_qopt prio = { .bands = FQ_BANDS, }; - u64 horizon = q->horizon; struct nlattr *opts; + u64 ce_threshold; s32 weights[3]; + u64 horizon; opts = nla_nest_start_noflag(skb, TCA_OPTIONS); if (opts == NULL) @@ -1174,35 +1191,48 @@ static int fq_dump(struct Qdisc *sch, struct sk_buff *skb) /* TCA_FQ_FLOW_DEFAULT_RATE is not used anymore */ + ce_threshold = READ_ONCE(q->ce_threshold); do_div(ce_threshold, NSEC_PER_USEC); + + horizon = READ_ONCE(q->horizon); do_div(horizon, NSEC_PER_USEC); - if (nla_put_u32(skb, TCA_FQ_PLIMIT, sch->limit) || - nla_put_u32(skb, TCA_FQ_FLOW_PLIMIT, q->flow_plimit) || - nla_put_u32(skb, TCA_FQ_QUANTUM, q->quantum) || - nla_put_u32(skb, TCA_FQ_INITIAL_QUANTUM, q->initial_quantum) || - nla_put_u32(skb, TCA_FQ_RATE_ENABLE, q->rate_enable) || + if (nla_put_u32(skb, TCA_FQ_PLIMIT, + READ_ONCE(sch->limit)) || + nla_put_u32(skb, TCA_FQ_FLOW_PLIMIT, + READ_ONCE(q->flow_plimit)) || + nla_put_u32(skb, TCA_FQ_QUANTUM, + READ_ONCE(q->quantum)) || + nla_put_u32(skb, TCA_FQ_INITIAL_QUANTUM, + READ_ONCE(q->initial_quantum)) || + nla_put_u32(skb, TCA_FQ_RATE_ENABLE, + READ_ONCE(q->rate_enable)) || nla_put_u32(skb, TCA_FQ_FLOW_MAX_RATE, - min_t(unsigned long, q->flow_max_rate, ~0U)) || + min_t(unsigned long, + READ_ONCE(q->flow_max_rate), ~0U)) || nla_put_u32(skb, TCA_FQ_FLOW_REFILL_DELAY, - jiffies_to_usecs(q->flow_refill_delay)) || - nla_put_u32(skb, TCA_FQ_ORPHAN_MASK, q->orphan_mask) || + jiffies_to_usecs(READ_ONCE(q->flow_refill_delay))) || + nla_put_u32(skb, TCA_FQ_ORPHAN_MASK, + READ_ONCE(q->orphan_mask)) || nla_put_u32(skb, TCA_FQ_LOW_RATE_THRESHOLD, - q->low_rate_threshold) || + READ_ONCE(q->low_rate_threshold)) || nla_put_u32(skb, TCA_FQ_CE_THRESHOLD, (u32)ce_threshold) || - nla_put_u32(skb, TCA_FQ_BUCKETS_LOG, q->fq_trees_log) || - nla_put_u32(skb, TCA_FQ_TIMER_SLACK, q->timer_slack) || + nla_put_u32(skb, TCA_FQ_BUCKETS_LOG, + READ_ONCE(q->fq_trees_log)) || + nla_put_u32(skb, TCA_FQ_TIMER_SLACK, + READ_ONCE(q->timer_slack)) || nla_put_u32(skb, TCA_FQ_HORIZON, (u32)horizon) || - nla_put_u8(skb, TCA_FQ_HORIZON_DROP, q->horizon_drop)) + nla_put_u8(skb, TCA_FQ_HORIZON_DROP, + READ_ONCE(q->horizon_drop))) goto nla_put_failure; fq_prio2band_decompress_crumb(q->prio2band, prio.priomap); if (nla_put(skb, TCA_FQ_PRIOMAP, sizeof(prio), &prio)) goto nla_put_failure; - weights[0] = q->band_flows[0].quantum; - weights[1] = q->band_flows[1].quantum; - weights[2] = q->band_flows[2].quantum; + weights[0] = READ_ONCE(q->band_flows[0].quantum); + weights[1] = READ_ONCE(q->band_flows[1].quantum); + weights[2] = READ_ONCE(q->band_flows[2].quantum); if (nla_put(skb, TCA_FQ_WEIGHTS, sizeof(weights), &weights)) goto nla_put_failure; -- cgit v1.2.3 From 9263650102bb8e4e6806587de395aae02ef76deb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:36 +0000 Subject: net_sched: cake: implement lockless cake_dump() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of relying on RTNL, cake_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in cake_change(). v2: addressed Simon feedback in V1: https://lore.kernel.org/netdev/20240417083549.GA3846178@kernel.org/ Signed-off-by: Eric Dumazet Cc: Toke Høiland-Jørgensen Reviewed-by: Simon Horman Acked-by: Toke Høiland-Jørgensen Signed-off-by: David S. Miller --- net/sched/sch_cake.c | 110 +++++++++++++++++++++++++++++---------------------- 1 file changed, 63 insertions(+), 47 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c index 2eabc4dc5b79..9602dafe32e6 100644 --- a/net/sched/sch_cake.c +++ b/net/sched/sch_cake.c @@ -2572,6 +2572,8 @@ static int cake_change(struct Qdisc *sch, struct nlattr *opt, { struct cake_sched_data *q = qdisc_priv(sch); struct nlattr *tb[TCA_CAKE_MAX + 1]; + u16 rate_flags; + u8 flow_mode; int err; err = nla_parse_nested_deprecated(tb, TCA_CAKE_MAX, opt, cake_policy, @@ -2579,10 +2581,11 @@ static int cake_change(struct Qdisc *sch, struct nlattr *opt, if (err < 0) return err; + flow_mode = q->flow_mode; if (tb[TCA_CAKE_NAT]) { #if IS_ENABLED(CONFIG_NF_CONNTRACK) - q->flow_mode &= ~CAKE_FLOW_NAT_FLAG; - q->flow_mode |= CAKE_FLOW_NAT_FLAG * + flow_mode &= ~CAKE_FLOW_NAT_FLAG; + flow_mode |= CAKE_FLOW_NAT_FLAG * !!nla_get_u32(tb[TCA_CAKE_NAT]); #else NL_SET_ERR_MSG_ATTR(extack, tb[TCA_CAKE_NAT], @@ -2592,29 +2595,34 @@ static int cake_change(struct Qdisc *sch, struct nlattr *opt, } if (tb[TCA_CAKE_BASE_RATE64]) - q->rate_bps = nla_get_u64(tb[TCA_CAKE_BASE_RATE64]); + WRITE_ONCE(q->rate_bps, + nla_get_u64(tb[TCA_CAKE_BASE_RATE64])); if (tb[TCA_CAKE_DIFFSERV_MODE]) - q->tin_mode = nla_get_u32(tb[TCA_CAKE_DIFFSERV_MODE]); + WRITE_ONCE(q->tin_mode, + nla_get_u32(tb[TCA_CAKE_DIFFSERV_MODE])); + rate_flags = q->rate_flags; if (tb[TCA_CAKE_WASH]) { if (!!nla_get_u32(tb[TCA_CAKE_WASH])) - q->rate_flags |= CAKE_FLAG_WASH; + rate_flags |= CAKE_FLAG_WASH; else - q->rate_flags &= ~CAKE_FLAG_WASH; + rate_flags &= ~CAKE_FLAG_WASH; } if (tb[TCA_CAKE_FLOW_MODE]) - q->flow_mode = ((q->flow_mode & CAKE_FLOW_NAT_FLAG) | + flow_mode = ((flow_mode & CAKE_FLOW_NAT_FLAG) | (nla_get_u32(tb[TCA_CAKE_FLOW_MODE]) & CAKE_FLOW_MASK)); if (tb[TCA_CAKE_ATM]) - q->atm_mode = nla_get_u32(tb[TCA_CAKE_ATM]); + WRITE_ONCE(q->atm_mode, + nla_get_u32(tb[TCA_CAKE_ATM])); if (tb[TCA_CAKE_OVERHEAD]) { - q->rate_overhead = nla_get_s32(tb[TCA_CAKE_OVERHEAD]); - q->rate_flags |= CAKE_FLAG_OVERHEAD; + WRITE_ONCE(q->rate_overhead, + nla_get_s32(tb[TCA_CAKE_OVERHEAD])); + rate_flags |= CAKE_FLAG_OVERHEAD; q->max_netlen = 0; q->max_adjlen = 0; @@ -2623,7 +2631,7 @@ static int cake_change(struct Qdisc *sch, struct nlattr *opt, } if (tb[TCA_CAKE_RAW]) { - q->rate_flags &= ~CAKE_FLAG_OVERHEAD; + rate_flags &= ~CAKE_FLAG_OVERHEAD; q->max_netlen = 0; q->max_adjlen = 0; @@ -2632,54 +2640,58 @@ static int cake_change(struct Qdisc *sch, struct nlattr *opt, } if (tb[TCA_CAKE_MPU]) - q->rate_mpu = nla_get_u32(tb[TCA_CAKE_MPU]); + WRITE_ONCE(q->rate_mpu, + nla_get_u32(tb[TCA_CAKE_MPU])); if (tb[TCA_CAKE_RTT]) { - q->interval = nla_get_u32(tb[TCA_CAKE_RTT]); + u32 interval = nla_get_u32(tb[TCA_CAKE_RTT]); - if (!q->interval) - q->interval = 1; + WRITE_ONCE(q->interval, max(interval, 1U)); } if (tb[TCA_CAKE_TARGET]) { - q->target = nla_get_u32(tb[TCA_CAKE_TARGET]); + u32 target = nla_get_u32(tb[TCA_CAKE_TARGET]); - if (!q->target) - q->target = 1; + WRITE_ONCE(q->target, max(target, 1U)); } if (tb[TCA_CAKE_AUTORATE]) { if (!!nla_get_u32(tb[TCA_CAKE_AUTORATE])) - q->rate_flags |= CAKE_FLAG_AUTORATE_INGRESS; + rate_flags |= CAKE_FLAG_AUTORATE_INGRESS; else - q->rate_flags &= ~CAKE_FLAG_AUTORATE_INGRESS; + rate_flags &= ~CAKE_FLAG_AUTORATE_INGRESS; } if (tb[TCA_CAKE_INGRESS]) { if (!!nla_get_u32(tb[TCA_CAKE_INGRESS])) - q->rate_flags |= CAKE_FLAG_INGRESS; + rate_flags |= CAKE_FLAG_INGRESS; else - q->rate_flags &= ~CAKE_FLAG_INGRESS; + rate_flags &= ~CAKE_FLAG_INGRESS; } if (tb[TCA_CAKE_ACK_FILTER]) - q->ack_filter = nla_get_u32(tb[TCA_CAKE_ACK_FILTER]); + WRITE_ONCE(q->ack_filter, + nla_get_u32(tb[TCA_CAKE_ACK_FILTER])); if (tb[TCA_CAKE_MEMORY]) - q->buffer_config_limit = nla_get_u32(tb[TCA_CAKE_MEMORY]); + WRITE_ONCE(q->buffer_config_limit, + nla_get_u32(tb[TCA_CAKE_MEMORY])); if (tb[TCA_CAKE_SPLIT_GSO]) { if (!!nla_get_u32(tb[TCA_CAKE_SPLIT_GSO])) - q->rate_flags |= CAKE_FLAG_SPLIT_GSO; + rate_flags |= CAKE_FLAG_SPLIT_GSO; else - q->rate_flags &= ~CAKE_FLAG_SPLIT_GSO; + rate_flags &= ~CAKE_FLAG_SPLIT_GSO; } if (tb[TCA_CAKE_FWMARK]) { - q->fwmark_mask = nla_get_u32(tb[TCA_CAKE_FWMARK]); - q->fwmark_shft = q->fwmark_mask ? __ffs(q->fwmark_mask) : 0; + WRITE_ONCE(q->fwmark_mask, nla_get_u32(tb[TCA_CAKE_FWMARK])); + WRITE_ONCE(q->fwmark_shft, + q->fwmark_mask ? __ffs(q->fwmark_mask) : 0); } + WRITE_ONCE(q->rate_flags, rate_flags); + WRITE_ONCE(q->flow_mode, flow_mode); if (q->tins) { sch_tree_lock(sch); cake_reconfigure(sch); @@ -2774,68 +2786,72 @@ static int cake_dump(struct Qdisc *sch, struct sk_buff *skb) { struct cake_sched_data *q = qdisc_priv(sch); struct nlattr *opts; + u16 rate_flags; + u8 flow_mode; opts = nla_nest_start_noflag(skb, TCA_OPTIONS); if (!opts) goto nla_put_failure; - if (nla_put_u64_64bit(skb, TCA_CAKE_BASE_RATE64, q->rate_bps, - TCA_CAKE_PAD)) + if (nla_put_u64_64bit(skb, TCA_CAKE_BASE_RATE64, + READ_ONCE(q->rate_bps), TCA_CAKE_PAD)) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_FLOW_MODE, - q->flow_mode & CAKE_FLOW_MASK)) + flow_mode = READ_ONCE(q->flow_mode); + if (nla_put_u32(skb, TCA_CAKE_FLOW_MODE, flow_mode & CAKE_FLOW_MASK)) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_RTT, q->interval)) + if (nla_put_u32(skb, TCA_CAKE_RTT, READ_ONCE(q->interval))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_TARGET, q->target)) + if (nla_put_u32(skb, TCA_CAKE_TARGET, READ_ONCE(q->target))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_MEMORY, q->buffer_config_limit)) + if (nla_put_u32(skb, TCA_CAKE_MEMORY, + READ_ONCE(q->buffer_config_limit))) goto nla_put_failure; + rate_flags = READ_ONCE(q->rate_flags); if (nla_put_u32(skb, TCA_CAKE_AUTORATE, - !!(q->rate_flags & CAKE_FLAG_AUTORATE_INGRESS))) + !!(rate_flags & CAKE_FLAG_AUTORATE_INGRESS))) goto nla_put_failure; if (nla_put_u32(skb, TCA_CAKE_INGRESS, - !!(q->rate_flags & CAKE_FLAG_INGRESS))) + !!(rate_flags & CAKE_FLAG_INGRESS))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_ACK_FILTER, q->ack_filter)) + if (nla_put_u32(skb, TCA_CAKE_ACK_FILTER, READ_ONCE(q->ack_filter))) goto nla_put_failure; if (nla_put_u32(skb, TCA_CAKE_NAT, - !!(q->flow_mode & CAKE_FLOW_NAT_FLAG))) + !!(flow_mode & CAKE_FLOW_NAT_FLAG))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_DIFFSERV_MODE, q->tin_mode)) + if (nla_put_u32(skb, TCA_CAKE_DIFFSERV_MODE, READ_ONCE(q->tin_mode))) goto nla_put_failure; if (nla_put_u32(skb, TCA_CAKE_WASH, - !!(q->rate_flags & CAKE_FLAG_WASH))) + !!(rate_flags & CAKE_FLAG_WASH))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_OVERHEAD, q->rate_overhead)) + if (nla_put_u32(skb, TCA_CAKE_OVERHEAD, READ_ONCE(q->rate_overhead))) goto nla_put_failure; - if (!(q->rate_flags & CAKE_FLAG_OVERHEAD)) + if (!(rate_flags & CAKE_FLAG_OVERHEAD)) if (nla_put_u32(skb, TCA_CAKE_RAW, 0)) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_ATM, q->atm_mode)) + if (nla_put_u32(skb, TCA_CAKE_ATM, READ_ONCE(q->atm_mode))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_MPU, q->rate_mpu)) + if (nla_put_u32(skb, TCA_CAKE_MPU, READ_ONCE(q->rate_mpu))) goto nla_put_failure; if (nla_put_u32(skb, TCA_CAKE_SPLIT_GSO, - !!(q->rate_flags & CAKE_FLAG_SPLIT_GSO))) + !!(rate_flags & CAKE_FLAG_SPLIT_GSO))) goto nla_put_failure; - if (nla_put_u32(skb, TCA_CAKE_FWMARK, q->fwmark_mask)) + if (nla_put_u32(skb, TCA_CAKE_FWMARK, READ_ONCE(q->fwmark_mask))) goto nla_put_failure; return nla_nest_end(skb, opts); -- cgit v1.2.3 From 8eb54a421a625a7797549d7bdad3f6c70c49180c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:37 +0000 Subject: net_sched: sch_cbs: implement lockless cbs_dump() Instead of relying on RTNL, cbs_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in cbs_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_cbs.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c index 69001eff0315..939425da1895 100644 --- a/net/sched/sch_cbs.c +++ b/net/sched/sch_cbs.c @@ -389,11 +389,11 @@ static int cbs_change(struct Qdisc *sch, struct nlattr *opt, } /* Everything went OK, save the parameters used. */ - q->hicredit = qopt->hicredit; - q->locredit = qopt->locredit; - q->idleslope = qopt->idleslope * BYTES_PER_KBIT; - q->sendslope = qopt->sendslope * BYTES_PER_KBIT; - q->offload = qopt->offload; + WRITE_ONCE(q->hicredit, qopt->hicredit); + WRITE_ONCE(q->locredit, qopt->locredit); + WRITE_ONCE(q->idleslope, qopt->idleslope * BYTES_PER_KBIT); + WRITE_ONCE(q->sendslope, qopt->sendslope * BYTES_PER_KBIT); + WRITE_ONCE(q->offload, qopt->offload); return 0; } @@ -459,11 +459,11 @@ static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb) if (!nest) goto nla_put_failure; - opt.hicredit = q->hicredit; - opt.locredit = q->locredit; - opt.sendslope = div64_s64(q->sendslope, BYTES_PER_KBIT); - opt.idleslope = div64_s64(q->idleslope, BYTES_PER_KBIT); - opt.offload = q->offload; + opt.hicredit = READ_ONCE(q->hicredit); + opt.locredit = READ_ONCE(q->locredit); + opt.sendslope = div64_s64(READ_ONCE(q->sendslope), BYTES_PER_KBIT); + opt.idleslope = div64_s64(READ_ONCE(q->idleslope), BYTES_PER_KBIT); + opt.offload = READ_ONCE(q->offload); if (nla_put(skb, TCA_CBS_PARMS, sizeof(opt), &opt)) goto nla_put_failure; -- cgit v1.2.3 From 7253c1d1e7a52504ffd08138e9843d0a80788690 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:38 +0000 Subject: net_sched: sch_choke: implement lockless choke_dump() Instead of relying on RTNL, choke_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in choke_change(). v2: added a WRITE_ONCE(p->Scell_log, Scell_log) per Simon feedback in V1 Removed the READ_ONCE(q->limit) in choke_enqueue() Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- include/net/red.h | 12 ++++++------ net/sched/sch_choke.c | 21 +++++++++++---------- 2 files changed, 17 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/net/red.h b/include/net/red.h index 425364de0df7..802287d52c9e 100644 --- a/include/net/red.h +++ b/include/net/red.h @@ -233,10 +233,10 @@ static inline void red_set_parms(struct red_parms *p, int delta = qth_max - qth_min; u32 max_p_delta; - p->qth_min = qth_min << Wlog; - p->qth_max = qth_max << Wlog; - p->Wlog = Wlog; - p->Plog = Plog; + WRITE_ONCE(p->qth_min, qth_min << Wlog); + WRITE_ONCE(p->qth_max, qth_max << Wlog); + WRITE_ONCE(p->Wlog, Wlog); + WRITE_ONCE(p->Plog, Plog); if (delta <= 0) delta = 1; p->qth_delta = delta; @@ -244,7 +244,7 @@ static inline void red_set_parms(struct red_parms *p, max_P = red_maxp(Plog); max_P *= delta; /* max_P = (qth_max - qth_min)/2^Plog */ } - p->max_P = max_P; + WRITE_ONCE(p->max_P, max_P); max_p_delta = max_P / delta; max_p_delta = max(max_p_delta, 1U); p->max_P_reciprocal = reciprocal_value(max_p_delta); @@ -257,7 +257,7 @@ static inline void red_set_parms(struct red_parms *p, p->target_min = qth_min + 2*delta; p->target_max = qth_min + 3*delta; - p->Scell_log = Scell_log; + WRITE_ONCE(p->Scell_log, Scell_log); p->Scell_max = (255 << Scell_log); if (stab) diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index ea108030c6b4..91072010923d 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -405,8 +405,8 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt, } else sch_tree_lock(sch); - q->flags = ctl->flags; - q->limit = ctl->limit; + WRITE_ONCE(q->flags, ctl->flags); + WRITE_ONCE(q->limit, ctl->limit); red_set_parms(&q->parms, ctl->qth_min, ctl->qth_max, ctl->Wlog, ctl->Plog, ctl->Scell_log, @@ -431,15 +431,16 @@ static int choke_init(struct Qdisc *sch, struct nlattr *opt, static int choke_dump(struct Qdisc *sch, struct sk_buff *skb) { struct choke_sched_data *q = qdisc_priv(sch); + u8 Wlog = READ_ONCE(q->parms.Wlog); struct nlattr *opts = NULL; struct tc_red_qopt opt = { - .limit = q->limit, - .flags = q->flags, - .qth_min = q->parms.qth_min >> q->parms.Wlog, - .qth_max = q->parms.qth_max >> q->parms.Wlog, - .Wlog = q->parms.Wlog, - .Plog = q->parms.Plog, - .Scell_log = q->parms.Scell_log, + .limit = READ_ONCE(q->limit), + .flags = READ_ONCE(q->flags), + .qth_min = READ_ONCE(q->parms.qth_min) >> Wlog, + .qth_max = READ_ONCE(q->parms.qth_max) >> Wlog, + .Wlog = Wlog, + .Plog = READ_ONCE(q->parms.Plog), + .Scell_log = READ_ONCE(q->parms.Scell_log), }; opts = nla_nest_start_noflag(skb, TCA_OPTIONS); @@ -447,7 +448,7 @@ static int choke_dump(struct Qdisc *sch, struct sk_buff *skb) goto nla_put_failure; if (nla_put(skb, TCA_CHOKE_PARMS, sizeof(opt), &opt) || - nla_put_u32(skb, TCA_CHOKE_MAX_P, q->parms.max_P)) + nla_put_u32(skb, TCA_CHOKE_MAX_P, READ_ONCE(q->parms.max_P))) goto nla_put_failure; return nla_nest_end(skb, opts); -- cgit v1.2.3 From c45bd26c829ea229c067f4096a746ea2f717bc20 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:39 +0000 Subject: net_sched: sch_codel: implement lockless codel_dump() Instead of relying on RTNL, codel_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in codel_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_codel.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c index ecb3f164bb25..3e8d4fe4d91e 100644 --- a/net/sched/sch_codel.c +++ b/net/sched/sch_codel.c @@ -118,26 +118,31 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt, if (tb[TCA_CODEL_TARGET]) { u32 target = nla_get_u32(tb[TCA_CODEL_TARGET]); - q->params.target = ((u64)target * NSEC_PER_USEC) >> CODEL_SHIFT; + WRITE_ONCE(q->params.target, + ((u64)target * NSEC_PER_USEC) >> CODEL_SHIFT); } if (tb[TCA_CODEL_CE_THRESHOLD]) { u64 val = nla_get_u32(tb[TCA_CODEL_CE_THRESHOLD]); - q->params.ce_threshold = (val * NSEC_PER_USEC) >> CODEL_SHIFT; + WRITE_ONCE(q->params.ce_threshold, + (val * NSEC_PER_USEC) >> CODEL_SHIFT); } if (tb[TCA_CODEL_INTERVAL]) { u32 interval = nla_get_u32(tb[TCA_CODEL_INTERVAL]); - q->params.interval = ((u64)interval * NSEC_PER_USEC) >> CODEL_SHIFT; + WRITE_ONCE(q->params.interval, + ((u64)interval * NSEC_PER_USEC) >> CODEL_SHIFT); } if (tb[TCA_CODEL_LIMIT]) - sch->limit = nla_get_u32(tb[TCA_CODEL_LIMIT]); + WRITE_ONCE(sch->limit, + nla_get_u32(tb[TCA_CODEL_LIMIT])); if (tb[TCA_CODEL_ECN]) - q->params.ecn = !!nla_get_u32(tb[TCA_CODEL_ECN]); + WRITE_ONCE(q->params.ecn, + !!nla_get_u32(tb[TCA_CODEL_ECN])); qlen = sch->q.qlen; while (sch->q.qlen > sch->limit) { @@ -183,6 +188,7 @@ static int codel_init(struct Qdisc *sch, struct nlattr *opt, static int codel_dump(struct Qdisc *sch, struct sk_buff *skb) { struct codel_sched_data *q = qdisc_priv(sch); + codel_time_t ce_threshold; struct nlattr *opts; opts = nla_nest_start_noflag(skb, TCA_OPTIONS); @@ -190,17 +196,18 @@ static int codel_dump(struct Qdisc *sch, struct sk_buff *skb) goto nla_put_failure; if (nla_put_u32(skb, TCA_CODEL_TARGET, - codel_time_to_us(q->params.target)) || + codel_time_to_us(READ_ONCE(q->params.target))) || nla_put_u32(skb, TCA_CODEL_LIMIT, - sch->limit) || + READ_ONCE(sch->limit)) || nla_put_u32(skb, TCA_CODEL_INTERVAL, - codel_time_to_us(q->params.interval)) || + codel_time_to_us(READ_ONCE(q->params.interval))) || nla_put_u32(skb, TCA_CODEL_ECN, - q->params.ecn)) + READ_ONCE(q->params.ecn))) goto nla_put_failure; - if (q->params.ce_threshold != CODEL_DISABLED_THRESHOLD && + ce_threshold = READ_ONCE(q->params.ce_threshold); + if (ce_threshold != CODEL_DISABLED_THRESHOLD && nla_put_u32(skb, TCA_CODEL_CE_THRESHOLD, - codel_time_to_us(q->params.ce_threshold))) + codel_time_to_us(ce_threshold))) goto nla_put_failure; return nla_nest_end(skb, opts); -- cgit v1.2.3 From a1ac3a7c3d1e1f80965510cec7f2c3d4eba57d37 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:40 +0000 Subject: net_sched: sch_tfs: implement lockless etf_dump() Instead of relying on RTNL, codel_dump() can use READ_ONCE() annotations. There is no etf_change() yet, this patch imply aligns this qdisc with others. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_etf.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sched/sch_etf.c b/net/sched/sch_etf.c index 2e4bef713b6a..c74d778c32a1 100644 --- a/net/sched/sch_etf.c +++ b/net/sched/sch_etf.c @@ -467,15 +467,15 @@ static int etf_dump(struct Qdisc *sch, struct sk_buff *skb) if (!nest) goto nla_put_failure; - opt.delta = q->delta; - opt.clockid = q->clockid; - if (q->offload) + opt.delta = READ_ONCE(q->delta); + opt.clockid = READ_ONCE(q->clockid); + if (READ_ONCE(q->offload)) opt.flags |= TC_ETF_OFFLOAD_ON; - if (q->deadline_mode) + if (READ_ONCE(q->deadline_mode)) opt.flags |= TC_ETF_DEADLINE_MODE_ON; - if (q->skip_sock_check) + if (READ_ONCE(q->skip_sock_check)) opt.flags |= TC_ETF_SKIP_SOCK_CHECK; if (nla_put(skb, TCA_ETF_PARMS, sizeof(opt), &opt)) -- cgit v1.2.3 From c5f1dde7f731e7bf2e7c169ca42cb4989fc2f8b9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:41 +0000 Subject: net_sched: sch_ets: implement lockless ets_dump() Instead of relying on RTNL, ets_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in ets_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_ets.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/sched/sch_ets.c b/net/sched/sch_ets.c index 835b4460b448..f80bc05d4c5a 100644 --- a/net/sched/sch_ets.c +++ b/net/sched/sch_ets.c @@ -646,7 +646,7 @@ static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt, sch_tree_lock(sch); - q->nbands = nbands; + WRITE_ONCE(q->nbands, nbands); for (i = nstrict; i < q->nstrict; i++) { if (q->classes[i].qdisc->q.qlen) { list_add_tail(&q->classes[i].alist, &q->active); @@ -658,11 +658,11 @@ static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt, list_del(&q->classes[i].alist); qdisc_tree_flush_backlog(q->classes[i].qdisc); } - q->nstrict = nstrict; + WRITE_ONCE(q->nstrict, nstrict); memcpy(q->prio2band, priomap, sizeof(priomap)); for (i = 0; i < q->nbands; i++) - q->classes[i].quantum = quanta[i]; + WRITE_ONCE(q->classes[i].quantum, quanta[i]); for (i = oldbands; i < q->nbands; i++) { q->classes[i].qdisc = queues[i]; @@ -676,7 +676,7 @@ static int ets_qdisc_change(struct Qdisc *sch, struct nlattr *opt, for (i = q->nbands; i < oldbands; i++) { qdisc_put(q->classes[i].qdisc); q->classes[i].qdisc = NULL; - q->classes[i].quantum = 0; + WRITE_ONCE(q->classes[i].quantum, 0); q->classes[i].deficit = 0; gnet_stats_basic_sync_init(&q->classes[i].bstats); memset(&q->classes[i].qstats, 0, sizeof(q->classes[i].qstats)); @@ -733,6 +733,7 @@ static int ets_qdisc_dump(struct Qdisc *sch, struct sk_buff *skb) struct ets_sched *q = qdisc_priv(sch); struct nlattr *opts; struct nlattr *nest; + u8 nbands, nstrict; int band; int prio; int err; @@ -745,21 +746,22 @@ static int ets_qdisc_dump(struct Qdisc *sch, struct sk_buff *skb) if (!opts) goto nla_err; - if (nla_put_u8(skb, TCA_ETS_NBANDS, q->nbands)) + nbands = READ_ONCE(q->nbands); + if (nla_put_u8(skb, TCA_ETS_NBANDS, nbands)) goto nla_err; - if (q->nstrict && - nla_put_u8(skb, TCA_ETS_NSTRICT, q->nstrict)) + nstrict = READ_ONCE(q->nstrict); + if (nstrict && nla_put_u8(skb, TCA_ETS_NSTRICT, nstrict)) goto nla_err; - if (q->nbands > q->nstrict) { + if (nbands > nstrict) { nest = nla_nest_start(skb, TCA_ETS_QUANTA); if (!nest) goto nla_err; - for (band = q->nstrict; band < q->nbands; band++) { + for (band = nstrict; band < nbands; band++) { if (nla_put_u32(skb, TCA_ETS_QUANTA_BAND, - q->classes[band].quantum)) + READ_ONCE(q->classes[band].quantum))) goto nla_err; } @@ -771,7 +773,8 @@ static int ets_qdisc_dump(struct Qdisc *sch, struct sk_buff *skb) goto nla_err; for (prio = 0; prio <= TC_PRIO_MAX; prio++) { - if (nla_put_u8(skb, TCA_ETS_PRIOMAP_BAND, q->prio2band[prio])) + if (nla_put_u8(skb, TCA_ETS_PRIOMAP_BAND, + READ_ONCE(q->prio2band[prio]))) goto nla_err; } -- cgit v1.2.3 From 01daf66b791e1ddd1a4aae2a65ea65fa1d4a96ef Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:42 +0000 Subject: net_sched: sch_fifo: implement lockless __fifo_dump() Instead of relying on RTNL, __fifo_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in __fifo_init(). Also add missing READ_ONCE(sh->limit) in bfifo_enqueue(), pfifo_enqueue() and pfifo_tail_enqueue(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_fifo.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 450f5c67ac49..b50b2c2cc09b 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -19,7 +19,8 @@ static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { - if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= sch->limit)) + if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= + READ_ONCE(sch->limit))) return qdisc_enqueue_tail(skb, sch); return qdisc_drop(skb, sch, to_free); @@ -28,7 +29,7 @@ static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { - if (likely(sch->q.qlen < sch->limit)) + if (likely(sch->q.qlen < READ_ONCE(sch->limit))) return qdisc_enqueue_tail(skb, sch); return qdisc_drop(skb, sch, to_free); @@ -39,7 +40,7 @@ static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch, { unsigned int prev_backlog; - if (likely(sch->q.qlen < sch->limit)) + if (likely(sch->q.qlen < READ_ONCE(sch->limit))) return qdisc_enqueue_tail(skb, sch); prev_backlog = sch->qstats.backlog; @@ -105,14 +106,14 @@ static int __fifo_init(struct Qdisc *sch, struct nlattr *opt, if (is_bfifo) limit *= psched_mtu(qdisc_dev(sch)); - sch->limit = limit; + WRITE_ONCE(sch->limit, limit); } else { struct tc_fifo_qopt *ctl = nla_data(opt); if (nla_len(opt) < sizeof(*ctl)) return -EINVAL; - sch->limit = ctl->limit; + WRITE_ONCE(sch->limit, ctl->limit); } if (is_bfifo) @@ -154,7 +155,7 @@ static void fifo_destroy(struct Qdisc *sch) static int __fifo_dump(struct Qdisc *sch, struct sk_buff *skb) { - struct tc_fifo_qopt opt = { .limit = sch->limit }; + struct tc_fifo_qopt opt = { .limit = READ_ONCE(sch->limit) }; if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) goto nla_put_failure; -- cgit v1.2.3 From 396a0038508aa7e380e317a80cc905b794e83191 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:43 +0000 Subject: net_sched: sch_fq_codel: implement lockless fq_codel_dump() Instead of relying on RTNL, fq_codel_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in fq_codel_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_fq_codel.c | 57 +++++++++++++++++++++++++++++------------------- 1 file changed, 35 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 79f9d6de6c85..4f908c11ba95 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -396,40 +396,49 @@ static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt, if (tb[TCA_FQ_CODEL_TARGET]) { u64 target = nla_get_u32(tb[TCA_FQ_CODEL_TARGET]); - q->cparams.target = (target * NSEC_PER_USEC) >> CODEL_SHIFT; + WRITE_ONCE(q->cparams.target, + (target * NSEC_PER_USEC) >> CODEL_SHIFT); } if (tb[TCA_FQ_CODEL_CE_THRESHOLD]) { u64 val = nla_get_u32(tb[TCA_FQ_CODEL_CE_THRESHOLD]); - q->cparams.ce_threshold = (val * NSEC_PER_USEC) >> CODEL_SHIFT; + WRITE_ONCE(q->cparams.ce_threshold, + (val * NSEC_PER_USEC) >> CODEL_SHIFT); } if (tb[TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR]) - q->cparams.ce_threshold_selector = nla_get_u8(tb[TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR]); + WRITE_ONCE(q->cparams.ce_threshold_selector, + nla_get_u8(tb[TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR])); if (tb[TCA_FQ_CODEL_CE_THRESHOLD_MASK]) - q->cparams.ce_threshold_mask = nla_get_u8(tb[TCA_FQ_CODEL_CE_THRESHOLD_MASK]); + WRITE_ONCE(q->cparams.ce_threshold_mask, + nla_get_u8(tb[TCA_FQ_CODEL_CE_THRESHOLD_MASK])); if (tb[TCA_FQ_CODEL_INTERVAL]) { u64 interval = nla_get_u32(tb[TCA_FQ_CODEL_INTERVAL]); - q->cparams.interval = (interval * NSEC_PER_USEC) >> CODEL_SHIFT; + WRITE_ONCE(q->cparams.interval, + (interval * NSEC_PER_USEC) >> CODEL_SHIFT); } if (tb[TCA_FQ_CODEL_LIMIT]) - sch->limit = nla_get_u32(tb[TCA_FQ_CODEL_LIMIT]); + WRITE_ONCE(sch->limit, + nla_get_u32(tb[TCA_FQ_CODEL_LIMIT])); if (tb[TCA_FQ_CODEL_ECN]) - q->cparams.ecn = !!nla_get_u32(tb[TCA_FQ_CODEL_ECN]); + WRITE_ONCE(q->cparams.ecn, + !!nla_get_u32(tb[TCA_FQ_CODEL_ECN])); if (quantum) - q->quantum = quantum; + WRITE_ONCE(q->quantum, quantum); if (tb[TCA_FQ_CODEL_DROP_BATCH_SIZE]) - q->drop_batch_size = max(1U, nla_get_u32(tb[TCA_FQ_CODEL_DROP_BATCH_SIZE])); + WRITE_ONCE(q->drop_batch_size, + max(1U, nla_get_u32(tb[TCA_FQ_CODEL_DROP_BATCH_SIZE]))); if (tb[TCA_FQ_CODEL_MEMORY_LIMIT]) - q->memory_limit = min(1U << 31, nla_get_u32(tb[TCA_FQ_CODEL_MEMORY_LIMIT])); + WRITE_ONCE(q->memory_limit, + min(1U << 31, nla_get_u32(tb[TCA_FQ_CODEL_MEMORY_LIMIT]))); while (sch->q.qlen > sch->limit || q->memory_usage > q->memory_limit) { @@ -522,6 +531,7 @@ init_failure: static int fq_codel_dump(struct Qdisc *sch, struct sk_buff *skb) { struct fq_codel_sched_data *q = qdisc_priv(sch); + codel_time_t ce_threshold; struct nlattr *opts; opts = nla_nest_start_noflag(skb, TCA_OPTIONS); @@ -529,30 +539,33 @@ static int fq_codel_dump(struct Qdisc *sch, struct sk_buff *skb) goto nla_put_failure; if (nla_put_u32(skb, TCA_FQ_CODEL_TARGET, - codel_time_to_us(q->cparams.target)) || + codel_time_to_us(READ_ONCE(q->cparams.target))) || nla_put_u32(skb, TCA_FQ_CODEL_LIMIT, - sch->limit) || + READ_ONCE(sch->limit)) || nla_put_u32(skb, TCA_FQ_CODEL_INTERVAL, - codel_time_to_us(q->cparams.interval)) || + codel_time_to_us(READ_ONCE(q->cparams.interval))) || nla_put_u32(skb, TCA_FQ_CODEL_ECN, - q->cparams.ecn) || + READ_ONCE(q->cparams.ecn)) || nla_put_u32(skb, TCA_FQ_CODEL_QUANTUM, - q->quantum) || + READ_ONCE(q->quantum)) || nla_put_u32(skb, TCA_FQ_CODEL_DROP_BATCH_SIZE, - q->drop_batch_size) || + READ_ONCE(q->drop_batch_size)) || nla_put_u32(skb, TCA_FQ_CODEL_MEMORY_LIMIT, - q->memory_limit) || + READ_ONCE(q->memory_limit)) || nla_put_u32(skb, TCA_FQ_CODEL_FLOWS, - q->flows_cnt)) + READ_ONCE(q->flows_cnt))) goto nla_put_failure; - if (q->cparams.ce_threshold != CODEL_DISABLED_THRESHOLD) { + ce_threshold = READ_ONCE(q->cparams.ce_threshold); + if (ce_threshold != CODEL_DISABLED_THRESHOLD) { if (nla_put_u32(skb, TCA_FQ_CODEL_CE_THRESHOLD, - codel_time_to_us(q->cparams.ce_threshold))) + codel_time_to_us(ce_threshold))) goto nla_put_failure; - if (nla_put_u8(skb, TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR, q->cparams.ce_threshold_selector)) + if (nla_put_u8(skb, TCA_FQ_CODEL_CE_THRESHOLD_SELECTOR, + READ_ONCE(q->cparams.ce_threshold_selector))) goto nla_put_failure; - if (nla_put_u8(skb, TCA_FQ_CODEL_CE_THRESHOLD_MASK, q->cparams.ce_threshold_mask)) + if (nla_put_u8(skb, TCA_FQ_CODEL_CE_THRESHOLD_MASK, + READ_ONCE(q->cparams.ce_threshold_mask))) goto nla_put_failure; } -- cgit v1.2.3 From 13a9965de32457d59aa3162d657aa4c5e512c146 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:44 +0000 Subject: net_sched: sch_fq_pie: implement lockless fq_pie_dump() Instead of relying on RTNL, fq_pie_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in fq_pie_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_fq_pie.c | 61 ++++++++++++++++++++++++++++---------------------- 1 file changed, 34 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/net/sched/sch_fq_pie.c b/net/sched/sch_fq_pie.c index 358cf304f4c9..c38f33ff80bd 100644 --- a/net/sched/sch_fq_pie.c +++ b/net/sched/sch_fq_pie.c @@ -299,8 +299,8 @@ static int fq_pie_change(struct Qdisc *sch, struct nlattr *opt, if (tb[TCA_FQ_PIE_LIMIT]) { u32 limit = nla_get_u32(tb[TCA_FQ_PIE_LIMIT]); - q->p_params.limit = limit; - sch->limit = limit; + WRITE_ONCE(q->p_params.limit, limit); + WRITE_ONCE(sch->limit, limit); } if (tb[TCA_FQ_PIE_FLOWS]) { if (q->flows) { @@ -322,39 +322,45 @@ static int fq_pie_change(struct Qdisc *sch, struct nlattr *opt, u32 target = nla_get_u32(tb[TCA_FQ_PIE_TARGET]); /* convert to pschedtime */ - q->p_params.target = - PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC); + WRITE_ONCE(q->p_params.target, + PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC)); } /* tupdate is in jiffies */ if (tb[TCA_FQ_PIE_TUPDATE]) - q->p_params.tupdate = - usecs_to_jiffies(nla_get_u32(tb[TCA_FQ_PIE_TUPDATE])); + WRITE_ONCE(q->p_params.tupdate, + usecs_to_jiffies(nla_get_u32(tb[TCA_FQ_PIE_TUPDATE]))); if (tb[TCA_FQ_PIE_ALPHA]) - q->p_params.alpha = nla_get_u32(tb[TCA_FQ_PIE_ALPHA]); + WRITE_ONCE(q->p_params.alpha, + nla_get_u32(tb[TCA_FQ_PIE_ALPHA])); if (tb[TCA_FQ_PIE_BETA]) - q->p_params.beta = nla_get_u32(tb[TCA_FQ_PIE_BETA]); + WRITE_ONCE(q->p_params.beta, + nla_get_u32(tb[TCA_FQ_PIE_BETA])); if (tb[TCA_FQ_PIE_QUANTUM]) - q->quantum = nla_get_u32(tb[TCA_FQ_PIE_QUANTUM]); + WRITE_ONCE(q->quantum, nla_get_u32(tb[TCA_FQ_PIE_QUANTUM])); if (tb[TCA_FQ_PIE_MEMORY_LIMIT]) - q->memory_limit = nla_get_u32(tb[TCA_FQ_PIE_MEMORY_LIMIT]); + WRITE_ONCE(q->memory_limit, + nla_get_u32(tb[TCA_FQ_PIE_MEMORY_LIMIT])); if (tb[TCA_FQ_PIE_ECN_PROB]) - q->ecn_prob = nla_get_u32(tb[TCA_FQ_PIE_ECN_PROB]); + WRITE_ONCE(q->ecn_prob, + nla_get_u32(tb[TCA_FQ_PIE_ECN_PROB])); if (tb[TCA_FQ_PIE_ECN]) - q->p_params.ecn = nla_get_u32(tb[TCA_FQ_PIE_ECN]); + WRITE_ONCE(q->p_params.ecn, + nla_get_u32(tb[TCA_FQ_PIE_ECN])); if (tb[TCA_FQ_PIE_BYTEMODE]) - q->p_params.bytemode = nla_get_u32(tb[TCA_FQ_PIE_BYTEMODE]); + WRITE_ONCE(q->p_params.bytemode, + nla_get_u32(tb[TCA_FQ_PIE_BYTEMODE])); if (tb[TCA_FQ_PIE_DQ_RATE_ESTIMATOR]) - q->p_params.dq_rate_estimator = - nla_get_u32(tb[TCA_FQ_PIE_DQ_RATE_ESTIMATOR]); + WRITE_ONCE(q->p_params.dq_rate_estimator, + nla_get_u32(tb[TCA_FQ_PIE_DQ_RATE_ESTIMATOR])); /* Drop excess packets if new limit is lower */ while (sch->q.qlen > sch->limit) { @@ -471,22 +477,23 @@ static int fq_pie_dump(struct Qdisc *sch, struct sk_buff *skb) return -EMSGSIZE; /* convert target from pschedtime to us */ - if (nla_put_u32(skb, TCA_FQ_PIE_LIMIT, sch->limit) || - nla_put_u32(skb, TCA_FQ_PIE_FLOWS, q->flows_cnt) || + if (nla_put_u32(skb, TCA_FQ_PIE_LIMIT, READ_ONCE(sch->limit)) || + nla_put_u32(skb, TCA_FQ_PIE_FLOWS, READ_ONCE(q->flows_cnt)) || nla_put_u32(skb, TCA_FQ_PIE_TARGET, - ((u32)PSCHED_TICKS2NS(q->p_params.target)) / + ((u32)PSCHED_TICKS2NS(READ_ONCE(q->p_params.target))) / NSEC_PER_USEC) || nla_put_u32(skb, TCA_FQ_PIE_TUPDATE, - jiffies_to_usecs(q->p_params.tupdate)) || - nla_put_u32(skb, TCA_FQ_PIE_ALPHA, q->p_params.alpha) || - nla_put_u32(skb, TCA_FQ_PIE_BETA, q->p_params.beta) || - nla_put_u32(skb, TCA_FQ_PIE_QUANTUM, q->quantum) || - nla_put_u32(skb, TCA_FQ_PIE_MEMORY_LIMIT, q->memory_limit) || - nla_put_u32(skb, TCA_FQ_PIE_ECN_PROB, q->ecn_prob) || - nla_put_u32(skb, TCA_FQ_PIE_ECN, q->p_params.ecn) || - nla_put_u32(skb, TCA_FQ_PIE_BYTEMODE, q->p_params.bytemode) || + jiffies_to_usecs(READ_ONCE(q->p_params.tupdate))) || + nla_put_u32(skb, TCA_FQ_PIE_ALPHA, READ_ONCE(q->p_params.alpha)) || + nla_put_u32(skb, TCA_FQ_PIE_BETA, READ_ONCE(q->p_params.beta)) || + nla_put_u32(skb, TCA_FQ_PIE_QUANTUM, READ_ONCE(q->quantum)) || + nla_put_u32(skb, TCA_FQ_PIE_MEMORY_LIMIT, + READ_ONCE(q->memory_limit)) || + nla_put_u32(skb, TCA_FQ_PIE_ECN_PROB, READ_ONCE(q->ecn_prob)) || + nla_put_u32(skb, TCA_FQ_PIE_ECN, READ_ONCE(q->p_params.ecn)) || + nla_put_u32(skb, TCA_FQ_PIE_BYTEMODE, READ_ONCE(q->p_params.bytemode)) || nla_put_u32(skb, TCA_FQ_PIE_DQ_RATE_ESTIMATOR, - q->p_params.dq_rate_estimator)) + READ_ONCE(q->p_params.dq_rate_estimator))) goto nla_put_failure; return nla_nest_end(skb, opts); -- cgit v1.2.3 From 49e8ae53700212fbb26302df219922752e81fbd2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:45 +0000 Subject: net_sched: sch_hfsc: implement lockless accesses to q->defcls Instead of relying on RTNL, hfsc_dump_qdisc() can use READ_ONCE() annotation, paired with WRITE_ONCE() one in hfsc_change_qdisc(). Use READ_ONCE(q->defcls) in hfsc_classify() to no longer acquire qdisc lock from hfsc_change_qdisc(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_hfsc.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 4e626df742d7..c287bf8423b4 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1174,7 +1174,8 @@ hfsc_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) } /* classification failed, try default class */ - cl = hfsc_find_class(TC_H_MAKE(TC_H_MAJ(sch->handle), q->defcls), sch); + cl = hfsc_find_class(TC_H_MAKE(TC_H_MAJ(sch->handle), + READ_ONCE(q->defcls)), sch); if (cl == NULL || cl->level > 0) return NULL; @@ -1443,9 +1444,7 @@ hfsc_change_qdisc(struct Qdisc *sch, struct nlattr *opt, return -EINVAL; qopt = nla_data(opt); - sch_tree_lock(sch); - q->defcls = qopt->defcls; - sch_tree_unlock(sch); + WRITE_ONCE(q->defcls, qopt->defcls); return 0; } @@ -1525,7 +1524,7 @@ hfsc_dump_qdisc(struct Qdisc *sch, struct sk_buff *skb) unsigned char *b = skb_tail_pointer(skb); struct tc_hfsc_qopt qopt; - qopt.defcls = q->defcls; + qopt.defcls = READ_ONCE(q->defcls); if (nla_put(skb, TCA_OPTIONS, sizeof(qopt), &qopt)) goto nla_put_failure; return skb->len; -- cgit v1.2.3 From 293c7e2b3e2fb7796778e3e61d94d58546fe96eb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:46 +0000 Subject: net_sched: sch_hhf: implement lockless hhf_dump() Instead of relying on RTNL, hhf_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in hhf_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_hhf.c | 35 +++++++++++++++++++++-------------- 1 file changed, 21 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index 3f906df1435b..44d9efe1a96a 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -534,27 +534,31 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt, sch_tree_lock(sch); if (tb[TCA_HHF_BACKLOG_LIMIT]) - sch->limit = nla_get_u32(tb[TCA_HHF_BACKLOG_LIMIT]); + WRITE_ONCE(sch->limit, nla_get_u32(tb[TCA_HHF_BACKLOG_LIMIT])); - q->quantum = new_quantum; - q->hhf_non_hh_weight = new_hhf_non_hh_weight; + WRITE_ONCE(q->quantum, new_quantum); + WRITE_ONCE(q->hhf_non_hh_weight, new_hhf_non_hh_weight); if (tb[TCA_HHF_HH_FLOWS_LIMIT]) - q->hh_flows_limit = nla_get_u32(tb[TCA_HHF_HH_FLOWS_LIMIT]); + WRITE_ONCE(q->hh_flows_limit, + nla_get_u32(tb[TCA_HHF_HH_FLOWS_LIMIT])); if (tb[TCA_HHF_RESET_TIMEOUT]) { u32 us = nla_get_u32(tb[TCA_HHF_RESET_TIMEOUT]); - q->hhf_reset_timeout = usecs_to_jiffies(us); + WRITE_ONCE(q->hhf_reset_timeout, + usecs_to_jiffies(us)); } if (tb[TCA_HHF_ADMIT_BYTES]) - q->hhf_admit_bytes = nla_get_u32(tb[TCA_HHF_ADMIT_BYTES]); + WRITE_ONCE(q->hhf_admit_bytes, + nla_get_u32(tb[TCA_HHF_ADMIT_BYTES])); if (tb[TCA_HHF_EVICT_TIMEOUT]) { u32 us = nla_get_u32(tb[TCA_HHF_EVICT_TIMEOUT]); - q->hhf_evict_timeout = usecs_to_jiffies(us); + WRITE_ONCE(q->hhf_evict_timeout, + usecs_to_jiffies(us)); } qlen = sch->q.qlen; @@ -657,15 +661,18 @@ static int hhf_dump(struct Qdisc *sch, struct sk_buff *skb) if (opts == NULL) goto nla_put_failure; - if (nla_put_u32(skb, TCA_HHF_BACKLOG_LIMIT, sch->limit) || - nla_put_u32(skb, TCA_HHF_QUANTUM, q->quantum) || - nla_put_u32(skb, TCA_HHF_HH_FLOWS_LIMIT, q->hh_flows_limit) || + if (nla_put_u32(skb, TCA_HHF_BACKLOG_LIMIT, READ_ONCE(sch->limit)) || + nla_put_u32(skb, TCA_HHF_QUANTUM, READ_ONCE(q->quantum)) || + nla_put_u32(skb, TCA_HHF_HH_FLOWS_LIMIT, + READ_ONCE(q->hh_flows_limit)) || nla_put_u32(skb, TCA_HHF_RESET_TIMEOUT, - jiffies_to_usecs(q->hhf_reset_timeout)) || - nla_put_u32(skb, TCA_HHF_ADMIT_BYTES, q->hhf_admit_bytes) || + jiffies_to_usecs(READ_ONCE(q->hhf_reset_timeout))) || + nla_put_u32(skb, TCA_HHF_ADMIT_BYTES, + READ_ONCE(q->hhf_admit_bytes)) || nla_put_u32(skb, TCA_HHF_EVICT_TIMEOUT, - jiffies_to_usecs(q->hhf_evict_timeout)) || - nla_put_u32(skb, TCA_HHF_NON_HH_WEIGHT, q->hhf_non_hh_weight)) + jiffies_to_usecs(READ_ONCE(q->hhf_evict_timeout))) || + nla_put_u32(skb, TCA_HHF_NON_HH_WEIGHT, + READ_ONCE(q->hhf_non_hh_weight))) goto nla_put_failure; return nla_nest_end(skb, opts); -- cgit v1.2.3 From 6c00dc4fdb40fa434cb1271fac3a1201449cb4d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:47 +0000 Subject: net_sched: sch_pie: implement lockless pie_dump() Instead of relying on RTNL, pie_dump() can use READ_ONCE() annotations, paired with WRITE_ONCE() ones in pie_change(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_pie.c | 39 +++++++++++++++++++++------------------ 1 file changed, 21 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c index 1764059b0635..b3dcb845b327 100644 --- a/net/sched/sch_pie.c +++ b/net/sched/sch_pie.c @@ -156,36 +156,38 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt, u32 target = nla_get_u32(tb[TCA_PIE_TARGET]); /* convert to pschedtime */ - q->params.target = PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC); + WRITE_ONCE(q->params.target, + PSCHED_NS2TICKS((u64)target * NSEC_PER_USEC)); } /* tupdate is in jiffies */ if (tb[TCA_PIE_TUPDATE]) - q->params.tupdate = - usecs_to_jiffies(nla_get_u32(tb[TCA_PIE_TUPDATE])); + WRITE_ONCE(q->params.tupdate, + usecs_to_jiffies(nla_get_u32(tb[TCA_PIE_TUPDATE]))); if (tb[TCA_PIE_LIMIT]) { u32 limit = nla_get_u32(tb[TCA_PIE_LIMIT]); - q->params.limit = limit; - sch->limit = limit; + WRITE_ONCE(q->params.limit, limit); + WRITE_ONCE(sch->limit, limit); } if (tb[TCA_PIE_ALPHA]) - q->params.alpha = nla_get_u32(tb[TCA_PIE_ALPHA]); + WRITE_ONCE(q->params.alpha, nla_get_u32(tb[TCA_PIE_ALPHA])); if (tb[TCA_PIE_BETA]) - q->params.beta = nla_get_u32(tb[TCA_PIE_BETA]); + WRITE_ONCE(q->params.beta, nla_get_u32(tb[TCA_PIE_BETA])); if (tb[TCA_PIE_ECN]) - q->params.ecn = nla_get_u32(tb[TCA_PIE_ECN]); + WRITE_ONCE(q->params.ecn, nla_get_u32(tb[TCA_PIE_ECN])); if (tb[TCA_PIE_BYTEMODE]) - q->params.bytemode = nla_get_u32(tb[TCA_PIE_BYTEMODE]); + WRITE_ONCE(q->params.bytemode, + nla_get_u32(tb[TCA_PIE_BYTEMODE])); if (tb[TCA_PIE_DQ_RATE_ESTIMATOR]) - q->params.dq_rate_estimator = - nla_get_u32(tb[TCA_PIE_DQ_RATE_ESTIMATOR]); + WRITE_ONCE(q->params.dq_rate_estimator, + nla_get_u32(tb[TCA_PIE_DQ_RATE_ESTIMATOR])); /* Drop excess packets if new limit is lower */ qlen = sch->q.qlen; @@ -469,17 +471,18 @@ static int pie_dump(struct Qdisc *sch, struct sk_buff *skb) /* convert target from pschedtime to us */ if (nla_put_u32(skb, TCA_PIE_TARGET, - ((u32)PSCHED_TICKS2NS(q->params.target)) / + ((u32)PSCHED_TICKS2NS(READ_ONCE(q->params.target))) / NSEC_PER_USEC) || - nla_put_u32(skb, TCA_PIE_LIMIT, sch->limit) || + nla_put_u32(skb, TCA_PIE_LIMIT, READ_ONCE(sch->limit)) || nla_put_u32(skb, TCA_PIE_TUPDATE, - jiffies_to_usecs(q->params.tupdate)) || - nla_put_u32(skb, TCA_PIE_ALPHA, q->params.alpha) || - nla_put_u32(skb, TCA_PIE_BETA, q->params.beta) || + jiffies_to_usecs(READ_ONCE(q->params.tupdate))) || + nla_put_u32(skb, TCA_PIE_ALPHA, READ_ONCE(q->params.alpha)) || + nla_put_u32(skb, TCA_PIE_BETA, READ_ONCE(q->params.beta)) || nla_put_u32(skb, TCA_PIE_ECN, q->params.ecn) || - nla_put_u32(skb, TCA_PIE_BYTEMODE, q->params.bytemode) || + nla_put_u32(skb, TCA_PIE_BYTEMODE, + READ_ONCE(q->params.bytemode)) || nla_put_u32(skb, TCA_PIE_DQ_RATE_ESTIMATOR, - q->params.dq_rate_estimator)) + READ_ONCE(q->params.dq_rate_estimator))) goto nla_put_failure; return nla_nest_end(skb, opts); -- cgit v1.2.3 From c85cedb38f41bebc3da08c4a7bc58c5f62a2a950 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 07:32:48 +0000 Subject: net_sched: sch_skbprio: implement lockless skbprio_dump() Instead of relying on RTNL, skbprio_dump() can use READ_ONCE() annotation, paired with WRITE_ONCE() one in skbprio_change(). Also add a READ_ONCE(sch->limit) in skbprio_enqueue(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_skbprio.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sched/sch_skbprio.c b/net/sched/sch_skbprio.c index b4dd626c309c..20ff7386b74b 100644 --- a/net/sched/sch_skbprio.c +++ b/net/sched/sch_skbprio.c @@ -79,7 +79,9 @@ static int skbprio_enqueue(struct sk_buff *skb, struct Qdisc *sch, prio = min(skb->priority, max_priority); qdisc = &q->qdiscs[prio]; - if (sch->q.qlen < sch->limit) { + + /* sch->limit can change under us from skbprio_change() */ + if (sch->q.qlen < READ_ONCE(sch->limit)) { __skb_queue_tail(qdisc, skb); qdisc_qstats_backlog_inc(sch, skb); q->qstats[prio].backlog += qdisc_pkt_len(skb); @@ -172,7 +174,7 @@ static int skbprio_change(struct Qdisc *sch, struct nlattr *opt, if (opt->nla_len != nla_attr_size(sizeof(*ctl))) return -EINVAL; - sch->limit = ctl->limit; + WRITE_ONCE(sch->limit, ctl->limit); return 0; } @@ -200,7 +202,7 @@ static int skbprio_dump(struct Qdisc *sch, struct sk_buff *skb) { struct tc_skbprio_qopt opt; - opt.limit = sch->limit; + opt.limit = READ_ONCE(sch->limit); if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) return -1; -- cgit v1.2.3 From 84b6823cd96b38c40b3b30beabbfa48d92990e1a Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 18 Apr 2024 15:36:01 +0800 Subject: net: rps: protect last_qtail with rps_input_queue_tail_save() helper Removing one unnecessary reader protection and add another writer protection to finish the locklessly proctection job. Note: the removed READ_ONCE() is not needed because we only have to protect the locklessly reader in the different context (rps_may_expire_flow()). Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index f6c6e494f0a9..6f027f676243 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4507,7 +4507,7 @@ set_rps_cpu(struct net_device *dev, struct sk_buff *skb, struct netdev_rx_queue *rxqueue; struct rps_dev_flow_table *flow_table; struct rps_dev_flow *old_rflow; - u32 flow_id; + u32 flow_id, head; u16 rxq_index; int rc; @@ -4535,8 +4535,8 @@ set_rps_cpu(struct net_device *dev, struct sk_buff *skb, old_rflow->filter = RPS_NO_FILTER; out: #endif - rflow->last_qtail = - READ_ONCE(per_cpu(softnet_data, next_cpu).input_queue_head); + head = READ_ONCE(per_cpu(softnet_data, next_cpu).input_queue_head); + rps_input_queue_tail_save(&rflow->last_qtail, head); } rflow->cpu = next_cpu; @@ -4619,7 +4619,7 @@ static int get_rps_cpu(struct net_device *dev, struct sk_buff *skb, if (unlikely(tcpu != next_cpu) && (tcpu >= nr_cpu_ids || !cpu_online(tcpu) || ((int)(READ_ONCE(per_cpu(softnet_data, tcpu).input_queue_head) - - READ_ONCE(rflow->last_qtail))) >= 0)) { + rflow->last_qtail)) >= 0)) { tcpu = next_cpu; rflow = set_rps_cpu(dev, skb, rflow, next_cpu); } -- cgit v1.2.3 From f00bf5dc83202fd9f75bde80e46c3a747c34cc6a Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 18 Apr 2024 15:36:02 +0800 Subject: net: rps: protect filter locklessly As we can see, rflow->filter can be written/read concurrently, so lockless access is needed. Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 6f027f676243..182fce2537d4 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4530,9 +4530,9 @@ set_rps_cpu(struct net_device *dev, struct sk_buff *skb, goto out; old_rflow = rflow; rflow = &flow_table->flows[flow_id]; - rflow->filter = rc; - if (old_rflow->filter == rflow->filter) - old_rflow->filter = RPS_NO_FILTER; + WRITE_ONCE(rflow->filter, rc); + if (old_rflow->filter == rc) + WRITE_ONCE(old_rflow->filter, RPS_NO_FILTER); out: #endif head = READ_ONCE(per_cpu(softnet_data, next_cpu).input_queue_head); @@ -4672,7 +4672,7 @@ bool rps_may_expire_flow(struct net_device *dev, u16 rxq_index, if (flow_table && flow_id <= flow_table->mask) { rflow = &flow_table->flows[flow_id]; cpu = READ_ONCE(rflow->cpu); - if (rflow->filter == filter_id && cpu < nr_cpu_ids && + if (READ_ONCE(rflow->filter) == filter_id && cpu < nr_cpu_ids && ((int)(READ_ONCE(per_cpu(softnet_data, cpu).input_queue_head) - READ_ONCE(rflow->last_qtail)) < (int)(10 * flow_table->mask))) -- cgit v1.2.3 From f7b60cce847036f4a639d44c675553a564d8e876 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 18 Apr 2024 15:36:03 +0800 Subject: net: rps: locklessly access rflow->cpu This is the last member in struct rps_dev_flow which should be protected locklessly. So finish it. Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 182fce2537d4..8bdc59074b29 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4539,7 +4539,7 @@ set_rps_cpu(struct net_device *dev, struct sk_buff *skb, rps_input_queue_tail_save(&rflow->last_qtail, head); } - rflow->cpu = next_cpu; + WRITE_ONCE(rflow->cpu, next_cpu); return rflow; } -- cgit v1.2.3 From f8f2eb9de69a1119117d198547c13d7a1123a5a9 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 09:51:04 +0000 Subject: neighbour: add RCU protection to neigh_tables[] In order to remove RTNL protection from neightbl_dump_info() and neigh_dump_info() later, we need to add RCU protection to neigh_tables[]. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/neighbour.c | 30 +++++++++++++++++++----------- 1 file changed, 19 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 47d07b122f7a..ccb770539f1a 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1771,7 +1771,7 @@ static void neigh_parms_destroy(struct neigh_parms *parms) static struct lock_class_key neigh_table_proxy_queue_class; -static struct neigh_table *neigh_tables[NEIGH_NR_TABLES] __read_mostly; +static struct neigh_table __rcu *neigh_tables[NEIGH_NR_TABLES] __read_mostly; void neigh_table_init(int index, struct neigh_table *tbl) { @@ -1828,13 +1828,19 @@ void neigh_table_init(int index, struct neigh_table *tbl) tbl->last_flush = now; tbl->last_rand = now + tbl->parms.reachable_time * 20; - neigh_tables[index] = tbl; + rcu_assign_pointer(neigh_tables[index], tbl); } EXPORT_SYMBOL(neigh_table_init); +/* + * Only called from ndisc_cleanup(), which means this is dead code + * because we no longer can unload IPv6 module. + */ int neigh_table_clear(int index, struct neigh_table *tbl) { - neigh_tables[index] = NULL; + RCU_INIT_POINTER(neigh_tables[index], NULL); + synchronize_rcu(); + /* It is not clean... Fix it to unload IPv6 module safely */ cancel_delayed_work_sync(&tbl->managed_work); cancel_delayed_work_sync(&tbl->gc_work); @@ -1866,10 +1872,10 @@ static struct neigh_table *neigh_find_table(int family) switch (family) { case AF_INET: - tbl = neigh_tables[NEIGH_ARP_TABLE]; + tbl = rcu_dereference_rtnl(neigh_tables[NEIGH_ARP_TABLE]); break; case AF_INET6: - tbl = neigh_tables[NEIGH_ND_TABLE]; + tbl = rcu_dereference_rtnl(neigh_tables[NEIGH_ND_TABLE]); break; } @@ -2333,7 +2339,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, ndtmsg = nlmsg_data(nlh); for (tidx = 0; tidx < NEIGH_NR_TABLES; tidx++) { - tbl = neigh_tables[tidx]; + tbl = rcu_dereference_rtnl(neigh_tables[tidx]); if (!tbl) continue; if (ndtmsg->ndtm_family && tbl->family != ndtmsg->ndtm_family) @@ -2521,7 +2527,7 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) for (tidx = 0; tidx < NEIGH_NR_TABLES; tidx++) { struct neigh_parms *p; - tbl = neigh_tables[tidx]; + tbl = rcu_dereference_rtnl(neigh_tables[tidx]); if (!tbl) continue; @@ -2881,7 +2887,7 @@ static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb) s_t = cb->args[0]; for (t = 0; t < NEIGH_NR_TABLES; t++) { - tbl = neigh_tables[t]; + tbl = rcu_dereference_rtnl(neigh_tables[t]); if (!tbl) continue; @@ -3145,14 +3151,15 @@ int neigh_xmit(int index, struct net_device *dev, const void *addr, struct sk_buff *skb) { int err = -EAFNOSUPPORT; + if (likely(index < NEIGH_NR_TABLES)) { struct neigh_table *tbl; struct neighbour *neigh; - tbl = neigh_tables[index]; - if (!tbl) - goto out; rcu_read_lock(); + tbl = rcu_dereference(neigh_tables[index]); + if (!tbl) + goto out_unlock; if (index == NEIGH_ARP_TABLE) { u32 key = *((u32 *)addr); @@ -3168,6 +3175,7 @@ int neigh_xmit(int index, struct net_device *dev, goto out_kfree_skb; } err = READ_ONCE(neigh->output)(neigh, skb); +out_unlock: rcu_read_unlock(); } else if (index == NEIGH_LINK_TABLE) { -- cgit v1.2.3 From 7e4975f7e7fb0eba3cbb69d9c467750a1c3ce131 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 09:51:05 +0000 Subject: neighbour: fix neigh_dump_info() return value Change neigh_dump_table() and pneigh_dump_table() to either return 0 or -EMSGSIZE if not enough space was available in the skb. Then neigh_dump_info() can do the same. This allows NLMSG_DONE to be appended to the current skb at the end of a dump, saving a couple of recvmsg() system calls. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/neighbour.c | 31 +++++++++++++------------------ 1 file changed, 13 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index ccb770539f1a..d8c3ffdee29f 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2715,7 +2715,7 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, { struct net *net = sock_net(skb->sk); struct neighbour *n; - int rc, h, s_h = cb->args[1]; + int err = 0, h, s_h = cb->args[1]; int idx, s_idx = idx = cb->args[2]; struct neigh_hash_table *nht; unsigned int flags = NLM_F_MULTI; @@ -2737,23 +2737,20 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, if (neigh_ifindex_filtered(n->dev, filter->dev_idx) || neigh_master_filtered(n->dev, filter->master_idx)) goto next; - if (neigh_fill_info(skb, n, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, - RTM_NEWNEIGH, - flags) < 0) { - rc = -1; + err = neigh_fill_info(skb, n, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, + RTM_NEWNEIGH, flags); + if (err < 0) goto out; - } next: idx++; } } - rc = skb->len; out: rcu_read_unlock(); cb->args[1] = h; cb->args[2] = idx; - return rc; + return err; } static int pneigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, @@ -2762,7 +2759,7 @@ static int pneigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, { struct pneigh_entry *n; struct net *net = sock_net(skb->sk); - int rc, h, s_h = cb->args[3]; + int err = 0, h, s_h = cb->args[3]; int idx, s_idx = idx = cb->args[4]; unsigned int flags = NLM_F_MULTI; @@ -2780,11 +2777,11 @@ static int pneigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, if (neigh_ifindex_filtered(n->dev, filter->dev_idx) || neigh_master_filtered(n->dev, filter->master_idx)) goto next; - if (pneigh_fill_info(skb, n, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, - RTM_NEWNEIGH, flags, tbl) < 0) { + err = pneigh_fill_info(skb, n, NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, + RTM_NEWNEIGH, flags, tbl); + if (err < 0) { read_unlock_bh(&tbl->lock); - rc = -1; goto out; } next: @@ -2793,12 +2790,10 @@ static int pneigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, } read_unlock_bh(&tbl->lock); - rc = skb->len; out: cb->args[3] = h; cb->args[4] = idx; - return rc; - + return err; } static int neigh_valid_dump_req(const struct nlmsghdr *nlh, @@ -2905,7 +2900,7 @@ static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb) } cb->args[0] = t; - return skb->len; + return err; } static int neigh_valid_get_req(const struct nlmsghdr *nlh, -- cgit v1.2.3 From ba0f780694237d96a1d6366f931cd716fb0f7ab5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 09:51:06 +0000 Subject: neighbour: no longer hold RTNL in neigh_dump_info() neigh_dump_table() is already relying on RCU protection. pneigh_dump_table() is using its own protection (tbl->lock) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/neighbour.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index d8c3ffdee29f..0805c00c63d4 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2723,7 +2723,6 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, if (filter->dev_idx || filter->master_idx) flags |= NLM_F_DUMP_FILTERED; - rcu_read_lock(); nht = rcu_dereference(tbl->nht); for (h = s_h; h < (1 << nht->hash_shift); h++) { @@ -2747,7 +2746,6 @@ next: } } out: - rcu_read_unlock(); cb->args[1] = h; cb->args[2] = idx; return err; @@ -2881,8 +2879,9 @@ static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb) s_t = cb->args[0]; + rcu_read_lock(); for (t = 0; t < NEIGH_NR_TABLES; t++) { - tbl = rcu_dereference_rtnl(neigh_tables[t]); + tbl = rcu_dereference(neigh_tables[t]); if (!tbl) continue; @@ -2898,6 +2897,7 @@ static int neigh_dump_info(struct sk_buff *skb, struct netlink_callback *cb) if (err < 0) break; } + rcu_read_unlock(); cb->args[0] = t; return err; @@ -3894,7 +3894,8 @@ static int __init neigh_init(void) { rtnl_register(PF_UNSPEC, RTM_NEWNEIGH, neigh_add, NULL, 0); rtnl_register(PF_UNSPEC, RTM_DELNEIGH, neigh_delete, NULL, 0); - rtnl_register(PF_UNSPEC, RTM_GETNEIGH, neigh_get, neigh_dump_info, 0); + rtnl_register(PF_UNSPEC, RTM_GETNEIGH, neigh_get, neigh_dump_info, + RTNL_FLAG_DUMP_UNLOCKED); rtnl_register(PF_UNSPEC, RTM_GETNEIGHTBL, NULL, neightbl_dump_info, 0); -- cgit v1.2.3 From bfa858f220ab8c950dd3e1310fee61950d0ecdae Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Thu, 18 Apr 2024 11:40:08 +0200 Subject: sysctl: treewide: constify ctl_table_header::ctl_table_arg MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit To be able to constify instances of struct ctl_tables it is necessary to remove ways through which non-const versions are exposed from the sysctl core. One of these is the ctl_table_arg member of struct ctl_table_header. Constify this reference as a prerequisite for the full constification of struct ctl_table instances. No functional change. Signed-off-by: Thomas Weißschuh Reviewed-by: Kees Cook Signed-off-by: David S. Miller --- drivers/net/vrf.c | 2 +- include/linux/sysctl.h | 2 +- ipc/ipc_sysctl.c | 2 +- ipc/mq_sysctl.c | 2 +- kernel/ucount.c | 2 +- net/ax25/sysctl_net_ax25.c | 2 +- net/bridge/br_netfilter_hooks.c | 2 +- net/core/sysctl_net_core.c | 2 +- net/ieee802154/6lowpan/reassembly.c | 2 +- net/ipv4/devinet.c | 2 +- net/ipv4/ip_fragment.c | 2 +- net/ipv4/route.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 2 +- net/ipv4/xfrm4_policy.c | 2 +- net/ipv6/addrconf.c | 2 +- net/ipv6/netfilter/nf_conntrack_reasm.c | 2 +- net/ipv6/reassembly.c | 2 +- net/ipv6/sysctl_net_ipv6.c | 6 +++--- net/ipv6/xfrm6_policy.c | 2 +- net/mpls/af_mpls.c | 4 ++-- net/mptcp/ctrl.c | 2 +- net/netfilter/nf_conntrack_standalone.c | 2 +- net/netfilter/nf_log.c | 2 +- net/sctp/sysctl.c | 2 +- net/smc/smc_sysctl.c | 2 +- net/unix/sysctl_net_unix.c | 2 +- net/xfrm/xfrm_sysctl.c | 2 +- 27 files changed, 30 insertions(+), 30 deletions(-) (limited to 'net') diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index bb95ce43cd97..66f8542f3b18 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -1971,7 +1971,7 @@ static int vrf_netns_init_sysctl(struct net *net, struct netns_vrf *nn_vrf) static void vrf_netns_exit_sysctl(struct net *net) { struct netns_vrf *nn_vrf = net_generic(net, vrf_net_id); - struct ctl_table *table; + const struct ctl_table *table; table = nn_vrf->ctl_hdr->ctl_table_arg; unregister_net_sysctl_table(nn_vrf->ctl_hdr); diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index ee7d33b89e9e..9413241df962 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -182,7 +182,7 @@ struct ctl_table_header { struct rcu_head rcu; }; struct completion *unregistering; - struct ctl_table *ctl_table_arg; + const struct ctl_table *ctl_table_arg; struct ctl_table_root *root; struct ctl_table_set *set; struct ctl_dir *parent; diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c index 45cb1dabce29..3c3755918d34 100644 --- a/ipc/ipc_sysctl.c +++ b/ipc/ipc_sysctl.c @@ -306,7 +306,7 @@ bool setup_ipc_sysctls(struct ipc_namespace *ns) void retire_ipc_sysctls(struct ipc_namespace *ns) { - struct ctl_table *tbl; + const struct ctl_table *tbl; tbl = ns->ipc_sysctls->ctl_table_arg; unregister_sysctl_table(ns->ipc_sysctls); diff --git a/ipc/mq_sysctl.c b/ipc/mq_sysctl.c index 21fba3a6edaf..69c709262f5a 100644 --- a/ipc/mq_sysctl.c +++ b/ipc/mq_sysctl.c @@ -160,7 +160,7 @@ bool setup_mq_sysctls(struct ipc_namespace *ns) void retire_mq_sysctls(struct ipc_namespace *ns) { - struct ctl_table *tbl; + const struct ctl_table *tbl; tbl = ns->mq_sysctls->ctl_table_arg; unregister_sysctl_table(ns->mq_sysctls); diff --git a/kernel/ucount.c b/kernel/ucount.c index 4aa6166cb856..d9e283600f5c 100644 --- a/kernel/ucount.c +++ b/kernel/ucount.c @@ -119,7 +119,7 @@ bool setup_userns_sysctls(struct user_namespace *ns) void retire_userns_sysctls(struct user_namespace *ns) { #ifdef CONFIG_SYSCTL - struct ctl_table *tbl; + const struct ctl_table *tbl; tbl = ns->sysctls->ctl_table_arg; unregister_sysctl_table(ns->sysctls); diff --git a/net/ax25/sysctl_net_ax25.c b/net/ax25/sysctl_net_ax25.c index db66e11e7fe8..e0128dc9def3 100644 --- a/net/ax25/sysctl_net_ax25.c +++ b/net/ax25/sysctl_net_ax25.c @@ -171,7 +171,7 @@ int ax25_register_dev_sysctl(ax25_dev *ax25_dev) void ax25_unregister_dev_sysctl(ax25_dev *ax25_dev) { struct ctl_table_header *header = ax25_dev->sysheader; - struct ctl_table *table; + const struct ctl_table *table; if (header) { ax25_dev->sysheader = NULL; diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index c683d8377862..7948a9e7542c 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -1275,7 +1275,7 @@ static int br_netfilter_sysctl_init_net(struct net *net) static void br_netfilter_sysctl_exit_net(struct net *net, struct brnf_net *brnet) { - struct ctl_table *table = brnet->ctl_hdr->ctl_table_arg; + const struct ctl_table *table = brnet->ctl_hdr->ctl_table_arg; unregister_net_sysctl_table(brnet->ctl_hdr); if (!net_eq(net, &init_net)) diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 6973dda3abda..903ab4a51c17 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -743,7 +743,7 @@ err_dup: static __net_exit void sysctl_core_net_exit(struct net *net) { - struct ctl_table *tbl; + const struct ctl_table *tbl; tbl = net->core.sysctl_hdr->ctl_table_arg; unregister_net_sysctl_table(net->core.sysctl_hdr); diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index 6dd960ec558c..2a983cf450da 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -399,7 +399,7 @@ err_alloc: static void __net_exit lowpan_frags_ns_sysctl_unregister(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; struct netns_ieee802154_lowpan *ieee802154_lowpan = net_ieee802154_lowpan(net); diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 7a437f0d4190..7592f242336b 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2749,7 +2749,7 @@ err_alloc_all: static __net_exit void devinet_exit_net(struct net *net) { #ifdef CONFIG_SYSCTL - struct ctl_table *tbl; + const struct ctl_table *tbl; tbl = net->ipv4.forw_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.forw_hdr); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index fb947d1613fe..534b98a0744a 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -632,7 +632,7 @@ err_alloc: static void __net_exit ip4_frags_ns_ctl_unregister(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->ipv4.frags_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.frags_hdr); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 6133f1fd848e..2e29ab819967 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3576,7 +3576,7 @@ err_dup: static __net_exit void sysctl_route_net_exit(struct net *net) { - struct ctl_table *tbl; + const struct ctl_table *tbl; tbl = net->ipv4.route_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.route_hdr); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 7e4f16a7dcc1..ce5d19978a26 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -1554,7 +1554,7 @@ err_alloc: static __net_exit void ipv4_sysctl_exit_net(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; kfree(net->ipv4.sysctl_local_reserved_ports); table = net->ipv4.ipv4_hdr->ctl_table_arg; diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index c33bca2c3841..1dda59e0aeab 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -186,7 +186,7 @@ err_alloc: static __net_exit void xfrm4_net_sysctl_exit(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; if (!net->ipv4.xfrm4_hdr) return; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 3a92d002abe4..9aa0900abfa1 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -7238,7 +7238,7 @@ out: static void __addrconf_sysctl_unregister(struct net *net, struct ipv6_devconf *p, int ifindex) { - struct ctl_table *table; + const struct ctl_table *table; if (!p->sysctl_header) return; diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index d0dcbaca1994..ce8c14d8aff5 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -105,7 +105,7 @@ err_alloc: static void __net_exit nf_ct_frags6_sysctl_unregister(struct net *net) { struct nft_ct_frag6_pernet *nf_frag = nf_frag_pernet(net); - struct ctl_table *table; + const struct ctl_table *table; table = nf_frag->nf_frag_frags_hdr->ctl_table_arg; unregister_net_sysctl_table(nf_frag->nf_frag_frags_hdr); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index acb4f119e11f..ee95cdcc8747 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -487,7 +487,7 @@ err_alloc: static void __net_exit ip6_frags_ns_sysctl_unregister(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->ipv6.sysctl.frags_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv6.sysctl.frags_hdr); diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index 888676163e90..75de55f907b0 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -313,9 +313,9 @@ out_ipv6_table: static void __net_exit ipv6_sysctl_net_exit(struct net *net) { - struct ctl_table *ipv6_table; - struct ctl_table *ipv6_route_table; - struct ctl_table *ipv6_icmp_table; + const struct ctl_table *ipv6_table; + const struct ctl_table *ipv6_route_table; + const struct ctl_table *ipv6_icmp_table; ipv6_table = net->ipv6.sysctl.hdr->ctl_table_arg; ipv6_route_table = net->ipv6.sysctl.route_hdr->ctl_table_arg; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 42fb6996b077..4891012b692f 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -218,7 +218,7 @@ err_alloc: static void __net_exit xfrm6_net_sysctl_exit(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; if (!net->ipv6.sysctl.xfrm6_hdr) return; diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 0315b8deed3f..5d2012d1cf4a 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1422,7 +1422,7 @@ static void mpls_dev_sysctl_unregister(struct net_device *dev, struct mpls_dev *mdev) { struct net *net = dev_net(dev); - struct ctl_table *table; + const struct ctl_table *table; if (!mdev->sysctl) return; @@ -2690,7 +2690,7 @@ static void mpls_net_exit(struct net *net) { struct mpls_route __rcu **platform_label; size_t platform_labels; - struct ctl_table *table; + const struct ctl_table *table; unsigned int index; table = net->mpls.ctl->ctl_table_arg; diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c index 13fe0748dde8..8d661156ab8c 100644 --- a/net/mptcp/ctrl.c +++ b/net/mptcp/ctrl.c @@ -198,7 +198,7 @@ err_alloc: static void mptcp_pernet_del_table(struct mptcp_pernet *pernet) { - struct ctl_table *table = pernet->ctl_table_hdr->ctl_table_arg; + const struct ctl_table *table = pernet->ctl_table_hdr->ctl_table_arg; unregister_net_sysctl_table(pernet->ctl_table_hdr); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 0ee98ce5b816..bb9dea676ec1 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -1122,7 +1122,7 @@ out_unregister_netfilter: static void nf_conntrack_standalone_fini_sysctl(struct net *net) { struct nf_conntrack_net *cnet = nf_ct_pernet(net); - struct ctl_table *table; + const struct ctl_table *table; table = cnet->sysctl_header->ctl_table_arg; unregister_net_sysctl_table(cnet->sysctl_header); diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 370f8231385c..efedd2f13ac7 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -514,7 +514,7 @@ err_alloc: static void netfilter_log_sysctl_exit(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->nf.nf_log_dir_header->ctl_table_arg; unregister_net_sysctl_table(net->nf.nf_log_dir_header); diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index f65d6f92afcb..25bdf17c7262 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -624,7 +624,7 @@ int sctp_sysctl_net_register(struct net *net) void sctp_sysctl_net_unregister(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->sctp.sysctl_header->ctl_table_arg; unregister_net_sysctl_table(net->sctp.sysctl_header); diff --git a/net/smc/smc_sysctl.c b/net/smc/smc_sysctl.c index a5946d1b9d60..4e8baa2e7ea4 100644 --- a/net/smc/smc_sysctl.c +++ b/net/smc/smc_sysctl.c @@ -133,7 +133,7 @@ err_alloc: void __net_exit smc_sysctl_net_exit(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->smc.smc_hdr->ctl_table_arg; unregister_net_sysctl_table(net->smc.smc_hdr); diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c index 3e84b31c355a..44996af61999 100644 --- a/net/unix/sysctl_net_unix.c +++ b/net/unix/sysctl_net_unix.c @@ -52,7 +52,7 @@ err_alloc: void unix_sysctl_unregister(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->unx.ctl->ctl_table_arg; unregister_net_sysctl_table(net->unx.ctl); diff --git a/net/xfrm/xfrm_sysctl.c b/net/xfrm/xfrm_sysctl.c index 7fdeafc838a7..e972930c292b 100644 --- a/net/xfrm/xfrm_sysctl.c +++ b/net/xfrm/xfrm_sysctl.c @@ -76,7 +76,7 @@ out_kmemdup: void __net_exit xfrm_sysctl_fini(struct net *net) { - struct ctl_table *table; + const struct ctl_table *table; table = net->xfrm.sysctl_hdr->ctl_table_arg; unregister_net_sysctl_table(net->xfrm.sysctl_hdr); -- cgit v1.2.3 From c51db4ac10d57c366f9a92121e3889bfc6c324cd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 19 Apr 2024 07:19:42 +0000 Subject: tcp: do not export tcp_twsk_purge() After commit 1eeb50435739 ("tcp/dccp: do not care about families in inet_twsk_purge()") tcp_twsk_purge() is no longer potentially called from a module. Signed-off-by: Eric Dumazet Cc: Kuniyuki Iwashima Reviewed-by: Kuniyuki Iwashima Signed-off-by: David S. Miller --- net/ipv4/tcp_minisocks.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index f53c7ada2ace..146c061145b4 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -403,7 +403,6 @@ void tcp_twsk_purge(struct list_head *net_exit_list) } } } -EXPORT_SYMBOL_GPL(tcp_twsk_purge); /* Warning : This function is called without sk_listener being locked. * Be sure to read socket fields once, as their value could change under us. -- cgit v1.2.3 From 5bcf0dcbf9066348058b88a510c57f70f384c92c Mon Sep 17 00:00:00 2001 From: Toke Høiland-Jørgensen Date: Thu, 18 Apr 2024 09:18:39 +0200 Subject: xdp: use flags field to disambiguate broadcast redirect MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When redirecting a packet using XDP, the bpf_redirect_map() helper will set up the redirect destination information in struct bpf_redirect_info (using the __bpf_xdp_redirect_map() helper function), and the xdp_do_redirect() function will read this information after the XDP program returns and pass the frame on to the right redirect destination. When using the BPF_F_BROADCAST flag to do multicast redirect to a whole map, __bpf_xdp_redirect_map() sets the 'map' pointer in struct bpf_redirect_info to point to the destination map to be broadcast. And xdp_do_redirect() reacts to the value of this map pointer to decide whether it's dealing with a broadcast or a single-value redirect. However, if the destination map is being destroyed before xdp_do_redirect() is called, the map pointer will be cleared out (by bpf_clear_redirect_map()) without waiting for any XDP programs to stop running. This causes xdp_do_redirect() to think that the redirect was to a single target, but the target pointer is also NULL (since broadcast redirects don't have a single target), so this causes a crash when a NULL pointer is passed to dev_map_enqueue(). To fix this, change xdp_do_redirect() to react directly to the presence of the BPF_F_BROADCAST flag in the 'flags' value in struct bpf_redirect_info to disambiguate between a single-target and a broadcast redirect. And only read the 'map' pointer if the broadcast flag is set, aborting if that has been cleared out in the meantime. This prevents the crash, while keeping the atomic (cmpxchg-based) clearing of the map pointer itself, and without adding any more checks in the non-broadcast fast path. Fixes: e624d4ed4aa8 ("xdp: Extend xdp_redirect_map with broadcast support") Reported-and-tested-by: syzbot+af9492708df9797198d6@syzkaller.appspotmail.com Signed-off-by: Toke Høiland-Jørgensen Acked-by: Stanislav Fomichev Reviewed-by: Hangbin Liu Acked-by: Jesper Dangaard Brouer Link: https://lore.kernel.org/r/20240418071840.156411-1-toke@redhat.com Signed-off-by: Martin KaFai Lau --- net/core/filter.c | 42 ++++++++++++++++++++++++++++++++---------- 1 file changed, 32 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/core/filter.c b/net/core/filter.c index 8adf95765cdd..ae5254f712c9 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4360,10 +4360,12 @@ static __always_inline int __xdp_do_redirect_frame(struct bpf_redirect_info *ri, enum bpf_map_type map_type = ri->map_type; void *fwd = ri->tgt_value; u32 map_id = ri->map_id; + u32 flags = ri->flags; struct bpf_map *map; int err; ri->map_id = 0; /* Valid map id idr range: [1,INT_MAX[ */ + ri->flags = 0; ri->map_type = BPF_MAP_TYPE_UNSPEC; if (unlikely(!xdpf)) { @@ -4375,11 +4377,20 @@ static __always_inline int __xdp_do_redirect_frame(struct bpf_redirect_info *ri, case BPF_MAP_TYPE_DEVMAP: fallthrough; case BPF_MAP_TYPE_DEVMAP_HASH: - map = READ_ONCE(ri->map); - if (unlikely(map)) { + if (unlikely(flags & BPF_F_BROADCAST)) { + map = READ_ONCE(ri->map); + + /* The map pointer is cleared when the map is being torn + * down by bpf_clear_redirect_map() + */ + if (unlikely(!map)) { + err = -ENOENT; + break; + } + WRITE_ONCE(ri->map, NULL); err = dev_map_enqueue_multi(xdpf, dev, map, - ri->flags & BPF_F_EXCLUDE_INGRESS); + flags & BPF_F_EXCLUDE_INGRESS); } else { err = dev_map_enqueue(fwd, xdpf, dev); } @@ -4442,9 +4453,9 @@ EXPORT_SYMBOL_GPL(xdp_do_redirect_frame); static int xdp_do_generic_redirect_map(struct net_device *dev, struct sk_buff *skb, struct xdp_buff *xdp, - struct bpf_prog *xdp_prog, - void *fwd, - enum bpf_map_type map_type, u32 map_id) + struct bpf_prog *xdp_prog, void *fwd, + enum bpf_map_type map_type, u32 map_id, + u32 flags) { struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info); struct bpf_map *map; @@ -4454,11 +4465,20 @@ static int xdp_do_generic_redirect_map(struct net_device *dev, case BPF_MAP_TYPE_DEVMAP: fallthrough; case BPF_MAP_TYPE_DEVMAP_HASH: - map = READ_ONCE(ri->map); - if (unlikely(map)) { + if (unlikely(flags & BPF_F_BROADCAST)) { + map = READ_ONCE(ri->map); + + /* The map pointer is cleared when the map is being torn + * down by bpf_clear_redirect_map() + */ + if (unlikely(!map)) { + err = -ENOENT; + break; + } + WRITE_ONCE(ri->map, NULL); err = dev_map_redirect_multi(dev, skb, xdp_prog, map, - ri->flags & BPF_F_EXCLUDE_INGRESS); + flags & BPF_F_EXCLUDE_INGRESS); } else { err = dev_map_generic_redirect(fwd, skb, xdp_prog); } @@ -4495,9 +4515,11 @@ int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb, enum bpf_map_type map_type = ri->map_type; void *fwd = ri->tgt_value; u32 map_id = ri->map_id; + u32 flags = ri->flags; int err; ri->map_id = 0; /* Valid map id idr range: [1,INT_MAX[ */ + ri->flags = 0; ri->map_type = BPF_MAP_TYPE_UNSPEC; if (map_type == BPF_MAP_TYPE_UNSPEC && map_id == INT_MAX) { @@ -4517,7 +4539,7 @@ int xdp_do_generic_redirect(struct net_device *dev, struct sk_buff *skb, return 0; } - return xdp_do_generic_redirect_map(dev, skb, xdp, xdp_prog, fwd, map_type, map_id); + return xdp_do_generic_redirect_map(dev, skb, xdp, xdp_prog, fwd, map_type, map_id, flags); err: _trace_xdp_redirect_err(dev, xdp_prog, ri->tgt_index, err); return err; -- cgit v1.2.3 From 5625ca5640caa3fb797f155601d56379d260d6ba Mon Sep 17 00:00:00 2001 From: Mateusz Polchlopek Date: Fri, 19 Apr 2024 04:08:49 -0400 Subject: devlink: extend devlink_param *set pointer Extend devlink_param *set function pointer to take extack as a param. Sometimes it is needed to pass information to the end user from set function. It is more proper to use for that netlink instead of passing message to dmesg. Reviewed-by: Jiri Pirko Reviewed-by: Przemek Kitszel Signed-off-by: Mateusz Polchlopek Signed-off-by: Tony Nguyen --- drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c | 9 ++++++--- drivers/net/ethernet/amd/pds_core/core.h | 3 ++- drivers/net/ethernet/amd/pds_core/devlink.c | 3 ++- drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c | 6 ++++-- drivers/net/ethernet/intel/ice/devlink/devlink.c | 12 ++++++------ drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c | 12 ++++++++---- drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c | 3 ++- drivers/net/ethernet/mellanox/mlx4/main.c | 6 ++++-- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 3 ++- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 3 ++- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 3 ++- drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c | 3 ++- drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 3 ++- drivers/net/ethernet/netronome/nfp/devlink_param.c | 3 ++- drivers/net/ethernet/qlogic/qed/qed_devlink.c | 3 ++- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 3 ++- drivers/net/ethernet/ti/cpsw_new.c | 6 ++++-- drivers/net/wwan/iosm/iosm_ipc_devlink.c | 3 ++- include/net/devlink.h | 3 ++- include/net/dsa.h | 3 ++- net/devlink/param.c | 7 ++++--- net/dsa/devlink.c | 3 ++- 22 files changed, 66 insertions(+), 37 deletions(-) (limited to 'net') diff --git a/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c b/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c index d2b8d26db968..215a1a8ba7e9 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cpt_devlink.c @@ -4,7 +4,8 @@ #include "otx2_cpt_devlink.h" static int otx2_cpt_dl_egrp_create(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_cpt_devlink *cpt_dl = devlink_priv(dl); struct otx2_cptpf_dev *cptpf = cpt_dl->cptpf; @@ -13,7 +14,8 @@ static int otx2_cpt_dl_egrp_create(struct devlink *dl, u32 id, } static int otx2_cpt_dl_egrp_delete(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_cpt_devlink *cpt_dl = devlink_priv(dl); struct otx2_cptpf_dev *cptpf = cpt_dl->cptpf; @@ -45,7 +47,8 @@ static int otx2_cpt_dl_t106_mode_get(struct devlink *dl, u32 id, } static int otx2_cpt_dl_t106_mode_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_cpt_devlink *cpt_dl = devlink_priv(dl); struct otx2_cptpf_dev *cptpf = cpt_dl->cptpf; diff --git a/drivers/net/ethernet/amd/pds_core/core.h b/drivers/net/ethernet/amd/pds_core/core.h index a3e17a0c187a..14522d6d5f86 100644 --- a/drivers/net/ethernet/amd/pds_core/core.h +++ b/drivers/net/ethernet/amd/pds_core/core.h @@ -256,7 +256,8 @@ int pdsc_dl_flash_update(struct devlink *dl, int pdsc_dl_enable_get(struct devlink *dl, u32 id, struct devlink_param_gset_ctx *ctx); int pdsc_dl_enable_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx); + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); int pdsc_dl_enable_validate(struct devlink *dl, u32 id, union devlink_param_value val, struct netlink_ext_ack *extack); diff --git a/drivers/net/ethernet/amd/pds_core/devlink.c b/drivers/net/ethernet/amd/pds_core/devlink.c index 54864f27c87a..2681889162a2 100644 --- a/drivers/net/ethernet/amd/pds_core/devlink.c +++ b/drivers/net/ethernet/amd/pds_core/devlink.c @@ -37,7 +37,8 @@ int pdsc_dl_enable_get(struct devlink *dl, u32 id, } int pdsc_dl_enable_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct pdsc *pdsc = devlink_priv(dl); struct pdsc_viftype *vt_entry; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c index ae4529c043f0..d9ea6fa23923 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_devlink.c @@ -1096,7 +1096,8 @@ static int bnxt_dl_nvm_param_get(struct devlink *dl, u32 id, } static int bnxt_dl_nvm_param_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct bnxt *bp = bnxt_get_bp_from_dl(dl); struct hwrm_nvm_set_variable_input *req; @@ -1145,7 +1146,8 @@ static int bnxt_remote_dev_reset_get(struct devlink *dl, u32 id, } static int bnxt_remote_dev_reset_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct bnxt *bp = bnxt_get_bp_from_dl(dl); int rc; diff --git a/drivers/net/ethernet/intel/ice/devlink/devlink.c b/drivers/net/ethernet/intel/ice/devlink/devlink.c index be9244bb8bbc..acbace240977 100644 --- a/drivers/net/ethernet/intel/ice/devlink/devlink.c +++ b/drivers/net/ethernet/intel/ice/devlink/devlink.c @@ -1144,9 +1144,9 @@ ice_devlink_enable_roce_get(struct devlink *devlink, u32 id, return 0; } -static int -ice_devlink_enable_roce_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) +static int ice_devlink_enable_roce_set(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); bool roce_ena = ctx->val.vbool; @@ -1195,9 +1195,9 @@ ice_devlink_enable_iw_get(struct devlink *devlink, u32 id, return 0; } -static int -ice_devlink_enable_iw_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) +static int ice_devlink_enable_iw_set(struct devlink *devlink, u32 id, + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct ice_pf *pf = devlink_priv(devlink); bool iw_ena = ctx->val.vbool; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c index 96c04f7d93f8..7498ab429963 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c @@ -1202,7 +1202,8 @@ static int rvu_af_dl_dwrr_mtu_validate(struct devlink *devlink, u32 id, } static int rvu_af_dl_dwrr_mtu_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1256,7 +1257,8 @@ static int rvu_af_npc_exact_feature_get(struct devlink *devlink, u32 id, } static int rvu_af_npc_exact_feature_disable(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1310,7 +1312,8 @@ static int rvu_af_dl_npc_mcam_high_zone_percent_get(struct devlink *devlink, u32 } static int rvu_af_dl_npc_mcam_high_zone_percent_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; @@ -1367,7 +1370,8 @@ static int rvu_af_dl_nix_maxlf_get(struct devlink *devlink, u32 id, } static int rvu_af_dl_nix_maxlf_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct rvu_devlink *rvu_dl = devlink_priv(devlink); struct rvu *rvu = rvu_dl->rvu; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c index 4e1130496573..99ddf31269d9 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_devlink.c @@ -32,7 +32,8 @@ static int otx2_dl_mcam_count_validate(struct devlink *devlink, u32 id, } static int otx2_dl_mcam_count_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct otx2_devlink *otx2_dl = devlink_priv(devlink); struct otx2_nic *pfvf = otx2_dl->pfvf; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 7b02ff61126d..98688e4dbec5 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -185,7 +185,8 @@ static int mlx4_devlink_ierr_reset_get(struct devlink *devlink, u32 id, } static int mlx4_devlink_ierr_reset_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { mlx4_internal_err_reset = ctx->val.vbool; return 0; @@ -202,7 +203,8 @@ static int mlx4_devlink_crdump_snapshot_get(struct devlink *devlink, u32 id, } static int mlx4_devlink_crdump_snapshot_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx4_priv *priv = devlink_priv(devlink); struct mlx4_dev *dev = &priv->dev; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 1789800faaeb..17f78091ad30 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -1805,7 +1805,8 @@ err: } static int mlx5_devlink_esw_multiport_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index dfcc7e85e0a3..f378b8176e47 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2413,7 +2413,8 @@ err: } static int esw_port_metadata_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); struct mlx5_eswitch *esw = dev->priv.eswitch; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index cf085a478e3e..32cdacc34a0d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -3332,7 +3332,8 @@ static int mlx5_fs_mode_validate(struct devlink *devlink, u32 id, } static int mlx5_fs_mode_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); enum mlx5_flow_steering_mode mode; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c index 2911aa34a5be..979c49ae6b5c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c @@ -52,7 +52,8 @@ static void mlx5_set_fw_rst_ack(struct mlx5_core_dev *dev) } static int mlx5_fw_reset_enable_remote_dev_reset_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlx5_core_dev *dev = devlink_priv(devlink); struct mlx5_fw_reset *fw_reset; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c index f20052776b3f..baedf0d45e85 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c @@ -1465,7 +1465,8 @@ mlxsw_sp_acl_tcam_region_rehash_intrvl_get(struct devlink *devlink, u32 id, static int mlxsw_sp_acl_tcam_region_rehash_intrvl_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct mlxsw_core *mlxsw_core = devlink_priv(devlink); struct mlxsw_sp_acl_tcam_vregion *vregion; diff --git a/drivers/net/ethernet/netronome/nfp/devlink_param.c b/drivers/net/ethernet/netronome/nfp/devlink_param.c index a655f9e69a7b..0e1a3800f371 100644 --- a/drivers/net/ethernet/netronome/nfp/devlink_param.c +++ b/drivers/net/ethernet/netronome/nfp/devlink_param.c @@ -132,7 +132,8 @@ exit_close_nsp: static int nfp_devlink_param_u8_set(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { const struct nfp_devlink_param_u8_arg *arg; struct nfp_pf *pf = devlink_priv(devlink); diff --git a/drivers/net/ethernet/qlogic/qed/qed_devlink.c b/drivers/net/ethernet/qlogic/qed/qed_devlink.c index dad8e617c393..1adc7fbb3f2f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_devlink.c +++ b/drivers/net/ethernet/qlogic/qed/qed_devlink.c @@ -132,7 +132,8 @@ static int qed_dl_param_get(struct devlink *dl, u32 id, } static int qed_dl_param_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct qed_devlink *qed_dl = devlink_priv(dl); struct qed_dev *cdev; diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index a7dbbbe3ff5d..bf3df86554d1 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -3061,7 +3061,8 @@ static void am65_cpsw_init_port_switch_ale(struct am65_cpsw_port *port) } static int am65_cpsw_dl_switch_mode_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct am65_cpsw_devlink *dl_priv = devlink_priv(dl); struct am65_cpsw_common *cpsw = dl_priv->common; diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c index 087dcb67505a..2baa198ebfa0 100644 --- a/drivers/net/ethernet/ti/cpsw_new.c +++ b/drivers/net/ethernet/ti/cpsw_new.c @@ -1625,7 +1625,8 @@ static int cpsw_dl_switch_mode_get(struct devlink *dl, u32 id, } static int cpsw_dl_switch_mode_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct cpsw_devlink *dl_priv = devlink_priv(dl); struct cpsw_common *cpsw = dl_priv->cpsw; @@ -1762,7 +1763,8 @@ static int cpsw_dl_ale_ctrl_get(struct devlink *dl, u32 id, } static int cpsw_dl_ale_ctrl_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct cpsw_devlink *dl_priv = devlink_priv(dl); struct cpsw_common *cpsw = dl_priv->cpsw; diff --git a/drivers/net/wwan/iosm/iosm_ipc_devlink.c b/drivers/net/wwan/iosm/iosm_ipc_devlink.c index 2fe724d623c0..bef6819986e9 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_devlink.c +++ b/drivers/net/wwan/iosm/iosm_ipc_devlink.c @@ -33,7 +33,8 @@ static int ipc_devlink_get_param(struct devlink *dl, u32 id, /* Set the param values for the specific param ID's */ static int ipc_devlink_set_param(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct iosm_devlink *ipc_devlink = devlink_priv(dl); diff --git a/include/net/devlink.h b/include/net/devlink.h index d31769a116ce..35eb0f884386 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -483,7 +483,8 @@ struct devlink_param { int (*get)(struct devlink *devlink, u32 id, struct devlink_param_gset_ctx *ctx); int (*set)(struct devlink *devlink, u32 id, - struct devlink_param_gset_ctx *ctx); + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); int (*validate)(struct devlink *devlink, u32 id, union devlink_param_value val, struct netlink_ext_ack *extack); diff --git a/include/net/dsa.h b/include/net/dsa.h index 7edfd8de8882..a6ef7e4c503f 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -1258,7 +1258,8 @@ struct dsa_switch_ops { int dsa_devlink_param_get(struct devlink *dl, u32 id, struct devlink_param_gset_ctx *ctx); int dsa_devlink_param_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx); + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack); int dsa_devlink_params_register(struct dsa_switch *ds, const struct devlink_param *params, size_t params_count); diff --git a/net/devlink/param.c b/net/devlink/param.c index 22bc3b500518..dcf0d1ccebba 100644 --- a/net/devlink/param.c +++ b/net/devlink/param.c @@ -158,11 +158,12 @@ static int devlink_param_get(struct devlink *devlink, static int devlink_param_set(struct devlink *devlink, const struct devlink_param *param, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { if (!param->set) return -EOPNOTSUPP; - return param->set(devlink, param->id, ctx); + return param->set(devlink, param->id, ctx, extack); } static int @@ -571,7 +572,7 @@ static int __devlink_nl_cmd_param_set_doit(struct devlink *devlink, return -EOPNOTSUPP; ctx.val = value; ctx.cmode = cmode; - err = devlink_param_set(devlink, param, &ctx); + err = devlink_param_set(devlink, param, &ctx, info->extack); if (err) return err; } diff --git a/net/dsa/devlink.c b/net/dsa/devlink.c index 431bf52290a1..0aac887d0098 100644 --- a/net/dsa/devlink.c +++ b/net/dsa/devlink.c @@ -194,7 +194,8 @@ int dsa_devlink_param_get(struct devlink *dl, u32 id, EXPORT_SYMBOL_GPL(dsa_devlink_param_get); int dsa_devlink_param_set(struct devlink *dl, u32 id, - struct devlink_param_gset_ctx *ctx) + struct devlink_param_gset_ctx *ctx, + struct netlink_ext_ack *extack) { struct dsa_switch *ds = dsa_devlink_to_ds(dl); -- cgit v1.2.3 From 22555032c513e62fe744d4cdd553539897e8e922 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 21:45:58 +0000 Subject: tcp: remove dubious FIN exception from tcp_cwnd_test() tcp_cwnd_test() has a special handing for the last packet in the write queue if it is smaller than one MSS and has the FIN flag. This is in violation of TCP RFC, and seems quite dubious. This packet can be sent only if the current CWND is bigger than the number of packets in flight. Making tcp_cwnd_test() result independent of the first skb in the write queue is needed for the last patch of the series. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240418214600.1291486-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_output.c | 18 +++++------------- 1 file changed, 5 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 61119d42b0fd..acbc76ca3e64 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2073,16 +2073,10 @@ static unsigned int tcp_mss_split_point(const struct sock *sk, /* Can at least one segment of SKB be sent right now, according to the * congestion window rules? If so, return how many segments are allowed. */ -static inline unsigned int tcp_cwnd_test(const struct tcp_sock *tp, - const struct sk_buff *skb) +static u32 tcp_cwnd_test(const struct tcp_sock *tp) { u32 in_flight, cwnd, halfcwnd; - /* Don't be strict about the congestion window for the final FIN. */ - if ((TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) && - tcp_skb_pcount(skb) == 1) - return 1; - in_flight = tcp_packets_in_flight(tp); cwnd = tcp_snd_cwnd(tp); if (in_flight >= cwnd) @@ -2706,10 +2700,9 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; unsigned int tso_segs, sent_pkts; - int cwnd_quota; + u32 cwnd_quota, max_segs; int result; bool is_cwnd_limited = false, is_rwnd_limited = false; - u32 max_segs; sent_pkts = 0; @@ -2743,7 +2736,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, tso_segs = tcp_init_tso_segs(skb, mss_now); BUG_ON(!tso_segs); - cwnd_quota = tcp_cwnd_test(tp, skb); + cwnd_quota = tcp_cwnd_test(tp); if (!cwnd_quota) { if (push_one == 2) /* Force out a loss probe pkt. */ @@ -2772,9 +2765,8 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, limit = mss_now; if (tso_segs > 1 && !tcp_urg_mode(tp)) limit = tcp_mss_split_point(sk, skb, mss_now, - min_t(unsigned int, - cwnd_quota, - max_segs), + min(cwnd_quota, + max_segs), nonagle); if (skb->len > limit && -- cgit v1.2.3 From d5b38a71d3334bc87931997f1b0a16464bbbe2ba Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 21:45:59 +0000 Subject: tcp: call tcp_set_skb_tso_segs() from tcp_write_xmit() tcp_write_xmit() calls tcp_init_tso_segs() to set gso_size and gso_segs on the packet. tcp_init_tso_segs() requires the stack to maintain an up to date tcp_skb_pcount(), and this makes sense for packets in rtx queue. Not so much for packets still in the write queue. In the following patch, we don't want to deal with tcp_skb_pcount() when moving payload from 2nd skb to 1st skb in the write queue. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240418214600.1291486-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_output.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index acbc76ca3e64..5e8665241f93 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1502,18 +1502,22 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb) } /* Initialize TSO segments for a packet. */ -static void tcp_set_skb_tso_segs(struct sk_buff *skb, unsigned int mss_now) +static int tcp_set_skb_tso_segs(struct sk_buff *skb, unsigned int mss_now) { + int tso_segs; + if (skb->len <= mss_now) { /* Avoid the costly divide in the normal * non-TSO case. */ - tcp_skb_pcount_set(skb, 1); TCP_SKB_CB(skb)->tcp_gso_size = 0; - } else { - tcp_skb_pcount_set(skb, DIV_ROUND_UP(skb->len, mss_now)); - TCP_SKB_CB(skb)->tcp_gso_size = mss_now; + tcp_skb_pcount_set(skb, 1); + return 1; } + TCP_SKB_CB(skb)->tcp_gso_size = mss_now; + tso_segs = DIV_ROUND_UP(skb->len, mss_now); + tcp_skb_pcount_set(skb, tso_segs); + return tso_segs; } /* Pcount in the middle of the write queue got changed, we need to do various @@ -2097,10 +2101,9 @@ static int tcp_init_tso_segs(struct sk_buff *skb, unsigned int mss_now) { int tso_segs = tcp_skb_pcount(skb); - if (!tso_segs || (tso_segs > 1 && tcp_skb_mss(skb) != mss_now)) { - tcp_set_skb_tso_segs(skb, mss_now); - tso_segs = tcp_skb_pcount(skb); - } + if (!tso_segs || (tso_segs > 1 && tcp_skb_mss(skb) != mss_now)) + return tcp_set_skb_tso_segs(skb, mss_now); + return tso_segs; } @@ -2733,9 +2736,6 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (tcp_pacing_check(sk)) break; - tso_segs = tcp_init_tso_segs(skb, mss_now); - BUG_ON(!tso_segs); - cwnd_quota = tcp_cwnd_test(tp); if (!cwnd_quota) { if (push_one == 2) @@ -2745,6 +2745,8 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, break; } + tso_segs = tcp_set_skb_tso_segs(skb, mss_now); + if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now))) { is_rwnd_limited = true; break; -- cgit v1.2.3 From 8ee602c635206ed012f979370094015857c02359 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 18 Apr 2024 21:46:00 +0000 Subject: tcp: try to send bigger TSO packets While investigating TCP performance, I found that TCP would sometimes send big skbs followed by a single MSS skb, in a 'locked' pattern. For instance, BIG TCP is enabled, MSS is set to have 4096 bytes of payload per segment. gso_max_size is set to 181000. This means that an optimal TCP packet size should contain 44 * 4096 = 180224 bytes of payload, However, I was seeing packets sizes interleaved in this pattern: 172032, 8192, 172032, 8192, 172032, 8192, tcp_tso_should_defer() heuristic is defeated, because after a split of a packet in write queue for whatever reason (this might be a too small CWND or a small enough pacing_rate), the leftover packet in the queue is smaller than the optimal size. It is time to try to make 'leftover packets' bigger so that tcp_tso_should_defer() can give its full potential. After this patch, we can see the following output: 14:13:34.009273 IP6 sender > receiver: Flags [P.], seq 4048380:4098360, ack 1, win 256, options [nop,nop,TS val 3425678144 ecr 1561784500], length 49980 14:13:34.010272 IP6 sender > receiver: Flags [P.], seq 4098360:4148340, ack 1, win 256, options [nop,nop,TS val 3425678145 ecr 1561784501], length 49980 14:13:34.011271 IP6 sender > receiver: Flags [P.], seq 4148340:4198320, ack 1, win 256, options [nop,nop,TS val 3425678146 ecr 1561784502], length 49980 14:13:34.012271 IP6 sender > receiver: Flags [P.], seq 4198320:4248300, ack 1, win 256, options [nop,nop,TS val 3425678147 ecr 1561784503], length 49980 14:13:34.013272 IP6 sender > receiver: Flags [P.], seq 4248300:4298280, ack 1, win 256, options [nop,nop,TS val 3425678148 ecr 1561784504], length 49980 14:13:34.014271 IP6 sender > receiver: Flags [P.], seq 4298280:4348260, ack 1, win 256, options [nop,nop,TS val 3425678149 ecr 1561784505], length 49980 14:13:34.015272 IP6 sender > receiver: Flags [P.], seq 4348260:4398240, ack 1, win 256, options [nop,nop,TS val 3425678150 ecr 1561784506], length 49980 14:13:34.016270 IP6 sender > receiver: Flags [P.], seq 4398240:4448220, ack 1, win 256, options [nop,nop,TS val 3425678151 ecr 1561784507], length 49980 14:13:34.017269 IP6 sender > receiver: Flags [P.], seq 4448220:4498200, ack 1, win 256, options [nop,nop,TS val 3425678152 ecr 1561784508], length 49980 14:13:34.018276 IP6 sender > receiver: Flags [P.], seq 4498200:4548180, ack 1, win 256, options [nop,nop,TS val 3425678153 ecr 1561784509], length 49980 14:13:34.019259 IP6 sender > receiver: Flags [P.], seq 4548180:4598160, ack 1, win 256, options [nop,nop,TS val 3425678154 ecr 1561784510], length 49980 With 200 concurrent flows on a 100Gbit NIC, we can see a reduction of TSO packets (and ACK packets) of about 30 %. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240418214600.1291486-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_output.c | 38 ++++++++++++++++++++++++++++++++++++-- 1 file changed, 36 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 5e8665241f93..99a1d88f7f47 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2683,6 +2683,36 @@ void tcp_chrono_stop(struct sock *sk, const enum tcp_chrono type) tcp_chrono_set(tp, TCP_CHRONO_BUSY); } +/* First skb in the write queue is smaller than ideal packet size. + * Check if we can move payload from the second skb in the queue. + */ +static void tcp_grow_skb(struct sock *sk, struct sk_buff *skb, int amount) +{ + struct sk_buff *next_skb = skb->next; + unsigned int nlen; + + if (tcp_skb_is_last(sk, skb)) + return; + + if (!tcp_skb_can_collapse(skb, next_skb)) + return; + + nlen = min_t(u32, amount, next_skb->len); + if (!nlen || !skb_shift(skb, next_skb, nlen)) + return; + + TCP_SKB_CB(skb)->end_seq += nlen; + TCP_SKB_CB(next_skb)->seq += nlen; + + if (!next_skb->len) { + TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(next_skb)->end_seq; + TCP_SKB_CB(skb)->eor = TCP_SKB_CB(next_skb)->eor; + TCP_SKB_CB(skb)->tcp_flags |= TCP_SKB_CB(next_skb)->tcp_flags; + tcp_unlink_write_queue(next_skb, sk); + tcp_wmem_free_skb(sk, next_skb); + } +} + /* This routine writes packets to the network. It advances the * send_head. This happens as incoming acks open up the remote * window for us. @@ -2723,6 +2753,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, max_segs = tcp_tso_segs(sk, mss_now); while ((skb = tcp_send_head(sk))) { unsigned int limit; + int missing_bytes; if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE) { /* "skb_mstamp_ns" is used as a start point for the retransmit timer */ @@ -2744,6 +2775,10 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, else break; } + cwnd_quota = min(cwnd_quota, max_segs); + missing_bytes = cwnd_quota * mss_now - skb->len; + if (missing_bytes > 0) + tcp_grow_skb(sk, skb, missing_bytes); tso_segs = tcp_set_skb_tso_segs(skb, mss_now); @@ -2767,8 +2802,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, limit = mss_now; if (tso_segs > 1 && !tcp_urg_mode(tp)) limit = tcp_mss_split_point(sk, skb, mss_now, - min(cwnd_quota, - max_segs), + cwnd_quota, nonagle); if (skb->len > limit && -- cgit v1.2.3 From 7ab4f16f9e2440e797eae88812f800458e5879d2 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Fri, 19 Apr 2024 12:08:39 +0100 Subject: net: extend ubuf_info callback to ops structure We'll need to associate additional callbacks with ubuf_info, introduce a structure holding ubuf_info callbacks. Apart from a more smarter io_uring notification management introduced in next patches, it can be used to generalise msg_zerocopy_put_abort() and also store ->sg_from_iter, which is currently passed in struct msghdr. Reviewed-by: Jens Axboe Reviewed-by: David Ahern Signed-off-by: Pavel Begunkov Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/all/a62015541de49c0e2a8a0377a1d5d0a5aeb07016.1713369317.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/tap.c | 2 +- drivers/net/tun.c | 2 +- drivers/net/xen-netback/common.h | 5 ++--- drivers/net/xen-netback/interface.c | 2 +- drivers/net/xen-netback/netback.c | 11 ++++++++--- drivers/vhost/net.c | 8 ++++++-- include/linux/skbuff.h | 19 +++++++++++-------- io_uring/notif.c | 18 +++++++++++++----- net/core/skbuff.c | 16 ++++++++++------ 9 files changed, 53 insertions(+), 30 deletions(-) (limited to 'net') diff --git a/drivers/net/tap.c b/drivers/net/tap.c index 9f0495e8df4d..bfdd3875fe86 100644 --- a/drivers/net/tap.c +++ b/drivers/net/tap.c @@ -754,7 +754,7 @@ static ssize_t tap_get_user(struct tap_queue *q, void *msg_control, skb_zcopy_init(skb, msg_control); } else if (msg_control) { struct ubuf_info *uarg = msg_control; - uarg->callback(NULL, uarg, false); + uarg->ops->complete(NULL, uarg, false); } dev_queue_xmit(skb); diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 0b3f21cba552..b7401d990680 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1906,7 +1906,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, skb_zcopy_init(skb, msg_control); } else if (msg_control) { struct ubuf_info *uarg = msg_control; - uarg->callback(NULL, uarg, false); + uarg->ops->complete(NULL, uarg, false); } skb_reset_network_header(skb); diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h index 1fcbd83f7ff2..17421da139f2 100644 --- a/drivers/net/xen-netback/common.h +++ b/drivers/net/xen-netback/common.h @@ -390,9 +390,8 @@ bool xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb); void xenvif_carrier_on(struct xenvif *vif); -/* Callback from stack when TX packet can be released */ -void xenvif_zerocopy_callback(struct sk_buff *skb, struct ubuf_info *ubuf, - bool zerocopy_success); +/* Callbacks from stack when TX packet can be released */ +extern const struct ubuf_info_ops xenvif_ubuf_ops; static inline pending_ring_idx_t nr_pending_reqs(struct xenvif_queue *queue) { diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index 7cff90aa8d24..65db5f14465f 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -593,7 +593,7 @@ int xenvif_init_queue(struct xenvif_queue *queue) for (i = 0; i < MAX_PENDING_REQS; i++) { queue->pending_tx_info[i].callback_struct = (struct ubuf_info_msgzc) - { { .callback = xenvif_zerocopy_callback }, + { { .ops = &xenvif_ubuf_ops }, { { .ctx = NULL, .desc = i } } }; queue->grant_tx_handle[i] = NETBACK_INVALID_HANDLE; diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index ef76850d9bcd..bab7e43ea05b 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -1156,7 +1156,7 @@ static int xenvif_handle_frag_list(struct xenvif_queue *queue, struct sk_buff *s uarg = skb_shinfo(skb)->destructor_arg; /* increase inflight counter to offset decrement in callback */ atomic_inc(&queue->inflight_packets); - uarg->callback(NULL, uarg, true); + uarg->ops->complete(NULL, uarg, true); skb_shinfo(skb)->destructor_arg = NULL; /* Fill the skb with the new (local) frags. */ @@ -1278,8 +1278,9 @@ static int xenvif_tx_submit(struct xenvif_queue *queue) return work_done; } -void xenvif_zerocopy_callback(struct sk_buff *skb, struct ubuf_info *ubuf_base, - bool zerocopy_success) +static void xenvif_zerocopy_callback(struct sk_buff *skb, + struct ubuf_info *ubuf_base, + bool zerocopy_success) { unsigned long flags; pending_ring_idx_t index; @@ -1312,6 +1313,10 @@ void xenvif_zerocopy_callback(struct sk_buff *skb, struct ubuf_info *ubuf_base, xenvif_skb_zerocopy_complete(queue); } +const struct ubuf_info_ops xenvif_ubuf_ops = { + .complete = xenvif_zerocopy_callback, +}; + static inline void xenvif_tx_dealloc_action(struct xenvif_queue *queue) { struct gnttab_unmap_grant_ref *gop; diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index c64ded183f8d..f16279351db5 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -380,7 +380,7 @@ static void vhost_zerocopy_signal_used(struct vhost_net *net, } } -static void vhost_zerocopy_callback(struct sk_buff *skb, +static void vhost_zerocopy_complete(struct sk_buff *skb, struct ubuf_info *ubuf_base, bool success) { struct ubuf_info_msgzc *ubuf = uarg_to_msgzc(ubuf_base); @@ -408,6 +408,10 @@ static void vhost_zerocopy_callback(struct sk_buff *skb, rcu_read_unlock_bh(); } +static const struct ubuf_info_ops vhost_ubuf_ops = { + .complete = vhost_zerocopy_complete, +}; + static inline unsigned long busy_clock(void) { return local_clock() >> 10; @@ -879,7 +883,7 @@ static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock) vq->heads[nvq->upend_idx].len = VHOST_DMA_IN_PROGRESS; ubuf->ctx = nvq->ubufs; ubuf->desc = nvq->upend_idx; - ubuf->ubuf.callback = vhost_zerocopy_callback; + ubuf->ubuf.ops = &vhost_ubuf_ops; ubuf->ubuf.flags = SKBFL_ZEROCOPY_FRAG; refcount_set(&ubuf->ubuf.refcnt, 1); msg.msg_control = &ctl; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9d24aec064e8..a110e97e074a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -527,6 +527,11 @@ enum { #define SKBFL_ALL_ZEROCOPY (SKBFL_ZEROCOPY_FRAG | SKBFL_PURE_ZEROCOPY | \ SKBFL_DONT_ORPHAN | SKBFL_MANAGED_FRAG_REFS) +struct ubuf_info_ops { + void (*complete)(struct sk_buff *, struct ubuf_info *, + bool zerocopy_success); +}; + /* * The callback notifies userspace to release buffers when skb DMA is done in * lower device, the skb last reference should be 0 when calling this. @@ -536,8 +541,7 @@ enum { * The desc field is used to track userspace buffer index. */ struct ubuf_info { - void (*callback)(struct sk_buff *, struct ubuf_info *, - bool zerocopy_success); + const struct ubuf_info_ops *ops; refcount_t refcnt; u8 flags; }; @@ -1662,14 +1666,13 @@ static inline void skb_set_end_offset(struct sk_buff *skb, unsigned int offset) } #endif +extern const struct ubuf_info_ops msg_zerocopy_ubuf_ops; + struct ubuf_info *msg_zerocopy_realloc(struct sock *sk, size_t size, struct ubuf_info *uarg); void msg_zerocopy_put_abort(struct ubuf_info *uarg, bool have_uref); -void msg_zerocopy_callback(struct sk_buff *skb, struct ubuf_info *uarg, - bool success); - int __zerocopy_sg_from_iter(struct msghdr *msg, struct sock *sk, struct sk_buff *skb, struct iov_iter *from, size_t length); @@ -1757,13 +1760,13 @@ static inline void *skb_zcopy_get_nouarg(struct sk_buff *skb) static inline void net_zcopy_put(struct ubuf_info *uarg) { if (uarg) - uarg->callback(NULL, uarg, true); + uarg->ops->complete(NULL, uarg, true); } static inline void net_zcopy_put_abort(struct ubuf_info *uarg, bool have_uref) { if (uarg) { - if (uarg->callback == msg_zerocopy_callback) + if (uarg->ops == &msg_zerocopy_ubuf_ops) msg_zerocopy_put_abort(uarg, have_uref); else if (have_uref) net_zcopy_put(uarg); @@ -1777,7 +1780,7 @@ static inline void skb_zcopy_clear(struct sk_buff *skb, bool zerocopy_success) if (uarg) { if (!skb_zcopy_is_nouarg(skb)) - uarg->callback(skb, uarg, zerocopy_success); + uarg->ops->complete(skb, uarg, zerocopy_success); skb_shinfo(skb)->flags &= ~SKBFL_ALL_ZEROCOPY; } diff --git a/io_uring/notif.c b/io_uring/notif.c index d3e703c37aba..204f0fc7d966 100644 --- a/io_uring/notif.c +++ b/io_uring/notif.c @@ -24,7 +24,7 @@ static void io_notif_complete_tw_ext(struct io_kiocb *notif, struct io_tw_state io_req_task_complete(notif, ts); } -static void io_tx_ubuf_callback(struct sk_buff *skb, struct ubuf_info *uarg, +static void io_tx_ubuf_complete(struct sk_buff *skb, struct ubuf_info *uarg, bool success) { struct io_notif_data *nd = container_of(uarg, struct io_notif_data, uarg); @@ -45,19 +45,27 @@ static void io_tx_ubuf_callback_ext(struct sk_buff *skb, struct ubuf_info *uarg, else if (!success && !nd->zc_copied) WRITE_ONCE(nd->zc_copied, true); } - io_tx_ubuf_callback(skb, uarg, success); + io_tx_ubuf_complete(skb, uarg, success); } +static const struct ubuf_info_ops io_ubuf_ops = { + .complete = io_tx_ubuf_complete, +}; + +static const struct ubuf_info_ops io_ubuf_ops_ext = { + .complete = io_tx_ubuf_callback_ext, +}; + void io_notif_set_extended(struct io_kiocb *notif) { struct io_notif_data *nd = io_notif_to_data(notif); - if (nd->uarg.callback != io_tx_ubuf_callback_ext) { + if (nd->uarg.ops != &io_ubuf_ops_ext) { nd->account_pages = 0; nd->zc_report = false; nd->zc_used = false; nd->zc_copied = false; - nd->uarg.callback = io_tx_ubuf_callback_ext; + nd->uarg.ops = &io_ubuf_ops_ext; notif->io_task_work.func = io_notif_complete_tw_ext; } } @@ -80,7 +88,7 @@ struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx) nd = io_notif_to_data(notif); nd->uarg.flags = IO_NOTIF_UBUF_FLAGS; - nd->uarg.callback = io_tx_ubuf_callback; + nd->uarg.ops = &io_ubuf_ops; refcount_set(&nd->uarg.refcnt, 1); return notif; } diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b99127712e67..9d36750459f3 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1708,7 +1708,7 @@ static struct ubuf_info *msg_zerocopy_alloc(struct sock *sk, size_t size) return NULL; } - uarg->ubuf.callback = msg_zerocopy_callback; + uarg->ubuf.ops = &msg_zerocopy_ubuf_ops; uarg->id = ((u32)atomic_inc_return(&sk->sk_zckey)) - 1; uarg->len = 1; uarg->bytelen = size; @@ -1734,7 +1734,7 @@ struct ubuf_info *msg_zerocopy_realloc(struct sock *sk, size_t size, u32 bytelen, next; /* there might be non MSG_ZEROCOPY users */ - if (uarg->callback != msg_zerocopy_callback) + if (uarg->ops != &msg_zerocopy_ubuf_ops) return NULL; /* realloc only when socket is locked (TCP, UDP cork), @@ -1845,8 +1845,8 @@ release: sock_put(sk); } -void msg_zerocopy_callback(struct sk_buff *skb, struct ubuf_info *uarg, - bool success) +static void msg_zerocopy_complete(struct sk_buff *skb, struct ubuf_info *uarg, + bool success) { struct ubuf_info_msgzc *uarg_zc = uarg_to_msgzc(uarg); @@ -1855,7 +1855,6 @@ void msg_zerocopy_callback(struct sk_buff *skb, struct ubuf_info *uarg, if (refcount_dec_and_test(&uarg->refcnt)) __msg_zerocopy_callback(uarg_zc); } -EXPORT_SYMBOL_GPL(msg_zerocopy_callback); void msg_zerocopy_put_abort(struct ubuf_info *uarg, bool have_uref) { @@ -1865,10 +1864,15 @@ void msg_zerocopy_put_abort(struct ubuf_info *uarg, bool have_uref) uarg_to_msgzc(uarg)->len--; if (have_uref) - msg_zerocopy_callback(NULL, uarg, true); + msg_zerocopy_complete(NULL, uarg, true); } EXPORT_SYMBOL_GPL(msg_zerocopy_put_abort); +const struct ubuf_info_ops msg_zerocopy_ubuf_ops = { + .complete = msg_zerocopy_complete, +}; +EXPORT_SYMBOL_GPL(msg_zerocopy_ubuf_ops); + int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb, struct msghdr *msg, int len, struct ubuf_info *uarg) -- cgit v1.2.3 From 65bada80dec1f2108a751644773b2120bd789934 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Fri, 19 Apr 2024 12:08:40 +0100 Subject: net: add callback for setting a ubuf_info to skb At the moment an skb can only have one ubuf_info associated with it, which might be a performance problem for zerocopy sends in cases like TCP via io_uring. Add a callback for assigning ubuf_info to skb, this way we will implement smarter assignment later like linking ubuf_info together. Note, it's an optional callback, which should be compatible with skb_zcopy_set(), that's because the net stack might potentially decide to clone an skb and take another reference to ubuf_info whenever it wishes. Also, a correct implementation should always be able to bind to an skb without prior ubuf_info, otherwise we could end up in a situation when the send would not be able to progress. Reviewed-by: Jens Axboe Reviewed-by: David Ahern Signed-off-by: Pavel Begunkov Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/all/b7918aadffeb787c84c9e72e34c729dc04f3a45d.1713369317.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 2 ++ net/core/skbuff.c | 20 ++++++++++++++------ 2 files changed, 16 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index a110e97e074a..ced69f37977f 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -530,6 +530,8 @@ enum { struct ubuf_info_ops { void (*complete)(struct sk_buff *, struct ubuf_info *, bool zerocopy_success); + /* has to be compatible with skb_zcopy_set() */ + int (*link_skb)(struct sk_buff *skb, struct ubuf_info *uarg); }; /* diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 9d36750459f3..4a0c10685cd2 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1880,11 +1880,18 @@ int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb, struct ubuf_info *orig_uarg = skb_zcopy(skb); int err, orig_len = skb->len; - /* An skb can only point to one uarg. This edge case happens when - * TCP appends to an skb, but zerocopy_realloc triggered a new alloc. - */ - if (orig_uarg && uarg != orig_uarg) - return -EEXIST; + if (uarg->ops->link_skb) { + err = uarg->ops->link_skb(skb, uarg); + if (err) + return err; + } else { + /* An skb can only point to one uarg. This edge case happens + * when TCP appends to an skb, but zerocopy_realloc triggered + * a new alloc. + */ + if (orig_uarg && uarg != orig_uarg) + return -EEXIST; + } err = __zerocopy_sg_from_iter(msg, sk, skb, &msg->msg_iter, len); if (err == -EFAULT || (err == -EMSGSIZE && skb->len == orig_len)) { @@ -1898,7 +1905,8 @@ int skb_zerocopy_iter_stream(struct sock *sk, struct sk_buff *skb, return err; } - skb_zcopy_set(skb, uarg, NULL); + if (!uarg->ops->link_skb) + skb_zcopy_set(skb, uarg, NULL); return skb->len - orig_len; } EXPORT_SYMBOL_GPL(skb_zerocopy_iter_stream); -- cgit v1.2.3 From bf2ac490d28c21a349e9eef81edc45320fca4a3c Mon Sep 17 00:00:00 2001 From: Donald Hunter Date: Thu, 18 Apr 2024 11:47:37 +0100 Subject: netfilter: nfnetlink: Handle ACK flags for batch messages The NLM_F_ACK flag is ignored for nfnetlink batch begin and end messages. This is a problem for ynl which wants to receive an ack for every message it sends, not just the commands in between the begin/end messages. Add processing for ACKs for begin/end messages and provide responses when requested. I have checked that iproute2, pyroute2 and systemd are unaffected by this change since none of them use NLM_F_ACK for batch begin/end. Signed-off-by: Donald Hunter Link: https://lore.kernel.org/r/20240418104737.77914-5-donald.hunter@gmail.com Signed-off-by: Jakub Kicinski --- net/netfilter/nfnetlink.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net') diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index c9fbe0f707b5..4abf660c7baf 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -427,6 +427,9 @@ replay_abort: nfnl_unlock(subsys_id); + if (nlh->nlmsg_flags & NLM_F_ACK) + nfnl_err_add(&err_list, nlh, 0, &extack); + while (skb->len >= nlmsg_total_size(0)) { int msglen, type; @@ -573,6 +576,8 @@ done: } else if (err) { ss->abort(net, oskb, NFNL_ABORT_NONE); netlink_ack(oskb, nlmsg_hdr(oskb), err, NULL); + } else if (nlh->nlmsg_flags & NLM_F_ACK) { + nfnl_err_add(&err_list, nlh, 0, &extack); } } else { enum nfnl_abort_action abort_action; -- cgit v1.2.3 From 2540088b836f9e208f7f722a67f9b168737036e8 Mon Sep 17 00:00:00 2001 From: Jun Gu Date: Fri, 19 Apr 2024 14:14:25 +0800 Subject: net: openvswitch: Check vport netdev name Ensure that the provided netdev name is not one of its aliases to prevent unnecessary creation and destruction of the vport by ovs-vswitchd. Signed-off-by: Jun Gu Acked-by: Eelco Chaudron Link: https://lore.kernel.org/r/20240419061425.132723-1-jun.gu@easystack.cn Signed-off-by: Jakub Kicinski --- net/openvswitch/vport-netdev.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 903537a5da22..618edc346c0f 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -78,7 +78,10 @@ struct vport *ovs_netdev_link(struct vport *vport, const char *name) int err; vport->dev = dev_get_by_name(ovs_dp_get_net(vport->dp), name); - if (!vport->dev) { + /* Ensure that the device exists and that the provided + * name is not one of its aliases. + */ + if (!vport->dev || strcmp(name, ovs_vport_name(vport))) { err = -ENODEV; goto error_free_vport; } -- cgit v1.2.3 From 1af2dface5d286dd1f2f3405a0d6fa9f2c8fb998 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 19 Apr 2024 16:51:02 -0700 Subject: af_unix: Don't access successor in unix_del_edges() during GC. syzbot reported use-after-free in unix_del_edges(). [0] What the repro does is basically repeat the following quickly. 1. pass a fd of an AF_UNIX socket to itself socketpair(AF_UNIX, SOCK_DGRAM, 0, [3, 4]) = 0 sendmsg(3, {..., msg_control=[{cmsg_len=20, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[4]}], ...}, 0) = 0 2. pass other fds of AF_UNIX sockets to the socket above socketpair(AF_UNIX, SOCK_SEQPACKET, 0, [5, 6]) = 0 sendmsg(3, {..., msg_control=[{cmsg_len=48, cmsg_level=SOL_SOCKET, cmsg_type=SCM_RIGHTS, cmsg_data=[5, 6]}], ...}, 0) = 0 3. close all sockets Here, two skb are created, and every unix_edge->successor is the first socket. Then, __unix_gc() will garbage-collect the two skb: (a) free skb with self-referencing fd (b) free skb holding other sockets After (a), the self-referencing socket will be scheduled to be freed later by the delayed_fput() task. syzbot repeated the sequences above (1. ~ 3.) quickly and triggered the task concurrently while GC was running. So, at (b), the socket was already freed, and accessing it was illegal. unix_del_edges() accesses the receiver socket as edge->successor to optimise GC. However, we should not do it during GC. Garbage-collecting sockets does not change the shape of the rest of the graph, so we need not call unix_update_graph() to update unix_graph_grouped when we purge skb. However, if we clean up all loops in the unix_walk_scc_fast() path, unix_graph_maybe_cyclic remains unchanged (true), and __unix_gc() will call unix_walk_scc_fast() continuously even though there is no socket to garbage-collect. To keep that optimisation while fixing UAF, let's add the same updating logic of unix_graph_maybe_cyclic in unix_walk_scc_fast() as done in unix_walk_scc() and __unix_walk_scc(). Note that when unix_del_edges() is called from other places, the receiver socket is always alive: - sendmsg: the successor's sk_refcnt is bumped by sock_hold() unix_find_other() for SOCK_DGRAM, connect() for SOCK_STREAM - recvmsg: the successor is the receiver, and its fd is alive [0]: BUG: KASAN: slab-use-after-free in unix_edge_successor net/unix/garbage.c:109 [inline] BUG: KASAN: slab-use-after-free in unix_del_edge net/unix/garbage.c:165 [inline] BUG: KASAN: slab-use-after-free in unix_del_edges+0x148/0x630 net/unix/garbage.c:237 Read of size 8 at addr ffff888079c6e640 by task kworker/u8:6/1099 CPU: 0 PID: 1099 Comm: kworker/u8:6 Not tainted 6.9.0-rc4-next-20240418-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 Workqueue: events_unbound __unix_gc Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 print_address_description mm/kasan/report.c:377 [inline] print_report+0x169/0x550 mm/kasan/report.c:488 kasan_report+0x143/0x180 mm/kasan/report.c:601 unix_edge_successor net/unix/garbage.c:109 [inline] unix_del_edge net/unix/garbage.c:165 [inline] unix_del_edges+0x148/0x630 net/unix/garbage.c:237 unix_destroy_fpl+0x59/0x210 net/unix/garbage.c:298 unix_detach_fds net/unix/af_unix.c:1811 [inline] unix_destruct_scm+0x13e/0x210 net/unix/af_unix.c:1826 skb_release_head_state+0x100/0x250 net/core/skbuff.c:1127 skb_release_all net/core/skbuff.c:1138 [inline] __kfree_skb net/core/skbuff.c:1154 [inline] kfree_skb_reason+0x16d/0x3b0 net/core/skbuff.c:1190 __skb_queue_purge_reason include/linux/skbuff.h:3251 [inline] __skb_queue_purge include/linux/skbuff.h:3256 [inline] __unix_gc+0x1732/0x1830 net/unix/garbage.c:575 process_one_work kernel/workqueue.c:3218 [inline] process_scheduled_works+0xa2c/0x1830 kernel/workqueue.c:3299 worker_thread+0x86d/0xd70 kernel/workqueue.c:3380 kthread+0x2f0/0x390 kernel/kthread.c:389 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 Allocated by task 14427: kasan_save_stack mm/kasan/common.c:47 [inline] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 unpoison_slab_object mm/kasan/common.c:312 [inline] __kasan_slab_alloc+0x66/0x80 mm/kasan/common.c:338 kasan_slab_alloc include/linux/kasan.h:201 [inline] slab_post_alloc_hook mm/slub.c:3897 [inline] slab_alloc_node mm/slub.c:3957 [inline] kmem_cache_alloc_noprof+0x135/0x290 mm/slub.c:3964 sk_prot_alloc+0x58/0x210 net/core/sock.c:2074 sk_alloc+0x38/0x370 net/core/sock.c:2133 unix_create1+0xb4/0x770 unix_create+0x14e/0x200 net/unix/af_unix.c:1034 __sock_create+0x490/0x920 net/socket.c:1571 sock_create net/socket.c:1622 [inline] __sys_socketpair+0x33e/0x720 net/socket.c:1773 __do_sys_socketpair net/socket.c:1822 [inline] __se_sys_socketpair net/socket.c:1819 [inline] __x64_sys_socketpair+0x9b/0xb0 net/socket.c:1819 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task 1805: kasan_save_stack mm/kasan/common.c:47 [inline] kasan_save_track+0x3f/0x80 mm/kasan/common.c:68 kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:579 poison_slab_object+0xe0/0x150 mm/kasan/common.c:240 __kasan_slab_free+0x37/0x60 mm/kasan/common.c:256 kasan_slab_free include/linux/kasan.h:184 [inline] slab_free_hook mm/slub.c:2190 [inline] slab_free mm/slub.c:4393 [inline] kmem_cache_free+0x145/0x340 mm/slub.c:4468 sk_prot_free net/core/sock.c:2114 [inline] __sk_destruct+0x467/0x5f0 net/core/sock.c:2208 sock_put include/net/sock.h:1948 [inline] unix_release_sock+0xa8b/0xd20 net/unix/af_unix.c:665 unix_release+0x91/0xc0 net/unix/af_unix.c:1049 __sock_release net/socket.c:659 [inline] sock_close+0xbc/0x240 net/socket.c:1421 __fput+0x406/0x8b0 fs/file_table.c:422 delayed_fput+0x59/0x80 fs/file_table.c:445 process_one_work kernel/workqueue.c:3218 [inline] process_scheduled_works+0xa2c/0x1830 kernel/workqueue.c:3299 worker_thread+0x86d/0xd70 kernel/workqueue.c:3380 kthread+0x2f0/0x390 kernel/kthread.c:389 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 The buggy address belongs to the object at ffff888079c6e000 which belongs to the cache UNIX of size 1920 The buggy address is located 1600 bytes inside of freed 1920-byte region [ffff888079c6e000, ffff888079c6e780) Reported-by: syzbot+f3f3eef1d2100200e593@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=f3f3eef1d2100200e593 Fixes: 77e5593aebba ("af_unix: Skip GC if no cycle exists.") Fixes: fd86344823b5 ("af_unix: Try not to hold unix_gc_lock during accept().") Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240419235102.31707-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni --- net/unix/garbage.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 95240a59808f..d76450133e4f 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -158,11 +158,14 @@ static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) unix_update_graph(unix_edge_successor(edge)); } +static bool gc_in_progress; + static void unix_del_edge(struct scm_fp_list *fpl, struct unix_edge *edge) { struct unix_vertex *vertex = edge->predecessor->vertex; - unix_update_graph(unix_edge_successor(edge)); + if (!gc_in_progress) + unix_update_graph(unix_edge_successor(edge)); list_del(&edge->vertex_entry); vertex->out_degree--; @@ -237,8 +240,10 @@ void unix_del_edges(struct scm_fp_list *fpl) unix_del_edge(fpl, edge); } while (i < fpl->count_unix); - receiver = fpl->edges[0].successor; - receiver->scm_stat.nr_unix_fds -= fpl->count_unix; + if (!gc_in_progress) { + receiver = fpl->edges[0].successor; + receiver->scm_stat.nr_unix_fds -= fpl->count_unix; + } WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - fpl->count_unix); out: WRITE_ONCE(fpl->user->unix_inflight, fpl->user->unix_inflight - fpl->count); @@ -526,6 +531,8 @@ static void unix_walk_scc(struct sk_buff_head *hitlist) static void unix_walk_scc_fast(struct sk_buff_head *hitlist) { + unix_graph_maybe_cyclic = false; + while (!list_empty(&unix_unvisited_vertices)) { struct unix_vertex *vertex; struct list_head scc; @@ -543,6 +550,8 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) if (scc_dead) unix_collect_skb(&scc, hitlist); + else if (!unix_graph_maybe_cyclic) + unix_graph_maybe_cyclic = unix_scc_cyclic(&scc); list_del(&scc); } @@ -550,8 +559,6 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); } -static bool gc_in_progress; - static void __unix_gc(struct work_struct *work) { struct sk_buff_head hitlist; -- cgit v1.2.3 From ce05d0f20368b583b43c99a7c8673e8a7187b76b Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 19 Apr 2024 19:35:39 -0700 Subject: netdev: support dumping a single netdev in qstats Having to filter the right ifindex in the tests is a bit tedious. Add support for dumping qstats for a single ifindex. Reviewed-by: David Ahern Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240420023543.3300306-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/netdev.yaml | 1 + net/core/netdev-genl-gen.c | 1 + net/core/netdev-genl.c | 52 ++++++++++++++++++++++++--------- 3 files changed, 41 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index 76352dbd2be4..679c4130707c 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -486,6 +486,7 @@ operations: dump: request: attributes: + - ifindex - scope reply: attributes: diff --git a/net/core/netdev-genl-gen.c b/net/core/netdev-genl-gen.c index 8d8ace9ef87f..8350a0afa9ec 100644 --- a/net/core/netdev-genl-gen.c +++ b/net/core/netdev-genl-gen.c @@ -70,6 +70,7 @@ static const struct nla_policy netdev_napi_get_dump_nl_policy[NETDEV_A_NAPI_IFIN /* NETDEV_CMD_QSTATS_GET - dump */ static const struct nla_policy netdev_qstats_get_nl_policy[NETDEV_A_QSTATS_SCOPE + 1] = { + [NETDEV_A_QSTATS_IFINDEX] = NLA_POLICY_MIN(NLA_U32, 1), [NETDEV_A_QSTATS_SCOPE] = NLA_POLICY_MASK(NLA_UINT, 0x1), }; diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 7004b3399c2b..dd6510f2c652 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -639,6 +639,24 @@ nla_put_failure: return -EMSGSIZE; } +static int +netdev_nl_qstats_get_dump_one(struct net_device *netdev, unsigned int scope, + struct sk_buff *skb, const struct genl_info *info, + struct netdev_nl_dump_ctx *ctx) +{ + if (!netdev->stat_ops) + return 0; + + switch (scope) { + case 0: + return netdev_nl_stats_by_netdev(netdev, skb, info); + case NETDEV_QSTATS_SCOPE_QUEUE: + return netdev_nl_stats_by_queue(netdev, skb, info, ctx); + } + + return -EINVAL; /* Should not happen, per netlink policy */ +} + int netdev_nl_qstats_get_dumpit(struct sk_buff *skb, struct netlink_callback *cb) { @@ -646,6 +664,7 @@ int netdev_nl_qstats_get_dumpit(struct sk_buff *skb, const struct genl_info *info = genl_info_dump(cb); struct net *net = sock_net(skb->sk); struct net_device *netdev; + unsigned int ifindex; unsigned int scope; int err = 0; @@ -653,21 +672,28 @@ int netdev_nl_qstats_get_dumpit(struct sk_buff *skb, if (info->attrs[NETDEV_A_QSTATS_SCOPE]) scope = nla_get_uint(info->attrs[NETDEV_A_QSTATS_SCOPE]); - rtnl_lock(); - for_each_netdev_dump(net, netdev, ctx->ifindex) { - if (!netdev->stat_ops) - continue; + ifindex = 0; + if (info->attrs[NETDEV_A_QSTATS_IFINDEX]) + ifindex = nla_get_u32(info->attrs[NETDEV_A_QSTATS_IFINDEX]); - switch (scope) { - case 0: - err = netdev_nl_stats_by_netdev(netdev, skb, info); - break; - case NETDEV_QSTATS_SCOPE_QUEUE: - err = netdev_nl_stats_by_queue(netdev, skb, info, ctx); - break; + rtnl_lock(); + if (ifindex) { + netdev = __dev_get_by_index(net, ifindex); + if (netdev && netdev->stat_ops) { + err = netdev_nl_qstats_get_dump_one(netdev, scope, skb, + info, ctx); + } else { + NL_SET_BAD_ATTR(info->extack, + info->attrs[NETDEV_A_QSTATS_IFINDEX]); + err = netdev ? -EOPNOTSUPP : -ENODEV; + } + } else { + for_each_netdev_dump(net, netdev, ctx->ifindex) { + err = netdev_nl_qstats_get_dump_one(netdev, scope, skb, + info, ctx); + if (err < 0) + break; } - if (err < 0) - break; } rtnl_unlock(); -- cgit v1.2.3 From 652332e3f1d6209dab372e0dfc7a5bbe209bf698 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 19 Apr 2024 19:35:40 -0700 Subject: netlink: move extack writing helpers Next change will need them in netlink_dump_done(), pure move. Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240420023543.3300306-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 126 +++++++++++++++++++++++------------------------ 1 file changed, 63 insertions(+), 63 deletions(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index dc8c3c01d51b..c5bb09597831 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2165,6 +2165,69 @@ __nlmsg_put(struct sk_buff *skb, u32 portid, u32 seq, int type, int len, int fla } EXPORT_SYMBOL(__nlmsg_put); +static size_t +netlink_ack_tlv_len(struct netlink_sock *nlk, int err, + const struct netlink_ext_ack *extack) +{ + size_t tlvlen; + + if (!extack || !test_bit(NETLINK_F_EXT_ACK, &nlk->flags)) + return 0; + + tlvlen = 0; + if (extack->_msg) + tlvlen += nla_total_size(strlen(extack->_msg) + 1); + if (extack->cookie_len) + tlvlen += nla_total_size(extack->cookie_len); + + /* Following attributes are only reported as error (not warning) */ + if (!err) + return tlvlen; + + if (extack->bad_attr) + tlvlen += nla_total_size(sizeof(u32)); + if (extack->policy) + tlvlen += netlink_policy_dump_attr_size_estimate(extack->policy); + if (extack->miss_type) + tlvlen += nla_total_size(sizeof(u32)); + if (extack->miss_nest) + tlvlen += nla_total_size(sizeof(u32)); + + return tlvlen; +} + +static void +netlink_ack_tlv_fill(struct sk_buff *in_skb, struct sk_buff *skb, + struct nlmsghdr *nlh, int err, + const struct netlink_ext_ack *extack) +{ + if (extack->_msg) + WARN_ON(nla_put_string(skb, NLMSGERR_ATTR_MSG, extack->_msg)); + if (extack->cookie_len) + WARN_ON(nla_put(skb, NLMSGERR_ATTR_COOKIE, + extack->cookie_len, extack->cookie)); + + if (!err) + return; + + if (extack->bad_attr && + !WARN_ON((u8 *)extack->bad_attr < in_skb->data || + (u8 *)extack->bad_attr >= in_skb->data + in_skb->len)) + WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_OFFS, + (u8 *)extack->bad_attr - (u8 *)nlh)); + if (extack->policy) + netlink_policy_dump_write_attr(skb, extack->policy, + NLMSGERR_ATTR_POLICY); + if (extack->miss_type) + WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_MISS_TYPE, + extack->miss_type)); + if (extack->miss_nest && + !WARN_ON((u8 *)extack->miss_nest < in_skb->data || + (u8 *)extack->miss_nest > in_skb->data + in_skb->len)) + WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_MISS_NEST, + (u8 *)extack->miss_nest - (u8 *)nlh)); +} + /* * It looks a bit ugly. * It would be better to create kernel thread. @@ -2406,69 +2469,6 @@ error_free: } EXPORT_SYMBOL(__netlink_dump_start); -static size_t -netlink_ack_tlv_len(struct netlink_sock *nlk, int err, - const struct netlink_ext_ack *extack) -{ - size_t tlvlen; - - if (!extack || !test_bit(NETLINK_F_EXT_ACK, &nlk->flags)) - return 0; - - tlvlen = 0; - if (extack->_msg) - tlvlen += nla_total_size(strlen(extack->_msg) + 1); - if (extack->cookie_len) - tlvlen += nla_total_size(extack->cookie_len); - - /* Following attributes are only reported as error (not warning) */ - if (!err) - return tlvlen; - - if (extack->bad_attr) - tlvlen += nla_total_size(sizeof(u32)); - if (extack->policy) - tlvlen += netlink_policy_dump_attr_size_estimate(extack->policy); - if (extack->miss_type) - tlvlen += nla_total_size(sizeof(u32)); - if (extack->miss_nest) - tlvlen += nla_total_size(sizeof(u32)); - - return tlvlen; -} - -static void -netlink_ack_tlv_fill(struct sk_buff *in_skb, struct sk_buff *skb, - struct nlmsghdr *nlh, int err, - const struct netlink_ext_ack *extack) -{ - if (extack->_msg) - WARN_ON(nla_put_string(skb, NLMSGERR_ATTR_MSG, extack->_msg)); - if (extack->cookie_len) - WARN_ON(nla_put(skb, NLMSGERR_ATTR_COOKIE, - extack->cookie_len, extack->cookie)); - - if (!err) - return; - - if (extack->bad_attr && - !WARN_ON((u8 *)extack->bad_attr < in_skb->data || - (u8 *)extack->bad_attr >= in_skb->data + in_skb->len)) - WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_OFFS, - (u8 *)extack->bad_attr - (u8 *)nlh)); - if (extack->policy) - netlink_policy_dump_write_attr(skb, extack->policy, - NLMSGERR_ATTR_POLICY); - if (extack->miss_type) - WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_MISS_TYPE, - extack->miss_type)); - if (extack->miss_nest && - !WARN_ON((u8 *)extack->miss_nest < in_skb->data || - (u8 *)extack->miss_nest > in_skb->data + in_skb->len)) - WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_MISS_NEST, - (u8 *)extack->miss_nest - (u8 *)nlh)); -} - void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err, const struct netlink_ext_ack *extack) { -- cgit v1.2.3 From 8af4f60472fce1f22db5068107b37bcc1a65eabd Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 19 Apr 2024 19:35:41 -0700 Subject: netlink: support all extack types in dumps Note that when this commit message refers to netlink dump it only means the actual dumping part, the parsing / dump start is handled by the same code as "doit". Commit 4a19edb60d02 ("netlink: Pass extack to dump handlers") added support for returning extack messages from dump handlers, but left out other extack info, e.g. bad attribute. This used to be fine because until YNL we had little practical use for the machine readable attributes, and only messages were used in practice. YNL flips the preference 180 degrees, it's now much more useful to point to a bad attr with NL_SET_BAD_ATTR() than type an English message saying "attribute XYZ is $reason-why-bad". Support all of extack. The fact that extack only gets added if it fits remains unaddressed. Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240420023543.3300306-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index c5bb09597831..fa9c090cf629 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2198,7 +2198,7 @@ netlink_ack_tlv_len(struct netlink_sock *nlk, int err, static void netlink_ack_tlv_fill(struct sk_buff *in_skb, struct sk_buff *skb, - struct nlmsghdr *nlh, int err, + const struct nlmsghdr *nlh, int err, const struct netlink_ext_ack *extack) { if (extack->_msg) @@ -2214,7 +2214,7 @@ netlink_ack_tlv_fill(struct sk_buff *in_skb, struct sk_buff *skb, !WARN_ON((u8 *)extack->bad_attr < in_skb->data || (u8 *)extack->bad_attr >= in_skb->data + in_skb->len)) WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_OFFS, - (u8 *)extack->bad_attr - (u8 *)nlh)); + (u8 *)extack->bad_attr - (const u8 *)nlh)); if (extack->policy) netlink_policy_dump_write_attr(skb, extack->policy, NLMSGERR_ATTR_POLICY); @@ -2225,7 +2225,7 @@ netlink_ack_tlv_fill(struct sk_buff *in_skb, struct sk_buff *skb, !WARN_ON((u8 *)extack->miss_nest < in_skb->data || (u8 *)extack->miss_nest > in_skb->data + in_skb->len)) WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_MISS_NEST, - (u8 *)extack->miss_nest - (u8 *)nlh)); + (u8 *)extack->miss_nest - (const u8 *)nlh)); } /* @@ -2238,6 +2238,7 @@ static int netlink_dump_done(struct netlink_sock *nlk, struct sk_buff *skb, struct netlink_ext_ack *extack) { struct nlmsghdr *nlh; + size_t extack_len; nlh = nlmsg_put_answer(skb, cb, NLMSG_DONE, sizeof(nlk->dump_done_errno), NLM_F_MULTI | cb->answer_flags); @@ -2247,10 +2248,14 @@ static int netlink_dump_done(struct netlink_sock *nlk, struct sk_buff *skb, nl_dump_check_consistent(cb, nlh); memcpy(nlmsg_data(nlh), &nlk->dump_done_errno, sizeof(nlk->dump_done_errno)); - if (extack->_msg && test_bit(NETLINK_F_EXT_ACK, &nlk->flags)) { + extack_len = netlink_ack_tlv_len(nlk, nlk->dump_done_errno, extack); + if (extack_len) { nlh->nlmsg_flags |= NLM_F_ACK_TLVS; - if (!nla_put_string(skb, NLMSGERR_ATTR_MSG, extack->_msg)) + if (skb_tailroom(skb) >= extack_len) { + netlink_ack_tlv_fill(cb->skb, skb, cb->nlh, + nlk->dump_done_errno, extack); nlmsg_end(skb, nlh); + } } return 0; -- cgit v1.2.3 From 1c04b46cbdddc7882eeb671521035ea884245b9f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 21 Apr 2024 18:57:53 +0000 Subject: neighbour: fix neigh_master_filtered() If we no longer hold RTNL, we must use netdev_master_upper_dev_get_rcu() instead of netdev_master_upper_dev_get(). Fixes: ba0f78069423 ("neighbour: no longer hold RTNL in neigh_dump_info()") Signed-off-by: Eric Dumazet Reviewed-by: Jiri Pirko Link: https://lore.kernel.org/r/20240421185753.1808077-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/neighbour.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 0805c00c63d4..af270c202d9a 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2682,7 +2682,7 @@ static bool neigh_master_filtered(struct net_device *dev, int master_idx) if (!master_idx) return false; - master = dev ? netdev_master_upper_dev_get(dev) : NULL; + master = dev ? netdev_master_upper_dev_get_rcu(dev) : NULL; /* 0 is already used to denote NDA_MASTER wasn't passed, therefore need another * invalid value for ifindex to denote "no master". -- cgit v1.2.3 From 520713a93d550406dae14d49cdb8778d70cecdfd Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Fri, 15 Mar 2024 19:11:30 +0100 Subject: sysctl: treewide: drop unused argument ctl_table_root::set_ownership(table) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove the 'table' argument from set_ownership as it is never used. This change is a step towards putting "struct ctl_table" into .rodata and eventually having sysctl core only use "const struct ctl_table". The patch was created with the following coccinelle script: @@ identifier func, head, table, uid, gid; @@ void func( struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { ... } No additional occurrences of 'set_ownership' were found after doing a tree-wide search. Reviewed-by: Joel Granados Signed-off-by: Thomas Weißschuh Signed-off-by: Joel Granados --- fs/proc/proc_sysctl.c | 2 +- include/linux/sysctl.h | 1 - ipc/ipc_sysctl.c | 3 +-- ipc/mq_sysctl.c | 3 +-- net/sysctl_net.c | 1 - 5 files changed, 3 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 37cde0efee57..ed3a41ed9705 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -480,7 +480,7 @@ static struct inode *proc_sys_make_inode(struct super_block *sb, } if (root->set_ownership) - root->set_ownership(head, table, &inode->i_uid, &inode->i_gid); + root->set_ownership(head, &inode->i_uid, &inode->i_gid); else { inode->i_uid = GLOBAL_ROOT_UID; inode->i_gid = GLOBAL_ROOT_GID; diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index ee7d33b89e9e..60333a6b9370 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -205,7 +205,6 @@ struct ctl_table_root { struct ctl_table_set default_set; struct ctl_table_set *(*lookup)(struct ctl_table_root *root); void (*set_ownership)(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid); int (*permissions)(struct ctl_table_header *head, struct ctl_table *table); }; diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c index 45cb1dabce29..1a5085e5b178 100644 --- a/ipc/ipc_sysctl.c +++ b/ipc/ipc_sysctl.c @@ -192,7 +192,6 @@ static int set_is_seen(struct ctl_table_set *set) } static void ipc_set_ownership(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { struct ipc_namespace *ns = @@ -224,7 +223,7 @@ static int ipc_permissions(struct ctl_table_header *head, struct ctl_table *tabl kuid_t ns_root_uid; kgid_t ns_root_gid; - ipc_set_ownership(head, table, &ns_root_uid, &ns_root_gid); + ipc_set_ownership(head, &ns_root_uid, &ns_root_gid); if (uid_eq(current_euid(), ns_root_uid)) mode >>= 6; diff --git a/ipc/mq_sysctl.c b/ipc/mq_sysctl.c index 21fba3a6edaf..6bb1c5397c69 100644 --- a/ipc/mq_sysctl.c +++ b/ipc/mq_sysctl.c @@ -78,7 +78,6 @@ static int set_is_seen(struct ctl_table_set *set) } static void mq_set_ownership(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { struct ipc_namespace *ns = @@ -97,7 +96,7 @@ static int mq_permissions(struct ctl_table_header *head, struct ctl_table *table kuid_t ns_root_uid; kgid_t ns_root_gid; - mq_set_ownership(head, table, &ns_root_uid, &ns_root_gid); + mq_set_ownership(head, &ns_root_uid, &ns_root_gid); if (uid_eq(current_euid(), ns_root_uid)) mode >>= 6; diff --git a/net/sysctl_net.c b/net/sysctl_net.c index 051ed5f6fc93..a0a7a79991f9 100644 --- a/net/sysctl_net.c +++ b/net/sysctl_net.c @@ -54,7 +54,6 @@ static int net_ctl_permissions(struct ctl_table_header *head, } static void net_ctl_set_ownership(struct ctl_table_header *head, - struct ctl_table *table, kuid_t *uid, kgid_t *gid) { struct net *net = container_of(head->set, struct net, sysctls); -- cgit v1.2.3 From 795f90c6f13c30484ff10355a6775979f57f78cb Mon Sep 17 00:00:00 2001 From: Thomas Weißschuh Date: Fri, 15 Mar 2024 19:11:31 +0100 Subject: sysctl: treewide: constify argument ctl_table_root::permissions(table) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The permissions callback should not modify the ctl_table. Enforce this expectation via the typesystem. This is a step to put "struct ctl_table" into .rodata throughout the kernel. The patch was created with the following coccinelle script: @@ identifier func, head, ctl; @@ int func( struct ctl_table_header *head, - struct ctl_table *ctl) + const struct ctl_table *ctl) { ... } (insert_entry() from fs/proc/proc_sysctl.c is a false-positive) No additional occurrences of '.permissions =' were found after a tree-wide search for places missed by the conccinelle script. Reviewed-by: Joel Granados Signed-off-by: Thomas Weißschuh Signed-off-by: Joel Granados --- include/linux/sysctl.h | 2 +- ipc/ipc_sysctl.c | 2 +- ipc/mq_sysctl.c | 2 +- kernel/ucount.c | 2 +- net/sysctl_net.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 60333a6b9370..f9214de0490c 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -206,7 +206,7 @@ struct ctl_table_root { struct ctl_table_set *(*lookup)(struct ctl_table_root *root); void (*set_ownership)(struct ctl_table_header *head, kuid_t *uid, kgid_t *gid); - int (*permissions)(struct ctl_table_header *head, struct ctl_table *table); + int (*permissions)(struct ctl_table_header *head, const struct ctl_table *table); }; #define register_sysctl(path, table) \ diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c index 1a5085e5b178..19b2a67aef40 100644 --- a/ipc/ipc_sysctl.c +++ b/ipc/ipc_sysctl.c @@ -204,7 +204,7 @@ static void ipc_set_ownership(struct ctl_table_header *head, *gid = gid_valid(ns_root_gid) ? ns_root_gid : GLOBAL_ROOT_GID; } -static int ipc_permissions(struct ctl_table_header *head, struct ctl_table *table) +static int ipc_permissions(struct ctl_table_header *head, const struct ctl_table *table) { int mode = table->mode; diff --git a/ipc/mq_sysctl.c b/ipc/mq_sysctl.c index 6bb1c5397c69..43c0825da9e8 100644 --- a/ipc/mq_sysctl.c +++ b/ipc/mq_sysctl.c @@ -90,7 +90,7 @@ static void mq_set_ownership(struct ctl_table_header *head, *gid = gid_valid(ns_root_gid) ? ns_root_gid : GLOBAL_ROOT_GID; } -static int mq_permissions(struct ctl_table_header *head, struct ctl_table *table) +static int mq_permissions(struct ctl_table_header *head, const struct ctl_table *table) { int mode = table->mode; kuid_t ns_root_uid; diff --git a/kernel/ucount.c b/kernel/ucount.c index e196da0204dc..4d5b9c12c014 100644 --- a/kernel/ucount.c +++ b/kernel/ucount.c @@ -38,7 +38,7 @@ static int set_is_seen(struct ctl_table_set *set) } static int set_permissions(struct ctl_table_header *head, - struct ctl_table *table) + const struct ctl_table *table) { struct user_namespace *user_ns = container_of(head->set, struct user_namespace, set); diff --git a/net/sysctl_net.c b/net/sysctl_net.c index a0a7a79991f9..f5017012a049 100644 --- a/net/sysctl_net.c +++ b/net/sysctl_net.c @@ -40,7 +40,7 @@ static int is_seen(struct ctl_table_set *set) /* Return standard mode bits for table entry. */ static int net_ctl_permissions(struct ctl_table_header *head, - struct ctl_table *table) + const struct ctl_table *table) { struct net *net = container_of(head->set, struct net, sysctls); -- cgit v1.2.3 From c6e7f276841d9a31009766f3e6f33e00cf3d93ed Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Mon, 22 Apr 2024 05:38:54 -0700 Subject: net: core: Fix documentation Fix bad grammar in description of init_dummy_netdev() function. This topic showed up in the review of the "allocate dummy device dynamically" patch set. Suggested-by: Ido Schimmel Signed-off-by: Breno Leitao Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 8bdc59074b29..44c2d698fd1b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -10424,7 +10424,7 @@ EXPORT_SYMBOL(register_netdevice); * init_dummy_netdev - init a dummy network device for NAPI * @dev: device to init * - * This takes a network device structure and initialize the minimum + * This takes a network device structure and initializes the minimum * amount of fields so it can be used to schedule NAPI polls without * registering a full blown interface. This is to be used by drivers * that need to tie several hardware interfaces to a single NAPI @@ -10433,7 +10433,7 @@ EXPORT_SYMBOL(register_netdevice); void init_dummy_netdev(struct net_device *dev) { /* Clear everything. Note we don't initialize spinlocks - * are they aren't supposed to be taken by any of the + * as they aren't supposed to be taken by any of the * NAPI code and this dummy netdev is supposed to be * only ever used for NAPI polls */ -- cgit v1.2.3 From f8d05679fb3faae478d604177b0c188b340371cd Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Mon, 22 Apr 2024 05:38:55 -0700 Subject: net: free_netdev: exit earlier if dummy For dummy devices, exit earlier at free_netdev() instead of executing the whole function. This is necessary, because dummy devices are special, and shouldn't have the second part of the function executed. Otherwise reg_state, which is NETREG_DUMMY, will be overwritten and there will be no way to identify that this is a dummy device. Also, this device do not need the final put_device(), since dummy devices are not registered (through register_netdevice()), where the device reference is increased (at netdev_register_kobject()/device_add()). Suggested-by: Jakub Kicinski Signed-off-by: Breno Leitao Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- net/core/dev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 44c2d698fd1b..62b39d6b1d8f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -11066,7 +11066,8 @@ void free_netdev(struct net_device *dev) phy_link_topo_destroy(dev->link_topo); /* Compatibility with error handling in drivers */ - if (dev->reg_state == NETREG_UNINITIALIZED) { + if (dev->reg_state == NETREG_UNINITIALIZED || + dev->reg_state == NETREG_DUMMY) { netdev_freemem(dev); return; } -- cgit v1.2.3 From c661050f93d3fd37a33c06041bb18a89688de7d2 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Mon, 22 Apr 2024 05:38:56 -0700 Subject: net: create a dummy net_device allocator It is impossible to use init_dummy_netdev together with alloc_netdev() as the 'setup' argument. This is because alloc_netdev() initializes some fields in the net_device structure, and later init_dummy_netdev() memzero them all. This causes some problems as reported here: https://lore.kernel.org/all/20240322082336.49f110cc@kernel.org/ Split the init_dummy_netdev() function in two. Create a new function called init_dummy_netdev_core() that does not memzero the net_device structure. Then have init_dummy_netdev() memzero-ing and calling init_dummy_netdev_core(), keeping the old behaviour. init_dummy_netdev_core() is the new function that could be called as an argument for alloc_netdev(). Also, create a helper to allocate and initialize dummy net devices, leveraging init_dummy_netdev_core() as the setup argument. This function basically simplify the allocation of dummy devices, by allocating and initializing it. Freeing the device continue to be done through free_netdev() Suggested-by: Jakub Kicinski Signed-off-by: Breno Leitao Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 +++ net/core/dev.c | 56 ++++++++++++++++++++++++++++++++--------------- 2 files changed, 41 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d45f330d083d..f849e7d110ed 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4519,6 +4519,9 @@ static inline void netif_addr_unlock_bh(struct net_device *dev) void ether_setup(struct net_device *dev); +/* Allocate dummy net_device */ +struct net_device *alloc_netdev_dummy(int sizeof_priv); + /* Support for loadable net-drivers */ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, unsigned char name_assign_type, diff --git a/net/core/dev.c b/net/core/dev.c index 62b39d6b1d8f..e09aa3785c15 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -10420,25 +10420,12 @@ err_free_name: } EXPORT_SYMBOL(register_netdevice); -/** - * init_dummy_netdev - init a dummy network device for NAPI - * @dev: device to init - * - * This takes a network device structure and initializes the minimum - * amount of fields so it can be used to schedule NAPI polls without - * registering a full blown interface. This is to be used by drivers - * that need to tie several hardware interfaces to a single NAPI - * poll scheduler due to HW limitations. +/* Initialize the core of a dummy net device. + * This is useful if you are calling this function after alloc_netdev(), + * since it does not memset the net_device fields. */ -void init_dummy_netdev(struct net_device *dev) +static void init_dummy_netdev_core(struct net_device *dev) { - /* Clear everything. Note we don't initialize spinlocks - * as they aren't supposed to be taken by any of the - * NAPI code and this dummy netdev is supposed to be - * only ever used for NAPI polls - */ - memset(dev, 0, sizeof(struct net_device)); - /* make sure we BUG if trying to hit standard * register/unregister code path */ @@ -10459,8 +10446,28 @@ void init_dummy_netdev(struct net_device *dev) * its refcount. */ } -EXPORT_SYMBOL_GPL(init_dummy_netdev); +/** + * init_dummy_netdev - init a dummy network device for NAPI + * @dev: device to init + * + * This takes a network device structure and initializes the minimum + * amount of fields so it can be used to schedule NAPI polls without + * registering a full blown interface. This is to be used by drivers + * that need to tie several hardware interfaces to a single NAPI + * poll scheduler due to HW limitations. + */ +void init_dummy_netdev(struct net_device *dev) +{ + /* Clear everything. Note we don't initialize spinlocks + * as they aren't supposed to be taken by any of the + * NAPI code and this dummy netdev is supposed to be + * only ever used for NAPI polls + */ + memset(dev, 0, sizeof(struct net_device)); + init_dummy_netdev_core(dev); +} +EXPORT_SYMBOL_GPL(init_dummy_netdev); /** * register_netdev - register a network device @@ -11080,6 +11087,19 @@ void free_netdev(struct net_device *dev) } EXPORT_SYMBOL(free_netdev); +/** + * alloc_netdev_dummy - Allocate and initialize a dummy net device. + * @sizeof_priv: size of private data to allocate space for + * + * Return: the allocated net_device on success, NULL otherwise + */ +struct net_device *alloc_netdev_dummy(int sizeof_priv) +{ + return alloc_netdev(sizeof_priv, "dummy#", NET_NAME_UNKNOWN, + init_dummy_netdev_core); +} +EXPORT_SYMBOL_GPL(alloc_netdev_dummy); + /** * synchronize_net - Synchronize with packet receive processing * -- cgit v1.2.3 From ef9226cd56b718c79184a3466d32984a51cb449c Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Thu, 18 Apr 2024 13:36:11 +0200 Subject: page_pool: constify some read-only function arguments There are several functions taking pointers to data they don't modify. This includes statistics fetching, page and page_pool parameters, etc. Constify the pointers, so that call sites will be able to pass const pointers as well. No functional changes, no visible changes in functions sizes. Reviewed-by: Ilias Apalodimas Signed-off-by: Alexander Lobakin Signed-off-by: Tony Nguyen --- include/net/page_pool/helpers.h | 10 +++++----- include/net/page_pool/types.h | 4 ++-- net/core/page_pool.c | 10 +++++----- 3 files changed, 12 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/include/net/page_pool/helpers.h b/include/net/page_pool/helpers.h index 1d397c1a0043..c7bb06750e85 100644 --- a/include/net/page_pool/helpers.h +++ b/include/net/page_pool/helpers.h @@ -58,7 +58,7 @@ /* Deprecated driver-facing API, use netlink instead */ int page_pool_ethtool_stats_get_count(void); u8 *page_pool_ethtool_stats_get_strings(u8 *data); -u64 *page_pool_ethtool_stats_get(u64 *data, void *stats); +u64 *page_pool_ethtool_stats_get(u64 *data, const void *stats); bool page_pool_get_stats(const struct page_pool *pool, struct page_pool_stats *stats); @@ -73,7 +73,7 @@ static inline u8 *page_pool_ethtool_stats_get_strings(u8 *data) return data; } -static inline u64 *page_pool_ethtool_stats_get(u64 *data, void *stats) +static inline u64 *page_pool_ethtool_stats_get(u64 *data, const void *stats) { return data; } @@ -204,8 +204,8 @@ static inline void *page_pool_dev_alloc_va(struct page_pool *pool, * Get the stored dma direction. A driver might decide to store this locally * and avoid the extra cache line from page_pool to determine the direction. */ -static -inline enum dma_data_direction page_pool_get_dma_dir(struct page_pool *pool) +static inline enum dma_data_direction +page_pool_get_dma_dir(const struct page_pool *pool) { return pool->p.dma_dir; } @@ -370,7 +370,7 @@ static inline void page_pool_free_va(struct page_pool *pool, void *va, * Fetch the DMA address of the page. The page pool to which the page belongs * must had been created with PP_FLAG_DMA_MAP. */ -static inline dma_addr_t page_pool_get_dma_addr(struct page *page) +static inline dma_addr_t page_pool_get_dma_addr(const struct page *page) { dma_addr_t ret = page->dma_addr; diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h index 5e43a08d3231..a6ebed002216 100644 --- a/include/net/page_pool/types.h +++ b/include/net/page_pool/types.h @@ -213,7 +213,7 @@ struct xdp_mem_info; #ifdef CONFIG_PAGE_POOL void page_pool_destroy(struct page_pool *pool); void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *), - struct xdp_mem_info *mem); + const struct xdp_mem_info *mem); void page_pool_put_page_bulk(struct page_pool *pool, void **data, int count); #else @@ -223,7 +223,7 @@ static inline void page_pool_destroy(struct page_pool *pool) static inline void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *), - struct xdp_mem_info *mem) + const struct xdp_mem_info *mem) { } diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 4c175091fc0a..273c24429bce 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -123,9 +123,9 @@ int page_pool_ethtool_stats_get_count(void) } EXPORT_SYMBOL(page_pool_ethtool_stats_get_count); -u64 *page_pool_ethtool_stats_get(u64 *data, void *stats) +u64 *page_pool_ethtool_stats_get(u64 *data, const void *stats) { - struct page_pool_stats *pool_stats = stats; + const struct page_pool_stats *pool_stats = stats; *data++ = pool_stats->alloc_stats.fast; *data++ = pool_stats->alloc_stats.slow; @@ -383,8 +383,8 @@ static struct page *__page_pool_get_cached(struct page_pool *pool) return page; } -static void page_pool_dma_sync_for_device(struct page_pool *pool, - struct page *page, +static void page_pool_dma_sync_for_device(const struct page_pool *pool, + const struct page *page, unsigned int dma_sync_size) { dma_addr_t dma_addr = page_pool_get_dma_addr(page); @@ -987,7 +987,7 @@ static void page_pool_release_retry(struct work_struct *wq) } void page_pool_use_xdp_mem(struct page_pool *pool, void (*disconnect)(void *), - struct xdp_mem_info *mem) + const struct xdp_mem_info *mem) { refcount_inc(&pool->user_cnt); pool->disconnect = disconnect; -- cgit v1.2.3 From 66270920f90fadb8f29477eb2b83cfb821bac621 Mon Sep 17 00:00:00 2001 From: Jun Gu Date: Tue, 23 Apr 2024 15:37:51 +0800 Subject: net: openvswitch: Release reference to netdev dev_get_by_name will provide a reference on the netdev. So ensure that the reference of netdev is released after completed. Fixes: 2540088b836f ("net: openvswitch: Check vport netdev name") Signed-off-by: Jun Gu Reviewed-by: Aaron Conole Link: https://lore.kernel.org/r/20240423073751.52706-1-jun.gu@easystack.cn Signed-off-by: Jakub Kicinski --- net/openvswitch/vport-netdev.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 618edc346c0f..91a11067e458 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -78,12 +78,16 @@ struct vport *ovs_netdev_link(struct vport *vport, const char *name) int err; vport->dev = dev_get_by_name(ovs_dp_get_net(vport->dp), name); + if (!vport->dev) { + err = -ENODEV; + goto error_free_vport; + } /* Ensure that the device exists and that the provided * name is not one of its aliases. */ - if (!vport->dev || strcmp(name, ovs_vport_name(vport))) { + if (strcmp(name, ovs_vport_name(vport))) { err = -ENODEV; - goto error_free_vport; + goto error_put; } netdev_tracker_alloc(vport->dev, &vport->dev_tracker, GFP_KERNEL); if (vport->dev->flags & IFF_LOOPBACK || -- cgit v1.2.3 From 2bf90a57f0e682872c5cfb66ffa45e432bb9c7ae Mon Sep 17 00:00:00 2001 From: Philo Lu Date: Sun, 21 Apr 2024 12:20:09 +0800 Subject: tcp: update sacked after tracepoint in __tcp_retransmit_skb Marking TCP_SKB_CB(skb)->sacked with TCPCB_EVER_RETRANS after the traceopint (trace_tcp_retransmit_skb), then we can get the retransmission efficiency by counting skbs w/ and w/o TCPCB_EVER_RETRANS mark in this tracepoint. We have discussed to achieve this with BPF_SOCK_OPS in [0], and using tracepoint is thought to be a better solution. [0] https://lore.kernel.org/all/20240417124622.35333-1-lulie@linux.alibaba.com/ Signed-off-by: Philo Lu Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 99a1d88f7f47..ce59e4499b66 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3418,11 +3418,6 @@ start: err = tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC); } - /* To avoid taking spuriously low RTT samples based on a timestamp - * for a transmit that never happened, always mark EVER_RETRANS - */ - TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS; - if (BPF_SOCK_OPS_TEST_FLAG(tp, BPF_SOCK_OPS_RETRANS_CB_FLAG)) tcp_call_bpf_3arg(sk, BPF_SOCK_OPS_RETRANS_CB, TCP_SKB_CB(skb)->seq, segs, err); @@ -3432,6 +3427,12 @@ start: } else if (err != -EBUSY) { NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPRETRANSFAIL, segs); } + + /* To avoid taking spuriously low RTT samples based on a timestamp + * for a transmit that never happened, always mark EVER_RETRANS + */ + TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS; + return err; } -- cgit v1.2.3 From e7d96e750f2bc9b951290c90df421ac5e9adc3ed Mon Sep 17 00:00:00 2001 From: linke li Date: Tue, 23 Apr 2024 18:53:26 +0800 Subject: net: bridge: remove redundant check of f->dst In br_fill_forward_path(), f->dst is checked not to be NULL, then immediately read using READ_ONCE and checked again. The first check is useless, so this patch aims to remove the redundant check of f->dst. Signed-off-by: linke li Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index c366ccc8b3db..ab4d33e02014 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -389,7 +389,7 @@ static int br_fill_forward_path(struct net_device_path_ctx *ctx, br_vlan_fill_forward_path_pvid(br, ctx, path); f = br_fdb_find_rcu(br, ctx->daddr, path->bridge.vlan_id); - if (!f || !f->dst) + if (!f) return -1; dst = READ_ONCE(f->dst); -- cgit v1.2.3 From ec00ed472bdb7d0af840da68c8c11bff9f4d9caa Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 23 Apr 2024 12:56:20 +0000 Subject: tcp: avoid premature drops in tcp_add_backlog() While testing TCP performance with latest trees, I saw suspect SOCKET_BACKLOG drops. tcp_add_backlog() computes its limit with : limit = (u32)READ_ONCE(sk->sk_rcvbuf) + (u32)(READ_ONCE(sk->sk_sndbuf) >> 1); limit += 64 * 1024; This does not take into account that sk->sk_backlog.len is reset only at the very end of __release_sock(). Both sk->sk_backlog.len and sk->sk_rmem_alloc could reach sk_rcvbuf in normal conditions. We should double sk->sk_rcvbuf contribution in the formula to absorb bubbles in the backlog, which happen more often for very fast flows. This change maintains decent protection against abuses. Fixes: c377411f2494 ("net: sk_add_backlog() take rmem_alloc into account") Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240423125620.3309458-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_ipv4.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 88c83ac42129..e06f0cd04f7e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1995,7 +1995,7 @@ int tcp_v4_early_demux(struct sk_buff *skb) bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb, enum skb_drop_reason *reason) { - u32 limit, tail_gso_size, tail_gso_segs; + u32 tail_gso_size, tail_gso_segs; struct skb_shared_info *shinfo; const struct tcphdr *th; struct tcphdr *thtail; @@ -2004,6 +2004,7 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb, bool fragstolen; u32 gso_segs; u32 gso_size; + u64 limit; int delta; /* In case all data was pulled from skb frags (in __pskb_pull_tail()), @@ -2099,7 +2100,13 @@ bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb, __skb_push(skb, hdrlen); no_coalesce: - limit = (u32)READ_ONCE(sk->sk_rcvbuf) + (u32)(READ_ONCE(sk->sk_sndbuf) >> 1); + /* sk->sk_backlog.len is reset only at the end of __release_sock(). + * Both sk->sk_backlog.len and sk->sk_rmem_alloc could reach + * sk_rcvbuf in normal conditions. + */ + limit = ((u64)READ_ONCE(sk->sk_rcvbuf)) << 1; + + limit += ((u32)READ_ONCE(sk->sk_sndbuf)) >> 1; /* Only socket owner can try to collapse/prune rx queues * to reduce memory overhead, so add a little headroom here. @@ -2107,6 +2114,8 @@ no_coalesce: */ limit += 64 * 1024; + limit = min_t(u64, limit, UINT_MAX); + if (unlikely(sk_add_backlog(sk, skb, limit))) { bh_unlock_sock(sk); *reason = SKB_DROP_REASON_SOCKET_BACKLOG; -- cgit v1.2.3 From 1479eaff1f16983d8fda7c5a08a586c21891087d Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 23 Apr 2024 18:28:17 -0700 Subject: bpf: mark bpf_dummy_struct_ops.test_1 parameter as nullable Test case dummy_st_ops/dummy_init_ret_value passes NULL as the first parameter of the test_1() function. Mark this parameter as nullable to make verifier aware of such possibility. Otherwise, NULL check in the test_1() code: SEC("struct_ops/test_1") int BPF_PROG(test_1, struct bpf_dummy_ops_state *state) { if (!state) return ...; ... access state ... } Might be removed by verifier, thus triggering NULL pointer dereference under certain conditions. Reported-by: Jose E. Marchesi Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20240424012821.595216-2-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- net/bpf/bpf_dummy_struct_ops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c index 25b75844891a..8f413cdfd91a 100644 --- a/net/bpf/bpf_dummy_struct_ops.c +++ b/net/bpf/bpf_dummy_struct_ops.c @@ -232,7 +232,7 @@ static void bpf_dummy_unreg(void *kdata) { } -static int bpf_dummy_test_1(struct bpf_dummy_ops_state *cb) +static int bpf_dummy_ops__test_1(struct bpf_dummy_ops_state *cb__nullable) { return 0; } @@ -249,7 +249,7 @@ static int bpf_dummy_test_sleepable(struct bpf_dummy_ops_state *cb) } static struct bpf_dummy_ops __bpf_bpf_dummy_ops = { - .test_1 = bpf_dummy_test_1, + .test_1 = bpf_dummy_ops__test_1, .test_2 = bpf_dummy_test_2, .test_sleepable = bpf_dummy_test_sleepable, }; -- cgit v1.2.3 From 980ca8ceeae69ddf362870ea9183f389ae26324a Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 23 Apr 2024 18:28:20 -0700 Subject: bpf: check bpf_dummy_struct_ops program params for test runs When doing BPF_PROG_TEST_RUN for bpf_dummy_struct_ops programs, reject execution when NULL is passed for non-nullable params. For programs with non-nullable params verifier assumes that such params are never NULL and thus might optimize out NULL checks. Suggested-by: Kui-Feng Lee Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20240424012821.595216-5-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- net/bpf/bpf_dummy_struct_ops.c | 51 +++++++++++++++++++++++++++++++++++++++++- 1 file changed, 50 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/bpf/bpf_dummy_struct_ops.c b/net/bpf/bpf_dummy_struct_ops.c index 8f413cdfd91a..891cdf61c65a 100644 --- a/net/bpf/bpf_dummy_struct_ops.c +++ b/net/bpf/bpf_dummy_struct_ops.c @@ -79,6 +79,51 @@ static int dummy_ops_call_op(void *image, struct bpf_dummy_ops_test_args *args) args->args[3], args->args[4]); } +static const struct bpf_ctx_arg_aux *find_ctx_arg_info(struct bpf_prog_aux *aux, int offset) +{ + int i; + + for (i = 0; i < aux->ctx_arg_info_size; i++) + if (aux->ctx_arg_info[i].offset == offset) + return &aux->ctx_arg_info[i]; + + return NULL; +} + +/* There is only one check at the moment: + * - zero should not be passed for pointer parameters not marked as nullable. + */ +static int check_test_run_args(struct bpf_prog *prog, struct bpf_dummy_ops_test_args *args) +{ + const struct btf_type *func_proto = prog->aux->attach_func_proto; + + for (u32 arg_no = 0; arg_no < btf_type_vlen(func_proto) ; ++arg_no) { + const struct btf_param *param = &btf_params(func_proto)[arg_no]; + const struct bpf_ctx_arg_aux *info; + const struct btf_type *t; + int offset; + + if (args->args[arg_no] != 0) + continue; + + /* Program is validated already, so there is no need + * to check if t is NULL. + */ + t = btf_type_skip_modifiers(bpf_dummy_ops_btf, param->type, NULL); + if (!btf_type_is_ptr(t)) + continue; + + offset = btf_ctx_arg_offset(bpf_dummy_ops_btf, func_proto, arg_no); + info = find_ctx_arg_info(prog->aux, offset); + if (info && (info->reg_type & PTR_MAYBE_NULL)) + continue; + + return -EINVAL; + } + + return 0; +} + extern const struct bpf_link_ops bpf_struct_ops_link_lops; int bpf_struct_ops_test_run(struct bpf_prog *prog, const union bpf_attr *kattr, @@ -87,7 +132,7 @@ int bpf_struct_ops_test_run(struct bpf_prog *prog, const union bpf_attr *kattr, const struct bpf_struct_ops *st_ops = &bpf_bpf_dummy_ops; const struct btf_type *func_proto; struct bpf_dummy_ops_test_args *args; - struct bpf_tramp_links *tlinks; + struct bpf_tramp_links *tlinks = NULL; struct bpf_tramp_link *link = NULL; void *image = NULL; unsigned int op_idx; @@ -109,6 +154,10 @@ int bpf_struct_ops_test_run(struct bpf_prog *prog, const union bpf_attr *kattr, if (IS_ERR(args)) return PTR_ERR(args); + err = check_test_run_args(prog, args); + if (err) + goto out; + tlinks = kcalloc(BPF_TRAMP_MAX, sizeof(*tlinks), GFP_KERNEL); if (!tlinks) { err = -ENOMEM; -- cgit v1.2.3 From 48e2cd3e3dcfe04f212df4fb189fa04c2a87b980 Mon Sep 17 00:00:00 2001 From: Philo Lu Date: Fri, 26 Apr 2024 00:17:23 +0800 Subject: bpf: add mrtt and srtt as BPF_SOCK_OPS_RTT_CB args Two important arguments in RTT estimation, mrtt and srtt, are passed to tcp_bpf_rtt(), so that bpf programs get more information about RTT computation in BPF_SOCK_OPS_RTT_CB. The difference between bpf_sock_ops->srtt_us and the srtt here is: the former is an old rtt before update, while srtt passed by tcp_bpf_rtt() is that after update. Signed-off-by: Philo Lu Link: https://lore.kernel.org/r/20240425161724.73707-2-lulie@linux.alibaba.com Signed-off-by: Martin KaFai Lau --- include/net/tcp.h | 4 ++-- include/uapi/linux/bpf.h | 2 ++ net/ipv4/tcp_input.c | 4 ++-- tools/include/uapi/linux/bpf.h | 2 ++ 4 files changed, 8 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 6ae35199d3b3..0f75d03287c2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2706,10 +2706,10 @@ static inline bool tcp_bpf_ca_needs_ecn(struct sock *sk) return (tcp_call_bpf(sk, BPF_SOCK_OPS_NEEDS_ECN, 0, NULL) == 1); } -static inline void tcp_bpf_rtt(struct sock *sk) +static inline void tcp_bpf_rtt(struct sock *sk, long mrtt, u32 srtt) { if (BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk), BPF_SOCK_OPS_RTT_CB_FLAG)) - tcp_call_bpf(sk, BPF_SOCK_OPS_RTT_CB, 0, NULL); + tcp_call_bpf_2arg(sk, BPF_SOCK_OPS_RTT_CB, mrtt, srtt); } #if IS_ENABLED(CONFIG_SMC) diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index e4ae83550fb3..d94a72593ead 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -6947,6 +6947,8 @@ enum { * socket transition to LISTEN state. */ BPF_SOCK_OPS_RTT_CB, /* Called on every RTT. + * Arg1: measured RTT input (mrtt) + * Arg2: updated srtt */ BPF_SOCK_OPS_PARSE_HDR_OPT_CB, /* Parse the header option. * It will be called to handle diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5d874817a78d..d1115d7c3936 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -911,7 +911,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us) tp->rtt_seq = tp->snd_nxt; tp->mdev_max_us = tcp_rto_min_us(sk); - tcp_bpf_rtt(sk); + tcp_bpf_rtt(sk, mrtt_us, srtt); } } else { /* no previous measure. */ @@ -921,7 +921,7 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us) tp->mdev_max_us = tp->rttvar_us; tp->rtt_seq = tp->snd_nxt; - tcp_bpf_rtt(sk); + tcp_bpf_rtt(sk, mrtt_us, srtt); } tp->srtt_us = max(1U, srtt); } diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index e4ae83550fb3..d94a72593ead 100644 --- a/tools/include/uapi/linux/bpf.h +++ b/tools/include/uapi/linux/bpf.h @@ -6947,6 +6947,8 @@ enum { * socket transition to LISTEN state. */ BPF_SOCK_OPS_RTT_CB, /* Called on every RTT. + * Arg1: measured RTT input (mrtt) + * Arg2: updated srtt */ BPF_SOCK_OPS_PARSE_HDR_OPT_CB, /* Parse the header option. * It will be called to handle -- cgit v1.2.3 From c4e86b4363ac8ecf836def8e46973707688fd98f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 23 Apr 2024 20:54:08 +0000 Subject: net: add two more call_rcu_hurry() I had failures with pmtu.sh selftests lately, with netns dismantles firing ref_tracking alerts [1]. After much debugging, I found that some queued rcu callbacks were delayed by minutes, because of CONFIG_RCU_LAZY=y option. Joel Fernandes had a similar issue in the past, fixed with commit 483c26ff63f4 ("net: Use call_rcu_hurry() for dst_release()") In this commit, I make sure nexthop_free_rcu() and free_fib_info_rcu() are not delayed too much because they both can release device references. tools/testing/selftests/net/pmtu.sh no longer fails. Traces were: [ 968.179860] ref_tracker: veth_A-R1@00000000d0ff3fe2 has 3/5 users at dst_alloc+0x76/0x160 ip6_dst_alloc+0x25/0x80 ip6_pol_route+0x2a8/0x450 ip6_pol_route_output+0x1f/0x30 fib6_rule_lookup+0x163/0x270 ip6_route_output_flags+0xda/0x190 ip6_dst_lookup_tail.constprop.0+0x1d0/0x260 ip6_dst_lookup_flow+0x47/0xa0 udp_tunnel6_dst_lookup+0x158/0x210 vxlan_xmit_one+0x4c2/0x1550 [vxlan] vxlan_xmit+0x52d/0x14f0 [vxlan] dev_hard_start_xmit+0x7b/0x1e0 __dev_queue_xmit+0x20b/0xe40 ip6_finish_output2+0x2ea/0x6e0 ip6_finish_output+0x143/0x320 ip6_output+0x74/0x140 [ 968.179860] ref_tracker: veth_A-R1@00000000d0ff3fe2 has 1/5 users at netdev_get_by_index+0xc0/0xe0 fib6_nh_init+0x1a9/0xa90 rtm_new_nexthop+0x6fa/0x1580 rtnetlink_rcv_msg+0x155/0x3e0 netlink_rcv_skb+0x61/0x110 rtnetlink_rcv+0x19/0x20 netlink_unicast+0x23f/0x380 netlink_sendmsg+0x1fc/0x430 ____sys_sendmsg+0x2ef/0x320 ___sys_sendmsg+0x86/0xd0 __sys_sendmsg+0x67/0xc0 __x64_sys_sendmsg+0x21/0x30 x64_sys_call+0x252/0x2030 do_syscall_64+0x6c/0x190 entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 968.179860] ref_tracker: veth_A-R1@00000000d0ff3fe2 has 1/5 users at ipv6_add_dev+0x136/0x530 addrconf_notify+0x19d/0x770 notifier_call_chain+0x65/0xd0 raw_notifier_call_chain+0x1a/0x20 call_netdevice_notifiers_info+0x54/0x90 register_netdevice+0x61e/0x790 veth_newlink+0x230/0x440 __rtnl_newlink+0x7d2/0xaa0 rtnl_newlink+0x4c/0x70 rtnetlink_rcv_msg+0x155/0x3e0 netlink_rcv_skb+0x61/0x110 rtnetlink_rcv+0x19/0x20 netlink_unicast+0x23f/0x380 netlink_sendmsg+0x1fc/0x430 ____sys_sendmsg+0x2ef/0x320 ___sys_sendmsg+0x86/0xd0 .... [ 1079.316024] ? show_regs+0x68/0x80 [ 1079.316087] ? __warn+0x8c/0x140 [ 1079.316103] ? ref_tracker_free+0x1a0/0x270 [ 1079.316117] ? report_bug+0x196/0x1c0 [ 1079.316135] ? handle_bug+0x42/0x80 [ 1079.316149] ? exc_invalid_op+0x1c/0x70 [ 1079.316162] ? asm_exc_invalid_op+0x1f/0x30 [ 1079.316193] ? ref_tracker_free+0x1a0/0x270 [ 1079.316208] ? _raw_spin_unlock+0x1a/0x40 [ 1079.316222] ? free_unref_page+0x126/0x1a0 [ 1079.316239] ? destroy_large_folio+0x69/0x90 [ 1079.316251] ? __folio_put+0x99/0xd0 [ 1079.316276] dst_dev_put+0x69/0xd0 [ 1079.316308] fib6_nh_release_dsts.part.0+0x3d/0x80 [ 1079.316327] fib6_nh_release+0x45/0x70 [ 1079.316340] nexthop_free_rcu+0x131/0x170 [ 1079.316356] rcu_do_batch+0x1ee/0x820 [ 1079.316370] ? rcu_do_batch+0x179/0x820 [ 1079.316388] rcu_core+0x1aa/0x4d0 [ 1079.316405] rcu_core_si+0x12/0x20 [ 1079.316417] __do_softirq+0x13a/0x3dc [ 1079.316435] __irq_exit_rcu+0xa3/0x110 [ 1079.316449] irq_exit_rcu+0x12/0x30 [ 1079.316462] sysvec_apic_timer_interrupt+0x5b/0xe0 [ 1079.316474] asm_sysvec_apic_timer_interrupt+0x1f/0x30 [ 1079.316569] RIP: 0033:0x7f06b65c63f0 Signed-off-by: Eric Dumazet Cc: Joel Fernandes (Google) Cc: Paul E. McKenney Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240423205408.39632-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/nexthop.h | 2 +- net/ipv4/fib_semantics.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/nexthop.h b/include/net/nexthop.h index 7ca315ad500e..68463aebcc05 100644 --- a/include/net/nexthop.h +++ b/include/net/nexthop.h @@ -267,7 +267,7 @@ static inline bool nexthop_get(struct nexthop *nh) static inline void nexthop_put(struct nexthop *nh) { if (refcount_dec_and_test(&nh->refcnt)) - call_rcu(&nh->rcu, nexthop_free_rcu); + call_rcu_hurry(&nh->rcu, nexthop_free_rcu); } static inline bool nexthop_cmp(const struct nexthop *nh1, diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 5eb1b8d302bb..f669da98d11d 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -254,7 +254,7 @@ void free_fib_info(struct fib_info *fi) return; } - call_rcu(&fi->rcu, free_fib_info_rcu); + call_rcu_hurry(&fi->rcu, free_fib_info_rcu); } EXPORT_SYMBOL_GPL(free_fib_info); -- cgit v1.2.3 From 2c321f3f70bc284510598f712b702ce8d60c4d14 Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Sun, 14 Apr 2024 19:07:31 -0700 Subject: mm: change inlined allocation helpers to account at the call site Main goal of memory allocation profiling patchset is to provide accounting that is cheap enough to run in production. To achieve that we inject counters using codetags at the allocation call sites to account every time allocation is made. This injection allows us to perform accounting efficiently because injected counters are immediately available as opposed to the alternative methods, such as using _RET_IP_, which would require counter lookup and appropriate locking that makes accounting much more expensive. This method requires all allocation functions to inject separate counters at their call sites so that their callers can be individually accounted. Counter injection is implemented by allocation hooks which should wrap all allocation functions. Inlined functions which perform allocations but do not use allocation hooks are directly charged for the allocations they perform. In most cases these functions are just specialized allocation wrappers used from multiple places to allocate objects of a specific type. It would be more useful to do the accounting at their call sites instead. Instrument these helpers to do accounting at the call site. Simple inlined allocation wrappers are converted directly into macros. More complex allocators or allocators with documentation are converted into _noprof versions and allocation hooks are added. This allows memory allocation profiling mechanism to charge allocations to the callers of these functions. Link: https://lkml.kernel.org/r/20240415020731.1152108-1-surenb@google.com Signed-off-by: Suren Baghdasaryan Acked-by: Jan Kara [jbd2] Cc: Anna Schumaker Cc: Arnd Bergmann Cc: Benjamin Tissoires Cc: Christoph Lameter Cc: David Rientjes Cc: David S. Miller Cc: Dennis Zhou Cc: Eric Dumazet Cc: Herbert Xu Cc: Jakub Kicinski Cc: Jakub Sitnicki Cc: Jiri Kosina Cc: Joerg Roedel Cc: Joonsoo Kim Cc: Kent Overstreet Cc: Matthew Wilcox (Oracle) Cc: Paolo Abeni Cc: Pekka Enberg Cc: Tejun Heo Cc: Theodore Ts'o Cc: Trond Myklebust Cc: Vlastimil Babka Cc: Will Deacon Signed-off-by: Andrew Morton --- drivers/iommu/amd/amd_iommu.h | 5 +++-- fs/nfs/iostat.h | 5 +---- include/acpi/platform/aclinuxex.h | 19 ++++++------------ include/asm-generic/pgalloc.h | 35 ++++++++++++++++++++------------- include/crypto/hash.h | 7 ++++--- include/crypto/internal/acompress.h | 5 +++-- include/crypto/skcipher.h | 7 ++++--- include/linux/bpf.h | 33 ++++++++----------------------- include/linux/bpfptr.h | 5 +++-- include/linux/dma-fence-chain.h | 6 ++---- include/linux/hid_bpf.h | 6 ++---- include/linux/jbd2.h | 12 ++++------- include/linux/mm.h | 5 +++-- include/linux/mm_types.h | 5 +++-- include/linux/percpu.h | 3 +++ include/linux/ptr_ring.h | 28 +++++++++++++++----------- include/linux/skb_array.h | 19 ++++++++++-------- include/linux/skbuff.h | 20 ++++++++----------- include/linux/skmsg.h | 8 +++----- include/linux/slab.h | 5 +++++ include/linux/sockptr.h | 10 ++++++---- include/net/netlabel.h | 16 +++++++++------ include/net/netlink.h | 5 +++-- include/net/request_sock.h | 5 +++-- include/net/tcx.h | 5 +++-- net/sunrpc/auth_gss/auth_gss_internal.h | 6 ++++-- 26 files changed, 142 insertions(+), 143 deletions(-) (limited to 'net') diff --git a/drivers/iommu/amd/amd_iommu.h b/drivers/iommu/amd/amd_iommu.h index f482aab420f7..52575ba9a141 100644 --- a/drivers/iommu/amd/amd_iommu.h +++ b/drivers/iommu/amd/amd_iommu.h @@ -134,13 +134,14 @@ static inline int get_pci_sbdf_id(struct pci_dev *pdev) return PCI_SEG_DEVID_TO_SBDF(seg, devid); } -static inline void *alloc_pgtable_page(int nid, gfp_t gfp) +static inline void *alloc_pgtable_page_noprof(int nid, gfp_t gfp) { struct page *page; - page = alloc_pages_node(nid, gfp | __GFP_ZERO, 0); + page = alloc_pages_node_noprof(nid, gfp | __GFP_ZERO, 0); return page ? page_address(page) : NULL; } +#define alloc_pgtable_page(...) alloc_hooks(alloc_pgtable_page_noprof(__VA_ARGS__)) /* * This must be called after device probe completes. During probe diff --git a/fs/nfs/iostat.h b/fs/nfs/iostat.h index 5aa776b5a3e7..b17a9eb9b148 100644 --- a/fs/nfs/iostat.h +++ b/fs/nfs/iostat.h @@ -46,10 +46,7 @@ static inline void nfs_add_stats(const struct inode *inode, nfs_add_server_stats(NFS_SERVER(inode), stat, addend); } -static inline struct nfs_iostats __percpu *nfs_alloc_iostats(void) -{ - return alloc_percpu(struct nfs_iostats); -} +#define nfs_alloc_iostats() alloc_percpu(struct nfs_iostats) static inline void nfs_free_iostats(struct nfs_iostats __percpu *stats) { diff --git a/include/acpi/platform/aclinuxex.h b/include/acpi/platform/aclinuxex.h index 600d4e2641da..62cac266a1c8 100644 --- a/include/acpi/platform/aclinuxex.h +++ b/include/acpi/platform/aclinuxex.h @@ -47,26 +47,19 @@ acpi_status acpi_os_terminate(void); * However, boot has (system_state != SYSTEM_RUNNING) * to quiet __might_sleep() in kmalloc() and resume does not. */ -static inline void *acpi_os_allocate(acpi_size size) -{ - return kmalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); -} +#define acpi_os_allocate(_size) \ + kmalloc(_size, irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL) -static inline void *acpi_os_allocate_zeroed(acpi_size size) -{ - return kzalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); -} +#define acpi_os_allocate_zeroed(_size) \ + kzalloc(_size, irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL) static inline void acpi_os_free(void *memory) { kfree(memory); } -static inline void *acpi_os_acquire_object(acpi_cache_t * cache) -{ - return kmem_cache_zalloc(cache, - irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); -} +#define acpi_os_acquire_object(_cache) \ + kmem_cache_zalloc(_cache, irqs_disabled() ? GFP_ATOMIC : GFP_KERNEL) static inline acpi_thread_id acpi_os_get_thread_id(void) { diff --git a/include/asm-generic/pgalloc.h b/include/asm-generic/pgalloc.h index 879e5f8aa5e9..7c48f5fbf8aa 100644 --- a/include/asm-generic/pgalloc.h +++ b/include/asm-generic/pgalloc.h @@ -16,15 +16,16 @@ * * Return: pointer to the allocated memory or %NULL on error */ -static inline pte_t *__pte_alloc_one_kernel(struct mm_struct *mm) +static inline pte_t *__pte_alloc_one_kernel_noprof(struct mm_struct *mm) { - struct ptdesc *ptdesc = pagetable_alloc(GFP_PGTABLE_KERNEL & + struct ptdesc *ptdesc = pagetable_alloc_noprof(GFP_PGTABLE_KERNEL & ~__GFP_HIGHMEM, 0); if (!ptdesc) return NULL; return ptdesc_address(ptdesc); } +#define __pte_alloc_one_kernel(...) alloc_hooks(__pte_alloc_one_kernel_noprof(__VA_ARGS__)) #ifndef __HAVE_ARCH_PTE_ALLOC_ONE_KERNEL /** @@ -33,10 +34,11 @@ static inline pte_t *__pte_alloc_one_kernel(struct mm_struct *mm) * * Return: pointer to the allocated memory or %NULL on error */ -static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm) +static inline pte_t *pte_alloc_one_kernel_noprof(struct mm_struct *mm) { - return __pte_alloc_one_kernel(mm); + return __pte_alloc_one_kernel_noprof(mm); } +#define pte_alloc_one_kernel(...) alloc_hooks(pte_alloc_one_kernel_noprof(__VA_ARGS__)) #endif /** @@ -61,11 +63,11 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) * * Return: `struct page` referencing the ptdesc or %NULL on error */ -static inline pgtable_t __pte_alloc_one(struct mm_struct *mm, gfp_t gfp) +static inline pgtable_t __pte_alloc_one_noprof(struct mm_struct *mm, gfp_t gfp) { struct ptdesc *ptdesc; - ptdesc = pagetable_alloc(gfp, 0); + ptdesc = pagetable_alloc_noprof(gfp, 0); if (!ptdesc) return NULL; if (!pagetable_pte_ctor(ptdesc)) { @@ -75,6 +77,7 @@ static inline pgtable_t __pte_alloc_one(struct mm_struct *mm, gfp_t gfp) return ptdesc_page(ptdesc); } +#define __pte_alloc_one(...) alloc_hooks(__pte_alloc_one_noprof(__VA_ARGS__)) #ifndef __HAVE_ARCH_PTE_ALLOC_ONE /** @@ -85,10 +88,11 @@ static inline pgtable_t __pte_alloc_one(struct mm_struct *mm, gfp_t gfp) * * Return: `struct page` referencing the ptdesc or %NULL on error */ -static inline pgtable_t pte_alloc_one(struct mm_struct *mm) +static inline pgtable_t pte_alloc_one_noprof(struct mm_struct *mm) { - return __pte_alloc_one(mm, GFP_PGTABLE_USER); + return __pte_alloc_one_noprof(mm, GFP_PGTABLE_USER); } +#define pte_alloc_one(...) alloc_hooks(pte_alloc_one_noprof(__VA_ARGS__)) #endif /* @@ -124,14 +128,14 @@ static inline void pte_free(struct mm_struct *mm, struct page *pte_page) * * Return: pointer to the allocated memory or %NULL on error */ -static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) +static inline pmd_t *pmd_alloc_one_noprof(struct mm_struct *mm, unsigned long addr) { struct ptdesc *ptdesc; gfp_t gfp = GFP_PGTABLE_USER; if (mm == &init_mm) gfp = GFP_PGTABLE_KERNEL; - ptdesc = pagetable_alloc(gfp, 0); + ptdesc = pagetable_alloc_noprof(gfp, 0); if (!ptdesc) return NULL; if (!pagetable_pmd_ctor(ptdesc)) { @@ -140,6 +144,7 @@ static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) } return ptdesc_address(ptdesc); } +#define pmd_alloc_one(...) alloc_hooks(pmd_alloc_one_noprof(__VA_ARGS__)) #endif #ifndef __HAVE_ARCH_PMD_FREE @@ -157,7 +162,7 @@ static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) #if CONFIG_PGTABLE_LEVELS > 3 -static inline pud_t *__pud_alloc_one(struct mm_struct *mm, unsigned long addr) +static inline pud_t *__pud_alloc_one_noprof(struct mm_struct *mm, unsigned long addr) { gfp_t gfp = GFP_PGTABLE_USER; struct ptdesc *ptdesc; @@ -166,13 +171,14 @@ static inline pud_t *__pud_alloc_one(struct mm_struct *mm, unsigned long addr) gfp = GFP_PGTABLE_KERNEL; gfp &= ~__GFP_HIGHMEM; - ptdesc = pagetable_alloc(gfp, 0); + ptdesc = pagetable_alloc_noprof(gfp, 0); if (!ptdesc) return NULL; pagetable_pud_ctor(ptdesc); return ptdesc_address(ptdesc); } +#define __pud_alloc_one(...) alloc_hooks(__pud_alloc_one_noprof(__VA_ARGS__)) #ifndef __HAVE_ARCH_PUD_ALLOC_ONE /** @@ -184,10 +190,11 @@ static inline pud_t *__pud_alloc_one(struct mm_struct *mm, unsigned long addr) * * Return: pointer to the allocated memory or %NULL on error */ -static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long addr) +static inline pud_t *pud_alloc_one_noprof(struct mm_struct *mm, unsigned long addr) { - return __pud_alloc_one(mm, addr); + return __pud_alloc_one_noprof(mm, addr); } +#define pud_alloc_one(...) alloc_hooks(pud_alloc_one_noprof(__VA_ARGS__)) #endif static inline void __pud_free(struct mm_struct *mm, pud_t *pud) diff --git a/include/crypto/hash.h b/include/crypto/hash.h index 5d61f576cfc8..e5181cc9b7c5 100644 --- a/include/crypto/hash.h +++ b/include/crypto/hash.h @@ -578,19 +578,20 @@ static inline void ahash_request_set_tfm(struct ahash_request *req, * * Return: allocated request handle in case of success, or NULL if out of memory */ -static inline struct ahash_request *ahash_request_alloc( +static inline struct ahash_request *ahash_request_alloc_noprof( struct crypto_ahash *tfm, gfp_t gfp) { struct ahash_request *req; - req = kmalloc(sizeof(struct ahash_request) + - crypto_ahash_reqsize(tfm), gfp); + req = kmalloc_noprof(sizeof(struct ahash_request) + + crypto_ahash_reqsize(tfm), gfp); if (likely(req)) ahash_request_set_tfm(req, tfm); return req; } +#define ahash_request_alloc(...) alloc_hooks(ahash_request_alloc_noprof(__VA_ARGS__)) /** * ahash_request_free() - zeroize and free the request data structure diff --git a/include/crypto/internal/acompress.h b/include/crypto/internal/acompress.h index 4ac46bafba9d..2a67793f52ad 100644 --- a/include/crypto/internal/acompress.h +++ b/include/crypto/internal/acompress.h @@ -69,15 +69,16 @@ static inline void acomp_request_complete(struct acomp_req *req, crypto_request_complete(&req->base, err); } -static inline struct acomp_req *__acomp_request_alloc(struct crypto_acomp *tfm) +static inline struct acomp_req *__acomp_request_alloc_noprof(struct crypto_acomp *tfm) { struct acomp_req *req; - req = kzalloc(sizeof(*req) + crypto_acomp_reqsize(tfm), GFP_KERNEL); + req = kzalloc_noprof(sizeof(*req) + crypto_acomp_reqsize(tfm), GFP_KERNEL); if (likely(req)) acomp_request_set_tfm(req, tfm); return req; } +#define __acomp_request_alloc(...) alloc_hooks(__acomp_request_alloc_noprof(__VA_ARGS__)) static inline void __acomp_request_free(struct acomp_req *req) { diff --git a/include/crypto/skcipher.h b/include/crypto/skcipher.h index c8857d7bdb37..6c5330e316b0 100644 --- a/include/crypto/skcipher.h +++ b/include/crypto/skcipher.h @@ -861,19 +861,20 @@ static inline struct skcipher_request *skcipher_request_cast( * * Return: allocated request handle in case of success, or NULL if out of memory */ -static inline struct skcipher_request *skcipher_request_alloc( +static inline struct skcipher_request *skcipher_request_alloc_noprof( struct crypto_skcipher *tfm, gfp_t gfp) { struct skcipher_request *req; - req = kmalloc(sizeof(struct skcipher_request) + - crypto_skcipher_reqsize(tfm), gfp); + req = kmalloc_noprof(sizeof(struct skcipher_request) + + crypto_skcipher_reqsize(tfm), gfp); if (likely(req)) skcipher_request_set_tfm(req, tfm); return req; } +#define skcipher_request_alloc(...) alloc_hooks(skcipher_request_alloc_noprof(__VA_ARGS__)) /** * skcipher_request_free() - zeroize and free request data structure diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 890e152d553e..a86da1f38b3c 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -2244,31 +2244,14 @@ void *bpf_map_kvcalloc(struct bpf_map *map, size_t n, size_t size, void __percpu *bpf_map_alloc_percpu(const struct bpf_map *map, size_t size, size_t align, gfp_t flags); #else -static inline void * -bpf_map_kmalloc_node(const struct bpf_map *map, size_t size, gfp_t flags, - int node) -{ - return kmalloc_node(size, flags, node); -} - -static inline void * -bpf_map_kzalloc(const struct bpf_map *map, size_t size, gfp_t flags) -{ - return kzalloc(size, flags); -} - -static inline void * -bpf_map_kvcalloc(struct bpf_map *map, size_t n, size_t size, gfp_t flags) -{ - return kvcalloc(n, size, flags); -} - -static inline void __percpu * -bpf_map_alloc_percpu(const struct bpf_map *map, size_t size, size_t align, - gfp_t flags) -{ - return __alloc_percpu_gfp(size, align, flags); -} +#define bpf_map_kmalloc_node(_map, _size, _flags, _node) \ + kmalloc_node(_size, _flags, _node) +#define bpf_map_kzalloc(_map, _size, _flags) \ + kzalloc(_size, _flags) +#define bpf_map_kvcalloc(_map, _n, _size, _flags) \ + kvcalloc(_n, _size, _flags) +#define bpf_map_alloc_percpu(_map, _size, _align, _flags) \ + __alloc_percpu_gfp(_size, _align, _flags) #endif static inline int diff --git a/include/linux/bpfptr.h b/include/linux/bpfptr.h index 79b2f78eec1a..1af241525a17 100644 --- a/include/linux/bpfptr.h +++ b/include/linux/bpfptr.h @@ -65,9 +65,9 @@ static inline int copy_to_bpfptr_offset(bpfptr_t dst, size_t offset, return copy_to_sockptr_offset((sockptr_t) dst, offset, src, size); } -static inline void *kvmemdup_bpfptr(bpfptr_t src, size_t len) +static inline void *kvmemdup_bpfptr_noprof(bpfptr_t src, size_t len) { - void *p = kvmalloc(len, GFP_USER | __GFP_NOWARN); + void *p = kvmalloc_noprof(len, GFP_USER | __GFP_NOWARN); if (!p) return ERR_PTR(-ENOMEM); @@ -77,6 +77,7 @@ static inline void *kvmemdup_bpfptr(bpfptr_t src, size_t len) } return p; } +#define kvmemdup_bpfptr(...) alloc_hooks(kvmemdup_bpfptr_noprof(__VA_ARGS__)) static inline long strncpy_from_bpfptr(char *dst, bpfptr_t src, size_t count) { diff --git a/include/linux/dma-fence-chain.h b/include/linux/dma-fence-chain.h index 4bdf0b96da28..ad9e2506c2f4 100644 --- a/include/linux/dma-fence-chain.h +++ b/include/linux/dma-fence-chain.h @@ -86,10 +86,8 @@ dma_fence_chain_contained(struct dma_fence *fence) * * Returns a new struct dma_fence_chain object or NULL on failure. */ -static inline struct dma_fence_chain *dma_fence_chain_alloc(void) -{ - return kmalloc(sizeof(struct dma_fence_chain), GFP_KERNEL); -}; +#define dma_fence_chain_alloc() \ + ((struct dma_fence_chain *)kmalloc(sizeof(struct dma_fence_chain), GFP_KERNEL)) /** * dma_fence_chain_free diff --git a/include/linux/hid_bpf.h b/include/linux/hid_bpf.h index 7118ac28d468..ca70eeb6393e 100644 --- a/include/linux/hid_bpf.h +++ b/include/linux/hid_bpf.h @@ -149,10 +149,8 @@ static inline int hid_bpf_connect_device(struct hid_device *hdev) { return 0; } static inline void hid_bpf_disconnect_device(struct hid_device *hdev) {} static inline void hid_bpf_destroy_device(struct hid_device *hid) {} static inline void hid_bpf_device_init(struct hid_device *hid) {} -static inline u8 *call_hid_bpf_rdesc_fixup(struct hid_device *hdev, u8 *rdesc, unsigned int *size) -{ - return kmemdup(rdesc, *size, GFP_KERNEL); -} +#define call_hid_bpf_rdesc_fixup(_hdev, _rdesc, _size) \ + ((u8 *)kmemdup(_rdesc, *(_size), GFP_KERNEL)) #endif /* CONFIG_HID_BPF */ diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 971f3e826e15..9512fe332668 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1586,10 +1586,8 @@ void jbd2_journal_put_journal_head(struct journal_head *jh); */ extern struct kmem_cache *jbd2_handle_cache; -static inline handle_t *jbd2_alloc_handle(gfp_t gfp_flags) -{ - return kmem_cache_zalloc(jbd2_handle_cache, gfp_flags); -} +#define jbd2_alloc_handle(_gfp_flags) \ + ((handle_t *)kmem_cache_zalloc(jbd2_handle_cache, _gfp_flags)) static inline void jbd2_free_handle(handle_t *handle) { @@ -1602,10 +1600,8 @@ static inline void jbd2_free_handle(handle_t *handle) */ extern struct kmem_cache *jbd2_inode_cache; -static inline struct jbd2_inode *jbd2_alloc_inode(gfp_t gfp_flags) -{ - return kmem_cache_alloc(jbd2_inode_cache, gfp_flags); -} +#define jbd2_alloc_inode(_gfp_flags) \ + ((struct jbd2_inode *)kmem_cache_alloc(jbd2_inode_cache, _gfp_flags)) static inline void jbd2_free_inode(struct jbd2_inode *jinode) { diff --git a/include/linux/mm.h b/include/linux/mm.h index 0dbd737cfabd..60eabc6c8e00 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2860,12 +2860,13 @@ static inline bool pagetable_is_reserved(struct ptdesc *pt) * * Return: The ptdesc describing the allocated page tables. */ -static inline struct ptdesc *pagetable_alloc(gfp_t gfp, unsigned int order) +static inline struct ptdesc *pagetable_alloc_noprof(gfp_t gfp, unsigned int order) { - struct page *page = alloc_pages(gfp | __GFP_COMP, order); + struct page *page = alloc_pages_noprof(gfp | __GFP_COMP, order); return page_ptdesc(page); } +#define pagetable_alloc(...) alloc_hooks(pagetable_alloc_noprof(__VA_ARGS__)) /** * pagetable_free - Free pagetables diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 4ae4684d1add..6fef56938a17 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -1170,14 +1170,15 @@ static inline void mm_init_cid(struct mm_struct *mm) cpumask_clear(mm_cidmask(mm)); } -static inline int mm_alloc_cid(struct mm_struct *mm) +static inline int mm_alloc_cid_noprof(struct mm_struct *mm) { - mm->pcpu_cid = alloc_percpu(struct mm_cid); + mm->pcpu_cid = alloc_percpu_noprof(struct mm_cid); if (!mm->pcpu_cid) return -ENOMEM; mm_init_cid(mm); return 0; } +#define mm_alloc_cid(...) alloc_hooks(mm_alloc_cid_noprof(__VA_ARGS__)) static inline void mm_destroy_cid(struct mm_struct *mm) { diff --git a/include/linux/percpu.h b/include/linux/percpu.h index 13a82f11e4fd..03053de557cf 100644 --- a/include/linux/percpu.h +++ b/include/linux/percpu.h @@ -151,6 +151,9 @@ extern size_t pcpu_alloc_size(void __percpu *__pdata); #define alloc_percpu(type) \ (typeof(type) __percpu *)__alloc_percpu(sizeof(type), \ __alignof__(type)) +#define alloc_percpu_noprof(type) \ + ((typeof(type) __percpu *)pcpu_alloc_noprof(sizeof(type), \ + __alignof__(type), false, GFP_KERNEL)) extern void free_percpu(void __percpu *__pdata); diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index 808f9d3ee546..fd037c127bb0 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -464,11 +464,11 @@ static inline int ptr_ring_consume_batched_bh(struct ptr_ring *r, /* Not all gfp_t flags (besides GFP_KERNEL) are allowed. See * documentation for vmalloc for which of them are legal. */ -static inline void **__ptr_ring_init_queue_alloc(unsigned int size, gfp_t gfp) +static inline void **__ptr_ring_init_queue_alloc_noprof(unsigned int size, gfp_t gfp) { if (size > KMALLOC_MAX_SIZE / sizeof(void *)) return NULL; - return kvmalloc_array(size, sizeof(void *), gfp | __GFP_ZERO); + return kvmalloc_array_noprof(size, sizeof(void *), gfp | __GFP_ZERO); } static inline void __ptr_ring_set_size(struct ptr_ring *r, int size) @@ -484,9 +484,9 @@ static inline void __ptr_ring_set_size(struct ptr_ring *r, int size) r->batch = 1; } -static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) +static inline int ptr_ring_init_noprof(struct ptr_ring *r, int size, gfp_t gfp) { - r->queue = __ptr_ring_init_queue_alloc(size, gfp); + r->queue = __ptr_ring_init_queue_alloc_noprof(size, gfp); if (!r->queue) return -ENOMEM; @@ -497,6 +497,7 @@ static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) return 0; } +#define ptr_ring_init(...) alloc_hooks(ptr_ring_init_noprof(__VA_ARGS__)) /* * Return entries into ring. Destroy entries that don't fit. @@ -587,11 +588,11 @@ static inline void **__ptr_ring_swap_queue(struct ptr_ring *r, void **queue, * In particular if you consume ring in interrupt or BH context, you must * disable interrupts/BH when doing so. */ -static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, +static inline int ptr_ring_resize_noprof(struct ptr_ring *r, int size, gfp_t gfp, void (*destroy)(void *)) { unsigned long flags; - void **queue = __ptr_ring_init_queue_alloc(size, gfp); + void **queue = __ptr_ring_init_queue_alloc_noprof(size, gfp); void **old; if (!queue) @@ -609,6 +610,7 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, return 0; } +#define ptr_ring_resize(...) alloc_hooks(ptr_ring_resize_noprof(__VA_ARGS__)) /* * Note: producer lock is nested within consumer lock, so if you @@ -616,21 +618,21 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, * In particular if you consume ring in interrupt or BH context, you must * disable interrupts/BH when doing so. */ -static inline int ptr_ring_resize_multiple(struct ptr_ring **rings, - unsigned int nrings, - int size, - gfp_t gfp, void (*destroy)(void *)) +static inline int ptr_ring_resize_multiple_noprof(struct ptr_ring **rings, + unsigned int nrings, + int size, + gfp_t gfp, void (*destroy)(void *)) { unsigned long flags; void ***queues; int i; - queues = kmalloc_array(nrings, sizeof(*queues), gfp); + queues = kmalloc_array_noprof(nrings, sizeof(*queues), gfp); if (!queues) goto noqueues; for (i = 0; i < nrings; ++i) { - queues[i] = __ptr_ring_init_queue_alloc(size, gfp); + queues[i] = __ptr_ring_init_queue_alloc_noprof(size, gfp); if (!queues[i]) goto nomem; } @@ -660,6 +662,8 @@ nomem: noqueues: return -ENOMEM; } +#define ptr_ring_resize_multiple(...) \ + alloc_hooks(ptr_ring_resize_multiple_noprof(__VA_ARGS__)) static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *)) { diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index e2d45b7cb619..926496c9cc9c 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -177,10 +177,11 @@ static inline int skb_array_peek_len_any(struct skb_array *a) return PTR_RING_PEEK_CALL_ANY(&a->ring, __skb_array_len_with_tag); } -static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) +static inline int skb_array_init_noprof(struct skb_array *a, int size, gfp_t gfp) { - return ptr_ring_init(&a->ring, size, gfp); + return ptr_ring_init_noprof(&a->ring, size, gfp); } +#define skb_array_init(...) alloc_hooks(skb_array_init_noprof(__VA_ARGS__)) static void __skb_array_destroy_skb(void *ptr) { @@ -198,15 +199,17 @@ static inline int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); } -static inline int skb_array_resize_multiple(struct skb_array **rings, - int nrings, unsigned int size, - gfp_t gfp) +static inline int skb_array_resize_multiple_noprof(struct skb_array **rings, + int nrings, unsigned int size, + gfp_t gfp) { BUILD_BUG_ON(offsetof(struct skb_array, ring)); - return ptr_ring_resize_multiple((struct ptr_ring **)rings, - nrings, size, gfp, - __skb_array_destroy_skb); + return ptr_ring_resize_multiple_noprof((struct ptr_ring **)rings, + nrings, size, gfp, + __skb_array_destroy_skb); } +#define skb_array_resize_multiple(...) \ + alloc_hooks(skb_array_resize_multiple_noprof(__VA_ARGS__)) static inline void skb_array_cleanup(struct skb_array *a) { diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9d24aec064e8..b72920c9887b 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3371,7 +3371,7 @@ void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason); * * %NULL is returned if there is no free memory. */ -static inline struct page *__dev_alloc_pages(gfp_t gfp_mask, +static inline struct page *__dev_alloc_pages_noprof(gfp_t gfp_mask, unsigned int order) { /* This piece of code contains several assumptions. @@ -3384,13 +3384,11 @@ static inline struct page *__dev_alloc_pages(gfp_t gfp_mask, */ gfp_mask |= __GFP_COMP | __GFP_MEMALLOC; - return alloc_pages_node(NUMA_NO_NODE, gfp_mask, order); + return alloc_pages_node_noprof(NUMA_NO_NODE, gfp_mask, order); } +#define __dev_alloc_pages(...) alloc_hooks(__dev_alloc_pages_noprof(__VA_ARGS__)) -static inline struct page *dev_alloc_pages(unsigned int order) -{ - return __dev_alloc_pages(GFP_ATOMIC | __GFP_NOWARN, order); -} +#define dev_alloc_pages(_order) __dev_alloc_pages(GFP_ATOMIC | __GFP_NOWARN, _order) /** * __dev_alloc_page - allocate a page for network Rx @@ -3400,15 +3398,13 @@ static inline struct page *dev_alloc_pages(unsigned int order) * * %NULL is returned if there is no free memory. */ -static inline struct page *__dev_alloc_page(gfp_t gfp_mask) +static inline struct page *__dev_alloc_page_noprof(gfp_t gfp_mask) { - return __dev_alloc_pages(gfp_mask, 0); + return __dev_alloc_pages_noprof(gfp_mask, 0); } +#define __dev_alloc_page(...) alloc_hooks(__dev_alloc_page_noprof(__VA_ARGS__)) -static inline struct page *dev_alloc_page(void) -{ - return dev_alloc_pages(0); -} +#define dev_alloc_page() dev_alloc_pages(0) /** * dev_page_is_reusable - check whether a page can be reused for network Rx diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index e65ec3fd2799..78efc5b20284 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -410,11 +410,9 @@ void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock); int sk_psock_msg_verdict(struct sock *sk, struct sk_psock *psock, struct sk_msg *msg); -static inline struct sk_psock_link *sk_psock_init_link(void) -{ - return kzalloc(sizeof(struct sk_psock_link), - GFP_ATOMIC | __GFP_NOWARN); -} +#define sk_psock_init_link() \ + ((struct sk_psock_link *)kzalloc(sizeof(struct sk_psock_link), \ + GFP_ATOMIC | __GFP_NOWARN)) static inline void sk_psock_free_link(struct sk_psock_link *link) { diff --git a/include/linux/slab.h b/include/linux/slab.h index c0be1cd03cf6..4cc37ef22aae 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -744,6 +744,9 @@ void *kmalloc_node_track_caller_noprof(size_t size, gfp_t flags, int node, */ #define kmalloc_track_caller(...) kmalloc_node_track_caller(__VA_ARGS__, NUMA_NO_NODE) +#define kmalloc_track_caller_noprof(...) \ + kmalloc_node_track_caller_noprof(__VA_ARGS__, NUMA_NO_NODE, _RET_IP_) + static inline __alloc_size(1, 2) void *kmalloc_array_node_noprof(size_t n, size_t size, gfp_t flags, int node) { @@ -781,6 +784,7 @@ extern void *kvmalloc_node_noprof(size_t size, gfp_t flags, int node) __alloc_si #define kvmalloc_node(...) alloc_hooks(kvmalloc_node_noprof(__VA_ARGS__)) #define kvmalloc(_size, _flags) kvmalloc_node(_size, _flags, NUMA_NO_NODE) +#define kvmalloc_noprof(_size, _flags) kvmalloc_node_noprof(_size, _flags, NUMA_NO_NODE) #define kvzalloc(_size, _flags) kvmalloc(_size, _flags|__GFP_ZERO) #define kvzalloc_node(_size, _flags, _node) kvmalloc_node(_size, _flags|__GFP_ZERO, _node) @@ -797,6 +801,7 @@ static inline __alloc_size(1, 2) void *kvmalloc_array_noprof(size_t n, size_t si #define kvmalloc_array(...) alloc_hooks(kvmalloc_array_noprof(__VA_ARGS__)) #define kvcalloc(_n, _size, _flags) kvmalloc_array(_n, _size, _flags|__GFP_ZERO) +#define kvcalloc_noprof(_n, _size, _flags) kvmalloc_array_noprof(_n, _size, _flags|__GFP_ZERO) extern void *kvrealloc_noprof(const void *p, size_t oldsize, size_t newsize, gfp_t flags) __realloc_size(3); diff --git a/include/linux/sockptr.h b/include/linux/sockptr.h index 317200cd3a60..fc5a206c4043 100644 --- a/include/linux/sockptr.h +++ b/include/linux/sockptr.h @@ -117,9 +117,9 @@ static inline int copy_to_sockptr(sockptr_t dst, const void *src, size_t size) return copy_to_sockptr_offset(dst, 0, src, size); } -static inline void *memdup_sockptr(sockptr_t src, size_t len) +static inline void *memdup_sockptr_noprof(sockptr_t src, size_t len) { - void *p = kmalloc_track_caller(len, GFP_USER | __GFP_NOWARN); + void *p = kmalloc_track_caller_noprof(len, GFP_USER | __GFP_NOWARN); if (!p) return ERR_PTR(-ENOMEM); @@ -129,10 +129,11 @@ static inline void *memdup_sockptr(sockptr_t src, size_t len) } return p; } +#define memdup_sockptr(...) alloc_hooks(memdup_sockptr_noprof(__VA_ARGS__)) -static inline void *memdup_sockptr_nul(sockptr_t src, size_t len) +static inline void *memdup_sockptr_nul_noprof(sockptr_t src, size_t len) { - char *p = kmalloc_track_caller(len + 1, GFP_KERNEL); + char *p = kmalloc_track_caller_noprof(len + 1, GFP_KERNEL); if (!p) return ERR_PTR(-ENOMEM); @@ -143,6 +144,7 @@ static inline void *memdup_sockptr_nul(sockptr_t src, size_t len) p[len] = '\0'; return p; } +#define memdup_sockptr_nul(...) alloc_hooks(memdup_sockptr_nul_noprof(__VA_ARGS__)) static inline long strncpy_from_sockptr(char *dst, sockptr_t src, size_t count) { diff --git a/include/net/netlabel.h b/include/net/netlabel.h index f3ab0b8a4b18..c31bd96dafdb 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -274,15 +274,17 @@ struct netlbl_calipso_ops { * on success, NULL on failure. * */ -static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc(gfp_t flags) +static inline struct netlbl_lsm_cache *netlbl_secattr_cache_alloc_noprof(gfp_t flags) { struct netlbl_lsm_cache *cache; - cache = kzalloc(sizeof(*cache), flags); + cache = kzalloc_noprof(sizeof(*cache), flags); if (cache) refcount_set(&cache->refcount, 1); return cache; } +#define netlbl_secattr_cache_alloc(...) \ + alloc_hooks(netlbl_secattr_cache_alloc_noprof(__VA_ARGS__)) /** * netlbl_secattr_cache_free - Frees a netlbl_lsm_cache struct @@ -311,10 +313,11 @@ static inline void netlbl_secattr_cache_free(struct netlbl_lsm_cache *cache) * on failure. * */ -static inline struct netlbl_lsm_catmap *netlbl_catmap_alloc(gfp_t flags) +static inline struct netlbl_lsm_catmap *netlbl_catmap_alloc_noprof(gfp_t flags) { - return kzalloc(sizeof(struct netlbl_lsm_catmap), flags); + return kzalloc_noprof(sizeof(struct netlbl_lsm_catmap), flags); } +#define netlbl_catmap_alloc(...) alloc_hooks(netlbl_catmap_alloc_noprof(__VA_ARGS__)) /** * netlbl_catmap_free - Free a LSM secattr catmap @@ -376,10 +379,11 @@ static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr) * pointer on success, or NULL on failure. * */ -static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc(gfp_t flags) +static inline struct netlbl_lsm_secattr *netlbl_secattr_alloc_noprof(gfp_t flags) { - return kzalloc(sizeof(struct netlbl_lsm_secattr), flags); + return kzalloc_noprof(sizeof(struct netlbl_lsm_secattr), flags); } +#define netlbl_secattr_alloc(...) alloc_hooks(netlbl_secattr_alloc_noprof(__VA_ARGS__)) /** * netlbl_secattr_free - Frees a netlbl_lsm_secattr struct diff --git a/include/net/netlink.h b/include/net/netlink.h index c19ff921b661..972b5484fa6f 100644 --- a/include/net/netlink.h +++ b/include/net/netlink.h @@ -1891,10 +1891,11 @@ static inline struct nla_bitfield32 nla_get_bitfield32(const struct nlattr *nla) * @src: netlink attribute to duplicate from * @gfp: GFP mask */ -static inline void *nla_memdup(const struct nlattr *src, gfp_t gfp) +static inline void *nla_memdup_noprof(const struct nlattr *src, gfp_t gfp) { - return kmemdup(nla_data(src), nla_len(src), gfp); + return kmemdup_noprof(nla_data(src), nla_len(src), gfp); } +#define nla_memdup(...) alloc_hooks(nla_memdup_noprof(__VA_ARGS__)) /** * nla_nest_start_noflag - Start a new level of nested attributes diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 004e651e6067..29495c331d20 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -127,12 +127,12 @@ static inline struct sock *skb_steal_sock(struct sk_buff *skb, } static inline struct request_sock * -reqsk_alloc(const struct request_sock_ops *ops, struct sock *sk_listener, +reqsk_alloc_noprof(const struct request_sock_ops *ops, struct sock *sk_listener, bool attach_listener) { struct request_sock *req; - req = kmem_cache_alloc(ops->slab, GFP_ATOMIC | __GFP_NOWARN); + req = kmem_cache_alloc_noprof(ops->slab, GFP_ATOMIC | __GFP_NOWARN); if (!req) return NULL; req->rsk_listener = NULL; @@ -157,6 +157,7 @@ reqsk_alloc(const struct request_sock_ops *ops, struct sock *sk_listener, return req; } +#define reqsk_alloc(...) alloc_hooks(reqsk_alloc_noprof(__VA_ARGS__)) static inline void __reqsk_free(struct request_sock *req) { diff --git a/include/net/tcx.h b/include/net/tcx.h index 04be9377785d..72a3e75e539f 100644 --- a/include/net/tcx.h +++ b/include/net/tcx.h @@ -75,9 +75,9 @@ tcx_entry_fetch(struct net_device *dev, bool ingress) return rcu_dereference_rtnl(dev->tcx_egress); } -static inline struct bpf_mprog_entry *tcx_entry_create(void) +static inline struct bpf_mprog_entry *tcx_entry_create_noprof(void) { - struct tcx_entry *tcx = kzalloc(sizeof(*tcx), GFP_KERNEL); + struct tcx_entry *tcx = kzalloc_noprof(sizeof(*tcx), GFP_KERNEL); if (tcx) { bpf_mprog_bundle_init(&tcx->bundle); @@ -85,6 +85,7 @@ static inline struct bpf_mprog_entry *tcx_entry_create(void) } return NULL; } +#define tcx_entry_create(...) alloc_hooks(tcx_entry_create_noprof(__VA_ARGS__)) static inline void tcx_entry_free(struct bpf_mprog_entry *entry) { diff --git a/net/sunrpc/auth_gss/auth_gss_internal.h b/net/sunrpc/auth_gss/auth_gss_internal.h index c53b329092d4..4ebc1b7043d9 100644 --- a/net/sunrpc/auth_gss/auth_gss_internal.h +++ b/net/sunrpc/auth_gss/auth_gss_internal.h @@ -23,7 +23,7 @@ simple_get_bytes(const void *p, const void *end, void *res, size_t len) } static inline const void * -simple_get_netobj(const void *p, const void *end, struct xdr_netobj *dest) +simple_get_netobj_noprof(const void *p, const void *end, struct xdr_netobj *dest) { const void *q; unsigned int len; @@ -35,7 +35,7 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *dest) if (unlikely(q > end || q < p)) return ERR_PTR(-EFAULT); if (len) { - dest->data = kmemdup(p, len, GFP_KERNEL); + dest->data = kmemdup_noprof(p, len, GFP_KERNEL); if (unlikely(dest->data == NULL)) return ERR_PTR(-ENOMEM); } else @@ -43,3 +43,5 @@ simple_get_netobj(const void *p, const void *end, struct xdr_netobj *dest) dest->len = len; return q; } + +#define simple_get_netobj(...) alloc_hooks(simple_get_netobj_noprof(__VA_ARGS__)) -- cgit v1.2.3 From 58fbfecab965014b6e3cc956a76b4a96265a1add Mon Sep 17 00:00:00 2001 From: Paul Davey Date: Tue, 23 Apr 2024 18:00:24 +1200 Subject: xfrm: Preserve vlan tags for transport mode software GRO The software GRO path for esp transport mode uses skb_mac_header_rebuild prior to re-injecting the packet via the xfrm_napi_dev. This only copies skb->mac_len bytes of header which may not be sufficient if the packet contains 802.1Q tags or other VLAN tags. Worse copying only the initial header will leave a packet marked as being VLAN tagged but without the corresponding tag leading to mangling when it is later untagged. The VLAN tags are important when receiving the decrypted esp transport mode packet after GRO processing to ensure it is received on the correct interface. Therefore record the full mac header length in xfrm*_transport_input for later use in corresponding xfrm*_transport_finish to copy the entire mac header when rebuilding the mac header for GRO. The skb->data pointer is left pointing skb->mac_header bytes after the start of the mac header as is expected by the network stack and network and transport header offsets reset to this location. Fixes: 7785bba299a8 ("esp: Add a software GRO codepath") Signed-off-by: Paul Davey Signed-off-by: Steffen Klassert --- include/linux/skbuff.h | 15 +++++++++++++++ include/net/xfrm.h | 3 +++ net/ipv4/xfrm4_input.c | 6 +++++- net/ipv6/xfrm6_input.c | 6 +++++- net/xfrm/xfrm_input.c | 8 ++++++++ 5 files changed, 36 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9d24aec064e8..4ff48eda3f64 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3031,6 +3031,21 @@ static inline void skb_mac_header_rebuild(struct sk_buff *skb) } } +/* Move the full mac header up to current network_header. + * Leaves skb->data pointing at offset skb->mac_len into the mac_header. + * Must be provided the complete mac header length. + */ +static inline void skb_mac_header_rebuild_full(struct sk_buff *skb, u32 full_mac_len) +{ + if (skb_mac_header_was_set(skb)) { + const unsigned char *old_mac = skb_mac_header(skb); + + skb_set_mac_header(skb, -full_mac_len); + memmove(skb_mac_header(skb), old_mac, full_mac_len); + __skb_push(skb, full_mac_len - skb->mac_len); + } +} + static inline int skb_checksum_start_offset(const struct sk_buff *skb) { return skb->csum_start - skb_headroom(skb); diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 57c743b7e4fe..cb4841a9fffd 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1049,6 +1049,9 @@ struct xfrm_offload { #define CRYPTO_INVALID_PACKET_SYNTAX 64 #define CRYPTO_INVALID_PROTOCOL 128 + /* Used to keep whole l2 header for transport mode GRO */ + __u32 orig_mac_len; + __u8 proto; __u8 inner_ipproto; }; diff --git a/net/ipv4/xfrm4_input.c b/net/ipv4/xfrm4_input.c index dae35101d189..86382e08140e 100644 --- a/net/ipv4/xfrm4_input.c +++ b/net/ipv4/xfrm4_input.c @@ -63,7 +63,11 @@ int xfrm4_transport_finish(struct sk_buff *skb, int async) ip_send_check(iph); if (xo && (xo->flags & XFRM_GRO)) { - skb_mac_header_rebuild(skb); + /* The full l2 header needs to be preserved so that re-injecting the packet at l2 + * works correctly in the presence of vlan tags. + */ + skb_mac_header_rebuild_full(skb, xo->orig_mac_len); + skb_reset_network_header(skb); skb_reset_transport_header(skb); return 0; } diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index a17d783dc7c0..c6b8e132e10a 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -58,7 +58,11 @@ int xfrm6_transport_finish(struct sk_buff *skb, int async) skb_postpush_rcsum(skb, skb_network_header(skb), nhlen); if (xo && (xo->flags & XFRM_GRO)) { - skb_mac_header_rebuild(skb); + /* The full l2 header needs to be preserved so that re-injecting the packet at l2 + * works correctly in the presence of vlan tags. + */ + skb_mac_header_rebuild_full(skb, xo->orig_mac_len); + skb_reset_network_header(skb); skb_reset_transport_header(skb); return 0; } diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 161f535c8b94..3a2982a72a6b 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -389,11 +389,15 @@ static int xfrm_prepare_input(struct xfrm_state *x, struct sk_buff *skb) */ static int xfrm4_transport_input(struct xfrm_state *x, struct sk_buff *skb) { + struct xfrm_offload *xo = xfrm_offload(skb); int ihl = skb->data - skb_transport_header(skb); if (skb->transport_header != skb->network_header) { memmove(skb_transport_header(skb), skb_network_header(skb), ihl); + if (xo) + xo->orig_mac_len = + skb_mac_header_was_set(skb) ? skb_mac_header_len(skb) : 0; skb->network_header = skb->transport_header; } ip_hdr(skb)->tot_len = htons(skb->len + ihl); @@ -404,11 +408,15 @@ static int xfrm4_transport_input(struct xfrm_state *x, struct sk_buff *skb) static int xfrm6_transport_input(struct xfrm_state *x, struct sk_buff *skb) { #if IS_ENABLED(CONFIG_IPV6) + struct xfrm_offload *xo = xfrm_offload(skb); int ihl = skb->data - skb_transport_header(skb); if (skb->transport_header != skb->network_header) { memmove(skb_transport_header(skb), skb_network_header(skb), ihl); + if (xo) + xo->orig_mac_len = + skb_mac_header_was_set(skb) ? skb_mac_header_len(skb) : 0; skb->network_header = skb->transport_header; } ipv6_hdr(skb)->payload_len = htons(skb->len + ihl - -- cgit v1.2.3 From af0cb3fa3f9ed258d14abab0152e28a0f9593084 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Thu, 18 Apr 2024 15:50:11 +0200 Subject: net/sched: fix false lockdep warning on qdisc root lock Xiumei and Christoph reported the following lockdep splat, complaining of the qdisc root lock being taken twice: ============================================ WARNING: possible recursive locking detected 6.7.0-rc3+ #598 Not tainted -------------------------------------------- swapper/2/0 is trying to acquire lock: ffff888177190110 (&sch->q.lock){+.-.}-{2:2}, at: __dev_queue_xmit+0x1560/0x2e70 but task is already holding lock: ffff88811995a110 (&sch->q.lock){+.-.}-{2:2}, at: __dev_queue_xmit+0x1560/0x2e70 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&sch->q.lock); lock(&sch->q.lock); *** DEADLOCK *** May be due to missing lock nesting notation 5 locks held by swapper/2/0: #0: ffff888135a09d98 ((&in_dev->mr_ifc_timer)){+.-.}-{0:0}, at: call_timer_fn+0x11a/0x510 #1: ffffffffaaee5260 (rcu_read_lock){....}-{1:2}, at: ip_finish_output2+0x2c0/0x1ed0 #2: ffffffffaaee5200 (rcu_read_lock_bh){....}-{1:2}, at: __dev_queue_xmit+0x209/0x2e70 #3: ffff88811995a110 (&sch->q.lock){+.-.}-{2:2}, at: __dev_queue_xmit+0x1560/0x2e70 #4: ffffffffaaee5200 (rcu_read_lock_bh){....}-{1:2}, at: __dev_queue_xmit+0x209/0x2e70 stack backtrace: CPU: 2 PID: 0 Comm: swapper/2 Not tainted 6.7.0-rc3+ #598 Hardware name: Red Hat KVM, BIOS 1.13.0-2.module+el8.3.0+7353+9de0a3cc 04/01/2014 Call Trace: dump_stack_lvl+0x4a/0x80 __lock_acquire+0xfdd/0x3150 lock_acquire+0x1ca/0x540 _raw_spin_lock+0x34/0x80 __dev_queue_xmit+0x1560/0x2e70 tcf_mirred_act+0x82e/0x1260 [act_mirred] tcf_action_exec+0x161/0x480 tcf_classify+0x689/0x1170 prio_enqueue+0x316/0x660 [sch_prio] dev_qdisc_enqueue+0x46/0x220 __dev_queue_xmit+0x1615/0x2e70 ip_finish_output2+0x1218/0x1ed0 __ip_finish_output+0x8b3/0x1350 ip_output+0x163/0x4e0 igmp_ifc_timer_expire+0x44b/0x930 call_timer_fn+0x1a2/0x510 run_timer_softirq+0x54d/0x11a0 __do_softirq+0x1b3/0x88f irq_exit_rcu+0x18f/0x1e0 sysvec_apic_timer_interrupt+0x6f/0x90 This happens when TC does a mirred egress redirect from the root qdisc of device A to the root qdisc of device B. As long as these two locks aren't protecting the same qdisc, they can be acquired in chain: add a per-qdisc lockdep key to silence false warnings. This dynamic key should safely replace the static key we have in sch_htb: it was added to allow enqueueing to the device "direct qdisc" while still holding the qdisc root lock. v2: don't use static keys anymore in HTB direct qdiscs (thanks Eric Dumazet) CC: Maxim Mikityanskiy CC: Xiumei Mu Reported-by: Christoph Paasch Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/451 Signed-off-by: Davide Caratti Link: https://lore.kernel.org/r/7dc06d6158f72053cf877a82e2a7a5bd23692faa.1713448007.git.dcaratti@redhat.com Signed-off-by: Paolo Abeni --- include/net/sch_generic.h | 1 + net/sched/sch_generic.c | 3 +++ net/sched/sch_htb.c | 22 +++------------------- 3 files changed, 7 insertions(+), 19 deletions(-) (limited to 'net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index f561dfb79743..79edd5b5e3c9 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -128,6 +128,7 @@ struct Qdisc { struct rcu_head rcu; netdevice_tracker dev_tracker; + struct lock_class_key root_lock_key; /* private data */ long privdata[] ____cacheline_aligned; }; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 4a2c763e2d11..31dfd6c7405b 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -945,7 +945,9 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, __skb_queue_head_init(&sch->gso_skb); __skb_queue_head_init(&sch->skb_bad_txq); gnet_stats_basic_sync_init(&sch->bstats); + lockdep_register_key(&sch->root_lock_key); spin_lock_init(&sch->q.lock); + lockdep_set_class(&sch->q.lock, &sch->root_lock_key); if (ops->static_flags & TCQ_F_CPUSTATS) { sch->cpu_bstats = @@ -1068,6 +1070,7 @@ static void __qdisc_destroy(struct Qdisc *qdisc) if (ops->destroy) ops->destroy(qdisc); + lockdep_unregister_key(&qdisc->root_lock_key); module_put(ops->owner); netdev_put(dev, &qdisc->dev_tracker); diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 93e6fb56f3b5..ff3de37874e4 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1039,13 +1039,6 @@ static void htb_work_func(struct work_struct *work) rcu_read_unlock(); } -static void htb_set_lockdep_class_child(struct Qdisc *q) -{ - static struct lock_class_key child_key; - - lockdep_set_class(qdisc_lock(q), &child_key); -} - static int htb_offload(struct net_device *dev, struct tc_htb_qopt_offload *opt) { return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_HTB, opt); @@ -1132,7 +1125,6 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt, return -ENOMEM; } - htb_set_lockdep_class_child(qdisc); q->direct_qdiscs[ntx] = qdisc; qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; } @@ -1468,7 +1460,6 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, } if (q->offload) { - htb_set_lockdep_class_child(new); /* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ qdisc_refcount_inc(new); old_q = htb_graft_helper(dev_queue, new); @@ -1733,11 +1724,8 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg, new_q = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, cl->parent->common.classid, NULL); - if (q->offload) { - if (new_q) - htb_set_lockdep_class_child(new_q); + if (q->offload) htb_parent_to_leaf_offload(sch, dev_queue, new_q); - } } sch_tree_lock(sch); @@ -1947,13 +1935,9 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, new_q = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, classid, NULL); if (q->offload) { - if (new_q) { - htb_set_lockdep_class_child(new_q); - /* One ref for cl->leaf.q, the other for - * dev_queue->qdisc. - */ + /* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ + if (new_q) qdisc_refcount_inc(new_q); - } old_q = htb_graft_helper(dev_queue, new_q); /* No qdisc_put needed. */ WARN_ON(!(old_q->flags & TCQ_F_BUILTIN)); -- cgit v1.2.3 From 5055cccfc2d1cc1a7306f6bcdcd0ee9521d707f5 Mon Sep 17 00:00:00 2001 From: Lukasz Majewski Date: Tue, 23 Apr 2024 14:49:04 +0200 Subject: net: hsr: Provide RedBox support (HSR-SAN) Introduce RedBox support (HSR-SAN to be more precise) for HSR networks. Following traffic reduction optimizations have been implemented: - Do not send HSR supervisory frames to Port C (interlink) - Do not forward to HSR ring frames addressed to Port C - Do not forward to Port C frames from HSR ring - Do not send duplicate HSR frame to HSR ring when destination is Port C The corresponding patch to modify iptable2 sources has already been sent: https://lore.kernel.org/netdev/20240308145729.490863-1-lukma@denx.de/T/ Testing procedure (veth and netns): ----------------------------------- One shall run: linux-vanila/tools/testing/selftests/net/hsr/hsr_redbox.sh (Detailed description of the setup one can find in the test script file). Testing procedure (real hardware): ---------------------------------- The EVB-KSZ9477 has been used for testing on net-next branch (SHA1: 5fc68320c1fb3c7d456ddcae0b4757326a043e6f). Ports 4/5 were used for SW managed HSR (hsr1) as first hsr0 for ports 1/2 (with HW offloading for ksz9477) was created. Port 3 has been used as interlink port (single USB-ETH dongle). Configuration - RedBox (EVB-KSZ9477): if link set lan1 down;ip link set lan2 down ip link add name hsr0 type hsr slave1 lan1 slave2 lan2 supervision 45 version 1 ip link add name hsr1 type hsr slave1 lan4 slave2 lan5 interlink lan3 supervision 45 version 1 ip link set lan4 up;ip link set lan5 up ip link set lan3 up ip addr add 192.168.0.11/24 dev hsr1 ip link set hsr1 up Configuration - DAN-H (EVB-KSZ9477): ip link set lan1 down;ip link set lan2 down ip link add name hsr0 type hsr slave1 lan1 slave2 lan2 supervision 45 version 1 ip link add name hsr1 type hsr slave1 lan4 slave2 lan5 supervision 45 version 1 ip link set lan4 up;ip link set lan5 up ip addr add 192.168.0.12/24 dev hsr1 ip link set hsr1 up This approach uses only SW based HSR devices (hsr1). -------------- ----------------- ------------ DAN-H Port5 | <------> | Port5 | | Port4 | <------> | Port4 Port3 | <---> | PC | | (RedBox) | | (USB-ETH) EVB-KSZ9477 | | EVB-KSZ9477 | | -------------- ----------------- ------------ Signed-off-by: Lukasz Majewski Signed-off-by: Paolo Abeni --- include/uapi/linux/if_link.h | 1 + net/hsr/hsr_device.c | 36 +++++++++++++++++-- net/hsr/hsr_device.h | 4 +-- net/hsr/hsr_forward.c | 85 +++++++++++++++++++++++++++++++++++++++----- net/hsr/hsr_framereg.c | 52 +++++++++++++++++++++++++++ net/hsr/hsr_framereg.h | 4 +++ net/hsr/hsr_main.h | 7 ++++ net/hsr/hsr_netlink.c | 30 ++++++++++++++-- net/hsr/hsr_slave.c | 1 + 9 files changed, 204 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index ffa637b38c93..e9f10860ec8e 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -1771,6 +1771,7 @@ enum { IFLA_HSR_PROTOCOL, /* Indicate different protocol than * HSR. For example PRP. */ + IFLA_HSR_INTERLINK, /* HSR interlink network device */ __IFLA_HSR_MAX, }; diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index e9d45133d641..cd1e7c6d2fc0 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -146,6 +146,9 @@ static int hsr_dev_open(struct net_device *dev) case HSR_PT_SLAVE_B: designation = "Slave B"; break; + case HSR_PT_INTERLINK: + designation = "Interlink"; + break; default: designation = "Unknown"; } @@ -285,6 +288,7 @@ static void send_hsr_supervision_frame(struct hsr_port *master, struct hsr_priv *hsr = master->hsr; __u8 type = HSR_TLV_LIFE_CHECK; struct hsr_sup_payload *hsr_sp; + struct hsr_sup_tlv *hsr_stlv; struct hsr_sup_tag *hsr_stag; struct sk_buff *skb; @@ -324,6 +328,16 @@ static void send_hsr_supervision_frame(struct hsr_port *master, hsr_sp = skb_put(skb, sizeof(struct hsr_sup_payload)); ether_addr_copy(hsr_sp->macaddress_A, master->dev->dev_addr); + if (hsr->redbox) { + hsr_stlv = skb_put(skb, sizeof(struct hsr_sup_tlv)); + hsr_stlv->HSR_TLV_type = PRP_TLV_REDBOX_MAC; + hsr_stlv->HSR_TLV_length = sizeof(struct hsr_sup_payload); + + /* Payload: MacAddressRedBox */ + hsr_sp = skb_put(skb, sizeof(struct hsr_sup_payload)); + ether_addr_copy(hsr_sp->macaddress_A, hsr->macaddress_redbox); + } + if (skb_put_padto(skb, ETH_ZLEN)) { spin_unlock_bh(&hsr->seqnr_lock); return; @@ -405,6 +419,10 @@ void hsr_del_ports(struct hsr_priv *hsr) if (port) hsr_del_port(port); + port = hsr_port_get_hsr(hsr, HSR_PT_INTERLINK); + if (port) + hsr_del_port(port); + port = hsr_port_get_hsr(hsr, HSR_PT_MASTER); if (port) hsr_del_port(port); @@ -534,8 +552,8 @@ static const unsigned char def_multicast_addr[ETH_ALEN] __aligned(2) = { }; int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], - unsigned char multicast_spec, u8 protocol_version, - struct netlink_ext_ack *extack) + struct net_device *interlink, unsigned char multicast_spec, + u8 protocol_version, struct netlink_ext_ack *extack) { bool unregister = false; struct hsr_priv *hsr; @@ -544,6 +562,7 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], hsr = netdev_priv(hsr_dev); INIT_LIST_HEAD(&hsr->ports); INIT_LIST_HEAD(&hsr->node_db); + INIT_LIST_HEAD(&hsr->proxy_node_db); spin_lock_init(&hsr->list_lock); eth_hw_addr_set(hsr_dev, slave[0]->dev_addr); @@ -569,6 +588,7 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], /* Overflow soon to find bugs easier: */ hsr->sequence_nr = HSR_SEQNR_START; hsr->sup_sequence_nr = HSR_SUP_SEQNR_START; + hsr->interlink_sequence_nr = HSR_SEQNR_START; timer_setup(&hsr->announce_timer, hsr_announce, 0); timer_setup(&hsr->prune_timer, hsr_prune_nodes, 0); @@ -604,6 +624,18 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], if (res) goto err_unregister; + if (interlink) { + res = hsr_add_port(hsr, interlink, HSR_PT_INTERLINK, extack); + if (res) + goto err_unregister; + + hsr->redbox = true; + ether_addr_copy(hsr->macaddress_redbox, interlink->dev_addr); + timer_setup(&hsr->prune_proxy_timer, hsr_prune_proxy_nodes, 0); + mod_timer(&hsr->prune_proxy_timer, + jiffies + msecs_to_jiffies(PRUNE_PROXY_PERIOD)); + } + hsr_debugfs_init(hsr, hsr_dev); mod_timer(&hsr->prune_timer, jiffies + msecs_to_jiffies(PRUNE_PERIOD)); diff --git a/net/hsr/hsr_device.h b/net/hsr/hsr_device.h index 9060c92168f9..655284095b78 100644 --- a/net/hsr/hsr_device.h +++ b/net/hsr/hsr_device.h @@ -16,8 +16,8 @@ void hsr_del_ports(struct hsr_priv *hsr); void hsr_dev_setup(struct net_device *dev); int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], - unsigned char multicast_spec, u8 protocol_version, - struct netlink_ext_ack *extack); + struct net_device *interlink, unsigned char multicast_spec, + u8 protocol_version, struct netlink_ext_ack *extack); void hsr_check_carrier_and_operstate(struct hsr_priv *hsr); int hsr_get_max_mtu(struct hsr_priv *hsr); #endif /* __HSR_DEVICE_H */ diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c index 5d68cb181695..05a61b8286ec 100644 --- a/net/hsr/hsr_forward.c +++ b/net/hsr/hsr_forward.c @@ -377,6 +377,15 @@ static int hsr_xmit(struct sk_buff *skb, struct hsr_port *port, */ ether_addr_copy(eth_hdr(skb)->h_source, port->dev->dev_addr); } + + /* When HSR node is used as RedBox - the frame received from HSR ring + * requires source MAC address (SA) replacement to one which can be + * recognized by SAN devices (otherwise, frames are dropped by switch) + */ + if (port->type == HSR_PT_INTERLINK) + ether_addr_copy(eth_hdr(skb)->h_source, + port->hsr->macaddress_redbox); + return dev_queue_xmit(skb); } @@ -390,9 +399,57 @@ bool prp_drop_frame(struct hsr_frame_info *frame, struct hsr_port *port) bool hsr_drop_frame(struct hsr_frame_info *frame, struct hsr_port *port) { + struct sk_buff *skb; + if (port->dev->features & NETIF_F_HW_HSR_FWD) return prp_drop_frame(frame, port); + /* RedBox specific frames dropping policies + * + * Do not send HSR supervisory frames to SAN devices + */ + if (frame->is_supervision && port->type == HSR_PT_INTERLINK) + return true; + + /* Do not forward to other HSR port (A or B) unicast frames which + * are addressed to interlink port (and are in the ProxyNodeTable). + */ + skb = frame->skb_hsr; + if (skb && prp_drop_frame(frame, port) && + is_unicast_ether_addr(eth_hdr(skb)->h_dest) && + hsr_is_node_in_db(&port->hsr->proxy_node_db, + eth_hdr(skb)->h_dest)) { + return true; + } + + /* Do not forward to port C (Interlink) frames from nodes A and B + * if DA is in NodeTable. + */ + if ((frame->port_rcv->type == HSR_PT_SLAVE_A || + frame->port_rcv->type == HSR_PT_SLAVE_B) && + port->type == HSR_PT_INTERLINK) { + skb = frame->skb_hsr; + if (skb && is_unicast_ether_addr(eth_hdr(skb)->h_dest) && + hsr_is_node_in_db(&port->hsr->node_db, + eth_hdr(skb)->h_dest)) { + return true; + } + } + + /* Do not forward to port A and B unicast frames received on the + * interlink port if it is addressed to one of nodes registered in + * the ProxyNodeTable. + */ + if ((port->type == HSR_PT_SLAVE_A || port->type == HSR_PT_SLAVE_B) && + frame->port_rcv->type == HSR_PT_INTERLINK) { + skb = frame->skb_std; + if (skb && is_unicast_ether_addr(eth_hdr(skb)->h_dest) && + hsr_is_node_in_db(&port->hsr->proxy_node_db, + eth_hdr(skb)->h_dest)) { + return true; + } + } + return false; } @@ -448,13 +505,14 @@ static void hsr_forward_do(struct hsr_frame_info *frame) } /* Check if frame is to be dropped. Eg. for PRP no forward - * between ports. + * between ports, or sending HSR supervision to RedBox. */ if (hsr->proto_ops->drop_frame && hsr->proto_ops->drop_frame(frame, port)) continue; - if (port->type != HSR_PT_MASTER) + if (port->type == HSR_PT_SLAVE_A || + port->type == HSR_PT_SLAVE_B) skb = hsr->proto_ops->create_tagged_frame(frame, port); else skb = hsr->proto_ops->get_untagged_frame(frame, port); @@ -469,7 +527,9 @@ static void hsr_forward_do(struct hsr_frame_info *frame) hsr_deliver_master(skb, port->dev, frame->node_src); } else { if (!hsr_xmit(skb, port, frame)) - sent = true; + if (port->type == HSR_PT_SLAVE_A || + port->type == HSR_PT_SLAVE_B) + sent = true; } } } @@ -503,10 +563,12 @@ static void handle_std_frame(struct sk_buff *skb, frame->skb_prp = NULL; frame->skb_std = skb; - if (port->type != HSR_PT_MASTER) { + if (port->type != HSR_PT_MASTER) frame->is_from_san = true; - } else { - /* Sequence nr for the master node */ + + if (port->type == HSR_PT_MASTER || + port->type == HSR_PT_INTERLINK) { + /* Sequence nr for the master/interlink node */ lockdep_assert_held(&hsr->seqnr_lock); frame->sequence_nr = hsr->sequence_nr; hsr->sequence_nr++; @@ -564,6 +626,7 @@ static int fill_frame_info(struct hsr_frame_info *frame, { struct hsr_priv *hsr = port->hsr; struct hsr_vlan_ethhdr *vlan_hdr; + struct list_head *n_db; struct ethhdr *ethhdr; __be16 proto; int ret; @@ -574,9 +637,13 @@ static int fill_frame_info(struct hsr_frame_info *frame, memset(frame, 0, sizeof(*frame)); frame->is_supervision = is_supervision_frame(port->hsr, skb); - frame->node_src = hsr_get_node(port, &hsr->node_db, skb, - frame->is_supervision, - port->type); + + n_db = &hsr->node_db; + if (port->type == HSR_PT_INTERLINK) + n_db = &hsr->proxy_node_db; + + frame->node_src = hsr_get_node(port, n_db, skb, + frame->is_supervision, port->type); if (!frame->node_src) return -1; /* Unknown node and !is_supervision, or no mem */ diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c index 26329db09210..614df9649794 100644 --- a/net/hsr/hsr_framereg.c +++ b/net/hsr/hsr_framereg.c @@ -71,6 +71,14 @@ static struct hsr_node *find_node_by_addr_A(struct list_head *node_db, return NULL; } +/* Check if node for a given MAC address is already present in data base + */ +bool hsr_is_node_in_db(struct list_head *node_db, + const unsigned char addr[ETH_ALEN]) +{ + return !!find_node_by_addr_A(node_db, addr); +} + /* Helper for device init; the self_node is used in hsr_rcv() to recognize * frames from self that's been looped over the HSR ring. */ @@ -223,6 +231,15 @@ struct hsr_node *hsr_get_node(struct hsr_port *port, struct list_head *node_db, } } + /* Check if required node is not in proxy nodes table */ + list_for_each_entry_rcu(node, &hsr->proxy_node_db, mac_list) { + if (ether_addr_equal(node->macaddress_A, ethhdr->h_source)) { + if (hsr->proto_ops->update_san_info) + hsr->proto_ops->update_san_info(node, is_sup); + return node; + } + } + /* Everyone may create a node entry, connected node to a HSR/PRP * device. */ @@ -418,6 +435,10 @@ void hsr_addr_subst_dest(struct hsr_node *node_src, struct sk_buff *skb, node_dst = find_node_by_addr_A(&port->hsr->node_db, eth_hdr(skb)->h_dest); + if (!node_dst && port->hsr->redbox) + node_dst = find_node_by_addr_A(&port->hsr->proxy_node_db, + eth_hdr(skb)->h_dest); + if (!node_dst) { if (port->hsr->prot_version != PRP_V1 && net_ratelimit()) netdev_err(skb->dev, "%s: Unknown node\n", __func__); @@ -561,6 +582,37 @@ void hsr_prune_nodes(struct timer_list *t) jiffies + msecs_to_jiffies(PRUNE_PERIOD)); } +void hsr_prune_proxy_nodes(struct timer_list *t) +{ + struct hsr_priv *hsr = from_timer(hsr, t, prune_proxy_timer); + unsigned long timestamp; + struct hsr_node *node; + struct hsr_node *tmp; + + spin_lock_bh(&hsr->list_lock); + list_for_each_entry_safe(node, tmp, &hsr->proxy_node_db, mac_list) { + timestamp = node->time_in[HSR_PT_INTERLINK]; + + /* Prune old entries */ + if (time_is_before_jiffies(timestamp + + msecs_to_jiffies(HSR_PROXY_NODE_FORGET_TIME))) { + hsr_nl_nodedown(hsr, node->macaddress_A); + if (!node->removed) { + list_del_rcu(&node->mac_list); + node->removed = true; + /* Note that we need to free this entry later: */ + kfree_rcu(node, rcu_head); + } + } + } + + spin_unlock_bh(&hsr->list_lock); + + /* Restart timer */ + mod_timer(&hsr->prune_proxy_timer, + jiffies + msecs_to_jiffies(PRUNE_PROXY_PERIOD)); +} + void *hsr_get_next_node(struct hsr_priv *hsr, void *_pos, unsigned char addr[ETH_ALEN]) { diff --git a/net/hsr/hsr_framereg.h b/net/hsr/hsr_framereg.h index b23556251d62..7619e31c1d2d 100644 --- a/net/hsr/hsr_framereg.h +++ b/net/hsr/hsr_framereg.h @@ -46,6 +46,7 @@ int hsr_register_frame_out(struct hsr_port *port, struct hsr_node *node, u16 sequence_nr); void hsr_prune_nodes(struct timer_list *t); +void hsr_prune_proxy_nodes(struct timer_list *t); int hsr_create_self_node(struct hsr_priv *hsr, const unsigned char addr_a[ETH_ALEN], @@ -67,6 +68,9 @@ void prp_handle_san_frame(bool san, enum hsr_port_type port, struct hsr_node *node); void prp_update_san_info(struct hsr_node *node, bool is_sup); +bool hsr_is_node_in_db(struct list_head *node_db, + const unsigned char addr[ETH_ALEN]); + struct hsr_node { struct list_head mac_list; /* Protect R/W access to seq_out */ diff --git a/net/hsr/hsr_main.h b/net/hsr/hsr_main.h index 18e01791ad79..23850b16d1ea 100644 --- a/net/hsr/hsr_main.h +++ b/net/hsr/hsr_main.h @@ -21,6 +21,7 @@ */ #define HSR_LIFE_CHECK_INTERVAL 2000 /* ms */ #define HSR_NODE_FORGET_TIME 60000 /* ms */ +#define HSR_PROXY_NODE_FORGET_TIME 60000 /* ms */ #define HSR_ANNOUNCE_INTERVAL 100 /* ms */ #define HSR_ENTRY_FORGET_TIME 400 /* ms */ @@ -35,6 +36,7 @@ * HSR_NODE_FORGET_TIME? */ #define PRUNE_PERIOD 3000 /* ms */ +#define PRUNE_PROXY_PERIOD 3000 /* ms */ #define HSR_TLV_EOT 0 /* End of TLVs */ #define HSR_TLV_ANNOUNCE 22 #define HSR_TLV_LIFE_CHECK 23 @@ -192,11 +194,14 @@ struct hsr_priv { struct rcu_head rcu_head; struct list_head ports; struct list_head node_db; /* Known HSR nodes */ + struct list_head proxy_node_db; /* RedBox HSR proxy nodes */ struct hsr_self_node __rcu *self_node; /* MACs of slaves */ struct timer_list announce_timer; /* Supervision frame dispatch */ struct timer_list prune_timer; + struct timer_list prune_proxy_timer; int announce_count; u16 sequence_nr; + u16 interlink_sequence_nr; /* Interlink port seq_nr */ u16 sup_sequence_nr; /* For HSRv1 separate seq_nr for supervision */ enum hsr_version prot_version; /* Indicate if HSRv0, HSRv1 or PRPv1 */ spinlock_t seqnr_lock; /* locking for sequence_nr */ @@ -209,6 +214,8 @@ struct hsr_priv { * of lan_id */ bool fwd_offloaded; /* Forwarding offloaded to HW */ + bool redbox; /* Device supports HSR RedBox */ + unsigned char macaddress_redbox[ETH_ALEN]; unsigned char sup_multicast_addr[ETH_ALEN] __aligned(sizeof(u16)); /* Align to u16 boundary to avoid unaligned access * in ether_addr_equal diff --git a/net/hsr/hsr_netlink.c b/net/hsr/hsr_netlink.c index 78fe40eb9f01..898f18c6da53 100644 --- a/net/hsr/hsr_netlink.c +++ b/net/hsr/hsr_netlink.c @@ -23,6 +23,7 @@ static const struct nla_policy hsr_policy[IFLA_HSR_MAX + 1] = { [IFLA_HSR_SUPERVISION_ADDR] = { .len = ETH_ALEN }, [IFLA_HSR_SEQ_NR] = { .type = NLA_U16 }, [IFLA_HSR_PROTOCOL] = { .type = NLA_U8 }, + [IFLA_HSR_INTERLINK] = { .type = NLA_U32 }, }; /* Here, it seems a netdevice has already been allocated for us, and the @@ -35,8 +36,8 @@ static int hsr_newlink(struct net *src_net, struct net_device *dev, enum hsr_version proto_version; unsigned char multicast_spec; u8 proto = HSR_PROTOCOL_HSR; - struct net_device *link[2]; + struct net_device *link[2], *interlink = NULL; if (!data) { NL_SET_ERR_MSG_MOD(extack, "No slave devices specified"); return -EINVAL; @@ -67,6 +68,20 @@ static int hsr_newlink(struct net *src_net, struct net_device *dev, return -EINVAL; } + if (data[IFLA_HSR_INTERLINK]) + interlink = __dev_get_by_index(src_net, + nla_get_u32(data[IFLA_HSR_INTERLINK])); + + if (interlink && interlink == link[0]) { + NL_SET_ERR_MSG_MOD(extack, "Interlink and Slave1 are the same"); + return -EINVAL; + } + + if (interlink && interlink == link[1]) { + NL_SET_ERR_MSG_MOD(extack, "Interlink and Slave2 are the same"); + return -EINVAL; + } + if (!data[IFLA_HSR_MULTICAST_SPEC]) multicast_spec = 0; else @@ -96,10 +111,17 @@ static int hsr_newlink(struct net *src_net, struct net_device *dev, } } - if (proto == HSR_PROTOCOL_PRP) + if (proto == HSR_PROTOCOL_PRP) { proto_version = PRP_V1; + if (interlink) { + NL_SET_ERR_MSG_MOD(extack, + "Interlink only works with HSR"); + return -EINVAL; + } + } - return hsr_dev_finalize(dev, link, multicast_spec, proto_version, extack); + return hsr_dev_finalize(dev, link, interlink, multicast_spec, + proto_version, extack); } static void hsr_dellink(struct net_device *dev, struct list_head *head) @@ -107,6 +129,7 @@ static void hsr_dellink(struct net_device *dev, struct list_head *head) struct hsr_priv *hsr = netdev_priv(dev); del_timer_sync(&hsr->prune_timer); + del_timer_sync(&hsr->prune_proxy_timer); del_timer_sync(&hsr->announce_timer); hsr_debugfs_term(hsr); @@ -114,6 +137,7 @@ static void hsr_dellink(struct net_device *dev, struct list_head *head) hsr_del_self_node(hsr); hsr_del_nodes(&hsr->node_db); + hsr_del_nodes(&hsr->proxy_node_db); unregister_netdevice_queue(dev, head); } diff --git a/net/hsr/hsr_slave.c b/net/hsr/hsr_slave.c index 1b6457f357bd..af6cf64a00e0 100644 --- a/net/hsr/hsr_slave.c +++ b/net/hsr/hsr_slave.c @@ -55,6 +55,7 @@ static rx_handler_result_t hsr_handle_frame(struct sk_buff **pskb) protocol = eth_hdr(skb)->h_proto; if (!(port->dev->features & NETIF_F_HW_HSR_TAG_RM) && + port->type != HSR_PT_INTERLINK && hsr->proto_ops->invalid_dan_ingress_frame && hsr->proto_ops->invalid_dan_ingress_frame(protocol)) goto finish_pass; -- cgit v1.2.3 From 4b911a9690d72641879ea6d13cce1de31d346d79 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Tue, 23 Apr 2024 19:35:49 -0700 Subject: nsh: Restore skb->{protocol,data,mac_header} for outer header in nsh_gso_segment(). syzbot triggered various splats (see [0] and links) by a crafted GSO packet of VIRTIO_NET_HDR_GSO_UDP layering the following protocols: ETH_P_8021AD + ETH_P_NSH + ETH_P_IPV6 + IPPROTO_UDP NSH can encapsulate IPv4, IPv6, Ethernet, NSH, and MPLS. As the inner protocol can be Ethernet, NSH GSO handler, nsh_gso_segment(), calls skb_mac_gso_segment() to invoke inner protocol GSO handlers. nsh_gso_segment() does the following for the original skb before calling skb_mac_gso_segment() 1. reset skb->network_header 2. save the original skb->{mac_heaeder,mac_len} in a local variable 3. pull the NSH header 4. resets skb->mac_header 5. set up skb->mac_len and skb->protocol for the inner protocol. and does the following for the segmented skb 6. set ntohs(ETH_P_NSH) to skb->protocol 7. push the NSH header 8. restore skb->mac_header 9. set skb->mac_header + mac_len to skb->network_header 10. restore skb->mac_len There are two problems in 6-7 and 8-9. (a) After 6 & 7, skb->data points to the NSH header, so the outer header (ETH_P_8021AD in this case) is stripped when skb is sent out of netdev. Also, if NSH is encapsulated by NSH + Ethernet (so NSH-Ethernet-NSH), skb_pull() in the first nsh_gso_segment() will make skb->data point to the middle of the outer NSH or Ethernet header because the Ethernet header is not pulled by the second nsh_gso_segment(). (b) While restoring skb->{mac_header,network_header} in 8 & 9, nsh_gso_segment() does not assume that the data in the linear buffer is shifted. However, udp6_ufo_fragment() could shift the data and change skb->mac_header accordingly as demonstrated by syzbot. If this happens, even the restored skb->mac_header points to the middle of the outer header. It seems nsh_gso_segment() has never worked with outer headers so far. At the end of nsh_gso_segment(), the outer header must be restored for the segmented skb, instead of the NSH header. To do that, let's calculate the outer header position relatively from the inner header and set skb->{data,mac_header,protocol} properly. [0]: BUG: KMSAN: uninit-value in ipvlan_process_outbound drivers/net/ipvlan/ipvlan_core.c:524 [inline] BUG: KMSAN: uninit-value in ipvlan_xmit_mode_l3 drivers/net/ipvlan/ipvlan_core.c:602 [inline] BUG: KMSAN: uninit-value in ipvlan_queue_xmit+0xf44/0x16b0 drivers/net/ipvlan/ipvlan_core.c:668 ipvlan_process_outbound drivers/net/ipvlan/ipvlan_core.c:524 [inline] ipvlan_xmit_mode_l3 drivers/net/ipvlan/ipvlan_core.c:602 [inline] ipvlan_queue_xmit+0xf44/0x16b0 drivers/net/ipvlan/ipvlan_core.c:668 ipvlan_start_xmit+0x5c/0x1a0 drivers/net/ipvlan/ipvlan_main.c:222 __netdev_start_xmit include/linux/netdevice.h:4989 [inline] netdev_start_xmit include/linux/netdevice.h:5003 [inline] xmit_one net/core/dev.c:3547 [inline] dev_hard_start_xmit+0x244/0xa10 net/core/dev.c:3563 __dev_queue_xmit+0x33ed/0x51c0 net/core/dev.c:4351 dev_queue_xmit include/linux/netdevice.h:3171 [inline] packet_xmit+0x9c/0x6b0 net/packet/af_packet.c:276 packet_snd net/packet/af_packet.c:3081 [inline] packet_sendmsg+0x8aef/0x9f10 net/packet/af_packet.c:3113 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg net/socket.c:745 [inline] __sys_sendto+0x735/0xa10 net/socket.c:2191 __do_sys_sendto net/socket.c:2203 [inline] __se_sys_sendto net/socket.c:2199 [inline] __x64_sys_sendto+0x125/0x1c0 net/socket.c:2199 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x63/0x6b Uninit was created at: slab_post_alloc_hook mm/slub.c:3819 [inline] slab_alloc_node mm/slub.c:3860 [inline] __do_kmalloc_node mm/slub.c:3980 [inline] __kmalloc_node_track_caller+0x705/0x1000 mm/slub.c:4001 kmalloc_reserve+0x249/0x4a0 net/core/skbuff.c:582 __alloc_skb+0x352/0x790 net/core/skbuff.c:651 skb_segment+0x20aa/0x7080 net/core/skbuff.c:4647 udp6_ufo_fragment+0xcab/0x1150 net/ipv6/udp_offload.c:109 ipv6_gso_segment+0x14be/0x2ca0 net/ipv6/ip6_offload.c:152 skb_mac_gso_segment+0x3e8/0x760 net/core/gso.c:53 nsh_gso_segment+0x6f4/0xf70 net/nsh/nsh.c:108 skb_mac_gso_segment+0x3e8/0x760 net/core/gso.c:53 __skb_gso_segment+0x4b0/0x730 net/core/gso.c:124 skb_gso_segment include/net/gso.h:83 [inline] validate_xmit_skb+0x107f/0x1930 net/core/dev.c:3628 __dev_queue_xmit+0x1f28/0x51c0 net/core/dev.c:4343 dev_queue_xmit include/linux/netdevice.h:3171 [inline] packet_xmit+0x9c/0x6b0 net/packet/af_packet.c:276 packet_snd net/packet/af_packet.c:3081 [inline] packet_sendmsg+0x8aef/0x9f10 net/packet/af_packet.c:3113 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg net/socket.c:745 [inline] __sys_sendto+0x735/0xa10 net/socket.c:2191 __do_sys_sendto net/socket.c:2203 [inline] __se_sys_sendto net/socket.c:2199 [inline] __x64_sys_sendto+0x125/0x1c0 net/socket.c:2199 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x63/0x6b CPU: 1 PID: 5101 Comm: syz-executor421 Not tainted 6.8.0-rc5-syzkaller-00297-gf2e367d6ad3b #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/25/2024 Fixes: c411ed854584 ("nsh: add GSO support") Reported-and-tested-by: syzbot+42a0dc856239de4de60e@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=42a0dc856239de4de60e Reported-and-tested-by: syzbot+c298c9f0e46a3c86332b@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=c298c9f0e46a3c86332b Link: https://lore.kernel.org/netdev/20240415222041.18537-1-kuniyu@amazon.com/ Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240424023549.21862-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni --- net/nsh/nsh.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/nsh/nsh.c b/net/nsh/nsh.c index f4a38bd6a7e0..bfb7758063f3 100644 --- a/net/nsh/nsh.c +++ b/net/nsh/nsh.c @@ -77,13 +77,15 @@ EXPORT_SYMBOL_GPL(nsh_pop); static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, netdev_features_t features) { + unsigned int outer_hlen, mac_len, nsh_len; struct sk_buff *segs = ERR_PTR(-EINVAL); u16 mac_offset = skb->mac_header; - unsigned int nsh_len, mac_len; - __be16 proto; + __be16 outer_proto, proto; skb_reset_network_header(skb); + outer_proto = skb->protocol; + outer_hlen = skb_mac_header_len(skb); mac_len = skb->mac_len; if (unlikely(!pskb_may_pull(skb, NSH_BASE_HDR_LEN))) @@ -113,10 +115,10 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, } for (skb = segs; skb; skb = skb->next) { - skb->protocol = htons(ETH_P_NSH); - __skb_push(skb, nsh_len); - skb->mac_header = mac_offset; - skb->network_header = skb->mac_header + mac_len; + skb->protocol = outer_proto; + __skb_push(skb, nsh_len + outer_hlen); + skb_reset_mac_header(skb); + skb_set_network_header(skb, outer_hlen); skb->mac_len = mac_len; } -- cgit v1.2.3 From 42f853b42899d9b445763b55c3c8adc72be0f0e1 Mon Sep 17 00:00:00 2001 From: David Bauer Date: Wed, 24 Apr 2024 19:11:10 +0200 Subject: net l2tp: drop flow hash on forward Drop the flow-hash of the skb when forwarding to the L2TP netdev. This avoids the L2TP qdisc from using the flow-hash from the outer packet, which is identical for every flow within the tunnel. This does not affect every platform but is specific for the ethernet driver. It depends on the platform including L4 information in the flow-hash. One such example is the Mediatek Filogic MT798x family of networking processors. Fixes: d9e31d17ceba ("l2tp: Add L2TP ethernet pseudowire support") Acked-by: James Chapman Signed-off-by: David Bauer Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240424171110.13701-1-mail@david-bauer.net Signed-off-by: Paolo Abeni --- net/l2tp/l2tp_eth.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index 39e487ccc468..8ba00ad433c2 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -127,6 +127,9 @@ static void l2tp_eth_dev_recv(struct l2tp_session *session, struct sk_buff *skb, /* checksums verified by L2TP */ skb->ip_summed = CHECKSUM_NONE; + /* drop outer flow-hash */ + skb_clear_hash(skb); + skb_dst_drop(skb); nf_reset_ct(skb); -- cgit v1.2.3 From 6be49deaa09576c141002a2e6f816a1709bc2c86 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 25 Apr 2024 11:13:35 +0800 Subject: rstreason: prepare for passive reset Adjust the parameter and support passing reason of reset which is for now NOT_SPECIFIED. No functional changes. Signed-off-by: Jason Xing Acked-by: Matthieu Baerts (NGI0) Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/net/request_sock.h | 4 +++- net/dccp/ipv4.c | 10 ++++++---- net/dccp/ipv6.c | 10 ++++++---- net/dccp/minisocks.c | 3 ++- net/ipv4/tcp_ipv4.c | 12 +++++++----- net/ipv4/tcp_minisocks.c | 3 ++- net/ipv6/tcp_ipv6.c | 15 +++++++++------ net/mptcp/subflow.c | 8 +++++--- 8 files changed, 40 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 004e651e6067..bdc737832da6 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -18,6 +18,7 @@ #include #include +#include struct request_sock; struct sk_buff; @@ -34,7 +35,8 @@ struct request_sock_ops { void (*send_ack)(const struct sock *sk, struct sk_buff *skb, struct request_sock *req); void (*send_reset)(const struct sock *sk, - struct sk_buff *skb); + struct sk_buff *skb, + enum sk_rst_reason reason); void (*destructor)(struct request_sock *req); void (*syn_ack_timeout)(const struct request_sock *req); }; diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 9fc9cea4c251..ff41bd6f99c3 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -24,6 +24,7 @@ #include #include #include +#include #include "ackvec.h" #include "ccid.h" @@ -521,7 +522,8 @@ out: return err; } -static void dccp_v4_ctl_send_reset(const struct sock *sk, struct sk_buff *rxskb) +static void dccp_v4_ctl_send_reset(const struct sock *sk, struct sk_buff *rxskb, + enum sk_rst_reason reason) { int err; const struct iphdr *rxiph; @@ -706,7 +708,7 @@ int dccp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - dccp_v4_ctl_send_reset(sk, skb); + dccp_v4_ctl_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); kfree_skb(skb); return 0; } @@ -869,7 +871,7 @@ lookup: if (nsk == sk) { reqsk_put(req); } else if (dccp_child_process(sk, nsk, skb)) { - dccp_v4_ctl_send_reset(sk, skb); + dccp_v4_ctl_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); goto discard_and_relse; } else { sock_put(sk); @@ -909,7 +911,7 @@ no_dccp_socket: if (dh->dccph_type != DCCP_PKT_RESET) { DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; - dccp_v4_ctl_send_reset(sk, skb); + dccp_v4_ctl_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); } discard_it: diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index c8ca703dc331..85f4b8fdbe5e 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -29,6 +29,7 @@ #include #include #include +#include #include "dccp.h" #include "ipv6.h" @@ -256,7 +257,8 @@ static void dccp_v6_reqsk_destructor(struct request_sock *req) kfree_skb(inet_rsk(req)->pktopts); } -static void dccp_v6_ctl_send_reset(const struct sock *sk, struct sk_buff *rxskb) +static void dccp_v6_ctl_send_reset(const struct sock *sk, struct sk_buff *rxskb, + enum sk_rst_reason reason) { const struct ipv6hdr *rxip6h; struct sk_buff *skb; @@ -656,7 +658,7 @@ static int dccp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - dccp_v6_ctl_send_reset(sk, skb); + dccp_v6_ctl_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); discard: if (opt_skb != NULL) __kfree_skb(opt_skb); @@ -762,7 +764,7 @@ lookup: if (nsk == sk) { reqsk_put(req); } else if (dccp_child_process(sk, nsk, skb)) { - dccp_v6_ctl_send_reset(sk, skb); + dccp_v6_ctl_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); goto discard_and_relse; } else { sock_put(sk); @@ -801,7 +803,7 @@ no_dccp_socket: if (dh->dccph_type != DCCP_PKT_RESET) { DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION; - dccp_v6_ctl_send_reset(sk, skb); + dccp_v6_ctl_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); } discard_it: diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 64d805b27add..251a57cf5822 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -15,6 +15,7 @@ #include #include #include +#include #include "ackvec.h" #include "ccid.h" @@ -202,7 +203,7 @@ struct sock *dccp_check_req(struct sock *sk, struct sk_buff *skb, DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY; drop: if (dccp_hdr(skb)->dccph_type != DCCP_PKT_RESET) - req->rsk_ops->send_reset(sk, skb); + req->rsk_ops->send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); inet_csk_reqsk_queue_drop(sk, req); out: diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e06f0cd04f7e..202f6dfa958b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -70,6 +70,7 @@ #include #include #include +#include #include #include @@ -723,7 +724,8 @@ out: * Exception: precedence violation. We do not implement it in any case. */ -static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb) +static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb, + enum sk_rst_reason reason) { const struct tcphdr *th = tcp_hdr(skb); struct { @@ -1934,7 +1936,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - tcp_v4_send_reset(rsk, skb); + tcp_v4_send_reset(rsk, skb, SK_RST_REASON_NOT_SPECIFIED); discard: kfree_skb_reason(skb, reason); /* Be careful here. If this function gets more complicated and @@ -2285,7 +2287,7 @@ lookup: } else { drop_reason = tcp_child_process(sk, nsk, skb); if (drop_reason) { - tcp_v4_send_reset(nsk, skb); + tcp_v4_send_reset(nsk, skb, SK_RST_REASON_NOT_SPECIFIED); goto discard_and_relse; } sock_put(sk); @@ -2364,7 +2366,7 @@ csum_error: bad_packet: __TCP_INC_STATS(net, TCP_MIB_INERRS); } else { - tcp_v4_send_reset(NULL, skb); + tcp_v4_send_reset(NULL, skb, SK_RST_REASON_NOT_SPECIFIED); } discard_it: @@ -2416,7 +2418,7 @@ do_time_wait: tcp_v4_timewait_ack(sk, skb); break; case TCP_TW_RST: - tcp_v4_send_reset(sk, skb); + tcp_v4_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); inet_twsk_deschedule_put(inet_twsk(sk)); goto discard_it; case TCP_TW_SUCCESS:; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 146c061145b4..7d543569a180 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -22,6 +22,7 @@ #include #include #include +#include static bool tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win) { @@ -878,7 +879,7 @@ embryonic_reset: * avoid becoming vulnerable to outside attack aiming at * resetting legit local connections. */ - req->rsk_ops->send_reset(sk, skb); + req->rsk_ops->send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); } else if (fastopen) { /* received a valid RST pkt */ reqsk_fastopen_remove(sk, req, true); tcp_reset(sk, skb); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index bb7c3caf4f85..017f6293b5f4 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -60,6 +60,7 @@ #include #include #include +#include #include #include @@ -69,7 +70,8 @@ #include -static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb); +static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb, + enum sk_rst_reason reason); static void tcp_v6_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb, struct request_sock *req); @@ -1008,7 +1010,8 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32 kfree_skb(buff); } -static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb) +static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb, + enum sk_rst_reason reason) { const struct tcphdr *th = tcp_hdr(skb); struct ipv6hdr *ipv6h = ipv6_hdr(skb); @@ -1677,7 +1680,7 @@ int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - tcp_v6_send_reset(sk, skb); + tcp_v6_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); discard: if (opt_skb) __kfree_skb(opt_skb); @@ -1862,7 +1865,7 @@ lookup: } else { drop_reason = tcp_child_process(sk, nsk, skb); if (drop_reason) { - tcp_v6_send_reset(nsk, skb); + tcp_v6_send_reset(nsk, skb, SK_RST_REASON_NOT_SPECIFIED); goto discard_and_relse; } sock_put(sk); @@ -1939,7 +1942,7 @@ csum_error: bad_packet: __TCP_INC_STATS(net, TCP_MIB_INERRS); } else { - tcp_v6_send_reset(NULL, skb); + tcp_v6_send_reset(NULL, skb, SK_RST_REASON_NOT_SPECIFIED); } discard_it: @@ -1995,7 +1998,7 @@ do_time_wait: tcp_v6_timewait_ack(sk, skb); break; case TCP_TW_RST: - tcp_v6_send_reset(sk, skb); + tcp_v6_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); inet_twsk_deschedule_put(inet_twsk(sk)); goto discard_it; case TCP_TW_SUCCESS: diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index b94d1dca1094..32fe2ef36d56 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -20,6 +20,8 @@ #include #endif #include +#include + #include "protocol.h" #include "mib.h" @@ -308,7 +310,7 @@ static struct dst_entry *subflow_v4_route_req(const struct sock *sk, dst_release(dst); if (!req->syncookie) - tcp_request_sock_ops.send_reset(sk, skb); + tcp_request_sock_ops.send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); return NULL; } @@ -376,7 +378,7 @@ static struct dst_entry *subflow_v6_route_req(const struct sock *sk, dst_release(dst); if (!req->syncookie) - tcp6_request_sock_ops.send_reset(sk, skb); + tcp6_request_sock_ops.send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); return NULL; } #endif @@ -911,7 +913,7 @@ dispose_child: tcp_rsk(req)->drop_req = true; inet_csk_prepare_for_destroy_sock(child); tcp_done(child); - req->rsk_ops->send_reset(sk, skb); + req->rsk_ops->send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); /* The last child reference will be released by the caller */ return child; -- cgit v1.2.3 From 5691276b39daf90294c6a81fb6d62d667f634c92 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 25 Apr 2024 11:13:36 +0800 Subject: rstreason: prepare for active reset Like what we did to passive reset: only passing possible reset reason in each active reset path. No functional changes. Signed-off-by: Jason Xing Acked-by: Matthieu Baerts (NGI0) Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/net/tcp.h | 3 ++- net/ipv4/tcp.c | 15 ++++++++++----- net/ipv4/tcp_output.c | 3 ++- net/ipv4/tcp_timer.c | 9 ++++++--- net/mptcp/protocol.c | 4 +++- net/mptcp/subflow.c | 5 +++-- 6 files changed, 26 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index ffc9371fe9de..a9eb21251195 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -670,7 +670,8 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, void tcp_send_probe0(struct sock *); int tcp_write_wakeup(struct sock *, int mib); void tcp_send_fin(struct sock *sk); -void tcp_send_active_reset(struct sock *sk, gfp_t priority); +void tcp_send_active_reset(struct sock *sk, gfp_t priority, + enum sk_rst_reason reason); int tcp_send_synack(struct sock *); void tcp_push_one(struct sock *, unsigned int mss_now); void __tcp_send_ack(struct sock *sk, u32 rcv_nxt); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f23b97777ea5..4ec0f4feee00 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -275,6 +275,7 @@ #include #include #include +#include #include #include @@ -2811,7 +2812,8 @@ void __tcp_close(struct sock *sk, long timeout) /* Unread data was tossed, zap the connection. */ NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONCLOSE); tcp_set_state(sk, TCP_CLOSE); - tcp_send_active_reset(sk, sk->sk_allocation); + tcp_send_active_reset(sk, sk->sk_allocation, + SK_RST_REASON_NOT_SPECIFIED); } else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { /* Check zero linger _after_ checking for unread data. */ sk->sk_prot->disconnect(sk, 0); @@ -2885,7 +2887,8 @@ adjudge_to_death: struct tcp_sock *tp = tcp_sk(sk); if (READ_ONCE(tp->linger2) < 0) { tcp_set_state(sk, TCP_CLOSE); - tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_send_active_reset(sk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONLINGER); } else { @@ -2903,7 +2906,8 @@ adjudge_to_death: if (sk->sk_state != TCP_CLOSE) { if (tcp_check_oom(sk, 0)) { tcp_set_state(sk, TCP_CLOSE); - tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_send_active_reset(sk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONMEMORY); } else if (!check_net(sock_net(sk))) { @@ -3007,7 +3011,7 @@ int tcp_disconnect(struct sock *sk, int flags) /* The last check adjusts for discrepancy of Linux wrt. RFC * states */ - tcp_send_active_reset(sk, gfp_any()); + tcp_send_active_reset(sk, gfp_any(), SK_RST_REASON_NOT_SPECIFIED); WRITE_ONCE(sk->sk_err, ECONNRESET); } else if (old_state == TCP_SYN_SENT) WRITE_ONCE(sk->sk_err, ECONNRESET); @@ -4564,7 +4568,8 @@ int tcp_abort(struct sock *sk, int err) smp_wmb(); sk_error_report(sk); if (tcp_need_reset(sk->sk_state)) - tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_send_active_reset(sk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); tcp_done(sk); } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index ce59e4499b66..41c352bf3394 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3615,7 +3615,8 @@ void tcp_send_fin(struct sock *sk) * was unread data in the receive queue. This behavior is recommended * by RFC 2525, section 2.17. -DaveM */ -void tcp_send_active_reset(struct sock *sk, gfp_t priority) +void tcp_send_active_reset(struct sock *sk, gfp_t priority, + enum sk_rst_reason reason) { struct sk_buff *skb; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 976db57b95d4..83fe7f62f7f1 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -22,6 +22,7 @@ #include #include #include +#include static u32 tcp_clamp_rto_to_user_timeout(const struct sock *sk) { @@ -127,7 +128,8 @@ static int tcp_out_of_resources(struct sock *sk, bool do_reset) (!tp->snd_wnd && !tp->packets_out)) do_reset = true; if (do_reset) - tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_send_active_reset(sk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); tcp_done(sk); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONMEMORY); return 1; @@ -768,7 +770,7 @@ static void tcp_keepalive_timer (struct timer_list *t) goto out; } } - tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_send_active_reset(sk, GFP_ATOMIC, SK_RST_REASON_NOT_SPECIFIED); goto death; } @@ -795,7 +797,8 @@ static void tcp_keepalive_timer (struct timer_list *t) icsk->icsk_probes_out > 0) || (user_timeout == 0 && icsk->icsk_probes_out >= keepalive_probes(tp))) { - tcp_send_active_reset(sk, GFP_ATOMIC); + tcp_send_active_reset(sk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); tcp_write_err(sk); goto out; } diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index f8bc34f0d973..065967086492 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -21,6 +21,7 @@ #endif #include #include +#include #include #include "protocol.h" #include "mib.h" @@ -2569,7 +2570,8 @@ static void mptcp_check_fastclose(struct mptcp_sock *msk) slow = lock_sock_fast(tcp_sk); if (tcp_sk->sk_state != TCP_CLOSE) { - tcp_send_active_reset(tcp_sk, GFP_ATOMIC); + tcp_send_active_reset(tcp_sk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); tcp_set_state(tcp_sk, TCP_CLOSE); } unlock_sock_fast(tcp_sk, slow); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 32fe2ef36d56..ac867d277860 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -414,7 +414,7 @@ void mptcp_subflow_reset(struct sock *ssk) /* must hold: tcp_done() could drop last reference on parent */ sock_hold(sk); - tcp_send_active_reset(ssk, GFP_ATOMIC); + tcp_send_active_reset(ssk, GFP_ATOMIC, SK_RST_REASON_NOT_SPECIFIED); tcp_done(ssk); if (!test_and_set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &mptcp_sk(sk)->flags)) mptcp_schedule_work(sk); @@ -1350,7 +1350,8 @@ reset: tcp_set_state(ssk, TCP_CLOSE); while ((skb = skb_peek(&ssk->sk_receive_queue))) sk_eat_skb(ssk, skb); - tcp_send_active_reset(ssk, GFP_ATOMIC); + tcp_send_active_reset(ssk, GFP_ATOMIC, + SK_RST_REASON_NOT_SPECIFIED); WRITE_ONCE(subflow->data_avail, false); return false; } -- cgit v1.2.3 From 120391ef9ca8fe8f82ea3f2961ad802043468226 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 25 Apr 2024 11:13:37 +0800 Subject: tcp: support rstreason for passive reset Reuse the dropreason logic to show the exact reason of tcp reset, so we can finally display the corresponding item in enum sk_reset_reason instead of reinventing new reset reasons. This patch replaces all the prior NOT_SPECIFIED reasons. Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/net/rstreason.h | 15 +++++++++++++++ net/ipv4/tcp_ipv4.c | 11 +++++++---- net/ipv6/tcp_ipv6.c | 11 +++++++---- 3 files changed, 29 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/include/net/rstreason.h b/include/net/rstreason.h index bc53b5a24505..df3b6ac0c9b3 100644 --- a/include/net/rstreason.h +++ b/include/net/rstreason.h @@ -103,4 +103,19 @@ enum sk_rst_reason { */ SK_RST_REASON_MAX, }; + +/* Convert skb drop reasons to enum sk_rst_reason type */ +static inline enum sk_rst_reason +sk_rst_convert_drop_reason(enum skb_drop_reason reason) +{ + switch (reason) { + case SKB_DROP_REASON_NOT_SPECIFIED: + return SK_RST_REASON_NOT_SPECIFIED; + case SKB_DROP_REASON_NO_SOCKET: + return SK_RST_REASON_NO_SOCKET; + default: + /* If we don't have our own corresponding reason */ + return SK_RST_REASON_NOT_SPECIFIED; + } +} #endif diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 202f6dfa958b..621879ba916d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1936,7 +1936,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - tcp_v4_send_reset(rsk, skb, SK_RST_REASON_NOT_SPECIFIED); + tcp_v4_send_reset(rsk, skb, sk_rst_convert_drop_reason(reason)); discard: kfree_skb_reason(skb, reason); /* Be careful here. If this function gets more complicated and @@ -2287,7 +2287,10 @@ lookup: } else { drop_reason = tcp_child_process(sk, nsk, skb); if (drop_reason) { - tcp_v4_send_reset(nsk, skb, SK_RST_REASON_NOT_SPECIFIED); + enum sk_rst_reason rst_reason; + + rst_reason = sk_rst_convert_drop_reason(drop_reason); + tcp_v4_send_reset(nsk, skb, rst_reason); goto discard_and_relse; } sock_put(sk); @@ -2366,7 +2369,7 @@ csum_error: bad_packet: __TCP_INC_STATS(net, TCP_MIB_INERRS); } else { - tcp_v4_send_reset(NULL, skb, SK_RST_REASON_NOT_SPECIFIED); + tcp_v4_send_reset(NULL, skb, sk_rst_convert_drop_reason(drop_reason)); } discard_it: @@ -2418,7 +2421,7 @@ do_time_wait: tcp_v4_timewait_ack(sk, skb); break; case TCP_TW_RST: - tcp_v4_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + tcp_v4_send_reset(sk, skb, sk_rst_convert_drop_reason(drop_reason)); inet_twsk_deschedule_put(inet_twsk(sk)); goto discard_it; case TCP_TW_SUCCESS:; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 017f6293b5f4..317d7a6e6b01 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1680,7 +1680,7 @@ int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; reset: - tcp_v6_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + tcp_v6_send_reset(sk, skb, sk_rst_convert_drop_reason(reason)); discard: if (opt_skb) __kfree_skb(opt_skb); @@ -1865,7 +1865,10 @@ lookup: } else { drop_reason = tcp_child_process(sk, nsk, skb); if (drop_reason) { - tcp_v6_send_reset(nsk, skb, SK_RST_REASON_NOT_SPECIFIED); + enum sk_rst_reason rst_reason; + + rst_reason = sk_rst_convert_drop_reason(drop_reason); + tcp_v6_send_reset(nsk, skb, rst_reason); goto discard_and_relse; } sock_put(sk); @@ -1942,7 +1945,7 @@ csum_error: bad_packet: __TCP_INC_STATS(net, TCP_MIB_INERRS); } else { - tcp_v6_send_reset(NULL, skb, SK_RST_REASON_NOT_SPECIFIED); + tcp_v6_send_reset(NULL, skb, sk_rst_convert_drop_reason(drop_reason)); } discard_it: @@ -1998,7 +2001,7 @@ do_time_wait: tcp_v6_timewait_ack(sk, skb); break; case TCP_TW_RST: - tcp_v6_send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + tcp_v6_send_reset(sk, skb, sk_rst_convert_drop_reason(drop_reason)); inet_twsk_deschedule_put(inet_twsk(sk)); goto discard_it; case TCP_TW_SUCCESS: -- cgit v1.2.3 From 3e140491dd80d8643261a21efde3ce2ff6fb9fdf Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 25 Apr 2024 11:13:38 +0800 Subject: mptcp: support rstreason for passive reset It relies on what reset options in the skb are as rfc8684 says. Reusing this logic can save us much energy. This patch replaces most of the prior NOT_SPECIFIED reasons. Signed-off-by: Jason Xing Reviewed-by: Matthieu Baerts (NGI0) Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- net/mptcp/protocol.h | 27 +++++++++++++++++++++++++++ net/mptcp/subflow.c | 22 +++++++++++++++++----- 2 files changed, 44 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index fdfa843e2d88..252618859ee8 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -581,6 +581,33 @@ mptcp_subflow_ctx_reset(struct mptcp_subflow_context *subflow) WRITE_ONCE(subflow->local_id, -1); } +/* Convert reset reasons in MPTCP to enum sk_rst_reason type */ +static inline enum sk_rst_reason +sk_rst_convert_mptcp_reason(u32 reason) +{ + switch (reason) { + case MPTCP_RST_EUNSPEC: + return SK_RST_REASON_MPTCP_RST_EUNSPEC; + case MPTCP_RST_EMPTCP: + return SK_RST_REASON_MPTCP_RST_EMPTCP; + case MPTCP_RST_ERESOURCE: + return SK_RST_REASON_MPTCP_RST_ERESOURCE; + case MPTCP_RST_EPROHIBIT: + return SK_RST_REASON_MPTCP_RST_EPROHIBIT; + case MPTCP_RST_EWQ2BIG: + return SK_RST_REASON_MPTCP_RST_EWQ2BIG; + case MPTCP_RST_EBADPERF: + return SK_RST_REASON_MPTCP_RST_EBADPERF; + case MPTCP_RST_EMIDDLEBOX: + return SK_RST_REASON_MPTCP_RST_EMIDDLEBOX; + default: + /* It should not happen, or else errors may occur + * in MPTCP layer + */ + return SK_RST_REASON_ERROR; + } +} + static inline u64 mptcp_subflow_get_map_offset(const struct mptcp_subflow_context *subflow) { diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index ac867d277860..fb7abf2d01ca 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -309,8 +309,13 @@ static struct dst_entry *subflow_v4_route_req(const struct sock *sk, return dst; dst_release(dst); - if (!req->syncookie) - tcp_request_sock_ops.send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + if (!req->syncookie) { + struct mptcp_ext *mpext = mptcp_get_ext(skb); + enum sk_rst_reason reason; + + reason = sk_rst_convert_mptcp_reason(mpext->reset_reason); + tcp_request_sock_ops.send_reset(sk, skb, reason); + } return NULL; } @@ -377,8 +382,13 @@ static struct dst_entry *subflow_v6_route_req(const struct sock *sk, return dst; dst_release(dst); - if (!req->syncookie) - tcp6_request_sock_ops.send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + if (!req->syncookie) { + struct mptcp_ext *mpext = mptcp_get_ext(skb); + enum sk_rst_reason reason; + + reason = sk_rst_convert_mptcp_reason(mpext->reset_reason); + tcp6_request_sock_ops.send_reset(sk, skb, reason); + } return NULL; } #endif @@ -783,6 +793,7 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk, struct mptcp_subflow_request_sock *subflow_req; struct mptcp_options_received mp_opt; bool fallback, fallback_is_fatal; + enum sk_rst_reason reason; struct mptcp_sock *owner; struct sock *child; @@ -913,7 +924,8 @@ dispose_child: tcp_rsk(req)->drop_req = true; inet_csk_prepare_for_destroy_sock(child); tcp_done(child); - req->rsk_ops->send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + reason = sk_rst_convert_mptcp_reason(mptcp_get_ext(skb)->reset_reason); + req->rsk_ops->send_reset(sk, skb, reason); /* The last child reference will be released by the caller */ return child; -- cgit v1.2.3 From 215d40248bde5562a21e4c6cdeaeca0495c9365a Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 25 Apr 2024 11:13:39 +0800 Subject: mptcp: introducing a helper into active reset logic Since we have mapped every mptcp reset reason definition in enum sk_rst_reason, introducing a new helper can cover some missing places where we have already set the subflow->reset_reason. Note: using SK_RST_REASON_NOT_SPECIFIED is the same as SK_RST_REASON_MPTCP_RST_EUNSPEC. They are both unknown. So we can convert it directly. Suggested-by: Paolo Abeni Signed-off-by: Jason Xing Reviewed-by: Matthieu Baerts (NGI0) Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- net/mptcp/protocol.c | 4 +--- net/mptcp/protocol.h | 11 +++++++++++ net/mptcp/subflow.c | 6 ++---- 3 files changed, 14 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 065967086492..4b13ca362efa 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -21,7 +21,6 @@ #endif #include #include -#include #include #include "protocol.h" #include "mib.h" @@ -2570,8 +2569,7 @@ static void mptcp_check_fastclose(struct mptcp_sock *msk) slow = lock_sock_fast(tcp_sk); if (tcp_sk->sk_state != TCP_CLOSE) { - tcp_send_active_reset(tcp_sk, GFP_ATOMIC, - SK_RST_REASON_NOT_SPECIFIED); + mptcp_send_active_reset_reason(tcp_sk); tcp_set_state(tcp_sk, TCP_CLOSE); } unlock_sock_fast(tcp_sk, slow); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 252618859ee8..cfc5f9c3f113 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -12,6 +12,7 @@ #include #include #include +#include #include "mptcp_pm_gen.h" @@ -608,6 +609,16 @@ sk_rst_convert_mptcp_reason(u32 reason) } } +static inline void +mptcp_send_active_reset_reason(struct sock *sk) +{ + struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); + enum sk_rst_reason reason; + + reason = sk_rst_convert_mptcp_reason(subflow->reset_reason); + tcp_send_active_reset(sk, GFP_ATOMIC, reason); +} + static inline u64 mptcp_subflow_get_map_offset(const struct mptcp_subflow_context *subflow) { diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index fb7abf2d01ca..97ec44d1df30 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -20,7 +20,6 @@ #include #endif #include -#include #include "protocol.h" #include "mib.h" @@ -424,7 +423,7 @@ void mptcp_subflow_reset(struct sock *ssk) /* must hold: tcp_done() could drop last reference on parent */ sock_hold(sk); - tcp_send_active_reset(ssk, GFP_ATOMIC, SK_RST_REASON_NOT_SPECIFIED); + mptcp_send_active_reset_reason(ssk); tcp_done(ssk); if (!test_and_set_bit(MPTCP_WORK_CLOSE_SUBFLOW, &mptcp_sk(sk)->flags)) mptcp_schedule_work(sk); @@ -1362,8 +1361,7 @@ reset: tcp_set_state(ssk, TCP_CLOSE); while ((skb = skb_peek(&ssk->sk_receive_queue))) sk_eat_skb(ssk, skb); - tcp_send_active_reset(ssk, GFP_ATOMIC, - SK_RST_REASON_NOT_SPECIFIED); + mptcp_send_active_reset_reason(ssk); WRITE_ONCE(subflow->data_avail, false); return false; } -- cgit v1.2.3 From b533fb9cf4f7c6ca2aa255a5a1fdcde49fff2b24 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Thu, 25 Apr 2024 11:13:40 +0800 Subject: rstreason: make it work in trace world At last, we should let it work by introducing this reset reason in trace world. One of the possible expected outputs is: ... tcp_send_reset: skbaddr=xxx skaddr=xxx src=xxx dest=xxx state=TCP_ESTABLISHED reason=NOT_SPECIFIED Signed-off-by: Jason Xing Reviewed-by: Steven Rostedt (Google) Reviewed-by: Eric Dumazet Signed-off-by: Paolo Abeni --- include/trace/events/tcp.h | 26 ++++++++++++++++++++++---- net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/tcp_output.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 4 files changed, 25 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h index 5c04a61a11c2..49b5ee091cf6 100644 --- a/include/trace/events/tcp.h +++ b/include/trace/events/tcp.h @@ -11,6 +11,7 @@ #include #include #include +#include /* * tcp event with arguments sk and skb @@ -74,20 +75,32 @@ DEFINE_EVENT(tcp_event_sk_skb, tcp_retransmit_skb, TP_ARGS(sk, skb) ); +#undef FN +#define FN(reason) TRACE_DEFINE_ENUM(SK_RST_REASON_##reason); +DEFINE_RST_REASON(FN, FN) + +#undef FN +#undef FNe +#define FN(reason) { SK_RST_REASON_##reason, #reason }, +#define FNe(reason) { SK_RST_REASON_##reason, #reason } + /* * skb of trace_tcp_send_reset is the skb that caused RST. In case of * active reset, skb should be NULL */ TRACE_EVENT(tcp_send_reset, - TP_PROTO(const struct sock *sk, const struct sk_buff *skb), + TP_PROTO(const struct sock *sk, + const struct sk_buff *skb, + const enum sk_rst_reason reason), - TP_ARGS(sk, skb), + TP_ARGS(sk, skb, reason), TP_STRUCT__entry( __field(const void *, skbaddr) __field(const void *, skaddr) __field(int, state) + __field(enum sk_rst_reason, reason) __array(__u8, saddr, sizeof(struct sockaddr_in6)) __array(__u8, daddr, sizeof(struct sockaddr_in6)) ), @@ -113,14 +126,19 @@ TRACE_EVENT(tcp_send_reset, */ TP_STORE_ADDR_PORTS_SKB(skb, th, entry->daddr, entry->saddr); } + __entry->reason = reason; ), - TP_printk("skbaddr=%p skaddr=%p src=%pISpc dest=%pISpc state=%s", + TP_printk("skbaddr=%p skaddr=%p src=%pISpc dest=%pISpc state=%s reason=%s", __entry->skbaddr, __entry->skaddr, __entry->saddr, __entry->daddr, - __entry->state ? show_tcp_state_name(__entry->state) : "UNKNOWN") + __entry->state ? show_tcp_state_name(__entry->state) : "UNKNOWN", + __print_symbolic(__entry->reason, DEFINE_RST_REASON(FN, FNe))) ); +#undef FN +#undef FNe + /* * tcp event with arguments sk * diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 621879ba916d..0427deca3e0e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -871,7 +871,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb, if (sk) arg.bound_dev_if = sk->sk_bound_dev_if; - trace_tcp_send_reset(sk, skb); + trace_tcp_send_reset(sk, skb, reason); BUILD_BUG_ON(offsetof(struct sock, sk_bound_dev_if) != offsetof(struct inet_timewait_sock, tw_bound_dev_if)); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 41c352bf3394..ece4726c8751 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3641,7 +3641,7 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority, /* skb of trace_tcp_send_reset() keeps the skb that caused RST, * skb here is different to the troublesome skb, so use NULL */ - trace_tcp_send_reset(sk, NULL); + trace_tcp_send_reset(sk, NULL, SK_RST_REASON_NOT_SPECIFIED); } /* Send a crossed SYN-ACK during socket establishment. diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 317d7a6e6b01..77958adf2e16 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1133,7 +1133,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb, label = ip6_flowlabel(ipv6h); } - trace_tcp_send_reset(sk, skb); + trace_tcp_send_reset(sk, skb, reason); tcp_v6_send_response(sk, skb, seq, ack_seq, 0, 0, 0, oif, 1, ipv6_get_dsfield(ipv6h), label, priority, txhash, -- cgit v1.2.3 From 1bede0a12d3a45bd366d3cf9e1c7611d86f1bc1f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 25 Apr 2024 19:34:50 +0000 Subject: tcp: fix tcp_grow_skb() vs tstamps I forgot to call tcp_skb_collapse_tstamp() in the case we consume the second skb in write queue. Neal suggested to create a common helper used by tcp_mtu_probe() and tcp_grow_skb(). Fixes: 8ee602c63520 ("tcp: try to send bigger TSO packets") Signed-off-by: Eric Dumazet Acked-by: Neal Cardwell Link: https://lore.kernel.org/r/20240425193450.411640-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_output.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index ece4726c8751..ea7ad7d99245 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2403,6 +2403,21 @@ commit: return 0; } +/* tcp_mtu_probe() and tcp_grow_skb() can both eat an skb (src) if + * all its payload was moved to another one (dst). + * Make sure to transfer tcp_flags, eor, and tstamp. + */ +static void tcp_eat_one_skb(struct sock *sk, + struct sk_buff *dst, + struct sk_buff *src) +{ + TCP_SKB_CB(dst)->tcp_flags |= TCP_SKB_CB(src)->tcp_flags; + TCP_SKB_CB(dst)->eor = TCP_SKB_CB(src)->eor; + tcp_skb_collapse_tstamp(dst, src); + tcp_unlink_write_queue(src, sk); + tcp_wmem_free_skb(sk, src); +} + /* Create a new MTU probe if we are ready. * MTU probe is regularly attempting to increase the path MTU by * deliberately sending larger packets. This discovers routing @@ -2508,16 +2523,7 @@ static int tcp_mtu_probe(struct sock *sk) copy = min_t(int, skb->len, probe_size - len); if (skb->len <= copy) { - /* We've eaten all the data from this skb. - * Throw it away. */ - TCP_SKB_CB(nskb)->tcp_flags |= TCP_SKB_CB(skb)->tcp_flags; - /* If this is the last SKB we copy and eor is set - * we need to propagate it to the new skb. - */ - TCP_SKB_CB(nskb)->eor = TCP_SKB_CB(skb)->eor; - tcp_skb_collapse_tstamp(nskb, skb); - tcp_unlink_write_queue(skb, sk); - tcp_wmem_free_skb(sk, skb); + tcp_eat_one_skb(sk, nskb, skb); } else { TCP_SKB_CB(nskb)->tcp_flags |= TCP_SKB_CB(skb)->tcp_flags & ~(TCPHDR_FIN|TCPHDR_PSH); @@ -2705,11 +2711,10 @@ static void tcp_grow_skb(struct sock *sk, struct sk_buff *skb, int amount) TCP_SKB_CB(next_skb)->seq += nlen; if (!next_skb->len) { + /* In case FIN is set, we need to update end_seq */ TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(next_skb)->end_seq; - TCP_SKB_CB(skb)->eor = TCP_SKB_CB(next_skb)->eor; - TCP_SKB_CB(skb)->tcp_flags |= TCP_SKB_CB(next_skb)->tcp_flags; - tcp_unlink_write_queue(next_skb, sk); - tcp_wmem_free_skb(sk, next_skb); + + tcp_eat_one_skb(sk, skb, next_skb); } } -- cgit v1.2.3 From cd42ba1c8ac9deb9032add6adf491110e7442040 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 26 Apr 2024 06:42:22 +0000 Subject: net: give more chances to rcu in netdev_wait_allrefs_any() This came while reviewing commit c4e86b4363ac ("net: add two more call_rcu_hurry()"). Paolo asked if adding one synchronize_rcu() would help. While synchronize_rcu() does not help, making sure to call rcu_barrier() before msleep(wait) is definitely helping to make sure lazy call_rcu() are completed. Instead of waiting ~100 seconds in my tests, the ref_tracker splats occurs one time only, and netdev_wait_allrefs_any() latency is reduced to the strict minimum. Ideally we should audit our call_rcu() users to make sure no refcount (or cascading call_rcu()) is held too long, because rcu_barrier() is quite expensive. Fixes: 0e4be9e57e8c ("net: use exponential backoff in netdev_wait_allrefs") Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/all/28bbf698-befb-42f6-b561-851c67f464aa@kernel.org/T/#m76d73ed6b03cd930778ac4d20a777f22a08d6824 Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- net/core/dev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index e09aa3785c15..c9e59eff8ec8 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -10566,8 +10566,9 @@ static struct net_device *netdev_wait_allrefs_any(struct list_head *list) rebroadcast_time = jiffies; } + rcu_barrier(); + if (!wait) { - rcu_barrier(); wait = WAIT_REFS_MIN_MSECS; } else { msleep(wait); -- cgit v1.2.3 From 61f5338d62673379ff2ce5a8c05682658a196980 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 26 Apr 2024 07:02:02 +0000 Subject: inet: use call_rcu_hurry() in inet_free_ifa() This is a followup of commit c4e86b4363ac ("net: add two more call_rcu_hurry()") Our reference to ifa->ifa_dev must be freed ASAP to release the reference to the netdev the same way. inet_rcu_free_ifa() in_dev_put() -> in_dev_finish_destroy() -> netdev_put() This should speedup device/netns dismantles when CONFIG_RCU_LAZY=y Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv4/devinet.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 7592f242336b..364dbf0cd9bf 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -224,6 +224,7 @@ static struct in_ifaddr *inet_alloc_ifa(void) static void inet_rcu_free_ifa(struct rcu_head *head) { struct in_ifaddr *ifa = container_of(head, struct in_ifaddr, rcu_head); + if (ifa->ifa_dev) in_dev_put(ifa->ifa_dev); kfree(ifa); @@ -231,7 +232,11 @@ static void inet_rcu_free_ifa(struct rcu_head *head) static void inet_free_ifa(struct in_ifaddr *ifa) { - call_rcu(&ifa->rcu_head, inet_rcu_free_ifa); + /* Our reference to ifa->ifa_dev must be freed ASAP + * to release the reference to the netdev the same way. + * in_dev_put() -> in_dev_finish_destroy() -> netdev_put() + */ + call_rcu_hurry(&ifa->rcu_head, inet_rcu_free_ifa); } static void in_dev_free_rcu(struct rcu_head *head) -- cgit v1.2.3 From e8dfd42c17faf183415323db1ef0c977be0d6489 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 26 Apr 2024 15:19:52 +0000 Subject: ipv6: introduce dst_rt6_info() helper Instead of (struct rt6_info *)dst casts, we can use : #define dst_rt6_info(_ptr) \ container_of_const(_ptr, struct rt6_info, dst) Some places needed missing const qualifiers : ip6_confirm_neigh(), ipv6_anycast_destination(), ipv6_unicast_destination(), has_gateway() v2: added missing parts (David Ahern) Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Signed-off-by: David S. Miller --- drivers/infiniband/core/addr.c | 6 ++--- .../net/ethernet/mellanox/mlxsw/spectrum_span.c | 2 +- drivers/net/vrf.c | 2 +- drivers/net/vxlan/vxlan_core.c | 2 +- drivers/s390/net/qeth_core.h | 4 ++-- include/net/ip6_fib.h | 6 +++-- include/net/ip6_route.h | 11 ++++----- net/bluetooth/6lowpan.c | 2 +- net/core/dst_cache.c | 2 +- net/core/filter.c | 2 +- net/ipv4/ip_tunnel.c | 2 +- net/ipv6/icmp.c | 8 +++---- net/ipv6/ila/ila_lwt.c | 4 ++-- net/ipv6/ip6_output.c | 18 +++++++------- net/ipv6/ip6mr.c | 2 +- net/ipv6/ndisc.c | 2 +- net/ipv6/ping.c | 2 +- net/ipv6/raw.c | 4 ++-- net/ipv6/route.c | 28 +++++++++++----------- net/ipv6/tcp_ipv6.c | 4 +--- net/ipv6/udp.c | 11 ++++----- net/ipv6/xfrm6_policy.c | 2 +- net/l2tp/l2tp_ip6.c | 2 +- net/mpls/mpls_iptunnel.c | 2 +- net/netfilter/ipvs/ip_vs_xmit.c | 14 +++++------ net/netfilter/nf_flow_table_core.c | 8 ++----- net/netfilter/nf_flow_table_ip.c | 4 ++-- net/netfilter/nft_rt.c | 2 +- net/sctp/ipv6.c | 2 +- net/xfrm/xfrm_policy.c | 3 +-- 30 files changed, 77 insertions(+), 86 deletions(-) (limited to 'net') diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index f253295795f0..f20dfe70fa0e 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -348,15 +348,15 @@ static int dst_fetch_ha(const struct dst_entry *dst, static bool has_gateway(const struct dst_entry *dst, sa_family_t family) { - struct rtable *rt; - struct rt6_info *rt6; + const struct rtable *rt; + const struct rt6_info *rt6; if (family == AF_INET) { rt = container_of(dst, struct rtable, dst); return rt->rt_uses_gateway; } - rt6 = container_of(dst, struct rt6_info, dst); + rt6 = dst_rt6_info(dst); return rt6->rt6i_flags & RTF_GATEWAY; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index 3de69b2eb5c4..4b5fd71c897d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -540,7 +540,7 @@ mlxsw_sp_span_gretap6_route(const struct net_device *to_dev, if (!dst || dst->error) goto out; - rt6 = container_of(dst, struct rt6_info, dst); + rt6 = dst_rt6_info(dst); dev = dst->dev; *saddrp = fl6.saddr; diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 66f8542f3b18..784b9b2d275e 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -653,7 +653,7 @@ static int vrf_finish_output6(struct net *net, struct sock *sk, skb->dev = dev; rcu_read_lock(); - nexthop = rt6_nexthop((struct rt6_info *)dst, &ipv6_hdr(skb)->daddr); + nexthop = rt6_nexthop(dst_rt6_info(dst), &ipv6_hdr(skb)->daddr); neigh = __ipv6_neigh_lookup_noref(dst->dev, nexthop); if (unlikely(!neigh)) neigh = __neigh_create(&nd_tbl, nexthop, dst->dev, false); diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index 4bbfb516ac05..a9a85d0f5a5d 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -2513,7 +2513,7 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, } if (!info) { - u32 rt6i_flags = ((struct rt6_info *)ndst)->rt6i_flags; + u32 rt6i_flags = dst_rt6_info(ndst)->rt6i_flags; err = encap_bypass_if_local(skb, dev, vxlan, AF_INET6, dst_port, ifindex, vni, diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index 613eab729704..5f17a2a5d0e3 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -956,7 +956,7 @@ static inline struct dst_entry *qeth_dst_check_rcu(struct sk_buff *skb, struct dst_entry *dst = skb_dst(skb); struct rt6_info *rt; - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); if (dst) { if (proto == htons(ETH_P_IPV6)) dst = dst_check(dst, rt6_get_cookie(rt)); @@ -978,7 +978,7 @@ static inline __be32 qeth_next_hop_v4_rcu(struct sk_buff *skb, static inline struct in6_addr *qeth_next_hop_v6_rcu(struct sk_buff *skb, struct dst_entry *dst) { - struct rt6_info *rt = (struct rt6_info *) dst; + struct rt6_info *rt = dst_rt6_info(dst); if (rt && !ipv6_addr_any(&rt->rt6i_gateway)) return &rt->rt6i_gateway; diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 7834f7f29d3c..6cb867ce4878 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -234,9 +234,11 @@ struct fib6_result { for (rt = (w)->leaf; rt; \ rt = rcu_dereference_protected(rt->fib6_next, 1)) -static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) +#define dst_rt6_info(_ptr) container_of_const(_ptr, struct rt6_info, dst) + +static inline struct inet6_dev *ip6_dst_idev(const struct dst_entry *dst) { - return ((struct rt6_info *)dst)->rt6i_idev; + return dst_rt6_info(dst)->rt6i_idev; } static inline bool fib6_requires_src(const struct fib6_info *rt) diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index a30c6aa9e5cf..a18ed24fed94 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -210,12 +210,11 @@ void rt6_uncached_list_del(struct rt6_info *rt); static inline const struct rt6_info *skb_rt6_info(const struct sk_buff *skb) { const struct dst_entry *dst = skb_dst(skb); - const struct rt6_info *rt6 = NULL; if (dst) - rt6 = container_of(dst, struct rt6_info, dst); + return dst_rt6_info(dst); - return rt6; + return NULL; } /* @@ -227,7 +226,7 @@ static inline void ip6_dst_store(struct sock *sk, struct dst_entry *dst, { struct ipv6_pinfo *np = inet6_sk(sk); - np->dst_cookie = rt6_get_cookie((struct rt6_info *)dst); + np->dst_cookie = rt6_get_cookie(dst_rt6_info(dst)); sk_setup_caps(sk, dst); np->daddr_cache = daddr; #ifdef CONFIG_IPV6_SUBTREES @@ -240,7 +239,7 @@ void ip6_sk_dst_store_flow(struct sock *sk, struct dst_entry *dst, static inline bool ipv6_unicast_destination(const struct sk_buff *skb) { - struct rt6_info *rt = (struct rt6_info *) skb_dst(skb); + const struct rt6_info *rt = dst_rt6_info(skb_dst(skb)); return rt->rt6i_flags & RTF_LOCAL; } @@ -248,7 +247,7 @@ static inline bool ipv6_unicast_destination(const struct sk_buff *skb) static inline bool ipv6_anycast_destination(const struct dst_entry *dst, const struct in6_addr *daddr) { - struct rt6_info *rt = (struct rt6_info *)dst; + const struct rt6_info *rt = dst_rt6_info(dst); return rt->rt6i_flags & RTF_ANYCAST || (rt->rt6i_dst.plen < 127 && diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 27520a8a486f..50cfec8ccac4 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -133,7 +133,7 @@ static inline struct lowpan_peer *peer_lookup_dst(struct lowpan_btle_dev *dev, struct in6_addr *daddr, struct sk_buff *skb) { - struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); + struct rt6_info *rt = dst_rt6_info(skb_dst(skb)); int count = atomic_read(&dev->peer_count); const struct in6_addr *nexthop; struct lowpan_peer *peer; diff --git a/net/core/dst_cache.c b/net/core/dst_cache.c index 0ccfd5fa5cb9..b17171345d64 100644 --- a/net/core/dst_cache.c +++ b/net/core/dst_cache.c @@ -112,7 +112,7 @@ void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst, idst = this_cpu_ptr(dst_cache->cache); dst_cache_per_cpu_dst_set(this_cpu_ptr(dst_cache->cache), dst, - rt6_get_cookie((struct rt6_info *)dst)); + rt6_get_cookie(dst_rt6_info(dst))); idst->in6_saddr = *saddr; } EXPORT_SYMBOL_GPL(dst_cache_set_ip6); diff --git a/net/core/filter.c b/net/core/filter.c index 294670d3850d..5662464e1abd 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2215,7 +2215,7 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb, rcu_read_lock(); if (!nh) { dst = skb_dst(skb); - nexthop = rt6_nexthop(container_of(dst, struct rt6_info, dst), + nexthop = rt6_nexthop(dst_rt6_info(dst), &ipv6_hdr(skb)->daddr); } else { nexthop = &nh->ipv6_nh; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 177f40c3a8e8..ba46cf7612f4 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -543,7 +543,7 @@ static int tnl_update_pmtu(struct net_device *dev, struct sk_buff *skb, struct rt6_info *rt6; __be32 daddr; - rt6 = skb_valid_dst(skb) ? (struct rt6_info *)skb_dst(skb) : + rt6 = skb_valid_dst(skb) ? dst_rt6_info(skb_dst(skb)) : NULL; daddr = md ? dst : tunnel->parms.iph.daddr; diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 1635da07285f..d285c1f6f1a6 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -212,7 +212,7 @@ static bool icmpv6_xrlim_allow(struct sock *sk, u8 type, } else if (dst->dev && (dst->dev->flags&IFF_LOOPBACK)) { res = true; } else { - struct rt6_info *rt = (struct rt6_info *)dst; + struct rt6_info *rt = dst_rt6_info(dst); int tmo = net->ipv6.sysctl.icmpv6_time; struct inet_peer *peer; @@ -241,7 +241,7 @@ static bool icmpv6_rt_has_prefsrc(struct sock *sk, u8 type, dst = ip6_route_output(net, sk, fl6); if (!dst->error) { - struct rt6_info *rt = (struct rt6_info *)dst; + struct rt6_info *rt = dst_rt6_info(dst); struct in6_addr prefsrc; rt6_get_prefsrc(rt, &prefsrc); @@ -616,7 +616,7 @@ void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, if (ip6_append_data(sk, icmpv6_getfrag, &msg, len + sizeof(struct icmp6hdr), sizeof(struct icmp6hdr), - &ipc6, &fl6, (struct rt6_info *)dst, + &ipc6, &fl6, dst_rt6_info(dst), MSG_DONTWAIT)) { ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS); ip6_flush_pending_frames(sk); @@ -803,7 +803,7 @@ static enum skb_drop_reason icmpv6_echo_reply(struct sk_buff *skb) if (ip6_append_data(sk, icmpv6_getfrag, &msg, skb->len + sizeof(struct icmp6hdr), sizeof(struct icmp6hdr), &ipc6, &fl6, - (struct rt6_info *)dst, MSG_DONTWAIT)) { + dst_rt6_info(dst), MSG_DONTWAIT)) { __ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS); ip6_flush_pending_frames(sk); } else { diff --git a/net/ipv6/ila/ila_lwt.c b/net/ipv6/ila/ila_lwt.c index 8c1ce78956ba..0601bad79822 100644 --- a/net/ipv6/ila/ila_lwt.c +++ b/net/ipv6/ila/ila_lwt.c @@ -38,7 +38,7 @@ static inline struct ila_params *ila_params_lwtunnel( static int ila_output(struct net *net, struct sock *sk, struct sk_buff *skb) { struct dst_entry *orig_dst = skb_dst(skb); - struct rt6_info *rt = (struct rt6_info *)orig_dst; + struct rt6_info *rt = dst_rt6_info(orig_dst); struct ila_lwt *ilwt = ila_lwt_lwtunnel(orig_dst->lwtstate); struct dst_entry *dst; int err = -EINVAL; @@ -70,7 +70,7 @@ static int ila_output(struct net *net, struct sock *sk, struct sk_buff *skb) memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_oif = orig_dst->dev->ifindex; fl6.flowi6_iif = LOOPBACK_IFINDEX; - fl6.daddr = *rt6_nexthop((struct rt6_info *)orig_dst, + fl6.daddr = *rt6_nexthop(dst_rt6_info(orig_dst), &ip6h->daddr); dst = ip6_route_output(net, NULL, &fl6); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index b9dd3a66e423..8f906e9fbc38 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -120,7 +120,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len); rcu_read_lock(); - nexthop = rt6_nexthop((struct rt6_info *)dst, daddr); + nexthop = rt6_nexthop(dst_rt6_info(dst), daddr); neigh = __ipv6_neigh_lookup_noref(dev, nexthop); if (unlikely(IS_ERR_OR_NULL(neigh))) { @@ -599,7 +599,7 @@ int ip6_forward(struct sk_buff *skb) * send a redirect. */ - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); if (rt->rt6i_flags & RTF_GATEWAY) target = &rt->rt6i_gateway; else @@ -856,7 +856,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, int (*output)(struct net *, struct sock *, struct sk_buff *)) { struct sk_buff *frag; - struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); + struct rt6_info *rt = dst_rt6_info(skb_dst(skb)); struct ipv6_pinfo *np = skb->sk && !dev_recursion_level() ? inet6_sk(skb->sk) : NULL; bool mono_delivery_time = skb->mono_delivery_time; @@ -1063,7 +1063,7 @@ static struct dst_entry *ip6_sk_dst_check(struct sock *sk, return NULL; } - rt = (struct rt6_info *)dst; + rt = dst_rt6_info(dst); /* Yes, checking route validity in not connected * case is not very simple. Take into account, * that we do not support routing by source, TOS, @@ -1118,7 +1118,7 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk, struct rt6_info *rt; *dst = ip6_route_output(net, sk, fl6); - rt = (*dst)->error ? NULL : (struct rt6_info *)*dst; + rt = (*dst)->error ? NULL : dst_rt6_info(*dst); rcu_read_lock(); from = rt ? rcu_dereference(rt->from) : NULL; @@ -1159,7 +1159,7 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk, * dst entry and replace it instead with the * dst entry of the nexthop router */ - rt = (struct rt6_info *) *dst; + rt = dst_rt6_info(*dst); rcu_read_lock(); n = __ipv6_neigh_lookup_noref(rt->dst.dev, rt6_nexthop(rt, &fl6->daddr)); @@ -1423,7 +1423,7 @@ static int __ip6_append_data(struct sock *sk, int offset = 0; bool zc = false; u32 tskey = 0; - struct rt6_info *rt = (struct rt6_info *)cork->dst; + struct rt6_info *rt = dst_rt6_info(cork->dst); bool paged, hold_tskey, extra_uref = false; struct ipv6_txoptions *opt = v6_cork->opt; int csummode = CHECKSUM_NONE; @@ -1877,7 +1877,7 @@ struct sk_buff *__ip6_make_skb(struct sock *sk, struct net *net = sock_net(sk); struct ipv6hdr *hdr; struct ipv6_txoptions *opt = v6_cork->opt; - struct rt6_info *rt = (struct rt6_info *)cork->base.dst; + struct rt6_info *rt = dst_rt6_info(cork->base.dst); struct flowi6 *fl6 = &cork->fl.u.ip6; unsigned char proto = fl6->flowi6_proto; @@ -1949,7 +1949,7 @@ out: int ip6_send_skb(struct sk_buff *skb) { struct net *net = sock_net(skb->sk); - struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); + struct rt6_info *rt = dst_rt6_info(skb_dst(skb)); int err; err = ip6_local_out(net, skb->sk, skb); diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index cb0ee81a068a..dd342e6ecf3f 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -2273,7 +2273,7 @@ int ip6mr_get_route(struct net *net, struct sk_buff *skb, struct rtmsg *rtm, int err; struct mr_table *mrt; struct mfc6_cache *cache; - struct rt6_info *rt = (struct rt6_info *)skb_dst(skb); + struct rt6_info *rt = dst_rt6_info(skb_dst(skb)); mrt = ip6mr_get_table(net, RT6_TABLE_DFLT); if (!mrt) diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index ae134634c323..d914b23256ce 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1722,7 +1722,7 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) if (IS_ERR(dst)) return; - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); if (rt->rt6i_flags & RTF_GATEWAY) { ND_PRINTK(2, warn, diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index ef2059c88955..88b3fcacd4f9 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -154,7 +154,7 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) dst = ip6_sk_dst_lookup_flow(sk, &fl6, daddr, false); if (IS_ERR(dst)) return PTR_ERR(dst); - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr)) fl6.flowi6_oif = READ_ONCE(np->mcast_oif); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 0d896ca7b589..2eedf255600b 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -598,7 +598,7 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length, struct ipv6hdr *iph; struct sk_buff *skb; int err; - struct rt6_info *rt = (struct rt6_info *)*dstp; + struct rt6_info *rt = dst_rt6_info(*dstp); int hlen = LL_RESERVED_SPACE(rt->dst.dev); int tlen = rt->dst.dev->needed_tailroom; @@ -917,7 +917,7 @@ back_from_confirm: ipc6.opt = opt; lock_sock(sk); err = ip6_append_data(sk, raw6_getfrag, &rfv, - len, 0, &ipc6, &fl6, (struct rt6_info *)dst, + len, 0, &ipc6, &fl6, dst_rt6_info(dst), msg->msg_flags); if (err) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 1f4b935a0e57..3e0b2cb20fd2 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -226,7 +226,7 @@ static struct neighbour *ip6_dst_neigh_lookup(const struct dst_entry *dst, struct sk_buff *skb, const void *daddr) { - const struct rt6_info *rt = container_of(dst, struct rt6_info, dst); + const struct rt6_info *rt = dst_rt6_info(dst); return ip6_neigh_lookup(rt6_nexthop(rt, &in6addr_any), dst->dev, skb, daddr); @@ -234,8 +234,8 @@ static struct neighbour *ip6_dst_neigh_lookup(const struct dst_entry *dst, static void ip6_confirm_neigh(const struct dst_entry *dst, const void *daddr) { + const struct rt6_info *rt = dst_rt6_info(dst); struct net_device *dev = dst->dev; - struct rt6_info *rt = (struct rt6_info *)dst; daddr = choose_neigh_daddr(rt6_nexthop(rt, &in6addr_any), NULL, daddr); if (!daddr) @@ -354,7 +354,7 @@ EXPORT_SYMBOL(ip6_dst_alloc); static void ip6_dst_destroy(struct dst_entry *dst) { - struct rt6_info *rt = (struct rt6_info *)dst; + struct rt6_info *rt = dst_rt6_info(dst); struct fib6_info *from; struct inet6_dev *idev; @@ -373,7 +373,7 @@ static void ip6_dst_destroy(struct dst_entry *dst) static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev) { - struct rt6_info *rt = (struct rt6_info *)dst; + struct rt6_info *rt = dst_rt6_info(dst); struct inet6_dev *idev = rt->rt6i_idev; if (idev && idev->dev != blackhole_netdev) { @@ -1288,7 +1288,7 @@ struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr, dst = fib6_rule_lookup(net, &fl6, skb, flags, ip6_pol_route_lookup); if (dst->error == 0) - return (struct rt6_info *) dst; + return dst_rt6_info(dst); dst_release(dst); @@ -2647,7 +2647,7 @@ struct dst_entry *ip6_route_output_flags(struct net *net, rcu_read_lock(); dst = ip6_route_output_flags_noref(net, sk, fl6, flags); - rt6 = (struct rt6_info *)dst; + rt6 = dst_rt6_info(dst); /* For dst cached in uncached_list, refcnt is already taken. */ if (list_empty(&rt6->dst.rt_uncached) && !dst_hold_safe(dst)) { dst = &net->ipv6.ip6_null_entry->dst; @@ -2661,7 +2661,7 @@ EXPORT_SYMBOL_GPL(ip6_route_output_flags); struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_orig) { - struct rt6_info *rt, *ort = (struct rt6_info *) dst_orig; + struct rt6_info *rt, *ort = dst_rt6_info(dst_orig); struct net_device *loopback_dev = net->loopback_dev; struct dst_entry *new = NULL; @@ -2744,7 +2744,7 @@ INDIRECT_CALLABLE_SCOPE struct dst_entry *ip6_dst_check(struct dst_entry *dst, struct fib6_info *from; struct rt6_info *rt; - rt = container_of(dst, struct rt6_info, dst); + rt = dst_rt6_info(dst); if (rt->sernum) return rt6_is_valid(rt) ? dst : NULL; @@ -2772,7 +2772,7 @@ EXPORT_INDIRECT_CALLABLE(ip6_dst_check); static struct dst_entry *ip6_negative_advice(struct dst_entry *dst) { - struct rt6_info *rt = (struct rt6_info *) dst; + struct rt6_info *rt = dst_rt6_info(dst); if (rt) { if (rt->rt6i_flags & RTF_CACHE) { @@ -2796,7 +2796,7 @@ static void ip6_link_failure(struct sk_buff *skb) icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); - rt = (struct rt6_info *) skb_dst(skb); + rt = dst_rt6_info(skb_dst(skb)); if (rt) { rcu_read_lock(); if (rt->rt6i_flags & RTF_CACHE) { @@ -2852,7 +2852,7 @@ static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk, bool confirm_neigh) { const struct in6_addr *daddr, *saddr; - struct rt6_info *rt6 = (struct rt6_info *)dst; + struct rt6_info *rt6 = dst_rt6_info(dst); /* Note: do *NOT* check dst_metric_locked(dst, RTAX_MTU) * IPv6 pmtu discovery isn't optional, so 'mtu lock' cannot disable it. @@ -4174,7 +4174,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu } } - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); if (rt->rt6i_flags & RTF_REJECT) { net_dbg_ratelimited("rt6_redirect: source isn't a valid nexthop for redirect target\n"); return; @@ -5608,7 +5608,7 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb, int iif, int type, u32 portid, u32 seq, unsigned int flags) { - struct rt6_info *rt6 = (struct rt6_info *)dst; + struct rt6_info *rt6 = dst_rt6_info(dst); struct rt6key *rt6_dst, *rt6_src; u32 *pmetrics, table, rt6_flags; unsigned char nh_flags = 0; @@ -6111,7 +6111,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, } - rt = container_of(dst, struct rt6_info, dst); + rt = dst_rt6_info(dst); if (rt->dst.error) { err = rt->dst.error; ip6_rt_put(rt); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 77958adf2e16..37201c4fb393 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -97,11 +97,9 @@ static void inet6_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) struct dst_entry *dst = skb_dst(skb); if (dst && dst_hold_safe(dst)) { - const struct rt6_info *rt = (const struct rt6_info *)dst; - rcu_assign_pointer(sk->sk_rx_dst, dst); sk->sk_rx_dst_ifindex = skb->skb_iif; - sk->sk_rx_dst_cookie = rt6_get_cookie(rt); + sk->sk_rx_dst_cookie = rt6_get_cookie(dst_rt6_info(dst)); } } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 085ee236d9a1..674eadfae569 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -910,11 +910,8 @@ start_lookup: static void udp6_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst) { - if (udp_sk_rx_dst_set(sk, dst)) { - const struct rt6_info *rt = (const struct rt6_info *)dst; - - sk->sk_rx_dst_cookie = rt6_get_cookie(rt); - } + if (udp_sk_rx_dst_set(sk, dst)) + sk->sk_rx_dst_cookie = rt6_get_cookie(dst_rt6_info(dst)); } /* wrapper for udp_queue_rcv_skb tacking care of csum conversion and @@ -1585,7 +1582,7 @@ back_from_confirm: skb = ip6_make_skb(sk, getfrag, msg, ulen, sizeof(struct udphdr), &ipc6, - (struct rt6_info *)dst, + dst_rt6_info(dst), msg->msg_flags, &cork); err = PTR_ERR(skb); if (!IS_ERR_OR_NULL(skb)) @@ -1612,7 +1609,7 @@ do_append_data: ipc6.dontfrag = inet6_test_bit(DONTFRAG, sk); up->len += ulen; err = ip6_append_data(sk, getfrag, msg, ulen, sizeof(struct udphdr), - &ipc6, fl6, (struct rt6_info *)dst, + &ipc6, fl6, dst_rt6_info(dst), corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); if (err) udp_v6_flush_pending_frames(sk); diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 4891012b692f..7924e08ee142 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -70,7 +70,7 @@ static int xfrm6_get_saddr(struct net *net, int oif, static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, const struct flowi *fl) { - struct rt6_info *rt = (struct rt6_info *)xdst->route; + struct rt6_info *rt = dst_rt6_info(xdst->route); xdst->u.dst.dev = dev; netdev_hold(dev, &xdst->u.dst.dev_tracker, GFP_ATOMIC); diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index 7bf14cf9ffaa..8780ec64f376 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -630,7 +630,7 @@ back_from_confirm: ulen = len + (skb_queue_empty(&sk->sk_write_queue) ? transhdrlen : 0); err = ip6_append_data(sk, ip_generic_getfrag, msg, ulen, transhdrlen, &ipc6, - &fl6, (struct rt6_info *)dst, + &fl6, dst_rt6_info(dst), msg->msg_flags); if (err) ip6_flush_pending_frames(sk); diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index 8fc790f2a01b..606349c8df0e 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -90,7 +90,7 @@ static int mpls_xmit(struct sk_buff *skb) ttl = net->mpls.default_ttl; else ttl = ipv6_hdr(skb)->hop_limit; - rt6 = (struct rt6_info *)dst; + rt6 = dst_rt6_info(dst); } else { goto drop; } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 39b5fd6bbf65..6e8b9d100ad2 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -180,7 +180,7 @@ static inline bool crosses_local_route_boundary(int skb_af, struct sk_buff *skb, (!skb->dev || skb->dev->flags & IFF_LOOPBACK) && (addr_type & IPV6_ADDR_LOOPBACK); old_rt_is_local = __ip_vs_is_local_route6( - (struct rt6_info *)skb_dst(skb)); + dst_rt6_info(skb_dst(skb))); } else #endif { @@ -481,7 +481,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, if (dest) { dest_dst = __ip_vs_dst_check(dest); if (likely(dest_dst)) - rt = (struct rt6_info *) dest_dst->dst_cache; + rt = dst_rt6_info(dest_dst->dst_cache); else { u32 cookie; @@ -501,7 +501,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, ip_vs_dest_dst_free(dest_dst); goto err_unreach; } - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); cookie = rt6_get_cookie(rt); __ip_vs_dst_set(dest, dest_dst, &rt->dst, cookie); spin_unlock_bh(&dest->dst_lock); @@ -517,7 +517,7 @@ __ip_vs_get_out_rt_v6(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, rt_mode); if (!dst) goto err_unreach; - rt = (struct rt6_info *) dst; + rt = dst_rt6_info(dst); } local = __ip_vs_is_local_route6(rt); @@ -862,7 +862,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_RT_MODE_RDR); if (local < 0) goto tx_error; - rt = (struct rt6_info *) skb_dst(skb); + rt = dst_rt6_info(skb_dst(skb)); /* * Avoid duplicate tuple in reply direction for NAT traffic * to local address when connection is sync-ed @@ -1288,7 +1288,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, if (local) return ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 1); - rt = (struct rt6_info *) skb_dst(skb); + rt = dst_rt6_info(skb_dst(skb)); tdev = rt->dst.dev; /* @@ -1590,7 +1590,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, &cp->daddr.in6, NULL, ipvsh, 0, rt_mode); if (local < 0) goto tx_error; - rt = (struct rt6_info *) skb_dst(skb); + rt = dst_rt6_info(skb_dst(skb)); /* * Avoid duplicate tuple in reply direction for NAT traffic * to local address when connection is sync-ed diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index a0571339239c..5c1ff07eaee0 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -77,12 +77,8 @@ EXPORT_SYMBOL_GPL(flow_offload_alloc); static u32 flow_offload_dst_cookie(struct flow_offload_tuple *flow_tuple) { - const struct rt6_info *rt; - - if (flow_tuple->l3proto == NFPROTO_IPV6) { - rt = (const struct rt6_info *)flow_tuple->dst_cache; - return rt6_get_cookie(rt); - } + if (flow_tuple->l3proto == NFPROTO_IPV6) + return rt6_get_cookie(dst_rt6_info(flow_tuple->dst_cache)); return 0; } diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 5383bed3d3e0..100887beed31 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -729,7 +729,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, return NF_ACCEPT; if (unlikely(tuplehash->tuple.xmit_type == FLOW_OFFLOAD_XMIT_XFRM)) { - rt = (struct rt6_info *)tuplehash->tuple.dst_cache; + rt = dst_rt6_info(tuplehash->tuple.dst_cache); memset(skb->cb, 0, sizeof(struct inet6_skb_parm)); IP6CB(skb)->iif = skb->dev->ifindex; IP6CB(skb)->flags = IP6SKB_FORWARDED; @@ -741,7 +741,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb, switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: - rt = (struct rt6_info *)tuplehash->tuple.dst_cache; + rt = dst_rt6_info(tuplehash->tuple.dst_cache); outdev = rt->dst.dev; skb->dev = outdev; nexthop = rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6); diff --git a/net/netfilter/nft_rt.c b/net/netfilter/nft_rt.c index 24d977138572..2434c624aafd 100644 --- a/net/netfilter/nft_rt.c +++ b/net/netfilter/nft_rt.c @@ -80,7 +80,7 @@ void nft_rt_get_eval(const struct nft_expr *expr, if (nft_pf(pkt) != NFPROTO_IPV6) goto err; - memcpy(dest, rt6_nexthop((struct rt6_info *)dst, + memcpy(dest, rt6_nexthop(dst_rt6_info(dst), &ipv6_hdr(skb)->daddr), sizeof(struct in6_addr)); break; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 24368f755ab1..f7b809c0d142 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -415,7 +415,7 @@ out: if (!IS_ERR_OR_NULL(dst)) { struct rt6_info *rt; - rt = (struct rt6_info *)dst; + rt = dst_rt6_info(dst); t->dst_cookie = rt6_get_cookie(rt); pr_debug("rt6_dst:%pI6/%d rt6_src:%pI6\n", &rt->rt6i_dst.addr, rt->rt6i_dst.plen, diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 6affe5cd85d8..1a41650d9d7b 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2598,8 +2598,7 @@ static void xfrm_init_path(struct xfrm_dst *path, struct dst_entry *dst, int nfheader_len) { if (dst->ops->family == AF_INET6) { - struct rt6_info *rt = (struct rt6_info *)dst; - path->path_cookie = rt6_get_cookie(rt); + path->path_cookie = rt6_get_cookie(dst_rt6_info(dst)); path->u.rt6.rt6i_nfheader_len = nfheader_len; } } -- cgit v1.2.3 From 3c668cef61adde9ddfb1b79708c9a08d9ac01bae Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 26 Apr 2024 16:33:55 +0000 Subject: net: hsr: init prune_proxy_timer sooner We must initialize prune_proxy_timer before we attempt a del_timer_sync() on it. syzbot reported the following splat: INFO: trying to register non-static key. The code is fine but needs lockdep annotation, or maybe you didn't initialize this object before use? turning off the locking correctness validator. CPU: 1 PID: 11 Comm: kworker/u8:1 Not tainted 6.9.0-rc5-syzkaller-01199-gfc48de77d69d #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 Workqueue: netns cleanup_net Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 assign_lock_key+0x238/0x270 kernel/locking/lockdep.c:976 register_lock_class+0x1cf/0x980 kernel/locking/lockdep.c:1289 __lock_acquire+0xda/0x1fd0 kernel/locking/lockdep.c:5014 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754 __timer_delete_sync+0x148/0x310 kernel/time/timer.c:1648 del_timer_sync include/linux/timer.h:185 [inline] hsr_dellink+0x33/0x80 net/hsr/hsr_netlink.c:132 default_device_exit_batch+0x956/0xa90 net/core/dev.c:11737 ops_exit_list net/core/net_namespace.c:175 [inline] cleanup_net+0x89d/0xcc0 net/core/net_namespace.c:637 process_one_work kernel/workqueue.c:3254 [inline] process_scheduled_works+0xa10/0x17c0 kernel/workqueue.c:3335 worker_thread+0x86d/0xd70 kernel/workqueue.c:3416 kthread+0x2f0/0x390 kernel/kthread.c:388 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 ODEBUG: assert_init not available (active state 0) object: ffff88806d3fcd88 object type: timer_list hint: 0x0 WARNING: CPU: 1 PID: 11 at lib/debugobjects.c:517 debug_print_object+0x17a/0x1f0 lib/debugobjects.c:514 Fixes: 5055cccfc2d1 ("net: hsr: Provide RedBox support (HSR-SAN)") Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: Lukasz Majewski Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240426163355.2613767-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/hsr/hsr_device.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index cd1e7c6d2fc0..86127300b102 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -592,6 +592,7 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], timer_setup(&hsr->announce_timer, hsr_announce, 0); timer_setup(&hsr->prune_timer, hsr_prune_nodes, 0); + timer_setup(&hsr->prune_proxy_timer, hsr_prune_proxy_nodes, 0); ether_addr_copy(hsr->sup_multicast_addr, def_multicast_addr); hsr->sup_multicast_addr[ETH_ALEN - 1] = multicast_spec; @@ -631,7 +632,6 @@ int hsr_dev_finalize(struct net_device *hsr_dev, struct net_device *slave[2], hsr->redbox = true; ether_addr_copy(hsr->macaddress_redbox, interlink->dev_addr); - timer_setup(&hsr->prune_proxy_timer, hsr_prune_proxy_nodes, 0); mod_timer(&hsr->prune_proxy_timer, jiffies + msecs_to_jiffies(PRUNE_PROXY_PERIOD)); } -- cgit v1.2.3 From 0cfe71f45f420e412fda2395807a56c453a6e0b6 Mon Sep 17 00:00:00 2001 From: Xuan Zhuo Date: Fri, 26 Apr 2024 11:39:27 +0800 Subject: netdev: add queue stats These stats are commonly. Support reporting those via netdev-genl queue stats. name: rx-hw-drops name: rx-hw-drop-overruns name: rx-csum-unnecessary name: rx-csum-none name: rx-csum-bad name: rx-hw-gro-packets name: rx-hw-gro-bytes name: rx-hw-gro-wire-packets name: rx-hw-gro-wire-bytes name: rx-hw-drop-ratelimits name: tx-hw-drops name: tx-hw-drop-errors name: tx-csum-none name: tx-needs-csum name: tx-hw-gso-packets name: tx-hw-gso-bytes name: tx-hw-gso-wire-packets name: tx-hw-gso-wire-bytes name: tx-hw-drop-ratelimits Signed-off-by: Xuan Zhuo Reviewed-by: Jakub Kicinski Signed-off-by: Paolo Abeni --- Documentation/netlink/specs/netdev.yaml | 104 ++++++++++++++++++++++++++++++++ include/net/netdev_queues.h | 27 +++++++++ include/uapi/linux/netdev.h | 19 ++++++ net/core/netdev-genl.c | 23 ++++++- tools/include/uapi/linux/netdev.h | 19 ++++++ 5 files changed, 190 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index 679c4130707c..2be4b3714d17 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -335,6 +335,110 @@ attribute-sets: Allocation failure may, or may not result in a packet drop, depending on driver implementation and whether system recovers quickly. type: uint + - + name: rx-hw-drops + doc: | + Number of all packets which entered the device, but never left it, + including but not limited to: packets dropped due to lack of buffer + space, processing errors, explicit or implicit policies and packet + filters. + type: uint + - + name: rx-hw-drop-overruns + doc: | + Number of packets dropped due to transient lack of resources, such as + buffer space, host descriptors etc. + type: uint + - + name: rx-csum-unnecessary + doc: Number of packets that were marked as CHECKSUM_UNNECESSARY. + type: uint + - + name: rx-csum-none + doc: Number of packets that were not checksummed by device. + type: uint + - + name: rx-csum-bad + doc: | + Number of packets with bad checksum. The packets are not discarded, + but still delivered to the stack. + type: uint + - + name: rx-hw-gro-packets + doc: | + Number of packets that were coalesced from smaller packets by the device. + Counts only packets coalesced with the HW-GRO netdevice feature, + LRO-coalesced packets are not counted. + type: uint + - + name: rx-hw-gro-bytes + doc: See `rx-hw-gro-packets`. + type: uint + - + name: rx-hw-gro-wire-packets + doc: | + Number of packets that were coalesced to bigger packetss with the HW-GRO + netdevice feature. LRO-coalesced packets are not counted. + type: uint + - + name: rx-hw-gro-wire-bytes + doc: See `rx-hw-gro-wire-packets`. + type: uint + - + name: rx-hw-drop-ratelimits + doc: | + Number of the packets dropped by the device due to the received + packets bitrate exceeding the device rate limit. + type: uint + - + name: tx-hw-drops + doc: | + Number of packets that arrived at the device but never left it, + encompassing packets dropped for reasons such as processing errors, as + well as those affected by explicitly defined policies and packet + filtering criteria. + type: uint + - + name: tx-hw-drop-errors + doc: Number of packets dropped because they were invalid or malformed. + type: uint + - + name: tx-csum-none + doc: | + Number of packets that did not require the device to calculate the + checksum. + type: uint + - + name: tx-needs-csum + doc: | + Number of packets that required the device to calculate the checksum. + type: uint + - + name: tx-hw-gso-packets + doc: | + Number of packets that necessitated segmentation into smaller packets + by the device. + type: uint + - + name: tx-hw-gso-bytes + doc: See `tx-hw-gso-packets`. + type: uint + - + name: tx-hw-gso-wire-packets + doc: | + Number of wire-sized packets generated by processing + `tx-hw-gso-packets` + type: uint + - + name: tx-hw-gso-wire-bytes + doc: See `tx-hw-gso-wire-packets`. + type: uint + - + name: tx-hw-drop-ratelimits + doc: | + Number of the packets dropped by the device due to the transmit + packets bitrate exceeding the device rate limit. + type: uint operations: list: diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h index 1ec408585373..c7ac4539eafc 100644 --- a/include/net/netdev_queues.h +++ b/include/net/netdev_queues.h @@ -9,11 +9,38 @@ struct netdev_queue_stats_rx { u64 bytes; u64 packets; u64 alloc_fail; + + u64 hw_drops; + u64 hw_drop_overruns; + + u64 csum_unnecessary; + u64 csum_none; + u64 csum_bad; + + u64 hw_gro_packets; + u64 hw_gro_bytes; + u64 hw_gro_wire_packets; + u64 hw_gro_wire_bytes; + + u64 hw_drop_ratelimits; }; struct netdev_queue_stats_tx { u64 bytes; u64 packets; + + u64 hw_drops; + u64 hw_drop_errors; + + u64 csum_none; + u64 needs_csum; + + u64 hw_gso_packets; + u64 hw_gso_bytes; + u64 hw_gso_wire_packets; + u64 hw_gso_wire_bytes; + + u64 hw_drop_ratelimits; }; /** diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h index bb65ee840cda..cf24f1d9adf8 100644 --- a/include/uapi/linux/netdev.h +++ b/include/uapi/linux/netdev.h @@ -146,6 +146,25 @@ enum { NETDEV_A_QSTATS_TX_PACKETS, NETDEV_A_QSTATS_TX_BYTES, NETDEV_A_QSTATS_RX_ALLOC_FAIL, + NETDEV_A_QSTATS_RX_HW_DROPS, + NETDEV_A_QSTATS_RX_HW_DROP_OVERRUNS, + NETDEV_A_QSTATS_RX_CSUM_UNNECESSARY, + NETDEV_A_QSTATS_RX_CSUM_NONE, + NETDEV_A_QSTATS_RX_CSUM_BAD, + NETDEV_A_QSTATS_RX_HW_GRO_PACKETS, + NETDEV_A_QSTATS_RX_HW_GRO_BYTES, + NETDEV_A_QSTATS_RX_HW_GRO_WIRE_PACKETS, + NETDEV_A_QSTATS_RX_HW_GRO_WIRE_BYTES, + NETDEV_A_QSTATS_RX_HW_DROP_RATELIMITS, + NETDEV_A_QSTATS_TX_HW_DROPS, + NETDEV_A_QSTATS_TX_HW_DROP_ERRORS, + NETDEV_A_QSTATS_TX_CSUM_NONE, + NETDEV_A_QSTATS_TX_NEEDS_CSUM, + NETDEV_A_QSTATS_TX_HW_GSO_PACKETS, + NETDEV_A_QSTATS_TX_HW_GSO_BYTES, + NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, + NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, + NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, __NETDEV_A_QSTATS_MAX, NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1) diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index dd6510f2c652..4b5054087309 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -489,7 +489,17 @@ netdev_nl_stats_write_rx(struct sk_buff *rsp, struct netdev_queue_stats_rx *rx) { if (netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_PACKETS, rx->packets) || netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_BYTES, rx->bytes) || - netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_ALLOC_FAIL, rx->alloc_fail)) + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_ALLOC_FAIL, rx->alloc_fail) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_DROPS, rx->hw_drops) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_DROP_OVERRUNS, rx->hw_drop_overruns) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_CSUM_UNNECESSARY, rx->csum_unnecessary) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_CSUM_NONE, rx->csum_none) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_CSUM_BAD, rx->csum_bad) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_GRO_PACKETS, rx->hw_gro_packets) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_GRO_BYTES, rx->hw_gro_bytes) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_GRO_WIRE_PACKETS, rx->hw_gro_wire_packets) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_GRO_WIRE_BYTES, rx->hw_gro_wire_bytes) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_RX_HW_DROP_RATELIMITS, rx->hw_drop_ratelimits)) return -EMSGSIZE; return 0; } @@ -498,7 +508,16 @@ static int netdev_nl_stats_write_tx(struct sk_buff *rsp, struct netdev_queue_stats_tx *tx) { if (netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_PACKETS, tx->packets) || - netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_BYTES, tx->bytes)) + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_BYTES, tx->bytes) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROPS, tx->hw_drops) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_ERRORS, tx->hw_drop_errors) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_CSUM_NONE, tx->csum_none) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_NEEDS_CSUM, tx->needs_csum) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_PACKETS, tx->hw_gso_packets) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_BYTES, tx->hw_gso_bytes) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, tx->hw_gso_wire_packets) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, tx->hw_gso_wire_bytes) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits)) return -EMSGSIZE; return 0; } diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h index bb65ee840cda..cf24f1d9adf8 100644 --- a/tools/include/uapi/linux/netdev.h +++ b/tools/include/uapi/linux/netdev.h @@ -146,6 +146,25 @@ enum { NETDEV_A_QSTATS_TX_PACKETS, NETDEV_A_QSTATS_TX_BYTES, NETDEV_A_QSTATS_RX_ALLOC_FAIL, + NETDEV_A_QSTATS_RX_HW_DROPS, + NETDEV_A_QSTATS_RX_HW_DROP_OVERRUNS, + NETDEV_A_QSTATS_RX_CSUM_UNNECESSARY, + NETDEV_A_QSTATS_RX_CSUM_NONE, + NETDEV_A_QSTATS_RX_CSUM_BAD, + NETDEV_A_QSTATS_RX_HW_GRO_PACKETS, + NETDEV_A_QSTATS_RX_HW_GRO_BYTES, + NETDEV_A_QSTATS_RX_HW_GRO_WIRE_PACKETS, + NETDEV_A_QSTATS_RX_HW_GRO_WIRE_BYTES, + NETDEV_A_QSTATS_RX_HW_DROP_RATELIMITS, + NETDEV_A_QSTATS_TX_HW_DROPS, + NETDEV_A_QSTATS_TX_HW_DROP_ERRORS, + NETDEV_A_QSTATS_TX_CSUM_NONE, + NETDEV_A_QSTATS_TX_NEEDS_CSUM, + NETDEV_A_QSTATS_TX_HW_GSO_PACKETS, + NETDEV_A_QSTATS_TX_HW_GSO_BYTES, + NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, + NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, + NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, __NETDEV_A_QSTATS_MAX, NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1) -- cgit v1.2.3 From e5c5f3596de224422561d48eba6ece5210d967b3 Mon Sep 17 00:00:00 2001 From: Erick Archer Date: Sat, 27 Apr 2024 19:23:36 +0200 Subject: sctp: prefer struct_size over open coded arithmetic This is an effort to get rid of all multiplications from allocation functions in order to prevent integer overflows [1][2]. As the "ids" variable is a pointer to "struct sctp_assoc_ids" and this structure ends in a flexible array: struct sctp_assoc_ids { [...] sctp_assoc_t gaids_assoc_id[]; }; the preferred way in the kernel is to use the struct_size() helper to do the arithmetic instead of the calculation "size + size * count" in the kmalloc() function. Also, refactor the code adding the "ids_size" variable to avoid sizing twice. This way, the code is more readable and safer. This code was detected with the help of Coccinelle, and audited and modified manually. Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [1] Link: https://github.com/KSPP/linux/issues/160 [2] Signed-off-by: Erick Archer Acked-by: Xin Long Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/PAXPR02MB724871DB78375AB06B5171C88B152@PAXPR02MB7248.eurprd02.prod.outlook.com Signed-off-by: Paolo Abeni --- net/sctp/socket.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index e416b6d3d270..64196b1dce1d 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -7119,6 +7119,7 @@ static int sctp_getsockopt_assoc_ids(struct sock *sk, int len, struct sctp_sock *sp = sctp_sk(sk); struct sctp_association *asoc; struct sctp_assoc_ids *ids; + size_t ids_size; u32 num = 0; if (sctp_style(sk, TCP)) @@ -7131,11 +7132,11 @@ static int sctp_getsockopt_assoc_ids(struct sock *sk, int len, num++; } - if (len < sizeof(struct sctp_assoc_ids) + sizeof(sctp_assoc_t) * num) + ids_size = struct_size(ids, gaids_assoc_id, num); + if (len < ids_size) return -EINVAL; - len = sizeof(struct sctp_assoc_ids) + sizeof(sctp_assoc_t) * num; - + len = ids_size; ids = kmalloc(len, GFP_USER | __GFP_NOWARN); if (unlikely(!ids)) return -ENOMEM; -- cgit v1.2.3 From 784c46f5467c3cd516e518de211e26611da5c0fb Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:28 +0800 Subject: net/smc: decouple ism_client from SMC-D DMB registration The struct 'ism_client' is specialized for s390 platform firmware ISM. So replace it with 'void' to make SMCD DMB registration helper generic for both Emulated-ISM and existing ISM. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- drivers/s390/net/ism_drv.c | 2 +- include/net/smc.h | 4 ++-- net/smc/smc_ism.c | 7 ++----- 3 files changed, 5 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/drivers/s390/net/ism_drv.c b/drivers/s390/net/ism_drv.c index 43778b088ffa..e36e3ea165d3 100644 --- a/drivers/s390/net/ism_drv.c +++ b/drivers/s390/net/ism_drv.c @@ -745,7 +745,7 @@ static int smcd_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid, } static int smcd_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb, - struct ism_client *client) + void *client) { return ism_register_dmb(smcd->priv, (struct ism_dmb *)dmb, client); } diff --git a/include/net/smc.h b/include/net/smc.h index 10684d0a33df..542d12372c18 100644 --- a/include/net/smc.h +++ b/include/net/smc.h @@ -47,7 +47,6 @@ struct smcd_dmb { #define ISM_ERROR 0xFFFF struct smcd_dev; -struct ism_client; struct smcd_gid { u64 gid; @@ -58,7 +57,7 @@ struct smcd_ops { int (*query_remote_gid)(struct smcd_dev *dev, struct smcd_gid *rgid, u32 vid_valid, u32 vid); int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb, - struct ism_client *client); + void *client); int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb); int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id); int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id); @@ -78,6 +77,7 @@ struct smcd_ops { struct smcd_dev { const struct smcd_ops *ops; void *priv; + void *client; struct list_head list; spinlock_t lock; struct smc_connection **conn; diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c index ac88de2a06a0..051726586730 100644 --- a/net/smc/smc_ism.c +++ b/net/smc/smc_ism.c @@ -222,7 +222,6 @@ int smc_ism_unregister_dmb(struct smcd_dev *smcd, struct smc_buf_desc *dmb_desc) int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len, struct smc_buf_desc *dmb_desc) { -#if IS_ENABLED(CONFIG_ISM) struct smcd_dmb dmb; int rc; @@ -231,7 +230,7 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len, dmb.sba_idx = dmb_desc->sba_idx; dmb.vlan_id = lgr->vlan_id; dmb.rgid = lgr->peer_gid.gid; - rc = lgr->smcd->ops->register_dmb(lgr->smcd, &dmb, &smc_ism_client); + rc = lgr->smcd->ops->register_dmb(lgr->smcd, &dmb, lgr->smcd->client); if (!rc) { dmb_desc->sba_idx = dmb.sba_idx; dmb_desc->token = dmb.dmb_tok; @@ -240,9 +239,6 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len, dmb_desc->len = dmb.dmb_len; } return rc; -#else - return 0; -#endif } static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd, @@ -446,6 +442,7 @@ static void smcd_register_dev(struct ism_dev *ism) if (!smcd) return; smcd->priv = ism; + smcd->client = &smc_ism_client; ism_set_priv(ism, &smc_ism_client, smcd); if (smc_pnetid_by_dev_port(&ism->pdev->dev, 0, smcd->pnetid)) smc_pnetid_by_table_smcd(smcd); -- cgit v1.2.3 From 46ac64419ded7bcbe1fb8d0f4df3258384a425c2 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:29 +0800 Subject: net/smc: introduce loopback-ism for SMC intra-OS shortcut This introduces a kind of Emulated-ISM device named loopback-ism for SMCv2.1. The loopback-ism device is currently exclusive for SMC usage, and aims to provide an SMC shortcut for sockets within the same kernel, leading to improved intra-OS traffic performance. Configuration of this feature is managed through the config SMC_LO. Signed-off-by: Wen Gu Reviewed-by: Gerd Bayer Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/Kconfig | 13 +++++ net/smc/Makefile | 1 + net/smc/af_smc.c | 12 +++- net/smc/smc_loopback.c | 156 +++++++++++++++++++++++++++++++++++++++++++++++++ net/smc/smc_loopback.h | 42 +++++++++++++ 5 files changed, 223 insertions(+), 1 deletion(-) create mode 100644 net/smc/smc_loopback.c create mode 100644 net/smc/smc_loopback.h (limited to 'net') diff --git a/net/smc/Kconfig b/net/smc/Kconfig index 746be3996768..ba5e6a2dd2fd 100644 --- a/net/smc/Kconfig +++ b/net/smc/Kconfig @@ -20,3 +20,16 @@ config SMC_DIAG smcss. if unsure, say Y. + +config SMC_LO + bool "SMC intra-OS shortcut with loopback-ism" + depends on SMC + default n + help + SMC_LO enables the creation of an Emulated-ISM device named + loopback-ism in SMC and makes use of it for transferring data + when communication occurs within the same OS. This helps in + convenient testing of SMC-D since loopback-ism is independent + of architecture or hardware. + + if unsure, say N. diff --git a/net/smc/Makefile b/net/smc/Makefile index 875efcd126a2..2c510d543058 100644 --- a/net/smc/Makefile +++ b/net/smc/Makefile @@ -6,3 +6,4 @@ smc-y := af_smc.o smc_pnet.o smc_ib.o smc_clc.o smc_core.o smc_wr.o smc_llc.o smc-y += smc_cdc.o smc_tx.o smc_rx.o smc_close.o smc_ism.o smc_netlink.o smc_stats.o smc-y += smc_tracepoint.o smc-$(CONFIG_SYSCTL) += smc_sysctl.o +smc-$(CONFIG_SMC_LO) += smc_loopback.o diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index e8dcd28a554c..47f3bc1470bc 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -53,6 +53,7 @@ #include "smc_stats.h" #include "smc_tracepoint.h" #include "smc_sysctl.h" +#include "smc_loopback.h" static DEFINE_MUTEX(smc_server_lgr_pending); /* serialize link group * creation on server @@ -3555,15 +3556,23 @@ static int __init smc_init(void) goto out_sock; } + rc = smc_loopback_init(); + if (rc) { + pr_err("%s: smc_loopback_init fails with %d\n", __func__, rc); + goto out_ib; + } + rc = tcp_register_ulp(&smc_ulp_ops); if (rc) { pr_err("%s: tcp_ulp_register fails with %d\n", __func__, rc); - goto out_ib; + goto out_lo; } static_branch_enable(&tcp_have_smc); return 0; +out_lo: + smc_loopback_exit(); out_ib: smc_ib_unregister_client(); out_sock: @@ -3601,6 +3610,7 @@ static void __exit smc_exit(void) tcp_unregister_ulp(&smc_ulp_ops); sock_unregister(PF_SMC); smc_core_exit(); + smc_loopback_exit(); smc_ib_unregister_client(); smc_ism_exit(); destroy_workqueue(smc_close_wq); diff --git a/net/smc/smc_loopback.c b/net/smc/smc_loopback.c new file mode 100644 index 000000000000..c364e3e6e3fb --- /dev/null +++ b/net/smc/smc_loopback.c @@ -0,0 +1,156 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Shared Memory Communications Direct over loopback-ism device. + * + * Functions for loopback-ism device. + * + * Copyright (c) 2024, Alibaba Inc. + * + * Author: Wen Gu + * Tony Lu + * + */ + +#include +#include +#include + +#include "smc_ism.h" +#include "smc_loopback.h" + +static const char smc_lo_dev_name[] = "loopback-ism"; +static struct smc_lo_dev *lo_dev; + +static const struct smcd_ops lo_ops = { + .query_remote_gid = NULL, + .register_dmb = NULL, + .unregister_dmb = NULL, + .add_vlan_id = NULL, + .del_vlan_id = NULL, + .set_vlan_required = NULL, + .reset_vlan_required = NULL, + .signal_event = NULL, + .move_data = NULL, + .supports_v2 = NULL, + .get_local_gid = NULL, + .get_chid = NULL, + .get_dev = NULL, +}; + +static struct smcd_dev *smcd_lo_alloc_dev(const struct smcd_ops *ops, + int max_dmbs) +{ + struct smcd_dev *smcd; + + smcd = kzalloc(sizeof(*smcd), GFP_KERNEL); + if (!smcd) + return NULL; + + smcd->conn = kcalloc(max_dmbs, sizeof(struct smc_connection *), + GFP_KERNEL); + if (!smcd->conn) + goto out_smcd; + + smcd->ops = ops; + + spin_lock_init(&smcd->lock); + spin_lock_init(&smcd->lgr_lock); + INIT_LIST_HEAD(&smcd->vlan); + INIT_LIST_HEAD(&smcd->lgr_list); + init_waitqueue_head(&smcd->lgrs_deleted); + return smcd; + +out_smcd: + kfree(smcd); + return NULL; +} + +static int smcd_lo_register_dev(struct smc_lo_dev *ldev) +{ + struct smcd_dev *smcd; + + smcd = smcd_lo_alloc_dev(&lo_ops, SMC_LO_MAX_DMBS); + if (!smcd) + return -ENOMEM; + ldev->smcd = smcd; + smcd->priv = ldev; + + /* TODO: + * register loopback-ism to smcd_dev list. + */ + return 0; +} + +static void smcd_lo_unregister_dev(struct smc_lo_dev *ldev) +{ + struct smcd_dev *smcd = ldev->smcd; + + /* TODO: + * unregister loopback-ism from smcd_dev list. + */ + kfree(smcd->conn); + kfree(smcd); +} + +static int smc_lo_dev_init(struct smc_lo_dev *ldev) +{ + return smcd_lo_register_dev(ldev); +} + +static void smc_lo_dev_exit(struct smc_lo_dev *ldev) +{ + smcd_lo_unregister_dev(ldev); +} + +static void smc_lo_dev_release(struct device *dev) +{ + struct smc_lo_dev *ldev = + container_of(dev, struct smc_lo_dev, dev); + + kfree(ldev); +} + +static int smc_lo_dev_probe(void) +{ + struct smc_lo_dev *ldev; + int ret; + + ldev = kzalloc(sizeof(*ldev), GFP_KERNEL); + if (!ldev) + return -ENOMEM; + + ldev->dev.parent = NULL; + ldev->dev.release = smc_lo_dev_release; + device_initialize(&ldev->dev); + dev_set_name(&ldev->dev, smc_lo_dev_name); + + ret = smc_lo_dev_init(ldev); + if (ret) + goto free_dev; + + lo_dev = ldev; /* global loopback device */ + return 0; + +free_dev: + put_device(&ldev->dev); + return ret; +} + +static void smc_lo_dev_remove(void) +{ + if (!lo_dev) + return; + + smc_lo_dev_exit(lo_dev); + put_device(&lo_dev->dev); /* device_initialize in smc_lo_dev_probe */ +} + +int smc_loopback_init(void) +{ + return smc_lo_dev_probe(); +} + +void smc_loopback_exit(void) +{ + smc_lo_dev_remove(); +} diff --git a/net/smc/smc_loopback.h b/net/smc/smc_loopback.h new file mode 100644 index 000000000000..7fc4b374bc82 --- /dev/null +++ b/net/smc/smc_loopback.h @@ -0,0 +1,42 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Shared Memory Communications Direct over loopback-ism device. + * + * SMC-D loopback-ism device structure definitions. + * + * Copyright (c) 2024, Alibaba Inc. + * + * Author: Wen Gu + * Tony Lu + * + */ + +#ifndef _SMC_LOOPBACK_H +#define _SMC_LOOPBACK_H + +#include +#include +#include + +#if IS_ENABLED(CONFIG_SMC_LO) +#define SMC_LO_MAX_DMBS 5000 + +struct smc_lo_dev { + struct smcd_dev *smcd; + struct device dev; +}; + +int smc_loopback_init(void); +void smc_loopback_exit(void); +#else +static inline int smc_loopback_init(void) +{ + return 0; +} + +static inline void smc_loopback_exit(void) +{ +} +#endif + +#endif /* _SMC_LOOPBACK_H */ -- cgit v1.2.3 From 45783ee85bf337693a99e03cd6142fdcfb06585d Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:30 +0800 Subject: net/smc: implement ID-related operations of loopback-ism This implements operations related to IDs for the loopback-ism device. loopback-ism uses an Extended GID that is a 128-bit GID instead of the existing ISM 64-bit GID, and uses the CHID defined with the reserved value 0xFFFF. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/smc_loopback.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++---- net/smc/smc_loopback.h | 3 +++ 2 files changed, 60 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/smc/smc_loopback.c b/net/smc/smc_loopback.c index c364e3e6e3fb..0349632a76c4 100644 --- a/net/smc/smc_loopback.c +++ b/net/smc/smc_loopback.c @@ -18,11 +18,62 @@ #include "smc_ism.h" #include "smc_loopback.h" +#define SMC_LO_V2_CAPABLE 0x1 /* loopback-ism acts as ISMv2 */ + static const char smc_lo_dev_name[] = "loopback-ism"; static struct smc_lo_dev *lo_dev; +static void smc_lo_generate_ids(struct smc_lo_dev *ldev) +{ + struct smcd_gid *lgid = &ldev->local_gid; + uuid_t uuid; + + uuid_gen(&uuid); + memcpy(&lgid->gid, &uuid, sizeof(lgid->gid)); + memcpy(&lgid->gid_ext, (u8 *)&uuid + sizeof(lgid->gid), + sizeof(lgid->gid_ext)); + + ldev->chid = SMC_LO_RESERVED_CHID; +} + +static int smc_lo_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid, + u32 vid_valid, u32 vid) +{ + struct smc_lo_dev *ldev = smcd->priv; + + /* rgid should be the same as lgid */ + if (!ldev || rgid->gid != ldev->local_gid.gid || + rgid->gid_ext != ldev->local_gid.gid_ext) + return -ENETUNREACH; + return 0; +} + +static int smc_lo_supports_v2(void) +{ + return SMC_LO_V2_CAPABLE; +} + +static void smc_lo_get_local_gid(struct smcd_dev *smcd, + struct smcd_gid *smcd_gid) +{ + struct smc_lo_dev *ldev = smcd->priv; + + smcd_gid->gid = ldev->local_gid.gid; + smcd_gid->gid_ext = ldev->local_gid.gid_ext; +} + +static u16 smc_lo_get_chid(struct smcd_dev *smcd) +{ + return ((struct smc_lo_dev *)smcd->priv)->chid; +} + +static struct device *smc_lo_get_dev(struct smcd_dev *smcd) +{ + return &((struct smc_lo_dev *)smcd->priv)->dev; +} + static const struct smcd_ops lo_ops = { - .query_remote_gid = NULL, + .query_remote_gid = smc_lo_query_rgid, .register_dmb = NULL, .unregister_dmb = NULL, .add_vlan_id = NULL, @@ -31,10 +82,10 @@ static const struct smcd_ops lo_ops = { .reset_vlan_required = NULL, .signal_event = NULL, .move_data = NULL, - .supports_v2 = NULL, - .get_local_gid = NULL, - .get_chid = NULL, - .get_dev = NULL, + .supports_v2 = smc_lo_supports_v2, + .get_local_gid = smc_lo_get_local_gid, + .get_chid = smc_lo_get_chid, + .get_dev = smc_lo_get_dev, }; static struct smcd_dev *smcd_lo_alloc_dev(const struct smcd_ops *ops, @@ -94,6 +145,7 @@ static void smcd_lo_unregister_dev(struct smc_lo_dev *ldev) static int smc_lo_dev_init(struct smc_lo_dev *ldev) { + smc_lo_generate_ids(ldev); return smcd_lo_register_dev(ldev); } diff --git a/net/smc/smc_loopback.h b/net/smc/smc_loopback.h index 7fc4b374bc82..d7df79f7bb64 100644 --- a/net/smc/smc_loopback.h +++ b/net/smc/smc_loopback.h @@ -20,10 +20,13 @@ #if IS_ENABLED(CONFIG_SMC_LO) #define SMC_LO_MAX_DMBS 5000 +#define SMC_LO_RESERVED_CHID 0xFFFF struct smc_lo_dev { struct smcd_dev *smcd; struct device dev; + u16 chid; + struct smcd_gid local_gid; }; int smc_loopback_init(void); -- cgit v1.2.3 From f7a22071dbf316c982fb44308874bd7ad9ac2091 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:31 +0800 Subject: net/smc: implement DMB-related operations of loopback-ism This implements DMB (un)registration and data move operations of loopback-ism device. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/smc_loopback.c | 129 +++++++++++++++++++++++++++++++++++++++++++++++-- net/smc/smc_loopback.h | 13 +++++ 2 files changed, 139 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/smc/smc_loopback.c b/net/smc/smc_loopback.c index 0349632a76c4..d75852549b00 100644 --- a/net/smc/smc_loopback.c +++ b/net/smc/smc_loopback.c @@ -15,10 +15,12 @@ #include #include +#include "smc_cdc.h" #include "smc_ism.h" #include "smc_loopback.h" #define SMC_LO_V2_CAPABLE 0x1 /* loopback-ism acts as ISMv2 */ +#define SMC_DMA_ADDR_INVALID (~(dma_addr_t)0) static const char smc_lo_dev_name[] = "loopback-ism"; static struct smc_lo_dev *lo_dev; @@ -48,6 +50,125 @@ static int smc_lo_query_rgid(struct smcd_dev *smcd, struct smcd_gid *rgid, return 0; } +static int smc_lo_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb, + void *client_priv) +{ + struct smc_lo_dmb_node *dmb_node, *tmp_node; + struct smc_lo_dev *ldev = smcd->priv; + int sba_idx, rc; + + /* check space for new dmb */ + for_each_clear_bit(sba_idx, ldev->sba_idx_mask, SMC_LO_MAX_DMBS) { + if (!test_and_set_bit(sba_idx, ldev->sba_idx_mask)) + break; + } + if (sba_idx == SMC_LO_MAX_DMBS) + return -ENOSPC; + + dmb_node = kzalloc(sizeof(*dmb_node), GFP_KERNEL); + if (!dmb_node) { + rc = -ENOMEM; + goto err_bit; + } + + dmb_node->sba_idx = sba_idx; + dmb_node->len = dmb->dmb_len; + dmb_node->cpu_addr = kzalloc(dmb_node->len, GFP_KERNEL | + __GFP_NOWARN | __GFP_NORETRY | + __GFP_NOMEMALLOC); + if (!dmb_node->cpu_addr) { + rc = -ENOMEM; + goto err_node; + } + dmb_node->dma_addr = SMC_DMA_ADDR_INVALID; + +again: + /* add new dmb into hash table */ + get_random_bytes(&dmb_node->token, sizeof(dmb_node->token)); + write_lock_bh(&ldev->dmb_ht_lock); + hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_node->token) { + if (tmp_node->token == dmb_node->token) { + write_unlock_bh(&ldev->dmb_ht_lock); + goto again; + } + } + hash_add(ldev->dmb_ht, &dmb_node->list, dmb_node->token); + write_unlock_bh(&ldev->dmb_ht_lock); + + dmb->sba_idx = dmb_node->sba_idx; + dmb->dmb_tok = dmb_node->token; + dmb->cpu_addr = dmb_node->cpu_addr; + dmb->dma_addr = dmb_node->dma_addr; + dmb->dmb_len = dmb_node->len; + + return 0; + +err_node: + kfree(dmb_node); +err_bit: + clear_bit(sba_idx, ldev->sba_idx_mask); + return rc; +} + +static int smc_lo_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb) +{ + struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node; + struct smc_lo_dev *ldev = smcd->priv; + + /* remove dmb from hash table */ + write_lock_bh(&ldev->dmb_ht_lock); + hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb->dmb_tok) { + if (tmp_node->token == dmb->dmb_tok) { + dmb_node = tmp_node; + break; + } + } + if (!dmb_node) { + write_unlock_bh(&ldev->dmb_ht_lock); + return -EINVAL; + } + hash_del(&dmb_node->list); + write_unlock_bh(&ldev->dmb_ht_lock); + + clear_bit(dmb_node->sba_idx, ldev->sba_idx_mask); + kfree(dmb_node->cpu_addr); + kfree(dmb_node); + + return 0; +} + +static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, + unsigned int idx, bool sf, unsigned int offset, + void *data, unsigned int size) +{ + struct smc_lo_dmb_node *rmb_node = NULL, *tmp_node; + struct smc_lo_dev *ldev = smcd->priv; + struct smc_connection *conn; + + read_lock_bh(&ldev->dmb_ht_lock); + hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) { + if (tmp_node->token == dmb_tok) { + rmb_node = tmp_node; + break; + } + } + if (!rmb_node) { + read_unlock_bh(&ldev->dmb_ht_lock); + return -EINVAL; + } + memcpy((char *)rmb_node->cpu_addr + offset, data, size); + read_unlock_bh(&ldev->dmb_ht_lock); + + if (sf) { + conn = smcd->conn[rmb_node->sba_idx]; + if (conn && !conn->killed) + tasklet_schedule(&conn->rx_tsklet); + else + return -EPIPE; + } + return 0; +} + static int smc_lo_supports_v2(void) { return SMC_LO_V2_CAPABLE; @@ -74,14 +195,14 @@ static struct device *smc_lo_get_dev(struct smcd_dev *smcd) static const struct smcd_ops lo_ops = { .query_remote_gid = smc_lo_query_rgid, - .register_dmb = NULL, - .unregister_dmb = NULL, + .register_dmb = smc_lo_register_dmb, + .unregister_dmb = smc_lo_unregister_dmb, .add_vlan_id = NULL, .del_vlan_id = NULL, .set_vlan_required = NULL, .reset_vlan_required = NULL, .signal_event = NULL, - .move_data = NULL, + .move_data = smc_lo_move_data, .supports_v2 = smc_lo_supports_v2, .get_local_gid = smc_lo_get_local_gid, .get_chid = smc_lo_get_chid, @@ -146,6 +267,8 @@ static void smcd_lo_unregister_dev(struct smc_lo_dev *ldev) static int smc_lo_dev_init(struct smc_lo_dev *ldev) { smc_lo_generate_ids(ldev); + rwlock_init(&ldev->dmb_ht_lock); + hash_init(ldev->dmb_ht); return smcd_lo_register_dev(ldev); } diff --git a/net/smc/smc_loopback.h b/net/smc/smc_loopback.h index d7df79f7bb64..b8206338309f 100644 --- a/net/smc/smc_loopback.h +++ b/net/smc/smc_loopback.h @@ -20,13 +20,26 @@ #if IS_ENABLED(CONFIG_SMC_LO) #define SMC_LO_MAX_DMBS 5000 +#define SMC_LO_DMBS_HASH_BITS 12 #define SMC_LO_RESERVED_CHID 0xFFFF +struct smc_lo_dmb_node { + struct hlist_node list; + u64 token; + u32 len; + u32 sba_idx; + void *cpu_addr; + dma_addr_t dma_addr; +}; + struct smc_lo_dev { struct smcd_dev *smcd; struct device dev; u16 chid; struct smcd_gid local_gid; + rwlock_t dmb_ht_lock; + DECLARE_BITMAP(sba_idx_mask, SMC_LO_MAX_DMBS); + DECLARE_HASHTABLE(dmb_ht, SMC_LO_DMBS_HASH_BITS); }; int smc_loopback_init(void); -- cgit v1.2.3 From d1d8d0b6c7c68b0665456831fa779174ebd78f90 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:32 +0800 Subject: net/smc: mark optional smcd_ops and check for support when called Some operations are not supported by new introduced Emulated-ISM, so mark them as optional and check if the device supports them when called. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- include/net/smc.h | 14 ++++++++------ net/smc/smc_ism.c | 9 ++++++++- 2 files changed, 16 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/smc.h b/include/net/smc.h index 542d12372c18..33b753115e43 100644 --- a/include/net/smc.h +++ b/include/net/smc.h @@ -59,12 +59,6 @@ struct smcd_ops { int (*register_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb, void *client); int (*unregister_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb); - int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id); - int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id); - int (*set_vlan_required)(struct smcd_dev *dev); - int (*reset_vlan_required)(struct smcd_dev *dev); - int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid, - u32 trigger_irq, u32 event_code, u64 info); int (*move_data)(struct smcd_dev *dev, u64 dmb_tok, unsigned int idx, bool sf, unsigned int offset, void *data, unsigned int size); @@ -72,6 +66,14 @@ struct smcd_ops { void (*get_local_gid)(struct smcd_dev *dev, struct smcd_gid *gid); u16 (*get_chid)(struct smcd_dev *dev); struct device* (*get_dev)(struct smcd_dev *dev); + + /* optional operations */ + int (*add_vlan_id)(struct smcd_dev *dev, u64 vlan_id); + int (*del_vlan_id)(struct smcd_dev *dev, u64 vlan_id); + int (*set_vlan_required)(struct smcd_dev *dev); + int (*reset_vlan_required)(struct smcd_dev *dev); + int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid, + u32 trigger_irq, u32 event_code, u64 info); }; struct smcd_dev { diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c index 051726586730..36459a3c739d 100644 --- a/net/smc/smc_ism.c +++ b/net/smc/smc_ism.c @@ -126,6 +126,8 @@ int smc_ism_get_vlan(struct smcd_dev *smcd, unsigned short vlanid) if (!vlanid) /* No valid vlan id */ return -EINVAL; + if (!smcd->ops->add_vlan_id) + return -EOPNOTSUPP; /* create new vlan entry, in case we need it */ new_vlan = kzalloc(sizeof(*new_vlan), GFP_KERNEL); @@ -171,6 +173,8 @@ int smc_ism_put_vlan(struct smcd_dev *smcd, unsigned short vlanid) if (!vlanid) /* No valid vlan id */ return -EINVAL; + if (!smcd->ops->del_vlan_id) + return -EOPNOTSUPP; spin_lock_irqsave(&smcd->lock, flags); list_for_each_entry(vlan, &smcd->vlan, list) { @@ -368,7 +372,8 @@ static void smcd_handle_sw_event(struct smc_ism_event_work *wrk) smc_smcd_terminate(wrk->smcd, &peer_gid, ev_info.vlan_id); break; case ISM_EVENT_CODE_TESTLINK: /* Activity timer */ - if (ev_info.code == ISM_EVENT_REQUEST) { + if (ev_info.code == ISM_EVENT_REQUEST && + wrk->smcd->ops->signal_event) { ev_info.code = ISM_EVENT_RESPONSE; wrk->smcd->ops->signal_event(wrk->smcd, &peer_gid, @@ -538,6 +543,8 @@ int smc_ism_signal_shutdown(struct smc_link_group *lgr) if (lgr->peer_shutdown) return 0; + if (!lgr->smcd->ops->signal_event) + return 0; memcpy(ev_info.uid, lgr->id, SMC_LGR_ID_SIZE); ev_info.vlan_id = lgr->vlan_id; -- cgit v1.2.3 From c8df2d449f645f90a67aaa05a617e6e86c3f1c5e Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:33 +0800 Subject: net/smc: ignore loopback-ism when dumping SMC-D devices Since loopback-ism is not a PCI device, the PCI information fed back by smc_nl_handle_smcd_dev() does not apply to loopback-ism. So currently ignore loopback-ism when dumping SMC-D devices. The netlink function of loopback-ism will be refactored when SMC netlink interface is updated. Link: https://lore.kernel.org/r/caab067b-f5c3-490f-9259-262624c236b4@linux.ibm.com/ Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/smc_ism.c | 2 ++ net/smc/smc_ism.h | 5 +++++ 2 files changed, 7 insertions(+) (limited to 'net') diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c index 36459a3c739d..188fd28423c2 100644 --- a/net/smc/smc_ism.c +++ b/net/smc/smc_ism.c @@ -322,6 +322,8 @@ static void smc_nl_prep_smcd_dev(struct smcd_dev_list *dev_list, list_for_each_entry(smcd, &dev_list->list, list) { if (num < snum) goto next; + if (smc_ism_is_loopback(smcd)) + goto next; if (smc_nl_handle_smcd_dev(smcd, skb, cb)) goto errout; next: diff --git a/net/smc/smc_ism.h b/net/smc/smc_ism.h index 165cd013404b..322973527c61 100644 --- a/net/smc/smc_ism.h +++ b/net/smc/smc_ism.h @@ -84,4 +84,9 @@ static inline bool smc_ism_is_emulated(struct smcd_dev *smcd) return __smc_ism_is_emulated(chid); } +static inline bool smc_ism_is_loopback(struct smcd_dev *smcd) +{ + return (smcd->ops->get_chid(smcd) == 0xFFFF); +} + #endif -- cgit v1.2.3 From 04791343d858242dc2bd25e05be480ad1591e900 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:34 +0800 Subject: net/smc: register loopback-ism into SMC-D device list After the loopback-ism device is ready, add it to the SMC-D device list as an ISMv2 device, and always keep it at the beginning to ensure it is preferred for providing a shortcut for data transfer within the same kernel. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/smc_ism.c | 30 +++++++++++++++++++++--------- net/smc/smc_ism.h | 1 + net/smc/smc_loopback.c | 20 +++++++++++++------- 3 files changed, 35 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c index 188fd28423c2..6bed0a61b746 100644 --- a/net/smc/smc_ism.c +++ b/net/smc/smc_ism.c @@ -91,6 +91,11 @@ bool smc_ism_is_v2_capable(void) return smc_ism_v2_capable; } +void smc_ism_set_v2_capable(void) +{ + smc_ism_v2_capable = true; +} + /* Set a connection using this DMBE. */ void smc_ism_set_conn(struct smc_connection *conn) { @@ -439,7 +444,7 @@ static struct smcd_dev *smcd_alloc_dev(struct device *parent, const char *name, static void smcd_register_dev(struct ism_dev *ism) { const struct smcd_ops *ops = ism_get_smcd_ops(); - struct smcd_dev *smcd; + struct smcd_dev *smcd, *fentry; if (!ops) return; @@ -454,16 +459,23 @@ static void smcd_register_dev(struct ism_dev *ism) if (smc_pnetid_by_dev_port(&ism->pdev->dev, 0, smcd->pnetid)) smc_pnetid_by_table_smcd(smcd); + if (smcd->ops->supports_v2()) + smc_ism_set_v2_capable(); mutex_lock(&smcd_dev_list.mutex); - if (list_empty(&smcd_dev_list.list)) { - if (smcd->ops->supports_v2()) - smc_ism_v2_capable = true; - } - /* sort list: devices without pnetid before devices with pnetid */ - if (smcd->pnetid[0]) + /* sort list: + * - devices without pnetid before devices with pnetid; + * - loopback-ism always at the very beginning; + */ + if (!smcd->pnetid[0]) { + fentry = list_first_entry_or_null(&smcd_dev_list.list, + struct smcd_dev, list); + if (fentry && smc_ism_is_loopback(fentry)) + list_add(&smcd->list, &fentry->list); + else + list_add(&smcd->list, &smcd_dev_list.list); + } else { list_add_tail(&smcd->list, &smcd_dev_list.list); - else - list_add(&smcd->list, &smcd_dev_list.list); + } mutex_unlock(&smcd_dev_list.mutex); pr_warn_ratelimited("smc: adding smcd device %s with pnetid %.16s%s\n", diff --git a/net/smc/smc_ism.h b/net/smc/smc_ism.h index 322973527c61..e6f57e5e1ef9 100644 --- a/net/smc/smc_ism.h +++ b/net/smc/smc_ism.h @@ -52,6 +52,7 @@ int smc_ism_signal_shutdown(struct smc_link_group *lgr); void smc_ism_get_system_eid(u8 **eid); u16 smc_ism_get_chid(struct smcd_dev *dev); bool smc_ism_is_v2_capable(void); +void smc_ism_set_v2_capable(void); int smc_ism_init(void); void smc_ism_exit(void); int smcd_nl_get_device(struct sk_buff *skb, struct netlink_callback *cb); diff --git a/net/smc/smc_loopback.c b/net/smc/smc_loopback.c index d75852549b00..94a57f4ee3f9 100644 --- a/net/smc/smc_loopback.c +++ b/net/smc/smc_loopback.c @@ -246,10 +246,12 @@ static int smcd_lo_register_dev(struct smc_lo_dev *ldev) return -ENOMEM; ldev->smcd = smcd; smcd->priv = ldev; - - /* TODO: - * register loopback-ism to smcd_dev list. - */ + smc_ism_set_v2_capable(); + mutex_lock(&smcd_dev_list.mutex); + list_add(&smcd->list, &smcd_dev_list.list); + mutex_unlock(&smcd_dev_list.mutex); + pr_warn_ratelimited("smc: adding smcd device %s\n", + dev_name(&ldev->dev)); return 0; } @@ -257,9 +259,13 @@ static void smcd_lo_unregister_dev(struct smc_lo_dev *ldev) { struct smcd_dev *smcd = ldev->smcd; - /* TODO: - * unregister loopback-ism from smcd_dev list. - */ + pr_warn_ratelimited("smc: removing smcd device %s\n", + dev_name(&ldev->dev)); + smcd->going_away = 1; + smc_smcd_terminate_all(smcd); + mutex_lock(&smcd_dev_list.mutex); + list_del_init(&smcd->list); + mutex_unlock(&smcd_dev_list.mutex); kfree(smcd->conn); kfree(smcd); } -- cgit v1.2.3 From 4398888268582cb51b69c6ee94f551bb8d37d12f Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:35 +0800 Subject: net/smc: add operations to merge sndbuf with peer DMB In some scenarios using Emulated-ISM device, sndbuf can share the same physical memory region with peer DMB to avoid data copy from one side to the other. In such case the sndbuf is only a descriptor that describes the shared memory and does not actually occupy memory, it's more like a ghost buffer. +----------+ +----------+ | socket A | | socket B | +----------+ +----------+ | | +--------+ +--------+ | sndbuf | | DMB | | desc | | desc | +--------+ +--------+ | | | +----v-----+ +--------------------------> memory | +----------+ So here introduces three new SMC-D device operations to check if this feature is supported by device, and to {attach|detach} ghost sndbuf to peer DMB. For now only loopback-ism supports this. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- include/net/smc.h | 3 +++ net/smc/smc_ism.c | 40 ++++++++++++++++++++++++++++++++++++++++ net/smc/smc_ism.h | 4 ++++ 3 files changed, 47 insertions(+) (limited to 'net') diff --git a/include/net/smc.h b/include/net/smc.h index 33b753115e43..db84e4e35080 100644 --- a/include/net/smc.h +++ b/include/net/smc.h @@ -74,6 +74,9 @@ struct smcd_ops { int (*reset_vlan_required)(struct smcd_dev *dev); int (*signal_event)(struct smcd_dev *dev, struct smcd_gid *rgid, u32 trigger_irq, u32 event_code, u64 info); + int (*support_dmb_nocopy)(struct smcd_dev *dev); + int (*attach_dmb)(struct smcd_dev *dev, struct smcd_dmb *dmb); + int (*detach_dmb)(struct smcd_dev *dev, u64 token); }; struct smcd_dev { diff --git a/net/smc/smc_ism.c b/net/smc/smc_ism.c index 6bed0a61b746..84f98e18c7db 100644 --- a/net/smc/smc_ism.c +++ b/net/smc/smc_ism.c @@ -250,6 +250,46 @@ int smc_ism_register_dmb(struct smc_link_group *lgr, int dmb_len, return rc; } +bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd) +{ + /* for now only loopback-ism supports + * merging sndbuf with peer DMB to avoid + * data copies between them. + */ + return (smcd->ops->support_dmb_nocopy && + smcd->ops->support_dmb_nocopy(smcd)); +} + +int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token, + struct smc_buf_desc *dmb_desc) +{ + struct smcd_dmb dmb; + int rc = 0; + + if (!dev->ops->attach_dmb) + return -EINVAL; + + memset(&dmb, 0, sizeof(dmb)); + dmb.dmb_tok = token; + rc = dev->ops->attach_dmb(dev, &dmb); + if (!rc) { + dmb_desc->sba_idx = dmb.sba_idx; + dmb_desc->token = dmb.dmb_tok; + dmb_desc->cpu_addr = dmb.cpu_addr; + dmb_desc->dma_addr = dmb.dma_addr; + dmb_desc->len = dmb.dmb_len; + } + return rc; +} + +int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token) +{ + if (!dev->ops->detach_dmb) + return -EINVAL; + + return dev->ops->detach_dmb(dev, token); +} + static int smc_nl_handle_smcd_dev(struct smcd_dev *smcd, struct sk_buff *skb, struct netlink_callback *cb) diff --git a/net/smc/smc_ism.h b/net/smc/smc_ism.h index e6f57e5e1ef9..6763133dd8d0 100644 --- a/net/smc/smc_ism.h +++ b/net/smc/smc_ism.h @@ -48,6 +48,10 @@ int smc_ism_put_vlan(struct smcd_dev *dev, unsigned short vlan_id); int smc_ism_register_dmb(struct smc_link_group *lgr, int buf_size, struct smc_buf_desc *dmb_desc); int smc_ism_unregister_dmb(struct smcd_dev *dev, struct smc_buf_desc *dmb_desc); +bool smc_ism_support_dmb_nocopy(struct smcd_dev *smcd); +int smc_ism_attach_dmb(struct smcd_dev *dev, u64 token, + struct smc_buf_desc *dmb_desc); +int smc_ism_detach_dmb(struct smcd_dev *dev, u64 token); int smc_ism_signal_shutdown(struct smc_link_group *lgr); void smc_ism_get_system_eid(u8 **eid); u16 smc_ism_get_chid(struct smcd_dev *dev); -- cgit v1.2.3 From ae2be35cbed2c8385e890147ea321a3fcc3ca5fa Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:36 +0800 Subject: net/smc: {at|de}tach sndbuf to peer DMB if supported If the device used by SMC-D supports merging local sndbuf to peer DMB, then create sndbuf descriptor and attach it to peer DMB once peer token is obtained, and detach and free the sndbuf descriptor when the connection is freed. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/af_smc.c | 16 ++++++++++++++ net/smc/smc_core.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++++++++- net/smc/smc_core.h | 1 + 3 files changed, 77 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 47f3bc1470bc..9389f0cfa374 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1436,6 +1436,14 @@ static int smc_connect_ism(struct smc_sock *smc, } smc_conn_save_peer_info(smc, aclc); + + if (smc_ism_support_dmb_nocopy(smc->conn.lgr->smcd)) { + rc = smcd_buf_attach(smc); + if (rc) { + rc = SMC_CLC_DECL_MEM; /* try to fallback */ + goto connect_abort; + } + } smc_close_init(smc); smc_rx_init(smc); smc_tx_init(smc); @@ -2540,6 +2548,14 @@ static void smc_listen_work(struct work_struct *work) mutex_unlock(&smc_server_lgr_pending); } smc_conn_save_peer_info(new_smc, cclc); + + if (ini->is_smcd && + smc_ism_support_dmb_nocopy(new_smc->conn.lgr->smcd)) { + rc = smcd_buf_attach(new_smc); + if (rc) + goto out_decl; + } + smc_listen_out_connected(new_smc); SMC_STAT_SERV_SUCC_INC(sock_net(newclcsock->sk), ini); goto out_free; diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 9b84d5897aa5..fafdb97adfad 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -1149,6 +1149,20 @@ static void smcr_buf_unuse(struct smc_buf_desc *buf_desc, bool is_rmb, } } +static void smcd_buf_detach(struct smc_connection *conn) +{ + struct smcd_dev *smcd = conn->lgr->smcd; + u64 peer_token = conn->peer_token; + + if (!conn->sndbuf_desc) + return; + + smc_ism_detach_dmb(smcd, peer_token); + + kfree(conn->sndbuf_desc); + conn->sndbuf_desc = NULL; +} + static void smc_buf_unuse(struct smc_connection *conn, struct smc_link_group *lgr) { @@ -1192,6 +1206,8 @@ void smc_conn_free(struct smc_connection *conn) if (lgr->is_smcd) { if (!list_empty(&lgr->list)) smc_ism_unset_conn(conn); + if (smc_ism_support_dmb_nocopy(lgr->smcd)) + smcd_buf_detach(conn); tasklet_kill(&conn->rx_tsklet); } else { smc_cdc_wait_pend_tx_wr(conn); @@ -1445,6 +1461,8 @@ static void smc_conn_kill(struct smc_connection *conn, bool soft) smc_sk_wake_ups(smc); if (conn->lgr->is_smcd) { smc_ism_unset_conn(conn); + if (smc_ism_support_dmb_nocopy(conn->lgr->smcd)) + smcd_buf_detach(conn); if (soft) tasklet_kill(&conn->rx_tsklet); else @@ -2464,12 +2482,18 @@ int smc_buf_create(struct smc_sock *smc, bool is_smcd) int rc; /* create send buffer */ + if (is_smcd && + smc_ism_support_dmb_nocopy(smc->conn.lgr->smcd)) + goto create_rmb; + rc = __smc_buf_create(smc, is_smcd, false); if (rc) return rc; + +create_rmb: /* create rmb */ rc = __smc_buf_create(smc, is_smcd, true); - if (rc) { + if (rc && smc->conn.sndbuf_desc) { down_write(&smc->conn.lgr->sndbufs_lock); list_del(&smc->conn.sndbuf_desc->list); up_write(&smc->conn.lgr->sndbufs_lock); @@ -2479,6 +2503,41 @@ int smc_buf_create(struct smc_sock *smc, bool is_smcd) return rc; } +int smcd_buf_attach(struct smc_sock *smc) +{ + struct smc_connection *conn = &smc->conn; + struct smcd_dev *smcd = conn->lgr->smcd; + u64 peer_token = conn->peer_token; + struct smc_buf_desc *buf_desc; + int rc; + + buf_desc = kzalloc(sizeof(*buf_desc), GFP_KERNEL); + if (!buf_desc) + return -ENOMEM; + + /* The ghost sndbuf_desc describes the same memory region as + * peer RMB. Its lifecycle is consistent with the connection's + * and it will be freed with the connections instead of the + * link group. + */ + rc = smc_ism_attach_dmb(smcd, peer_token, buf_desc); + if (rc) + goto free; + + smc->sk.sk_sndbuf = buf_desc->len; + buf_desc->cpu_addr = + (u8 *)buf_desc->cpu_addr + sizeof(struct smcd_cdc_msg); + buf_desc->len -= sizeof(struct smcd_cdc_msg); + conn->sndbuf_desc = buf_desc; + conn->sndbuf_desc->used = 1; + atomic_set(&conn->sndbuf_space, conn->sndbuf_desc->len); + return 0; + +free: + kfree(buf_desc); + return rc; +} + static inline int smc_rmb_reserve_rtoken_idx(struct smc_link_group *lgr) { int i; diff --git a/net/smc/smc_core.h b/net/smc/smc_core.h index 1f175376037b..d93cf51dbd7c 100644 --- a/net/smc/smc_core.h +++ b/net/smc/smc_core.h @@ -557,6 +557,7 @@ void smc_smcd_terminate(struct smcd_dev *dev, struct smcd_gid *peer_gid, void smc_smcd_terminate_all(struct smcd_dev *dev); void smc_smcr_terminate_all(struct smc_ib_device *smcibdev); int smc_buf_create(struct smc_sock *smc, bool is_smcd); +int smcd_buf_attach(struct smc_sock *smc); int smc_uncompress_bufsize(u8 compressed); int smc_rmb_rtoken_handling(struct smc_connection *conn, struct smc_link *link, struct smc_clc_msg_accept_confirm *clc); -- cgit v1.2.3 From cc0ab806fc52e77f961b275ebb58024bd0e7adf2 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:37 +0800 Subject: net/smc: adapt cursor update when sndbuf and peer DMB are merged If the local sndbuf shares the same physical memory with peer DMB, the cursor update processing needs to be adapted to ensure that the data to be consumed won't be overwritten. So in this case, the fin_curs and sndbuf_space that were originally updated after sending the CDC message should be modified to not be update until the peer updates cons_curs. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/smc_cdc.c | 36 ++++++++++++++++++++++++++++++++++-- 1 file changed, 34 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c index 3c06625ceb20..619b3bab3824 100644 --- a/net/smc/smc_cdc.c +++ b/net/smc/smc_cdc.c @@ -18,6 +18,7 @@ #include "smc_tx.h" #include "smc_rx.h" #include "smc_close.h" +#include "smc_ism.h" /********************************** send *************************************/ @@ -255,6 +256,14 @@ int smcd_cdc_msg_send(struct smc_connection *conn) return rc; smc_curs_copy(&conn->rx_curs_confirmed, &curs, conn); conn->local_rx_ctrl.prod_flags.cons_curs_upd_req = 0; + + if (smc_ism_support_dmb_nocopy(conn->lgr->smcd)) + /* if local sndbuf shares the same memory region with + * peer DMB, then don't update the tx_curs_fin + * and sndbuf_space until peer has consumed the data. + */ + return 0; + /* Calculate transmitted data and increment free send buffer space */ diff = smc_curs_diff(conn->sndbuf_desc->len, &conn->tx_curs_fin, &conn->tx_curs_sent); @@ -266,7 +275,7 @@ int smcd_cdc_msg_send(struct smc_connection *conn) smc_curs_copy(&conn->tx_curs_fin, &conn->tx_curs_sent, conn); smc_tx_sndbuf_nonfull(smc); - return rc; + return 0; } /********************************* receive ***********************************/ @@ -323,7 +332,7 @@ static void smc_cdc_msg_recv_action(struct smc_sock *smc, { union smc_host_cursor cons_old, prod_old; struct smc_connection *conn = &smc->conn; - int diff_cons, diff_prod; + int diff_cons, diff_prod, diff_tx; smc_curs_copy(&prod_old, &conn->local_rx_ctrl.prod, conn); smc_curs_copy(&cons_old, &conn->local_rx_ctrl.cons, conn); @@ -339,6 +348,29 @@ static void smc_cdc_msg_recv_action(struct smc_sock *smc, atomic_add(diff_cons, &conn->peer_rmbe_space); /* guarantee 0 <= peer_rmbe_space <= peer_rmbe_size */ smp_mb__after_atomic(); + + /* if local sndbuf shares the same memory region with + * peer RMB, then update tx_curs_fin and sndbuf_space + * here since peer has already consumed the data. + */ + if (conn->lgr->is_smcd && + smc_ism_support_dmb_nocopy(conn->lgr->smcd)) { + /* Calculate consumed data and + * increment free send buffer space. + */ + diff_tx = smc_curs_diff(conn->sndbuf_desc->len, + &conn->tx_curs_fin, + &conn->local_rx_ctrl.cons); + /* increase local sndbuf space and fin_curs */ + smp_mb__before_atomic(); + atomic_add(diff_tx, &conn->sndbuf_space); + /* guarantee 0 <= sndbuf_space <= sndbuf_desc->len */ + smp_mb__after_atomic(); + smc_curs_copy(&conn->tx_curs_fin, + &conn->local_rx_ctrl.cons, conn); + + smc_tx_sndbuf_nonfull(smc); + } } diff_prod = smc_curs_diff(conn->rmb_desc->len, &prod_old, -- cgit v1.2.3 From c3a910f2380fe294d14e42af66af3d3eed8fecbf Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Sun, 28 Apr 2024 14:07:38 +0800 Subject: net/smc: implement DMB-merged operations of loopback-ism This implements operations related to merging sndbuf with peer DMB in loopback-ism. The DMB won't be freed until no sndbuf is attached to it. Signed-off-by: Wen Gu Reviewed-by: Wenjia Zhang Reviewed-and-tested-by: Jan Karcher Signed-off-by: Paolo Abeni --- net/smc/smc_loopback.c | 120 ++++++++++++++++++++++++++++++++++++++++++------- net/smc/smc_loopback.h | 3 ++ 2 files changed, 108 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/smc/smc_loopback.c b/net/smc/smc_loopback.c index 94a57f4ee3f9..3c5f64ca4115 100644 --- a/net/smc/smc_loopback.c +++ b/net/smc/smc_loopback.c @@ -20,6 +20,7 @@ #include "smc_loopback.h" #define SMC_LO_V2_CAPABLE 0x1 /* loopback-ism acts as ISMv2 */ +#define SMC_LO_SUPPORT_NOCOPY 0x1 #define SMC_DMA_ADDR_INVALID (~(dma_addr_t)0) static const char smc_lo_dev_name[] = "loopback-ism"; @@ -81,6 +82,7 @@ static int smc_lo_register_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb, goto err_node; } dmb_node->dma_addr = SMC_DMA_ADDR_INVALID; + refcount_set(&dmb_node->refcnt, 1); again: /* add new dmb into hash table */ @@ -94,6 +96,7 @@ again: } hash_add(ldev->dmb_ht, &dmb_node->list, dmb_node->token); write_unlock_bh(&ldev->dmb_ht_lock); + atomic_inc(&ldev->dmb_cnt); dmb->sba_idx = dmb_node->sba_idx; dmb->dmb_tok = dmb_node->token; @@ -110,13 +113,29 @@ err_bit: return rc; } +static void __smc_lo_unregister_dmb(struct smc_lo_dev *ldev, + struct smc_lo_dmb_node *dmb_node) +{ + /* remove dmb from hash table */ + write_lock_bh(&ldev->dmb_ht_lock); + hash_del(&dmb_node->list); + write_unlock_bh(&ldev->dmb_ht_lock); + + clear_bit(dmb_node->sba_idx, ldev->sba_idx_mask); + kvfree(dmb_node->cpu_addr); + kfree(dmb_node); + + if (atomic_dec_and_test(&ldev->dmb_cnt)) + wake_up(&ldev->ldev_release); +} + static int smc_lo_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb) { struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node; struct smc_lo_dev *ldev = smcd->priv; - /* remove dmb from hash table */ - write_lock_bh(&ldev->dmb_ht_lock); + /* find dmb from hash table */ + read_lock_bh(&ldev->dmb_ht_lock); hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb->dmb_tok) { if (tmp_node->token == dmb->dmb_tok) { dmb_node = tmp_node; @@ -124,16 +143,76 @@ static int smc_lo_unregister_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb) } } if (!dmb_node) { - write_unlock_bh(&ldev->dmb_ht_lock); + read_unlock_bh(&ldev->dmb_ht_lock); return -EINVAL; } - hash_del(&dmb_node->list); - write_unlock_bh(&ldev->dmb_ht_lock); + read_unlock_bh(&ldev->dmb_ht_lock); - clear_bit(dmb_node->sba_idx, ldev->sba_idx_mask); - kfree(dmb_node->cpu_addr); - kfree(dmb_node); + if (refcount_dec_and_test(&dmb_node->refcnt)) + __smc_lo_unregister_dmb(ldev, dmb_node); + return 0; +} + +static int smc_lo_support_dmb_nocopy(struct smcd_dev *smcd) +{ + return SMC_LO_SUPPORT_NOCOPY; +} + +static int smc_lo_attach_dmb(struct smcd_dev *smcd, struct smcd_dmb *dmb) +{ + struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node; + struct smc_lo_dev *ldev = smcd->priv; + + /* find dmb_node according to dmb->dmb_tok */ + read_lock_bh(&ldev->dmb_ht_lock); + hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb->dmb_tok) { + if (tmp_node->token == dmb->dmb_tok) { + dmb_node = tmp_node; + break; + } + } + if (!dmb_node) { + read_unlock_bh(&ldev->dmb_ht_lock); + return -EINVAL; + } + read_unlock_bh(&ldev->dmb_ht_lock); + + if (!refcount_inc_not_zero(&dmb_node->refcnt)) + /* the dmb is being unregistered, but has + * not been removed from the hash table. + */ + return -EINVAL; + /* provide dmb information */ + dmb->sba_idx = dmb_node->sba_idx; + dmb->dmb_tok = dmb_node->token; + dmb->cpu_addr = dmb_node->cpu_addr; + dmb->dma_addr = dmb_node->dma_addr; + dmb->dmb_len = dmb_node->len; + return 0; +} + +static int smc_lo_detach_dmb(struct smcd_dev *smcd, u64 token) +{ + struct smc_lo_dmb_node *dmb_node = NULL, *tmp_node; + struct smc_lo_dev *ldev = smcd->priv; + + /* find dmb_node according to dmb->dmb_tok */ + read_lock_bh(&ldev->dmb_ht_lock); + hash_for_each_possible(ldev->dmb_ht, tmp_node, list, token) { + if (tmp_node->token == token) { + dmb_node = tmp_node; + break; + } + } + if (!dmb_node) { + read_unlock_bh(&ldev->dmb_ht_lock); + return -EINVAL; + } + read_unlock_bh(&ldev->dmb_ht_lock); + + if (refcount_dec_and_test(&dmb_node->refcnt)) + __smc_lo_unregister_dmb(ldev, dmb_node); return 0; } @@ -145,6 +224,12 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, struct smc_lo_dev *ldev = smcd->priv; struct smc_connection *conn; + if (!sf) + /* since sndbuf is merged with peer DMB, there is + * no need to copy data from sndbuf to peer DMB. + */ + return 0; + read_lock_bh(&ldev->dmb_ht_lock); hash_for_each_possible(ldev->dmb_ht, tmp_node, list, dmb_tok) { if (tmp_node->token == dmb_tok) { @@ -159,13 +244,10 @@ static int smc_lo_move_data(struct smcd_dev *smcd, u64 dmb_tok, memcpy((char *)rmb_node->cpu_addr + offset, data, size); read_unlock_bh(&ldev->dmb_ht_lock); - if (sf) { - conn = smcd->conn[rmb_node->sba_idx]; - if (conn && !conn->killed) - tasklet_schedule(&conn->rx_tsklet); - else - return -EPIPE; - } + conn = smcd->conn[rmb_node->sba_idx]; + if (!conn || conn->killed) + return -EPIPE; + tasklet_schedule(&conn->rx_tsklet); return 0; } @@ -197,6 +279,9 @@ static const struct smcd_ops lo_ops = { .query_remote_gid = smc_lo_query_rgid, .register_dmb = smc_lo_register_dmb, .unregister_dmb = smc_lo_unregister_dmb, + .support_dmb_nocopy = smc_lo_support_dmb_nocopy, + .attach_dmb = smc_lo_attach_dmb, + .detach_dmb = smc_lo_detach_dmb, .add_vlan_id = NULL, .del_vlan_id = NULL, .set_vlan_required = NULL, @@ -275,12 +360,17 @@ static int smc_lo_dev_init(struct smc_lo_dev *ldev) smc_lo_generate_ids(ldev); rwlock_init(&ldev->dmb_ht_lock); hash_init(ldev->dmb_ht); + atomic_set(&ldev->dmb_cnt, 0); + init_waitqueue_head(&ldev->ldev_release); + return smcd_lo_register_dev(ldev); } static void smc_lo_dev_exit(struct smc_lo_dev *ldev) { smcd_lo_unregister_dev(ldev); + if (atomic_read(&ldev->dmb_cnt)) + wait_event(ldev->ldev_release, !atomic_read(&ldev->dmb_cnt)); } static void smc_lo_dev_release(struct device *dev) diff --git a/net/smc/smc_loopback.h b/net/smc/smc_loopback.h index b8206338309f..6dd4292dae56 100644 --- a/net/smc/smc_loopback.h +++ b/net/smc/smc_loopback.h @@ -30,6 +30,7 @@ struct smc_lo_dmb_node { u32 sba_idx; void *cpu_addr; dma_addr_t dma_addr; + refcount_t refcnt; }; struct smc_lo_dev { @@ -37,9 +38,11 @@ struct smc_lo_dev { struct device dev; u16 chid; struct smcd_gid local_gid; + atomic_t dmb_cnt; rwlock_t dmb_ht_lock; DECLARE_BITMAP(sba_idx_mask, SMC_LO_MAX_DMBS); DECLARE_HASHTABLE(dmb_ht, SMC_LO_DMBS_HASH_BITS); + wait_queue_head_t ldev_release; }; int smc_loopback_init(void); -- cgit v1.2.3 From 12b6c3a0380a220edb03691689167be13faa4f45 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 29 Apr 2024 07:44:21 -0700 Subject: net: page_pool: support error injection Because of caching / recycling using the general page allocation failures to induce errors in page pool allocation is very hard. Add direct error injection support to page_pool_alloc_pages(). Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/20240429144426.743476-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/core/page_pool.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 273c24429bce..8bcc7014a61a 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -5,6 +5,7 @@ * Copyright (C) 2016 Red Hat, Inc. */ +#include #include #include #include @@ -550,6 +551,7 @@ struct page *page_pool_alloc_pages(struct page_pool *pool, gfp_t gfp) return page; } EXPORT_SYMBOL(page_pool_alloc_pages); +ALLOW_ERROR_INJECTION(page_pool_alloc_pages, NULL); /* Calculate distance between two u32 values, valid if distance is below 2^(31) * https://en.wikipedia.org/wiki/Serial_number_arithmetic#General_Solution -- cgit v1.2.3 From 9f8eeea1643c213c0e1ad2e546a15536200d216b Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Sun, 28 Apr 2024 19:16:38 +0800 Subject: rxrpc: Fix using alignmask being zero for __page_frag_alloc_align() rxrpc_alloc_data_txbuf() may be called with data_align being zero in none_alloc_txbuf() and rxkad_alloc_txbuf(), data_align is supposed to be an order-based alignment value, but zero is not a valid order-based alignment value, and '~(data_align - 1)' doesn't result in a valid mask-based alignment value for __page_frag_alloc_align(). Fix it by passing a valid order-based alignment value in none_alloc_txbuf() and rxkad_alloc_txbuf(). Also use page_frag_alloc_align() expecting an order-based alignment value in rxrpc_alloc_data_txbuf() to avoid doing the alignment converting operation and to catch possible invalid alignment value in the future. Remove the 'if (data_align)' checking too, as it is always true for a valid order-based alignment value. Fixes: 6b2536462fd4 ("rxrpc: Fix use of changed alignment param to page_frag_alloc_align()") Fixes: 49489bb03a50 ("rxrpc: Do zerocopy using MSG_SPLICE_PAGES and page frags") CC: Alexander Duyck Signed-off-by: Yunsheng Lin Acked-by: David Howells Link: https://lore.kernel.org/r/20240428111640.27306-1-linyunsheng@huawei.com Signed-off-by: Jakub Kicinski --- net/rxrpc/insecure.c | 2 +- net/rxrpc/rxkad.c | 2 +- net/rxrpc/txbuf.c | 10 +++++----- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/rxrpc/insecure.c b/net/rxrpc/insecure.c index f2701068ed9e..6716c021a532 100644 --- a/net/rxrpc/insecure.c +++ b/net/rxrpc/insecure.c @@ -19,7 +19,7 @@ static int none_init_connection_security(struct rxrpc_connection *conn, */ static struct rxrpc_txbuf *none_alloc_txbuf(struct rxrpc_call *call, size_t remain, gfp_t gfp) { - return rxrpc_alloc_data_txbuf(call, min_t(size_t, remain, RXRPC_JUMBO_DATALEN), 0, gfp); + return rxrpc_alloc_data_txbuf(call, min_t(size_t, remain, RXRPC_JUMBO_DATALEN), 1, gfp); } static int none_secure_packet(struct rxrpc_call *call, struct rxrpc_txbuf *txb) diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index f1a68270862d..48a1475e6b06 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -155,7 +155,7 @@ static struct rxrpc_txbuf *rxkad_alloc_txbuf(struct rxrpc_call *call, size_t rem switch (call->conn->security_level) { default: space = min_t(size_t, remain, RXRPC_JUMBO_DATALEN); - return rxrpc_alloc_data_txbuf(call, space, 0, gfp); + return rxrpc_alloc_data_txbuf(call, space, 1, gfp); case RXRPC_SECURITY_AUTH: shdr = sizeof(struct rxkad_level1_hdr); break; diff --git a/net/rxrpc/txbuf.c b/net/rxrpc/txbuf.c index e0679658d9de..c3913d8a50d3 100644 --- a/net/rxrpc/txbuf.c +++ b/net/rxrpc/txbuf.c @@ -21,20 +21,20 @@ struct rxrpc_txbuf *rxrpc_alloc_data_txbuf(struct rxrpc_call *call, size_t data_ { struct rxrpc_wire_header *whdr; struct rxrpc_txbuf *txb; - size_t total, hoff = 0; + size_t total, hoff; void *buf; txb = kmalloc(sizeof(*txb), gfp); if (!txb) return NULL; - if (data_align) - hoff = round_up(sizeof(*whdr), data_align) - sizeof(*whdr); + hoff = round_up(sizeof(*whdr), data_align) - sizeof(*whdr); total = hoff + sizeof(*whdr) + data_size; + data_align = umax(data_align, L1_CACHE_BYTES); mutex_lock(&call->conn->tx_data_alloc_lock); - buf = __page_frag_alloc_align(&call->conn->tx_data_alloc, total, gfp, - ~(data_align - 1) & ~(L1_CACHE_BYTES - 1)); + buf = page_frag_alloc_align(&call->conn->tx_data_alloc, total, gfp, + data_align); mutex_unlock(&call->conn->tx_data_alloc_lock); if (!buf) { kfree(txb); -- cgit v1.2.3 From 05d6d492097c55f2d153fc3fd33cbe78e1e28e0a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 13:30:09 +0000 Subject: inet: introduce dst_rtable() helper I added dst_rt6_info() in commit e8dfd42c17fa ("ipv6: introduce dst_rt6_info() helper") This patch does a similar change for IPv4. Instead of (struct rtable *)dst casts, we can use : #define dst_rtable(_ptr) \ container_of_const(_ptr, struct rtable, dst) Patch is smaller than IPv6 one, because IPv4 has skb_rtable() helper. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Reviewed-by: Sabrina Dubroca Link: https://lore.kernel.org/r/20240429133009.1227754-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/infiniband/core/addr.c | 12 +++--------- drivers/net/vrf.c | 2 +- drivers/s390/net/qeth_core.h | 5 ++--- include/linux/skbuff.h | 9 --------- include/net/ip.h | 4 ++-- include/net/route.h | 11 +++++++++++ net/atm/clip.c | 2 +- net/core/dst_cache.c | 2 +- net/core/filter.c | 3 +-- net/ipv4/af_inet.c | 2 +- net/ipv4/icmp.c | 26 ++++++++++++++------------ net/ipv4/ip_input.c | 2 +- net/ipv4/ip_output.c | 8 ++++---- net/ipv4/route.c | 24 +++++++++++------------- net/ipv4/udp.c | 2 +- net/ipv4/xfrm4_policy.c | 2 +- net/l2tp/l2tp_ip.c | 2 +- net/mpls/mpls_iptunnel.c | 2 +- net/netfilter/ipvs/ip_vs_xmit.c | 2 +- net/netfilter/nf_flow_table_ip.c | 4 ++-- net/netfilter/nft_rt.c | 2 +- net/sctp/protocol.c | 4 ++-- net/tipc/udp_media.c | 2 +- 23 files changed, 64 insertions(+), 70 deletions(-) (limited to 'net') diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index f20dfe70fa0e..be0743dac3ff 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -348,16 +348,10 @@ static int dst_fetch_ha(const struct dst_entry *dst, static bool has_gateway(const struct dst_entry *dst, sa_family_t family) { - const struct rtable *rt; - const struct rt6_info *rt6; + if (family == AF_INET) + return dst_rtable(dst)->rt_uses_gateway; - if (family == AF_INET) { - rt = container_of(dst, struct rtable, dst); - return rt->rt_uses_gateway; - } - - rt6 = dst_rt6_info(dst); - return rt6->rt6i_flags & RTF_GATEWAY; + return dst_rt6_info(dst)->rt6i_flags & RTF_GATEWAY; } static int fetch_ha(const struct dst_entry *dst, struct rdma_dev_addr *dev_addr, diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 784b9b2d275e..3a252ac5dd28 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -860,7 +860,7 @@ static int vrf_rt6_create(struct net_device *dev) static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); - struct rtable *rt = (struct rtable *)dst; + struct rtable *rt = dst_rtable(dst); struct net_device *dev = dst->dev; unsigned int hh_len = LL_RESERVED_SPACE(dev); struct neighbour *neigh; diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index 5f17a2a5d0e3..41fe8a043d61 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -970,9 +970,8 @@ static inline struct dst_entry *qeth_dst_check_rcu(struct sk_buff *skb, static inline __be32 qeth_next_hop_v4_rcu(struct sk_buff *skb, struct dst_entry *dst) { - struct rtable *rt = (struct rtable *) dst; - - return (rt) ? rt_nexthop(rt, ip_hdr(skb)->daddr) : ip_hdr(skb)->daddr; + return (dst) ? rt_nexthop(dst_rtable(dst), ip_hdr(skb)->daddr) : + ip_hdr(skb)->daddr; } static inline struct in6_addr *qeth_next_hop_v6_rcu(struct sk_buff *skb, diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index f76825e5b92a..adf75d69770c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1180,15 +1180,6 @@ static inline bool skb_dst_is_noref(const struct sk_buff *skb) return (skb->_skb_refdst & SKB_DST_NOREF) && skb_dst(skb); } -/** - * skb_rtable - Returns the skb &rtable - * @skb: buffer - */ -static inline struct rtable *skb_rtable(const struct sk_buff *skb) -{ - return (struct rtable *)skb_dst(skb); -} - /* For mangling skb->pkt_type from user space side from applications * such as nft, tc, etc, we only allow a conservative subset of * possible pkt_types to be set. diff --git a/include/net/ip.h b/include/net/ip.h index 25cb688bdc62..6d735e00d3f3 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -423,7 +423,7 @@ int ip_decrease_ttl(struct iphdr *iph) static inline int ip_mtu_locked(const struct dst_entry *dst) { - const struct rtable *rt = (const struct rtable *)dst; + const struct rtable *rt = dst_rtable(dst); return rt->rt_mtu_locked || dst_metric_locked(dst, RTAX_MTU); } @@ -461,7 +461,7 @@ static inline bool ip_sk_ignore_df(const struct sock *sk) static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, bool forwarding) { - const struct rtable *rt = container_of(dst, struct rtable, dst); + const struct rtable *rt = dst_rtable(dst); struct net *net = dev_net(dst->dev); unsigned int mtu; diff --git a/include/net/route.h b/include/net/route.h index 630d1ef6868a..93833cfe9c96 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -75,6 +75,17 @@ struct rtable { rt_pmtu:31; }; +#define dst_rtable(_ptr) container_of_const(_ptr, struct rtable, dst) + +/** + * skb_rtable - Returns the skb &rtable + * @skb: buffer + */ +static inline struct rtable *skb_rtable(const struct sk_buff *skb) +{ + return dst_rtable(skb_dst(skb)); +} + static inline bool rt_is_input_route(const struct rtable *rt) { return rt->rt_is_input != 0; diff --git a/net/atm/clip.c b/net/atm/clip.c index 362e8d25a79e..42b910cb4e8e 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -345,7 +345,7 @@ static netdev_tx_t clip_start_xmit(struct sk_buff *skb, dev->stats.tx_dropped++; return NETDEV_TX_OK; } - rt = (struct rtable *) dst; + rt = dst_rtable(dst); if (rt->rt_gw_family == AF_INET) daddr = &rt->rt_gw4; else diff --git a/net/core/dst_cache.c b/net/core/dst_cache.c index b17171345d64..0c0bdb058c5b 100644 --- a/net/core/dst_cache.c +++ b/net/core/dst_cache.c @@ -83,7 +83,7 @@ struct rtable *dst_cache_get_ip4(struct dst_cache *dst_cache, __be32 *saddr) return NULL; *saddr = idst->in_saddr.s_addr; - return container_of(dst, struct rtable, dst); + return dst_rtable(dst); } EXPORT_SYMBOL_GPL(dst_cache_get_ip4); diff --git a/net/core/filter.c b/net/core/filter.c index 6d319c76188b..29165744c505 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2317,8 +2317,7 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb, rcu_read_lock(); if (!nh) { - struct dst_entry *dst = skb_dst(skb); - struct rtable *rt = container_of(dst, struct rtable, dst); + struct rtable *rt = skb_rtable(skb); neigh = ip_neigh_for_gw(rt, skb, &is_v6gw); } else if (nh->nh_family == AF_INET6) { diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a7cfeda28bb2..486a8d4f53b1 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1307,8 +1307,8 @@ static int inet_sk_reselect_saddr(struct sock *sk) int inet_sk_rebuild_header(struct sock *sk) { + struct rtable *rt = dst_rtable(__sk_dst_check(sk, 0)); struct inet_sock *inet = inet_sk(sk); - struct rtable *rt = (struct rtable *)__sk_dst_check(sk, 0); __be32 daddr; struct ip_options_rcu *inet_opt; struct flowi4 *fl4; diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 437e782b9663..207482d30dc7 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -483,6 +483,7 @@ static struct rtable *icmp_route_lookup(struct net *net, struct icmp_bxm *param) { struct net_device *route_lookup_dev; + struct dst_entry *dst, *dst2; struct rtable *rt, *rt2; struct flowi4 fl4_dec; int err; @@ -508,16 +509,17 @@ static struct rtable *icmp_route_lookup(struct net *net, /* No need to clone since we're just using its address. */ rt2 = rt; - rt = (struct rtable *) xfrm_lookup(net, &rt->dst, - flowi4_to_flowi(fl4), NULL, 0); - if (!IS_ERR(rt)) { + dst = xfrm_lookup(net, &rt->dst, + flowi4_to_flowi(fl4), NULL, 0); + rt = dst_rtable(dst); + if (!IS_ERR(dst)) { if (rt != rt2) return rt; - } else if (PTR_ERR(rt) == -EPERM) { + } else if (PTR_ERR(dst) == -EPERM) { rt = NULL; - } else + } else { return rt; - + } err = xfrm_decode_session_reverse(net, skb_in, flowi4_to_flowi(&fl4_dec), AF_INET); if (err) goto relookup_failed; @@ -551,19 +553,19 @@ static struct rtable *icmp_route_lookup(struct net *net, if (err) goto relookup_failed; - rt2 = (struct rtable *) xfrm_lookup(net, &rt2->dst, - flowi4_to_flowi(&fl4_dec), NULL, - XFRM_LOOKUP_ICMP); - if (!IS_ERR(rt2)) { + dst2 = xfrm_lookup(net, &rt2->dst, flowi4_to_flowi(&fl4_dec), NULL, + XFRM_LOOKUP_ICMP); + rt2 = dst_rtable(dst2); + if (!IS_ERR(dst2)) { dst_release(&rt->dst); memcpy(fl4, &fl4_dec, sizeof(*fl4)); rt = rt2; - } else if (PTR_ERR(rt2) == -EPERM) { + } else if (PTR_ERR(dst2) == -EPERM) { if (rt) dst_release(&rt->dst); return rt2; } else { - err = PTR_ERR(rt2); + err = PTR_ERR(dst2); goto relookup_failed; } return rt; diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 5e9c8156656a..d6fbcbd2358a 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -616,7 +616,7 @@ static void ip_list_rcv_finish(struct net *net, struct sock *sk, dst = skb_dst(skb); if (curr_dst != dst) { hint = ip_extract_route_hint(net, skb, - ((struct rtable *)dst)->rt_type); + dst_rtable(dst)->rt_type); /* dispatch old sublist */ if (!list_empty(&sublist)) diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 1fe794967211..b455bd05a7d5 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -198,7 +198,7 @@ EXPORT_SYMBOL_GPL(ip_build_and_send_pkt); static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); - struct rtable *rt = (struct rtable *)dst; + struct rtable *rt = dst_rtable(dst); struct net_device *dev = dst->dev; unsigned int hh_len = LL_RESERVED_SPACE(dev); struct neighbour *neigh; @@ -475,7 +475,7 @@ int __ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, goto packet_routed; /* Make sure we can route this packet. */ - rt = (struct rtable *)__sk_dst_check(sk, 0); + rt = dst_rtable(__sk_dst_check(sk, 0)); if (!rt) { __be32 daddr; @@ -971,7 +971,7 @@ static int __ip_append_data(struct sock *sk, bool zc = false; unsigned int maxfraglen, fragheaderlen, maxnonfragsize; int csummode = CHECKSUM_NONE; - struct rtable *rt = (struct rtable *)cork->dst; + struct rtable *rt = dst_rtable(cork->dst); bool paged, hold_tskey, extra_uref = false; unsigned int wmem_alloc_delta = 0; u32 tskey = 0; @@ -1390,7 +1390,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk, struct inet_sock *inet = inet_sk(sk); struct net *net = sock_net(sk); struct ip_options *opt = NULL; - struct rtable *rt = (struct rtable *)cork->dst; + struct rtable *rt = dst_rtable(cork->dst); struct iphdr *iph; u8 pmtudisc, ttl; __be16 df = 0; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index f89ff2e5a05b..0fd9a3d7ac4a 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -819,7 +819,7 @@ static void ip_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buf u32 mark = skb->mark; __u8 tos = iph->tos; - rt = (struct rtable *) dst; + rt = dst_rtable(dst); __build_flow_key(net, &fl4, sk, iph, oif, tos, prot, mark, 0); __ip_do_redirect(rt, skb, &fl4, true); @@ -827,7 +827,7 @@ static void ip_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buf static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst) { - struct rtable *rt = (struct rtable *)dst; + struct rtable *rt = dst_rtable(dst); struct dst_entry *ret = dst; if (rt) { @@ -1044,7 +1044,7 @@ static void ip_rt_update_pmtu(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb, u32 mtu, bool confirm_neigh) { - struct rtable *rt = (struct rtable *) dst; + struct rtable *rt = dst_rtable(dst); struct flowi4 fl4; ip_rt_build_flow_key(&fl4, sk, skb); @@ -1115,7 +1115,7 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) __build_flow_key(net, &fl4, sk, iph, 0, 0, 0, 0, 0); - rt = (struct rtable *)odst; + rt = dst_rtable(odst); if (odst->obsolete && !odst->ops->check(odst, 0)) { rt = ip_route_output_flow(sock_net(sk), &fl4, sk); if (IS_ERR(rt)) @@ -1124,7 +1124,7 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) new = true; } - __ip_rt_update_pmtu((struct rtable *)xfrm_dst_path(&rt->dst), &fl4, mtu); + __ip_rt_update_pmtu(dst_rtable(xfrm_dst_path(&rt->dst)), &fl4, mtu); if (!dst_check(&rt->dst, 0)) { if (new) @@ -1181,7 +1181,7 @@ EXPORT_SYMBOL_GPL(ipv4_sk_redirect); INDIRECT_CALLABLE_SCOPE struct dst_entry *ipv4_dst_check(struct dst_entry *dst, u32 cookie) { - struct rtable *rt = (struct rtable *) dst; + struct rtable *rt = dst_rtable(dst); /* All IPV4 dsts are created with ->obsolete set to the value * DST_OBSOLETE_FORCE_CHK which forces validation calls down @@ -1516,10 +1516,8 @@ void rt_del_uncached_list(struct rtable *rt) static void ipv4_dst_destroy(struct dst_entry *dst) { - struct rtable *rt = (struct rtable *)dst; - ip_dst_metrics_put(dst); - rt_del_uncached_list(rt); + rt_del_uncached_list(dst_rtable(dst)); } void rt_flush_dev(struct net_device *dev) @@ -2820,7 +2818,7 @@ static struct dst_ops ipv4_dst_blackhole_ops = { struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_orig) { - struct rtable *ort = (struct rtable *) dst_orig; + struct rtable *ort = dst_rtable(dst_orig); struct rtable *rt; rt = dst_alloc(&ipv4_dst_blackhole_ops, NULL, DST_OBSOLETE_DEAD, 0); @@ -2865,9 +2863,9 @@ struct rtable *ip_route_output_flow(struct net *net, struct flowi4 *flp4, if (flp4->flowi4_proto) { flp4->flowi4_oif = rt->dst.dev->ifindex; - rt = (struct rtable *)xfrm_lookup_route(net, &rt->dst, - flowi4_to_flowi(flp4), - sk, 0); + rt = dst_rtable(xfrm_lookup_route(net, &rt->dst, + flowi4_to_flowi(flp4), + sk, 0)); } return rt; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 6e2446295089..fe55ff5d379b 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1217,7 +1217,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) } if (connected) - rt = (struct rtable *)sk_dst_check(sk, 0); + rt = dst_rtable(sk_dst_check(sk, 0)); if (!rt) { struct net *net = sock_net(sk); diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 1dda59e0aeab..fccbbd3e1a4b 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -69,7 +69,7 @@ static int xfrm4_get_saddr(struct net *net, int oif, static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev, const struct flowi *fl) { - struct rtable *rt = (struct rtable *)xdst->route; + struct rtable *rt = dst_rtable(xdst->route); const struct flowi4 *fl4 = &fl->u.ip4; xdst->u.rt.rt_iif = fl4->flowi4_iif; diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 970af3983d11..19c8cc5289d5 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -459,7 +459,7 @@ static int l2tp_ip_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) fl4 = &inet->cork.fl.u.ip4; if (connected) - rt = (struct rtable *)__sk_dst_check(sk, 0); + rt = dst_rtable(__sk_dst_check(sk, 0)); rcu_read_lock(); if (!rt) { diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index 606349c8df0e..4385fd3b13be 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -81,7 +81,7 @@ static int mpls_xmit(struct sk_buff *skb) ttl = net->mpls.default_ttl; else ttl = ip_hdr(skb)->ttl; - rt = (struct rtable *)dst; + rt = dst_rtable(dst); } else if (dst->ops->family == AF_INET6) { if (tun_encap_info->ttl_propagate == MPLS_TTL_PROP_DISABLED) ttl = tun_encap_info->default_ttl; diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 6e8b9d100ad2..3313bceb6cc9 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -318,7 +318,7 @@ __ip_vs_get_out_rt(struct netns_ipvs *ipvs, int skb_af, struct sk_buff *skb, if (dest) { dest_dst = __ip_vs_dst_check(dest); if (likely(dest_dst)) - rt = (struct rtable *) dest_dst->dst_cache; + rt = dst_rtable(dest_dst->dst_cache); else { dest_dst = ip_vs_dest_dst_alloc(); spin_lock_bh(&dest->dst_lock); diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c index 100887beed31..c2c005234dcd 100644 --- a/net/netfilter/nf_flow_table_ip.c +++ b/net/netfilter/nf_flow_table_ip.c @@ -434,7 +434,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, return NF_ACCEPT; if (unlikely(tuplehash->tuple.xmit_type == FLOW_OFFLOAD_XMIT_XFRM)) { - rt = (struct rtable *)tuplehash->tuple.dst_cache; + rt = dst_rtable(tuplehash->tuple.dst_cache); memset(skb->cb, 0, sizeof(struct inet_skb_parm)); IPCB(skb)->iif = skb->dev->ifindex; IPCB(skb)->flags = IPSKB_FORWARDED; @@ -446,7 +446,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb, switch (tuplehash->tuple.xmit_type) { case FLOW_OFFLOAD_XMIT_NEIGH: - rt = (struct rtable *)tuplehash->tuple.dst_cache; + rt = dst_rtable(tuplehash->tuple.dst_cache); outdev = rt->dst.dev; skb->dev = outdev; nexthop = rt_nexthop(rt, flow->tuplehash[!dir].tuple.src_v4.s_addr); diff --git a/net/netfilter/nft_rt.c b/net/netfilter/nft_rt.c index 2434c624aafd..14d88394bcb7 100644 --- a/net/netfilter/nft_rt.c +++ b/net/netfilter/nft_rt.c @@ -73,7 +73,7 @@ void nft_rt_get_eval(const struct nft_expr *expr, if (nft_pf(pkt) != NFPROTO_IPV4) goto err; - *dest = (__force u32)rt_nexthop((const struct rtable *)dst, + *dest = (__force u32)rt_nexthop(dst_rtable(dst), ip_hdr(skb)->daddr); break; case NFT_RT_NEXTHOP6: diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index e849f368ed91..5a7436a13b74 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -552,7 +552,7 @@ static void sctp_v4_get_saddr(struct sctp_sock *sk, struct flowi *fl) { union sctp_addr *saddr = &t->saddr; - struct rtable *rt = (struct rtable *)t->dst; + struct rtable *rt = dst_rtable(t->dst); if (rt) { saddr->v4.sin_family = AF_INET; @@ -1085,7 +1085,7 @@ static inline int sctp_v4_xmit(struct sk_buff *skb, struct sctp_transport *t) skb_reset_inner_mac_header(skb); skb_reset_inner_transport_header(skb); skb_set_inner_ipproto(skb, IPPROTO_SCTP); - udp_tunnel_xmit_skb((struct rtable *)dst, sk, skb, fl4->saddr, + udp_tunnel_xmit_skb(dst_rtable(dst), sk, skb, fl4->saddr, fl4->daddr, dscp, ip4_dst_hoplimit(dst), df, sctp_sk(sk)->udp_port, t->encap_port, false, false); return 0; diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c index f892b0903dba..b849a3d133a0 100644 --- a/net/tipc/udp_media.c +++ b/net/tipc/udp_media.c @@ -174,7 +174,7 @@ static int tipc_udp_xmit(struct net *net, struct sk_buff *skb, local_bh_disable(); ndst = dst_cache_get(cache); if (dst->proto == htons(ETH_P_IP)) { - struct rtable *rt = (struct rtable *)ndst; + struct rtable *rt = dst_rtable(ndst); if (!rt) { struct flowi4 fl = { -- cgit v1.2.3 From a86a0661b86f310c0b73a30c829648864f0b2619 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 13:40:21 +0000 Subject: net: move sysctl_max_skb_frags to net_hotdata sysctl_max_skb_frags is used in TCP and MPTCP fast paths, move it to net_hodata for better cache locality. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240429134025.1233626-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/linux/skbuff.h | 2 -- include/net/hotdata.h | 1 + net/core/hotdata.c | 1 + net/core/skbuff.c | 5 +---- net/core/sysctl_net_core.c | 2 +- net/ipv4/tcp.c | 3 ++- net/mptcp/protocol.c | 3 ++- 7 files changed, 8 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index adf75d69770c..36b133f04d30 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -353,8 +353,6 @@ struct sk_buff; #define MAX_SKB_FRAGS CONFIG_MAX_SKB_FRAGS -extern int sysctl_max_skb_frags; - /* Set skb_shinfo(skb)->gso_size to this in case you want skb_segment to * segment using its current segmentation instead. */ diff --git a/include/net/hotdata.h b/include/net/hotdata.h index 003667a1efd6..a6cff6590426 100644 --- a/include/net/hotdata.h +++ b/include/net/hotdata.h @@ -38,6 +38,7 @@ struct net_hotdata { int max_backlog; int dev_tx_weight; int dev_rx_weight; + int sysctl_max_skb_frags; }; #define inet_ehash_secret net_hotdata.tcp_protocol.secret diff --git a/net/core/hotdata.c b/net/core/hotdata.c index c8a7a451c18a..f17cbb4807b9 100644 --- a/net/core/hotdata.c +++ b/net/core/hotdata.c @@ -18,5 +18,6 @@ struct net_hotdata net_hotdata __cacheline_aligned = { .max_backlog = 1000, .dev_tx_weight = 64, .dev_rx_weight = 64, + .sysctl_max_skb_frags = MAX_SKB_FRAGS, }; EXPORT_SYMBOL(net_hotdata); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 0c8b82750000..65779b8f0b12 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -109,9 +109,6 @@ static struct kmem_cache *skbuff_ext_cache __ro_after_init; #define SKB_SMALL_HEAD_HEADROOM \ SKB_WITH_OVERHEAD(SKB_SMALL_HEAD_CACHE_SIZE) -int sysctl_max_skb_frags __read_mostly = MAX_SKB_FRAGS; -EXPORT_SYMBOL(sysctl_max_skb_frags); - /* kcm_write_msgs() relies on casting paged frags to bio_vec to use * iov_iter_bvec(). These static asserts ensure the cast is valid is long as the * netmem is a page. @@ -7040,7 +7037,7 @@ static void skb_splice_csum_page(struct sk_buff *skb, struct page *page, ssize_t skb_splice_from_iter(struct sk_buff *skb, struct iov_iter *iter, ssize_t maxsize, gfp_t gfp) { - size_t frag_limit = READ_ONCE(sysctl_max_skb_frags); + size_t frag_limit = READ_ONCE(net_hotdata.sysctl_max_skb_frags); struct page *pages[8], **ppages = pages; ssize_t spliced = 0, ret = 0; unsigned int i; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 903ab4a51c17..e75375d54b9e 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -595,7 +595,7 @@ static struct ctl_table net_core_table[] = { }, { .procname = "max_skb_frags", - .data = &sysctl_max_skb_frags, + .data = &net_hotdata.sysctl_max_skb_frags, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4ec0f4feee00..388f6e115bf1 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -280,6 +280,7 @@ #include #include #include +#include #include /* Track pending CMSGs. */ @@ -1188,7 +1189,7 @@ new_segment: if (!skb_can_coalesce(skb, i, pfrag->page, pfrag->offset)) { - if (i >= READ_ONCE(sysctl_max_skb_frags)) { + if (i >= READ_ONCE(net_hotdata.sysctl_max_skb_frags)) { tcp_mark_push(tp, skb); goto new_segment; } diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 4b13ca362efa..aff17597e6a7 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -20,6 +20,7 @@ #include #endif #include +#include #include #include #include "protocol.h" @@ -1272,7 +1273,7 @@ static int mptcp_sendmsg_frag(struct sock *sk, struct sock *ssk, i = skb_shinfo(skb)->nr_frags; can_coalesce = skb_can_coalesce(skb, i, dfrag->page, offset); - if (!can_coalesce && i >= READ_ONCE(sysctl_max_skb_frags)) { + if (!can_coalesce && i >= READ_ONCE(net_hotdata.sysctl_max_skb_frags)) { tcp_mark_push(tcp_sk(ssk), skb); goto alloc_skb; } -- cgit v1.2.3 From d480dc76d9f8a72671a6f9a7d987d2389c2e2ee4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 13:40:22 +0000 Subject: net: move sysctl_skb_defer_max to net_hotdata sysctl_skb_defer_max is used in TCP fast path, move it to net_hodata. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240429134025.1233626-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/hotdata.h | 1 + net/core/dev.c | 1 - net/core/dev.h | 1 - net/core/hotdata.c | 1 + net/core/skbuff.c | 2 +- net/core/sysctl_net_core.c | 2 +- 6 files changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/hotdata.h b/include/net/hotdata.h index a6cff6590426..290499f72e18 100644 --- a/include/net/hotdata.h +++ b/include/net/hotdata.h @@ -39,6 +39,7 @@ struct net_hotdata { int dev_tx_weight; int dev_rx_weight; int sysctl_max_skb_frags; + int sysctl_skb_defer_max; }; #define inet_ehash_secret net_hotdata.tcp_protocol.secret diff --git a/net/core/dev.c b/net/core/dev.c index c9e59eff8ec8..cd7ba50eac15 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4450,7 +4450,6 @@ EXPORT_SYMBOL(__dev_direct_xmit); *************************************************************************/ static DEFINE_PER_CPU(struct task_struct *, backlog_napi); -unsigned int sysctl_skb_defer_max __read_mostly = 64; int weight_p __read_mostly = 64; /* old backlog weight */ int dev_weight_rx_bias __read_mostly = 1; /* bias for backlog weight */ int dev_weight_tx_bias __read_mostly = 1; /* bias for output_queue quota */ diff --git a/net/core/dev.h b/net/core/dev.h index 8572d2c8dc4a..b7b518bc2be5 100644 --- a/net/core/dev.h +++ b/net/core/dev.h @@ -36,7 +36,6 @@ int dev_addr_init(struct net_device *dev); void dev_addr_check(struct net_device *dev); /* sysctls not referred to from outside net/core/ */ -extern unsigned int sysctl_skb_defer_max; extern int netdev_unregister_timeout_secs; extern int weight_p; extern int dev_weight_rx_bias; diff --git a/net/core/hotdata.c b/net/core/hotdata.c index f17cbb4807b9..a359ff160d54 100644 --- a/net/core/hotdata.c +++ b/net/core/hotdata.c @@ -19,5 +19,6 @@ struct net_hotdata net_hotdata __cacheline_aligned = { .dev_tx_weight = 64, .dev_rx_weight = 64, .sysctl_max_skb_frags = MAX_SKB_FRAGS, + .sysctl_skb_defer_max = 64, }; EXPORT_SYMBOL(net_hotdata); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 65779b8f0b12..5f382e94b4d1 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6985,7 +6985,7 @@ nodefer: kfree_skb_napi_cache(skb); DEBUG_NET_WARN_ON_ONCE(skb->destructor); sd = &per_cpu(softnet_data, cpu); - defer_max = READ_ONCE(sysctl_skb_defer_max); + defer_max = READ_ONCE(net_hotdata.sysctl_skb_defer_max); if (READ_ONCE(sd->defer_count) >= defer_max) goto nodefer; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index e75375d54b9e..118c78615543 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -654,7 +654,7 @@ static struct ctl_table net_core_table[] = { }, { .procname = "skb_defer_max", - .data = &sysctl_skb_defer_max, + .data = &net_hotdata.sysctl_skb_defer_max, .maxlen = sizeof(unsigned int), .mode = 0644, .proc_handler = proc_dointvec_minmax, -- cgit v1.2.3 From dda4d96acb20c02920f6d9a20fdc3f4846192aeb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 13:40:23 +0000 Subject: tcp: move tcp_out_of_memory() to net/ipv4/tcp.c tcp_out_of_memory() has a single caller: tcp_check_oom(). Following patch will also make sk_memory_allocated() not anymore visible from and Add const qualifier to sock argument of tcp_out_of_memory() and tcp_check_oom(). Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240429134025.1233626-4-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 10 +--------- net/ipv4/tcp.c | 10 +++++++++- 2 files changed, 10 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index fe98fb01879b..0a51e6a45bce 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -296,14 +296,6 @@ static inline bool between(__u32 seq1, __u32 seq2, __u32 seq3) return seq3 - seq2 >= seq1 - seq2; } -static inline bool tcp_out_of_memory(struct sock *sk) -{ - if (sk->sk_wmem_queued > SOCK_MIN_SNDBUF && - sk_memory_allocated(sk) > sk_prot_mem_limits(sk, 2)) - return true; - return false; -} - static inline void tcp_wmem_free_skb(struct sock *sk, struct sk_buff *skb) { sk_wmem_queued_add(sk, -skb->truesize); @@ -316,7 +308,7 @@ static inline void tcp_wmem_free_skb(struct sock *sk, struct sk_buff *skb) void sk_forced_mem_schedule(struct sock *sk, int size); -bool tcp_check_oom(struct sock *sk, int shift); +bool tcp_check_oom(const struct sock *sk, int shift); extern struct proto tcp_prot; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 388f6e115bf1..0a3aa3047083 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2752,7 +2752,15 @@ static bool tcp_too_many_orphans(int shift) READ_ONCE(sysctl_tcp_max_orphans); } -bool tcp_check_oom(struct sock *sk, int shift) +static bool tcp_out_of_memory(const struct sock *sk) +{ + if (sk->sk_wmem_queued > SOCK_MIN_SNDBUF && + sk_memory_allocated(sk) > sk_prot_mem_limits(sk, 2)) + return true; + return false; +} + +bool tcp_check_oom(const struct sock *sk, int shift) { bool too_many_orphans, out_of_socket_memory; -- cgit v1.2.3 From f3d93817fba30a8d3508fa990405039c0820dca3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 13:40:24 +0000 Subject: net: add Move some proto memory definitions out of Very few files need them, and following patch will include from Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240429134025.1233626-5-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/proto_memory.h | 83 ++++++++++++++++++++++++++++++++++++++++++++++ include/net/sock.h | 78 ------------------------------------------- net/core/sock.c | 1 + net/core/sysctl_net_core.c | 1 + net/ipv4/proc.c | 1 + net/ipv4/tcp.c | 1 + net/ipv4/tcp_input.c | 1 + net/ipv4/tcp_output.c | 1 + net/sctp/sm_statefuns.c | 1 + 9 files changed, 90 insertions(+), 78 deletions(-) create mode 100644 include/net/proto_memory.h (limited to 'net') diff --git a/include/net/proto_memory.h b/include/net/proto_memory.h new file mode 100644 index 000000000000..41404d4bb6f0 --- /dev/null +++ b/include/net/proto_memory.h @@ -0,0 +1,83 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef _PROTO_MEMORY_H +#define _PROTO_MEMORY_H + +#include + +/* 1 MB per cpu, in page units */ +#define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT)) +extern int sysctl_mem_pcpu_rsv; + +static inline bool sk_has_memory_pressure(const struct sock *sk) +{ + return sk->sk_prot->memory_pressure != NULL; +} + +static inline bool +proto_memory_pressure(const struct proto *prot) +{ + if (!prot->memory_pressure) + return false; + return !!READ_ONCE(*prot->memory_pressure); +} + +static inline bool sk_under_global_memory_pressure(const struct sock *sk) +{ + return proto_memory_pressure(sk->sk_prot); +} + +static inline bool sk_under_memory_pressure(const struct sock *sk) +{ + if (!sk->sk_prot->memory_pressure) + return false; + + if (mem_cgroup_sockets_enabled && sk->sk_memcg && + mem_cgroup_under_socket_pressure(sk->sk_memcg)) + return true; + + return !!READ_ONCE(*sk->sk_prot->memory_pressure); +} + +static inline long +proto_memory_allocated(const struct proto *prot) +{ + return max(0L, atomic_long_read(prot->memory_allocated)); +} + +static inline long +sk_memory_allocated(const struct sock *sk) +{ + return proto_memory_allocated(sk->sk_prot); +} + +static inline void proto_memory_pcpu_drain(struct proto *proto) +{ + int val = this_cpu_xchg(*proto->per_cpu_fw_alloc, 0); + + if (val) + atomic_long_add(val, proto->memory_allocated); +} + +static inline void +sk_memory_allocated_add(const struct sock *sk, int val) +{ + struct proto *proto = sk->sk_prot; + + val = this_cpu_add_return(*proto->per_cpu_fw_alloc, val); + + if (unlikely(val >= READ_ONCE(sysctl_mem_pcpu_rsv))) + proto_memory_pcpu_drain(proto); +} + +static inline void +sk_memory_allocated_sub(const struct sock *sk, int val) +{ + struct proto *proto = sk->sk_prot; + + val = this_cpu_sub_return(*proto->per_cpu_fw_alloc, val); + + if (unlikely(val <= -READ_ONCE(sysctl_mem_pcpu_rsv))) + proto_memory_pcpu_drain(proto); +} + +#endif /* _PROTO_MEMORY_H */ diff --git a/include/net/sock.h b/include/net/sock.h index 48bcc845202f..0450494a1766 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1371,75 +1371,6 @@ static inline int sk_under_cgroup_hierarchy(struct sock *sk, #endif } -static inline bool sk_has_memory_pressure(const struct sock *sk) -{ - return sk->sk_prot->memory_pressure != NULL; -} - -static inline bool sk_under_global_memory_pressure(const struct sock *sk) -{ - return sk->sk_prot->memory_pressure && - !!READ_ONCE(*sk->sk_prot->memory_pressure); -} - -static inline bool sk_under_memory_pressure(const struct sock *sk) -{ - if (!sk->sk_prot->memory_pressure) - return false; - - if (mem_cgroup_sockets_enabled && sk->sk_memcg && - mem_cgroup_under_socket_pressure(sk->sk_memcg)) - return true; - - return !!READ_ONCE(*sk->sk_prot->memory_pressure); -} - -static inline long -proto_memory_allocated(const struct proto *prot) -{ - return max(0L, atomic_long_read(prot->memory_allocated)); -} - -static inline long -sk_memory_allocated(const struct sock *sk) -{ - return proto_memory_allocated(sk->sk_prot); -} - -/* 1 MB per cpu, in page units */ -#define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT)) -extern int sysctl_mem_pcpu_rsv; - -static inline void proto_memory_pcpu_drain(struct proto *proto) -{ - int val = this_cpu_xchg(*proto->per_cpu_fw_alloc, 0); - - if (val) - atomic_long_add(val, proto->memory_allocated); -} - -static inline void -sk_memory_allocated_add(const struct sock *sk, int val) -{ - struct proto *proto = sk->sk_prot; - - val = this_cpu_add_return(*proto->per_cpu_fw_alloc, val); - - if (unlikely(val >= READ_ONCE(sysctl_mem_pcpu_rsv))) - proto_memory_pcpu_drain(proto); -} - -static inline void -sk_memory_allocated_sub(const struct sock *sk, int val) -{ - struct proto *proto = sk->sk_prot; - - val = this_cpu_sub_return(*proto->per_cpu_fw_alloc, val); - - if (unlikely(val <= -READ_ONCE(sysctl_mem_pcpu_rsv))) - proto_memory_pcpu_drain(proto); -} - #define SK_ALLOC_PERCPU_COUNTER_BATCH 16 static inline void sk_sockets_allocated_dec(struct sock *sk) @@ -1466,15 +1397,6 @@ proto_sockets_allocated_sum_positive(struct proto *prot) return percpu_counter_sum_positive(prot->sockets_allocated); } -static inline bool -proto_memory_pressure(struct proto *prot) -{ - if (!prot->memory_pressure) - return false; - return !!READ_ONCE(*prot->memory_pressure); -} - - #ifdef CONFIG_PROC_FS #define PROTO_INUSE_NR 64 /* should be enough for the first time */ struct prot_inuse { diff --git a/net/core/sock.c b/net/core/sock.c index fe9195186c13..e0692b752369 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -127,6 +127,7 @@ #include #include #include +#include #include #include #include diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 118c78615543..a452a330d0ed 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "dev.h" diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 914bc9c35cc7..6c4664c681ca 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0a3aa3047083..e1f0efbb29d6 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -272,6 +272,7 @@ #include #include #include +#include #include #include #include diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 53e1150f706f..ad8fa129fcfe 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -72,6 +72,7 @@ #include #include #include +#include #include #include #include diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index ea7ad7d99245..57edf66ff91b 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -39,6 +39,7 @@ #include #include +#include #include #include diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 08fdf1251f46..5adf0c0a6c1a 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From c204fef97ee62ca3310f43d12bbda4eb10266c7c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 13:40:25 +0000 Subject: net: move sysctl_mem_pcpu_rsv to net_hotdata sysctl_mem_pcpu_rsv is used in TCP fast path, move it to net_hodata for better cache locality. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240429134025.1233626-6-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/hotdata.h | 1 + include/net/proto_memory.h | 6 +++--- net/core/hotdata.c | 5 +++-- net/core/sock.c | 1 - net/core/sysctl_net_core.c | 2 +- 5 files changed, 8 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/hotdata.h b/include/net/hotdata.h index 290499f72e18..30e9570beb2a 100644 --- a/include/net/hotdata.h +++ b/include/net/hotdata.h @@ -40,6 +40,7 @@ struct net_hotdata { int dev_rx_weight; int sysctl_max_skb_frags; int sysctl_skb_defer_max; + int sysctl_mem_pcpu_rsv; }; #define inet_ehash_secret net_hotdata.tcp_protocol.secret diff --git a/include/net/proto_memory.h b/include/net/proto_memory.h index 41404d4bb6f0..a6ab2f4f5e28 100644 --- a/include/net/proto_memory.h +++ b/include/net/proto_memory.h @@ -3,10 +3,10 @@ #define _PROTO_MEMORY_H #include +#include /* 1 MB per cpu, in page units */ #define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT)) -extern int sysctl_mem_pcpu_rsv; static inline bool sk_has_memory_pressure(const struct sock *sk) { @@ -65,7 +65,7 @@ sk_memory_allocated_add(const struct sock *sk, int val) val = this_cpu_add_return(*proto->per_cpu_fw_alloc, val); - if (unlikely(val >= READ_ONCE(sysctl_mem_pcpu_rsv))) + if (unlikely(val >= READ_ONCE(net_hotdata.sysctl_mem_pcpu_rsv))) proto_memory_pcpu_drain(proto); } @@ -76,7 +76,7 @@ sk_memory_allocated_sub(const struct sock *sk, int val) val = this_cpu_sub_return(*proto->per_cpu_fw_alloc, val); - if (unlikely(val <= -READ_ONCE(sysctl_mem_pcpu_rsv))) + if (unlikely(val <= -READ_ONCE(net_hotdata.sysctl_mem_pcpu_rsv))) proto_memory_pcpu_drain(proto); } diff --git a/net/core/hotdata.c b/net/core/hotdata.c index a359ff160d54..d0aaaaa556f2 100644 --- a/net/core/hotdata.c +++ b/net/core/hotdata.c @@ -1,9 +1,9 @@ // SPDX-License-Identifier: GPL-2.0-or-later -#include #include #include #include - +#include +#include struct net_hotdata net_hotdata __cacheline_aligned = { .offload_base = LIST_HEAD_INIT(net_hotdata.offload_base), @@ -20,5 +20,6 @@ struct net_hotdata net_hotdata __cacheline_aligned = { .dev_rx_weight = 64, .sysctl_max_skb_frags = MAX_SKB_FRAGS, .sysctl_skb_defer_max = 64, + .sysctl_mem_pcpu_rsv = SK_MEMORY_PCPU_RESERVE }; EXPORT_SYMBOL(net_hotdata); diff --git a/net/core/sock.c b/net/core/sock.c index e0692b752369..8d6e638b5426 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -284,7 +284,6 @@ __u32 sysctl_rmem_max __read_mostly = SK_RMEM_MAX; EXPORT_SYMBOL(sysctl_rmem_max); __u32 sysctl_wmem_default __read_mostly = SK_WMEM_MAX; __u32 sysctl_rmem_default __read_mostly = SK_RMEM_MAX; -int sysctl_mem_pcpu_rsv __read_mostly = SK_MEMORY_PCPU_RESERVE; int sysctl_tstamp_allow_data __read_mostly = 1; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index a452a330d0ed..6da5995ac86a 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -416,7 +416,7 @@ static struct ctl_table net_core_table[] = { }, { .procname = "mem_pcpu_rsv", - .data = &sysctl_mem_pcpu_rsv, + .data = &net_hotdata.sysctl_mem_pcpu_rsv, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, -- cgit v1.2.3 From fb7a0d334894206ae35f023a82cad5a290fd7386 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Mon, 29 Apr 2024 20:00:31 +0200 Subject: mptcp: ensure snd_nxt is properly initialized on connect Christoph reported a splat hinting at a corrupted snd_una: WARNING: CPU: 1 PID: 38 at net/mptcp/protocol.c:1005 __mptcp_clean_una+0x4b3/0x620 net/mptcp/protocol.c:1005 Modules linked in: CPU: 1 PID: 38 Comm: kworker/1:1 Not tainted 6.9.0-rc1-gbbeac67456c9 #59 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 Workqueue: events mptcp_worker RIP: 0010:__mptcp_clean_una+0x4b3/0x620 net/mptcp/protocol.c:1005 Code: be 06 01 00 00 bf 06 01 00 00 e8 a8 12 e7 fe e9 00 fe ff ff e8 8e 1a e7 fe 0f b7 ab 3e 02 00 00 e9 d3 fd ff ff e8 7d 1a e7 fe <0f> 0b 4c 8b bb e0 05 00 00 e9 74 fc ff ff e8 6a 1a e7 fe 0f 0b e9 RSP: 0018:ffffc9000013fd48 EFLAGS: 00010293 RAX: 0000000000000000 RBX: ffff8881029bd280 RCX: ffffffff82382fe4 RDX: ffff8881003cbd00 RSI: ffffffff823833c3 RDI: 0000000000000001 RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000000 R11: fefefefefefefeff R12: ffff888138ba8000 R13: 0000000000000106 R14: ffff8881029bd908 R15: ffff888126560000 FS: 0000000000000000(0000) GS:ffff88813bd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f604a5dae38 CR3: 0000000101dac002 CR4: 0000000000170ef0 Call Trace: __mptcp_clean_una_wakeup net/mptcp/protocol.c:1055 [inline] mptcp_clean_una_wakeup net/mptcp/protocol.c:1062 [inline] __mptcp_retrans+0x7f/0x7e0 net/mptcp/protocol.c:2615 mptcp_worker+0x434/0x740 net/mptcp/protocol.c:2767 process_one_work+0x1e0/0x560 kernel/workqueue.c:3254 process_scheduled_works kernel/workqueue.c:3335 [inline] worker_thread+0x3c7/0x640 kernel/workqueue.c:3416 kthread+0x121/0x170 kernel/kthread.c:388 ret_from_fork+0x44/0x50 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:243 When fallback to TCP happens early on a client socket, snd_nxt is not yet initialized and any incoming ack will copy such value into snd_una. If the mptcp worker (dumbly) tries mptcp-level re-injection after such ack, that would unconditionally trigger a send buffer cleanup using 'bad' snd_una values. We could easily disable re-injection for fallback sockets, but such dumb behavior already helped catching a few subtle issues and a very low to zero impact in practice. Instead address the issue always initializing snd_nxt (and write_seq, for consistency) at connect time. Fixes: 8fd738049ac3 ("mptcp: fallback in case of simultaneous connect") Cc: stable@vger.kernel.org Reported-by: Christoph Paasch Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/485 Tested-by: Christoph Paasch Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240429-upstream-net-20240429-mptcp-snd_nxt-init-connect-v1-1-59ceac0a7dcb@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 7e74b812e366..965eb69dc5de 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3723,6 +3723,9 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_TOKENFALLBACKINIT); mptcp_subflow_early_fallback(msk, subflow); } + + WRITE_ONCE(msk->write_seq, subflow->idsn); + WRITE_ONCE(msk->snd_nxt, subflow->idsn); if (likely(!__mptcp_check_fallback(msk))) MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPCAPABLEACTIVE); -- cgit v1.2.3 From c2e6a872bde9912f1a7579639c5ca3adf1003916 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Mon, 29 Apr 2024 03:04:33 -0700 Subject: netpoll: Fix race condition in netpoll_owner_active KCSAN detected a race condition in netpoll: BUG: KCSAN: data-race in net_rx_action / netpoll_send_skb write (marked) to 0xffff8881164168b0 of 4 bytes by interrupt on cpu 10: net_rx_action (./include/linux/netpoll.h:90 net/core/dev.c:6712 net/core/dev.c:6822) read to 0xffff8881164168b0 of 4 bytes by task 1 on cpu 2: netpoll_send_skb (net/core/netpoll.c:319 net/core/netpoll.c:345 net/core/netpoll.c:393) netpoll_send_udp (net/core/netpoll.c:?) value changed: 0x0000000a -> 0xffffffff This happens because netpoll_owner_active() needs to check if the current CPU is the owner of the lock, touching napi->poll_owner non atomically. The ->poll_owner field contains the current CPU holding the lock. Use an atomic read to check if the poll owner is the current CPU. Signed-off-by: Breno Leitao Link: https://lore.kernel.org/r/20240429100437.3487432-1-leitao@debian.org Signed-off-by: Jakub Kicinski --- net/core/netpoll.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 543007f159f9..55bcacf67df3 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -316,7 +316,7 @@ static int netpoll_owner_active(struct net_device *dev) struct napi_struct *napi; list_for_each_entry_rcu(napi, &dev->napi_list, dev_list) { - if (napi->poll_owner == smp_processor_id()) + if (READ_ONCE(napi->poll_owner) == smp_processor_id()) return 1; } return 0; -- cgit v1.2.3 From a4a87fa4e96c7746e009de06a567688fd9af6013 Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Tue, 30 Apr 2024 09:08:52 +0200 Subject: xfrm: Add Direction to the SA in or out This patch introduces the 'dir' attribute, 'in' or 'out', to the xfrm_state, SA, enhancing usability by delineating the scope of values based on direction. An input SA will restrict values pertinent to input, effectively segregating them from output-related values. And an output SA will restrict attributes for output. This change aims to streamline the configuration process and improve the overall consistency of SA attributes during configuration. This feature sets the groundwork for future patches, including the upcoming IP-TFS patch. Signed-off-by: Antony Antony Reviewed-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 1 + include/uapi/linux/xfrm.h | 6 ++ net/xfrm/xfrm_compat.c | 7 ++- net/xfrm/xfrm_device.c | 6 ++ net/xfrm/xfrm_replay.c | 3 +- net/xfrm/xfrm_state.c | 8 +++ net/xfrm/xfrm_user.c | 138 ++++++++++++++++++++++++++++++++++++++++++++-- 7 files changed, 160 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 57c743b7e4fe..7c9be06f8302 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -291,6 +291,7 @@ struct xfrm_state { /* Private data of this transformer, format is opaque, * interpreted by xfrm_type methods. */ void *data; + u8 dir; }; static inline struct net *xs_net(struct xfrm_state *x) diff --git a/include/uapi/linux/xfrm.h b/include/uapi/linux/xfrm.h index 6a77328be114..18ceaba8486e 100644 --- a/include/uapi/linux/xfrm.h +++ b/include/uapi/linux/xfrm.h @@ -141,6 +141,11 @@ enum { XFRM_POLICY_MAX = 3 }; +enum xfrm_sa_dir { + XFRM_SA_DIR_IN = 1, + XFRM_SA_DIR_OUT = 2 +}; + enum { XFRM_SHARE_ANY, /* No limitations */ XFRM_SHARE_SESSION, /* For this session only */ @@ -315,6 +320,7 @@ enum xfrm_attr_type_t { XFRMA_SET_MARK_MASK, /* __u32 */ XFRMA_IF_ID, /* __u32 */ XFRMA_MTIMER_THRESH, /* __u32 in seconds for input SA */ + XFRMA_SA_DIR, /* __u8 */ __XFRMA_MAX #define XFRMA_OUTPUT_MARK XFRMA_SET_MARK /* Compatibility */ diff --git a/net/xfrm/xfrm_compat.c b/net/xfrm/xfrm_compat.c index 655fe4ff8621..703d4172c7d7 100644 --- a/net/xfrm/xfrm_compat.c +++ b/net/xfrm/xfrm_compat.c @@ -98,6 +98,7 @@ static const int compat_msg_min[XFRM_NR_MSGTYPES] = { }; static const struct nla_policy compat_policy[XFRMA_MAX+1] = { + [XFRMA_UNSPEC] = { .strict_start_type = XFRMA_SA_DIR }, [XFRMA_SA] = { .len = XMSGSIZE(compat_xfrm_usersa_info)}, [XFRMA_POLICY] = { .len = XMSGSIZE(compat_xfrm_userpolicy_info)}, [XFRMA_LASTUSED] = { .type = NLA_U64}, @@ -129,6 +130,7 @@ static const struct nla_policy compat_policy[XFRMA_MAX+1] = { [XFRMA_SET_MARK_MASK] = { .type = NLA_U32 }, [XFRMA_IF_ID] = { .type = NLA_U32 }, [XFRMA_MTIMER_THRESH] = { .type = NLA_U32 }, + [XFRMA_SA_DIR] = NLA_POLICY_RANGE(NLA_U8, XFRM_SA_DIR_IN, XFRM_SA_DIR_OUT), }; static struct nlmsghdr *xfrm_nlmsg_put_compat(struct sk_buff *skb, @@ -277,9 +279,10 @@ static int xfrm_xlate64_attr(struct sk_buff *dst, const struct nlattr *src) case XFRMA_SET_MARK_MASK: case XFRMA_IF_ID: case XFRMA_MTIMER_THRESH: + case XFRMA_SA_DIR: return xfrm_nla_cpy(dst, src, nla_len(src)); default: - BUILD_BUG_ON(XFRMA_MAX != XFRMA_MTIMER_THRESH); + BUILD_BUG_ON(XFRMA_MAX != XFRMA_SA_DIR); pr_warn_once("unsupported nla_type %d\n", src->nla_type); return -EOPNOTSUPP; } @@ -434,7 +437,7 @@ static int xfrm_xlate32_attr(void *dst, const struct nlattr *nla, int err; if (type > XFRMA_MAX) { - BUILD_BUG_ON(XFRMA_MAX != XFRMA_MTIMER_THRESH); + BUILD_BUG_ON(XFRMA_MAX != XFRMA_SA_DIR); NL_SET_ERR_MSG(extack, "Bad attribute"); return -EOPNOTSUPP; } diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c index 6346690d5c69..2455a76a1cff 100644 --- a/net/xfrm/xfrm_device.c +++ b/net/xfrm/xfrm_device.c @@ -253,6 +253,12 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x, return -EINVAL; } + if ((xuo->flags & XFRM_OFFLOAD_INBOUND && x->dir == XFRM_SA_DIR_OUT) || + (!(xuo->flags & XFRM_OFFLOAD_INBOUND) && x->dir == XFRM_SA_DIR_IN)) { + NL_SET_ERR_MSG(extack, "Mismatched SA and offload direction"); + return -EINVAL; + } + is_packet_offload = xuo->flags & XFRM_OFFLOAD_PACKET; /* We don't yet support UDP encapsulation and TFC padding. */ diff --git a/net/xfrm/xfrm_replay.c b/net/xfrm/xfrm_replay.c index ce56d659c55a..bc56c6305725 100644 --- a/net/xfrm/xfrm_replay.c +++ b/net/xfrm/xfrm_replay.c @@ -778,7 +778,8 @@ int xfrm_init_replay(struct xfrm_state *x, struct netlink_ext_ack *extack) } if (x->props.flags & XFRM_STATE_ESN) { - if (replay_esn->replay_window == 0) { + if (replay_esn->replay_window == 0 && + (!x->dir || x->dir == XFRM_SA_DIR_IN)) { NL_SET_ERR_MSG(extack, "ESN replay window must be > 0"); return -EINVAL; } diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 0c306473a79d..649bb739df0d 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1292,6 +1292,7 @@ found: if (km_query(x, tmpl, pol) == 0) { spin_lock_bh(&net->xfrm.xfrm_state_lock); x->km.state = XFRM_STATE_ACQ; + x->dir = XFRM_SA_DIR_OUT; list_add(&x->km.all, &net->xfrm.state_all); XFRM_STATE_INSERT(bydst, &x->bydst, net->xfrm.state_bydst + h, @@ -1744,6 +1745,7 @@ static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig, x->lastused = orig->lastused; x->new_mapping = 0; x->new_mapping_sport = 0; + x->dir = orig->dir; return x; @@ -1864,8 +1866,14 @@ int xfrm_state_update(struct xfrm_state *x) } if (x1->km.state == XFRM_STATE_ACQ) { + if (x->dir && x1->dir != x->dir) + goto out; + __xfrm_state_insert(x); x = NULL; + } else { + if (x1->dir != x->dir) + goto out; } err = 0; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 810b520493f3..f5eb3af4fb81 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -130,7 +130,7 @@ static inline int verify_sec_ctx_len(struct nlattr **attrs, struct netlink_ext_a } static inline int verify_replay(struct xfrm_usersa_info *p, - struct nlattr **attrs, + struct nlattr **attrs, u8 sa_dir, struct netlink_ext_ack *extack) { struct nlattr *rt = attrs[XFRMA_REPLAY_ESN_VAL]; @@ -168,6 +168,30 @@ static inline int verify_replay(struct xfrm_usersa_info *p, return -EINVAL; } + if (sa_dir == XFRM_SA_DIR_OUT) { + if (rs->replay_window) { + NL_SET_ERR_MSG(extack, "Replay window should be 0 for output SA"); + return -EINVAL; + } + if (rs->seq || rs->seq_hi) { + NL_SET_ERR_MSG(extack, + "Replay seq and seq_hi should be 0 for output SA"); + return -EINVAL; + } + if (rs->bmp_len) { + NL_SET_ERR_MSG(extack, "Replay bmp_len should 0 for output SA"); + return -EINVAL; + } + } + + if (sa_dir == XFRM_SA_DIR_IN) { + if (rs->oseq || rs->oseq_hi) { + NL_SET_ERR_MSG(extack, + "Replay oseq and oseq_hi should be 0 for input SA"); + return -EINVAL; + } + } + return 0; } @@ -176,6 +200,7 @@ static int verify_newsa_info(struct xfrm_usersa_info *p, struct netlink_ext_ack *extack) { int err; + u8 sa_dir = attrs[XFRMA_SA_DIR] ? nla_get_u8(attrs[XFRMA_SA_DIR]) : 0; err = -EINVAL; switch (p->family) { @@ -334,7 +359,7 @@ static int verify_newsa_info(struct xfrm_usersa_info *p, goto out; if ((err = verify_sec_ctx_len(attrs, extack))) goto out; - if ((err = verify_replay(p, attrs, extack))) + if ((err = verify_replay(p, attrs, sa_dir, extack))) goto out; err = -EINVAL; @@ -358,6 +383,77 @@ static int verify_newsa_info(struct xfrm_usersa_info *p, err = -EINVAL; goto out; } + + if (sa_dir == XFRM_SA_DIR_OUT) { + NL_SET_ERR_MSG(extack, + "MTIMER_THRESH attribute should not be set on output SA"); + err = -EINVAL; + goto out; + } + } + + if (sa_dir == XFRM_SA_DIR_OUT) { + if (p->flags & XFRM_STATE_DECAP_DSCP) { + NL_SET_ERR_MSG(extack, "Flag DECAP_DSCP should not be set for output SA"); + err = -EINVAL; + goto out; + } + + if (p->flags & XFRM_STATE_ICMP) { + NL_SET_ERR_MSG(extack, "Flag ICMP should not be set for output SA"); + err = -EINVAL; + goto out; + } + + if (p->flags & XFRM_STATE_WILDRECV) { + NL_SET_ERR_MSG(extack, "Flag WILDRECV should not be set for output SA"); + err = -EINVAL; + goto out; + } + + if (p->replay_window) { + NL_SET_ERR_MSG(extack, "Replay window should be 0 for output SA"); + err = -EINVAL; + goto out; + } + + if (attrs[XFRMA_REPLAY_VAL]) { + struct xfrm_replay_state *replay; + + replay = nla_data(attrs[XFRMA_REPLAY_VAL]); + + if (replay->seq || replay->bitmap) { + NL_SET_ERR_MSG(extack, + "Replay seq and bitmap should be 0 for output SA"); + err = -EINVAL; + goto out; + } + } + } + + if (sa_dir == XFRM_SA_DIR_IN) { + if (p->flags & XFRM_STATE_NOPMTUDISC) { + NL_SET_ERR_MSG(extack, "Flag NOPMTUDISC should not be set for input SA"); + err = -EINVAL; + goto out; + } + + if (attrs[XFRMA_SA_EXTRA_FLAGS]) { + u32 xflags = nla_get_u32(attrs[XFRMA_SA_EXTRA_FLAGS]); + + if (xflags & XFRM_SA_XFLAG_DONT_ENCAP_DSCP) { + NL_SET_ERR_MSG(extack, "Flag DONT_ENCAP_DSCP should not be set for input SA"); + err = -EINVAL; + goto out; + } + + if (xflags & XFRM_SA_XFLAG_OSEQ_MAY_WRAP) { + NL_SET_ERR_MSG(extack, "Flag OSEQ_MAY_WRAP should not be set for input SA"); + err = -EINVAL; + goto out; + } + + } } out: @@ -734,6 +830,9 @@ static struct xfrm_state *xfrm_state_construct(struct net *net, if (attrs[XFRMA_IF_ID]) x->if_id = nla_get_u32(attrs[XFRMA_IF_ID]); + if (attrs[XFRMA_SA_DIR]) + x->dir = nla_get_u8(attrs[XFRMA_SA_DIR]); + err = __xfrm_init_state(x, false, attrs[XFRMA_OFFLOAD_DEV], extack); if (err) goto error; @@ -1182,8 +1281,13 @@ static int copy_to_user_state_extra(struct xfrm_state *x, if (ret) goto out; } - if (x->mapping_maxage) + if (x->mapping_maxage) { ret = nla_put_u32(skb, XFRMA_MTIMER_THRESH, x->mapping_maxage); + if (ret) + goto out; + } + if (x->dir) + ret = nla_put_u8(skb, XFRMA_SA_DIR, x->dir); out: return ret; } @@ -1618,6 +1722,9 @@ static int xfrm_alloc_userspi(struct sk_buff *skb, struct nlmsghdr *nlh, if (err) goto out; + if (attrs[XFRMA_SA_DIR]) + x->dir = nla_get_u8(attrs[XFRMA_SA_DIR]); + resp_skb = xfrm_state_netlink(skb, x, nlh->nlmsg_seq); if (IS_ERR(resp_skb)) { err = PTR_ERR(resp_skb); @@ -2402,7 +2509,8 @@ static inline unsigned int xfrm_aevent_msgsize(struct xfrm_state *x) + nla_total_size_64bit(sizeof(struct xfrm_lifetime_cur)) + nla_total_size(sizeof(struct xfrm_mark)) + nla_total_size(4) /* XFRM_AE_RTHR */ - + nla_total_size(4); /* XFRM_AE_ETHR */ + + nla_total_size(4) /* XFRM_AE_ETHR */ + + nla_total_size(sizeof(x->dir)); /* XFRMA_SA_DIR */ } static int build_aevent(struct sk_buff *skb, struct xfrm_state *x, const struct km_event *c) @@ -2459,6 +2567,12 @@ static int build_aevent(struct sk_buff *skb, struct xfrm_state *x, const struct if (err) goto out_cancel; + if (x->dir) { + err = nla_put_u8(skb, XFRMA_SA_DIR, x->dir); + if (err) + goto out_cancel; + } + nlmsg_end(skb, nlh); return 0; @@ -3018,6 +3132,7 @@ EXPORT_SYMBOL_GPL(xfrm_msg_min); #undef XMSGSIZE const struct nla_policy xfrma_policy[XFRMA_MAX+1] = { + [XFRMA_UNSPEC] = { .strict_start_type = XFRMA_SA_DIR }, [XFRMA_SA] = { .len = sizeof(struct xfrm_usersa_info)}, [XFRMA_POLICY] = { .len = sizeof(struct xfrm_userpolicy_info)}, [XFRMA_LASTUSED] = { .type = NLA_U64}, @@ -3049,6 +3164,7 @@ const struct nla_policy xfrma_policy[XFRMA_MAX+1] = { [XFRMA_SET_MARK_MASK] = { .type = NLA_U32 }, [XFRMA_IF_ID] = { .type = NLA_U32 }, [XFRMA_MTIMER_THRESH] = { .type = NLA_U32 }, + [XFRMA_SA_DIR] = NLA_POLICY_RANGE(NLA_U8, XFRM_SA_DIR_IN, XFRM_SA_DIR_OUT), }; EXPORT_SYMBOL_GPL(xfrma_policy); @@ -3189,8 +3305,9 @@ static void xfrm_netlink_rcv(struct sk_buff *skb) static inline unsigned int xfrm_expire_msgsize(void) { - return NLMSG_ALIGN(sizeof(struct xfrm_user_expire)) - + nla_total_size(sizeof(struct xfrm_mark)); + return NLMSG_ALIGN(sizeof(struct xfrm_user_expire)) + + nla_total_size(sizeof(struct xfrm_mark)) + + nla_total_size(sizeof_field(struct xfrm_state, dir)); } static int build_expire(struct sk_buff *skb, struct xfrm_state *x, const struct km_event *c) @@ -3217,6 +3334,12 @@ static int build_expire(struct sk_buff *skb, struct xfrm_state *x, const struct if (err) return err; + if (x->dir) { + err = nla_put_u8(skb, XFRMA_SA_DIR, x->dir); + if (err) + return err; + } + nlmsg_end(skb, nlh); return 0; } @@ -3324,6 +3447,9 @@ static inline unsigned int xfrm_sa_len(struct xfrm_state *x) if (x->mapping_maxage) l += nla_total_size(sizeof(x->mapping_maxage)); + if (x->dir) + l += nla_total_size(sizeof(x->dir)); + return l; } -- cgit v1.2.3 From 601a0867f86cbb5e137ce485a7eb60cbf9fc5180 Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Tue, 30 Apr 2024 09:09:09 +0200 Subject: xfrm: Add dir validation to "out" data path lookup Introduces validation for the x->dir attribute within the XFRM output data lookup path. If the configured direction does not match the expected direction, output, increment the XfrmOutStateDirError counter and drop the packet to ensure data integrity and correct flow handling. grep -vw 0 /proc/net/xfrm_stat XfrmOutPolError 1 XfrmOutStateDirError 1 Signed-off-by: Antony Antony Reviewed-by: Sabrina Dubroca Reviewed-by: Nicolas Dichtel Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_proc.rst | 3 +++ include/uapi/linux/snmp.h | 1 + net/xfrm/xfrm_policy.c | 6 ++++++ net/xfrm/xfrm_proc.c | 1 + 4 files changed, 11 insertions(+) (limited to 'net') diff --git a/Documentation/networking/xfrm_proc.rst b/Documentation/networking/xfrm_proc.rst index 0a771c5a7399..5ac3acf4cf51 100644 --- a/Documentation/networking/xfrm_proc.rst +++ b/Documentation/networking/xfrm_proc.rst @@ -111,3 +111,6 @@ XfrmOutPolError: XfrmOutStateInvalid: State is invalid, perhaps expired + +XfrmOutStateDirError: + State direction mismatch (lookup found an input state on the output path, expected output or no direction) diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h index a0819c6a5988..23792b8412bd 100644 --- a/include/uapi/linux/snmp.h +++ b/include/uapi/linux/snmp.h @@ -337,6 +337,7 @@ enum LINUX_MIB_XFRMFWDHDRERROR, /* XfrmFwdHdrError*/ LINUX_MIB_XFRMOUTSTATEINVALID, /* XfrmOutStateInvalid */ LINUX_MIB_XFRMACQUIREERROR, /* XfrmAcquireError */ + LINUX_MIB_XFRMOUTSTATEDIRERROR, /* XfrmOutStateDirError */ __LINUX_MIB_XFRMMAX }; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 6affe5cd85d8..298b3a9eb48d 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2489,6 +2489,12 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl, x = xfrm_state_find(remote, local, fl, tmpl, policy, &error, family, policy->if_id); + if (x && x->dir && x->dir != XFRM_SA_DIR_OUT) { + XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTSTATEDIRERROR); + xfrm_state_put(x); + error = -EINVAL; + goto fail; + } if (x && x->km.state == XFRM_STATE_VALID) { xfrm[nx++] = x; diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c index 5f9bf8e5c933..98606f1078f7 100644 --- a/net/xfrm/xfrm_proc.c +++ b/net/xfrm/xfrm_proc.c @@ -41,6 +41,7 @@ static const struct snmp_mib xfrm_mib_list[] = { SNMP_MIB_ITEM("XfrmFwdHdrError", LINUX_MIB_XFRMFWDHDRERROR), SNMP_MIB_ITEM("XfrmOutStateInvalid", LINUX_MIB_XFRMOUTSTATEINVALID), SNMP_MIB_ITEM("XfrmAcquireError", LINUX_MIB_XFRMACQUIREERROR), + SNMP_MIB_ITEM("XfrmOutStateDirError", LINUX_MIB_XFRMOUTSTATEDIRERROR), SNMP_MIB_SENTINEL }; -- cgit v1.2.3 From 304b44f0d5a4c2f91f82f7c31538d00485fb484c Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Tue, 30 Apr 2024 09:09:29 +0200 Subject: xfrm: Add dir validation to "in" data path lookup Introduces validation for the x->dir attribute within the XFRM input data lookup path. If the configured direction does not match the expected direction, input, increment the XfrmInStateDirError counter and drop the packet to ensure data integrity and correct flow handling. grep -vw 0 /proc/net/xfrm_stat XfrmInStateDirError 1 Signed-off-by: Antony Antony Reviewed-by: Sabrina Dubroca Reviewed-by: Nicolas Dichtel Signed-off-by: Steffen Klassert --- Documentation/networking/xfrm_proc.rst | 3 +++ include/uapi/linux/snmp.h | 1 + net/ipv6/xfrm6_input.c | 7 +++++++ net/xfrm/xfrm_input.c | 11 +++++++++++ net/xfrm/xfrm_proc.c | 1 + 5 files changed, 23 insertions(+) (limited to 'net') diff --git a/Documentation/networking/xfrm_proc.rst b/Documentation/networking/xfrm_proc.rst index 5ac3acf4cf51..973d1571acac 100644 --- a/Documentation/networking/xfrm_proc.rst +++ b/Documentation/networking/xfrm_proc.rst @@ -73,6 +73,9 @@ XfrmAcquireError: XfrmFwdHdrError: Forward routing of a packet is not allowed +XfrmInStateDirError: + State direction mismatch (lookup found an output state on the input path, expected input or no direction) + Outbound errors ~~~~~~~~~~~~~~~ XfrmOutError: diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h index 23792b8412bd..adf5fd78dd50 100644 --- a/include/uapi/linux/snmp.h +++ b/include/uapi/linux/snmp.h @@ -338,6 +338,7 @@ enum LINUX_MIB_XFRMOUTSTATEINVALID, /* XfrmOutStateInvalid */ LINUX_MIB_XFRMACQUIREERROR, /* XfrmAcquireError */ LINUX_MIB_XFRMOUTSTATEDIRERROR, /* XfrmOutStateDirError */ + LINUX_MIB_XFRMINSTATEDIRERROR, /* XfrmInStateDirError */ __LINUX_MIB_XFRMMAX }; diff --git a/net/ipv6/xfrm6_input.c b/net/ipv6/xfrm6_input.c index 2c6aeb090b7a..d5bac0d76b6e 100644 --- a/net/ipv6/xfrm6_input.c +++ b/net/ipv6/xfrm6_input.c @@ -266,6 +266,13 @@ int xfrm6_input_addr(struct sk_buff *skb, xfrm_address_t *daddr, if (!x) continue; + if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) { + XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR); + xfrm_state_put(x); + x = NULL; + continue; + } + spin_lock(&x->lock); if ((!i || (x->props.flags & XFRM_STATE_WILDRECV)) && diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 161f535c8b94..71b42de6e3c9 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -466,6 +466,11 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) if (encap_type < 0 || (xo && xo->flags & XFRM_GRO)) { x = xfrm_input_state(skb); + if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) { + XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR); + goto drop; + } + if (unlikely(x->km.state != XFRM_STATE_VALID)) { if (x->km.state == XFRM_STATE_ACQ) XFRM_INC_STATS(net, LINUX_MIB_XFRMACQUIREERROR); @@ -571,6 +576,12 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type) goto drop; } + if (unlikely(x->dir && x->dir != XFRM_SA_DIR_IN)) { + XFRM_INC_STATS(net, LINUX_MIB_XFRMINSTATEDIRERROR); + xfrm_state_put(x); + goto drop; + } + skb->mark = xfrm_smark_get(skb->mark, x); sp->xvec[sp->len++] = x; diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c index 98606f1078f7..eeb984be03a7 100644 --- a/net/xfrm/xfrm_proc.c +++ b/net/xfrm/xfrm_proc.c @@ -42,6 +42,7 @@ static const struct snmp_mib xfrm_mib_list[] = { SNMP_MIB_ITEM("XfrmOutStateInvalid", LINUX_MIB_XFRMOUTSTATEINVALID), SNMP_MIB_ITEM("XfrmAcquireError", LINUX_MIB_XFRMACQUIREERROR), SNMP_MIB_ITEM("XfrmOutStateDirError", LINUX_MIB_XFRMOUTSTATEDIRERROR), + SNMP_MIB_ITEM("XfrmInStateDirError", LINUX_MIB_XFRMINSTATEDIRERROR), SNMP_MIB_SENTINEL }; -- cgit v1.2.3 From 451b509678978250702bd833c69276e70531f07a Mon Sep 17 00:00:00 2001 From: Antony Antony Date: Tue, 30 Apr 2024 09:09:45 +0200 Subject: xfrm: Restrict SA direction attribute to specific netlink message types Reject the usage of the SA_DIR attribute in xfrm netlink messages when it's not applicable. This ensures that SA_DIR is only accepted for certain message types (NEWSA, UPDSA, and ALLOCSPI) Signed-off-by: Antony Antony Reviewed-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index f5eb3af4fb81..e83c687bd64e 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -3213,6 +3213,24 @@ static const struct xfrm_link { [XFRM_MSG_GETDEFAULT - XFRM_MSG_BASE] = { .doit = xfrm_get_default }, }; +static int xfrm_reject_unused_attr(int type, struct nlattr **attrs, + struct netlink_ext_ack *extack) +{ + if (attrs[XFRMA_SA_DIR]) { + switch (type) { + case XFRM_MSG_NEWSA: + case XFRM_MSG_UPDSA: + case XFRM_MSG_ALLOCSPI: + break; + default: + NL_SET_ERR_MSG(extack, "Invalid attribute SA_DIR"); + return -EINVAL; + } + } + + return 0; +} + static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, struct netlink_ext_ack *extack) { @@ -3272,6 +3290,12 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, if (err < 0) goto err; + if (!link->nla_pol || link->nla_pol == xfrma_policy) { + err = xfrm_reject_unused_attr((type + XFRM_MSG_BASE), attrs, extack); + if (err < 0) + goto err; + } + if (link->doit == NULL) { err = -EINVAL; goto err; -- cgit v1.2.3 From 59c878cbcdd80ed39315573b3511d0acfd3501b5 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Sat, 27 Apr 2024 20:24:18 +0200 Subject: net: bridge: fix multicast-to-unicast with fraglist GSO Calling skb_copy on a SKB_GSO_FRAGLIST skb is not valid, since it returns an invalid linearized skb. This code only needs to change the ethernet header, so pskb_copy is the right function to call here. Fixes: 6db6f0eae605 ("bridge: multicast to unicast") Signed-off-by: Felix Fietkau Acked-by: Paolo Abeni Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_forward.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 7431f89e897b..d7c35f55bd69 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -266,7 +266,7 @@ static void maybe_deliver_addr(struct net_bridge_port *p, struct sk_buff *skb, if (skb->dev == p->dev && ether_addr_equal(src, addr)) return; - skb = skb_copy(skb, GFP_ATOMIC); + skb = pskb_copy(skb, GFP_ATOMIC); if (!skb) { DEV_STATS_INC(dev, tx_dropped); return; -- cgit v1.2.3 From d091e579b864fa790dd6a0cd537a22c383126681 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Sat, 27 Apr 2024 20:24:19 +0200 Subject: net: core: reject skb_copy(_expand) for fraglist GSO skbs SKB_GSO_FRAGLIST skbs must not be linearized, otherwise they become invalid. Return NULL if such an skb is passed to skb_copy or skb_copy_expand, in order to prevent a crash on a potential later call to skb_gso_segment. Fixes: 3a1296a38d0c ("net: Support GRO/GSO fraglist chaining.") Signed-off-by: Felix Fietkau Signed-off-by: David S. Miller --- net/core/skbuff.c | 27 +++++++++++++++++++-------- 1 file changed, 19 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b99127712e67..4096e679f61c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2123,11 +2123,17 @@ static inline int skb_alloc_rx_flag(const struct sk_buff *skb) struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t gfp_mask) { - int headerlen = skb_headroom(skb); - unsigned int size = skb_end_offset(skb) + skb->data_len; - struct sk_buff *n = __alloc_skb(size, gfp_mask, - skb_alloc_rx_flag(skb), NUMA_NO_NODE); + struct sk_buff *n; + unsigned int size; + int headerlen; + + if (WARN_ON_ONCE(skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST)) + return NULL; + headerlen = skb_headroom(skb); + size = skb_end_offset(skb) + skb->data_len; + n = __alloc_skb(size, gfp_mask, + skb_alloc_rx_flag(skb), NUMA_NO_NODE); if (!n) return NULL; @@ -2455,12 +2461,17 @@ struct sk_buff *skb_copy_expand(const struct sk_buff *skb, /* * Allocate the copy buffer */ - struct sk_buff *n = __alloc_skb(newheadroom + skb->len + newtailroom, - gfp_mask, skb_alloc_rx_flag(skb), - NUMA_NO_NODE); - int oldheadroom = skb_headroom(skb); int head_copy_len, head_copy_off; + struct sk_buff *n; + int oldheadroom; + + if (WARN_ON_ONCE(skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST)) + return NULL; + oldheadroom = skb_headroom(skb); + n = __alloc_skb(newheadroom + skb->len + newtailroom, + gfp_mask, skb_alloc_rx_flag(skb), + NUMA_NO_NODE); if (!n) return NULL; -- cgit v1.2.3 From fff6e6accdb71e5dd36e50478bd82d5409ac32d2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 29 Apr 2024 18:36:43 +0000 Subject: ipv6: anycast: use call_rcu_hurry() in aca_put() This is a followup of commit b5327b9a300e ("ipv6: use call_rcu_hurry() in fib6_info_release()"). I had another pmtu.sh failure, and found another lazy call_rcu() causing this failure. aca_free_rcu() calls fib6_info_release() which releases devices references. We must not delay it too much or risk unregister_netdevice/ref_tracker traces because references to netdev are not released in time. This should speedup device/netns dismantles when CONFIG_RCU_LAZY=y Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/anycast.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 0f2506e35359..0627c4c18d1a 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -252,9 +252,8 @@ static void aca_free_rcu(struct rcu_head *h) static void aca_put(struct ifacaddr6 *ac) { - if (refcount_dec_and_test(&ac->aca_refcnt)) { - call_rcu(&ac->rcu, aca_free_rcu); - } + if (refcount_dec_and_test(&ac->aca_refcnt)) + call_rcu_hurry(&ac->rcu, aca_free_rcu); } static struct ifacaddr6 *aca_alloc(struct fib6_info *f6i, -- cgit v1.2.3 From 5fb70e7275a61dd404f684370e1add7fe0ebe9c5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 18 Mar 2024 20:29:53 +0000 Subject: netfs, 9p: Implement helpers for new write code Implement the helpers for the new write code in 9p. There's now an optional ->prepare_write() that allows the filesystem to set the parameters for the next write, such as maximum size and maximum segment count, and an ->issue_write() that is called to initiate an (asynchronous) write operation. Signed-off-by: David Howells Reviewed-by: Jeff Layton cc: Eric Van Hensbergen cc: Latchesar Ionkov cc: Dominique Martinet cc: Christian Schoenebeck cc: v9fs@lists.linux.dev cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org --- fs/9p/vfs_addr.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ include/net/9p/client.h | 2 ++ net/9p/Kconfig | 1 + net/9p/client.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 100 insertions(+) (limited to 'net') diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 5a943c122d83..07d03efdd594 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -26,6 +26,40 @@ #include "cache.h" #include "fid.h" +/* + * Writeback calls this when it finds a folio that needs uploading. This isn't + * called if writeback only has copy-to-cache to deal with. + */ +static void v9fs_begin_writeback(struct netfs_io_request *wreq) +{ + struct p9_fid *fid; + + fid = v9fs_fid_find_inode(wreq->inode, true, INVALID_UID, true); + if (!fid) { + WARN_ONCE(1, "folio expected an open fid inode->i_ino=%lx\n", + wreq->inode->i_ino); + return; + } + + wreq->wsize = fid->clnt->msize - P9_IOHDRSZ; + if (fid->iounit) + wreq->wsize = min(wreq->wsize, fid->iounit); + wreq->netfs_priv = fid; + wreq->io_streams[0].avail = true; +} + +/* + * Issue a subrequest to write to the server. + */ +static void v9fs_issue_write(struct netfs_io_subrequest *subreq) +{ + struct p9_fid *fid = subreq->rreq->netfs_priv; + int err, len; + + len = p9_client_write(fid, subreq->start, &subreq->io_iter, &err); + netfs_write_subrequest_terminated(subreq, len ?: err, false); +} + static void v9fs_upload_to_server(struct netfs_io_subrequest *subreq) { struct p9_fid *fid = subreq->rreq->netfs_priv; @@ -92,6 +126,14 @@ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) rreq->origin == NETFS_UNBUFFERED_WRITE || rreq->origin == NETFS_DIO_WRITE); +#if 0 // TODO: Cut over + if (rreq->origin == NETFS_WRITEBACK) + return 0; /* We don't get the write handle until we find we + * have actually dirty data and not just + * copy-to-cache data. + */ +#endif + if (file) { fid = file->private_data; if (!fid) @@ -103,6 +145,10 @@ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) goto no_fid; } + rreq->wsize = fid->clnt->msize - P9_IOHDRSZ; + if (fid->iounit) + rreq->wsize = min(rreq->wsize, fid->iounit); + /* we might need to read from a fid that was opened write-only * for read-modify-write of page cache, use the writeback fid * for that */ @@ -131,6 +177,8 @@ const struct netfs_request_ops v9fs_req_ops = { .init_request = v9fs_init_request, .free_request = v9fs_free_request, .issue_read = v9fs_issue_read, + .begin_writeback = v9fs_begin_writeback, + .issue_write = v9fs_issue_write, .create_write_requests = v9fs_create_write_requests, }; diff --git a/include/net/9p/client.h b/include/net/9p/client.h index 78ebcf782ce5..4f785098c67a 100644 --- a/include/net/9p/client.h +++ b/include/net/9p/client.h @@ -207,6 +207,8 @@ int p9_client_read(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err int p9_client_read_once(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err); int p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err); +struct netfs_io_subrequest; +void p9_client_write_subreq(struct netfs_io_subrequest *subreq); int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset); int p9dirent_read(struct p9_client *clnt, char *buf, int len, struct p9_dirent *dirent); diff --git a/net/9p/Kconfig b/net/9p/Kconfig index 00ebce9e5a65..bcdab9c23b40 100644 --- a/net/9p/Kconfig +++ b/net/9p/Kconfig @@ -5,6 +5,7 @@ menuconfig NET_9P tristate "Plan 9 Resource Sharing Support (9P2000)" + select NETFS_SUPPORT help If you say Y here, you will get experimental support for Plan 9 resource sharing via the 9P2000 protocol. diff --git a/net/9p/client.c b/net/9p/client.c index f7e90b4769bb..00774656eeac 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -1661,6 +1662,54 @@ p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err) } EXPORT_SYMBOL(p9_client_write); +void +p9_client_write_subreq(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *wreq = subreq->rreq; + struct p9_fid *fid = wreq->netfs_priv; + struct p9_client *clnt = fid->clnt; + struct p9_req_t *req; + unsigned long long start = subreq->start + subreq->transferred; + int written, len = subreq->len - subreq->transferred; + int err; + + p9_debug(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu len %d\n", + fid->fid, start, len); + + /* Don't bother zerocopy for small IO (< 1024) */ + if (clnt->trans_mod->zc_request && len > 1024) { + req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, &subreq->io_iter, + 0, wreq->len, P9_ZC_HDR_SZ, "dqd", + fid->fid, start, len); + } else { + req = p9_client_rpc(clnt, P9_TWRITE, "dqV", fid->fid, + start, len, &subreq->io_iter); + } + if (IS_ERR(req)) { + netfs_write_subrequest_terminated(subreq, PTR_ERR(req), false); + return; + } + + err = p9pdu_readf(&req->rc, clnt->proto_version, "d", &written); + if (err) { + trace_9p_protocol_dump(clnt, &req->rc); + p9_req_put(clnt, req); + netfs_write_subrequest_terminated(subreq, err, false); + return; + } + + if (written > len) { + pr_err("bogus RWRITE count (%d > %u)\n", written, len); + written = len; + } + + p9_debug(P9_DEBUG_9P, "<<< RWRITE count %d\n", len); + + p9_req_put(clnt, req); + netfs_write_subrequest_terminated(subreq, written, false); +} +EXPORT_SYMBOL(p9_client_write_subreq); + struct p9_wstat *p9_client_stat(struct p9_fid *fid) { int err; -- cgit v1.2.3 From 8953285d7bd63c12b007432a9b4587fa2fad49fb Mon Sep 17 00:00:00 2001 From: Jeffrey Altman Date: Fri, 19 Apr 2024 13:30:57 -0300 Subject: rxrpc: Clients must accept conn from any address The find connection logic of Transarc's Rx was modified in the mid-1990s to support multi-homed servers which might send a response packet from an address other than the destination address in the received packet. The rules for accepting a packet by an Rx initiator (RX_CLIENT_CONNECTION) were altered to permit acceptance of a packet from any address provided that the port number was unchanged and all of the connection identifiers matched (Epoch, CID, SecurityClass, ...). This change applies the same rules to the Linux implementation which makes it consistent with IBM AFS 3.6, Arla, OpenAFS and AuriStorFS. Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Signed-off-by: Jeffrey Altman Acked-by: David Howells Signed-off-by: Marc Dionne Link: https://lore.kernel.org/r/20240419163057.4141728-1-marc.dionne@auristor.com Signed-off-by: Jakub Kicinski --- net/rxrpc/conn_object.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 0af4642aeec4..1539d315afe7 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -119,18 +119,13 @@ struct rxrpc_connection *rxrpc_find_client_connection_rcu(struct rxrpc_local *lo switch (srx->transport.family) { case AF_INET: if (peer->srx.transport.sin.sin_port != - srx->transport.sin.sin_port || - peer->srx.transport.sin.sin_addr.s_addr != - srx->transport.sin.sin_addr.s_addr) + srx->transport.sin.sin_port) goto not_found; break; #ifdef CONFIG_AF_RXRPC_IPV6 case AF_INET6: if (peer->srx.transport.sin6.sin6_port != - srx->transport.sin6.sin6_port || - memcmp(&peer->srx.transport.sin6.sin6_addr, - &srx->transport.sin6.sin6_addr, - sizeof(struct in6_addr)) != 0) + srx->transport.sin6.sin6_port) goto not_found; break; #endif -- cgit v1.2.3 From 42033d0cfc8606adc9cb7e18d4c73d2e9a6289e5 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:07 -0700 Subject: arp: Move ATF_COM setting in arp_req_set(). In arp_req_set(), if ATF_PERM is set in arpreq.arp_flags, ATF_COM is set automatically. The flag will be used later for neigh_update() only when a neighbour entry is found. Let's set ATF_COM just before calling neigh_update(). Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-2-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/arp.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index ab82ca104496..3093374165fa 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1054,8 +1054,7 @@ static int arp_req_set(struct net *net, struct arpreq *r, return arp_req_set_public(net, r, dev); ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; - if (r->arp_flags & ATF_PERM) - r->arp_flags |= ATF_COM; + if (!dev) { struct rtable *rt = ip_route_output(net, ip, 0, 0, 0, RT_SCOPE_LINK); @@ -1092,8 +1091,12 @@ static int arp_req_set(struct net *net, struct arpreq *r, err = PTR_ERR(neigh); if (!IS_ERR(neigh)) { unsigned int state = NUD_STALE; - if (r->arp_flags & ATF_PERM) + + if (r->arp_flags & ATF_PERM) { + r->arp_flags |= ATF_COM; state = NUD_PERMANENT; + } + err = neigh_update(neigh, (r->arp_flags & ATF_COM) ? r->arp_ha.sa_data : NULL, state, NEIGH_UPDATE_F_OVERRIDE | -- cgit v1.2.3 From 0592367424bbbdef506ef1450f5b8beec148f8a4 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:08 -0700 Subject: arp: Validate netmask earlier for SIOCDARP and SIOCSARP in arp_ioctl(). When ioctl(SIOCDARP/SIOCSARP) is issued with ATF_PUBL, r.arp_netmask must be 0.0.0.0 or 255.255.255.255. Currently, the netmask is validated in arp_req_delete_public() or arp_req_set_public() under rtnl_lock(). We have ATF_NETMASK test in arp_ioctl() before holding rtnl_lock(), so let's move the netmask validation there. Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-3-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/arp.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 3093374165fa..b20a5771d069 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1023,11 +1023,8 @@ static int arp_req_set_proxy(struct net *net, struct net_device *dev, int on) static int arp_req_set_public(struct net *net, struct arpreq *r, struct net_device *dev) { - __be32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; __be32 mask = ((struct sockaddr_in *)&r->arp_netmask)->sin_addr.s_addr; - if (mask && mask != htonl(0xFFFFFFFF)) - return -EINVAL; if (!dev && (r->arp_flags & ATF_COM)) { dev = dev_getbyhwaddr_rcu(net, r->arp_ha.sa_family, r->arp_ha.sa_data); @@ -1035,6 +1032,8 @@ static int arp_req_set_public(struct net *net, struct arpreq *r, return -ENODEV; } if (mask) { + __be32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; + if (!pneigh_lookup(&arp_tbl, net, &ip, dev, 1)) return -ENOBUFS; return 0; @@ -1171,14 +1170,13 @@ int arp_invalidate(struct net_device *dev, __be32 ip, bool force) static int arp_req_delete_public(struct net *net, struct arpreq *r, struct net_device *dev) { - __be32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr; __be32 mask = ((struct sockaddr_in *)&r->arp_netmask)->sin_addr.s_addr; - if (mask == htonl(0xFFFFFFFF)) - return pneigh_delete(&arp_tbl, net, &ip, dev); + if (mask) { + __be32 ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; - if (mask) - return -EINVAL; + return pneigh_delete(&arp_tbl, net, &ip, dev); + } return arp_req_set_proxy(net, dev, 0); } @@ -1211,9 +1209,10 @@ static int arp_req_delete(struct net *net, struct arpreq *r, int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) { - int err; - struct arpreq r; struct net_device *dev = NULL; + struct arpreq r; + __be32 *netmask; + int err; switch (cmd) { case SIOCDARP: @@ -1236,9 +1235,13 @@ int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) if (!(r.arp_flags & ATF_PUBL) && (r.arp_flags & (ATF_NETMASK | ATF_DONTPUB))) return -EINVAL; + + netmask = &((struct sockaddr_in *)&r.arp_netmask)->sin_addr.s_addr; if (!(r.arp_flags & ATF_NETMASK)) - ((struct sockaddr_in *)&r.arp_netmask)->sin_addr.s_addr = - htonl(0xFFFFFFFFUL); + *netmask = htonl(0xFFFFFFFFUL); + else if (*netmask && *netmask != htonl(0xFFFFFFFFUL)) + return -EINVAL; + rtnl_lock(); if (r.arp_dev[0]) { err = -ENODEV; -- cgit v1.2.3 From f8696133f6aa4e6a83c9fb2d9dddc6d194a2ba1f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:09 -0700 Subject: arp: Factorise ip_route_output() call in arp_req_set() and arp_req_delete(). When ioctl(SIOCDARP/SIOCSARP) is issued for non-proxy entry (no ATF_COM) without arpreq.arp_dev[] set, arp_req_set() and arp_req_delete() looks up dev based on IPv4 address by ip_route_output(). Let's factorise the same code as arp_req_dev(). Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-4-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/arp.c | 50 ++++++++++++++++++++++++++++++-------------------- 1 file changed, 30 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index b20a5771d069..ac3e15799c2f 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1003,6 +1003,27 @@ out_of_mem: * User level interface (ioctl) */ +static struct net_device *arp_req_dev(struct net *net, struct arpreq *r) +{ + struct net_device *dev; + struct rtable *rt; + __be32 ip; + + ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; + + rt = ip_route_output(net, ip, 0, 0, 0, RT_SCOPE_LINK); + if (IS_ERR(rt)) + return ERR_CAST(rt); + + dev = rt->dst.dev; + ip_rt_put(rt); + + if (!dev) + return ERR_PTR(-EINVAL); + + return dev; +} + /* * Set (create) an ARP cache entry. */ @@ -1045,25 +1066,17 @@ static int arp_req_set_public(struct net *net, struct arpreq *r, static int arp_req_set(struct net *net, struct arpreq *r, struct net_device *dev) { - __be32 ip; struct neighbour *neigh; + __be32 ip; int err; if (r->arp_flags & ATF_PUBL) return arp_req_set_public(net, r, dev); - ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; - if (!dev) { - struct rtable *rt = ip_route_output(net, ip, 0, 0, 0, - RT_SCOPE_LINK); - - if (IS_ERR(rt)) - return PTR_ERR(rt); - dev = rt->dst.dev; - ip_rt_put(rt); - if (!dev) - return -EINVAL; + dev = arp_req_dev(net, r); + if (IS_ERR(dev)) + return PTR_ERR(dev); } switch (dev->type) { #if IS_ENABLED(CONFIG_FDDI) @@ -1086,6 +1099,8 @@ static int arp_req_set(struct net *net, struct arpreq *r, break; } + ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; + neigh = __neigh_lookup_errno(&arp_tbl, &ip, dev); err = PTR_ERR(neigh); if (!IS_ERR(neigh)) { @@ -1191,14 +1206,9 @@ static int arp_req_delete(struct net *net, struct arpreq *r, ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; if (!dev) { - struct rtable *rt = ip_route_output(net, ip, 0, 0, 0, - RT_SCOPE_LINK); - if (IS_ERR(rt)) - return PTR_ERR(rt); - dev = rt->dst.dev; - ip_rt_put(rt); - if (!dev) - return -EINVAL; + dev = arp_req_dev(net, r); + if (IS_ERR(dev)) + return PTR_ERR(dev); } return arp_invalidate(dev, ip, true); } -- cgit v1.2.3 From 51e9ba48d48786da89d2695be9a1cab40b2afc31 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:10 -0700 Subject: arp: Remove a nest in arp_req_get(). This is a prep patch to make the following changes tidy. No functional change intended. Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-5-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/arp.c | 31 ++++++++++++++++++------------- 1 file changed, 18 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index ac3e15799c2f..60f633b24ec8 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1138,23 +1138,28 @@ static int arp_req_get(struct arpreq *r, struct net_device *dev) { __be32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr; struct neighbour *neigh; - int err = -ENXIO; neigh = neigh_lookup(&arp_tbl, &ip, dev); - if (neigh) { - if (!(READ_ONCE(neigh->nud_state) & NUD_NOARP)) { - read_lock_bh(&neigh->lock); - memcpy(r->arp_ha.sa_data, neigh->ha, - min(dev->addr_len, sizeof(r->arp_ha.sa_data_min))); - r->arp_flags = arp_state_to_flags(neigh); - read_unlock_bh(&neigh->lock); - r->arp_ha.sa_family = dev->type; - strscpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); - err = 0; - } + if (!neigh) + return -ENXIO; + + if (READ_ONCE(neigh->nud_state) & NUD_NOARP) { neigh_release(neigh); + return -ENXIO; } - return err; + + read_lock_bh(&neigh->lock); + memcpy(r->arp_ha.sa_data, neigh->ha, + min(dev->addr_len, sizeof(r->arp_ha.sa_data_min))); + r->arp_flags = arp_state_to_flags(neigh); + read_unlock_bh(&neigh->lock); + + neigh_release(neigh); + + r->arp_ha.sa_family = dev->type; + strscpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); + + return 0; } int arp_invalidate(struct net_device *dev, __be32 ip, bool force) -- cgit v1.2.3 From a428bfc77a4dd4ba19b7646e887fa655fcfee5a0 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:11 -0700 Subject: arp: Get dev after calling arp_req_(delete|set|get)(). arp_ioctl() holds rtnl_lock() first regardless of cmd (SIOCDARP, SIOCSARP, and SIOCGARP) to get net_device by __dev_get_by_name() and copy dev->name safely. In the SIOCGARP path, arp_req_get() calls neigh_lookup(), which looks up a neighbour entry under RCU. We will extend the RCU section not to take rtnl_lock() and instead use dev_get_by_name_rcu() for SIOCGARP. As a preparation, let's move __dev_get_by_name() into another function and call it from arp_req_delete(), arp_req_set(), and arp_req_get(). Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-6-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/arp.c | 86 ++++++++++++++++++++++++++++++++++------------------------ 1 file changed, 50 insertions(+), 36 deletions(-) (limited to 'net') diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 60f633b24ec8..5034920be85a 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1003,12 +1003,36 @@ out_of_mem: * User level interface (ioctl) */ +static struct net_device *arp_req_dev_by_name(struct net *net, struct arpreq *r) +{ + struct net_device *dev; + + dev = __dev_get_by_name(net, r->arp_dev); + if (!dev) + return ERR_PTR(-ENODEV); + + /* Mmmm... It is wrong... ARPHRD_NETROM == 0 */ + if (!r->arp_ha.sa_family) + r->arp_ha.sa_family = dev->type; + + if ((r->arp_flags & ATF_COM) && r->arp_ha.sa_family != dev->type) + return ERR_PTR(-EINVAL); + + return dev; +} + static struct net_device *arp_req_dev(struct net *net, struct arpreq *r) { struct net_device *dev; struct rtable *rt; __be32 ip; + if (r->arp_dev[0]) + return arp_req_dev_by_name(net, r); + + if (r->arp_flags & ATF_PUBL) + return NULL; + ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; rt = ip_route_output(net, ip, 0, 0, 0, RT_SCOPE_LINK); @@ -1063,21 +1087,20 @@ static int arp_req_set_public(struct net *net, struct arpreq *r, return arp_req_set_proxy(net, dev, 1); } -static int arp_req_set(struct net *net, struct arpreq *r, - struct net_device *dev) +static int arp_req_set(struct net *net, struct arpreq *r) { struct neighbour *neigh; + struct net_device *dev; __be32 ip; int err; + dev = arp_req_dev(net, r); + if (IS_ERR(dev)) + return PTR_ERR(dev); + if (r->arp_flags & ATF_PUBL) return arp_req_set_public(net, r, dev); - if (!dev) { - dev = arp_req_dev(net, r); - if (IS_ERR(dev)) - return PTR_ERR(dev); - } switch (dev->type) { #if IS_ENABLED(CONFIG_FDDI) case ARPHRD_FDDI: @@ -1134,10 +1157,18 @@ static unsigned int arp_state_to_flags(struct neighbour *neigh) * Get an ARP cache entry. */ -static int arp_req_get(struct arpreq *r, struct net_device *dev) +static int arp_req_get(struct net *net, struct arpreq *r) { __be32 ip = ((struct sockaddr_in *) &r->arp_pa)->sin_addr.s_addr; struct neighbour *neigh; + struct net_device *dev; + + if (!r->arp_dev[0]) + return -ENODEV; + + dev = arp_req_dev_by_name(net, r); + if (IS_ERR(dev)) + return PTR_ERR(dev); neigh = neigh_lookup(&arp_tbl, &ip, dev); if (!neigh) @@ -1201,20 +1232,20 @@ static int arp_req_delete_public(struct net *net, struct arpreq *r, return arp_req_set_proxy(net, dev, 0); } -static int arp_req_delete(struct net *net, struct arpreq *r, - struct net_device *dev) +static int arp_req_delete(struct net *net, struct arpreq *r) { + struct net_device *dev; __be32 ip; + dev = arp_req_dev(net, r); + if (IS_ERR(dev)) + return PTR_ERR(dev); + if (r->arp_flags & ATF_PUBL) return arp_req_delete_public(net, r, dev); ip = ((struct sockaddr_in *)&r->arp_pa)->sin_addr.s_addr; - if (!dev) { - dev = arp_req_dev(net, r); - if (IS_ERR(dev)) - return PTR_ERR(dev); - } + return arp_invalidate(dev, ip, true); } @@ -1224,7 +1255,6 @@ static int arp_req_delete(struct net *net, struct arpreq *r, int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) { - struct net_device *dev = NULL; struct arpreq r; __be32 *netmask; int err; @@ -1258,35 +1288,19 @@ int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) return -EINVAL; rtnl_lock(); - if (r.arp_dev[0]) { - err = -ENODEV; - dev = __dev_get_by_name(net, r.arp_dev); - if (!dev) - goto out; - - /* Mmmm... It is wrong... ARPHRD_NETROM==0 */ - if (!r.arp_ha.sa_family) - r.arp_ha.sa_family = dev->type; - err = -EINVAL; - if ((r.arp_flags & ATF_COM) && r.arp_ha.sa_family != dev->type) - goto out; - } else if (cmd == SIOCGARP) { - err = -ENODEV; - goto out; - } switch (cmd) { case SIOCDARP: - err = arp_req_delete(net, &r, dev); + err = arp_req_delete(net, &r); break; case SIOCSARP: - err = arp_req_set(net, &r, dev); + err = arp_req_set(net, &r); break; case SIOCGARP: - err = arp_req_get(&r, dev); + err = arp_req_get(net, &r); break; } -out: + rtnl_unlock(); if (cmd == SIOCGARP && !err && copy_to_user(arg, &r, sizeof(r))) err = -EFAULT; -- cgit v1.2.3 From 0840556e5a3a331b6932ef17dd4bc94445df3297 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:12 -0700 Subject: net: Protect dev->name by seqlock. We will convert ioctl(SIOCGARP) to RCU, and then we need to copy dev->name which is currently protected by rtnl_lock(). This patch does the following: 1) Add seqlock netdev_rename_lock to protect dev->name 2) Add netdev_copy_name() that copies dev->name to buffer under netdev_rename_lock 3) Use netdev_copy_name() in netdev_get_name() and drop devnet_rename_sem Suggested-by: Eric Dumazet Link: https://lore.kernel.org/netdev/CANn89iJEWs7AYSJqGCUABeVqOCTkErponfZdT5kV-iD=-SajnQ@mail.gmail.com/ Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-7-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 1 + net/core/dev.c | 27 +++++++++++++++++++++++---- 2 files changed, 24 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f849e7d110ed..41853424b41d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3136,6 +3136,7 @@ struct net_device *netdev_get_by_name(struct net *net, const char *name, netdevice_tracker *tracker, gfp_t gfp); struct net_device *dev_get_by_index_rcu(struct net *net, int ifindex); struct net_device *dev_get_by_napi_id(unsigned int napi_id); +void netdev_copy_name(struct net_device *dev, char *name); static inline int dev_hard_header(struct sk_buff *skb, struct net_device *dev, unsigned short type, diff --git a/net/core/dev.c b/net/core/dev.c index cd7ba50eac15..e02d2363347e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -940,6 +940,18 @@ struct net_device *dev_get_by_napi_id(unsigned int napi_id) } EXPORT_SYMBOL(dev_get_by_napi_id); +static DEFINE_SEQLOCK(netdev_rename_lock); + +void netdev_copy_name(struct net_device *dev, char *name) +{ + unsigned int seq; + + do { + seq = read_seqbegin(&netdev_rename_lock); + strscpy(name, dev->name, IFNAMSIZ); + } while (read_seqretry(&netdev_rename_lock, seq)); +} + /** * netdev_get_name - get a netdevice name, knowing its ifindex. * @net: network namespace @@ -951,7 +963,6 @@ int netdev_get_name(struct net *net, char *name, int ifindex) struct net_device *dev; int ret; - down_read(&devnet_rename_sem); rcu_read_lock(); dev = dev_get_by_index_rcu(net, ifindex); @@ -960,12 +971,11 @@ int netdev_get_name(struct net *net, char *name, int ifindex) goto out; } - strcpy(name, dev->name); + netdev_copy_name(dev, name); ret = 0; out: rcu_read_unlock(); - up_read(&devnet_rename_sem); return ret; } @@ -1217,7 +1227,10 @@ int dev_change_name(struct net_device *dev, const char *newname) memcpy(oldname, dev->name, IFNAMSIZ); + write_seqlock(&netdev_rename_lock); err = dev_get_valid_name(net, dev, newname); + write_sequnlock(&netdev_rename_lock); + if (err < 0) { up_write(&devnet_rename_sem); return err; @@ -1257,7 +1270,9 @@ rollback: if (err >= 0) { err = ret; down_write(&devnet_rename_sem); + write_seqlock(&netdev_rename_lock); memcpy(dev->name, oldname, IFNAMSIZ); + write_sequnlock(&netdev_rename_lock); memcpy(oldname, newname, IFNAMSIZ); WRITE_ONCE(dev->name_assign_type, old_assign_type); old_assign_type = NET_NAME_RENAMED; @@ -11403,8 +11418,12 @@ int __dev_change_net_namespace(struct net_device *dev, struct net *net, dev_net_set(dev, net); dev->ifindex = new_ifindex; - if (new_name[0]) /* Rename the netdev to prepared name */ + if (new_name[0]) { + /* Rename the netdev to prepared name */ + write_seqlock(&netdev_rename_lock); strscpy(dev->name, new_name, IFNAMSIZ); + write_sequnlock(&netdev_rename_lock); + } /* Fixup kobjects */ dev_set_uevent_suppress(&dev->dev, 1); -- cgit v1.2.3 From bf4ea58874df3d43f7264709cec7fe320616552c Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 29 Apr 2024 18:58:13 -0700 Subject: arp: Convert ioctl(SIOCGARP) to RCU. ioctl(SIOCGARP) holds rtnl_lock() to get netdev by __dev_get_by_name() and copy dev->name safely and calls neigh_lookup() later, which looks up a neighbour entry under RCU. Let's replace __dev_get_by_name() with dev_get_by_name_rcu() and strscpy() with netdev_copy_name() to avoid locking rtnl_lock(). Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240430015813.71143-8-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/arp.c | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 5034920be85a..11c1519b3699 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -1003,11 +1003,15 @@ out_of_mem: * User level interface (ioctl) */ -static struct net_device *arp_req_dev_by_name(struct net *net, struct arpreq *r) +static struct net_device *arp_req_dev_by_name(struct net *net, struct arpreq *r, + bool getarp) { struct net_device *dev; - dev = __dev_get_by_name(net, r->arp_dev); + if (getarp) + dev = dev_get_by_name_rcu(net, r->arp_dev); + else + dev = __dev_get_by_name(net, r->arp_dev); if (!dev) return ERR_PTR(-ENODEV); @@ -1028,7 +1032,7 @@ static struct net_device *arp_req_dev(struct net *net, struct arpreq *r) __be32 ip; if (r->arp_dev[0]) - return arp_req_dev_by_name(net, r); + return arp_req_dev_by_name(net, r, false); if (r->arp_flags & ATF_PUBL) return NULL; @@ -1166,7 +1170,7 @@ static int arp_req_get(struct net *net, struct arpreq *r) if (!r->arp_dev[0]) return -ENODEV; - dev = arp_req_dev_by_name(net, r); + dev = arp_req_dev_by_name(net, r, true); if (IS_ERR(dev)) return PTR_ERR(dev); @@ -1188,7 +1192,7 @@ static int arp_req_get(struct net *net, struct arpreq *r) neigh_release(neigh); r->arp_ha.sa_family = dev->type; - strscpy(r->arp_dev, dev->name, sizeof(r->arp_dev)); + netdev_copy_name(dev, r->arp_dev); return 0; } @@ -1287,23 +1291,27 @@ int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg) else if (*netmask && *netmask != htonl(0xFFFFFFFFUL)) return -EINVAL; - rtnl_lock(); - switch (cmd) { case SIOCDARP: + rtnl_lock(); err = arp_req_delete(net, &r); + rtnl_unlock(); break; case SIOCSARP: + rtnl_lock(); err = arp_req_set(net, &r); + rtnl_unlock(); break; case SIOCGARP: + rcu_read_lock(); err = arp_req_get(net, &r); + rcu_read_unlock(); + + if (!err && copy_to_user(arg, &r, sizeof(r))) + err = -EFAULT; break; } - rtnl_unlock(); - if (cmd == SIOCGARP && !err && copy_to_user(arg, &r, sizeof(r))) - err = -EFAULT; return err; } -- cgit v1.2.3 From 080cbb890286cd794f1ee788bbc5463e2deb7c2b Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 30 Apr 2024 15:53:37 +0200 Subject: tipc: fix UAF in error path Sam Page (sam4k) working with Trend Micro Zero Day Initiative reported a UAF in the tipc_buf_append() error path: BUG: KASAN: slab-use-after-free in kfree_skb_list_reason+0x47e/0x4c0 linux/net/core/skbuff.c:1183 Read of size 8 at addr ffff88804d2a7c80 by task poc/8034 CPU: 1 PID: 8034 Comm: poc Not tainted 6.8.2 #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-debian-1.16.0-5 04/01/2014 Call Trace: __dump_stack linux/lib/dump_stack.c:88 dump_stack_lvl+0xd9/0x1b0 linux/lib/dump_stack.c:106 print_address_description linux/mm/kasan/report.c:377 print_report+0xc4/0x620 linux/mm/kasan/report.c:488 kasan_report+0xda/0x110 linux/mm/kasan/report.c:601 kfree_skb_list_reason+0x47e/0x4c0 linux/net/core/skbuff.c:1183 skb_release_data+0x5af/0x880 linux/net/core/skbuff.c:1026 skb_release_all linux/net/core/skbuff.c:1094 __kfree_skb linux/net/core/skbuff.c:1108 kfree_skb_reason+0x12d/0x210 linux/net/core/skbuff.c:1144 kfree_skb linux/./include/linux/skbuff.h:1244 tipc_buf_append+0x425/0xb50 linux/net/tipc/msg.c:186 tipc_link_input+0x224/0x7c0 linux/net/tipc/link.c:1324 tipc_link_rcv+0x76e/0x2d70 linux/net/tipc/link.c:1824 tipc_rcv+0x45f/0x10f0 linux/net/tipc/node.c:2159 tipc_udp_recv+0x73b/0x8f0 linux/net/tipc/udp_media.c:390 udp_queue_rcv_one_skb+0xad2/0x1850 linux/net/ipv4/udp.c:2108 udp_queue_rcv_skb+0x131/0xb00 linux/net/ipv4/udp.c:2186 udp_unicast_rcv_skb+0x165/0x3b0 linux/net/ipv4/udp.c:2346 __udp4_lib_rcv+0x2594/0x3400 linux/net/ipv4/udp.c:2422 ip_protocol_deliver_rcu+0x30c/0x4e0 linux/net/ipv4/ip_input.c:205 ip_local_deliver_finish+0x2e4/0x520 linux/net/ipv4/ip_input.c:233 NF_HOOK linux/./include/linux/netfilter.h:314 NF_HOOK linux/./include/linux/netfilter.h:308 ip_local_deliver+0x18e/0x1f0 linux/net/ipv4/ip_input.c:254 dst_input linux/./include/net/dst.h:461 ip_rcv_finish linux/net/ipv4/ip_input.c:449 NF_HOOK linux/./include/linux/netfilter.h:314 NF_HOOK linux/./include/linux/netfilter.h:308 ip_rcv+0x2c5/0x5d0 linux/net/ipv4/ip_input.c:569 __netif_receive_skb_one_core+0x199/0x1e0 linux/net/core/dev.c:5534 __netif_receive_skb+0x1f/0x1c0 linux/net/core/dev.c:5648 process_backlog+0x101/0x6b0 linux/net/core/dev.c:5976 __napi_poll.constprop.0+0xba/0x550 linux/net/core/dev.c:6576 napi_poll linux/net/core/dev.c:6645 net_rx_action+0x95a/0xe90 linux/net/core/dev.c:6781 __do_softirq+0x21f/0x8e7 linux/kernel/softirq.c:553 do_softirq linux/kernel/softirq.c:454 do_softirq+0xb2/0xf0 linux/kernel/softirq.c:441 __local_bh_enable_ip+0x100/0x120 linux/kernel/softirq.c:381 local_bh_enable linux/./include/linux/bottom_half.h:33 rcu_read_unlock_bh linux/./include/linux/rcupdate.h:851 __dev_queue_xmit+0x871/0x3ee0 linux/net/core/dev.c:4378 dev_queue_xmit linux/./include/linux/netdevice.h:3169 neigh_hh_output linux/./include/net/neighbour.h:526 neigh_output linux/./include/net/neighbour.h:540 ip_finish_output2+0x169f/0x2550 linux/net/ipv4/ip_output.c:235 __ip_finish_output linux/net/ipv4/ip_output.c:313 __ip_finish_output+0x49e/0x950 linux/net/ipv4/ip_output.c:295 ip_finish_output+0x31/0x310 linux/net/ipv4/ip_output.c:323 NF_HOOK_COND linux/./include/linux/netfilter.h:303 ip_output+0x13b/0x2a0 linux/net/ipv4/ip_output.c:433 dst_output linux/./include/net/dst.h:451 ip_local_out linux/net/ipv4/ip_output.c:129 ip_send_skb+0x3e5/0x560 linux/net/ipv4/ip_output.c:1492 udp_send_skb+0x73f/0x1530 linux/net/ipv4/udp.c:963 udp_sendmsg+0x1a36/0x2b40 linux/net/ipv4/udp.c:1250 inet_sendmsg+0x105/0x140 linux/net/ipv4/af_inet.c:850 sock_sendmsg_nosec linux/net/socket.c:730 __sock_sendmsg linux/net/socket.c:745 __sys_sendto+0x42c/0x4e0 linux/net/socket.c:2191 __do_sys_sendto linux/net/socket.c:2203 __se_sys_sendto linux/net/socket.c:2199 __x64_sys_sendto+0xe0/0x1c0 linux/net/socket.c:2199 do_syscall_x64 linux/arch/x86/entry/common.c:52 do_syscall_64+0xd8/0x270 linux/arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x6f/0x77 linux/arch/x86/entry/entry_64.S:120 RIP: 0033:0x7f3434974f29 Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 37 8f 0d 00 f7 d8 64 89 01 48 RSP: 002b:00007fff9154f2b8 EFLAGS: 00000212 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f3434974f29 RDX: 00000000000032c8 RSI: 00007fff9154f300 RDI: 0000000000000003 RBP: 00007fff915532e0 R08: 00007fff91553360 R09: 0000000000000010 R10: 0000000000000000 R11: 0000000000000212 R12: 000055ed86d261d0 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 In the critical scenario, either the relevant skb is freed or its ownership is transferred into a frag_lists. In both cases, the cleanup code must not free it again: we need to clear the skb reference earlier. Fixes: 1149557d64c9 ("tipc: eliminate unnecessary linearization of incoming buffers") Cc: stable@vger.kernel.org Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-23852 Acked-by: Xin Long Signed-off-by: Paolo Abeni Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/752f1ccf762223d109845365d07f55414058e5a3.1714484273.git.pabeni@redhat.com Signed-off-by: Jakub Kicinski --- net/tipc/msg.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 5c9fd4791c4b..9a6e9bcbf694 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -156,6 +156,11 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) if (!head) goto err; + /* Either the input skb ownership is transferred to headskb + * or the input skb is freed, clear the reference to avoid + * bad access on error path. + */ + *buf = NULL; if (skb_try_coalesce(head, frag, &headstolen, &delta)) { kfree_skb_partial(frag, headstolen); } else { @@ -179,7 +184,6 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) *headbuf = NULL; return 1; } - *buf = NULL; return 0; err: kfree_skb(*buf); -- cgit v1.2.3 From 97bf6f81b29a8efaf5d0983251a7450e5794370d Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 30 Apr 2024 10:03:38 -0400 Subject: tipc: fix a possible memleak in tipc_buf_append __skb_linearize() doesn't free the skb when it fails, so move '*buf = NULL' after __skb_linearize(), so that the skb can be freed on the err path. Fixes: b7df21cf1b79 ("tipc: skb_linearize the head skb when reassembling msgs") Reported-by: Paolo Abeni Signed-off-by: Xin Long Reviewed-by: Simon Horman Reviewed-by: Tung Nguyen Link: https://lore.kernel.org/r/90710748c29a1521efac4f75ea01b3b7e61414cf.1714485818.git.lucien.xin@gmail.com Signed-off-by: Jakub Kicinski --- net/tipc/msg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 9a6e9bcbf694..76284fc538eb 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -142,9 +142,9 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) if (fragid == FIRST_FRAGMENT) { if (unlikely(head)) goto err; - *buf = NULL; if (skb_has_frag_list(frag) && __skb_linearize(frag)) goto err; + *buf = NULL; frag = skb_unshare(frag, GFP_ATOMIC); if (unlikely(!frag)) goto err; -- cgit v1.2.3 From 8a021a863a7dc9f4d3d00c5571b79e0961eeabc8 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 30 Apr 2024 09:48:16 -0700 Subject: net: dsa: Remove adjust_link paths Now that we no longer any drivers using PHYLIB's adjust_link callback, remove all paths that made use of adjust_link as well as the associated functions. Signed-off-by: Florian Fainelli Reviewed-by: Russell King (Oracle) Link: https://lore.kernel.org/r/20240430164816.2400606-3-florian.fainelli@broadcom.com Signed-off-by: Jakub Kicinski --- include/net/dsa.h | 7 --- net/dsa/dsa.c | 3 +- net/dsa/port.c | 139 +++++------------------------------------------------- 3 files changed, 12 insertions(+), 137 deletions(-) (limited to 'net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 678f1fd8b189..eef702dbea78 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -24,8 +24,6 @@ struct dsa_8021q_context; struct tc_action; -struct phy_device; -struct phylink_link_state; #define DSA_TAG_PROTO_NONE_VALUE 0 #define DSA_TAG_PROTO_BRCM_VALUE 1 @@ -867,11 +865,6 @@ struct dsa_switch_ops { int (*phy_write)(struct dsa_switch *ds, int port, int regnum, u16 val); - /* - * Link state adjustment (called from libphy) - */ - void (*adjust_link)(struct dsa_switch *ds, int port, - struct phy_device *phydev); /* * PHYLINK integration */ diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 2f347cd37316..12521a7d4048 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -1511,8 +1511,7 @@ static int dsa_switch_probe(struct dsa_switch *ds) ds->ops->phylink_mac_config || ds->ops->phylink_mac_finish || ds->ops->phylink_mac_link_down || - ds->ops->phylink_mac_link_up || - ds->ops->adjust_link) + ds->ops->phylink_mac_link_up) return -EINVAL; } diff --git a/net/dsa/port.c b/net/dsa/port.c index c6febc3d96d9..9a249d4ac3a5 100644 --- a/net/dsa/port.c +++ b/net/dsa/port.c @@ -1535,25 +1535,6 @@ void dsa_port_set_tag_protocol(struct dsa_port *cpu_dp, cpu_dp->tag_ops = tag_ops; } -static struct phy_device *dsa_port_get_phy_device(struct dsa_port *dp) -{ - struct device_node *phy_dn; - struct phy_device *phydev; - - phy_dn = of_parse_phandle(dp->dn, "phy-handle", 0); - if (!phy_dn) - return NULL; - - phydev = of_phy_find_device(phy_dn); - if (!phydev) { - of_node_put(phy_dn); - return ERR_PTR(-EPROBE_DEFER); - } - - of_node_put(phy_dn); - return phydev; -} - static struct phylink_pcs * dsa_port_phylink_mac_select_pcs(struct phylink_config *config, phy_interface_t interface) @@ -1616,17 +1597,10 @@ static void dsa_port_phylink_mac_link_down(struct phylink_config *config, phy_interface_t interface) { struct dsa_port *dp = dsa_phylink_to_port(config); - struct phy_device *phydev = NULL; struct dsa_switch *ds = dp->ds; - if (dsa_port_is_user(dp)) - phydev = dp->user->phydev; - - if (!ds->ops->phylink_mac_link_down) { - if (ds->ops->adjust_link && phydev) - ds->ops->adjust_link(ds, dp->index, phydev); + if (!ds->ops->phylink_mac_link_down) return; - } ds->ops->phylink_mac_link_down(ds, dp->index, mode, interface); } @@ -1641,11 +1615,8 @@ static void dsa_port_phylink_mac_link_up(struct phylink_config *config, struct dsa_port *dp = dsa_phylink_to_port(config); struct dsa_switch *ds = dp->ds; - if (!ds->ops->phylink_mac_link_up) { - if (ds->ops->adjust_link && phydev) - ds->ops->adjust_link(ds, dp->index, phydev); + if (!ds->ops->phylink_mac_link_up) return; - } ds->ops->phylink_mac_link_up(ds, dp->index, mode, interface, phydev, speed, duplex, tx_pause, rx_pause); @@ -1708,78 +1679,6 @@ void dsa_port_phylink_destroy(struct dsa_port *dp) dp->pl = NULL; } -static int dsa_shared_port_setup_phy_of(struct dsa_port *dp, bool enable) -{ - struct dsa_switch *ds = dp->ds; - struct phy_device *phydev; - int port = dp->index; - int err = 0; - - phydev = dsa_port_get_phy_device(dp); - if (!phydev) - return 0; - - if (IS_ERR(phydev)) - return PTR_ERR(phydev); - - if (enable) { - err = genphy_resume(phydev); - if (err < 0) - goto err_put_dev; - - err = genphy_read_status(phydev); - if (err < 0) - goto err_put_dev; - } else { - err = genphy_suspend(phydev); - if (err < 0) - goto err_put_dev; - } - - if (ds->ops->adjust_link) - ds->ops->adjust_link(ds, port, phydev); - - dev_dbg(ds->dev, "enabled port's phy: %s", phydev_name(phydev)); - -err_put_dev: - put_device(&phydev->mdio.dev); - return err; -} - -static int dsa_shared_port_fixed_link_register_of(struct dsa_port *dp) -{ - struct device_node *dn = dp->dn; - struct dsa_switch *ds = dp->ds; - struct phy_device *phydev; - int port = dp->index; - phy_interface_t mode; - int err; - - err = of_phy_register_fixed_link(dn); - if (err) { - dev_err(ds->dev, - "failed to register the fixed PHY of port %d\n", - port); - return err; - } - - phydev = of_phy_find_device(dn); - - err = of_get_phy_mode(dn, &mode); - if (err) - mode = PHY_INTERFACE_MODE_NA; - phydev->interface = mode; - - genphy_read_status(phydev); - - if (ds->ops->adjust_link) - ds->ops->adjust_link(ds, port, phydev); - - put_device(&phydev->mdio.dev); - - return 0; -} - static int dsa_shared_port_phylink_register(struct dsa_port *dp) { struct dsa_switch *ds = dp->ds; @@ -1983,44 +1882,28 @@ int dsa_shared_port_link_register_of(struct dsa_port *dp) dsa_switches_apply_workarounds)) return -EINVAL; - if (!ds->ops->adjust_link) { - if (missing_link_description) { - dev_warn(ds->dev, - "Skipping phylink registration for %s port %d\n", - dsa_port_is_cpu(dp) ? "CPU" : "DSA", dp->index); - } else { - dsa_shared_port_link_down(dp); + if (missing_link_description) { + dev_warn(ds->dev, + "Skipping phylink registration for %s port %d\n", + dsa_port_is_cpu(dp) ? "CPU" : "DSA", dp->index); + } else { + dsa_shared_port_link_down(dp); - return dsa_shared_port_phylink_register(dp); - } - return 0; + return dsa_shared_port_phylink_register(dp); } - dev_warn(ds->dev, - "Using legacy PHYLIB callbacks. Please migrate to PHYLINK!\n"); - - if (of_phy_is_fixed_link(dp->dn)) - return dsa_shared_port_fixed_link_register_of(dp); - else - return dsa_shared_port_setup_phy_of(dp, true); + return 0; } void dsa_shared_port_link_unregister_of(struct dsa_port *dp) { - struct dsa_switch *ds = dp->ds; - - if (!ds->ops->adjust_link && dp->pl) { + if (dp->pl) { rtnl_lock(); phylink_disconnect_phy(dp->pl); rtnl_unlock(); dsa_port_phylink_destroy(dp); return; } - - if (of_phy_is_fixed_link(dp->dn)) - of_phy_deregister_fixed_link(dp->dn); - else - dsa_shared_port_setup_phy_of(dp, false); } int dsa_port_hsr_join(struct dsa_port *dp, struct net_device *hsr, -- cgit v1.2.3 From fc1092f51567277509563800a3c56732070b6aa4 Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Tue, 30 Apr 2024 21:39:45 +0900 Subject: ipv4: Fix uninit-value access in __ip_make_skb() KMSAN reported uninit-value access in __ip_make_skb() [1]. __ip_make_skb() tests HDRINCL to know if the skb has icmphdr. However, HDRINCL can cause a race condition. If calling setsockopt(2) with IP_HDRINCL changes HDRINCL while __ip_make_skb() is running, the function will access icmphdr in the skb even if it is not included. This causes the issue reported by KMSAN. Check FLOWI_FLAG_KNOWN_NH on fl4->flowi4_flags instead of testing HDRINCL on the socket. Also, fl4->fl4_icmp_type and fl4->fl4_icmp_code are not initialized. These are union in struct flowi4 and are implicitly initialized by flowi4_init_output(), but we should not rely on specific union layout. Initialize these explicitly in raw_sendmsg(). [1] BUG: KMSAN: uninit-value in __ip_make_skb+0x2b74/0x2d20 net/ipv4/ip_output.c:1481 __ip_make_skb+0x2b74/0x2d20 net/ipv4/ip_output.c:1481 ip_finish_skb include/net/ip.h:243 [inline] ip_push_pending_frames+0x4c/0x5c0 net/ipv4/ip_output.c:1508 raw_sendmsg+0x2381/0x2690 net/ipv4/raw.c:654 inet_sendmsg+0x27b/0x2a0 net/ipv4/af_inet.c:851 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x274/0x3c0 net/socket.c:745 __sys_sendto+0x62c/0x7b0 net/socket.c:2191 __do_sys_sendto net/socket.c:2203 [inline] __se_sys_sendto net/socket.c:2199 [inline] __x64_sys_sendto+0x130/0x200 net/socket.c:2199 do_syscall_64+0xd8/0x1f0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x6d/0x75 Uninit was created at: slab_post_alloc_hook mm/slub.c:3804 [inline] slab_alloc_node mm/slub.c:3845 [inline] kmem_cache_alloc_node+0x5f6/0xc50 mm/slub.c:3888 kmalloc_reserve+0x13c/0x4a0 net/core/skbuff.c:577 __alloc_skb+0x35a/0x7c0 net/core/skbuff.c:668 alloc_skb include/linux/skbuff.h:1318 [inline] __ip_append_data+0x49ab/0x68c0 net/ipv4/ip_output.c:1128 ip_append_data+0x1e7/0x260 net/ipv4/ip_output.c:1365 raw_sendmsg+0x22b1/0x2690 net/ipv4/raw.c:648 inet_sendmsg+0x27b/0x2a0 net/ipv4/af_inet.c:851 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x274/0x3c0 net/socket.c:745 __sys_sendto+0x62c/0x7b0 net/socket.c:2191 __do_sys_sendto net/socket.c:2203 [inline] __se_sys_sendto net/socket.c:2199 [inline] __x64_sys_sendto+0x130/0x200 net/socket.c:2199 do_syscall_64+0xd8/0x1f0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x6d/0x75 CPU: 1 PID: 15709 Comm: syz-executor.7 Not tainted 6.8.0-11567-gb3603fcb79b1 #25 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-1.fc39 04/01/2014 Fixes: 99e5acae193e ("ipv4: Fix potential uninit variable access bug in __ip_make_skb()") Reported-by: syzkaller Signed-off-by: Shigeru Yoshida Link: https://lore.kernel.org/r/20240430123945.2057348-1-syoshida@redhat.com Signed-off-by: Paolo Abeni --- net/ipv4/ip_output.c | 2 +- net/ipv4/raw.c | 3 +++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 1fe794967211..39229fd0601a 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1473,7 +1473,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk, * by icmp_hdr(skb)->type. */ if (sk->sk_type == SOCK_RAW && - !inet_test_bit(HDRINCL, sk)) + !(fl4->flowi4_flags & FLOWI_FLAG_KNOWN_NH)) icmp_type = fl4->fl4_icmp_type; else icmp_type = icmp_hdr(skb)->type; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index dcb11f22cbf2..4cb43401e0e0 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -612,6 +612,9 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) (hdrincl ? FLOWI_FLAG_KNOWN_NH : 0), daddr, saddr, 0, 0, sk->sk_uid); + fl4.fl4_icmp_type = 0; + fl4.fl4_icmp_code = 0; + if (!hdrincl) { rfv.msg = msg; rfv.hlen = 0; -- cgit v1.2.3 From 5ef31ea5d053a8f493a772ebad3f3ce82c35d845 Mon Sep 17 00:00:00 2001 From: Richard Gobert Date: Tue, 30 Apr 2024 16:35:54 +0200 Subject: net: gro: fix udp bad offset in socket lookup by adding {inner_}network_offset to napi_gro_cb Commits a602456 ("udp: Add GRO functions to UDP socket") and 57c67ff ("udp: additional GRO support") introduce incorrect usage of {ip,ipv6}_hdr in the complete phase of gro. The functions always return skb->network_header, which in the case of encapsulated packets at the gro complete phase, is always set to the innermost L3 of the packet. That means that calling {ip,ipv6}_hdr for skbs which completed the GRO receive phase (both in gro_list and *_gro_complete) when parsing an encapsulated packet's _outer_ L3/L4 may return an unexpected value. This incorrect usage leads to a bug in GRO's UDP socket lookup. udp{4,6}_lib_lookup_skb functions use ip_hdr/ipv6_hdr respectively. These *_hdr functions return network_header which will point to the innermost L3, resulting in the wrong offset being used in __udp{4,6}_lib_lookup with encapsulated packets. This patch adds network_offset and inner_network_offset to napi_gro_cb, and makes sure both are set correctly. To fix the issue, network_offsets union is used inside napi_gro_cb, in which both the outer and the inner network offsets are saved. Reproduction example: Endpoint configuration example (fou + local address bind) # ip fou add port 6666 ipproto 4 # ip link add name tun1 type ipip remote 2.2.2.1 local 2.2.2.2 encap fou encap-dport 5555 encap-sport 6666 mode ipip # ip link set tun1 up # ip a add 1.1.1.2/24 dev tun1 Netperf TCP_STREAM result on net-next before patch is applied: net-next main, GRO enabled: $ netperf -H 1.1.1.2 -t TCP_STREAM -l 5 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 131072 16384 16384 5.28 2.37 net-next main, GRO disabled: $ netperf -H 1.1.1.2 -t TCP_STREAM -l 5 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 131072 16384 16384 5.01 2745.06 patch applied, GRO enabled: $ netperf -H 1.1.1.2 -t TCP_STREAM -l 5 Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 131072 16384 16384 5.01 2877.38 Fixes: a6024562ffd7 ("udp: Add GRO functions to UDP socket") Signed-off-by: Richard Gobert Reviewed-by: Eric Dumazet Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- include/net/gro.h | 9 +++++++++ net/8021q/vlan_core.c | 2 ++ net/core/gro.c | 1 + net/ipv4/af_inet.c | 1 + net/ipv4/udp.c | 3 ++- net/ipv4/udp_offload.c | 3 ++- net/ipv6/ip6_offload.c | 1 + net/ipv6/udp.c | 3 ++- net/ipv6/udp_offload.c | 3 ++- 9 files changed, 22 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/gro.h b/include/net/gro.h index 50f1e403dbbb..c1d4ca0463a1 100644 --- a/include/net/gro.h +++ b/include/net/gro.h @@ -87,6 +87,15 @@ struct napi_gro_cb { /* used to support CHECKSUM_COMPLETE for tunneling protocols */ __wsum csum; + + /* L3 offsets */ + union { + struct { + u16 network_offset; + u16 inner_network_offset; + }; + u16 network_offsets[2]; + }; }; #define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb) diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index f00158234505..9404dd551dfd 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -478,6 +478,8 @@ static struct sk_buff *vlan_gro_receive(struct list_head *head, if (unlikely(!vhdr)) goto out; + NAPI_GRO_CB(skb)->network_offsets[NAPI_GRO_CB(skb)->encap_mark] = hlen; + type = vhdr->h_vlan_encapsulated_proto; ptype = gro_find_receive_by_type(type); diff --git a/net/core/gro.c b/net/core/gro.c index 83f35d99a682..c7901253a1a8 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -371,6 +371,7 @@ static inline void skb_gro_reset_offset(struct sk_buff *skb, u32 nhoff) const skb_frag_t *frag0; unsigned int headlen; + NAPI_GRO_CB(skb)->network_offset = 0; NAPI_GRO_CB(skb)->data_offset = 0; headlen = skb_headlen(skb); NAPI_GRO_CB(skb)->frag0 = skb->data; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 55bd72997b31..fafb123f798b 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1572,6 +1572,7 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) /* The above will be needed by the transport layer if there is one * immediately following this IP hdr. */ + NAPI_GRO_CB(skb)->inner_network_offset = off; /* Note : No need to call skb_gro_postpull_rcsum() here, * as we already checked checksum over ipv4 header was 0 diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 420905be5f30..b32cf2eeeb41 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -532,7 +532,8 @@ static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb, struct sock *udp4_lib_lookup_skb(const struct sk_buff *skb, __be16 sport, __be16 dport) { - const struct iphdr *iph = ip_hdr(skb); + const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation]; + const struct iphdr *iph = (struct iphdr *)(skb->data + offset); struct net *net = dev_net(skb->dev); int iif, sdif; diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 3498dd1d0694..fd29d21d579c 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -718,7 +718,8 @@ EXPORT_SYMBOL(udp_gro_complete); INDIRECT_CALLABLE_SCOPE int udp4_gro_complete(struct sk_buff *skb, int nhoff) { - const struct iphdr *iph = ip_hdr(skb); + const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation]; + const struct iphdr *iph = (struct iphdr *)(skb->data + offset); struct udphdr *uh = (struct udphdr *)(skb->data + nhoff); /* do fraglist only if there is no outer UDP encap (or we already processed it) */ diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c index b41e35af69ea..c8b909a9904f 100644 --- a/net/ipv6/ip6_offload.c +++ b/net/ipv6/ip6_offload.c @@ -237,6 +237,7 @@ INDIRECT_CALLABLE_SCOPE struct sk_buff *ipv6_gro_receive(struct list_head *head, goto out; skb_set_network_header(skb, off); + NAPI_GRO_CB(skb)->inner_network_offset = off; flush += ntohs(iph->payload_len) != skb->len - hlen; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 1a4cccdd40c9..8f7aa8bac1e7 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -272,7 +272,8 @@ static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb, struct sock *udp6_lib_lookup_skb(const struct sk_buff *skb, __be16 sport, __be16 dport) { - const struct ipv6hdr *iph = ipv6_hdr(skb); + const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation]; + const struct ipv6hdr *iph = (struct ipv6hdr *)(skb->data + offset); struct net *net = dev_net(skb->dev); int iif, sdif; diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c index bbd347de00b4..b41152dd4246 100644 --- a/net/ipv6/udp_offload.c +++ b/net/ipv6/udp_offload.c @@ -164,7 +164,8 @@ flush: INDIRECT_CALLABLE_SCOPE int udp6_gro_complete(struct sk_buff *skb, int nhoff) { - const struct ipv6hdr *ipv6h = ipv6_hdr(skb); + const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation]; + const struct ipv6hdr *ipv6h = (struct ipv6hdr *)(skb->data + offset); struct udphdr *uh = (struct udphdr *)(skb->data + nhoff); /* do fraglist only if there is no outer UDP encap (or we already processed it) */ -- cgit v1.2.3 From 5babae777c61aa8a8679d59d3cdc54165ad96d42 Mon Sep 17 00:00:00 2001 From: Richard Gobert Date: Tue, 30 Apr 2024 16:35:55 +0200 Subject: net: gro: add flush check in udp_gro_receive_segment GRO-GSO path is supposed to be transparent and as such L3 flush checks are relevant to all UDP flows merging in GRO. This patch uses the same logic and code from tcp_gro_receive, terminating merge if flush is non zero. Fixes: e20cf8d3f1f7 ("udp: implement GRO for plain UDP sockets.") Signed-off-by: Richard Gobert Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- net/ipv4/udp_offload.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index fd29d21d579c..8721fe5beca2 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -471,6 +471,7 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head, struct sk_buff *p; unsigned int ulen; int ret = 0; + int flush; /* requires non zero csum, for symmetry with GSO */ if (!uh->check) { @@ -504,13 +505,22 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head, return p; } + flush = NAPI_GRO_CB(p)->flush; + + if (NAPI_GRO_CB(p)->flush_id != 1 || + NAPI_GRO_CB(p)->count != 1 || + !NAPI_GRO_CB(p)->is_atomic) + flush |= NAPI_GRO_CB(p)->flush_id; + else + NAPI_GRO_CB(p)->is_atomic = false; + /* Terminate the flow on len mismatch or if it grow "too much". * Under small packet flood GRO count could elsewhere grow a lot * leading to excessive truesize values. * On len mismatch merge the first packet shorter than gso_size, * otherwise complete the GRO packet. */ - if (ulen > ntohs(uh2->len)) { + if (ulen > ntohs(uh2->len) || flush) { pp = p; } else { if (NAPI_GRO_CB(skb)->is_flist) { -- cgit v1.2.3 From 86735b57c905e775f05de995df35379366b72168 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Tue, 30 Apr 2024 19:11:13 +0200 Subject: net/sched: unregister lockdep keys in qdisc_create/qdisc_alloc error path Naresh and Eric report several errors (corrupted elements in the dynamic key hash list), when running tdc.py or syzbot. The error path of qdisc_alloc() and qdisc_create() frees the qdisc memory, but it forgets to unregister the lockdep key, thus causing use-after-free like the following one: ================================================================== BUG: KASAN: slab-use-after-free in lockdep_register_key+0x5f2/0x700 Read of size 8 at addr ffff88811236f2a8 by task ip/7925 CPU: 26 PID: 7925 Comm: ip Kdump: loaded Not tainted 6.9.0-rc2+ #648 Hardware name: Supermicro SYS-6027R-72RF/X9DRH-7TF/7F/iTF/iF, BIOS 3.0 07/26/2013 Call Trace: dump_stack_lvl+0x7c/0xc0 print_report+0xc9/0x610 kasan_report+0x89/0xc0 lockdep_register_key+0x5f2/0x700 qdisc_alloc+0x21d/0xb60 qdisc_create_dflt+0x63/0x3c0 attach_one_default_qdisc.constprop.37+0x8e/0x170 dev_activate+0x4bd/0xc30 __dev_open+0x275/0x380 __dev_change_flags+0x3f1/0x570 dev_change_flags+0x7c/0x160 do_setlink+0x1ea1/0x34b0 __rtnl_newlink+0x8c9/0x1510 rtnl_newlink+0x61/0x90 rtnetlink_rcv_msg+0x2f0/0xbc0 netlink_rcv_skb+0x120/0x380 netlink_unicast+0x420/0x630 netlink_sendmsg+0x732/0xbc0 __sock_sendmsg+0x1ea/0x280 ____sys_sendmsg+0x5a9/0x990 ___sys_sendmsg+0xf1/0x180 __sys_sendmsg+0xd3/0x180 do_syscall_64+0x96/0x180 entry_SYSCALL_64_after_hwframe+0x71/0x79 RIP: 0033:0x7f9503f4fa07 Code: 0a 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10 RSP: 002b:00007fff6c729068 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 000000006630c681 RCX: 00007f9503f4fa07 RDX: 0000000000000000 RSI: 00007fff6c7290d0 RDI: 0000000000000003 RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000078 R10: 000000000000009b R11: 0000000000000246 R12: 0000000000000001 R13: 00007fff6c729180 R14: 0000000000000000 R15: 000055bf67dd9040 Allocated by task 7745: kasan_save_stack+0x1c/0x40 kasan_save_track+0x10/0x30 __kasan_kmalloc+0x7b/0x90 __kmalloc_node+0x1ff/0x460 qdisc_alloc+0xae/0xb60 qdisc_create+0xdd/0xfb0 tc_modify_qdisc+0x37e/0x1960 rtnetlink_rcv_msg+0x2f0/0xbc0 netlink_rcv_skb+0x120/0x380 netlink_unicast+0x420/0x630 netlink_sendmsg+0x732/0xbc0 __sock_sendmsg+0x1ea/0x280 ____sys_sendmsg+0x5a9/0x990 ___sys_sendmsg+0xf1/0x180 __sys_sendmsg+0xd3/0x180 do_syscall_64+0x96/0x180 entry_SYSCALL_64_after_hwframe+0x71/0x79 Freed by task 7745: kasan_save_stack+0x1c/0x40 kasan_save_track+0x10/0x30 kasan_save_free_info+0x36/0x60 __kasan_slab_free+0xfe/0x180 kfree+0x113/0x380 qdisc_create+0xafb/0xfb0 tc_modify_qdisc+0x37e/0x1960 rtnetlink_rcv_msg+0x2f0/0xbc0 netlink_rcv_skb+0x120/0x380 netlink_unicast+0x420/0x630 netlink_sendmsg+0x732/0xbc0 __sock_sendmsg+0x1ea/0x280 ____sys_sendmsg+0x5a9/0x990 ___sys_sendmsg+0xf1/0x180 __sys_sendmsg+0xd3/0x180 do_syscall_64+0x96/0x180 entry_SYSCALL_64_after_hwframe+0x71/0x79 Fix this ensuring that lockdep_unregister_key() is called before the qdisc struct is freed, also in the error path of qdisc_create() and qdisc_alloc(). Fixes: af0cb3fa3f9e ("net/sched: fix false lockdep warning on qdisc root lock") Reported-by: Linux Kernel Functional Testing Closes: https://lore.kernel.org/netdev/20240429221706.1492418-1-naresh.kamboju@linaro.org/ Signed-off-by: Davide Caratti Reviewed-by: Eric Dumazet Reviewed-by: Ido Schimmel Tested-by: Naresh Kamboju Tested-by: Ido Schimmel Link: https://lore.kernel.org/r/2aa1ca0c0a3aa0acc15925c666c777a4b5de553c.1714496886.git.dcaratti@redhat.com Signed-off-by: Jakub Kicinski --- net/sched/sch_api.c | 1 + net/sched/sch_generic.c | 1 + 2 files changed, 2 insertions(+) (limited to 'net') diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 60239378d43f..6292d6d73b72 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1389,6 +1389,7 @@ err_out4: ops->destroy(sch); qdisc_put_stab(rtnl_dereference(sch->stab)); err_out3: + lockdep_unregister_key(&sch->root_lock_key); netdev_put(dev, &sch->dev_tracker); qdisc_free(sch); err_out2: diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 31dfd6c7405b..d3f6006b563c 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -982,6 +982,7 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, return sch; errout1: + lockdep_unregister_key(&sch->root_lock_key); kfree(sch); errout: return ERR_PTR(err); -- cgit v1.2.3 From 57bfc7605ca5b102ba336779ae9adbc5bbba1d96 Mon Sep 17 00:00:00 2001 From: Miao Xu Date: Wed, 1 May 2024 21:23:16 -0700 Subject: tcp: Add new args for cong_control in tcp_congestion_ops This patch adds two new arguments for cong_control of struct tcp_congestion_ops: - ack - flag These two arguments are inherited from the caller tcp_cong_control in tcp_intput.c. One use case of them is to update cwnd and pacing rate inside cong_control based on the info they provide. For example, the flag can be used to decide if it is the right time to raise or reduce a sender's cwnd. Reviewed-by: Eric Dumazet Signed-off-by: Miao Xu Link: https://lore.kernel.org/r/20240502042318.801932-2-miaxu@meta.com Signed-off-by: Martin KaFai Lau --- include/net/tcp.h | 2 +- net/ipv4/bpf_tcp_ca.c | 3 ++- net/ipv4/tcp_bbr.c | 2 +- net/ipv4/tcp_input.c | 2 +- tools/testing/selftests/bpf/progs/tcp_ca_kfunc.c | 6 +++--- 5 files changed, 8 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index fe98fb01879b..7294da8fb780 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1172,7 +1172,7 @@ struct tcp_congestion_ops { /* call when packets are delivered to update cwnd and pacing rate, * after all the ca_state processing. (optional) */ - void (*cong_control)(struct sock *sk, const struct rate_sample *rs); + void (*cong_control)(struct sock *sk, u32 ack, int flag, const struct rate_sample *rs); /* new value of cwnd after loss (required) */ diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index 7f518ea5f4ac..6bd7f8db189a 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -307,7 +307,8 @@ static u32 bpf_tcp_ca_min_tso_segs(struct sock *sk) return 0; } -static void bpf_tcp_ca_cong_control(struct sock *sk, const struct rate_sample *rs) +static void bpf_tcp_ca_cong_control(struct sock *sk, u32 ack, int flag, + const struct rate_sample *rs) { } diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp_bbr.c index 7e52ab24e40a..760941e55153 100644 --- a/net/ipv4/tcp_bbr.c +++ b/net/ipv4/tcp_bbr.c @@ -1024,7 +1024,7 @@ static void bbr_update_model(struct sock *sk, const struct rate_sample *rs) bbr_update_gains(sk); } -__bpf_kfunc static void bbr_main(struct sock *sk, const struct rate_sample *rs) +__bpf_kfunc static void bbr_main(struct sock *sk, u32 ack, int flag, const struct rate_sample *rs) { struct bbr *bbr = inet_csk_ca(sk); u32 bw; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 53e1150f706f..23ccfc7b1d3c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3541,7 +3541,7 @@ static void tcp_cong_control(struct sock *sk, u32 ack, u32 acked_sacked, const struct inet_connection_sock *icsk = inet_csk(sk); if (icsk->icsk_ca_ops->cong_control) { - icsk->icsk_ca_ops->cong_control(sk, rs); + icsk->icsk_ca_ops->cong_control(sk, ack, flag, rs); return; } diff --git a/tools/testing/selftests/bpf/progs/tcp_ca_kfunc.c b/tools/testing/selftests/bpf/progs/tcp_ca_kfunc.c index fcfbfe0336b4..52b610357309 100644 --- a/tools/testing/selftests/bpf/progs/tcp_ca_kfunc.c +++ b/tools/testing/selftests/bpf/progs/tcp_ca_kfunc.c @@ -5,7 +5,7 @@ #include extern void bbr_init(struct sock *sk) __ksym; -extern void bbr_main(struct sock *sk, const struct rate_sample *rs) __ksym; +extern void bbr_main(struct sock *sk, u32 ack, int flag, const struct rate_sample *rs) __ksym; extern u32 bbr_sndbuf_expand(struct sock *sk) __ksym; extern u32 bbr_undo_cwnd(struct sock *sk) __ksym; extern void bbr_cwnd_event(struct sock *sk, enum tcp_ca_event event) __ksym; @@ -42,9 +42,9 @@ void BPF_PROG(in_ack_event, struct sock *sk, u32 flags) } SEC("struct_ops/cong_control") -void BPF_PROG(cong_control, struct sock *sk, const struct rate_sample *rs) +void BPF_PROG(cong_control, struct sock *sk, u32 ack, int flag, const struct rate_sample *rs) { - bbr_main(sk, rs); + bbr_main(sk, ack, flag, rs); } SEC("struct_ops/cong_avoid") -- cgit v1.2.3 From 0325cbd21e3c26eff88bc7da303ffb46b4f5d294 Mon Sep 17 00:00:00 2001 From: Miao Xu Date: Wed, 1 May 2024 21:23:17 -0700 Subject: bpf: tcp: Allow to write tp->snd_cwnd_stamp in bpf_tcp_ca This patch allows the write of tp->snd_cwnd_stamp in a bpf tcp ca program. An use case of writing this field is to keep track of the time whenever tp->snd_cwnd is raised or reduced inside the `cong_control` callback. Reviewed-by: Eric Dumazet Signed-off-by: Miao Xu Link: https://lore.kernel.org/r/20240502042318.801932-3-miaxu@meta.com Signed-off-by: Martin KaFai Lau --- net/ipv4/bpf_tcp_ca.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/ipv4/bpf_tcp_ca.c b/net/ipv4/bpf_tcp_ca.c index 6bd7f8db189a..18227757ec0c 100644 --- a/net/ipv4/bpf_tcp_ca.c +++ b/net/ipv4/bpf_tcp_ca.c @@ -107,6 +107,9 @@ static int bpf_tcp_ca_btf_struct_access(struct bpf_verifier_log *log, case offsetof(struct tcp_sock, snd_cwnd_cnt): end = offsetofend(struct tcp_sock, snd_cwnd_cnt); break; + case offsetof(struct tcp_sock, snd_cwnd_stamp): + end = offsetofend(struct tcp_sock, snd_cwnd_stamp); + break; case offsetof(struct tcp_sock, snd_ssthresh): end = offsetofend(struct tcp_sock, snd_ssthresh); break; -- cgit v1.2.3 From a17ef9e6c2c1cf0fc6cd6ca6a9ce525c67d1da7f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 30 Apr 2024 18:00:15 +0000 Subject: net_sched: sch_sfq: annotate data-races around q->perturb_period sfq_perturbation() reads q->perturb_period locklessly. Add annotations to fix potential issues. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240430180015.3111398-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/sched/sch_sfq.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index e66f4afb920d..3b9245a3c767 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -608,6 +608,7 @@ static void sfq_perturbation(struct timer_list *t) struct Qdisc *sch = q->sch; spinlock_t *root_lock; siphash_key_t nkey; + int period; get_random_bytes(&nkey, sizeof(nkey)); rcu_read_lock(); @@ -618,8 +619,12 @@ static void sfq_perturbation(struct timer_list *t) sfq_rehash(sch); spin_unlock(root_lock); - if (q->perturb_period) - mod_timer(&q->perturb_timer, jiffies + q->perturb_period); + /* q->perturb_period can change under us from + * sfq_change() and sfq_destroy(). + */ + period = READ_ONCE(q->perturb_period); + if (period) + mod_timer(&q->perturb_timer, jiffies + period); rcu_read_unlock(); } @@ -662,7 +667,7 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt) q->quantum = ctl->quantum; q->scaled_quantum = SFQ_ALLOT_SIZE(q->quantum); } - q->perturb_period = ctl->perturb_period * HZ; + WRITE_ONCE(q->perturb_period, ctl->perturb_period * HZ); if (ctl->flows) q->maxflows = min_t(u32, ctl->flows, SFQ_MAX_FLOWS); if (ctl->divisor) { @@ -724,7 +729,7 @@ static void sfq_destroy(struct Qdisc *sch) struct sfq_sched_data *q = qdisc_priv(sch); tcf_block_put(q->block); - q->perturb_period = 0; + WRITE_ONCE(q->perturb_period, 0); del_timer_sync(&q->perturb_timer); sfq_free(q->ht); sfq_free(q->slots); -- cgit v1.2.3 From 94062790aedb505bdda209b10bea47b294d6394f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 1 May 2024 12:54:48 +0000 Subject: tcp: defer shutdown(SEND_SHUTDOWN) for TCP_SYN_RECV sockets TCP_SYN_RECV state is really special, it is only used by cross-syn connections, mostly used by fuzzers. In the following crash [1], syzbot managed to trigger a divide by zero in tcp_rcv_space_adjust() A socket makes the following state transitions, without ever calling tcp_init_transfer(), meaning tcp_init_buffer_space() is also not called. TCP_CLOSE connect() TCP_SYN_SENT TCP_SYN_RECV shutdown() -> tcp_shutdown(sk, SEND_SHUTDOWN) TCP_FIN_WAIT1 To fix this issue, change tcp_shutdown() to not perform a TCP_SYN_RECV -> TCP_FIN_WAIT1 transition, which makes no sense anyway. When tcp_rcv_state_process() later changes socket state from TCP_SYN_RECV to TCP_ESTABLISH, then look at sk->sk_shutdown to finally enter TCP_FIN_WAIT1 state, and send a FIN packet from a sane socket state. This means tcp_send_fin() can now be called from BH context, and must use GFP_ATOMIC allocations. [1] divide error: 0000 [#1] PREEMPT SMP KASAN NOPTI CPU: 1 PID: 5084 Comm: syz-executor358 Not tainted 6.9.0-rc6-syzkaller-00022-g98369dccd2f8 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 RIP: 0010:tcp_rcv_space_adjust+0x2df/0x890 net/ipv4/tcp_input.c:767 Code: e3 04 4c 01 eb 48 8b 44 24 38 0f b6 04 10 84 c0 49 89 d5 0f 85 a5 03 00 00 41 8b 8e c8 09 00 00 89 e8 29 c8 48 0f af c3 31 d2 <48> f7 f1 48 8d 1c 43 49 8d 96 76 08 00 00 48 89 d0 48 c1 e8 03 48 RSP: 0018:ffffc900031ef3f0 EFLAGS: 00010246 RAX: 0c677a10441f8f42 RBX: 000000004fb95e7e RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: 0000000027d4b11f R08: ffffffff89e535a4 R09: 1ffffffff25e6ab7 R10: dffffc0000000000 R11: ffffffff8135e920 R12: ffff88802a9f8d30 R13: dffffc0000000000 R14: ffff88802a9f8d00 R15: 1ffff1100553f2da FS: 00005555775c0380(0000) GS:ffff8880b9500000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f1155bf2304 CR3: 000000002b9f2000 CR4: 0000000000350ef0 Call Trace: tcp_recvmsg_locked+0x106d/0x25a0 net/ipv4/tcp.c:2513 tcp_recvmsg+0x25d/0x920 net/ipv4/tcp.c:2578 inet6_recvmsg+0x16a/0x730 net/ipv6/af_inet6.c:680 sock_recvmsg_nosec net/socket.c:1046 [inline] sock_recvmsg+0x109/0x280 net/socket.c:1068 ____sys_recvmsg+0x1db/0x470 net/socket.c:2803 ___sys_recvmsg net/socket.c:2845 [inline] do_recvmmsg+0x474/0xae0 net/socket.c:2939 __sys_recvmmsg net/socket.c:3018 [inline] __do_sys_recvmmsg net/socket.c:3041 [inline] __se_sys_recvmmsg net/socket.c:3034 [inline] __x64_sys_recvmmsg+0x199/0x250 net/socket.c:3034 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7faeb6363db9 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 c1 17 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffcc1997168 EFLAGS: 00000246 ORIG_RAX: 000000000000012b RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007faeb6363db9 RDX: 0000000000000001 RSI: 0000000020000bc0 RDI: 0000000000000005 RBP: 0000000000000000 R08: 0000000000000000 R09: 000000000000001c R10: 0000000000000122 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000001 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Eric Dumazet Acked-by: Neal Cardwell Link: https://lore.kernel.org/r/20240501125448.896529-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 4 ++-- net/ipv4/tcp_input.c | 2 ++ net/ipv4/tcp_output.c | 4 +++- 3 files changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index e767721b3a58..66d77faca64f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2710,7 +2710,7 @@ void tcp_shutdown(struct sock *sk, int how) /* If we've already sent a FIN, or it's a closed state, skip this. */ if ((1 << sk->sk_state) & (TCPF_ESTABLISHED | TCPF_SYN_SENT | - TCPF_SYN_RECV | TCPF_CLOSE_WAIT)) { + TCPF_CLOSE_WAIT)) { /* Clear out any half completed packets. FIN if needed. */ if (tcp_close_state(sk)) tcp_send_fin(sk); @@ -2819,7 +2819,7 @@ void __tcp_close(struct sock *sk, long timeout) * machine. State transitions: * * TCP_ESTABLISHED -> TCP_FIN_WAIT1 - * TCP_SYN_RECV -> TCP_FIN_WAIT1 (forget it, it's impossible) + * TCP_SYN_RECV -> TCP_FIN_WAIT1 (it is difficult) * TCP_CLOSE_WAIT -> TCP_LAST_ACK * * are legal only when FIN has been sent (i.e. in window), diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5d874817a78d..a140d9f7a0a3 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6761,6 +6761,8 @@ tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) tcp_initialize_rcv_mss(sk); tcp_fast_path_on(tp); + if (sk->sk_shutdown & SEND_SHUTDOWN) + tcp_shutdown(sk, SEND_SHUTDOWN); break; case TCP_FIN_WAIT1: { diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index e3167ad96567..02caeb7bcf63 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3563,7 +3563,9 @@ void tcp_send_fin(struct sock *sk) return; } } else { - skb = alloc_skb_fclone(MAX_TCP_HEADER, sk->sk_allocation); + skb = alloc_skb_fclone(MAX_TCP_HEADER, + sk_gfp_mask(sk, GFP_ATOMIC | + __GFP_NOWARN)); if (unlikely(!skb)) return; -- cgit v1.2.3 From f2db7230f73a80dbb179deab78f88a7947f0ab7e Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 1 May 2024 14:31:45 -0700 Subject: tcp: Use refcount_inc_not_zero() in tcp_twsk_unique(). Anderson Nascimento reported a use-after-free splat in tcp_twsk_unique() with nice analysis. Since commit ec94c2696f0b ("tcp/dccp: avoid one atomic operation for timewait hashdance"), inet_twsk_hashdance() sets TIME-WAIT socket's sk_refcnt after putting it into ehash and releasing the bucket lock. Thus, there is a small race window where other threads could try to reuse the port during connect() and call sock_hold() in tcp_twsk_unique() for the TIME-WAIT socket with zero refcnt. If that happens, the refcnt taken by tcp_twsk_unique() is overwritten and sock_put() will cause underflow, triggering a real use-after-free somewhere else. To avoid the use-after-free, we need to use refcount_inc_not_zero() in tcp_twsk_unique() and give up on reusing the port if it returns false. [0]: refcount_t: addition on 0; use-after-free. WARNING: CPU: 0 PID: 1039313 at lib/refcount.c:25 refcount_warn_saturate+0xe5/0x110 CPU: 0 PID: 1039313 Comm: trigger Not tainted 6.8.6-200.fc39.x86_64 #1 Hardware name: VMware, Inc. VMware20,1/440BX Desktop Reference Platform, BIOS VMW201.00V.21805430.B64.2305221830 05/22/2023 RIP: 0010:refcount_warn_saturate+0xe5/0x110 Code: 42 8e ff 0f 0b c3 cc cc cc cc 80 3d aa 13 ea 01 00 0f 85 5e ff ff ff 48 c7 c7 f8 8e b7 82 c6 05 96 13 ea 01 01 e8 7b 42 8e ff <0f> 0b c3 cc cc cc cc 48 c7 c7 50 8f b7 82 c6 05 7a 13 ea 01 01 e8 RSP: 0018:ffffc90006b43b60 EFLAGS: 00010282 RAX: 0000000000000000 RBX: ffff888009bb3ef0 RCX: 0000000000000027 RDX: ffff88807be218c8 RSI: 0000000000000001 RDI: ffff88807be218c0 RBP: 0000000000069d70 R08: 0000000000000000 R09: ffffc90006b439f0 R10: ffffc90006b439e8 R11: 0000000000000003 R12: ffff8880029ede84 R13: 0000000000004e20 R14: ffffffff84356dc0 R15: ffff888009bb3ef0 FS: 00007f62c10926c0(0000) GS:ffff88807be00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020ccb000 CR3: 000000004628c005 CR4: 0000000000f70ef0 PKRU: 55555554 Call Trace: ? refcount_warn_saturate+0xe5/0x110 ? __warn+0x81/0x130 ? refcount_warn_saturate+0xe5/0x110 ? report_bug+0x171/0x1a0 ? refcount_warn_saturate+0xe5/0x110 ? handle_bug+0x3c/0x80 ? exc_invalid_op+0x17/0x70 ? asm_exc_invalid_op+0x1a/0x20 ? refcount_warn_saturate+0xe5/0x110 tcp_twsk_unique+0x186/0x190 __inet_check_established+0x176/0x2d0 __inet_hash_connect+0x74/0x7d0 ? __pfx___inet_check_established+0x10/0x10 tcp_v4_connect+0x278/0x530 __inet_stream_connect+0x10f/0x3d0 inet_stream_connect+0x3a/0x60 __sys_connect+0xa8/0xd0 __x64_sys_connect+0x18/0x20 do_syscall_64+0x83/0x170 entry_SYSCALL_64_after_hwframe+0x78/0x80 RIP: 0033:0x7f62c11a885d Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a3 45 0c 00 f7 d8 64 89 01 48 RSP: 002b:00007f62c1091e58 EFLAGS: 00000296 ORIG_RAX: 000000000000002a RAX: ffffffffffffffda RBX: 0000000020ccb004 RCX: 00007f62c11a885d RDX: 0000000000000010 RSI: 0000000020ccb000 RDI: 0000000000000003 RBP: 00007f62c1091e90 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000296 R12: 00007f62c10926c0 R13: ffffffffffffff88 R14: 0000000000000000 R15: 00007ffe237885b0 Fixes: ec94c2696f0b ("tcp/dccp: avoid one atomic operation for timewait hashdance") Reported-by: Anderson Nascimento Closes: https://lore.kernel.org/netdev/37a477a6-d39e-486b-9577-3463f655a6b7@allelesecurity.com/ Suggested-by: Eric Dumazet Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/20240501213145.62261-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp_ipv4.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a22ee5838751..e0cef75f85fb 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -154,6 +154,12 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp) if (tcptw->tw_ts_recent_stamp && (!twp || (reuse && time_after32(ktime_get_seconds(), tcptw->tw_ts_recent_stamp)))) { + /* inet_twsk_hashdance() sets sk_refcnt after putting twsk + * and releasing the bucket lock. + */ + if (unlikely(!refcount_inc_not_zero(&sktw->sk_refcnt))) + return 0; + /* In case of repair and re-using TIME-WAIT sockets we still * want to be sure that it is safe as above but honor the * sequence numbers and time stamps set as part of the repair @@ -174,7 +180,7 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp) tp->rx_opt.ts_recent = tcptw->tw_ts_recent; tp->rx_opt.ts_recent_stamp = tcptw->tw_ts_recent_stamp; } - sock_hold(sktw); + return 1; } -- cgit v1.2.3 From 838c7b8f1f278404d9d684c34a8cb26dc41aaaa1 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 24 Apr 2024 15:01:01 -0700 Subject: wifi: nl80211: Avoid address calculations via out of bounds array indexing Before request->channels[] can be used, request->n_channels must be set. Additionally, address calculations for memory after the "channels" array need to be calculated from the allocation base ("request") rather than via the first "out of bounds" index of "channels", otherwise run-time bounds checking will throw a warning. Reported-by: Nathan Chancellor Fixes: e3eac9f32ec0 ("wifi: cfg80211: Annotate struct cfg80211_scan_request with __counted_by") Signed-off-by: Kees Cook Tested-by: Nathan Chancellor Link: https://msgid.link/20240424220057.work.819-kees@kernel.org Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 30ff9a470813..65c416e8d25e 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -9162,6 +9162,7 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) struct wiphy *wiphy; int err, tmp, n_ssids = 0, n_channels, i; size_t ie_len, size; + size_t ssids_offset, ie_offset; wiphy = &rdev->wiphy; @@ -9207,21 +9208,20 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) return -EINVAL; size = struct_size(request, channels, n_channels); + ssids_offset = size; size = size_add(size, array_size(sizeof(*request->ssids), n_ssids)); + ie_offset = size; size = size_add(size, ie_len); request = kzalloc(size, GFP_KERNEL); if (!request) return -ENOMEM; + request->n_channels = n_channels; if (n_ssids) - request->ssids = (void *)&request->channels[n_channels]; + request->ssids = (void *)request + ssids_offset; request->n_ssids = n_ssids; - if (ie_len) { - if (n_ssids) - request->ie = (void *)(request->ssids + n_ssids); - else - request->ie = (void *)(request->channels + n_channels); - } + if (ie_len) + request->ie = (void *)request + ie_offset; i = 0; if (scan_freqs) { -- cgit v1.2.3 From 9f6d4b8d149af8dc3f9a1e3000168b99ca576390 Mon Sep 17 00:00:00 2001 From: Xin Deng Date: Fri, 26 Apr 2024 02:25:01 -0700 Subject: wifi: cfg80211: Clear mlo_links info when STA disconnects wdev->valid_links is not cleared when upper layer disconnect from a wdev->AP MLD. It has been observed that this would prevent offchannel operations like remain-on-channel which would be needed for user space operations with Public Action frame. Clear the wdev->valid_links when STA disconnects. Signed-off-by: Xin Deng Link: https://msgid.link/20240426092501.8592-1-quic_deng@quicinc.com Signed-off-by: Johannes Berg --- net/wireless/sme.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/wireless/sme.c b/net/wireless/sme.c index 82e3ce42206c..a8ad55f11133 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -1353,6 +1353,7 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, return; cfg80211_wdev_release_bsses(wdev); + wdev->valid_links = 0; wdev->connected = false; wdev->u.client.ssid_len = 0; wdev->conn_owner_nlportid = 0; -- cgit v1.2.3 From 91d2b6ee137dc709c8012d4e02c28a308f6753a2 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Mon, 22 Apr 2024 11:04:06 +0530 Subject: wifi: cfg80211: handle color change per link Currently, during color change, no link id information is passed down. In order to support color change during Multi Link Operation, it is required to pass link id as well. Additionally, update notification APIs to allow drivers/mac80211 to pass the link ID. Signed-off-by: Aditya Kumar Singh Link: https://msgid.link/20240422053412.2024075-2-quic_adisi@quicinc.com Link: https://msgid.link/20240422053412.2024075-3-quic_adisi@quicinc.com [squash, actually only pass 0 from mac80211] Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 29 ++++++++++++++++++++--------- net/mac80211/cfg.c | 10 +++++----- net/wireless/nl80211.c | 10 ++++++++-- net/wireless/trace.h | 6 ++++-- 4 files changed, 37 insertions(+), 18 deletions(-) (limited to 'net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 0112d645a2b4..cbf1664dc569 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1577,6 +1577,8 @@ struct cfg80211_csa_settings { * @beacon_next: beacon data to be used after the color change * @count: number of beacons until the color change * @color: the color used after the change + * @link_id: defines the link on which color change is expected during MLO. + * 0 in case of non-MLO. */ struct cfg80211_color_change_settings { struct cfg80211_beacon_data beacon_color_change; @@ -1585,6 +1587,7 @@ struct cfg80211_color_change_settings { struct cfg80211_beacon_data beacon_next; u8 count; u8 color; + u8 link_id; }; /** @@ -9425,70 +9428,78 @@ void cfg80211_bss_flush(struct wiphy *wiphy); * @cmd: the actual event we want to notify * @count: the number of TBTTs until the color change happens * @color_bitmap: representations of the colors that the local BSS is aware of + * @link_id: valid link_id in case of MLO or 0 for non-MLO. * * Return: 0 on success. Non-zero on error. */ int cfg80211_bss_color_notify(struct net_device *dev, enum nl80211_commands cmd, u8 count, - u64 color_bitmap); + u64 color_bitmap, u8 link_id); /** * cfg80211_obss_color_collision_notify - notify about bss color collision * @dev: network device * @color_bitmap: representations of the colors that the local BSS is aware of + * @link_id: valid link_id in case of MLO or 0 for non-MLO. * * Return: 0 on success. Non-zero on error. */ static inline int cfg80211_obss_color_collision_notify(struct net_device *dev, - u64 color_bitmap) + u64 color_bitmap, + u8 link_id) { return cfg80211_bss_color_notify(dev, NL80211_CMD_OBSS_COLOR_COLLISION, - 0, color_bitmap); + 0, color_bitmap, link_id); } /** * cfg80211_color_change_started_notify - notify color change start * @dev: the device on which the color is switched * @count: the number of TBTTs until the color change happens + * @link_id: valid link_id in case of MLO or 0 for non-MLO. * * Inform the userspace about the color change that has started. * * Return: 0 on success. Non-zero on error. */ static inline int cfg80211_color_change_started_notify(struct net_device *dev, - u8 count) + u8 count, u8 link_id) { return cfg80211_bss_color_notify(dev, NL80211_CMD_COLOR_CHANGE_STARTED, - count, 0); + count, 0, link_id); } /** * cfg80211_color_change_aborted_notify - notify color change abort * @dev: the device on which the color is switched + * @link_id: valid link_id in case of MLO or 0 for non-MLO. * * Inform the userspace about the color change that has aborted. * * Return: 0 on success. Non-zero on error. */ -static inline int cfg80211_color_change_aborted_notify(struct net_device *dev) +static inline int cfg80211_color_change_aborted_notify(struct net_device *dev, + u8 link_id) { return cfg80211_bss_color_notify(dev, NL80211_CMD_COLOR_CHANGE_ABORTED, - 0, 0); + 0, 0, link_id); } /** * cfg80211_color_change_notify - notify color change completion * @dev: the device on which the color was switched + * @link_id: valid link_id in case of MLO or 0 for non-MLO. * * Inform the userspace about the color change that has completed. * * Return: 0 on success. Non-zero on error. */ -static inline int cfg80211_color_change_notify(struct net_device *dev) +static inline int cfg80211_color_change_notify(struct net_device *dev, + u8 link_id) { return cfg80211_bss_color_notify(dev, NL80211_CMD_COLOR_CHANGE_COMPLETED, - 0, 0); + 0, 0, link_id); } /** diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 44971a62d798..4131ad6c1de3 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -3924,7 +3924,7 @@ static void ieee80211_color_change_abort(struct ieee80211_sub_if_data *sdata) ieee80211_free_next_beacon(&sdata->deflink); - cfg80211_color_change_aborted_notify(sdata->dev); + cfg80211_color_change_aborted_notify(sdata->dev, 0); } static int @@ -4772,14 +4772,14 @@ static int ieee80211_color_change_finalize(struct ieee80211_sub_if_data *sdata) err = ieee80211_set_after_color_change_beacon(sdata, &changed); if (err) { - cfg80211_color_change_aborted_notify(sdata->dev); + cfg80211_color_change_aborted_notify(sdata->dev, 0); return err; } ieee80211_color_change_bss_config_notify(sdata, sdata->vif.bss_conf.color_change_color, 1, changed); - cfg80211_color_change_notify(sdata->dev); + cfg80211_color_change_notify(sdata->dev, 0); return 0; } @@ -4812,7 +4812,7 @@ void ieee80211_color_collision_detection_work(struct work_struct *work) color_collision_detect_work); struct ieee80211_sub_if_data *sdata = link->sdata; - cfg80211_obss_color_collision_notify(sdata->dev, link->color_bitmap); + cfg80211_obss_color_collision_notify(sdata->dev, link->color_bitmap, 0); } void ieee80211_color_change_finish(struct ieee80211_vif *vif) @@ -4876,7 +4876,7 @@ ieee80211_color_change(struct wiphy *wiphy, struct net_device *dev, sdata->vif.bss_conf.color_change_active = true; sdata->vif.bss_conf.color_change_color = params->color; - cfg80211_color_change_started_notify(sdata->dev, params->count); + cfg80211_color_change_started_notify(sdata->dev, params->count, 0); if (changed) ieee80211_color_change_bss_config_notify(sdata, 0, 0, changed); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index f391b4055944..1886f78d9c3c 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -16032,6 +16032,7 @@ static int nl80211_color_change(struct sk_buff *skb, struct genl_info *info) params.counter_offset_presp = offset; } + params.link_id = nl80211_link_id(info->attrs); err = rdev_color_change(rdev, dev, ¶ms); out: @@ -17434,7 +17435,8 @@ static const struct genl_small_ops nl80211_small_ops[] = { .validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP, .doit = nl80211_color_change, .flags = GENL_UNS_ADMIN_PERM, - .internal_flags = IFLAGS(NL80211_FLAG_NEED_NETDEV_UP), + .internal_flags = IFLAGS(NL80211_FLAG_NEED_NETDEV_UP | + NL80211_FLAG_MLO_VALID_LINK_ID), }, { .cmd = NL80211_CMD_SET_FILS_AAD, @@ -19456,7 +19458,7 @@ EXPORT_SYMBOL(cfg80211_ch_switch_started_notify); int cfg80211_bss_color_notify(struct net_device *dev, enum nl80211_commands cmd, u8 count, - u64 color_bitmap) + u64 color_bitmap, u8 link_id) { struct wireless_dev *wdev = dev->ieee80211_ptr; struct wiphy *wiphy = wdev->wiphy; @@ -19479,6 +19481,10 @@ int cfg80211_bss_color_notify(struct net_device *dev, if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex)) goto nla_put_failure; + if (wdev->valid_links && + nla_put_u8(msg, NL80211_ATTR_MLO_LINK_ID, link_id)) + goto nla_put_failure; + if (cmd == NL80211_CMD_COLOR_CHANGE_STARTED && nla_put_u32(msg, NL80211_ATTR_COLOR_CHANGE_COUNT, count)) goto nla_put_failure; diff --git a/net/wireless/trace.h b/net/wireless/trace.h index df013c98b80d..9bf987519811 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -2842,6 +2842,7 @@ TRACE_EVENT(rdev_color_change, __field(u8, count) __field(u16, bcn_ofs) __field(u16, pres_ofs) + __field(u8, link_id) ), TP_fast_assign( WIPHY_ASSIGN; @@ -2849,11 +2850,12 @@ TRACE_EVENT(rdev_color_change, __entry->count = params->count; __entry->bcn_ofs = params->counter_offset_beacon; __entry->pres_ofs = params->counter_offset_presp; + __entry->link_id = params->link_id; ), TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT - ", count: %u", + ", count: %u, link_id: %d", WIPHY_PR_ARG, NETDEV_PR_ARG, - __entry->count) + __entry->count, __entry->link_id) ); TRACE_EVENT(rdev_set_radar_background, -- cgit v1.2.3 From 414e736c3db032733d20d0ada8200af4159d2940 Mon Sep 17 00:00:00 2001 From: Aditya Kumar Singh Date: Mon, 22 Apr 2024 11:04:08 +0530 Subject: wifi: mac80211: handle color change per link In order to support color change with MLO, handle the link ID now passed from cfg80211, adjust the code to do everything per link and call the notifications to cfg80211 correctly. Signed-off-by: Aditya Kumar Singh Link: https://msgid.link/20240422053412.2024075-4-quic_adisi@quicinc.com Link: https://msgid.link/20240422053412.2024075-5-quic_adisi@quicinc.com Link: https://msgid.link/20240422053412.2024075-6-quic_adisi@quicinc.com Link: https://msgid.link/20240422053412.2024075-7-quic_adisi@quicinc.com [squash, move API call updates to this patch] Signed-off-by: Johannes Berg --- drivers/net/wireless/ath/ath11k/mac.c | 2 +- drivers/net/wireless/ath/ath11k/wmi.c | 3 +- drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +- drivers/net/wireless/mediatek/mt76/mt7996/mcu.c | 2 +- include/net/mac80211.h | 6 +- net/mac80211/cfg.c | 144 ++++++++++++++++-------- net/mac80211/rx.c | 7 +- 7 files changed, 111 insertions(+), 55 deletions(-) (limited to 'net') diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c index 4fdd56cd30a1..7fa760ef674e 100644 --- a/drivers/net/wireless/ath/ath11k/mac.c +++ b/drivers/net/wireless/ath/ath11k/mac.c @@ -1659,7 +1659,7 @@ void ath11k_mac_bcn_tx_event(struct ath11k_vif *arvif) if (vif->bss_conf.color_change_active && ieee80211_beacon_cntdwn_is_complete(vif, 0)) { arvif->bcca_zero_sent = true; - ieee80211_color_change_finish(vif); + ieee80211_color_change_finish(vif, 0); return; } diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c index c74aa3f95658..3761dfce1f64 100644 --- a/drivers/net/wireless/ath/ath11k/wmi.c +++ b/drivers/net/wireless/ath/ath11k/wmi.c @@ -4064,7 +4064,8 @@ ath11k_wmi_obss_color_collision_event(struct ath11k_base *ab, struct sk_buff *sk switch (ev->evt_type) { case WMI_BSS_COLOR_COLLISION_DETECTION: - ieee80211_obss_color_collision_notify(arvif->vif, ev->obss_color_bitmap); + ieee80211_obss_color_collision_notify(arvif->vif, ev->obss_color_bitmap, + 0); ath11k_dbg(ab, ATH11K_DBG_WMI, "OBSS color collision detected vdev:%d, event:%d, bitmap:%08llx\n", ev->vdev_id, ev->evt_type, ev->obss_color_bitmap); diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c index d90f98c50039..6442e3e5d45b 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c @@ -331,7 +331,7 @@ mt7915_mcu_cca_finish(void *priv, u8 *mac, struct ieee80211_vif *vif) if (!vif->bss_conf.color_change_active || vif->type == NL80211_IFTYPE_STATION) return; - ieee80211_color_change_finish(vif); + ieee80211_color_change_finish(vif, 0); } static void diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c index b44abe2acc81..163b822a477c 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c +++ b/drivers/net/wireless/mediatek/mt76/mt7996/mcu.c @@ -418,7 +418,7 @@ mt7996_mcu_cca_finish(void *priv, u8 *mac, struct ieee80211_vif *vif) if (!vif->bss_conf.color_change_active || vif->type == NL80211_IFTYPE_STATION) return; - ieee80211_color_change_finish(vif); + ieee80211_color_change_finish(vif, 0); } static void diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 5be28b45049b..cafc664ee531 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -5615,12 +5615,13 @@ bool ieee80211_beacon_cntdwn_is_complete(struct ieee80211_vif *vif, /** * ieee80211_color_change_finish - notify mac80211 about color change * @vif: &struct ieee80211_vif pointer from the add_interface callback. + * @link_id: valid link_id during MLO or 0 for non-MLO * * After a color change announcement was scheduled and the counter in this * announcement hits 1, this function must be called by the driver to * notify mac80211 that the color can be changed */ -void ieee80211_color_change_finish(struct ieee80211_vif *vif); +void ieee80211_color_change_finish(struct ieee80211_vif *vif, u8 link_id); /** * ieee80211_proberesp_get - retrieve a Probe Response template @@ -7533,6 +7534,7 @@ ieee80211_get_unsol_bcast_probe_resp_tmpl(struct ieee80211_hw *hw, /** * ieee80211_obss_color_collision_notify - notify userland about a BSS color * collision. + * @link_id: valid link_id during MLO or 0 for non-MLO * * @vif: &struct ieee80211_vif pointer from the add_interface callback. * @color_bitmap: a 64 bit bitmap representing the colors that the local BSS is @@ -7540,7 +7542,7 @@ ieee80211_get_unsol_bcast_probe_resp_tmpl(struct ieee80211_hw *hw, */ void ieee80211_obss_color_collision_notify(struct ieee80211_vif *vif, - u64 color_bitmap); + u64 color_bitmap, u8 link_id); /** * ieee80211_is_tx_data - check if frame is a data frame diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 4131ad6c1de3..b08e5d7687e3 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -3918,13 +3918,13 @@ static int ieee80211_set_csa_beacon(struct ieee80211_link_data *link_data, return 0; } -static void ieee80211_color_change_abort(struct ieee80211_sub_if_data *sdata) +static void ieee80211_color_change_abort(struct ieee80211_link_data *link) { - sdata->vif.bss_conf.color_change_active = false; + link->conf->color_change_active = false; - ieee80211_free_next_beacon(&sdata->deflink); + ieee80211_free_next_beacon(link); - cfg80211_color_change_aborted_notify(sdata->dev, 0); + cfg80211_color_change_aborted_notify(link->sdata->dev, link->link_id); } static int @@ -4008,7 +4008,7 @@ __ieee80211_channel_switch(struct wiphy *wiphy, struct net_device *dev, /* if there is a color change in progress, abort it */ if (link_conf->color_change_active) - ieee80211_color_change_abort(sdata); + ieee80211_color_change_abort(link_data); err = ieee80211_set_csa_beacon(link_data, params, &changed); if (err) { @@ -4666,20 +4666,22 @@ static int ieee80211_set_sar_specs(struct wiphy *wiphy, } static int -ieee80211_set_after_color_change_beacon(struct ieee80211_sub_if_data *sdata, +ieee80211_set_after_color_change_beacon(struct ieee80211_link_data *link, u64 *changed) { + struct ieee80211_sub_if_data *sdata = link->sdata; + switch (sdata->vif.type) { case NL80211_IFTYPE_AP: { int ret; - if (!sdata->deflink.u.ap.next_beacon) + if (!link->u.ap.next_beacon) return -EINVAL; - ret = ieee80211_assign_beacon(sdata, &sdata->deflink, - sdata->deflink.u.ap.next_beacon, + ret = ieee80211_assign_beacon(sdata, link, + link->u.ap.next_beacon, NULL, NULL, changed); - ieee80211_free_next_beacon(&sdata->deflink); + ieee80211_free_next_beacon(link); if (ret < 0) return ret; @@ -4695,18 +4697,19 @@ ieee80211_set_after_color_change_beacon(struct ieee80211_sub_if_data *sdata, } static int -ieee80211_set_color_change_beacon(struct ieee80211_sub_if_data *sdata, +ieee80211_set_color_change_beacon(struct ieee80211_link_data *link, struct cfg80211_color_change_settings *params, u64 *changed) { + struct ieee80211_sub_if_data *sdata = link->sdata; struct ieee80211_color_change_settings color_change = {}; int err; switch (sdata->vif.type) { case NL80211_IFTYPE_AP: - sdata->deflink.u.ap.next_beacon = + link->u.ap.next_beacon = cfg80211_beacon_dup(¶ms->beacon_next); - if (!sdata->deflink.u.ap.next_beacon) + if (!link->u.ap.next_beacon) return -ENOMEM; if (params->count <= 1) @@ -4718,11 +4721,11 @@ ieee80211_set_color_change_beacon(struct ieee80211_sub_if_data *sdata, params->counter_offset_presp; color_change.count = params->count; - err = ieee80211_assign_beacon(sdata, &sdata->deflink, + err = ieee80211_assign_beacon(sdata, link, ¶ms->beacon_color_change, NULL, &color_change, changed); if (err < 0) { - ieee80211_free_next_beacon(&sdata->deflink); + ieee80211_free_next_beacon(link); return err; } break; @@ -4734,16 +4737,18 @@ ieee80211_set_color_change_beacon(struct ieee80211_sub_if_data *sdata, } static void -ieee80211_color_change_bss_config_notify(struct ieee80211_sub_if_data *sdata, +ieee80211_color_change_bss_config_notify(struct ieee80211_link_data *link, u8 color, int enable, u64 changed) { + struct ieee80211_sub_if_data *sdata = link->sdata; + lockdep_assert_wiphy(sdata->local->hw.wiphy); - sdata->vif.bss_conf.he_bss_color.color = color; - sdata->vif.bss_conf.he_bss_color.enabled = enable; + link->conf->he_bss_color.color = color; + link->conf->he_bss_color.enabled = enable; changed |= BSS_CHANGED_HE_BSS_COLOR; - ieee80211_link_info_change_notify(sdata, &sdata->deflink, changed); + ieee80211_link_info_change_notify(sdata, link, changed); if (!sdata->vif.bss_conf.nontransmitted && sdata->vif.mbssid_tx_vif) { struct ieee80211_sub_if_data *child; @@ -4760,26 +4765,27 @@ ieee80211_color_change_bss_config_notify(struct ieee80211_sub_if_data *sdata, } } -static int ieee80211_color_change_finalize(struct ieee80211_sub_if_data *sdata) +static int ieee80211_color_change_finalize(struct ieee80211_link_data *link) { + struct ieee80211_sub_if_data *sdata = link->sdata; struct ieee80211_local *local = sdata->local; u64 changed = 0; int err; lockdep_assert_wiphy(local->hw.wiphy); - sdata->vif.bss_conf.color_change_active = false; + link->conf->color_change_active = false; - err = ieee80211_set_after_color_change_beacon(sdata, &changed); + err = ieee80211_set_after_color_change_beacon(link, &changed); if (err) { - cfg80211_color_change_aborted_notify(sdata->dev, 0); + cfg80211_color_change_aborted_notify(sdata->dev, link->link_id); return err; } - ieee80211_color_change_bss_config_notify(sdata, - sdata->vif.bss_conf.color_change_color, + ieee80211_color_change_bss_config_notify(link, + link->conf->color_change_color, 1, changed); - cfg80211_color_change_notify(sdata->dev, 0); + cfg80211_color_change_notify(sdata->dev, link->link_id); return 0; } @@ -4787,21 +4793,23 @@ static int ieee80211_color_change_finalize(struct ieee80211_sub_if_data *sdata) void ieee80211_color_change_finalize_work(struct wiphy *wiphy, struct wiphy_work *work) { - struct ieee80211_sub_if_data *sdata = - container_of(work, struct ieee80211_sub_if_data, - deflink.color_change_finalize_work); + struct ieee80211_link_data *link = + container_of(work, struct ieee80211_link_data, + color_change_finalize_work); + struct ieee80211_sub_if_data *sdata = link->sdata; + struct ieee80211_bss_conf *link_conf = link->conf; struct ieee80211_local *local = sdata->local; lockdep_assert_wiphy(local->hw.wiphy); /* AP might have been stopped while waiting for the lock. */ - if (!sdata->vif.bss_conf.color_change_active) + if (!link_conf->color_change_active) return; if (!ieee80211_sdata_running(sdata)) return; - ieee80211_color_change_finalize(sdata); + ieee80211_color_change_finalize(link); } void ieee80211_color_collision_detection_work(struct work_struct *work) @@ -4812,30 +4820,60 @@ void ieee80211_color_collision_detection_work(struct work_struct *work) color_collision_detect_work); struct ieee80211_sub_if_data *sdata = link->sdata; - cfg80211_obss_color_collision_notify(sdata->dev, link->color_bitmap, 0); + cfg80211_obss_color_collision_notify(sdata->dev, link->color_bitmap, + link->link_id); } -void ieee80211_color_change_finish(struct ieee80211_vif *vif) +void ieee80211_color_change_finish(struct ieee80211_vif *vif, u8 link_id) { struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); + struct ieee80211_link_data *link; + + if (WARN_ON(link_id >= IEEE80211_MLD_MAX_NUM_LINKS)) + return; + + rcu_read_lock(); + + link = rcu_dereference(sdata->link[link_id]); + if (WARN_ON(!link)) { + rcu_read_unlock(); + return; + } wiphy_work_queue(sdata->local->hw.wiphy, - &sdata->deflink.color_change_finalize_work); + &link->color_change_finalize_work); + + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(ieee80211_color_change_finish); void ieee80211_obss_color_collision_notify(struct ieee80211_vif *vif, - u64 color_bitmap) + u64 color_bitmap, u8 link_id) { struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); - struct ieee80211_link_data *link = &sdata->deflink; + struct ieee80211_link_data *link; - if (sdata->vif.bss_conf.color_change_active || sdata->vif.bss_conf.csa_active) + if (WARN_ON(link_id >= IEEE80211_MLD_MAX_NUM_LINKS)) return; - if (delayed_work_pending(&link->color_collision_detect_work)) + rcu_read_lock(); + + link = rcu_dereference(sdata->link[link_id]); + if (WARN_ON(!link)) { + rcu_read_unlock(); return; + } + + if (link->conf->color_change_active || link->conf->csa_active) { + rcu_read_unlock(); + return; + } + + if (delayed_work_pending(&link->color_collision_detect_work)) { + rcu_read_unlock(); + return; + } link->color_bitmap = color_bitmap; /* queue the color collision detection event every 500 ms in order to @@ -4844,6 +4882,8 @@ ieee80211_obss_color_collision_notify(struct ieee80211_vif *vif, ieee80211_queue_delayed_work(&sdata->local->hw, &link->color_collision_detect_work, msecs_to_jiffies(500)); + + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(ieee80211_obss_color_collision_notify); @@ -4853,36 +4893,48 @@ ieee80211_color_change(struct wiphy *wiphy, struct net_device *dev, { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = sdata->local; + struct ieee80211_bss_conf *link_conf; + struct ieee80211_link_data *link; + u8 link_id = params->link_id; u64 changed = 0; int err; lockdep_assert_wiphy(local->hw.wiphy); - if (sdata->vif.bss_conf.nontransmitted) + if (WARN_ON(link_id >= IEEE80211_MLD_MAX_NUM_LINKS)) + return -EINVAL; + + link = wiphy_dereference(wiphy, sdata->link[link_id]); + if (!link) + return -ENOLINK; + + link_conf = link->conf; + + if (link_conf->nontransmitted) return -EINVAL; /* don't allow another color change if one is already active or if csa * is active */ - if (sdata->vif.bss_conf.color_change_active || sdata->vif.bss_conf.csa_active) { + if (link_conf->color_change_active || link_conf->csa_active) { err = -EBUSY; goto out; } - err = ieee80211_set_color_change_beacon(sdata, params, &changed); + err = ieee80211_set_color_change_beacon(link, params, &changed); if (err) goto out; - sdata->vif.bss_conf.color_change_active = true; - sdata->vif.bss_conf.color_change_color = params->color; + link_conf->color_change_active = true; + link_conf->color_change_color = params->color; - cfg80211_color_change_started_notify(sdata->dev, params->count, 0); + cfg80211_color_change_started_notify(sdata->dev, params->count, link_id); if (changed) - ieee80211_color_change_bss_config_notify(sdata, 0, 0, changed); + ieee80211_color_change_bss_config_notify(link, 0, 0, changed); else /* if the beacon didn't change, we can finalize immediately */ - ieee80211_color_change_finalize(sdata); + ieee80211_color_change_finalize(link); out: diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 8274dba76206..4914692750e5 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -3368,7 +3368,7 @@ ieee80211_rx_check_bss_color_collision(struct ieee80211_rx_data *rx) if (ieee80211_hw_check(&rx->local->hw, DETECTS_COLOR_COLLISION)) return; - if (rx->sdata->vif.bss_conf.csa_active) + if (rx->link->conf->csa_active) return; baselen = mgmt->u.beacon.variable - rx->skb->data; @@ -3380,7 +3380,7 @@ ieee80211_rx_check_bss_color_collision(struct ieee80211_rx_data *rx) rx->skb->len - baselen); if (ie && ie->datalen >= sizeof(struct ieee80211_he_operation) && ie->datalen >= ieee80211_he_oper_size(ie->data + 1)) { - struct ieee80211_bss_conf *bss_conf = &rx->sdata->vif.bss_conf; + struct ieee80211_bss_conf *bss_conf = rx->link->conf; const struct ieee80211_he_operation *he_oper; u8 color; @@ -3393,7 +3393,8 @@ ieee80211_rx_check_bss_color_collision(struct ieee80211_rx_data *rx) IEEE80211_HE_OPERATION_BSS_COLOR_MASK); if (color == bss_conf->he_bss_color.color) ieee80211_obss_color_collision_notify(&rx->sdata->vif, - BIT_ULL(color)); + BIT_ULL(color), + bss_conf->link_id); } } -- cgit v1.2.3 From ce218712b0f6f97f9e838634548044b970eeca63 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:25 +0200 Subject: net: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) * Remove sentinel element from ctl_table structs. * Remove the zeroing out of an array element (to make it look like a sentinel) in neigh_sysctl_register and lowpan_frags_ns_sysctl_register This is not longer needed and is safe after commit c899710fe7f9 ("networking: Update to register_net_sysctl_sz") added the array size to the ctl_table registration. * Replace the for loop stop condition in sysctl_core_net_init that tests for procname == NULL with one that depends on array size * Removed the "-1" in mpls_net_init that adjusted for having an extra empty element when looping over ctl_table arrays * Use a table_size variable to keep the value of ARRAY_SIZE Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/core/neighbour.c | 5 +---- net/core/sysctl_net_core.c | 13 ++++++------- net/dccp/sysctl.c | 2 -- net/ieee802154/6lowpan/reassembly.c | 6 +----- net/mpls/af_mpls.c | 13 ++++++------- net/unix/sysctl_net_unix.c | 1 - 6 files changed, 14 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/net/core/neighbour.c b/net/core/neighbour.c index af270c202d9a..45fd88405b6b 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -3733,7 +3733,7 @@ static int neigh_proc_base_reachable_time(struct ctl_table *ctl, int write, static struct neigh_sysctl_table { struct ctl_table_header *sysctl_header; - struct ctl_table neigh_vars[NEIGH_VAR_MAX + 1]; + struct ctl_table neigh_vars[NEIGH_VAR_MAX]; } neigh_sysctl_template __read_mostly = { .neigh_vars = { NEIGH_SYSCTL_ZERO_INTMAX_ENTRY(MCAST_PROBES, "mcast_solicit"), @@ -3784,7 +3784,6 @@ static struct neigh_sysctl_table { .extra2 = SYSCTL_INT_MAX, .proc_handler = proc_dointvec_minmax, }, - {}, }, }; @@ -3812,8 +3811,6 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, if (dev) { dev_name_source = dev->name; /* Terminate the table early */ - memset(&t->neigh_vars[NEIGH_VAR_GC_INTERVAL], 0, - sizeof(t->neigh_vars[NEIGH_VAR_GC_INTERVAL])); neigh_vars_size = NEIGH_VAR_BASE_REACHABLE_TIME_MS + 1; } else { struct neigh_table *tbl = p->tbl; diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 6da5995ac86a..c9fb9ad87485 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -661,7 +661,6 @@ static struct ctl_table net_core_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, - { } }; static struct ctl_table netns_core_table[] = { @@ -698,7 +697,6 @@ static struct ctl_table netns_core_table[] = { .extra2 = SYSCTL_ONE, .proc_handler = proc_dou8vec_minmax, }, - { } }; static int __init fb_tunnels_only_for_init_net_sysctl_setup(char *str) @@ -716,20 +714,21 @@ __setup("fb_tunnels=", fb_tunnels_only_for_init_net_sysctl_setup); static __net_init int sysctl_core_net_init(struct net *net) { - struct ctl_table *tbl, *tmp; + size_t table_size = ARRAY_SIZE(netns_core_table); + struct ctl_table *tbl; tbl = netns_core_table; if (!net_eq(net, &init_net)) { + int i; tbl = kmemdup(tbl, sizeof(netns_core_table), GFP_KERNEL); if (tbl == NULL) goto err_dup; - for (tmp = tbl; tmp->procname; tmp++) - tmp->data += (char *)net - (char *)&init_net; + for (i = 0; i < table_size; ++i) + tbl[i].data += (char *)net - (char *)&init_net; } - net->core.sysctl_hdr = register_net_sysctl_sz(net, "net/core", tbl, - ARRAY_SIZE(netns_core_table)); + net->core.sysctl_hdr = register_net_sysctl_sz(net, "net/core", tbl, table_size); if (net->core.sysctl_hdr == NULL) goto err_reg; diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index ee8d4f5afa72..3fc474d6e57d 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -90,8 +90,6 @@ static struct ctl_table dccp_default_table[] = { .mode = 0644, .proc_handler = proc_dointvec_ms_jiffies, }, - - { } }; static struct ctl_table_header *dccp_table_header; diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index 2a983cf450da..56ef873828f4 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -338,7 +338,6 @@ static struct ctl_table lowpan_frags_ns_ctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; /* secret interval has been deprecated */ @@ -351,7 +350,6 @@ static struct ctl_table lowpan_frags_ctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; static int __net_init lowpan_frags_ns_sysctl_register(struct net *net) @@ -370,10 +368,8 @@ static int __net_init lowpan_frags_ns_sysctl_register(struct net *net) goto err_alloc; /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) { - table[0].procname = NULL; + if (net->user_ns != &init_user_ns) table_size = 0; - } } table[0].data = &ieee802154_lowpan->fqdir->high_thresh; diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 5d2012d1cf4a..2dc7a908a6bb 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1377,13 +1377,13 @@ static const struct ctl_table mpls_dev_table[] = { .proc_handler = mpls_conf_proc, .data = MPLS_PERDEV_SYSCTL_OFFSET(input_enabled), }, - { } }; static int mpls_dev_sysctl_register(struct net_device *dev, struct mpls_dev *mdev) { char path[sizeof("net/mpls/conf/") + IFNAMSIZ]; + size_t table_size = ARRAY_SIZE(mpls_dev_table); struct net *net = dev_net(dev); struct ctl_table *table; int i; @@ -1395,7 +1395,7 @@ static int mpls_dev_sysctl_register(struct net_device *dev, /* Table data contains only offsets relative to the base of * the mdev at this point, so make them absolute. */ - for (i = 0; i < ARRAY_SIZE(mpls_dev_table); i++) { + for (i = 0; i < table_size; i++) { table[i].data = (char *)mdev + (uintptr_t)table[i].data; table[i].extra1 = mdev; table[i].extra2 = net; @@ -1403,8 +1403,7 @@ static int mpls_dev_sysctl_register(struct net_device *dev, snprintf(path, sizeof(path), "net/mpls/conf/%s", dev->name); - mdev->sysctl = register_net_sysctl_sz(net, path, table, - ARRAY_SIZE(mpls_dev_table)); + mdev->sysctl = register_net_sysctl_sz(net, path, table, table_size); if (!mdev->sysctl) goto free; @@ -2653,11 +2652,11 @@ static const struct ctl_table mpls_table[] = { .extra1 = SYSCTL_ONE, .extra2 = &ttl_max, }, - { } }; static int mpls_net_init(struct net *net) { + size_t table_size = ARRAY_SIZE(mpls_table); struct ctl_table *table; int i; @@ -2673,11 +2672,11 @@ static int mpls_net_init(struct net *net) /* Table data contains only offsets relative to the base of * the mdev at this point, so make them absolute. */ - for (i = 0; i < ARRAY_SIZE(mpls_table) - 1; i++) + for (i = 0; i < table_size; i++) table[i].data = (char *)net + (uintptr_t)table[i].data; net->mpls.ctl = register_net_sysctl_sz(net, "net/mpls", table, - ARRAY_SIZE(mpls_table)); + table_size); if (net->mpls.ctl == NULL) { kfree(table); return -ENOMEM; diff --git a/net/unix/sysctl_net_unix.c b/net/unix/sysctl_net_unix.c index 44996af61999..357b3e5f3847 100644 --- a/net/unix/sysctl_net_unix.c +++ b/net/unix/sysctl_net_unix.c @@ -19,7 +19,6 @@ static struct ctl_table unix_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { } }; int __net_init unix_sysctl_register(struct net *net) -- cgit v1.2.3 From 1c106eb01ceefa1ceec56af497e628593fb5fd9f Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:26 +0200 Subject: net: ipv{6,4}: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) * Remove sentinel element from ctl_table structs. * Remove the zeroing out of an array element (to make it look like a sentinel) in sysctl_route_net_init And ipv6_route_sysctl_init. This is not longer needed and is safe after commit c899710fe7f9 ("networking: Update to register_net_sysctl_sz") added the array size to the ctl_table registration. * Remove extra sentinel element in the declaration of devinet_vars. * Removed the "-1" in __devinet_sysctl_register, sysctl_route_net_init, ipv6_sysctl_net_init and ipv4_sysctl_init_net that adjusted for having an extra empty element when looping over ctl_table arrays * Replace the for loop stop condition in __addrconf_sysctl_register that tests for procname == NULL with one that depends on array size * Removing the unprivileged user check in ipv6_route_sysctl_init is safe as it is replaced by calling ipv6_route_sysctl_table_size; introduced in commit c899710fe7f9 ("networking: Update to register_net_sysctl_sz") * Use a table_size variable to keep the value of ARRAY_SIZE Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/ipv4/devinet.c | 5 ++--- net/ipv4/ip_fragment.c | 2 -- net/ipv4/route.c | 8 ++------ net/ipv4/sysctl_net_ipv4.c | 7 +++---- net/ipv4/xfrm4_policy.c | 1 - net/ipv6/addrconf.c | 8 +++----- net/ipv6/icmp.c | 1 - net/ipv6/reassembly.c | 2 -- net/ipv6/route.c | 5 ----- net/ipv6/sysctl_net_ipv6.c | 8 +++----- net/ipv6/xfrm6_policy.c | 1 - 11 files changed, 13 insertions(+), 35 deletions(-) (limited to 'net') diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 364dbf0cd9bf..a612c57b61c5 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2520,7 +2520,7 @@ static int ipv4_doint_and_flush(struct ctl_table *ctl, int write, static struct devinet_sysctl_table { struct ctl_table_header *sysctl_header; - struct ctl_table devinet_vars[__IPV4_DEVCONF_MAX]; + struct ctl_table devinet_vars[IPV4_DEVCONF_MAX]; } devinet_sysctl = { .devinet_vars = { DEVINET_SYSCTL_COMPLEX_ENTRY(FORWARDING, "forwarding", @@ -2583,7 +2583,7 @@ static int __devinet_sysctl_register(struct net *net, char *dev_name, if (!t) goto out; - for (i = 0; i < ARRAY_SIZE(t->devinet_vars) - 1; i++) { + for (i = 0; i < ARRAY_SIZE(t->devinet_vars); i++) { t->devinet_vars[i].data += (char *)p - (char *)&ipv4_devconf; t->devinet_vars[i].extra1 = p; t->devinet_vars[i].extra2 = net; @@ -2657,7 +2657,6 @@ static struct ctl_table ctl_forward_entry[] = { .extra1 = &ipv4_devconf, .extra2 = &init_net, }, - { }, }; #endif diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 534b98a0744a..08e2c92e25ab 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -580,7 +580,6 @@ static struct ctl_table ip4_frags_ns_ctl_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = &dist_min, }, - { } }; /* secret interval has been deprecated */ @@ -593,7 +592,6 @@ static struct ctl_table ip4_frags_ctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; static int __net_init ip4_frags_ns_ctl_register(struct net *net) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 0fd9a3d7ac4a..5fd54103174f 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3496,7 +3496,6 @@ static struct ctl_table ipv4_route_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static const char ipv4_route_flush_procname[] = "flush"; @@ -3530,7 +3529,6 @@ static struct ctl_table ipv4_route_netns_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { }, }; static __net_init int sysctl_route_net_init(struct net *net) @@ -3548,16 +3546,14 @@ static __net_init int sysctl_route_net_init(struct net *net) /* Don't export non-whitelisted sysctls to unprivileged users */ if (net->user_ns != &init_user_ns) { - if (tbl[0].procname != ipv4_route_flush_procname) { - tbl[0].procname = NULL; + if (tbl[0].procname != ipv4_route_flush_procname) table_size = 0; - } } /* Update the variables to point into the current struct net * except for the first element flush */ - for (i = 1; i < ARRAY_SIZE(ipv4_route_netns_table) - 1; i++) + for (i = 1; i < table_size; i++) tbl[i].data += (void *)net - (void *)&init_net; } tbl[0].extra1 = net; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index ce5d19978a26..162a0a3b6ba5 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -575,7 +575,6 @@ static struct ctl_table ipv4_table[] = { .extra1 = &sysctl_fib_sync_mem_min, .extra2 = &sysctl_fib_sync_mem_max, }, - { } }; static struct ctl_table ipv4_net_table[] = { @@ -1502,11 +1501,11 @@ static struct ctl_table ipv4_net_table[] = { .proc_handler = proc_dou8vec_minmax, .extra1 = SYSCTL_ONE, }, - { } }; static __net_init int ipv4_sysctl_init_net(struct net *net) { + size_t table_size = ARRAY_SIZE(ipv4_net_table); struct ctl_table *table; table = ipv4_net_table; @@ -1517,7 +1516,7 @@ static __net_init int ipv4_sysctl_init_net(struct net *net) if (!table) goto err_alloc; - for (i = 0; i < ARRAY_SIZE(ipv4_net_table) - 1; i++) { + for (i = 0; i < table_size; i++) { if (table[i].data) { /* Update the variables to point into * the current struct net @@ -1533,7 +1532,7 @@ static __net_init int ipv4_sysctl_init_net(struct net *net) } net->ipv4.ipv4_hdr = register_net_sysctl_sz(net, "net/ipv4", table, - ARRAY_SIZE(ipv4_net_table)); + table_size); if (!net->ipv4.ipv4_hdr) goto err_reg; diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index fccbbd3e1a4b..0294fef577fa 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -152,7 +152,6 @@ static struct ctl_table xfrm4_policy_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static __net_init int xfrm4_net_sysctl_init(struct net *net) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 9aa0900abfa1..5c424a0e7232 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -7184,14 +7184,12 @@ static const struct ctl_table addrconf_sysctl[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_TWO, }, - { - /* sentinel */ - } }; static int __addrconf_sysctl_register(struct net *net, char *dev_name, struct inet6_dev *idev, struct ipv6_devconf *p) { + size_t table_size = ARRAY_SIZE(addrconf_sysctl); int i, ifindex; struct ctl_table *table; char path[sizeof("net/ipv6/conf/") + IFNAMSIZ]; @@ -7200,7 +7198,7 @@ static int __addrconf_sysctl_register(struct net *net, char *dev_name, if (!table) goto out; - for (i = 0; table[i].data; i++) { + for (i = 0; i < table_size; i++) { table[i].data += (char *)p - (char *)&ipv6_devconf; /* If one of these is already set, then it is not safe to * overwrite either of them: this makes proc_dointvec_minmax @@ -7215,7 +7213,7 @@ static int __addrconf_sysctl_register(struct net *net, char *dev_name, snprintf(path, sizeof(path), "net/ipv6/conf/%s", dev_name); p->sysctl_header = register_net_sysctl_sz(net, path, table, - ARRAY_SIZE(addrconf_sysctl)); + table_size); if (!p->sysctl_header) goto free; diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index d285c1f6f1a6..7b31674644ef 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -1206,7 +1206,6 @@ static struct ctl_table ipv6_icmp_table_template[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, - { }, }; struct ctl_table * __net_init ipv6_icmp_sysctl_init(struct net *net) diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index ee95cdcc8747..439f93512b0a 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -436,7 +436,6 @@ static struct ctl_table ip6_frags_ns_ctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; /* secret interval has been deprecated */ @@ -449,7 +448,6 @@ static struct ctl_table ip6_frags_ctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; static int __net_init ip6_frags_ns_sysctl_register(struct net *net) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 3e0b2cb20fd2..c43b0616742e 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -6428,7 +6428,6 @@ static struct ctl_table ipv6_route_table_template[] = { .extra1 = SYSCTL_ZERO, .extra2 = SYSCTL_ONE, }, - { } }; struct ctl_table * __net_init ipv6_route_sysctl_init(struct net *net) @@ -6452,10 +6451,6 @@ struct ctl_table * __net_init ipv6_route_sysctl_init(struct net *net) table[8].data = &net->ipv6.sysctl.ip6_rt_min_advmss; table[9].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval; table[10].data = &net->ipv6.sysctl.skip_notify_on_dev_down; - - /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) - table[1].procname = NULL; } return table; diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index 75de55f907b0..c060285ff47f 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -213,7 +213,6 @@ static struct ctl_table ipv6_table_template[] = { .proc_handler = proc_doulongvec_minmax, .extra2 = &ioam6_id_wide_max, }, - { } }; static struct ctl_table ipv6_rotable[] = { @@ -248,11 +247,11 @@ static struct ctl_table ipv6_rotable[] = { .proc_handler = proc_dointvec, }, #endif /* CONFIG_NETLABEL */ - { } }; static int __net_init ipv6_sysctl_net_init(struct net *net) { + size_t table_size = ARRAY_SIZE(ipv6_table_template); struct ctl_table *ipv6_table; struct ctl_table *ipv6_route_table; struct ctl_table *ipv6_icmp_table; @@ -264,7 +263,7 @@ static int __net_init ipv6_sysctl_net_init(struct net *net) if (!ipv6_table) goto out; /* Update the variables to point into the current struct net */ - for (i = 0; i < ARRAY_SIZE(ipv6_table_template) - 1; i++) + for (i = 0; i < table_size; i++) ipv6_table[i].data += (void *)net - (void *)&init_net; ipv6_route_table = ipv6_route_sysctl_init(net); @@ -276,8 +275,7 @@ static int __net_init ipv6_sysctl_net_init(struct net *net) goto out_ipv6_route_table; net->ipv6.sysctl.hdr = register_net_sysctl_sz(net, "net/ipv6", - ipv6_table, - ARRAY_SIZE(ipv6_table_template)); + ipv6_table, table_size); if (!net->ipv6.sysctl.hdr) goto out_ipv6_icmp_table; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 7924e08ee142..cc885d3aa9e5 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -184,7 +184,6 @@ static struct ctl_table xfrm6_policy_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static int __net_init xfrm6_net_sysctl_init(struct net *net) -- cgit v1.2.3 From 92bedf07836bf0971dc18a97307bc3d5dc9db787 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:27 +0200 Subject: net: rds: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) * Remove sentinel element from ctl_table structs. Signed-off-by: Joel Granados Acked-by: Allison Henderson Signed-off-by: David S. Miller --- net/rds/ib_sysctl.c | 1 - net/rds/sysctl.c | 1 - net/rds/tcp.c | 1 - 3 files changed, 3 deletions(-) (limited to 'net') diff --git a/net/rds/ib_sysctl.c b/net/rds/ib_sysctl.c index e4e41b3afce7..2af678e71e3c 100644 --- a/net/rds/ib_sysctl.c +++ b/net/rds/ib_sysctl.c @@ -103,7 +103,6 @@ static struct ctl_table rds_ib_sysctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; void rds_ib_sysctl_exit(void) diff --git a/net/rds/sysctl.c b/net/rds/sysctl.c index e381bbcd9cc1..025f518a4349 100644 --- a/net/rds/sysctl.c +++ b/net/rds/sysctl.c @@ -89,7 +89,6 @@ static struct ctl_table rds_sysctl_rds_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; void rds_sysctl_exit(void) diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 2dba7505b414..d8111ac83bb6 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -86,7 +86,6 @@ static struct ctl_table rds_tcp_sysctl_table[] = { .proc_handler = rds_tcp_skbuf_handler, .extra1 = &rds_tcp_min_rcvbuf, }, - { } }; u32 rds_tcp_write_seq(struct rds_tcp_connection *tc) -- cgit v1.2.3 From ca5d1fce7994e224f2701882cd6f0299b77834a0 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:28 +0200 Subject: net: sunrpc: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) * Remove sentinel element from ctl_table structs. Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/sunrpc/sysctl.c | 1 - net/sunrpc/xprtrdma/svc_rdma.c | 1 - net/sunrpc/xprtrdma/transport.c | 1 - net/sunrpc/xprtsock.c | 1 - 4 files changed, 4 deletions(-) (limited to 'net') diff --git a/net/sunrpc/sysctl.c b/net/sunrpc/sysctl.c index 93941ab12549..5f3170a1c9bb 100644 --- a/net/sunrpc/sysctl.c +++ b/net/sunrpc/sysctl.c @@ -160,7 +160,6 @@ static struct ctl_table debug_table[] = { .mode = 0444, .proc_handler = proc_do_xprt, }, - { } }; void diff --git a/net/sunrpc/xprtrdma/svc_rdma.c b/net/sunrpc/xprtrdma/svc_rdma.c index f86970733eb0..474f7a98fe9e 100644 --- a/net/sunrpc/xprtrdma/svc_rdma.c +++ b/net/sunrpc/xprtrdma/svc_rdma.c @@ -209,7 +209,6 @@ static struct ctl_table svcrdma_parm_table[] = { .extra1 = &zero, .extra2 = &zero, }, - { }, }; static void svc_rdma_proc_cleanup(void) diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index 29b0562d62e7..9a8ce5df83ca 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -137,7 +137,6 @@ static struct ctl_table xr_tunables_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { }, }; #endif diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index ce18716491c8..dfc353eea8ed 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -160,7 +160,6 @@ static struct ctl_table xs_tunables_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { }, }; /* -- cgit v1.2.3 From 73dbd8cf7947f7da47b23f4adc07593c4dc452b4 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:29 +0200 Subject: net: Remove ctl_table sentinel elements from several networking subsystems This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) To avoid lots of small commits, this commit brings together network changes from (as they appear in MAINTAINERS) LLC, MPTCP, NETROM NETWORK LAYER, PHONET PROTOCOL, ROSE NETWORK LAYER, RXRPC SOCKETS, SCTP PROTOCOL, SHARED MEMORY COMMUNICATIONS (SMC), TIPC NETWORK LAYER and NETWORKING [IPSEC] * Remove sentinel element from ctl_table structs. * Replace empty array registration with the register_net_sysctl_sz call in llc_sysctl_init * Replace the for loop stop condition that tests for procname == NULL with one that depends on array size in sctp_sysctl_net_register * Remove instances where an array element is zeroed out to make it look like a sentinel in xfrm_sysctl_init. This is not longer needed and is safe after commit c899710fe7f9 ("networking: Update to register_net_sysctl_sz") added the array size to the ctl_table registration * Use a table_size variable to keep the value of ARRAY_SIZE Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/llc/sysctl_net_llc.c | 8 ++------ net/mptcp/ctrl.c | 1 - net/netrom/sysctl_net_netrom.c | 1 - net/phonet/sysctl.c | 1 - net/rose/sysctl_net_rose.c | 1 - net/rxrpc/sysctl.c | 1 - net/sctp/sysctl.c | 10 +++------- net/smc/smc_sysctl.c | 6 +++--- net/tipc/sysctl.c | 1 - net/xfrm/xfrm_sysctl.c | 5 +---- 10 files changed, 9 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/net/llc/sysctl_net_llc.c b/net/llc/sysctl_net_llc.c index 8443a6d841b0..72e101135f8c 100644 --- a/net/llc/sysctl_net_llc.c +++ b/net/llc/sysctl_net_llc.c @@ -44,11 +44,6 @@ static struct ctl_table llc2_timeout_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { }, -}; - -static struct ctl_table llc_station_table[] = { - { }, }; static struct ctl_table_header *llc2_timeout_header; @@ -56,8 +51,9 @@ static struct ctl_table_header *llc_station_header; int __init llc_sysctl_init(void) { + struct ctl_table empty[1] = {}; llc2_timeout_header = register_net_sysctl(&init_net, "net/llc/llc2/timeout", llc2_timeout_table); - llc_station_header = register_net_sysctl(&init_net, "net/llc/station", llc_station_table); + llc_station_header = register_net_sysctl_sz(&init_net, "net/llc/station", empty, 0); if (!llc2_timeout_header || !llc_station_header) { llc_sysctl_exit(); diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c index 8d661156ab8c..f4e7a53acc5a 100644 --- a/net/mptcp/ctrl.c +++ b/net/mptcp/ctrl.c @@ -156,7 +156,6 @@ static struct ctl_table mptcp_sysctl_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - {} }; static int mptcp_pernet_new_table(struct net *net, struct mptcp_pernet *pernet) diff --git a/net/netrom/sysctl_net_netrom.c b/net/netrom/sysctl_net_netrom.c index 79fb2d3f477b..7dc0fa628f2e 100644 --- a/net/netrom/sysctl_net_netrom.c +++ b/net/netrom/sysctl_net_netrom.c @@ -140,7 +140,6 @@ static struct ctl_table nr_table[] = { .extra1 = &min_reset, .extra2 = &max_reset }, - { } }; int __init nr_register_sysctl(void) diff --git a/net/phonet/sysctl.c b/net/phonet/sysctl.c index 0d0bf41381c2..82fc22467a09 100644 --- a/net/phonet/sysctl.c +++ b/net/phonet/sysctl.c @@ -81,7 +81,6 @@ static struct ctl_table phonet_table[] = { .mode = 0644, .proc_handler = proc_local_port_range, }, - { } }; int __init phonet_sysctl_init(void) diff --git a/net/rose/sysctl_net_rose.c b/net/rose/sysctl_net_rose.c index d391d7758f52..d801315b7083 100644 --- a/net/rose/sysctl_net_rose.c +++ b/net/rose/sysctl_net_rose.c @@ -112,7 +112,6 @@ static struct ctl_table rose_table[] = { .extra1 = &min_window, .extra2 = &max_window }, - { } }; void __init rose_register_sysctl(void) diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c index c9bedd0e2d86..9bf9a1f6e4cb 100644 --- a/net/rxrpc/sysctl.c +++ b/net/rxrpc/sysctl.c @@ -127,7 +127,6 @@ static struct ctl_table rxrpc_sysctl_table[] = { .extra1 = (void *)SYSCTL_ONE, .extra2 = (void *)&four, }, - { } }; int __init rxrpc_sysctl_init(void) diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 25bdf17c7262..61c6f3027e7f 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -80,8 +80,6 @@ static struct ctl_table sctp_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - - { /* sentinel */ } }; /* The following index defines are used in sctp_sysctl_net_register(). @@ -384,8 +382,6 @@ static struct ctl_table sctp_net_table[] = { .extra1 = SYSCTL_ZERO, .extra2 = &pf_expose_max, }, - - { /* sentinel */ } }; static int proc_sctp_do_hmac_alg(struct ctl_table *ctl, int write, @@ -597,6 +593,7 @@ static int proc_sctp_do_probe_interval(struct ctl_table *ctl, int write, int sctp_sysctl_net_register(struct net *net) { + size_t table_size = ARRAY_SIZE(sctp_net_table); struct ctl_table *table; int i; @@ -604,7 +601,7 @@ int sctp_sysctl_net_register(struct net *net) if (!table) return -ENOMEM; - for (i = 0; table[i].data; i++) + for (i = 0; i < table_size; i++) table[i].data += (char *)(&net->sctp) - (char *)&init_net.sctp; table[SCTP_RTO_MIN_IDX].extra2 = &net->sctp.rto_max; @@ -613,8 +610,7 @@ int sctp_sysctl_net_register(struct net *net) table[SCTP_PS_RETRANS_IDX].extra1 = &net->sctp.pf_retrans; net->sctp.sysctl_header = register_net_sysctl_sz(net, "net/sctp", - table, - ARRAY_SIZE(sctp_net_table)); + table, table_size); if (net->sctp.sysctl_header == NULL) { kfree(table); return -ENOMEM; diff --git a/net/smc/smc_sysctl.c b/net/smc/smc_sysctl.c index 4e8baa2e7ea4..13f2bc092db1 100644 --- a/net/smc/smc_sysctl.c +++ b/net/smc/smc_sysctl.c @@ -90,11 +90,11 @@ static struct ctl_table smc_table[] = { .extra1 = &conns_per_lgr_min, .extra2 = &conns_per_lgr_max, }, - { } }; int __net_init smc_sysctl_net_init(struct net *net) { + size_t table_size = ARRAY_SIZE(smc_table); struct ctl_table *table; table = smc_table; @@ -105,12 +105,12 @@ int __net_init smc_sysctl_net_init(struct net *net) if (!table) goto err_alloc; - for (i = 0; i < ARRAY_SIZE(smc_table) - 1; i++) + for (i = 0; i < table_size; i++) table[i].data += (void *)net - (void *)&init_net; } net->smc.smc_hdr = register_net_sysctl_sz(net, "net/smc", table, - ARRAY_SIZE(smc_table)); + table_size); if (!net->smc.smc_hdr) goto err_reg; diff --git a/net/tipc/sysctl.c b/net/tipc/sysctl.c index 9fb65c988f7f..30d2e06e3d8c 100644 --- a/net/tipc/sysctl.c +++ b/net/tipc/sysctl.c @@ -91,7 +91,6 @@ static struct ctl_table tipc_table[] = { .mode = 0644, .proc_handler = proc_doulongvec_minmax, }, - {} }; int tipc_register_sysctl(void) diff --git a/net/xfrm/xfrm_sysctl.c b/net/xfrm/xfrm_sysctl.c index e972930c292b..ca003e8a0376 100644 --- a/net/xfrm/xfrm_sysctl.c +++ b/net/xfrm/xfrm_sysctl.c @@ -38,7 +38,6 @@ static struct ctl_table xfrm_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - {} }; int __net_init xfrm_sysctl_init(struct net *net) @@ -57,10 +56,8 @@ int __net_init xfrm_sysctl_init(struct net *net) table[3].data = &net->xfrm.sysctl_acq_expires; /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) { - table[0].procname = NULL; + if (net->user_ns != &init_user_ns) table_size = 0; - } net->xfrm.sysctl_hdr = register_net_sysctl_sz(net, "net/core", table, table_size); -- cgit v1.2.3 From 635470eb0aa71ba41c47593c66f65ac1e5d59dd7 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:30 +0200 Subject: netfilter: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) * Remove sentinel elements from ctl_table structs * Remove instances where an array element is zeroed out to make it look like a sentinel. This is not longer needed and is safe after commit c899710fe7f9 ("networking: Update to register_net_sysctl_sz") added the array size to the ctl_table registration * Remove the need for having __NF_SYSCTL_CT_LAST_SYSCTL as the sysctl array size is now in NF_SYSCTL_CT_LAST_SYSCTL * Remove extra element in ctl_table arrays declarations Acked-by: Kees Cook # loadpin & yama Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/bridge/br_netfilter_hooks.c | 1 - net/ipv6/netfilter/nf_conntrack_reasm.c | 1 - net/netfilter/ipvs/ip_vs_ctl.c | 5 +---- net/netfilter/ipvs/ip_vs_lblc.c | 5 +---- net/netfilter/ipvs/ip_vs_lblcr.c | 5 +---- net/netfilter/nf_conntrack_standalone.c | 6 +----- net/netfilter/nf_log.c | 3 +-- 7 files changed, 5 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/bridge/br_netfilter_hooks.c b/net/bridge/br_netfilter_hooks.c index 7948a9e7542c..bf30c50b5689 100644 --- a/net/bridge/br_netfilter_hooks.c +++ b/net/bridge/br_netfilter_hooks.c @@ -1226,7 +1226,6 @@ static struct ctl_table brnf_table[] = { .mode = 0644, .proc_handler = brnf_sysctl_call_tables, }, - { } }; static inline void br_netfilter_sysctl_default(struct brnf_net *brnf) diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index ce8c14d8aff5..5e1b50c6a44d 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -62,7 +62,6 @@ static struct ctl_table nf_ct_frag6_sysctl_table[] = { .mode = 0644, .proc_handler = proc_doulongvec_minmax, }, - { } }; static int nf_ct_frag6_sysctl_register(struct net *net) diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 143a341bbc0a..50b5dbe40eb8 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2263,7 +2263,6 @@ static struct ctl_table vs_vars[] = { .proc_handler = proc_dointvec, }, #endif - { } }; #endif @@ -4286,10 +4285,8 @@ static int __net_init ip_vs_control_net_init_sysctl(struct netns_ipvs *ipvs) return -ENOMEM; /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) { - tbl[0].procname = NULL; + if (net->user_ns != &init_user_ns) ctl_table_size = 0; - } } else tbl = vs_vars; /* Initialize sysctl defaults */ diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 8ceec7a2fa8f..2423513d701d 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -123,7 +123,6 @@ static struct ctl_table vs_vars_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; #endif @@ -563,10 +562,8 @@ static int __net_init __ip_vs_lblc_init(struct net *net) return -ENOMEM; /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) { - ipvs->lblc_ctl_table[0].procname = NULL; + if (net->user_ns != &init_user_ns) vars_table_size = 0; - } } else ipvs->lblc_ctl_table = vs_vars_table; diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index 0fb64707213f..cdb1d4bf6761 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -294,7 +294,6 @@ static struct ctl_table vs_vars_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { } }; #endif @@ -749,10 +748,8 @@ static int __net_init __ip_vs_lblcr_init(struct net *net) return -ENOMEM; /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) { - ipvs->lblcr_ctl_table[0].procname = NULL; + if (net->user_ns != &init_user_ns) vars_table_size = 0; - } } else ipvs->lblcr_ctl_table = vs_vars_table; ipvs->sysctl_lblcr_expiration = DEFAULT_EXPIRATION; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index bb9dea676ec1..74112e9c5dab 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -616,11 +616,9 @@ enum nf_ct_sysctl_index { NF_SYSCTL_CT_LWTUNNEL, #endif - __NF_SYSCTL_CT_LAST_SYSCTL, + NF_SYSCTL_CT_LAST_SYSCTL, }; -#define NF_SYSCTL_CT_LAST_SYSCTL (__NF_SYSCTL_CT_LAST_SYSCTL + 1) - static struct ctl_table nf_ct_sysctl_table[] = { [NF_SYSCTL_CT_MAX] = { .procname = "nf_conntrack_max", @@ -957,7 +955,6 @@ static struct ctl_table nf_ct_sysctl_table[] = { .proc_handler = nf_hooks_lwtunnel_sysctl_handler, }, #endif - {} }; static struct ctl_table nf_ct_netfilter_table[] = { @@ -968,7 +965,6 @@ static struct ctl_table nf_ct_netfilter_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static void nf_conntrack_standalone_init_tcp_sysctl(struct net *net, diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index efedd2f13ac7..769fd7680fac 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -395,7 +395,7 @@ static const struct seq_operations nflog_seq_ops = { #ifdef CONFIG_SYSCTL static char nf_log_sysctl_fnames[NFPROTO_NUMPROTO-NFPROTO_UNSPEC][3]; -static struct ctl_table nf_log_sysctl_table[NFPROTO_NUMPROTO+1]; +static struct ctl_table nf_log_sysctl_table[NFPROTO_NUMPROTO]; static struct ctl_table_header *nf_log_sysctl_fhdr; static struct ctl_table nf_log_sysctl_ftable[] = { @@ -406,7 +406,6 @@ static struct ctl_table nf_log_sysctl_ftable[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { } }; static int nf_log_proc_dostring(struct ctl_table *table, int write, -- cgit v1.2.3 From e00e35e217c0fb5f614cdeb86f28fd3a1c9c936e Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:31 +0200 Subject: appletalk: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) Remove sentinel from atalk_table ctl_table array. Acked-by: Kees Cook # loadpin & yama Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/appletalk/sysctl_net_atalk.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/appletalk/sysctl_net_atalk.c b/net/appletalk/sysctl_net_atalk.c index d945b7c0176d..7aebfe903242 100644 --- a/net/appletalk/sysctl_net_atalk.c +++ b/net/appletalk/sysctl_net_atalk.c @@ -40,7 +40,6 @@ static struct ctl_table atalk_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { }, }; static struct ctl_table_header *atalk_table_header; -- cgit v1.2.3 From 78a7b5dbc0609b8c395f754953a45db687cb05e6 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Wed, 1 May 2024 11:29:32 +0200 Subject: ax.25: x.25: Remove the now superfluous sentinel elements from ctl_table array This commit comes at the tail end of a greater effort to remove the empty elements at the end of the ctl_table arrays (sentinels) which will reduce the overall build time size of the kernel and run time memory bloat by ~64 bytes per sentinel (further information Link : https://lore.kernel.org/all/ZO5Yx5JFogGi%2FcBo@bombadil.infradead.org/) Avoid a buffer overflow when traversing the ctl_table by ensuring that AX25_MAX_VALUES is the same as the size of ax25_param_table. This is done with a BUILD_BUG_ON where ax25_param_table is defined and a CONFIG_AX25_DAMA_SLAVE guard in the unnamed enum definition as well as in the ax25_dev_device_up and ax25_ds_set_timer functions. The overflow happened when the sentinel was removed from ax25_param_table. The sentinel's data element was changed when CONFIG_AX25_DAMA_SLAVE was undefined. This had no adverse effects as it still stopped on the sentinel's null procname but needed to be addressed once the sentinel was removed. Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- include/net/ax25.h | 2 ++ net/ax25/ax25_dev.c | 3 +++ net/ax25/ax25_ds_timer.c | 1 + net/ax25/sysctl_net_ax25.c | 3 +-- net/x25/sysctl_net_x25.c | 1 - 5 files changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/ax25.h b/include/net/ax25.h index 0d939e5aee4e..eb9cee8252c8 100644 --- a/include/net/ax25.h +++ b/include/net/ax25.h @@ -139,7 +139,9 @@ enum { AX25_VALUES_N2, /* Default N2 value */ AX25_VALUES_PACLEN, /* AX.25 MTU */ AX25_VALUES_PROTOCOL, /* Std AX.25, DAMA Slave, DAMA Master */ +#ifdef CONFIG_AX25_DAMA_SLAVE AX25_VALUES_DS_TIMEOUT, /* DAMA Slave timeout */ +#endif AX25_MAX_VALUES /* THIS MUST REMAIN THE LAST ENTRY OF THIS LIST */ }; diff --git a/net/ax25/ax25_dev.c b/net/ax25/ax25_dev.c index 282ec581c072..0bc682ffae9c 100644 --- a/net/ax25/ax25_dev.c +++ b/net/ax25/ax25_dev.c @@ -78,7 +78,10 @@ void ax25_dev_device_up(struct net_device *dev) ax25_dev->values[AX25_VALUES_N2] = AX25_DEF_N2; ax25_dev->values[AX25_VALUES_PACLEN] = AX25_DEF_PACLEN; ax25_dev->values[AX25_VALUES_PROTOCOL] = AX25_DEF_PROTOCOL; + +#ifdef CONFIG_AX25_DAMA_SLAVE ax25_dev->values[AX25_VALUES_DS_TIMEOUT]= AX25_DEF_DS_TIMEOUT; +#endif #if defined(CONFIG_AX25_DAMA_SLAVE) || defined(CONFIG_AX25_DAMA_MASTER) ax25_ds_setup_timer(ax25_dev); diff --git a/net/ax25/ax25_ds_timer.c b/net/ax25/ax25_ds_timer.c index c4f8adbf8144..c50a58d9e368 100644 --- a/net/ax25/ax25_ds_timer.c +++ b/net/ax25/ax25_ds_timer.c @@ -55,6 +55,7 @@ void ax25_ds_set_timer(ax25_dev *ax25_dev) ax25_dev->dama.slave_timeout = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_DS_TIMEOUT]) / 10; mod_timer(&ax25_dev->dama.slave_timer, jiffies + HZ); + return; } /* diff --git a/net/ax25/sysctl_net_ax25.c b/net/ax25/sysctl_net_ax25.c index e0128dc9def3..68753aa30334 100644 --- a/net/ax25/sysctl_net_ax25.c +++ b/net/ax25/sysctl_net_ax25.c @@ -141,8 +141,6 @@ static const struct ctl_table ax25_param_table[] = { .extra2 = &max_ds_timeout }, #endif - - { } /* that's all, folks! */ }; int ax25_register_dev_sysctl(ax25_dev *ax25_dev) @@ -155,6 +153,7 @@ int ax25_register_dev_sysctl(ax25_dev *ax25_dev) if (!table) return -ENOMEM; + BUILD_BUG_ON(ARRAY_SIZE(ax25_param_table) != AX25_MAX_VALUES); for (k = 0; k < AX25_MAX_VALUES; k++) table[k].data = &ax25_dev->values[k]; diff --git a/net/x25/sysctl_net_x25.c b/net/x25/sysctl_net_x25.c index e9802afa43d0..643f50874dfe 100644 --- a/net/x25/sysctl_net_x25.c +++ b/net/x25/sysctl_net_x25.c @@ -71,7 +71,6 @@ static struct ctl_table x25_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, - { }, }; int __init x25_register_sysctl(void) -- cgit v1.2.3 From 483bc08181827fc475643272ffb69c533007e546 Mon Sep 17 00:00:00 2001 From: Duoming Zhou Date: Thu, 25 Apr 2024 22:23:45 +0800 Subject: Bluetooth: Fix use-after-free bugs caused by sco_sock_timeout When the sco connection is established and then, the sco socket is releasing, timeout_work will be scheduled to judge whether the sco disconnection is timeout. The sock will be deallocated later, but it is dereferenced again in sco_sock_timeout. As a result, the use-after-free bugs will happen. The root cause is shown below: Cleanup Thread | Worker Thread sco_sock_release | sco_sock_close | __sco_sock_close | sco_sock_set_timer | schedule_delayed_work | sco_sock_kill | (wait a time) sock_put(sk) //FREE | sco_sock_timeout | sock_hold(sk) //USE The KASAN report triggered by POC is shown below: [ 95.890016] ================================================================== [ 95.890496] BUG: KASAN: slab-use-after-free in sco_sock_timeout+0x5e/0x1c0 [ 95.890755] Write of size 4 at addr ffff88800c388080 by task kworker/0:0/7 ... [ 95.890755] Workqueue: events sco_sock_timeout [ 95.890755] Call Trace: [ 95.890755] [ 95.890755] dump_stack_lvl+0x45/0x110 [ 95.890755] print_address_description+0x78/0x390 [ 95.890755] print_report+0x11b/0x250 [ 95.890755] ? __virt_addr_valid+0xbe/0xf0 [ 95.890755] ? sco_sock_timeout+0x5e/0x1c0 [ 95.890755] kasan_report+0x139/0x170 [ 95.890755] ? update_load_avg+0xe5/0x9f0 [ 95.890755] ? sco_sock_timeout+0x5e/0x1c0 [ 95.890755] kasan_check_range+0x2c3/0x2e0 [ 95.890755] sco_sock_timeout+0x5e/0x1c0 [ 95.890755] process_one_work+0x561/0xc50 [ 95.890755] worker_thread+0xab2/0x13c0 [ 95.890755] ? pr_cont_work+0x490/0x490 [ 95.890755] kthread+0x279/0x300 [ 95.890755] ? pr_cont_work+0x490/0x490 [ 95.890755] ? kthread_blkcg+0xa0/0xa0 [ 95.890755] ret_from_fork+0x34/0x60 [ 95.890755] ? kthread_blkcg+0xa0/0xa0 [ 95.890755] ret_from_fork_asm+0x11/0x20 [ 95.890755] [ 95.890755] [ 95.890755] Allocated by task 506: [ 95.890755] kasan_save_track+0x3f/0x70 [ 95.890755] __kasan_kmalloc+0x86/0x90 [ 95.890755] __kmalloc+0x17f/0x360 [ 95.890755] sk_prot_alloc+0xe1/0x1a0 [ 95.890755] sk_alloc+0x31/0x4e0 [ 95.890755] bt_sock_alloc+0x2b/0x2a0 [ 95.890755] sco_sock_create+0xad/0x320 [ 95.890755] bt_sock_create+0x145/0x320 [ 95.890755] __sock_create+0x2e1/0x650 [ 95.890755] __sys_socket+0xd0/0x280 [ 95.890755] __x64_sys_socket+0x75/0x80 [ 95.890755] do_syscall_64+0xc4/0x1b0 [ 95.890755] entry_SYSCALL_64_after_hwframe+0x67/0x6f [ 95.890755] [ 95.890755] Freed by task 506: [ 95.890755] kasan_save_track+0x3f/0x70 [ 95.890755] kasan_save_free_info+0x40/0x50 [ 95.890755] poison_slab_object+0x118/0x180 [ 95.890755] __kasan_slab_free+0x12/0x30 [ 95.890755] kfree+0xb2/0x240 [ 95.890755] __sk_destruct+0x317/0x410 [ 95.890755] sco_sock_release+0x232/0x280 [ 95.890755] sock_close+0xb2/0x210 [ 95.890755] __fput+0x37f/0x770 [ 95.890755] task_work_run+0x1ae/0x210 [ 95.890755] get_signal+0xe17/0xf70 [ 95.890755] arch_do_signal_or_restart+0x3f/0x520 [ 95.890755] syscall_exit_to_user_mode+0x55/0x120 [ 95.890755] do_syscall_64+0xd1/0x1b0 [ 95.890755] entry_SYSCALL_64_after_hwframe+0x67/0x6f [ 95.890755] [ 95.890755] The buggy address belongs to the object at ffff88800c388000 [ 95.890755] which belongs to the cache kmalloc-1k of size 1024 [ 95.890755] The buggy address is located 128 bytes inside of [ 95.890755] freed 1024-byte region [ffff88800c388000, ffff88800c388400) [ 95.890755] [ 95.890755] The buggy address belongs to the physical page: [ 95.890755] page: refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88800c38a800 pfn:0xc388 [ 95.890755] head: order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0 [ 95.890755] anon flags: 0x100000000000840(slab|head|node=0|zone=1) [ 95.890755] page_type: 0xffffffff() [ 95.890755] raw: 0100000000000840 ffff888006842dc0 0000000000000000 0000000000000001 [ 95.890755] raw: ffff88800c38a800 000000000010000a 00000001ffffffff 0000000000000000 [ 95.890755] head: 0100000000000840 ffff888006842dc0 0000000000000000 0000000000000001 [ 95.890755] head: ffff88800c38a800 000000000010000a 00000001ffffffff 0000000000000000 [ 95.890755] head: 0100000000000003 ffffea000030e201 ffffea000030e248 00000000ffffffff [ 95.890755] head: 0000000800000000 0000000000000000 00000000ffffffff 0000000000000000 [ 95.890755] page dumped because: kasan: bad access detected [ 95.890755] [ 95.890755] Memory state around the buggy address: [ 95.890755] ffff88800c387f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 95.890755] ffff88800c388000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 95.890755] >ffff88800c388080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 95.890755] ^ [ 95.890755] ffff88800c388100: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 95.890755] ffff88800c388180: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb [ 95.890755] ================================================================== Fix this problem by adding a check protected by sco_conn_lock to judget whether the conn->hcon is null. Because the conn->hcon will be set to null, when the sock is releasing. Fixes: ba316be1b6a0 ("Bluetooth: schedule SCO timeouts with delayed_work") Signed-off-by: Duoming Zhou Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index 5d03c5440b06..e0ad30862ee4 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -83,6 +83,10 @@ static void sco_sock_timeout(struct work_struct *work) struct sock *sk; sco_conn_lock(conn); + if (!conn->hcon) { + sco_conn_unlock(conn); + return; + } sk = conn->sk; if (sk) sock_hold(sk); -- cgit v1.2.3 From 4d7b41c0e43995b0e992b9f8903109275744b658 Mon Sep 17 00:00:00 2001 From: Sungwoo Kim Date: Tue, 30 Apr 2024 02:32:10 -0400 Subject: Bluetooth: L2CAP: Fix slab-use-after-free in l2cap_connect() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Extend a critical section to prevent chan from early freeing. Also make the l2cap_connect() return type void. Nothing is using the returned value but it is ugly to return a potentially freed pointer. Making it void will help with backports because earlier kernels did use the return value. Now the compile will break for kernels where this patch is not a complete fix. Call stack summary: [use] l2cap_bredr_sig_cmd l2cap_connect ┌ mutex_lock(&conn->chan_lock); │ chan = pchan->ops->new_connection(pchan); <- alloc chan │ __l2cap_chan_add(conn, chan); │ l2cap_chan_hold(chan); │ list_add(&chan->list, &conn->chan_l); ... (1) └ mutex_unlock(&conn->chan_lock); chan->conf_state ... (4) <- use after free [free] l2cap_conn_del ┌ mutex_lock(&conn->chan_lock); │ foreach chan in conn->chan_l: ... (2) │ l2cap_chan_put(chan); │ l2cap_chan_destroy │ kfree(chan) ... (3) <- chan freed └ mutex_unlock(&conn->chan_lock); ================================================================== BUG: KASAN: slab-use-after-free in instrument_atomic_read include/linux/instrumented.h:68 [inline] BUG: KASAN: slab-use-after-free in _test_bit include/asm-generic/bitops/instrumented-non-atomic.h:141 [inline] BUG: KASAN: slab-use-after-free in l2cap_connect+0xa67/0x11a0 net/bluetooth/l2cap_core.c:4260 Read of size 8 at addr ffff88810bf040a0 by task kworker/u3:1/311 Fixes: 73ffa904b782 ("Bluetooth: Move conf_{req,rsp} stuff to struct l2cap_chan") Signed-off-by: Sungwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 84fc70862d78..868a370a16aa 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -3902,13 +3902,12 @@ static inline int l2cap_command_rej(struct l2cap_conn *conn, return 0; } -static struct l2cap_chan *l2cap_connect(struct l2cap_conn *conn, - struct l2cap_cmd_hdr *cmd, - u8 *data, u8 rsp_code, u8 amp_id) +static void l2cap_connect(struct l2cap_conn *conn, struct l2cap_cmd_hdr *cmd, + u8 *data, u8 rsp_code, u8 amp_id) { struct l2cap_conn_req *req = (struct l2cap_conn_req *) data; struct l2cap_conn_rsp rsp; - struct l2cap_chan *chan = NULL, *pchan; + struct l2cap_chan *chan = NULL, *pchan = NULL; int result, status = L2CAP_CS_NO_INFO; u16 dcid = 0, scid = __le16_to_cpu(req->scid); @@ -3921,7 +3920,7 @@ static struct l2cap_chan *l2cap_connect(struct l2cap_conn *conn, &conn->hcon->dst, ACL_LINK); if (!pchan) { result = L2CAP_CR_BAD_PSM; - goto sendresp; + goto response; } mutex_lock(&conn->chan_lock); @@ -4008,17 +4007,15 @@ static struct l2cap_chan *l2cap_connect(struct l2cap_conn *conn, } response: - l2cap_chan_unlock(pchan); - mutex_unlock(&conn->chan_lock); - l2cap_chan_put(pchan); - -sendresp: rsp.scid = cpu_to_le16(scid); rsp.dcid = cpu_to_le16(dcid); rsp.result = cpu_to_le16(result); rsp.status = cpu_to_le16(status); l2cap_send_cmd(conn, cmd->ident, rsp_code, sizeof(rsp), &rsp); + if (!pchan) + return; + if (result == L2CAP_CR_PEND && status == L2CAP_CS_NO_INFO) { struct l2cap_info_req info; info.type = cpu_to_le16(L2CAP_IT_FEAT_MASK); @@ -4041,7 +4038,9 @@ sendresp: chan->num_conf_req++; } - return chan; + l2cap_chan_unlock(pchan); + mutex_unlock(&conn->chan_lock); + l2cap_chan_put(pchan); } static int l2cap_connect_req(struct l2cap_conn *conn, -- cgit v1.2.3 From 10f9f426ac6e752c8d87bf4346930ba347aaabac Mon Sep 17 00:00:00 2001 From: Sungwoo Kim Date: Tue, 30 Apr 2024 12:20:51 -0400 Subject: Bluetooth: msft: fix slab-use-after-free in msft_do_close() Tying the msft->data lifetime to hdev by freeing it in hci_release_dev() to fix the following case: [use] msft_do_close() msft = hdev->msft_data; if (!msft) ...(1) <- passed. return; mutex_lock(&msft->filter_lock); ...(4) <- used after freed. [free] msft_unregister() msft = hdev->msft_data; hdev->msft_data = NULL; ...(2) kfree(msft); ...(3) <- msft is freed. ================================================================== BUG: KASAN: slab-use-after-free in __mutex_lock_common kernel/locking/mutex.c:587 [inline] BUG: KASAN: slab-use-after-free in __mutex_lock+0x8f/0xc30 kernel/locking/mutex.c:752 Read of size 8 at addr ffff888106cbbca8 by task kworker/u5:2/309 Fixes: bf6a4e30ffbd ("Bluetooth: disable advertisement filters during suspend") Signed-off-by: Sungwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 3 +-- net/bluetooth/msft.c | 2 +- net/bluetooth/msft.h | 4 ++-- 3 files changed, 4 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index a7028d38c1f5..bc5086423ab8 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -2768,8 +2768,6 @@ void hci_unregister_dev(struct hci_dev *hdev) hci_unregister_suspend_notifier(hdev); - msft_unregister(hdev); - hci_dev_do_close(hdev); if (!test_bit(HCI_INIT, &hdev->flags) && @@ -2823,6 +2821,7 @@ void hci_release_dev(struct hci_dev *hdev) hci_discovery_filter_clear(hdev); hci_blocked_keys_clear(hdev); hci_codec_list_clear(&hdev->local_codecs); + msft_release(hdev); hci_dev_unlock(hdev); ida_destroy(&hdev->unset_handle_ida); diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c index 9612c5d1b13f..d039683d3bdd 100644 --- a/net/bluetooth/msft.c +++ b/net/bluetooth/msft.c @@ -769,7 +769,7 @@ void msft_register(struct hci_dev *hdev) mutex_init(&msft->filter_lock); } -void msft_unregister(struct hci_dev *hdev) +void msft_release(struct hci_dev *hdev) { struct msft_data *msft = hdev->msft_data; diff --git a/net/bluetooth/msft.h b/net/bluetooth/msft.h index 2a63205b377b..fe538e9c91c0 100644 --- a/net/bluetooth/msft.h +++ b/net/bluetooth/msft.h @@ -14,7 +14,7 @@ bool msft_monitor_supported(struct hci_dev *hdev); void msft_register(struct hci_dev *hdev); -void msft_unregister(struct hci_dev *hdev); +void msft_release(struct hci_dev *hdev); void msft_do_open(struct hci_dev *hdev); void msft_do_close(struct hci_dev *hdev); void msft_vendor_evt(struct hci_dev *hdev, void *data, struct sk_buff *skb); @@ -35,7 +35,7 @@ static inline bool msft_monitor_supported(struct hci_dev *hdev) } static inline void msft_register(struct hci_dev *hdev) {} -static inline void msft_unregister(struct hci_dev *hdev) {} +static inline void msft_release(struct hci_dev *hdev) {} static inline void msft_do_open(struct hci_dev *hdev) {} static inline void msft_do_close(struct hci_dev *hdev) {} static inline void msft_vendor_evt(struct hci_dev *hdev, void *data, -- cgit v1.2.3 From d2706004a1b8b526592e823d7e52551b518a7941 Mon Sep 17 00:00:00 2001 From: Sungwoo Kim Date: Thu, 2 May 2024 12:09:31 -0400 Subject: Bluetooth: HCI: Fix potential null-ptr-deref Fix potential null-ptr-deref in hci_le_big_sync_established_evt(). Fixes: f777d8827817 (Bluetooth: ISO: Notify user space about failed bis connections) Signed-off-by: Sungwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_event.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 4a27e4a17a67..d72d238c1656 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -7037,6 +7037,8 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data, u16 handle = le16_to_cpu(ev->bis[i]); bis = hci_conn_hash_lookup_handle(hdev, handle); + if (!bis) + continue; set_bit(HCI_CONN_BIG_SYNC_FAILED, &bis->flags); hci_connect_cfm(bis, ev->status); -- cgit v1.2.3 From adf0398cee86643b8eacde95f17d073d022f782c Mon Sep 17 00:00:00 2001 From: Duoming Zhou Date: Thu, 2 May 2024 20:57:36 +0800 Subject: Bluetooth: l2cap: fix null-ptr-deref in l2cap_chan_timeout There is a race condition between l2cap_chan_timeout() and l2cap_chan_del(). When we use l2cap_chan_del() to delete the channel, the chan->conn will be set to null. But the conn could be dereferenced again in the mutex_lock() of l2cap_chan_timeout(). As a result the null pointer dereference bug will happen. The KASAN report triggered by POC is shown below: [ 472.074580] ================================================================== [ 472.075284] BUG: KASAN: null-ptr-deref in mutex_lock+0x68/0xc0 [ 472.075308] Write of size 8 at addr 0000000000000158 by task kworker/0:0/7 [ 472.075308] [ 472.075308] CPU: 0 PID: 7 Comm: kworker/0:0 Not tainted 6.9.0-rc5-00356-g78c0094a146b #36 [ 472.075308] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu4 [ 472.075308] Workqueue: events l2cap_chan_timeout [ 472.075308] Call Trace: [ 472.075308] [ 472.075308] dump_stack_lvl+0x137/0x1a0 [ 472.075308] print_report+0x101/0x250 [ 472.075308] ? __virt_addr_valid+0x77/0x160 [ 472.075308] ? mutex_lock+0x68/0xc0 [ 472.075308] kasan_report+0x139/0x170 [ 472.075308] ? mutex_lock+0x68/0xc0 [ 472.075308] kasan_check_range+0x2c3/0x2e0 [ 472.075308] mutex_lock+0x68/0xc0 [ 472.075308] l2cap_chan_timeout+0x181/0x300 [ 472.075308] process_one_work+0x5d2/0xe00 [ 472.075308] worker_thread+0xe1d/0x1660 [ 472.075308] ? pr_cont_work+0x5e0/0x5e0 [ 472.075308] kthread+0x2b7/0x350 [ 472.075308] ? pr_cont_work+0x5e0/0x5e0 [ 472.075308] ? kthread_blkcg+0xd0/0xd0 [ 472.075308] ret_from_fork+0x4d/0x80 [ 472.075308] ? kthread_blkcg+0xd0/0xd0 [ 472.075308] ret_from_fork_asm+0x11/0x20 [ 472.075308] [ 472.075308] ================================================================== [ 472.094860] Disabling lock debugging due to kernel taint [ 472.096136] BUG: kernel NULL pointer dereference, address: 0000000000000158 [ 472.096136] #PF: supervisor write access in kernel mode [ 472.096136] #PF: error_code(0x0002) - not-present page [ 472.096136] PGD 0 P4D 0 [ 472.096136] Oops: 0002 [#1] PREEMPT SMP KASAN NOPTI [ 472.096136] CPU: 0 PID: 7 Comm: kworker/0:0 Tainted: G B 6.9.0-rc5-00356-g78c0094a146b #36 [ 472.096136] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu4 [ 472.096136] Workqueue: events l2cap_chan_timeout [ 472.096136] RIP: 0010:mutex_lock+0x88/0xc0 [ 472.096136] Code: be 08 00 00 00 e8 f8 23 1f fd 4c 89 f7 be 08 00 00 00 e8 eb 23 1f fd 42 80 3c 23 00 74 08 48 88 [ 472.096136] RSP: 0018:ffff88800744fc78 EFLAGS: 00000246 [ 472.096136] RAX: 0000000000000000 RBX: 1ffff11000e89f8f RCX: ffffffff8457c865 [ 472.096136] RDX: 0000000000000001 RSI: 0000000000000008 RDI: ffff88800744fc78 [ 472.096136] RBP: 0000000000000158 R08: ffff88800744fc7f R09: 1ffff11000e89f8f [ 472.096136] R10: dffffc0000000000 R11: ffffed1000e89f90 R12: dffffc0000000000 [ 472.096136] R13: 0000000000000158 R14: ffff88800744fc78 R15: ffff888007405a00 [ 472.096136] FS: 0000000000000000(0000) GS:ffff88806d200000(0000) knlGS:0000000000000000 [ 472.096136] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 472.096136] CR2: 0000000000000158 CR3: 000000000da32000 CR4: 00000000000006f0 [ 472.096136] Call Trace: [ 472.096136] [ 472.096136] ? __die_body+0x8d/0xe0 [ 472.096136] ? page_fault_oops+0x6b8/0x9a0 [ 472.096136] ? kernelmode_fixup_or_oops+0x20c/0x2a0 [ 472.096136] ? do_user_addr_fault+0x1027/0x1340 [ 472.096136] ? _printk+0x7a/0xa0 [ 472.096136] ? mutex_lock+0x68/0xc0 [ 472.096136] ? add_taint+0x42/0xd0 [ 472.096136] ? exc_page_fault+0x6a/0x1b0 [ 472.096136] ? asm_exc_page_fault+0x26/0x30 [ 472.096136] ? mutex_lock+0x75/0xc0 [ 472.096136] ? mutex_lock+0x88/0xc0 [ 472.096136] ? mutex_lock+0x75/0xc0 [ 472.096136] l2cap_chan_timeout+0x181/0x300 [ 472.096136] process_one_work+0x5d2/0xe00 [ 472.096136] worker_thread+0xe1d/0x1660 [ 472.096136] ? pr_cont_work+0x5e0/0x5e0 [ 472.096136] kthread+0x2b7/0x350 [ 472.096136] ? pr_cont_work+0x5e0/0x5e0 [ 472.096136] ? kthread_blkcg+0xd0/0xd0 [ 472.096136] ret_from_fork+0x4d/0x80 [ 472.096136] ? kthread_blkcg+0xd0/0xd0 [ 472.096136] ret_from_fork_asm+0x11/0x20 [ 472.096136] [ 472.096136] Modules linked in: [ 472.096136] CR2: 0000000000000158 [ 472.096136] ---[ end trace 0000000000000000 ]--- [ 472.096136] RIP: 0010:mutex_lock+0x88/0xc0 [ 472.096136] Code: be 08 00 00 00 e8 f8 23 1f fd 4c 89 f7 be 08 00 00 00 e8 eb 23 1f fd 42 80 3c 23 00 74 08 48 88 [ 472.096136] RSP: 0018:ffff88800744fc78 EFLAGS: 00000246 [ 472.096136] RAX: 0000000000000000 RBX: 1ffff11000e89f8f RCX: ffffffff8457c865 [ 472.096136] RDX: 0000000000000001 RSI: 0000000000000008 RDI: ffff88800744fc78 [ 472.096136] RBP: 0000000000000158 R08: ffff88800744fc7f R09: 1ffff11000e89f8f [ 472.132932] R10: dffffc0000000000 R11: ffffed1000e89f90 R12: dffffc0000000000 [ 472.132932] R13: 0000000000000158 R14: ffff88800744fc78 R15: ffff888007405a00 [ 472.132932] FS: 0000000000000000(0000) GS:ffff88806d200000(0000) knlGS:0000000000000000 [ 472.132932] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 472.132932] CR2: 0000000000000158 CR3: 000000000da32000 CR4: 00000000000006f0 [ 472.132932] Kernel panic - not syncing: Fatal exception [ 472.132932] Kernel Offset: disabled [ 472.132932] ---[ end Kernel panic - not syncing: Fatal exception ]--- Add a check to judge whether the conn is null in l2cap_chan_timeout() in order to mitigate the bug. Fixes: 3df91ea20e74 ("Bluetooth: Revert to mutexes from RCU list") Signed-off-by: Duoming Zhou Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net') diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 868a370a16aa..9223b1a698e3 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -415,6 +415,9 @@ static void l2cap_chan_timeout(struct work_struct *work) BT_DBG("chan %p state %s", chan, state_to_string(chan->state)); + if (!conn) + return; + mutex_lock(&conn->chan_lock); /* __set_chan_timer() calls l2cap_chan_hold(chan) while scheduling * this work. No need to call l2cap_chan_hold(chan) here again. -- cgit v1.2.3 From 136c2a9a2a8760d8dae83ae7c882c50be02bdb63 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 2 May 2024 11:37:47 +0000 Subject: rtnetlink: change rtnl_stats_dump() return value By returning 0 (or an error) instead of skb->len, we allow NLMSG_DONE to be appended to the current skb at the end of a dump, saving a couple of recvmsg() system calls. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240502113748.1622637-2-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 283e42f48af6..88980c8bcf33 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -6024,7 +6024,7 @@ out: cb->args[1] = idx; cb->args[0] = h; - return skb->len; + return err; } void rtnl_offload_xstats_notify(struct net_device *dev) -- cgit v1.2.3 From 0feb396f7428b95710ea72c1dc33ae363019fae5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 2 May 2024 11:37:48 +0000 Subject: rtnetlink: use for_each_netdev_dump() in rtnl_stats_dump() Switch rtnl_stats_dump() to use for_each_netdev_dump() instead of net->dev_index_head[] hash table. This makes the code much easier to read, and fixes scalability issues. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240502113748.1622637-3-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 59 ++++++++++++++++++++-------------------------------- 1 file changed, 22 insertions(+), 37 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 88980c8bcf33..28050e53ecb0 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -5961,19 +5961,17 @@ static int rtnl_stats_get(struct sk_buff *skb, struct nlmsghdr *nlh, static int rtnl_stats_dump(struct sk_buff *skb, struct netlink_callback *cb) { struct netlink_ext_ack *extack = cb->extack; - int h, s_h, err, s_idx, s_idxattr, s_prividx; struct rtnl_stats_dump_filters filters; struct net *net = sock_net(skb->sk); unsigned int flags = NLM_F_MULTI; struct if_stats_msg *ifsm; - struct hlist_head *head; + struct { + unsigned long ifindex; + int idxattr; + int prividx; + } *ctx = (void *)cb->ctx; struct net_device *dev; - int idx = 0; - - s_h = cb->args[0]; - s_idx = cb->args[1]; - s_idxattr = cb->args[2]; - s_prividx = cb->args[3]; + int err; cb->seq = net->dev_base_seq; @@ -5992,37 +5990,24 @@ static int rtnl_stats_dump(struct sk_buff *skb, struct netlink_callback *cb) if (err) return err; - for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) { - idx = 0; - head = &net->dev_index_head[h]; - hlist_for_each_entry(dev, head, index_hlist) { - if (idx < s_idx) - goto cont; - err = rtnl_fill_statsinfo(skb, dev, RTM_NEWSTATS, - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, 0, - flags, &filters, - &s_idxattr, &s_prividx, - extack); - /* If we ran out of room on the first message, - * we're in trouble - */ - WARN_ON((err == -EMSGSIZE) && (skb->len == 0)); + for_each_netdev_dump(net, dev, ctx->ifindex) { + err = rtnl_fill_statsinfo(skb, dev, RTM_NEWSTATS, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, 0, + flags, &filters, + &ctx->idxattr, &ctx->prividx, + extack); + /* If we ran out of room on the first message, + * we're in trouble. + */ + WARN_ON((err == -EMSGSIZE) && (skb->len == 0)); - if (err < 0) - goto out; - s_prividx = 0; - s_idxattr = 0; - nl_dump_check_consistent(cb, nlmsg_hdr(skb)); -cont: - idx++; - } + if (err < 0) + break; + ctx->prividx = 0; + ctx->idxattr = 0; + nl_dump_check_consistent(cb, nlmsg_hdr(skb)); } -out: - cb->args[3] = s_prividx; - cb->args[2] = s_idxattr; - cb->args[1] = idx; - cb->args[0] = h; return err; } -- cgit v1.2.3 From c1742dcb6bda5fd535fbaa2145f0a180bc329aa6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 2 May 2024 17:39:26 +0000 Subject: net: no longer acquire RTNL in threaded_show() dev->threaded can be read locklessly, if we add corresponding READ_ONCE()/WRITE_ONCE() annotations. Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20240502173926.2010646-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/linux/netdevice.h | 2 +- net/core/dev.c | 4 ++-- net/core/net-sysfs.c | 8 ++++---- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 41853424b41d..2814a15eed73 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2370,8 +2370,8 @@ struct net_device { struct sfp_bus *sfp_bus; struct lock_class_key *qdisc_tx_busylock; bool proto_down; + bool threaded; unsigned wol_enabled:1; - unsigned threaded:1; struct list_head net_notifier_list; diff --git a/net/core/dev.c b/net/core/dev.c index e02d2363347e..d6b24749eb2e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6531,7 +6531,7 @@ int dev_set_threaded(struct net_device *dev, bool threaded) } } - dev->threaded = threaded; + WRITE_ONCE(dev->threaded, threaded); /* Make sure kthread is created before THREADED bit * is set. @@ -6622,7 +6622,7 @@ void netif_napi_add_weight(struct net_device *dev, struct napi_struct *napi, * threaded mode will not be enabled in napi_enable(). */ if (dev->threaded && napi_kthread_create(napi)) - dev->threaded = 0; + dev->threaded = false; netif_napi_set_irq(napi, -1); } EXPORT_SYMBOL(netif_napi_add_weight); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 1f7f09e56771..4c27a360c294 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -605,13 +605,13 @@ static ssize_t threaded_show(struct device *dev, struct net_device *netdev = to_net_dev(dev); ssize_t ret = -EINVAL; - if (!rtnl_trylock()) - return restart_syscall(); + rcu_read_lock(); if (dev_isalive(netdev)) - ret = sysfs_emit(buf, fmt_dec, netdev->threaded); + ret = sysfs_emit(buf, fmt_dec, READ_ONCE(netdev->threaded)); + + rcu_read_unlock(); - rtnl_unlock(); return ret; } -- cgit v1.2.3 From 1aec77b2bb2ed1db0f5efc61c4c1ca3813307489 Mon Sep 17 00:00:00 2001 From: Roded Zats Date: Thu, 2 May 2024 18:57:51 +0300 Subject: rtnetlink: Correct nested IFLA_VF_VLAN_LIST attribute validation Each attribute inside a nested IFLA_VF_VLAN_LIST is assumed to be a struct ifla_vf_vlan_info so the size of such attribute needs to be at least of sizeof(struct ifla_vf_vlan_info) which is 14 bytes. The current size validation in do_setvfinfo is against NLA_HDRLEN (4 bytes) which is less than sizeof(struct ifla_vf_vlan_info) so this validation is not enough and a too small attribute might be cast to a struct ifla_vf_vlan_info, this might result in an out of bands read access when accessing the saved (casted) entry in ivvl. Fixes: 79aab093a0b5 ("net: Update API for VF vlan protocol 802.1ad support") Signed-off-by: Roded Zats Reviewed-by: Donald Hunter Link: https://lore.kernel.org/r/20240502155751.75705-1-rzats@paloaltonetworks.com Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a3d7847ce69d..8ba6a4e4be26 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2530,7 +2530,7 @@ static int do_setvfinfo(struct net_device *dev, struct nlattr **tb) nla_for_each_nested(attr, tb[IFLA_VF_VLAN_LIST], rem) { if (nla_type(attr) != IFLA_VF_VLAN_INFO || - nla_len(attr) < NLA_HDRLEN) { + nla_len(attr) < sizeof(struct ifla_vf_vlan_info)) { return -EINVAL; } if (len >= MAX_VLAN_LIST_LEN) -- cgit v1.2.3 From 173e7622ccb3f46834bd4176ed363f435e142942 Mon Sep 17 00:00:00 2001 From: Mina Almasry Date: Thu, 2 May 2024 10:54:22 -0700 Subject: Revert "net: mirror skb frag ref/unref helpers" This reverts commit a580ea994fd37f4105028f5a85c38ff6508a2b25. This revert is to resolve Dragos's report of page_pool leak here: https://lore.kernel.org/lkml/20240424165646.1625690-2-dtatulea@nvidia.com/ The reverted patch interacts very badly with commit 2cc3aeb5eccc ("skbuff: Fix a potential race while recycling page_pool packets"). The reverted commit hopes that the pp_recycle + is_pp_page variables do not change between the skb_frag_ref and skb_frag_unref operation. If such a change occurs, the skb_frag_ref/unref will not operate on the same reference type. In the case of Dragos's report, the grabbed ref was a pp ref, but the unref was a page ref, because the pp_recycle setting on the skb was changed. Attempting to fix this issue on the fly is risky. Lets revert and I hope to reland this with better understanding and testing to ensure we don't regress some edge case while streamlining skb reffing. Fixes: a580ea994fd3 ("net: mirror skb frag ref/unref helpers") Reported-by: Dragos Tatulea Signed-off-by: Mina Almasry Link: https://lore.kernel.org/r/20240502175423.2456544-1-almasrymina@google.com Signed-off-by: Jakub Kicinski --- .../chelsio/inline_crypto/ch_ktls/chcr_ktls.c | 2 +- drivers/net/ethernet/sun/cassini.c | 4 +- drivers/net/veth.c | 2 +- include/linux/skbuff_ref.h | 39 ++---------------- net/core/skbuff.c | 46 ++++++++++++++++++++-- net/tls/tls_device_fallback.c | 2 +- 6 files changed, 51 insertions(+), 44 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c index 3832c2e8ea5a..e8e460a92e0e 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c @@ -1659,7 +1659,7 @@ static void chcr_ktls_copy_record_in_skb(struct sk_buff *nskb, for (i = 0; i < record->num_frags; i++) { skb_shinfo(nskb)->frags[i] = record->frags[i]; /* increase the frag ref count */ - __skb_frag_ref(&skb_shinfo(nskb)->frags[i], nskb->pp_recycle); + __skb_frag_ref(&skb_shinfo(nskb)->frags[i]); } skb_shinfo(nskb)->nr_frags = record->num_frags; diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index f058e154a3bc..8f1f43dbb76d 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -2000,7 +2000,7 @@ static int cas_rx_process_pkt(struct cas *cp, struct cas_rx_comp *rxc, skb->len += hlen - swivel; skb_frag_fill_page_desc(frag, page->buffer, off, hlen - swivel); - __skb_frag_ref(frag, skb->pp_recycle); + __skb_frag_ref(frag); /* any more data? */ if ((words[0] & RX_COMP1_SPLIT_PKT) && ((dlen -= hlen) > 0)) { @@ -2024,7 +2024,7 @@ static int cas_rx_process_pkt(struct cas *cp, struct cas_rx_comp *rxc, frag++; skb_frag_fill_page_desc(frag, page->buffer, 0, hlen); - __skb_frag_ref(frag, skb->pp_recycle); + __skb_frag_ref(frag); RX_USED_ADD(page, hlen + cp->crc_size); } diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 0b0293629329..426e68a95067 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -717,7 +717,7 @@ static void veth_xdp_get(struct xdp_buff *xdp) return; for (i = 0; i < sinfo->nr_frags; i++) - __skb_frag_ref(&sinfo->frags[i], false); + __skb_frag_ref(&sinfo->frags[i]); } static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq, diff --git a/include/linux/skbuff_ref.h b/include/linux/skbuff_ref.h index 4dcdbe9fbc5f..11f0a4063403 100644 --- a/include/linux/skbuff_ref.h +++ b/include/linux/skbuff_ref.h @@ -8,47 +8,16 @@ #define _LINUX_SKBUFF_REF_H #include -#include - -#ifdef CONFIG_PAGE_POOL -static inline bool is_pp_page(struct page *page) -{ - return (page->pp_magic & ~0x3UL) == PP_SIGNATURE; -} - -static inline bool napi_pp_get_page(struct page *page) -{ - page = compound_head(page); - - if (!is_pp_page(page)) - return false; - - page_pool_ref_page(page); - return true; -} -#endif - -static inline void skb_page_ref(struct page *page, bool recycle) -{ -#ifdef CONFIG_PAGE_POOL - if (recycle && napi_pp_get_page(page)) - return; -#endif - get_page(page); -} /** * __skb_frag_ref - take an addition reference on a paged fragment. * @frag: the paged fragment - * @recycle: skb->pp_recycle param of the parent skb. False if no parent skb. * - * Takes an additional reference on the paged fragment @frag. Obtains the - * correct reference count depending on whether skb->pp_recycle is set and - * whether the frag is a page pool frag. + * Takes an additional reference on the paged fragment @frag. */ -static inline void __skb_frag_ref(skb_frag_t *frag, bool recycle) +static inline void __skb_frag_ref(skb_frag_t *frag) { - skb_page_ref(skb_frag_page(frag), recycle); + get_page(skb_frag_page(frag)); } /** @@ -60,7 +29,7 @@ static inline void __skb_frag_ref(skb_frag_t *frag, bool recycle) */ static inline void skb_frag_ref(struct sk_buff *skb, int f) { - __skb_frag_ref(&skb_shinfo(skb)->frags[f], skb->pp_recycle); + __skb_frag_ref(&skb_shinfo(skb)->frags[f]); } bool napi_pp_put_page(struct page *page); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 28cd640a6ea9..466999a7515e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -904,6 +904,11 @@ static void skb_clone_fraglist(struct sk_buff *skb) skb_get(list); } +static bool is_pp_page(struct page *page) +{ + return (page->pp_magic & ~0x3UL) == PP_SIGNATURE; +} + int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb, unsigned int headroom) { @@ -1025,6 +1030,37 @@ static bool skb_pp_recycle(struct sk_buff *skb, void *data) return napi_pp_put_page(virt_to_page(data)); } +/** + * skb_pp_frag_ref() - Increase fragment references of a page pool aware skb + * @skb: page pool aware skb + * + * Increase the fragment reference count (pp_ref_count) of a skb. This is + * intended to gain fragment references only for page pool aware skbs, + * i.e. when skb->pp_recycle is true, and not for fragments in a + * non-pp-recycling skb. It has a fallback to increase references on normal + * pages, as page pool aware skbs may also have normal page fragments. + */ +static int skb_pp_frag_ref(struct sk_buff *skb) +{ + struct skb_shared_info *shinfo; + struct page *head_page; + int i; + + if (!skb->pp_recycle) + return -EINVAL; + + shinfo = skb_shinfo(skb); + + for (i = 0; i < shinfo->nr_frags; i++) { + head_page = compound_head(skb_frag_page(&shinfo->frags[i])); + if (likely(is_pp_page(head_page))) + page_pool_ref_page(head_page); + else + page_ref_inc(head_page); + } + return 0; +} + static void skb_kfree_head(void *head, unsigned int end_offset) { if (end_offset == SKB_SMALL_HEAD_HEADROOM) @@ -4160,7 +4196,7 @@ int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen) to++; } else { - __skb_frag_ref(fragfrom, skb->pp_recycle); + __skb_frag_ref(fragfrom); skb_frag_page_copy(fragto, fragfrom); skb_frag_off_copy(fragto, fragfrom); skb_frag_size_set(fragto, todo); @@ -4810,7 +4846,7 @@ normal: } *nskb_frag = (i < 0) ? skb_head_frag_to_page_desc(frag_skb) : *frag; - __skb_frag_ref(nskb_frag, nskb->pp_recycle); + __skb_frag_ref(nskb_frag); size = skb_frag_size(nskb_frag); if (pos < offset) { @@ -5941,8 +5977,10 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, /* if the skb is not cloned this does nothing * since we set nr_frags to 0. */ - for (i = 0; i < from_shinfo->nr_frags; i++) - __skb_frag_ref(&from_shinfo->frags[i], from->pp_recycle); + if (skb_pp_frag_ref(from)) { + for (i = 0; i < from_shinfo->nr_frags; i++) + __skb_frag_ref(&from_shinfo->frags[i]); + } to->truesize += delta; to->len += len; diff --git a/net/tls/tls_device_fallback.c b/net/tls/tls_device_fallback.c index 9237dded4467..f9e3d3d90dcf 100644 --- a/net/tls/tls_device_fallback.c +++ b/net/tls/tls_device_fallback.c @@ -278,7 +278,7 @@ static int fill_sg_in(struct scatterlist *sg_in, for (i = 0; remaining > 0; i++) { skb_frag_t *frag = &record->frags[i]; - __skb_frag_ref(frag, false); + __skb_frag_ref(frag); sg_set_page(sg_in + i, skb_frag_page(frag), skb_frag_size(frag), skb_frag_off(frag)); -- cgit v1.2.3 From 84b1a0c0140a9a92ea108576c0002210f224ce59 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 5 Mar 2024 09:35:48 +0100 Subject: netfilter: nf_tables: skip transaction if update object is not implemented Turn update into noop as a follow up for: 9fedd894b4e1 ("netfilter: nf_tables: fix unexpected EOPNOTSUPP error") instead of adding a transaction object which is simply discarded at a later stage of the commit protocol. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 167074283ea9..a7f54eb68d9a 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -7776,6 +7776,9 @@ static int nf_tables_newobj(struct sk_buff *skb, const struct nfnl_info *info, if (WARN_ON_ONCE(!type)) return -ENOENT; + if (!obj->ops->update) + return 0; + nft_ctx_init(&ctx, net, skb, info->nlh, family, table, NULL, nla); return nf_tables_updobj(&ctx, type, nla[NFTA_OBJ_DATA], obj); @@ -9467,9 +9470,10 @@ static void nft_obj_commit_update(struct nft_trans *trans) obj = nft_trans_obj(trans); newobj = nft_trans_obj_newobj(trans); - if (obj->ops->update) - obj->ops->update(obj, newobj); + if (WARN_ON_ONCE(!obj->ops->update)) + return; + obj->ops->update(obj, newobj); nft_obj_destroy(&trans->ctx, newobj); } -- cgit v1.2.3 From 6e20eef413d5aa8ea0b19165e40efc8d47c681db Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 5 Mar 2024 09:38:04 +0100 Subject: netfilter: nf_tables: remove NETDEV_CHANGENAME from netdev chain event handler Originally, device name used to be stored in the basechain, but it is not the case anymore. Remove check for NETDEV_CHANGENAME. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_chain_filter.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_chain_filter.c b/net/netfilter/nft_chain_filter.c index d170758a1eb5..7010541fcca6 100644 --- a/net/netfilter/nft_chain_filter.c +++ b/net/netfilter/nft_chain_filter.c @@ -325,9 +325,6 @@ static void nft_netdev_event(unsigned long event, struct net_device *dev, struct nft_hook *hook, *found = NULL; int n = 0; - if (event != NETDEV_UNREGISTER) - return; - list_for_each_entry(hook, &basechain->hook_list, list) { if (hook->ops.dev == dev) found = hook; @@ -367,8 +364,7 @@ static int nf_tables_netdev_event(struct notifier_block *this, .net = dev_net(dev), }; - if (event != NETDEV_UNREGISTER && - event != NETDEV_CHANGENAME) + if (event != NETDEV_UNREGISTER) return NOTIFY_DONE; nft_net = nft_pernet(ctx.net); -- cgit v1.2.3 From 4a3540a8bf3c13dc3955f0c0895332b9c653be3f Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Wed, 6 Mar 2024 15:18:04 +0100 Subject: netfilter: conntrack: fix ct-state for ICMPv6 Multicast Router Discovery MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit So far Multicast Router Advertisements and Multicast Router Solicitations from the Multicast Router Discovery protocol (RFC4286) would be marked as INVALID for IPv6, even if they are in fact intact and adhering to RFC4286. This broke MRA reception and by that multicast reception on IPv6 multicast routers in a Proxmox managed setup, where Proxmox would install a rule like "-m conntrack --ctstate INVALID -j DROP" at the top of the FORWARD chain with br-nf-call-ip6tables enabled by default. Similar to as it's done for MLDv1, MLDv2 and IPv6 Neighbor Discovery already, fix this issue by excluding MRD from connection tracking handling as MRD always uses predefined multicast destinations for its messages, too. This changes the ct-state for ICMPv6 MRD messages from INVALID to UNTRACKED. This issue was found and fixed with the help of the mrdisc tool (https://github.com/troglobit/mrdisc). Signed-off-by: Linus Lüssing Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/icmpv6.h | 1 + net/netfilter/nf_conntrack_proto_icmpv6.c | 4 +++- 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/uapi/linux/icmpv6.h b/include/uapi/linux/icmpv6.h index ecaece3af38d..4eaab89e2856 100644 --- a/include/uapi/linux/icmpv6.h +++ b/include/uapi/linux/icmpv6.h @@ -112,6 +112,7 @@ struct icmp6hdr { #define ICMPV6_MOBILE_PREFIX_ADV 147 #define ICMPV6_MRDISC_ADV 151 +#define ICMPV6_MRDISC_SOL 152 #define ICMPV6_MSG_MAX 255 diff --git a/net/netfilter/nf_conntrack_proto_icmpv6.c b/net/netfilter/nf_conntrack_proto_icmpv6.c index 1020d67600a9..327b8059025d 100644 --- a/net/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/netfilter/nf_conntrack_proto_icmpv6.c @@ -62,7 +62,9 @@ static const u_int8_t noct_valid_new[] = { [NDISC_ROUTER_ADVERTISEMENT - 130] = 1, [NDISC_NEIGHBOUR_SOLICITATION - 130] = 1, [NDISC_NEIGHBOUR_ADVERTISEMENT - 130] = 1, - [ICMPV6_MLD2_REPORT - 130] = 1 + [ICMPV6_MLD2_REPORT - 130] = 1, + [ICMPV6_MRDISC_ADV - 130] = 1, + [ICMPV6_MRDISC_SOL - 130] = 1 }; bool nf_conntrack_invert_icmpv6_tuple(struct nf_conntrack_tuple *tuple, -- cgit v1.2.3 From 40616789ec462df0e97ee1d5aa4e4fed4249cae0 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Mon, 25 Mar 2024 10:59:38 +0800 Subject: netfilter: conntrack: dccp: try not to drop skb in conntrack It would be better not to drop skb in conntrack unless we have good alternatives. So we can treat the result of testing skb's header pointer as nf_conntrack_tcp_packet() does. Signed-off-by: Jason Xing Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_proto_dccp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index e2db1f4ec2df..ebc4f733bb2e 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -525,7 +525,7 @@ int nf_conntrack_dccp_packet(struct nf_conn *ct, struct sk_buff *skb, dh = skb_header_pointer(skb, dataoff, sizeof(*dh), &_dh.dh); if (!dh) - return NF_DROP; + return -NF_ACCEPT; if (dccp_error(dh, skb, dataoff, state)) return -NF_ACCEPT; @@ -533,7 +533,7 @@ int nf_conntrack_dccp_packet(struct nf_conn *ct, struct sk_buff *skb, /* pull again, including possible 48 bit sequences and subtype header */ dh = dccp_header_pointer(skb, dataoff, dh, &_dh); if (!dh) - return NF_DROP; + return -NF_ACCEPT; type = dh->dccph_type; if (!nf_ct_is_confirmed(ct) && !dccp_new(ct, skb, dh, state)) -- cgit v1.2.3 From 8928756d53d5b99dcd18073dc7738b8ebdbe7d96 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 2 May 2024 10:44:42 +0200 Subject: net: move skb_gro_receive_list from udp to core This helper function will be used for TCP fraglist GRO support Acked-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: Felix Fietkau Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- include/net/gro.h | 1 + net/core/gro.c | 27 +++++++++++++++++++++++++++ net/ipv4/udp_offload.c | 27 --------------------------- 3 files changed, 28 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/include/net/gro.h b/include/net/gro.h index c1d4ca0463a1..5df8bf318197 100644 --- a/include/net/gro.h +++ b/include/net/gro.h @@ -438,6 +438,7 @@ static inline __wsum ip6_gro_compute_pseudo(const struct sk_buff *skb, } int skb_gro_receive(struct sk_buff *p, struct sk_buff *skb); +int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb); /* Pass the currently batched GRO_NORMAL SKBs up to the stack. */ static inline void gro_normal_list(struct napi_struct *napi) diff --git a/net/core/gro.c b/net/core/gro.c index 99a45a5211c9..e2f84947fb74 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -231,6 +231,33 @@ done: return 0; } +int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb) +{ + if (unlikely(p->len + skb->len >= 65536)) + return -E2BIG; + + if (NAPI_GRO_CB(p)->last == p) + skb_shinfo(p)->frag_list = skb; + else + NAPI_GRO_CB(p)->last->next = skb; + + skb_pull(skb, skb_gro_offset(skb)); + + NAPI_GRO_CB(p)->last = skb; + NAPI_GRO_CB(p)->count++; + p->data_len += skb->len; + + /* sk ownership - if any - completely transferred to the aggregated packet */ + skb->destructor = NULL; + skb->sk = NULL; + p->truesize += skb->truesize; + p->len += skb->len; + + NAPI_GRO_CB(skb)->same_flow = 1; + + return 0; +} + static void napi_gro_complete(struct napi_struct *napi, struct sk_buff *skb) { diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 8721fe5beca2..f6d3c442e260 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -433,33 +433,6 @@ out: return segs; } -static int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb) -{ - if (unlikely(p->len + skb->len >= 65536)) - return -E2BIG; - - if (NAPI_GRO_CB(p)->last == p) - skb_shinfo(p)->frag_list = skb; - else - NAPI_GRO_CB(p)->last->next = skb; - - skb_pull(skb, skb_gro_offset(skb)); - - NAPI_GRO_CB(p)->last = skb; - NAPI_GRO_CB(p)->count++; - p->data_len += skb->len; - - /* sk ownership - if any - completely transferred to the aggregated packet */ - skb->destructor = NULL; - skb->sk = NULL; - p->truesize += skb->truesize; - p->len += skb->len; - - NAPI_GRO_CB(skb)->same_flow = 1; - - return 0; -} - #define UDP_GRO_CNT_MAX 64 static struct sk_buff *udp_gro_receive_segment(struct list_head *head, -- cgit v1.2.3 From bee88cd5bd83d40b8aec4d6cb729378f707f6197 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 2 May 2024 10:44:43 +0200 Subject: net: add support for segmenting TCP fraglist GSO packets Preparation for adding TCP fraglist GRO support. It expects packets to be combined in a similar way as UDP fraglist GSO packets. For IPv4 packets, NAT is handled in the same way as UDP fraglist GSO. Acked-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: Felix Fietkau Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- net/ipv4/tcp_offload.c | 67 ++++++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/tcpv6_offload.c | 58 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 125 insertions(+) (limited to 'net') diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index fab0973f995b..affd4ed28cfe 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -28,6 +28,70 @@ static void tcp_gso_tstamp(struct sk_buff *skb, unsigned int ts_seq, } } +static void __tcpv4_gso_segment_csum(struct sk_buff *seg, + __be32 *oldip, __be32 newip, + __be16 *oldport, __be16 newport) +{ + struct tcphdr *th; + struct iphdr *iph; + + if (*oldip == newip && *oldport == newport) + return; + + th = tcp_hdr(seg); + iph = ip_hdr(seg); + + inet_proto_csum_replace4(&th->check, seg, *oldip, newip, true); + inet_proto_csum_replace2(&th->check, seg, *oldport, newport, false); + *oldport = newport; + + csum_replace4(&iph->check, *oldip, newip); + *oldip = newip; +} + +static struct sk_buff *__tcpv4_gso_segment_list_csum(struct sk_buff *segs) +{ + const struct tcphdr *th; + const struct iphdr *iph; + struct sk_buff *seg; + struct tcphdr *th2; + struct iphdr *iph2; + + seg = segs; + th = tcp_hdr(seg); + iph = ip_hdr(seg); + th2 = tcp_hdr(seg->next); + iph2 = ip_hdr(seg->next); + + if (!(*(const u32 *)&th->source ^ *(const u32 *)&th2->source) && + iph->daddr == iph2->daddr && iph->saddr == iph2->saddr) + return segs; + + while ((seg = seg->next)) { + th2 = tcp_hdr(seg); + iph2 = ip_hdr(seg); + + __tcpv4_gso_segment_csum(seg, + &iph2->saddr, iph->saddr, + &th2->source, th->source); + __tcpv4_gso_segment_csum(seg, + &iph2->daddr, iph->daddr, + &th2->dest, th->dest); + } + + return segs; +} + +static struct sk_buff *__tcp4_gso_segment_list(struct sk_buff *skb, + netdev_features_t features) +{ + skb = skb_segment_list(skb, features, skb_mac_header_len(skb)); + if (IS_ERR(skb)) + return skb; + + return __tcpv4_gso_segment_list_csum(skb); +} + static struct sk_buff *tcp4_gso_segment(struct sk_buff *skb, netdev_features_t features) { @@ -37,6 +101,9 @@ static struct sk_buff *tcp4_gso_segment(struct sk_buff *skb, if (!pskb_may_pull(skb, sizeof(struct tcphdr))) return ERR_PTR(-EINVAL); + if (skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST) + return __tcp4_gso_segment_list(skb, features); + if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) { const struct iphdr *iph = ip_hdr(skb); struct tcphdr *th = tcp_hdr(skb); diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index 4b07d1e6c952..7180c30dbbef 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -40,6 +40,61 @@ INDIRECT_CALLABLE_SCOPE int tcp6_gro_complete(struct sk_buff *skb, int thoff) return 0; } +static void __tcpv6_gso_segment_csum(struct sk_buff *seg, + __be16 *oldport, __be16 newport) +{ + struct tcphdr *th; + + if (*oldport == newport) + return; + + th = tcp_hdr(seg); + inet_proto_csum_replace2(&th->check, seg, *oldport, newport, false); + *oldport = newport; +} + +static struct sk_buff *__tcpv6_gso_segment_list_csum(struct sk_buff *segs) +{ + const struct tcphdr *th; + const struct ipv6hdr *iph; + struct sk_buff *seg; + struct tcphdr *th2; + struct ipv6hdr *iph2; + + seg = segs; + th = tcp_hdr(seg); + iph = ipv6_hdr(seg); + th2 = tcp_hdr(seg->next); + iph2 = ipv6_hdr(seg->next); + + if (!(*(const u32 *)&th->source ^ *(const u32 *)&th2->source) && + ipv6_addr_equal(&iph->saddr, &iph2->saddr) && + ipv6_addr_equal(&iph->daddr, &iph2->daddr)) + return segs; + + while ((seg = seg->next)) { + th2 = tcp_hdr(seg); + iph2 = ipv6_hdr(seg); + + iph2->saddr = iph->saddr; + iph2->daddr = iph->daddr; + __tcpv6_gso_segment_csum(seg, &th2->source, th->source); + __tcpv6_gso_segment_csum(seg, &th2->dest, th->dest); + } + + return segs; +} + +static struct sk_buff *__tcp6_gso_segment_list(struct sk_buff *skb, + netdev_features_t features) +{ + skb = skb_segment_list(skb, features, skb_mac_header_len(skb)); + if (IS_ERR(skb)) + return skb; + + return __tcpv6_gso_segment_list_csum(skb); +} + static struct sk_buff *tcp6_gso_segment(struct sk_buff *skb, netdev_features_t features) { @@ -51,6 +106,9 @@ static struct sk_buff *tcp6_gso_segment(struct sk_buff *skb, if (!pskb_may_pull(skb, sizeof(*th))) return ERR_PTR(-EINVAL); + if (skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST) + return __tcp6_gso_segment_list(skb, features); + if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) { const struct ipv6hdr *ipv6h = ipv6_hdr(skb); struct tcphdr *th = tcp_hdr(skb); -- cgit v1.2.3 From 8d95dc474f85481652a0e422d2f1f079de81f63c Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 2 May 2024 10:44:44 +0200 Subject: net: add code for TCP fraglist GRO This implements fraglist GRO similar to how it's handled in UDP, however no functional changes are added yet. The next change adds a heuristic for using fraglist GRO instead of regular GRO. Acked-by: Paolo Abeni Signed-off-by: Felix Fietkau Reviewed-by: Eric Dumazet Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- net/ipv4/tcp_offload.c | 21 +++++++++++++++++++++ net/ipv6/tcpv6_offload.c | 9 +++++++++ 2 files changed, 30 insertions(+) (limited to 'net') diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index affd4ed28cfe..aa7508676315 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -334,6 +334,18 @@ found: flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq); flush |= skb_cmp_decrypted(p, skb); + if (unlikely(NAPI_GRO_CB(p)->is_flist)) { + flush |= (__force int)(flags ^ tcp_flag_word(th2)); + flush |= skb->ip_summed != p->ip_summed; + flush |= skb->csum_level != p->csum_level; + flush |= NAPI_GRO_CB(p)->count >= 64; + + if (flush || skb_gro_receive_list(p, skb)) + mss = 1; + + goto out_check_final; + } + if (flush || skb_gro_receive(p, skb)) { mss = 1; goto out_check_final; @@ -400,6 +412,15 @@ INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff) const struct iphdr *iph = ip_hdr(skb); struct tcphdr *th = tcp_hdr(skb); + if (unlikely(NAPI_GRO_CB(skb)->is_flist)) { + skb_shinfo(skb)->gso_type |= SKB_GSO_FRAGLIST | SKB_GSO_TCPV4; + skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count; + + __skb_incr_checksum_unnecessary(skb); + + return 0; + } + th->check = ~tcp_v4_check(skb->len - thoff, iph->saddr, iph->daddr, 0); diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index 7180c30dbbef..575e2743e331 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -32,6 +32,15 @@ INDIRECT_CALLABLE_SCOPE int tcp6_gro_complete(struct sk_buff *skb, int thoff) const struct ipv6hdr *iph = ipv6_hdr(skb); struct tcphdr *th = tcp_hdr(skb); + if (unlikely(NAPI_GRO_CB(skb)->is_flist)) { + skb_shinfo(skb)->gso_type |= SKB_GSO_FRAGLIST | SKB_GSO_TCPV6; + skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count; + + __skb_incr_checksum_unnecessary(skb); + + return 0; + } + th->check = ~tcp_v6_check(skb->len - thoff, &iph->saddr, &iph->daddr, 0); skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV6; -- cgit v1.2.3 From 80e85fbdf19ecc4dfa31ecf639adb55555db02fe Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 2 May 2024 10:44:45 +0200 Subject: net: create tcp_gro_lookup helper function This pulls the flow port matching out of tcp_gro_receive, so that it can be reused for the next change, which adds the TCP fraglist GRO heuristic. Acked-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: Felix Fietkau Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- include/net/tcp.h | 1 + net/ipv4/tcp_offload.c | 41 +++++++++++++++++++++++++---------------- 2 files changed, 26 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 0a51e6a45bce..79c4e8947b5d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2191,6 +2191,7 @@ void tcp_v4_destroy_sock(struct sock *sk); struct sk_buff *tcp_gso_segment(struct sk_buff *skb, netdev_features_t features); +struct sk_buff *tcp_gro_lookup(struct list_head *head, struct tcphdr *th); struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb); INDIRECT_CALLABLE_DECLARE(int tcp4_gro_complete(struct sk_buff *skb, int thoff)); INDIRECT_CALLABLE_DECLARE(struct sk_buff *tcp4_gro_receive(struct list_head *head, struct sk_buff *skb)); diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index aa7508676315..4a194a9d36cd 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -245,6 +245,27 @@ out: return segs; } +struct sk_buff *tcp_gro_lookup(struct list_head *head, struct tcphdr *th) +{ + struct tcphdr *th2; + struct sk_buff *p; + + list_for_each_entry(p, head, list) { + if (!NAPI_GRO_CB(p)->same_flow) + continue; + + th2 = tcp_hdr(p); + if (*(u32 *)&th->source ^ *(u32 *)&th2->source) { + NAPI_GRO_CB(p)->same_flow = 0; + continue; + } + + return p; + } + + return NULL; +} + struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb) { struct sk_buff *pp = NULL; @@ -282,24 +303,12 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb) len = skb_gro_len(skb); flags = tcp_flag_word(th); - list_for_each_entry(p, head, list) { - if (!NAPI_GRO_CB(p)->same_flow) - continue; - - th2 = tcp_hdr(p); - - if (*(u32 *)&th->source ^ *(u32 *)&th2->source) { - NAPI_GRO_CB(p)->same_flow = 0; - continue; - } - - goto found; - } - p = NULL; - goto out_check_final; + p = tcp_gro_lookup(head, th); + if (!p) + goto out_check_final; -found: /* Include the IP ID check below from the inner most IP hdr */ + th2 = tcp_hdr(p); flush = NAPI_GRO_CB(p)->flush; flush |= (__force int)(flags & TCP_FLAG_CWR); flush |= (__force int)((flags ^ tcp_flag_word(th2)) & -- cgit v1.2.3 From 7516b27c555c1711ec17a5d891befb6986e573a3 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 2 May 2024 10:44:46 +0200 Subject: net: create tcp_gro_header_pull helper function Pull the code out of tcp_gro_receive in order to access the tcp header from tcp4/6_gro_receive. Acked-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: Felix Fietkau Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- include/net/tcp.h | 4 +++- net/ipv4/tcp_offload.c | 55 ++++++++++++++++++++++++++++++------------------ net/ipv6/tcpv6_offload.c | 18 +++++++++++----- 3 files changed, 50 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 79c4e8947b5d..8f63a163c7de 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2191,8 +2191,10 @@ void tcp_v4_destroy_sock(struct sock *sk); struct sk_buff *tcp_gso_segment(struct sk_buff *skb, netdev_features_t features); +struct tcphdr *tcp_gro_pull_header(struct sk_buff *skb); struct sk_buff *tcp_gro_lookup(struct list_head *head, struct tcphdr *th); -struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb); +struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, + struct tcphdr *th); INDIRECT_CALLABLE_DECLARE(int tcp4_gro_complete(struct sk_buff *skb, int thoff)); INDIRECT_CALLABLE_DECLARE(struct sk_buff *tcp4_gro_receive(struct list_head *head, struct sk_buff *skb)); INDIRECT_CALLABLE_DECLARE(int tcp6_gro_complete(struct sk_buff *skb, int thoff)); diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index 4a194a9d36cd..87ae9808e260 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -266,40 +266,46 @@ struct sk_buff *tcp_gro_lookup(struct list_head *head, struct tcphdr *th) return NULL; } -struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb) +struct tcphdr *tcp_gro_pull_header(struct sk_buff *skb) { - struct sk_buff *pp = NULL; - struct sk_buff *p; + unsigned int thlen, hlen, off; struct tcphdr *th; - struct tcphdr *th2; - unsigned int len; - unsigned int thlen; - __be32 flags; - unsigned int mss = 1; - unsigned int hlen; - unsigned int off; - int flush = 1; - int i; off = skb_gro_offset(skb); hlen = off + sizeof(*th); th = skb_gro_header(skb, hlen, off); if (unlikely(!th)) - goto out; + return NULL; thlen = th->doff * 4; if (thlen < sizeof(*th)) - goto out; + return NULL; hlen = off + thlen; if (!skb_gro_may_pull(skb, hlen)) { th = skb_gro_header_slow(skb, hlen, off); if (unlikely(!th)) - goto out; + return NULL; } skb_gro_pull(skb, thlen); + return th; +} + +struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, + struct tcphdr *th) +{ + unsigned int thlen = th->doff * 4; + struct sk_buff *pp = NULL; + struct sk_buff *p; + struct tcphdr *th2; + unsigned int len; + __be32 flags; + unsigned int mss = 1; + int flush = 1; + int i; + len = skb_gro_len(skb); flags = tcp_flag_word(th); @@ -376,7 +382,6 @@ out_check_final: if (p && (!NAPI_GRO_CB(skb)->same_flow || flush)) pp = p; -out: NAPI_GRO_CB(skb)->flush |= (flush != 0); return pp; @@ -405,15 +410,23 @@ EXPORT_SYMBOL(tcp_gro_complete); INDIRECT_CALLABLE_SCOPE struct sk_buff *tcp4_gro_receive(struct list_head *head, struct sk_buff *skb) { + struct tcphdr *th; + /* Don't bother verifying checksum if we're going to flush anyway. */ if (!NAPI_GRO_CB(skb)->flush && skb_gro_checksum_validate(skb, IPPROTO_TCP, - inet_gro_compute_pseudo)) { - NAPI_GRO_CB(skb)->flush = 1; - return NULL; - } + inet_gro_compute_pseudo)) + goto flush; + + th = tcp_gro_pull_header(skb); + if (!th) + goto flush; - return tcp_gro_receive(head, skb); + return tcp_gro_receive(head, skb, th); + +flush: + NAPI_GRO_CB(skb)->flush = 1; + return NULL; } INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff) diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index 575e2743e331..e73a4f74fd96 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -16,15 +16,23 @@ INDIRECT_CALLABLE_SCOPE struct sk_buff *tcp6_gro_receive(struct list_head *head, struct sk_buff *skb) { + struct tcphdr *th; + /* Don't bother verifying checksum if we're going to flush anyway. */ if (!NAPI_GRO_CB(skb)->flush && skb_gro_checksum_validate(skb, IPPROTO_TCP, - ip6_gro_compute_pseudo)) { - NAPI_GRO_CB(skb)->flush = 1; - return NULL; - } + ip6_gro_compute_pseudo)) + goto flush; + + th = tcp_gro_pull_header(skb); + if (!th) + goto flush; + + return tcp_gro_receive(head, skb, th); - return tcp_gro_receive(head, skb); +flush: + NAPI_GRO_CB(skb)->flush = 1; + return NULL; } INDIRECT_CALLABLE_SCOPE int tcp6_gro_complete(struct sk_buff *skb, int thoff) -- cgit v1.2.3 From c9d1d23e5239f41700be69133a5769ac5ebc88a8 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Thu, 2 May 2024 10:44:47 +0200 Subject: net: add heuristic for enabling TCP fraglist GRO When forwarding TCP after GRO, software segmentation is very expensive, especially when the checksum needs to be recalculated. One case where that's currently unavoidable is when routing packets over PPPoE. Performance improves significantly when using fraglist GRO implemented in the same way as for UDP. When NETIF_F_GRO_FRAGLIST is enabled, perform a lookup for an established socket in the same netns as the receiving device. While this may not cover all relevant use cases in multi-netns configurations, it should be good enough for most configurations that need this. Here's a measurement of running 2 TCP streams through a MediaTek MT7622 device (2-core Cortex-A53), which runs NAT with flow offload enabled from one ethernet port to PPPoE on another ethernet port + cake qdisc set to 1Gbps. rx-gro-list off: 630 Mbit/s, CPU 35% idle rx-gro-list on: 770 Mbit/s, CPU 40% idle Acked-by: Paolo Abeni Reviewed-by: Eric Dumazet Signed-off-by: Felix Fietkau Reviewed-by: David Ahern Reviewed-by: Willem de Bruijn Signed-off-by: Paolo Abeni --- net/ipv4/tcp_offload.c | 32 ++++++++++++++++++++++++++++++++ net/ipv6/tcpv6_offload.c | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 67 insertions(+) (limited to 'net') diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index 87ae9808e260..c90704befd7b 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -407,6 +407,36 @@ void tcp_gro_complete(struct sk_buff *skb) } EXPORT_SYMBOL(tcp_gro_complete); +static void tcp4_check_fraglist_gro(struct list_head *head, struct sk_buff *skb, + struct tcphdr *th) +{ + const struct iphdr *iph; + struct sk_buff *p; + struct sock *sk; + struct net *net; + int iif, sdif; + + if (likely(!(skb->dev->features & NETIF_F_GRO_FRAGLIST))) + return; + + p = tcp_gro_lookup(head, th); + if (p) { + NAPI_GRO_CB(skb)->is_flist = NAPI_GRO_CB(p)->is_flist; + return; + } + + inet_get_iif_sdif(skb, &iif, &sdif); + iph = skb_gro_network_header(skb); + net = dev_net(skb->dev); + sk = __inet_lookup_established(net, net->ipv4.tcp_death_row.hashinfo, + iph->saddr, th->source, + iph->daddr, ntohs(th->dest), + iif, sdif); + NAPI_GRO_CB(skb)->is_flist = !sk; + if (sk) + sock_put(sk); +} + INDIRECT_CALLABLE_SCOPE struct sk_buff *tcp4_gro_receive(struct list_head *head, struct sk_buff *skb) { @@ -422,6 +452,8 @@ struct sk_buff *tcp4_gro_receive(struct list_head *head, struct sk_buff *skb) if (!th) goto flush; + tcp4_check_fraglist_gro(head, skb, th); + return tcp_gro_receive(head, skb, th); flush: diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index e73a4f74fd96..d59f58cbd306 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -7,12 +7,45 @@ */ #include #include +#include #include #include #include #include #include "ip6_offload.h" +static void tcp6_check_fraglist_gro(struct list_head *head, struct sk_buff *skb, + struct tcphdr *th) +{ +#if IS_ENABLED(CONFIG_IPV6) + const struct ipv6hdr *hdr; + struct sk_buff *p; + struct sock *sk; + struct net *net; + int iif, sdif; + + if (likely(!(skb->dev->features & NETIF_F_GRO_FRAGLIST))) + return; + + p = tcp_gro_lookup(head, th); + if (p) { + NAPI_GRO_CB(skb)->is_flist = NAPI_GRO_CB(p)->is_flist; + return; + } + + inet6_get_iif_sdif(skb, &iif, &sdif); + hdr = skb_gro_network_header(skb); + net = dev_net(skb->dev); + sk = __inet6_lookup_established(net, net->ipv4.tcp_death_row.hashinfo, + &hdr->saddr, th->source, + &hdr->daddr, ntohs(th->dest), + iif, sdif); + NAPI_GRO_CB(skb)->is_flist = !sk; + if (sk) + sock_put(sk); +#endif /* IS_ENABLED(CONFIG_IPV6) */ +} + INDIRECT_CALLABLE_SCOPE struct sk_buff *tcp6_gro_receive(struct list_head *head, struct sk_buff *skb) { @@ -28,6 +61,8 @@ struct sk_buff *tcp6_gro_receive(struct list_head *head, struct sk_buff *skb) if (!th) goto flush; + tcp6_check_fraglist_gro(head, skb, th); + return tcp_gro_receive(head, skb, th); flush: -- cgit v1.2.3 From a26ff37e624d12e28077e5b24d2b264f62764ad6 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Thu, 2 May 2024 10:20:06 -0300 Subject: net: fix out-of-bounds access in ops_init net_alloc_generic is called by net_alloc, which is called without any locking. It reads max_gen_ptrs, which is changed under pernet_ops_rwsem. It is read twice, first to allocate an array, then to set s.len, which is later used to limit the bounds of the array access. It is possible that the array is allocated and another thread is registering a new pernet ops, increments max_gen_ptrs, which is then used to set s.len with a larger than allocated length for the variable array. Fix it by reading max_gen_ptrs only once in net_alloc_generic. If max_gen_ptrs is later incremented, it will be caught in net_assign_generic. Signed-off-by: Thadeu Lima de Souza Cascardo Fixes: 073862ba5d24 ("netns: fix net_alloc_generic()") Reviewed-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20240502132006.3430840-1-cascardo@igalia.com Signed-off-by: Paolo Abeni --- net/core/net_namespace.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index f0540c557515..9d690d32da33 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -69,12 +69,15 @@ DEFINE_COOKIE(net_cookie); static struct net_generic *net_alloc_generic(void) { + unsigned int gen_ptrs = READ_ONCE(max_gen_ptrs); + unsigned int generic_size; struct net_generic *ng; - unsigned int generic_size = offsetof(struct net_generic, ptr[max_gen_ptrs]); + + generic_size = offsetof(struct net_generic, ptr[gen_ptrs]); ng = kzalloc(generic_size, GFP_KERNEL); if (ng) - ng->s.len = max_gen_ptrs; + ng->s.len = gen_ptrs; return ng; } @@ -1307,7 +1310,11 @@ static int register_pernet_operations(struct list_head *list, if (error < 0) return error; *ops->id = error; - max_gen_ptrs = max(max_gen_ptrs, *ops->id + 1); + /* This does not require READ_ONCE as writers already hold + * pernet_ops_rwsem. But WRITE_ONCE is needed to protect + * net_alloc_generic. + */ + WRITE_ONCE(max_gen_ptrs, max(max_gen_ptrs, *ops->id + 1)); } error = __register_pernet_operations(list, ops); if (error) { -- cgit v1.2.3 From a576f36971ab4097b6aa76433532aa1fb5ee2d3b Mon Sep 17 00:00:00 2001 From: Aleksandr Aprelkov Date: Wed, 27 Mar 2024 14:10:44 +0700 Subject: sunrpc: removed redundant procp check since vs_proc pointer is dereferenced before getting it's address there's no need to check for NULL. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: 8e5b67731d08 ("SUNRPC: Add a callback to initialise server requests") Signed-off-by: Aleksandr Aprelkov Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- net/sunrpc/svc.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net') diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index b33e429336fb..2b4b1276d4e8 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -1265,8 +1265,6 @@ svc_generic_init_request(struct svc_rqst *rqstp, if (rqstp->rq_proc >= versp->vs_nproc) goto err_bad_proc; rqstp->rq_procinfo = procp = &versp->vs_proc[rqstp->rq_proc]; - if (!procp) - goto err_bad_proc; /* Initialize storage for argp and resp */ memset(rqstp->rq_argp, 0, procp->pc_argzero); -- cgit v1.2.3 From a79ec2aecb1b3d9c59241a6155a5554b236de9b7 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 23 Apr 2024 15:25:42 +0200 Subject: SUNRPC: introduce svc_xprt_create_from_sa utility routine Add svc_xprt_create_from_sa utility routine and refactor svc_xprt_create() codebase in order to introduce the capability to create a svc port from socket address. Reviewed-by: Jeff Layton Tested-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- include/linux/sunrpc/svc_xprt.h | 3 + net/sunrpc/svc_xprt.c | 133 ++++++++++++++++++++++------------------ 2 files changed, 78 insertions(+), 58 deletions(-) (limited to 'net') diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 8e20cd60e2e7..0d9b10dbe07d 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -135,6 +135,9 @@ int svc_reg_xprt_class(struct svc_xprt_class *); void svc_unreg_xprt_class(struct svc_xprt_class *); void svc_xprt_init(struct net *, struct svc_xprt_class *, struct svc_xprt *, struct svc_serv *); +int svc_xprt_create_from_sa(struct svc_serv *serv, const char *xprt_name, + struct net *net, struct sockaddr *sap, + int flags, const struct cred *cred); int svc_xprt_create(struct svc_serv *serv, const char *xprt_name, struct net *net, const int family, const unsigned short port, int flags, diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index b4a85a227bd7..463fe544ae28 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -211,51 +211,6 @@ void svc_xprt_init(struct net *net, struct svc_xprt_class *xcl, } EXPORT_SYMBOL_GPL(svc_xprt_init); -static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, - struct svc_serv *serv, - struct net *net, - const int family, - const unsigned short port, - int flags) -{ - struct sockaddr_in sin = { - .sin_family = AF_INET, - .sin_addr.s_addr = htonl(INADDR_ANY), - .sin_port = htons(port), - }; -#if IS_ENABLED(CONFIG_IPV6) - struct sockaddr_in6 sin6 = { - .sin6_family = AF_INET6, - .sin6_addr = IN6ADDR_ANY_INIT, - .sin6_port = htons(port), - }; -#endif - struct svc_xprt *xprt; - struct sockaddr *sap; - size_t len; - - switch (family) { - case PF_INET: - sap = (struct sockaddr *)&sin; - len = sizeof(sin); - break; -#if IS_ENABLED(CONFIG_IPV6) - case PF_INET6: - sap = (struct sockaddr *)&sin6; - len = sizeof(sin6); - break; -#endif - default: - return ERR_PTR(-EAFNOSUPPORT); - } - - xprt = xcl->xcl_ops->xpo_create(serv, net, sap, len, flags); - if (IS_ERR(xprt)) - trace_svc_xprt_create_err(serv->sv_program->pg_name, - xcl->xcl_name, sap, len, xprt); - return xprt; -} - /** * svc_xprt_received - start next receiver thread * @xprt: controlling transport @@ -294,9 +249,8 @@ void svc_add_new_perm_xprt(struct svc_serv *serv, struct svc_xprt *new) } static int _svc_xprt_create(struct svc_serv *serv, const char *xprt_name, - struct net *net, const int family, - const unsigned short port, int flags, - const struct cred *cred) + struct net *net, struct sockaddr *sap, + size_t len, int flags, const struct cred *cred) { struct svc_xprt_class *xcl; @@ -312,8 +266,11 @@ static int _svc_xprt_create(struct svc_serv *serv, const char *xprt_name, goto err; spin_unlock(&svc_xprt_class_lock); - newxprt = __svc_xpo_create(xcl, serv, net, family, port, flags); + newxprt = xcl->xcl_ops->xpo_create(serv, net, sap, len, flags); if (IS_ERR(newxprt)) { + trace_svc_xprt_create_err(serv->sv_program->pg_name, + xcl->xcl_name, sap, len, + newxprt); module_put(xcl->xcl_owner); return PTR_ERR(newxprt); } @@ -329,6 +286,48 @@ static int _svc_xprt_create(struct svc_serv *serv, const char *xprt_name, return -EPROTONOSUPPORT; } +/** + * svc_xprt_create_from_sa - Add a new listener to @serv from socket address + * @serv: target RPC service + * @xprt_name: transport class name + * @net: network namespace + * @sap: socket address pointer + * @flags: SVC_SOCK flags + * @cred: credential to bind to this transport + * + * Return local xprt port on success or %-EPROTONOSUPPORT on failure + */ +int svc_xprt_create_from_sa(struct svc_serv *serv, const char *xprt_name, + struct net *net, struct sockaddr *sap, + int flags, const struct cred *cred) +{ + size_t len; + int err; + + switch (sap->sa_family) { + case AF_INET: + len = sizeof(struct sockaddr_in); + break; +#if IS_ENABLED(CONFIG_IPV6) + case AF_INET6: + len = sizeof(struct sockaddr_in6); + break; +#endif + default: + return -EAFNOSUPPORT; + } + + err = _svc_xprt_create(serv, xprt_name, net, sap, len, flags, cred); + if (err == -EPROTONOSUPPORT) { + request_module("svc%s", xprt_name); + err = _svc_xprt_create(serv, xprt_name, net, sap, len, flags, + cred); + } + + return err; +} +EXPORT_SYMBOL_GPL(svc_xprt_create_from_sa); + /** * svc_xprt_create - Add a new listener to @serv * @serv: target RPC service @@ -339,23 +338,41 @@ static int _svc_xprt_create(struct svc_serv *serv, const char *xprt_name, * @flags: SVC_SOCK flags * @cred: credential to bind to this transport * - * Return values: - * %0: New listener added successfully - * %-EPROTONOSUPPORT: Requested transport type not supported + * Return local xprt port on success or %-EPROTONOSUPPORT on failure */ int svc_xprt_create(struct svc_serv *serv, const char *xprt_name, struct net *net, const int family, const unsigned short port, int flags, const struct cred *cred) { - int err; + struct sockaddr_in sin = { + .sin_family = AF_INET, + .sin_addr.s_addr = htonl(INADDR_ANY), + .sin_port = htons(port), + }; +#if IS_ENABLED(CONFIG_IPV6) + struct sockaddr_in6 sin6 = { + .sin6_family = AF_INET6, + .sin6_addr = IN6ADDR_ANY_INIT, + .sin6_port = htons(port), + }; +#endif + struct sockaddr *sap; - err = _svc_xprt_create(serv, xprt_name, net, family, port, flags, cred); - if (err == -EPROTONOSUPPORT) { - request_module("svc%s", xprt_name); - err = _svc_xprt_create(serv, xprt_name, net, family, port, flags, cred); + switch (family) { + case PF_INET: + sap = (struct sockaddr *)&sin; + break; +#if IS_ENABLED(CONFIG_IPV6) + case PF_INET6: + sap = (struct sockaddr *)&sin6; + break; +#endif + default: + return -EAFNOSUPPORT; } - return err; + + return svc_xprt_create_from_sa(serv, xprt_name, net, sap, flags, cred); } EXPORT_SYMBOL_GPL(svc_xprt_create); -- cgit v1.2.3 From cf619507ae8d30b5735e9356d85a8df2bcf2b6ca Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 23 Apr 2024 15:25:43 +0200 Subject: SUNRPC: add a new svc_find_listener helper svc_find_listener will return the transport instance pointer for the endpoint accepting connections/peer traffic from the specified transport class and matching sockaddr. Signed-off-by: Jeff Layton Signed-off-by: Lorenzo Bianconi Signed-off-by: Chuck Lever --- include/linux/sunrpc/svc_xprt.h | 2 ++ net/sunrpc/svc_xprt.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+) (limited to 'net') diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 0d9b10dbe07d..0981e35a9fed 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -150,6 +150,8 @@ void svc_xprt_copy_addrs(struct svc_rqst *rqstp, struct svc_xprt *xprt); void svc_xprt_close(struct svc_xprt *xprt); int svc_port_is_privileged(struct sockaddr *sin); int svc_print_xprts(char *buf, int maxlen); +struct svc_xprt *svc_find_listener(struct svc_serv *serv, const char *xcl_name, + struct net *net, const struct sockaddr *sa); struct svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name, struct net *net, const sa_family_t af, const unsigned short port); diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 463fe544ae28..34a3626c56b1 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -1276,6 +1276,40 @@ static struct svc_deferred_req *svc_deferred_dequeue(struct svc_xprt *xprt) return dr; } +/** + * svc_find_listener - find an RPC transport instance + * @serv: pointer to svc_serv to search + * @xcl_name: C string containing transport's class name + * @net: owner net pointer + * @sa: sockaddr containing address + * + * Return the transport instance pointer for the endpoint accepting + * connections/peer traffic from the specified transport class, + * and matching sockaddr. + */ +struct svc_xprt *svc_find_listener(struct svc_serv *serv, const char *xcl_name, + struct net *net, const struct sockaddr *sa) +{ + struct svc_xprt *xprt; + struct svc_xprt *found = NULL; + + spin_lock_bh(&serv->sv_lock); + list_for_each_entry(xprt, &serv->sv_permsocks, xpt_list) { + if (xprt->xpt_net != net) + continue; + if (strcmp(xprt->xpt_class->xcl_name, xcl_name)) + continue; + if (!rpc_cmp_addr_port(sa, (struct sockaddr *)&xprt->xpt_local)) + continue; + found = xprt; + svc_xprt_get(xprt); + break; + } + spin_unlock_bh(&serv->sv_lock); + return found; +} +EXPORT_SYMBOL_GPL(svc_find_listener); + /** * svc_find_xprt - find an RPC transport instance * @serv: pointer to svc_serv to search -- cgit v1.2.3 From f2ad13ad08e14ed04395b5bfe2ced5cdbd69b174 Mon Sep 17 00:00:00 2001 From: Guoqing Jiang Date: Fri, 26 Apr 2024 11:47:50 +0800 Subject: SUNRPC: Remove comment for sp_lock It is obsolete since sp_lock was discarded in commit 580a25756a9f ("SUNRPC: discard sp_lock"). Signed-off-by: Guoqing Jiang Reviewed-by: NeilBrown Signed-off-by: Chuck Lever --- net/sunrpc/svc_xprt.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 34a3626c56b1..dd86d7f1e97e 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -46,7 +46,6 @@ static LIST_HEAD(svc_xprt_class_list); /* SMP locking strategy: * - * svc_pool->sp_lock protects most of the fields of that pool. * svc_serv->sv_lock protects sv_tempsocks, sv_permsocks, sv_tmpcnt. * when both need to be taken (rare), svc_serv->sv_lock is first. * The "service mutex" protects svc_serv->sv_nrthread. -- cgit v1.2.3 From 8edc27fc4f220179189932cb5f537f2e90a32b87 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Mon, 25 Mar 2024 20:36:14 +0800 Subject: netfilter: use NF_DROP instead of -NF_DROP At the beginning in 2009 one patch [1] introduced collecting drop counter in nf_conntrack_in() by returning -NF_DROP. Later, another patch [2] changed the return value of tcp_packet() which now is renamed to nf_conntrack_tcp_packet() from -NF_DROP to NF_DROP. As we can see, that -NF_DROP should be corrected. Similarly, there are other two points where the -NF_DROP is used. Well, as NF_DROP is equal to 0, inverting NF_DROP makes no sense as patch [2] said many years ago. [1] commit 7d1e04598e5e ("netfilter: nf_conntrack: account packets drop by tcp_packet()") [2] commit ec8d540969da ("netfilter: conntrack: fix dropping packet after l4proto->packet()") Signed-off-by: Jason Xing Signed-off-by: Pablo Neira Ayuso --- net/ipv4/netfilter/iptable_filter.c | 2 +- net/ipv6/netfilter/ip6table_filter.c | 2 +- net/netfilter/nf_conntrack_core.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index b9062f4552ac..3ab908b74795 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -44,7 +44,7 @@ static int iptable_filter_table_init(struct net *net) return -ENOMEM; /* Entry 1 is the FORWARD hook */ ((struct ipt_standard *)repl->entries)[1].target.verdict = - forward ? -NF_ACCEPT - 1 : -NF_DROP - 1; + forward ? -NF_ACCEPT - 1 : NF_DROP - 1; err = ipt_register_table(net, &packet_filter, repl, filter_ops); kfree(repl); diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index df785ebda0ca..e8992693e14a 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -43,7 +43,7 @@ static int ip6table_filter_table_init(struct net *net) return -ENOMEM; /* Entry 1 is the FORWARD hook */ ((struct ip6t_standard *)repl->entries)[1].target.verdict = - forward ? -NF_ACCEPT - 1 : -NF_DROP - 1; + forward ? -NF_ACCEPT - 1 : NF_DROP - 1; err = ip6t_register_table(net, &packet_filter, repl, filter_ops); kfree(repl); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index c63868666bd9..6102dc09cdd3 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -2024,7 +2024,7 @@ repeat: goto repeat; NF_CT_STAT_INC_ATOMIC(state->net, invalid); - if (ret == -NF_DROP) + if (ret == NF_DROP) NF_CT_STAT_INC_ATOMIC(state->net, drop); ret = -ret; -- cgit v1.2.3 From 119c790a271de02dadb36ba0c1fc31a7d5d4c62b Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 23 Apr 2024 15:44:28 +0200 Subject: netfilter: conntrack: remove flowtable early-drop test Not sure why this special case exists. Early drop logic (which kicks in when conntrack table is full) should be independent of flowtable offload and only consider assured bit (i.e., two-way traffic was seen). flowtable entries hold a reference to the conntrack entry (struct nf_conn) that has been offloaded. The conntrack use count is not decremented until after the entry is free'd. This change therefore will not result in exceeding the conntrack table limit. It does allow early-drop of tcp flows even when they've been offloaded, but only if they have been offloaded before syn-ack was received or after at least one peer has sent a fin. Currently 'fin' packet reception already stops offloading, so this should not impact offloading either. Cc: Vlad Buslov Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_core.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 6102dc09cdd3..7ac20750c127 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1440,8 +1440,6 @@ static bool gc_worker_can_early_drop(const struct nf_conn *ct) const struct nf_conntrack_l4proto *l4proto; u8 protonum = nf_ct_protonum(ct); - if (test_bit(IPS_OFFLOAD_BIT, &ct->status) && protonum != IPPROTO_UDP) - return false; if (!test_bit(IPS_ASSURED_BIT, &ct->status)) return true; -- cgit v1.2.3 From a590f4760922acaa2d2b55a88004a38eecdd6412 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:40 +0200 Subject: netfilter: nft_set_pipapo: move prove_locking helper around Preparation patch, the helper will soon get called from insert function too. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index 187138afac45..b8205d961ba4 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -1247,6 +1247,17 @@ static int pipapo_realloc_scratch(struct nft_pipapo_match *clone, return 0; } +static bool nft_pipapo_transaction_mutex_held(const struct nft_set *set) +{ +#ifdef CONFIG_PROVE_LOCKING + const struct net *net = read_pnet(&set->net); + + return lockdep_is_held(&nft_pernet(net)->commit_mutex); +#else + return true; +#endif +} + /** * nft_pipapo_insert() - Validate and insert ranged elements * @net: Network namespace @@ -1799,17 +1810,6 @@ static void nft_pipapo_commit(struct nft_set *set) priv->clone = new_clone; } -static bool nft_pipapo_transaction_mutex_held(const struct nft_set *set) -{ -#ifdef CONFIG_PROVE_LOCKING - const struct net *net = read_pnet(&set->net); - - return lockdep_is_held(&nft_pernet(net)->commit_mutex); -#else - return true; -#endif -} - static void nft_pipapo_abort(const struct nft_set *set) { struct nft_pipapo *priv = nft_set_priv(set); -- cgit v1.2.3 From 80efd2997fb9343a0283cf3cac5524a4595c8ff4 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:41 +0200 Subject: netfilter: nft_set_pipapo: make pipapo_clone helper return NULL Currently it returns an error pointer, but the only possible failure is ENOMEM. After a followup patch, we'd need to discard the errno code, i.e. x = pipapo_clone() if (IS_ERR(x)) return NULL or make more changes to fix up callers to expect IS_ERR() code from set->ops->deactivate(). So simplify this and make it return ptr-or-null. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index b8205d961ba4..7b6d5d2d0d54 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -1395,7 +1395,7 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, * pipapo_clone() - Clone matching data to create new working copy * @old: Existing matching data * - * Return: copy of matching data passed as 'old', error pointer on failure + * Return: copy of matching data passed as 'old' or NULL. */ static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old) { @@ -1405,7 +1405,7 @@ static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old) new = kmalloc(struct_size(new, f, old->field_count), GFP_KERNEL); if (!new) - return ERR_PTR(-ENOMEM); + return NULL; new->field_count = old->field_count; new->bsize_max = old->bsize_max; @@ -1477,7 +1477,7 @@ out_scratch: free_percpu(new->scratch); kfree(new); - return ERR_PTR(-ENOMEM); + return NULL; } /** @@ -1797,7 +1797,7 @@ static void nft_pipapo_commit(struct nft_set *set) return; new_clone = pipapo_clone(priv->clone); - if (IS_ERR(new_clone)) + if (!new_clone) return; priv->dirty = false; @@ -1821,7 +1821,7 @@ static void nft_pipapo_abort(const struct nft_set *set) m = rcu_dereference_protected(priv->match, nft_pipapo_transaction_mutex_held(set)); new_clone = pipapo_clone(m); - if (IS_ERR(new_clone)) + if (!new_clone) return; priv->dirty = false; @@ -2269,8 +2269,8 @@ static int nft_pipapo_init(const struct nft_set *set, /* Create an initial clone of matching data for next insertion */ priv->clone = pipapo_clone(m); - if (IS_ERR(priv->clone)) { - err = PTR_ERR(priv->clone); + if (!priv->clone) { + err = -ENOMEM; goto out_free; } -- cgit v1.2.3 From 8b8a2417558c632f249abebde97adf8c46540de2 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:42 +0200 Subject: netfilter: nft_set_pipapo: prepare destroy function for on-demand clone Once priv->clone can be NULL in case no insertions/removals occurred in the last transaction we need to drop set elements from priv->match if priv->clone is NULL. While at it, condense this function by reusing the pipapo_free_match helper instead of open-coded version. The rcu_barrier() is removed, its not needed: old call_rcu instances for pipapo_reclaim_match do not access struct nft_set. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 27 ++++++--------------------- 1 file changed, 6 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index 7b6d5d2d0d54..459e2dd5050c 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -2326,33 +2326,18 @@ static void nft_pipapo_destroy(const struct nft_ctx *ctx, { struct nft_pipapo *priv = nft_set_priv(set); struct nft_pipapo_match *m; - int cpu; m = rcu_dereference_protected(priv->match, true); - if (m) { - rcu_barrier(); - - for_each_possible_cpu(cpu) - pipapo_free_scratch(m, cpu); - free_percpu(m->scratch); - pipapo_free_fields(m); - kfree(m); - priv->match = NULL; - } if (priv->clone) { - m = priv->clone; - - nft_set_pipapo_match_destroy(ctx, set, m); - - for_each_possible_cpu(cpu) - pipapo_free_scratch(priv->clone, cpu); - free_percpu(priv->clone->scratch); - - pipapo_free_fields(priv->clone); - kfree(priv->clone); + nft_set_pipapo_match_destroy(ctx, set, priv->clone); + pipapo_free_match(priv->clone); priv->clone = NULL; + } else { + nft_set_pipapo_match_destroy(ctx, set, m); } + + pipapo_free_match(m); } /** -- cgit v1.2.3 From 6c108d9bee448a850b03e682836bfe91fca645cb Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:43 +0200 Subject: netfilter: nft_set_pipapo: prepare walk function for on-demand clone The existing code uses iter->type to figure out what data is needed, the live copy (READ) or clone (UPDATE). Without pending updates, priv->clone and priv->match will point to different memory locations, but they have identical content. Future patch will make priv->clone == NULL if there are no pending changes, in this case we must copy the live data for the UPDATE case. Currently this would require GFP_ATOMIC allocation. Split the walk function in two parts: one that does the walk and one that decides which data is needed. In the UPDATE case, callers hold the transaction mutex so we do not need the rcu read lock. This allows to use GFP_KERNEL allocation while cloning. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 60 +++++++++++++++++++++++++++--------------- 1 file changed, 39 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index 459e2dd5050c..543d6cf01de6 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -2106,35 +2106,23 @@ static void nft_pipapo_remove(const struct net *net, const struct nft_set *set, } /** - * nft_pipapo_walk() - Walk over elements + * nft_pipapo_do_walk() - Walk over elements in m * @ctx: nftables API context * @set: nftables API set representation + * @m: matching data pointing to key mapping array * @iter: Iterator * * As elements are referenced in the mapping array for the last field, directly * scan that array: there's no need to follow rule mappings from the first - * field. + * field. @m is protected either by RCU read lock or by transaction mutex. */ -static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set, - struct nft_set_iter *iter) +static void nft_pipapo_do_walk(const struct nft_ctx *ctx, struct nft_set *set, + const struct nft_pipapo_match *m, + struct nft_set_iter *iter) { - struct nft_pipapo *priv = nft_set_priv(set); - const struct nft_pipapo_match *m; const struct nft_pipapo_field *f; unsigned int i, r; - WARN_ON_ONCE(iter->type != NFT_ITER_READ && - iter->type != NFT_ITER_UPDATE); - - rcu_read_lock(); - if (iter->type == NFT_ITER_READ) - m = rcu_dereference(priv->match); - else - m = priv->clone; - - if (unlikely(!m)) - goto out; - for (i = 0, f = m->f; i < m->field_count - 1; i++, f++) ; @@ -2151,14 +2139,44 @@ static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set, iter->err = iter->fn(ctx, set, iter, &e->priv); if (iter->err < 0) - goto out; + return; cont: iter->count++; } +} -out: - rcu_read_unlock(); +/** + * nft_pipapo_walk() - Walk over elements + * @ctx: nftables API context + * @set: nftables API set representation + * @iter: Iterator + * + * Test if destructive action is needed or not, clone active backend if needed + * and call the real function to work on the data. + */ +static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set, + struct nft_set_iter *iter) +{ + struct nft_pipapo *priv = nft_set_priv(set); + const struct nft_pipapo_match *m; + + switch (iter->type) { + case NFT_ITER_UPDATE: + m = priv->clone; + nft_pipapo_do_walk(ctx, set, m, iter); + break; + case NFT_ITER_READ: + rcu_read_lock(); + m = rcu_dereference(priv->match); + nft_pipapo_do_walk(ctx, set, m, iter); + rcu_read_unlock(); + break; + default: + iter->err = -EINVAL; + WARN_ON_ONCE(1); + break; + } } /** -- cgit v1.2.3 From c5444786d0ea2417a5e2cee7bd67137fc8bad687 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:44 +0200 Subject: netfilter: nft_set_pipapo: merge deactivate helper into caller Its the only remaining call site so there is no need for this to be separated anymore. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 39 +++++++++------------------------------ 1 file changed, 9 insertions(+), 30 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index 543d6cf01de6..7c11f568069c 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -1851,52 +1851,31 @@ static void nft_pipapo_activate(const struct net *net, } /** - * pipapo_deactivate() - Check that element is in set, mark as inactive + * nft_pipapo_deactivate() - Search for element and make it inactive * @net: Network namespace * @set: nftables API set representation - * @data: Input key data - * @ext: nftables API extension pointer, used to check for end element - * - * This is a convenience function that can be called from both - * nft_pipapo_deactivate() and nft_pipapo_flush(), as they are in fact the same - * operation. + * @elem: nftables API element representation containing key data * * Return: deactivated element if found, NULL otherwise. */ -static void *pipapo_deactivate(const struct net *net, const struct nft_set *set, - const u8 *data, const struct nft_set_ext *ext) +static struct nft_elem_priv * +nft_pipapo_deactivate(const struct net *net, const struct nft_set *set, + const struct nft_set_elem *elem) { struct nft_pipapo_elem *e; - e = pipapo_get(net, set, data, nft_genmask_next(net), - nft_net_tstamp(net), GFP_KERNEL); + e = pipapo_get(net, set, (const u8 *)elem->key.val.data, + nft_genmask_next(net), nft_net_tstamp(net), GFP_KERNEL); if (IS_ERR(e)) return NULL; nft_set_elem_change_active(net, set, &e->ext); - return e; -} - -/** - * nft_pipapo_deactivate() - Call pipapo_deactivate() to make element inactive - * @net: Network namespace - * @set: nftables API set representation - * @elem: nftables API element representation containing key data - * - * Return: deactivated element if found, NULL otherwise. - */ -static struct nft_elem_priv * -nft_pipapo_deactivate(const struct net *net, const struct nft_set *set, - const struct nft_set_elem *elem) -{ - const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv); - - return pipapo_deactivate(net, set, (const u8 *)elem->key.val.data, ext); + return &e->priv; } /** - * nft_pipapo_flush() - Call pipapo_deactivate() to make element inactive + * nft_pipapo_flush() - make element inactive * @net: Network namespace * @set: nftables API set representation * @elem_priv: nftables API element representation containing key data -- cgit v1.2.3 From d8cac8568618dcb8a51af3db1103e8d4cc4aeea7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 2 May 2024 16:17:00 +0000 Subject: phonet: fix rtm_phonet_notify() skb allocation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit fill_route() stores three components in the skb: - struct rtmsg - RTA_DST (u8) - RTA_OIF (u32) Therefore, rtm_phonet_notify() should use NLMSG_ALIGN(sizeof(struct rtmsg)) + nla_total_size(1) + nla_total_size(4) Fixes: f062f41d0657 ("Phonet: routing table Netlink interface") Signed-off-by: Eric Dumazet Acked-by: Rémi Denis-Courmont Link: https://lore.kernel.org/r/20240502161700.1804476-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/phonet/pn_netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c index 59aebe296890..dd4c7e9a634f 100644 --- a/net/phonet/pn_netlink.c +++ b/net/phonet/pn_netlink.c @@ -193,7 +193,7 @@ void rtm_phonet_notify(int event, struct net_device *dev, u8 dst) struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(NLMSG_ALIGN(sizeof(struct ifaddrmsg)) + + skb = nlmsg_new(NLMSG_ALIGN(sizeof(struct rtmsg)) + nla_total_size(1) + nla_total_size(4), GFP_KERNEL); if (skb == NULL) goto errout; -- cgit v1.2.3 From 46a5d3abedbeb78640376257ee5c3aa690f6ee41 Mon Sep 17 00:00:00 2001 From: Shi-Sheng Yang Date: Thu, 2 May 2024 23:47:40 +0800 Subject: mptcp: fix typos in comments This patch fixes the spelling mistakes in comments. The changes were generated using codespell and reviewed manually. eariler -> earlier greceful -> graceful Signed-off-by: Shi-Sheng Yang Reviewed-by: Simon Horman Reviewed-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240502154740.249839-1-fourcolor4c@gmail.com Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 97ec44d1df30..0ac8d4850751 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -911,7 +911,7 @@ create_child: } /* check for expected invariant - should never trigger, just help - * catching eariler subtle bugs + * catching earlier subtle bugs */ WARN_ON_ONCE(child && *own_req && tcp_sk(child)->is_mptcp && (!mptcp_subflow_ctx(child) || @@ -1259,7 +1259,7 @@ static void mptcp_subflow_fail(struct mptcp_sock *msk, struct sock *ssk) struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk); unsigned long fail_tout; - /* greceful failure can happen only on the MPC subflow */ + /* graceful failure can happen only on the MPC subflow */ if (WARN_ON_ONCE(ssk != READ_ONCE(msk->first))) return; -- cgit v1.2.3 From 698419ffb6fc83dd7b0359d9e8476e732967eed2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:52 +0000 Subject: rtnetlink: do not depend on RTNL for IFLA_QDISC output dev->qdisc can be read using RCU protection. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni --- net/core/rtnetlink.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 28050e53ecb0..ba5301d6df84 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1832,7 +1832,6 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, if (tgt_netnsid >= 0 && nla_put_s32(skb, IFLA_TARGET_NETNSID, tgt_netnsid)) goto nla_put_failure; - qdisc = rtnl_dereference(dev->qdisc); if (nla_put_string(skb, IFLA_IFNAME, dev->name) || nla_put_u32(skb, IFLA_TXQLEN, dev->tx_queue_len) || nla_put_u8(skb, IFLA_OPERSTATE, @@ -1857,8 +1856,6 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, #endif put_master_ifindex(skb, dev) || nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) || - (qdisc && - nla_put_string(skb, IFLA_QDISC, qdisc->ops->id)) || nla_put_ifalias(skb, dev) || nla_put_u32(skb, IFLA_CARRIER_CHANGES, atomic_read(&dev->carrier_up_count) + @@ -1924,6 +1921,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, goto nla_put_failure; rcu_read_lock(); + qdisc = rcu_dereference(dev->qdisc); + if (qdisc && nla_put_string(skb, IFLA_QDISC, qdisc->ops->id)) + goto nla_put_failure_rcu; if (rtnl_fill_link_af(skb, dev, ext_filter_mask)) goto nla_put_failure_rcu; if (rtnl_fill_link_ifmap(skb, dev)) -- cgit v1.2.3 From 8a58268133622c3d50155ac5798ad1d51d6bd3be Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:53 +0000 Subject: rtnetlink: do not depend on RTNL for IFLA_IFNAME output We can use netdev_copy_name() to no longer rely on RTNL to fetch dev->name. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni --- net/core/rtnetlink.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index ba5301d6df84..eb1a151c6910 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1812,6 +1812,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, u32 event, int *new_nsid, int new_ifindex, int tgt_netnsid, gfp_t gfp) { + char devname[IFNAMSIZ]; struct ifinfomsg *ifm; struct nlmsghdr *nlh; struct Qdisc *qdisc; @@ -1832,8 +1833,11 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, if (tgt_netnsid >= 0 && nla_put_s32(skb, IFLA_TARGET_NETNSID, tgt_netnsid)) goto nla_put_failure; - if (nla_put_string(skb, IFLA_IFNAME, dev->name) || - nla_put_u32(skb, IFLA_TXQLEN, dev->tx_queue_len) || + netdev_copy_name(dev, devname); + if (nla_put_string(skb, IFLA_IFNAME, devname)) + goto nla_put_failure; + + if (nla_put_u32(skb, IFLA_TXQLEN, dev->tx_queue_len) || nla_put_u8(skb, IFLA_OPERSTATE, netif_running(dev) ? dev->operstate : IF_OPER_DOWN) || nla_put_u8(skb, IFLA_LINKMODE, dev->link_mode) || -- cgit v1.2.3 From ad13b5b0d1f9eb8e048394919e6393e520b14552 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:54 +0000 Subject: rtnetlink: do not depend on RTNL for IFLA_TXQLEN output rtnl_fill_ifinfo() can read dev->tx_queue_len locklessly, granted we add corresponding READ_ONCE()/WRITE_ONCE() annotations. Add missing READ_ONCE(dev->tx_queue_len) in teql_enqueue() Signed-off-by: Eric Dumazet Signed-off-by: Paolo Abeni --- net/core/dev.c | 4 ++-- net/core/rtnetlink.c | 2 +- net/sched/sch_api.c | 2 +- net/sched/sch_teql.c | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index d6b24749eb2e..882ce6375b2e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -8959,7 +8959,7 @@ int dev_change_tx_queue_len(struct net_device *dev, unsigned long new_len) return -ERANGE; if (new_len != orig_len) { - dev->tx_queue_len = new_len; + WRITE_ONCE(dev->tx_queue_len, new_len); res = call_netdevice_notifiers(NETDEV_CHANGE_TX_QUEUE_LEN, dev); res = notifier_to_errno(res); if (res) @@ -8973,7 +8973,7 @@ int dev_change_tx_queue_len(struct net_device *dev, unsigned long new_len) err_rollback: netdev_err(dev, "refused to change device tx_queue_len\n"); - dev->tx_queue_len = orig_len; + WRITE_ONCE(dev->tx_queue_len, orig_len); return res; } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index eb1a151c6910..7e156f535bd8 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1837,7 +1837,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, if (nla_put_string(skb, IFLA_IFNAME, devname)) goto nla_put_failure; - if (nla_put_u32(skb, IFLA_TXQLEN, dev->tx_queue_len) || + if (nla_put_u32(skb, IFLA_TXQLEN, READ_ONCE(dev->tx_queue_len)) || nla_put_u8(skb, IFLA_OPERSTATE, netif_running(dev) ? dev->operstate : IF_OPER_DOWN) || nla_put_u8(skb, IFLA_LINKMODE, dev->link_mode) || diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 6292d6d73b72..74afc210527d 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1334,7 +1334,7 @@ static struct Qdisc *qdisc_create(struct net_device *dev, * before again attaching a qdisc. */ if ((dev->priv_flags & IFF_NO_QUEUE) && (dev->tx_queue_len == 0)) { - dev->tx_queue_len = DEFAULT_TX_QUEUE_LEN; + WRITE_ONCE(dev->tx_queue_len, DEFAULT_TX_QUEUE_LEN); netdev_info(dev, "Caught tx_queue_len zero misconfig\n"); } diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 59304611dc00..29850d0f0733 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -78,7 +78,7 @@ teql_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) struct net_device *dev = qdisc_dev(sch); struct teql_sched_data *q = qdisc_priv(sch); - if (q->q.qlen < dev->tx_queue_len) { + if (q->q.qlen < READ_ONCE(dev->tx_queue_len)) { __skb_queue_tail(&q->q, skb); return NET_XMIT_SUCCESS; } -- cgit v1.2.3 From 55a2c86c8db3d7aa2c1967efd37ed47d5ae37f43 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:55 +0000 Subject: net: write once on dev->allmulti and dev->promiscuity In the following patch we want to read dev->allmulti and dev->promiscuity locklessly from rtnl_fill_ifinfo() In this patch I change __dev_set_promiscuity() and __dev_set_allmulti() to write these fields (and dev->flags) only if they succeed, with WRITE_ONCE() annotations. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni --- net/core/dev.c | 36 ++++++++++++++++++++---------------- 1 file changed, 20 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index 882ce6375b2e..d9d9a0d91593 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -8544,27 +8544,29 @@ static void dev_change_rx_flags(struct net_device *dev, int flags) static int __dev_set_promiscuity(struct net_device *dev, int inc, bool notify) { unsigned int old_flags = dev->flags; + unsigned int promiscuity, flags; kuid_t uid; kgid_t gid; ASSERT_RTNL(); - dev->flags |= IFF_PROMISC; - dev->promiscuity += inc; - if (dev->promiscuity == 0) { + promiscuity = dev->promiscuity + inc; + if (promiscuity == 0) { /* * Avoid overflow. * If inc causes overflow, untouch promisc and return error. */ - if (inc < 0) - dev->flags &= ~IFF_PROMISC; - else { - dev->promiscuity -= inc; + if (unlikely(inc > 0)) { netdev_warn(dev, "promiscuity touches roof, set promiscuity failed. promiscuity feature of device might be broken.\n"); return -EOVERFLOW; } + flags = old_flags & ~IFF_PROMISC; + } else { + flags = old_flags | IFF_PROMISC; } - if (dev->flags != old_flags) { + WRITE_ONCE(dev->promiscuity, promiscuity); + if (flags != old_flags) { + WRITE_ONCE(dev->flags, flags); netdev_info(dev, "%s promiscuous mode\n", dev->flags & IFF_PROMISC ? "entered" : "left"); if (audit_enabled) { @@ -8615,25 +8617,27 @@ EXPORT_SYMBOL(dev_set_promiscuity); static int __dev_set_allmulti(struct net_device *dev, int inc, bool notify) { unsigned int old_flags = dev->flags, old_gflags = dev->gflags; + unsigned int allmulti, flags; ASSERT_RTNL(); - dev->flags |= IFF_ALLMULTI; - dev->allmulti += inc; - if (dev->allmulti == 0) { + allmulti = dev->allmulti + inc; + if (allmulti == 0) { /* * Avoid overflow. * If inc causes overflow, untouch allmulti and return error. */ - if (inc < 0) - dev->flags &= ~IFF_ALLMULTI; - else { - dev->allmulti -= inc; + if (unlikely(inc > 0)) { netdev_warn(dev, "allmulti touches roof, set allmulti failed. allmulti feature of device might be broken.\n"); return -EOVERFLOW; } + flags = old_flags & ~IFF_ALLMULTI; + } else { + flags = old_flags | IFF_ALLMULTI; } - if (dev->flags ^ old_flags) { + WRITE_ONCE(dev->allmulti, allmulti); + if (flags != old_flags) { + WRITE_ONCE(dev->flags, flags); netdev_info(dev, "%s allmulticast mode\n", dev->flags & IFF_ALLMULTI ? "entered" : "left"); dev_change_rx_flags(dev, IFF_ALLMULTI); -- cgit v1.2.3 From 6747a5d4990b8c8d7392f7a06b7a4bb5f4ada80e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:56 +0000 Subject: rtnetlink: do not depend on RTNL for many attributes Following device fields can be read locklessly in rtnl_fill_ifinfo() : type, ifindex, operstate, link_mode, mtu, min_mtu, max_mtu, group, promiscuity, allmulti, num_tx_queues, gso_max_segs, gso_max_size, gro_max_size, gso_ipv4_max_size, gro_ipv4_max_size, tso_max_size, tso_max_segs, num_rx_queues. Signed-off-by: Eric Dumazet Signed-off-by: Paolo Abeni --- net/core/rtnetlink.c | 51 +++++++++++++++++++++++++++++++-------------------- 1 file changed, 31 insertions(+), 20 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 7e156f535bd8..16b5f9c68bbf 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1603,7 +1603,8 @@ static int put_master_ifindex(struct sk_buff *skb, struct net_device *dev) upper_dev = netdev_master_upper_dev_get_rcu(dev); if (upper_dev) - ret = nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex); + ret = nla_put_u32(skb, IFLA_MASTER, + READ_ONCE(upper_dev->ifindex)); rcu_read_unlock(); return ret; @@ -1825,8 +1826,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, ifm = nlmsg_data(nlh); ifm->ifi_family = AF_UNSPEC; ifm->__ifi_pad = 0; - ifm->ifi_type = dev->type; - ifm->ifi_index = dev->ifindex; + ifm->ifi_type = READ_ONCE(dev->type); + ifm->ifi_index = READ_ONCE(dev->ifindex); ifm->ifi_flags = dev_get_flags(dev); ifm->ifi_change = change; @@ -1839,24 +1840,34 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, if (nla_put_u32(skb, IFLA_TXQLEN, READ_ONCE(dev->tx_queue_len)) || nla_put_u8(skb, IFLA_OPERSTATE, - netif_running(dev) ? dev->operstate : IF_OPER_DOWN) || - nla_put_u8(skb, IFLA_LINKMODE, dev->link_mode) || - nla_put_u32(skb, IFLA_MTU, dev->mtu) || - nla_put_u32(skb, IFLA_MIN_MTU, dev->min_mtu) || - nla_put_u32(skb, IFLA_MAX_MTU, dev->max_mtu) || - nla_put_u32(skb, IFLA_GROUP, dev->group) || - nla_put_u32(skb, IFLA_PROMISCUITY, dev->promiscuity) || - nla_put_u32(skb, IFLA_ALLMULTI, dev->allmulti) || - nla_put_u32(skb, IFLA_NUM_TX_QUEUES, dev->num_tx_queues) || - nla_put_u32(skb, IFLA_GSO_MAX_SEGS, dev->gso_max_segs) || - nla_put_u32(skb, IFLA_GSO_MAX_SIZE, dev->gso_max_size) || - nla_put_u32(skb, IFLA_GRO_MAX_SIZE, dev->gro_max_size) || - nla_put_u32(skb, IFLA_GSO_IPV4_MAX_SIZE, dev->gso_ipv4_max_size) || - nla_put_u32(skb, IFLA_GRO_IPV4_MAX_SIZE, dev->gro_ipv4_max_size) || - nla_put_u32(skb, IFLA_TSO_MAX_SIZE, dev->tso_max_size) || - nla_put_u32(skb, IFLA_TSO_MAX_SEGS, dev->tso_max_segs) || + netif_running(dev) ? READ_ONCE(dev->operstate) : + IF_OPER_DOWN) || + nla_put_u8(skb, IFLA_LINKMODE, READ_ONCE(dev->link_mode)) || + nla_put_u32(skb, IFLA_MTU, READ_ONCE(dev->mtu)) || + nla_put_u32(skb, IFLA_MIN_MTU, READ_ONCE(dev->min_mtu)) || + nla_put_u32(skb, IFLA_MAX_MTU, READ_ONCE(dev->max_mtu)) || + nla_put_u32(skb, IFLA_GROUP, READ_ONCE(dev->group)) || + nla_put_u32(skb, IFLA_PROMISCUITY, READ_ONCE(dev->promiscuity)) || + nla_put_u32(skb, IFLA_ALLMULTI, READ_ONCE(dev->allmulti)) || + nla_put_u32(skb, IFLA_NUM_TX_QUEUES, + READ_ONCE(dev->num_tx_queues)) || + nla_put_u32(skb, IFLA_GSO_MAX_SEGS, + READ_ONCE(dev->gso_max_segs)) || + nla_put_u32(skb, IFLA_GSO_MAX_SIZE, + READ_ONCE(dev->gso_max_size)) || + nla_put_u32(skb, IFLA_GRO_MAX_SIZE, + READ_ONCE(dev->gro_max_size)) || + nla_put_u32(skb, IFLA_GSO_IPV4_MAX_SIZE, + READ_ONCE(dev->gso_ipv4_max_size)) || + nla_put_u32(skb, IFLA_GRO_IPV4_MAX_SIZE, + READ_ONCE(dev->gro_ipv4_max_size)) || + nla_put_u32(skb, IFLA_TSO_MAX_SIZE, + READ_ONCE(dev->tso_max_size)) || + nla_put_u32(skb, IFLA_TSO_MAX_SEGS, + READ_ONCE(dev->tso_max_segs)) || #ifdef CONFIG_RPS - nla_put_u32(skb, IFLA_NUM_RX_QUEUES, dev->num_rx_queues) || + nla_put_u32(skb, IFLA_NUM_RX_QUEUES, + READ_ONCE(dev->num_rx_queues)) || #endif put_master_ifindex(skb, dev) || nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) || -- cgit v1.2.3 From 6890ab31d1a35444741e6150db19d64797db2919 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:57 +0000 Subject: rtnetlink: do not depend on RTNL in rtnl_fill_proto_down() Change dev_change_proto_down() and dev_change_proto_down_reason() to write once on dev->proto_down and dev->proto_down_reason. Then rtnl_fill_proto_down() can use READ_ONCE() annotations and run locklessly. rtnl_proto_down_size() should assume worst case, because readng dev->proto_down_reason multiple times would be racy without RTNL in the future. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni --- net/core/dev.c | 11 +++++++---- net/core/rtnetlink.c | 8 ++++---- 2 files changed, 11 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/core/dev.c b/net/core/dev.c index d9d9a0d91593..d2ce91a334c1 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -9223,7 +9223,7 @@ int dev_change_proto_down(struct net_device *dev, bool proto_down) netif_carrier_off(dev); else netif_carrier_on(dev); - dev->proto_down = proto_down; + WRITE_ONCE(dev->proto_down, proto_down); return 0; } @@ -9237,18 +9237,21 @@ int dev_change_proto_down(struct net_device *dev, bool proto_down) void dev_change_proto_down_reason(struct net_device *dev, unsigned long mask, u32 value) { + u32 proto_down_reason; int b; if (!mask) { - dev->proto_down_reason = value; + proto_down_reason = value; } else { + proto_down_reason = dev->proto_down_reason; for_each_set_bit(b, &mask, 32) { if (value & (1 << b)) - dev->proto_down_reason |= BIT(b); + proto_down_reason |= BIT(b); else - dev->proto_down_reason &= ~BIT(b); + proto_down_reason &= ~BIT(b); } } + WRITE_ONCE(dev->proto_down_reason, proto_down_reason); } struct bpf_xdp_link { diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 16b5f9c68bbf..dfdd08006fae 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1036,8 +1036,8 @@ static size_t rtnl_proto_down_size(const struct net_device *dev) { size_t size = nla_total_size(1); - if (dev->proto_down_reason) - size += nla_total_size(0) + nla_total_size(4); + /* Assume dev->proto_down_reason is not zero. */ + size += nla_total_size(0) + nla_total_size(4); return size; } @@ -1737,10 +1737,10 @@ static int rtnl_fill_proto_down(struct sk_buff *skb, struct nlattr *pr; u32 preason; - if (nla_put_u8(skb, IFLA_PROTO_DOWN, dev->proto_down)) + if (nla_put_u8(skb, IFLA_PROTO_DOWN, READ_ONCE(dev->proto_down))) goto nla_put_failure; - preason = dev->proto_down_reason; + preason = READ_ONCE(dev->proto_down_reason); if (!preason) return 0; -- cgit v1.2.3 From 979aad40da9217d5e907ee4ad7c7f0dc555944a7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:58 +0000 Subject: rtnetlink: do not depend on RTNL in rtnl_xdp_prog_skb() dev->xdp_prog is protected by RCU, we can lift RTNL requirement from rtnl_xdp_prog_skb(). Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni --- net/core/rtnetlink.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index dfdd08006fae..41eb8bca5305 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1477,13 +1477,15 @@ static int rtnl_fill_link_ifmap(struct sk_buff *skb, static u32 rtnl_xdp_prog_skb(struct net_device *dev) { const struct bpf_prog *generic_xdp_prog; + u32 res = 0; - ASSERT_RTNL(); + rcu_read_lock(); + generic_xdp_prog = rcu_dereference(dev->xdp_prog); + if (generic_xdp_prog) + res = generic_xdp_prog->aux->id; + rcu_read_unlock(); - generic_xdp_prog = rtnl_dereference(dev->xdp_prog); - if (!generic_xdp_prog) - return 0; - return generic_xdp_prog->aux->id; + return res; } static u32 rtnl_xdp_prog_drv(struct net_device *dev) -- cgit v1.2.3 From 9cf621bd5fcbeadc2804951d13d487e22e95b363 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 May 2024 19:20:59 +0000 Subject: rtnetlink: allow rtnl_fill_link_netnsid() to run under RCU protection We want to be able to run rtnl_fill_ifinfo() under RCU protection instead of RTNL in the future. All rtnl_link_ops->get_link_net() methods already using dev_net() are ready. I added READ_ONCE() annotations on others. Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Signed-off-by: Paolo Abeni --- drivers/net/ppp/ppp_generic.c | 2 +- drivers/net/vxlan/vxlan_core.c | 2 +- net/core/rtnetlink.c | 5 ++--- net/ipv4/ip_tunnel.c | 2 +- net/ipv6/ip6_tunnel.c | 2 +- net/xfrm/xfrm_interface_core.c | 2 +- 6 files changed, 7 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index fe380fe196e7..0a65b6d690fe 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1357,7 +1357,7 @@ static struct net *ppp_nl_get_link_net(const struct net_device *dev) { struct ppp *ppp = netdev_priv(dev); - return ppp->ppp_net; + return READ_ONCE(ppp->ppp_net); } static struct rtnl_link_ops ppp_link_ops __read_mostly = { diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index 8884913e0473..7e3a7d1f2018 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -4569,7 +4569,7 @@ static struct net *vxlan_get_link_net(const struct net_device *dev) { struct vxlan_dev *vxlan = netdev_priv(dev); - return vxlan->net; + return READ_ONCE(vxlan->net); } static struct rtnl_link_ops vxlan_link_ops __read_mostly = { diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 41eb8bca5305..af8da8aeed39 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1923,9 +1923,6 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, goto nla_put_failure; } - if (rtnl_fill_link_netnsid(skb, dev, src_net, gfp)) - goto nla_put_failure; - if (new_nsid && nla_put_s32(skb, IFLA_NEW_NETNSID, *new_nsid) < 0) goto nla_put_failure; @@ -1938,6 +1935,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, goto nla_put_failure; rcu_read_lock(); + if (rtnl_fill_link_netnsid(skb, dev, src_net, GFP_ATOMIC)) + goto nla_put_failure_rcu; qdisc = rcu_dereference(dev->qdisc); if (qdisc && nla_put_string(skb, IFLA_QDISC, qdisc->ops->id)) goto nla_put_failure_rcu; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index ba46cf7612f4..f1c5f6c3f2f8 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -1120,7 +1120,7 @@ struct net *ip_tunnel_get_link_net(const struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); - return tunnel->net; + return READ_ONCE(tunnel->net); } EXPORT_SYMBOL(ip_tunnel_get_link_net); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 57bb3b3ea0c5..5aec79c2af1a 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -2146,7 +2146,7 @@ struct net *ip6_tnl_get_link_net(const struct net_device *dev) { struct ip6_tnl *tunnel = netdev_priv(dev); - return tunnel->net; + return READ_ONCE(tunnel->net); } EXPORT_SYMBOL(ip6_tnl_get_link_net); diff --git a/net/xfrm/xfrm_interface_core.c b/net/xfrm/xfrm_interface_core.c index 4df5c06e3ece..e50e4bf993fa 100644 --- a/net/xfrm/xfrm_interface_core.c +++ b/net/xfrm/xfrm_interface_core.c @@ -926,7 +926,7 @@ static struct net *xfrmi_get_link_net(const struct net_device *dev) { struct xfrm_if *xi = netdev_priv(dev); - return xi->net; + return READ_ONCE(xi->net); } static const struct nla_policy xfrmi_policy[IFLA_XFRM_MAX + 1] = { -- cgit v1.2.3 From 1f20a5769446a1acae67ac9e63d07a594829a789 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Tue, 7 May 2024 13:20:23 +0200 Subject: page_pool: make sure frag API fields don't span between cachelines After commit 5027ec19f104 ("net: page_pool: split the page_pool_params into fast and slow") that made &page_pool contain only "hot" params at the start, cacheline boundary chops frag API fields group in the middle again. To not bother with this each time fast params get expanded or shrunk, let's just align them to `4 * sizeof(long)`, the closest upper pow-2 to their actual size (2 longs + 1 int). This ensures 16-byte alignment for the 32-bit architectures and 32-byte alignment for the 64-bit ones, excluding unnecessary false-sharing. ::page_state_hold_cnt is used quite intensively on hotpath no matter if frag API is used, so move it to the newly created hole in the first cacheline. Signed-off-by: Alexander Lobakin Signed-off-by: Christoph Hellwig --- include/net/page_pool/types.h | 12 +++++++++++- net/core/page_pool.c | 10 ++++++++++ 2 files changed, 21 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h index 5e43a08d3231..5460cbab5de0 100644 --- a/include/net/page_pool/types.h +++ b/include/net/page_pool/types.h @@ -130,12 +130,22 @@ struct page_pool { struct page_pool_params_fast p; int cpuid; + u32 pages_state_hold_cnt; bool has_init_callback; + /* The following block must stay within one cacheline. On 32-bit + * systems, sizeof(long) == sizeof(int), so that the block size is + * ``3 * sizeof(long)``. On 64-bit systems, the actual size is + * ``2 * sizeof(long) + sizeof(int)``. The closest pow-2 to both of + * them is ``4 * sizeof(long)``, so just use that one for simplicity. + * Having it aligned to a cacheline boundary may be excessive and + * doesn't bring any good. + */ + __cacheline_group_begin(frag) __aligned(4 * sizeof(long)); long frag_users; struct page *frag_page; unsigned int frag_offset; - u32 pages_state_hold_cnt; + __cacheline_group_end(frag); struct delayed_work release_dw; void (*disconnect)(void *pool); diff --git a/net/core/page_pool.c b/net/core/page_pool.c index dd364d738c00..95eac12e8790 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -172,12 +172,22 @@ static void page_pool_producer_unlock(struct page_pool *pool, spin_unlock_bh(&pool->ring.producer_lock); } +static void page_pool_struct_check(void) +{ + CACHELINE_ASSERT_GROUP_MEMBER(struct page_pool, frag, frag_users); + CACHELINE_ASSERT_GROUP_MEMBER(struct page_pool, frag, frag_page); + CACHELINE_ASSERT_GROUP_MEMBER(struct page_pool, frag, frag_offset); + CACHELINE_ASSERT_GROUP_SIZE(struct page_pool, frag, 4 * sizeof(long)); +} + static int page_pool_init(struct page_pool *pool, const struct page_pool_params *params, int cpuid) { unsigned int ring_qsize = 1024; /* Default */ + page_pool_struct_check(); + memcpy(&pool->p, ¶ms->fast, sizeof(pool->p)); memcpy(&pool->slow, ¶ms->slow, sizeof(pool->slow)); -- cgit v1.2.3 From 403f11ac9ab72fc3bee0b8c80c16e33212ea8cd9 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Tue, 7 May 2024 13:20:24 +0200 Subject: page_pool: don't use driver-set flags field directly page_pool::p is driver-defined params, copied directly from the structure passed to page_pool_create(). The structure isn't meant to be modified by the Page Pool core code and this even might look confusing[0][1]. In order to be able to alter some flags, let's define our own, internal fields the same way as the already existing one (::has_init_callback). They are defined as bits in the driver-set params, leave them so here as well, to not waste byte-per-bit or so. Almost 30 bits are still free for future extensions. We could've defined only new flags here or only the ones we may need to alter, but checking some flags in one place while others in another doesn't sound convenient or intuitive. ::flags passed by the driver can now go to the "slow" PP params. Suggested-by: Jakub Kicinski Link[0]: https://lore.kernel.org/netdev/20230703133207.4f0c54ce@kernel.org Suggested-by: Alexander Duyck Link[1]: https://lore.kernel.org/netdev/CAKgT0UfZCGnWgOH96E4GV3ZP6LLbROHM7SHE8NKwq+exX+Gk_Q@mail.gmail.com Signed-off-by: Alexander Lobakin Signed-off-by: Christoph Hellwig --- include/net/page_pool/types.h | 13 ++++++++++--- net/core/page_pool.c | 41 ++++++++++++++++++++++------------------- 2 files changed, 32 insertions(+), 22 deletions(-) (limited to 'net') diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h index 5460cbab5de0..8c6d9f16bf65 100644 --- a/include/net/page_pool/types.h +++ b/include/net/page_pool/types.h @@ -45,7 +45,6 @@ struct pp_alloc_cache { /** * struct page_pool_params - page pool parameters - * @flags: PP_FLAG_DMA_MAP, PP_FLAG_DMA_SYNC_DEV * @order: 2^order pages on allocation * @pool_size: size of the ptr_ring * @nid: NUMA node id to allocate from pages from @@ -55,10 +54,11 @@ struct pp_alloc_cache { * @dma_dir: DMA mapping direction * @max_len: max DMA sync memory size for PP_FLAG_DMA_SYNC_DEV * @offset: DMA sync address offset for PP_FLAG_DMA_SYNC_DEV + * @netdev: corresponding &net_device for Netlink introspection + * @flags: PP_FLAG_DMA_MAP, PP_FLAG_DMA_SYNC_DEV, PP_FLAG_SYSTEM_POOL */ struct page_pool_params { struct_group_tagged(page_pool_params_fast, fast, - unsigned int flags; unsigned int order; unsigned int pool_size; int nid; @@ -70,6 +70,7 @@ struct page_pool_params { ); struct_group_tagged(page_pool_params_slow, slow, struct net_device *netdev; + unsigned int flags; /* private: used by test code only */ void (*init_callback)(struct page *page, void *arg); void *init_arg; @@ -131,7 +132,13 @@ struct page_pool { int cpuid; u32 pages_state_hold_cnt; - bool has_init_callback; + + bool has_init_callback:1; /* slow::init_callback is set */ + bool dma_map:1; /* Perform DMA mapping */ + bool dma_sync:1; /* Perform DMA sync */ +#ifdef CONFIG_PAGE_POOL_STATS + bool system:1; /* This is a global percpu pool */ +#endif /* The following block must stay within one cacheline. On 32-bit * systems, sizeof(long) == sizeof(int), so that the block size is diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 95eac12e8790..c2819ff03dd2 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -194,7 +194,7 @@ static int page_pool_init(struct page_pool *pool, pool->cpuid = cpuid; /* Validate only known flags were used */ - if (pool->p.flags & ~(PP_FLAG_ALL)) + if (pool->slow.flags & ~PP_FLAG_ALL) return -EINVAL; if (pool->p.pool_size) @@ -208,22 +208,26 @@ static int page_pool_init(struct page_pool *pool, * DMA_BIDIRECTIONAL is for allowing page used for DMA sending, * which is the XDP_TX use-case. */ - if (pool->p.flags & PP_FLAG_DMA_MAP) { + if (pool->slow.flags & PP_FLAG_DMA_MAP) { if ((pool->p.dma_dir != DMA_FROM_DEVICE) && (pool->p.dma_dir != DMA_BIDIRECTIONAL)) return -EINVAL; + + pool->dma_map = true; } - if (pool->p.flags & PP_FLAG_DMA_SYNC_DEV) { + if (pool->slow.flags & PP_FLAG_DMA_SYNC_DEV) { /* In order to request DMA-sync-for-device the page * needs to be mapped */ - if (!(pool->p.flags & PP_FLAG_DMA_MAP)) + if (!(pool->slow.flags & PP_FLAG_DMA_MAP)) return -EINVAL; if (!pool->p.max_len) return -EINVAL; + pool->dma_sync = true; + /* pool->p.offset has to be set according to the address * offset used by the DMA engine to start copying rx data */ @@ -232,7 +236,7 @@ static int page_pool_init(struct page_pool *pool, pool->has_init_callback = !!pool->slow.init_callback; #ifdef CONFIG_PAGE_POOL_STATS - if (!(pool->p.flags & PP_FLAG_SYSTEM_POOL)) { + if (!(pool->slow.flags & PP_FLAG_SYSTEM_POOL)) { pool->recycle_stats = alloc_percpu(struct page_pool_recycle_stats); if (!pool->recycle_stats) return -ENOMEM; @@ -242,12 +246,13 @@ static int page_pool_init(struct page_pool *pool, * (also percpu) page pool instance. */ pool->recycle_stats = &pp_system_recycle_stats; + pool->system = true; } #endif if (ptr_ring_init(&pool->ring, ring_qsize, GFP_KERNEL) < 0) { #ifdef CONFIG_PAGE_POOL_STATS - if (!(pool->p.flags & PP_FLAG_SYSTEM_POOL)) + if (!pool->system) free_percpu(pool->recycle_stats); #endif return -ENOMEM; @@ -258,7 +263,7 @@ static int page_pool_init(struct page_pool *pool, /* Driver calling page_pool_create() also call page_pool_destroy() */ refcount_set(&pool->user_cnt, 1); - if (pool->p.flags & PP_FLAG_DMA_MAP) + if (pool->dma_map) get_device(pool->p.dev); return 0; @@ -268,11 +273,11 @@ static void page_pool_uninit(struct page_pool *pool) { ptr_ring_cleanup(&pool->ring, NULL); - if (pool->p.flags & PP_FLAG_DMA_MAP) + if (pool->dma_map) put_device(pool->p.dev); #ifdef CONFIG_PAGE_POOL_STATS - if (!(pool->p.flags & PP_FLAG_SYSTEM_POOL)) + if (!pool->system) free_percpu(pool->recycle_stats); #endif } @@ -424,7 +429,7 @@ static bool page_pool_dma_map(struct page_pool *pool, struct page *page) if (page_pool_set_dma_addr(page, dma)) goto unmap_failed; - if (pool->p.flags & PP_FLAG_DMA_SYNC_DEV) + if (pool->dma_sync) page_pool_dma_sync_for_device(pool, page, pool->p.max_len); return true; @@ -470,8 +475,7 @@ static struct page *__page_pool_alloc_page_order(struct page_pool *pool, if (unlikely(!page)) return NULL; - if ((pool->p.flags & PP_FLAG_DMA_MAP) && - unlikely(!page_pool_dma_map(pool, page))) { + if (pool->dma_map && unlikely(!page_pool_dma_map(pool, page))) { put_page(page); return NULL; } @@ -491,8 +495,8 @@ static struct page *__page_pool_alloc_pages_slow(struct page_pool *pool, gfp_t gfp) { const int bulk = PP_ALLOC_CACHE_REFILL; - unsigned int pp_flags = pool->p.flags; unsigned int pp_order = pool->p.order; + bool dma_map = pool->dma_map; struct page *page; int i, nr_pages; @@ -517,8 +521,7 @@ static struct page *__page_pool_alloc_pages_slow(struct page_pool *pool, */ for (i = 0; i < nr_pages; i++) { page = pool->alloc.cache[i]; - if ((pp_flags & PP_FLAG_DMA_MAP) && - unlikely(!page_pool_dma_map(pool, page))) { + if (dma_map && unlikely(!page_pool_dma_map(pool, page))) { put_page(page); continue; } @@ -590,7 +593,7 @@ void __page_pool_release_page_dma(struct page_pool *pool, struct page *page) { dma_addr_t dma; - if (!(pool->p.flags & PP_FLAG_DMA_MAP)) + if (!pool->dma_map) /* Always account for inflight pages, even if we didn't * map them */ @@ -673,7 +676,7 @@ static bool __page_pool_page_can_be_recycled(const struct page *page) } /* If the page refcnt == 1, this will try to recycle the page. - * if PP_FLAG_DMA_SYNC_DEV is set, we'll try to sync the DMA area for + * If pool->dma_sync is set, we'll try to sync the DMA area for * the configured size min(dma_sync_size, pool->max_len). * If the page refcnt != 1, then the page will be returned to memory * subsystem. @@ -696,7 +699,7 @@ __page_pool_put_page(struct page_pool *pool, struct page *page, if (likely(__page_pool_page_can_be_recycled(page))) { /* Read barrier done in page_ref_count / READ_ONCE */ - if (pool->p.flags & PP_FLAG_DMA_SYNC_DEV) + if (pool->dma_sync) page_pool_dma_sync_for_device(pool, page, dma_sync_size); @@ -809,7 +812,7 @@ static struct page *page_pool_drain_frag(struct page_pool *pool, return NULL; if (__page_pool_page_can_be_recycled(page)) { - if (pool->p.flags & PP_FLAG_DMA_SYNC_DEV) + if (pool->dma_sync) page_pool_dma_sync_for_device(pool, page, -1); return page; -- cgit v1.2.3 From feb8c2b76eb3fae59a1f62451508c98c36db5e3a Mon Sep 17 00:00:00 2001 From: Jeff Johnson Date: Sun, 5 May 2024 13:09:31 -0700 Subject: net: dccp: Fix ccid2_rtt_estimator() kernel-doc make C=1 reports: warning: Function parameter or struct member 'mrtt' not described in 'ccid2_rtt_estimator' So document the 'mrtt' parameter. Signed-off-by: Jeff Johnson Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240505-ccid2_rtt_estimator-kdoc-v1-1-09231fcb9145@quicinc.com Signed-off-by: Jakub Kicinski --- net/dccp/ccids/ccid2.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 4d9823d6dced..d6b30700af67 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -353,6 +353,7 @@ static void ccid2_hc_tx_packet_sent(struct sock *sk, unsigned int len) /** * ccid2_rtt_estimator - Sample RTT and compute RTO using RFC2988 algorithm * @sk: socket to perform estimator on + * @mrtt: measured RTT * * This code is almost identical with TCP's tcp_rtt_estimator(), since * - it has a higher sampling frequency (recommended by RFC 1323), -- cgit v1.2.3 From 1eb2cded45b35816085c1f962933c187d970f9dc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 May 2024 10:28:12 +0000 Subject: net: annotate writes on dev->mtu from ndo_change_mtu() Simon reported that ndo_change_mtu() methods were never updated to use WRITE_ONCE(dev->mtu, new_mtu) as hinted in commit 501a90c94510 ("inet: protect against too small mtu values.") We read dev->mtu without holding RTNL in many places, with READ_ONCE() annotations. It is time to take care of ndo_change_mtu() methods to use corresponding WRITE_ONCE() Signed-off-by: Eric Dumazet Reported-by: Simon Horman Closes: https://lore.kernel.org/netdev/20240505144608.GB67882@kernel.org/ Reviewed-by: Jacob Keller Reviewed-by: Sabrina Dubroca Reviewed-by: Simon Horman Acked-by: Shannon Nelson Link: https://lore.kernel.org/r/20240506102812.3025432-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/infiniband/ulp/ipoib/ipoib_main.c | 4 ++-- drivers/net/bonding/bond_main.c | 2 +- drivers/net/can/dev/dev.c | 2 +- drivers/net/can/vcan.c | 2 +- drivers/net/can/vxcan.c | 2 +- drivers/net/ethernet/agere/et131x.c | 2 +- drivers/net/ethernet/alteon/acenic.c | 2 +- drivers/net/ethernet/altera/altera_tse_main.c | 2 +- drivers/net/ethernet/amazon/ena/ena_netdev.c | 2 +- drivers/net/ethernet/amd/amd8111e.c | 6 +++--- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 2 +- drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 2 +- drivers/net/ethernet/aquantia/atlantic/aq_main.c | 2 +- drivers/net/ethernet/atheros/ag71xx.c | 2 +- drivers/net/ethernet/atheros/alx/main.c | 2 +- drivers/net/ethernet/atheros/atl1c/atl1c_main.c | 2 +- drivers/net/ethernet/atheros/atl1e/atl1e_main.c | 2 +- drivers/net/ethernet/atheros/atlx/atl1.c | 2 +- drivers/net/ethernet/atheros/atlx/atl2.c | 2 +- drivers/net/ethernet/broadcom/b44.c | 4 ++-- drivers/net/ethernet/broadcom/bcm63xx_enet.c | 2 +- drivers/net/ethernet/broadcom/bnx2.c | 2 +- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- drivers/net/ethernet/broadcom/tg3.c | 2 +- drivers/net/ethernet/brocade/bna/bnad.c | 2 +- drivers/net/ethernet/cadence/macb_main.c | 2 +- drivers/net/ethernet/calxeda/xgmac.c | 2 +- drivers/net/ethernet/cavium/liquidio/lio_core.c | 2 +- drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c | 2 +- drivers/net/ethernet/cavium/octeon/octeon_mgmt.c | 2 +- drivers/net/ethernet/cavium/thunder/nicvf_main.c | 2 +- drivers/net/ethernet/chelsio/cxgb/cxgb2.c | 2 +- drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c | 2 +- drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 2 +- drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c | 2 +- drivers/net/ethernet/cisco/enic/enic_main.c | 2 +- drivers/net/ethernet/cortina/gemini.c | 2 +- drivers/net/ethernet/dlink/sundance.c | 2 +- drivers/net/ethernet/faraday/ftmac100.c | 2 +- drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 2 +- drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c | 2 +- drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c | 2 +- drivers/net/ethernet/freescale/gianfar.c | 2 +- drivers/net/ethernet/fungible/funeth/funeth_main.c | 2 +- drivers/net/ethernet/hisilicon/hns/hns_enet.c | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 2 +- drivers/net/ethernet/huawei/hinic/hinic_main.c | 2 +- drivers/net/ethernet/ibm/emac/core.c | 4 ++-- drivers/net/ethernet/ibm/ibmveth.c | 2 +- drivers/net/ethernet/intel/e1000/e1000_main.c | 2 +- drivers/net/ethernet/intel/e1000e/netdev.c | 2 +- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +- drivers/net/ethernet/intel/iavf/iavf_main.c | 2 +- drivers/net/ethernet/intel/ice/ice_main.c | 2 +- drivers/net/ethernet/intel/idpf/idpf_lib.c | 2 +- drivers/net/ethernet/intel/igb/igb_main.c | 2 +- drivers/net/ethernet/intel/igbvf/netdev.c | 2 +- drivers/net/ethernet/intel/igc/igc_main.c | 2 +- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 +- drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 2 +- drivers/net/ethernet/jme.c | 2 +- drivers/net/ethernet/lantiq_etop.c | 2 +- drivers/net/ethernet/lantiq_xrx200.c | 4 ++-- drivers/net/ethernet/marvell/mv643xx_eth.c | 2 +- drivers/net/ethernet/marvell/mvneta.c | 2 +- drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 2 +- drivers/net/ethernet/marvell/octeon_ep/octep_main.c | 2 +- drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c | 2 +- drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 2 +- drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c | 2 +- drivers/net/ethernet/marvell/prestera/prestera_main.c | 2 +- drivers/net/ethernet/marvell/pxa168_eth.c | 2 +- drivers/net/ethernet/marvell/skge.c | 4 ++-- drivers/net/ethernet/marvell/sky2.c | 4 ++-- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c | 2 +- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 2 +- drivers/net/ethernet/micrel/ksz884x.c | 2 +- drivers/net/ethernet/microchip/lan743x_main.c | 2 +- drivers/net/ethernet/microchip/lan966x/lan966x_main.c | 2 +- drivers/net/ethernet/microsoft/mana/mana_en.c | 4 ++-- drivers/net/ethernet/myricom/myri10ge/myri10ge.c | 8 ++++---- drivers/net/ethernet/natsemi/natsemi.c | 2 +- drivers/net/ethernet/neterion/s2io.c | 2 +- drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 2 +- drivers/net/ethernet/netronome/nfp/nfp_net_repr.c | 2 +- drivers/net/ethernet/ni/nixge.c | 2 +- drivers/net/ethernet/nvidia/forcedeth.c | 2 +- drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c | 2 +- drivers/net/ethernet/pasemi/pasemi_mac.c | 2 +- drivers/net/ethernet/pensando/ionic/ionic_lif.c | 4 ++-- drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c | 2 +- drivers/net/ethernet/qlogic/qede/qede_ethtool.c | 2 +- drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c | 2 +- drivers/net/ethernet/qualcomm/emac/emac.c | 2 +- drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c | 2 +- drivers/net/ethernet/realtek/8139cp.c | 4 ++-- drivers/net/ethernet/realtek/r8169_main.c | 2 +- drivers/net/ethernet/renesas/ravb_main.c | 2 +- drivers/net/ethernet/renesas/sh_eth.c | 2 +- drivers/net/ethernet/rocker/rocker_main.c | 2 +- drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c | 2 +- drivers/net/ethernet/sfc/efx_common.c | 2 +- drivers/net/ethernet/sfc/falcon/efx.c | 2 +- drivers/net/ethernet/sfc/siena/efx_common.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- drivers/net/ethernet/sun/cassini.c | 2 +- drivers/net/ethernet/sun/niu.c | 2 +- drivers/net/ethernet/sun/sungem.c | 2 +- drivers/net/ethernet/synopsys/dwc-xlgmac-net.c | 2 +- drivers/net/ethernet/tehuti/tehuti.c | 2 +- drivers/net/ethernet/via/via-velocity.c | 4 ++-- drivers/net/ethernet/wangxun/libwx/wx_hw.c | 2 +- drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 2 +- drivers/net/ethernet/xscale/ixp4xx_eth.c | 2 +- drivers/net/fjes/fjes_main.c | 2 +- drivers/net/geneve.c | 2 +- drivers/net/hyperv/netvsc_drv.c | 4 ++-- drivers/net/macsec.c | 2 +- drivers/net/macvlan.c | 2 +- drivers/net/net_failover.c | 2 +- drivers/net/netdevsim/netdev.c | 2 +- drivers/net/ntb_netdev.c | 4 ++-- drivers/net/slip/slip.c | 2 +- drivers/net/team/team_core.c | 2 +- drivers/net/usb/aqc111.c | 2 +- drivers/net/usb/asix_devices.c | 2 +- drivers/net/usb/ax88179_178a.c | 2 +- drivers/net/usb/cdc_ncm.c | 2 +- drivers/net/usb/lan78xx.c | 2 +- drivers/net/usb/r8152.c | 4 ++-- drivers/net/usb/usbnet.c | 2 +- drivers/net/vmxnet3/vmxnet3_drv.c | 2 +- drivers/net/vsockmon.c | 2 +- drivers/net/vxlan/vxlan_core.c | 2 +- drivers/net/xen-netback/interface.c | 2 +- drivers/net/xen-netfront.c | 2 +- drivers/s390/net/ctcm_main.c | 2 +- net/8021q/vlan_dev.c | 2 +- net/batman-adv/soft-interface.c | 2 +- net/bridge/br_device.c | 2 +- net/dsa/user.c | 2 +- net/hsr/hsr_device.c | 2 +- net/hsr/hsr_main.c | 2 +- net/ipv4/ip_gre.c | 2 +- net/ipv4/ip_tunnel.c | 4 ++-- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/ip6_vti.c | 3 ++- net/sched/sch_teql.c | 2 +- 153 files changed, 174 insertions(+), 173 deletions(-) (limited to 'net') diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 6f2a688fccbf..32f6a28b02b1 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -238,7 +238,7 @@ static int ipoib_change_mtu(struct net_device *dev, int new_mtu) ipoib_warn(priv, "mtu > %d will cause multicast packet drops.\n", priv->mcast_mtu); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } @@ -265,7 +265,7 @@ static int ipoib_change_mtu(struct net_device *dev, int new_mtu) if (carrier_status) netif_carrier_on(dev); } else { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); } return ret; diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index b3a7d60c3a5c..3b0c7758ea4f 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4710,7 +4710,7 @@ static int bond_change_mtu(struct net_device *bond_dev, int new_mtu) } } - bond_dev->mtu = new_mtu; + WRITE_ONCE(bond_dev->mtu, new_mtu); return 0; diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c index 3a3be5cdfc1f..83e724e0ab87 100644 --- a/drivers/net/can/dev/dev.c +++ b/drivers/net/can/dev/dev.c @@ -338,7 +338,7 @@ int can_change_mtu(struct net_device *dev, int new_mtu) return -EINVAL; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } EXPORT_SYMBOL_GPL(can_change_mtu); diff --git a/drivers/net/can/vcan.c b/drivers/net/can/vcan.c index 285635c23443..f67e85807100 100644 --- a/drivers/net/can/vcan.c +++ b/drivers/net/can/vcan.c @@ -140,7 +140,7 @@ static int vcan_change_mtu(struct net_device *dev, int new_mtu) !can_is_canxl_dev_mtu(new_mtu)) return -EINVAL; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/can/vxcan.c b/drivers/net/can/vxcan.c index f7fabba707ea..9e1b7d41005f 100644 --- a/drivers/net/can/vxcan.c +++ b/drivers/net/can/vxcan.c @@ -135,7 +135,7 @@ static int vxcan_change_mtu(struct net_device *dev, int new_mtu) !can_is_canxl_dev_mtu(new_mtu)) return -EINVAL; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index 3d9220f9c9fe..b325e0cef120 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -3852,7 +3852,7 @@ static int et131x_change_mtu(struct net_device *netdev, int new_mtu) et131x_disable_txrx(netdev); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); et131x_adapter_memory_free(adapter); diff --git a/drivers/net/ethernet/alteon/acenic.c b/drivers/net/ethernet/alteon/acenic.c index eafef84fe3be..3d8ac63132fb 100644 --- a/drivers/net/ethernet/alteon/acenic.c +++ b/drivers/net/ethernet/alteon/acenic.c @@ -2539,7 +2539,7 @@ static int ace_change_mtu(struct net_device *dev, int new_mtu) struct ace_regs __iomem *regs = ap->regs; writel(new_mtu + ETH_HLEN + 4, ®s->IfMtu); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (new_mtu > ACE_STD_MTU) { if (!(ap->jumbo)) { diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 1c8763be0e4b..3c112c18ae6a 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -788,7 +788,7 @@ static int tse_change_mtu(struct net_device *dev, int new_mtu) return -EBUSY; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); netdev_update_features(dev); return 0; diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index be5acfa41ee0..28eaedaf713d 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -104,7 +104,7 @@ static int ena_change_mtu(struct net_device *dev, int new_mtu) if (!ret) { netif_dbg(adapter, drv, dev, "Set MTU to %d\n", new_mtu); update_rx_ring_mtu(adapter, new_mtu); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); } else { netif_err(adapter, drv, dev, "Failed to set MTU to %d\n", new_mtu); diff --git a/drivers/net/ethernet/amd/amd8111e.c b/drivers/net/ethernet/amd/amd8111e.c index c1b5e9a94308..f64f96fa17cf 100644 --- a/drivers/net/ethernet/amd/amd8111e.c +++ b/drivers/net/ethernet/amd/amd8111e.c @@ -1520,9 +1520,9 @@ static int amd8111e_change_mtu(struct net_device *dev, int new_mtu) if (!netif_running(dev)) { /* new_mtu will be used - * when device starts netxt time + * when device starts next time */ - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } @@ -1531,7 +1531,7 @@ static int amd8111e_change_mtu(struct net_device *dev, int new_mtu) /* stop the chip */ writel(RUN, lp->mmio + CMD0); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); err = amd8111e_restart(dev); spin_unlock_irq(&lp->lock); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index 6b73648b3779..c4a4e316683f 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -2070,7 +2070,7 @@ static int xgbe_change_mtu(struct net_device *netdev, int mtu) return ret; pdata->rx_buf_size = ret; - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); xgbe_restart_dev(pdata); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 44900026d11b..4af9d89d5f88 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -1530,7 +1530,7 @@ static int xgene_change_mtu(struct net_device *ndev, int new_mtu) frame_size = (new_mtu > ETH_DATA_LEN) ? (new_mtu + 18) : 0x600; xgene_enet_close(ndev); - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); pdata->mac_ops->set_framesize(pdata, frame_size); xgene_enet_open(ndev); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c index 0b2a52199914..c1d1673c5749 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c @@ -146,7 +146,7 @@ static int aq_ndev_change_mtu(struct net_device *ndev, int new_mtu) if (err < 0) goto err_exit; - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); err_exit: return err; diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c index 0f2f400b5bc4..a38be924cdaa 100644 --- a/drivers/net/ethernet/atheros/ag71xx.c +++ b/drivers/net/ethernet/atheros/ag71xx.c @@ -1788,7 +1788,7 @@ static int ag71xx_change_mtu(struct net_device *ndev, int new_mtu) { struct ag71xx *ag = netdev_priv(ndev); - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); ag71xx_wr(ag, AG71XX_REG_MAC_MFL, ag71xx_max_frame_len(ndev->mtu)); diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c index 49bb9a8f00e6..3d28654e5df7 100644 --- a/drivers/net/ethernet/atheros/alx/main.c +++ b/drivers/net/ethernet/atheros/alx/main.c @@ -1176,7 +1176,7 @@ static int alx_change_mtu(struct net_device *netdev, int mtu) struct alx_priv *alx = netdev_priv(netdev); int max_frame = ALX_MAX_FRAME_LEN(mtu); - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); alx->hw.mtu = mtu; alx->rxbuf_size = max(max_frame, ALX_DEF_RXBUF_SIZE); netdev_update_features(netdev); diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c index 46cdc32b4e31..c571614b1d50 100644 --- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c +++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c @@ -561,7 +561,7 @@ static int atl1c_change_mtu(struct net_device *netdev, int new_mtu) if (netif_running(netdev)) { while (test_and_set_bit(__AT_RESETTING, &adapter->flags)) msleep(1); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); adapter->hw.max_frame_size = new_mtu; atl1c_set_rxbufsize(adapter, netdev); atl1c_down(adapter); diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c index 5f2a6fcba967..9b778b34b67e 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c +++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c @@ -428,7 +428,7 @@ static int atl1e_change_mtu(struct net_device *netdev, int new_mtu) if (netif_running(netdev)) { while (test_and_set_bit(__AT_RESETTING, &adapter->flags)) msleep(1); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); adapter->hw.max_frame_size = new_mtu; adapter->hw.rx_jumbo_th = (max_frame + 7) >> 3; atl1e_down(adapter); diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c index a9014d7932db..3afd3627ce48 100644 --- a/drivers/net/ethernet/atheros/atlx/atl1.c +++ b/drivers/net/ethernet/atheros/atlx/atl1.c @@ -2687,7 +2687,7 @@ static int atl1_change_mtu(struct net_device *netdev, int new_mtu) adapter->rx_buffer_len = (max_frame + 7) & ~7; adapter->hw.rx_jumbo_th = adapter->rx_buffer_len / 8; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) { atl1_down(adapter); atl1_up(adapter); diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c index bcfc9488125b..fa9a4919f25d 100644 --- a/drivers/net/ethernet/atheros/atlx/atl2.c +++ b/drivers/net/ethernet/atheros/atlx/atl2.c @@ -905,7 +905,7 @@ static int atl2_change_mtu(struct net_device *netdev, int new_mtu) struct atl2_hw *hw = &adapter->hw; /* set MTU */ - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); hw->max_frame_size = new_mtu; ATL2_WRITE_REG(hw, REG_MTU, new_mtu + ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN); diff --git a/drivers/net/ethernet/broadcom/b44.c b/drivers/net/ethernet/broadcom/b44.c index 1be6d14030bc..e5809ad5eb82 100644 --- a/drivers/net/ethernet/broadcom/b44.c +++ b/drivers/net/ethernet/broadcom/b44.c @@ -1042,13 +1042,13 @@ static int b44_change_mtu(struct net_device *dev, int new_mtu) /* We'll just catch it later when the * device is up'd. */ - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } spin_lock_irq(&bp->lock); b44_halt(bp); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); b44_init_rings(bp); b44_init_hw(bp, B44_FULL_RESET); spin_unlock_irq(&bp->lock); diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c index 3196c4dea076..3c0e3b9828be 100644 --- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c +++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c @@ -1652,7 +1652,7 @@ static int bcm_enet_change_mtu(struct net_device *dev, int new_mtu) priv->rx_frag_size = SKB_DATA_ALIGN(priv->rx_buf_offset + priv->rx_buf_size) + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c index b65b8592ad75..6ec773e61182 100644 --- a/drivers/net/ethernet/broadcom/bnx2.c +++ b/drivers/net/ethernet/broadcom/bnx2.c @@ -7912,7 +7912,7 @@ bnx2_change_mtu(struct net_device *dev, int new_mtu) { struct bnx2 *bp = netdev_priv(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return bnx2_change_ring_size(bp, bp->rx_ring_size, bp->tx_ring_size, false); } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index c9b6acd8c892..a8e07e51418f 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -4902,7 +4902,7 @@ int bnx2x_change_mtu(struct net_device *dev, int new_mtu) * because the actual alloc size is * only updated as part of load */ - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (!bnx2x_mtu_allows_gro(new_mtu)) dev->features &= ~NETIF_F_GRO_HW; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 0d1ed6e5dfbd..c437ca1c0fd3 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -14280,7 +14280,7 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu) if (netif_running(dev)) bnxt_close_nic(bp, true, false); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); bnxt_set_ring_params(bp); if (netif_running(dev)) diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index e6ff3c9bd7e5..1589a49b876c 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -14295,7 +14295,7 @@ static void tg3_set_rx_mode(struct net_device *dev) static inline void tg3_set_mtu(struct net_device *dev, struct tg3 *tp, int new_mtu) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (new_mtu > ETH_DATA_LEN) { if (tg3_flag(tp, 5780_CLASS)) { diff --git a/drivers/net/ethernet/brocade/bna/bnad.c b/drivers/net/ethernet/brocade/bna/bnad.c index c32174484a96..fe121d36112d 100644 --- a/drivers/net/ethernet/brocade/bna/bnad.c +++ b/drivers/net/ethernet/brocade/bna/bnad.c @@ -3276,7 +3276,7 @@ bnad_change_mtu(struct net_device *netdev, int new_mtu) mutex_lock(&bnad->conf_mutex); mtu = netdev->mtu; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); frame = BNAD_FRAME_SIZE(mtu); new_frame = BNAD_FRAME_SIZE(new_mtu); diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 898debfd4db3..241ce9a2fa99 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -3022,7 +3022,7 @@ static int macb_change_mtu(struct net_device *dev, int new_mtu) if (netif_running(dev)) return -EBUSY; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/calxeda/xgmac.c b/drivers/net/ethernet/calxeda/xgmac.c index 5e97f1e4e38e..a71b320fd030 100644 --- a/drivers/net/ethernet/calxeda/xgmac.c +++ b/drivers/net/ethernet/calxeda/xgmac.c @@ -1358,7 +1358,7 @@ static int xgmac_change_mtu(struct net_device *dev, int new_mtu) /* Bring interface down, change mtu and bring interface back up */ xgmac_stop(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return xgmac_open(dev); } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_core.c b/drivers/net/ethernet/cavium/liquidio/lio_core.c index f38d31bfab1b..674c54831875 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_core.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_core.c @@ -1262,7 +1262,7 @@ int liquidio_change_mtu(struct net_device *netdev, int new_mtu) return -EINVAL; } - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); lio->mtu = new_mtu; WRITE_ONCE(sc->caller_is_done, true); diff --git a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c index aa6c0dfb6f1c..96c6ea12279f 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_vf_rep.c @@ -218,7 +218,7 @@ lio_vf_rep_change_mtu(struct net_device *ndev, int new_mtu) return -EIO; } - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index 007d4b06819e..744f2434f7fa 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -649,7 +649,7 @@ static int octeon_mgmt_change_mtu(struct net_device *netdev, int new_mtu) struct octeon_mgmt *p = netdev_priv(netdev); int max_packet = new_mtu + ETH_HLEN + ETH_FCS_LEN; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); /* HW lifts the limit if the frame is VLAN tagged * (+4 bytes per each tag, up to two tags) diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index eff350e0bc2a..aebb9fef3f6e 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -1589,7 +1589,7 @@ static int nicvf_change_mtu(struct net_device *netdev, int new_mtu) return -EINVAL; } - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (!netif_running(netdev)) return 0; diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c index d2286adf09fe..7d7d3e0098df 100644 --- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c +++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c @@ -844,7 +844,7 @@ static int t1_change_mtu(struct net_device *dev, int new_mtu) return -EOPNOTSUPP; if ((ret = mac->ops->set_mtu(mac, new_mtu))) return ret; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c index 2236f1d35f2b..f92a3550e480 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c +++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c @@ -2559,7 +2559,7 @@ static int cxgb_change_mtu(struct net_device *dev, int new_mtu) if ((ret = t3_mac_set_mtu(&pi->mac, new_mtu))) return ret; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); init_port_mtus(adapter); if (adapter->params.rev == 0 && offload_running(adapter)) t3_load_mtus(adapter, adapter->params.mtus, diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 2eb33a727bba..2418645c8823 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -3180,7 +3180,7 @@ static int cxgb_change_mtu(struct net_device *dev, int new_mtu) ret = t4_set_rxmode(pi->adapter, pi->adapter->mbox, pi->viid, pi->viid_mirror, new_mtu, -1, -1, -1, -1, true); if (!ret) - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return ret; } diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 9ba0864592e8..2fbe0f059a0b 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -1169,7 +1169,7 @@ static int cxgb4vf_change_mtu(struct net_device *dev, int new_mtu) ret = t4vf_set_rxmode(pi->adapter, pi->viid, new_mtu, -1, -1, -1, -1, true); if (!ret) - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return ret; } diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index d266a87297a5..f604119efc80 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -2027,7 +2027,7 @@ static int _enic_change_mtu(struct net_device *netdev, int new_mtu) return err; } - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (running) { err = enic_open(netdev); diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c index 705c3eb19cd3..2f98f644b9d7 100644 --- a/drivers/net/ethernet/cortina/gemini.c +++ b/drivers/net/ethernet/cortina/gemini.c @@ -1978,7 +1978,7 @@ static int gmac_change_mtu(struct net_device *netdev, int new_mtu) gmac_disable_tx_rx(netdev); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); gmac_update_config0_reg(netdev, max_len << CONFIG0_MAXLEN_SHIFT, CONFIG0_MAXLEN_MASK); diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c index aaf0eda96292..8af5ecec7d61 100644 --- a/drivers/net/ethernet/dlink/sundance.c +++ b/drivers/net/ethernet/dlink/sundance.c @@ -708,7 +708,7 @@ static int change_mtu(struct net_device *dev, int new_mtu) { if (netif_running(dev)) return -EBUSY; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c index 003bc9a45c65..1047c805054e 100644 --- a/drivers/net/ethernet/faraday/ftmac100.c +++ b/drivers/net/ethernet/faraday/ftmac100.c @@ -1092,7 +1092,7 @@ static int ftmac100_change_mtu(struct net_device *netdev, int mtu) } iowrite32(maccr, priv->base + FTMAC100_OFFSET_MACCR); - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); return 0; } diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index dcbc598b11c6..baa0b3c2ce6f 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -2995,7 +2995,7 @@ static int dpaa_change_mtu(struct net_device *net_dev, int new_mtu) if (priv->xdp_prog && !xdp_validate_mtu(priv, new_mtu)) return -EINVAL; - net_dev->mtu = new_mtu; + WRITE_ONCE(net_dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index 40e881829595..6866807973da 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -2698,7 +2698,7 @@ static int dpaa2_eth_change_mtu(struct net_device *dev, int new_mtu) return err; out: - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c index f3543a2df68d..a71f848adc05 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-switch.c @@ -590,7 +590,7 @@ static int dpaa2_switch_port_change_mtu(struct net_device *netdev, int mtu) return err; } - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); return 0; } diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index a811238c018d..2baef59f741d 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -2026,7 +2026,7 @@ static int gfar_change_mtu(struct net_device *dev, int new_mtu) if (dev->flags & IFF_UP) stop_gfar(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (dev->flags & IFF_UP) startup_gfar(dev); diff --git a/drivers/net/ethernet/fungible/funeth/funeth_main.c b/drivers/net/ethernet/fungible/funeth/funeth_main.c index df86770731ad..ac86179a0a81 100644 --- a/drivers/net/ethernet/fungible/funeth/funeth_main.c +++ b/drivers/net/ethernet/fungible/funeth/funeth_main.c @@ -927,7 +927,7 @@ static int fun_change_mtu(struct net_device *netdev, int new_mtu) rc = fun_port_write_cmd(fp, FUN_ADMIN_PORT_KEY_MTU, new_mtu); if (!rc) - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); return rc; } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 8a713eed4465..fd32e15cadcb 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1777,7 +1777,7 @@ static int hns_nic_change_mtu(struct net_device *ndev, int new_mtu) } /* finally, set new mtu to netdevice */ - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); out: if (if_running) { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 19668a8d22f7..dfdc0e032c07 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -2761,7 +2761,7 @@ static int hns3_nic_change_mtu(struct net_device *netdev, int new_mtu) netdev_err(netdev, "failed to change MTU in hardware %d\n", ret); else - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); return ret; } diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index 499c657d37a9..890f213da8d1 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -581,7 +581,7 @@ static int hinic_change_mtu(struct net_device *netdev, int new_mtu) if (err) netif_err(nic_dev, drv, netdev, "Failed to set port mtu\n"); else - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); return err; } diff --git a/drivers/net/ethernet/ibm/emac/core.c b/drivers/net/ethernet/ibm/emac/core.c index e6e47b1842ea..a19d098f2e2b 100644 --- a/drivers/net/ethernet/ibm/emac/core.c +++ b/drivers/net/ethernet/ibm/emac/core.c @@ -1098,7 +1098,7 @@ static int emac_resize_rx_ring(struct emac_instance *dev, int new_mtu) /* This is to prevent starting RX channel in emac_rx_enable() */ set_bit(MAL_COMMAC_RX_STOPPED, &dev->commac.flags); - dev->ndev->mtu = new_mtu; + WRITE_ONCE(dev->ndev->mtu, new_mtu); emac_full_tx_reset(dev); } @@ -1130,7 +1130,7 @@ static int emac_change_mtu(struct net_device *ndev, int new_mtu) } if (!ret) { - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); dev->rx_skb_size = emac_rx_skb_size(new_mtu); dev->rx_sync_size = emac_rx_sync_size(new_mtu); } diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index b5aef0b29efe..4c9d9badd698 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -1537,7 +1537,7 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu) adapter->rx_buff_pool[i].active = 1; if (new_mtu_oh <= adapter->rx_buff_pool[i].buff_size) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); vio_cmo_set_dev_desired(viodev, ibmveth_get_desired_dma (viodev)); diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c index 5b43f9b194fc..60fff9a6c53e 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_main.c +++ b/drivers/net/ethernet/intel/e1000/e1000_main.c @@ -3569,7 +3569,7 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu) netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) e1000_up(adapter); diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index bef65ee4c549..220d62fca55d 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -6038,7 +6038,7 @@ static int e1000_change_mtu(struct net_device *netdev, int new_mtu) adapter->max_frame_size = max_frame; netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); pm_runtime_get_sync(netdev->dev.parent); diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 2cc7bec0557b..143e37ae88ef 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -2961,7 +2961,7 @@ static int i40e_change_mtu(struct net_device *netdev, int new_mtu) netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) i40e_vsi_reinit_locked(vsi); set_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state); diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c index d4f4fd6a1001..d4699c4c535d 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_main.c +++ b/drivers/net/ethernet/intel/iavf/iavf_main.c @@ -4296,7 +4296,7 @@ static int iavf_change_mtu(struct net_device *netdev, int new_mtu) netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) { iavf_schedule_reset(adapter, IAVF_FLAG_RESET_NEEDED); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 06549dae4cca..125c53c51898 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -7770,7 +7770,7 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) return -EBUSY; } - netdev->mtu = (unsigned int)new_mtu; + WRITE_ONCE(netdev->mtu, (unsigned int)new_mtu); err = ice_down_up(vsi); if (err) return err; diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c index 5d3532c27d57..52ceda6306a3 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_lib.c +++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c @@ -2234,7 +2234,7 @@ static int idpf_change_mtu(struct net_device *netdev, int new_mtu) idpf_vport_ctrl_lock(netdev); vport = idpf_netdev_to_vport(netdev); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); err = idpf_initiate_soft_reset(vport, IDPF_SR_MTU_CHANGE); diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index 74a998fcaa6f..7d9389040e40 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -6641,7 +6641,7 @@ static int igb_change_mtu(struct net_device *netdev, int new_mtu) netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) igb_up(adapter); diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c index 40ccd24ffc53..7661edd7d0f2 100644 --- a/drivers/net/ethernet/intel/igbvf/netdev.c +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -2434,7 +2434,7 @@ static int igbvf_change_mtu(struct net_device *netdev, int new_mtu) netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) igbvf_up(adapter); diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 34c257a51ed1..c30f0f572600 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -5275,7 +5275,7 @@ static int igc_change_mtu(struct net_device *netdev, int new_mtu) igc_down(adapter); netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) igc_up(adapter); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 43e7e75ae18c..094653e81b97 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -6847,7 +6847,7 @@ static int ixgbe_change_mtu(struct net_device *netdev, int new_mtu) netdev->mtu, new_mtu); /* must set new MTU before calling down or up */ - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) ixgbe_reinit_locked(adapter); diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 3161a13079fe..b938dc06045d 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -4292,7 +4292,7 @@ static int ixgbevf_change_mtu(struct net_device *netdev, int new_mtu) netdev->mtu, new_mtu); /* must set new MTU before calling down or up */ - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) ixgbevf_reinit_locked(adapter); diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c index 1732ec3c3dbd..b06e24562973 100644 --- a/drivers/net/ethernet/jme.c +++ b/drivers/net/ethernet/jme.c @@ -2301,7 +2301,7 @@ jme_change_mtu(struct net_device *netdev, int new_mtu) { struct jme_adapter *jme = netdev_priv(netdev); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); netdev_update_features(netdev); jme_restart_rx_engine(jme); diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 1d5b7bb6380f..5352fee62d2b 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -519,7 +519,7 @@ ltq_etop_change_mtu(struct net_device *dev, int new_mtu) struct ltq_etop_priv *priv = netdev_priv(dev); unsigned long flags; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); spin_lock_irqsave(&priv->lock, flags); ltq_etop_w32((ETOP_PLEN_UNDER << 16) | new_mtu, LTQ_ETOP_IGPLEN); diff --git a/drivers/net/ethernet/lantiq_xrx200.c b/drivers/net/ethernet/lantiq_xrx200.c index 8bd4def3622e..07904a528f21 100644 --- a/drivers/net/ethernet/lantiq_xrx200.c +++ b/drivers/net/ethernet/lantiq_xrx200.c @@ -419,7 +419,7 @@ xrx200_change_mtu(struct net_device *net_dev, int new_mtu) int curr_desc; int ret = 0; - net_dev->mtu = new_mtu; + WRITE_ONCE(net_dev->mtu, new_mtu); priv->rx_buf_size = xrx200_buffer_size(new_mtu); priv->rx_skb_size = xrx200_skb_size(priv->rx_buf_size); @@ -440,7 +440,7 @@ xrx200_change_mtu(struct net_device *net_dev, int new_mtu) buff = ch_rx->rx_buff[ch_rx->dma.desc]; ret = xrx200_alloc_buf(ch_rx, netdev_alloc_frag); if (ret) { - net_dev->mtu = old_mtu; + WRITE_ONCE(net_dev->mtu, old_mtu); priv->rx_buf_size = xrx200_buffer_size(old_mtu); priv->rx_skb_size = xrx200_skb_size(priv->rx_buf_size); break; diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index f0bdc06d253d..f35ae2c88091 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -2562,7 +2562,7 @@ static int mv643xx_eth_change_mtu(struct net_device *dev, int new_mtu) { struct mv643xx_eth_private *mp = netdev_priv(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); mv643xx_eth_recalc_skb_size(mp); tx_set_rate(mp, 1000000000, 16777216); diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 26bf5d47ba02..41894834fb53 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -3861,7 +3861,7 @@ static int mvneta_change_mtu(struct net_device *dev, int mtu) return -EINVAL; } - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); if (!netif_running(dev)) { if (pp->bm_priv) diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c index 19253a0fb4fe..e91486c48de3 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c @@ -1375,7 +1375,7 @@ static int mvpp2_bm_update_mtu(struct net_device *dev, int mtu) } out_set: - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); dev->wanted_features = dev->features; netdev_update_features(dev); diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c index 7c9faa714a10..549436efc204 100644 --- a/drivers/net/ethernet/marvell/octeon_ep/octep_main.c +++ b/drivers/net/ethernet/marvell/octeon_ep/octep_main.c @@ -1096,7 +1096,7 @@ static int octep_change_mtu(struct net_device *netdev, int new_mtu) true); if (!err) { oct->link_info.mtu = new_mtu; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); } return err; diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c index dd49d0b8b494..7e6771c9cdbb 100644 --- a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c @@ -881,7 +881,7 @@ static int octep_vf_change_mtu(struct net_device *netdev, int new_mtu) err = octep_vf_mbox_set_mtu(oct, new_mtu); if (!err) { oct->link_info.mtu = new_mtu; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); } return err; } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index 14bccff0ee5c..f5bce3e326cc 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -67,7 +67,7 @@ static int otx2_change_mtu(struct net_device *netdev, int new_mtu) netdev_info(netdev, "Changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (if_up) err = otx2_open(netdev); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c index cf0aa16d7540..99fcc5661674 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c @@ -456,7 +456,7 @@ static int otx2vf_change_mtu(struct net_device *netdev, int new_mtu) netdev_info(netdev, "Changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (if_up) err = otx2vf_open(netdev); diff --git a/drivers/net/ethernet/marvell/prestera/prestera_main.c b/drivers/net/ethernet/marvell/prestera/prestera_main.c index ba6d53ac7f55..63ae01954dfc 100644 --- a/drivers/net/ethernet/marvell/prestera/prestera_main.c +++ b/drivers/net/ethernet/marvell/prestera/prestera_main.c @@ -489,7 +489,7 @@ static int prestera_port_change_mtu(struct net_device *dev, int mtu) if (err) return err; - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); return 0; } diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c index dd6ca2e4fd51..1a59c952aa01 100644 --- a/drivers/net/ethernet/marvell/pxa168_eth.c +++ b/drivers/net/ethernet/marvell/pxa168_eth.c @@ -1188,7 +1188,7 @@ static int pxa168_eth_change_mtu(struct net_device *dev, int mtu) { struct pxa168_eth_private *pep = netdev_priv(dev); - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); set_port_config_ext(pep); if (!netif_running(dev)) diff --git a/drivers/net/ethernet/marvell/skge.c b/drivers/net/ethernet/marvell/skge.c index 1b43704baceb..fcfb34561882 100644 --- a/drivers/net/ethernet/marvell/skge.c +++ b/drivers/net/ethernet/marvell/skge.c @@ -2905,13 +2905,13 @@ static int skge_change_mtu(struct net_device *dev, int new_mtu) int err; if (!netif_running(dev)) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } skge_down(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); err = skge_up(dev); if (err) diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c index f3f7f4cc27b3..a7a16eac1891 100644 --- a/drivers/net/ethernet/marvell/sky2.c +++ b/drivers/net/ethernet/marvell/sky2.c @@ -2384,7 +2384,7 @@ static int sky2_change_mtu(struct net_device *dev, int new_mtu) u32 imask; if (!netif_running(dev)) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); netdev_update_features(dev); return 0; } @@ -2407,7 +2407,7 @@ static int sky2_change_mtu(struct net_device *dev, int new_mtu) sky2_rx_stop(sky2); sky2_rx_clean(sky2); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); netdev_update_features(dev); mode = DATA_BLIND_VAL(DATA_BLIND_DEF) | GM_SMOD_VLAN_ENA; diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index d7a96dc11c07..179c0230655a 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -4055,7 +4055,7 @@ static int mtk_change_mtu(struct net_device *dev, int new_mtu) } mtk_set_mcr_max_rx(mac, length); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 5d3fde63b273..4c089cfa027a 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1649,7 +1649,7 @@ int mlx4_en_start_port(struct net_device *dev) sizeof(struct ethtool_flow_id) * MAX_NUM_OF_FS_RULES); /* Calculate Rx buf size */ - dev->mtu = min(dev->mtu, priv->max_mtu); + WRITE_ONCE(dev->mtu, min(dev->mtu, priv->max_mtu)); mlx4_en_calc_rx_buf(dev); en_dbg(DRV, priv, "Rx buf size:%d\n", priv->rx_skb_size); @@ -2394,7 +2394,7 @@ static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu) !mlx4_en_check_xdp_mtu(dev, new_mtu)) return -EOPNOTSUPP; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (netif_running(dev)) { mutex_lock(&mdev->state_lock); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 3bd0695845c7..ffe8919494d5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -4525,7 +4525,7 @@ int mlx5e_change_mtu(struct net_device *netdev, int new_mtu, err = mlx5e_safe_switch_params(priv, &new_params, preactivate, NULL, reset); out: - netdev->mtu = params->sw_mtu; + WRITE_ONCE(netdev->mtu, params->sw_mtu); mutex_unlock(&priv->state_lock); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index d77be1b4dd9c..8e0404c0d1ca 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -531,7 +531,7 @@ static int mlx5i_change_mtu(struct net_device *netdev, int new_mtu) if (err) goto out; - netdev->mtu = new_params.sw_mtu; + WRITE_ONCE(netdev->mtu, new_params.sw_mtu); out: mutex_unlock(&priv->state_lock); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c index f87471306f6b..028a76944d82 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c @@ -280,7 +280,7 @@ static int mlx5i_pkey_change_mtu(struct net_device *netdev, int new_mtu) struct mlx5e_priv *priv = mlx5i_epriv(netdev); mutex_lock(&priv->state_lock); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); mutex_unlock(&priv->state_lock); return 0; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index bb642e9bb6cf..030ed71f945d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -825,7 +825,7 @@ static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, mtu); if (err) goto err_port_mtu_set; - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); return 0; err_port_mtu_set: diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c index c5aeeb964c17..dc1d9f774565 100644 --- a/drivers/net/ethernet/micrel/ksz884x.c +++ b/drivers/net/ethernet/micrel/ksz884x.c @@ -5427,7 +5427,7 @@ static int netdev_change_mtu(struct net_device *dev, int new_mtu) } hw_mtu = (hw_mtu + 3) & ~3; hw_priv->mtu = hw_mtu; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index cee47729d022..6be8a43c908a 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -3184,7 +3184,7 @@ static int lan743x_netdev_change_mtu(struct net_device *netdev, int new_mtu) ret = lan743x_mac_set_mtu(adapter, new_mtu); if (!ret) - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); return ret; } diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c index b7e75da65834..b12d3b8a64fd 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_main.c +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_main.c @@ -402,7 +402,7 @@ static int lan966x_port_change_mtu(struct net_device *dev, int new_mtu) lan_wr(DEV_MAC_MAXLEN_CFG_MAX_LEN_SET(LAN966X_HW_MTU(new_mtu)), lan966x, DEV_MAC_MAXLEN_CFG(port->chip_port)); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (!lan966x->fdma) return 0; diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index f2fae659bf3b..d087cf954f75 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -690,12 +690,12 @@ static int mana_change_mtu(struct net_device *ndev, int new_mtu) goto out; } - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); err = mana_attach(ndev); if (err) { netdev_err(ndev, "mana_attach failed: %d\n", err); - ndev->mtu = old_mtu; + WRITE_ONCE(ndev->mtu, old_mtu); } out: diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c index 7b7e1c5b00f4..b7d9657a7af3 100644 --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c @@ -3036,11 +3036,11 @@ static int myri10ge_change_mtu(struct net_device *dev, int new_mtu) /* if we change the mtu on an active device, we must * reset the device so the firmware sees the change */ myri10ge_close(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); myri10ge_open(dev); - } else - dev->mtu = new_mtu; - + } else { + WRITE_ONCE(dev->mtu, new_mtu); + } return 0; } diff --git a/drivers/net/ethernet/natsemi/natsemi.c b/drivers/net/ethernet/natsemi/natsemi.c index 650a5a166070..ad0c14849115 100644 --- a/drivers/net/ethernet/natsemi/natsemi.c +++ b/drivers/net/ethernet/natsemi/natsemi.c @@ -2526,7 +2526,7 @@ static void __set_rx_mode(struct net_device *dev) static int natsemi_change_mtu(struct net_device *dev, int new_mtu) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); /* synchronized against open : rtnl_lock() held by caller */ if (netif_running(dev)) { diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c index 55408f16fbbc..f235e76e4ce9 100644 --- a/drivers/net/ethernet/neterion/s2io.c +++ b/drivers/net/ethernet/neterion/s2io.c @@ -6637,7 +6637,7 @@ static int s2io_change_mtu(struct net_device *dev, int new_mtu) struct s2io_nic *sp = netdev_priv(dev); int ret = 0; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (netif_running(dev)) { s2io_stop_all_tx_queue(sp); s2io_card_down(sp); diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 997cc4fcffdb..182ba0a8b095 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1526,7 +1526,7 @@ static void nfp_net_dp_swap(struct nfp_net *nn, struct nfp_net_dp *dp) *dp = nn->dp; nn->dp = new_dp; - nn->dp.netdev->mtu = new_dp.mtu; + WRITE_ONCE(nn->dp.netdev->mtu, new_dp.mtu); if (!netif_is_rxfh_configured(nn->dp.netdev)) nfp_net_rss_init_itbl(nn); diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c index 3af1229a3f08..eee0bfc41074 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c @@ -177,7 +177,7 @@ static int nfp_repr_change_mtu(struct net_device *netdev, int new_mtu) if (err) return err; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c index fa1f78b03cb2..2aa4ad9cf96e 100644 --- a/drivers/net/ethernet/ni/nixge.c +++ b/drivers/net/ethernet/ni/nixge.c @@ -946,7 +946,7 @@ static int nixge_change_mtu(struct net_device *ndev, int new_mtu) NIXGE_MAX_JUMBO_FRAME_SIZE) return -EINVAL; - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c index 31f896c4aa26..720f577929db 100644 --- a/drivers/net/ethernet/nvidia/forcedeth.c +++ b/drivers/net/ethernet/nvidia/forcedeth.c @@ -3098,7 +3098,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu) int old_mtu; old_mtu = dev->mtu; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); /* return early if the buffer sizes will not change */ if (old_mtu <= ETH_DATA_LEN && new_mtu <= ETH_DATA_LEN) diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c index 28b7cec485ef..4ac29cd59f2b 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c @@ -2184,7 +2184,7 @@ static int pch_gbe_change_mtu(struct net_device *netdev, int new_mtu) } } else { pch_gbe_reset(adapter); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); adapter->hw.mac.max_frame_size = max_frame; } diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.c b/drivers/net/ethernet/pasemi/pasemi_mac.c index ed7dd0a04235..62ba269da902 100644 --- a/drivers/net/ethernet/pasemi/pasemi_mac.c +++ b/drivers/net/ethernet/pasemi/pasemi_mac.c @@ -1639,7 +1639,7 @@ static int pasemi_mac_change_mtu(struct net_device *dev, int new_mtu) reg |= PAS_MAC_CFG_MACCFG_MAXF(new_mtu + ETH_HLEN + 4); write_mac_reg(mac, PAS_MAC_CFG_MACCFG, reg); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); /* MTU + ETH_HLEN + VLAN_HLEN + 2 64B cachelines */ mac->bufsz = new_mtu + ETH_HLEN + ETH_FCS_LEN + LOCAL_SKB_ALIGN + 128; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 7f0c6cdc375e..24870da3f484 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -1761,13 +1761,13 @@ static int ionic_change_mtu(struct net_device *netdev, int new_mtu) /* if we're not running, nothing more to do */ if (!netif_running(netdev)) { - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); return 0; } mutex_lock(&lif->queue_lock); ionic_stop_queues_reconfig(lif); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); err = ionic_start_queues_reconfig(lif); mutex_unlock(&lif->queue_lock); diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c index 6e12cd21ac90..89c8b2349694 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c @@ -960,7 +960,7 @@ int netxen_nic_change_mtu(struct net_device *netdev, int mtu) rc = adapter->set_mtu(adapter, mtu); if (!rc) - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); return rc; } diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index ae3ebf0cf999..f497f6ca1018 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -1026,7 +1026,7 @@ static int qede_get_regs_len(struct net_device *ndev) static void qede_update_mtu(struct qede_dev *edev, struct qede_reload_args *args) { - edev->ndev->mtu = args->u.mtu; + WRITE_ONCE(edev->ndev->mtu, args->u.mtu); } /* Netdevice NDOs */ diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c index 4b8bc46f55c2..ae4ee0326ee1 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_hw.c @@ -1015,7 +1015,7 @@ int qlcnic_change_mtu(struct net_device *netdev, int mtu) rc = qlcnic_fw_cmd_set_mtu(adapter, mtu); if (!rc) - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); return rc; } diff --git a/drivers/net/ethernet/qualcomm/emac/emac.c b/drivers/net/ethernet/qualcomm/emac/emac.c index 4c06f55878de..99d4647bf245 100644 --- a/drivers/net/ethernet/qualcomm/emac/emac.c +++ b/drivers/net/ethernet/qualcomm/emac/emac.c @@ -216,7 +216,7 @@ static int emac_change_mtu(struct net_device *netdev, int new_mtu) netif_dbg(adpt, hw, adpt->netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (netif_running(netdev)) return emac_reinit_locked(adpt); diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c index 9d2a9562c96f..f1e40aade127 100644 --- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c +++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c @@ -90,7 +90,7 @@ static int rmnet_vnd_change_mtu(struct net_device *rmnet_dev, int new_mtu) new_mtu > (priv->real_dev->mtu - headroom)) return -EINVAL; - rmnet_dev->mtu = new_mtu; + WRITE_ONCE(rmnet_dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index f5786d78ed23..5652da8a178c 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -1277,14 +1277,14 @@ static int cp_change_mtu(struct net_device *dev, int new_mtu) /* if network interface not up, no need for complexity */ if (!netif_running(dev)) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); cp_set_rxbufsize(cp); /* set new rx buf size */ return 0; } /* network IS up, close it, reset MTU, and come up again. */ cp_close(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); cp_set_rxbufsize(cp); return cp_open(dev); } diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index fab21d2bc4ff..5abbea91bc07 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -3924,7 +3924,7 @@ static int rtl8169_change_mtu(struct net_device *dev, int new_mtu) { struct rtl8169_private *tp = netdev_priv(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); netdev_update_features(dev); rtl_jumbo_config(tp); rtl_set_eee_txidle_timer(tp); diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 384ddad65aaf..4d100283c30f 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2423,7 +2423,7 @@ static int ravb_change_mtu(struct net_device *ndev, int new_mtu) { struct ravb_private *priv = netdev_priv(ndev); - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); if (netif_running(ndev)) { synchronize_irq(priv->emac_irq); diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 0786eb0da391..7a25903e35c3 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -2624,7 +2624,7 @@ static int sh_eth_change_mtu(struct net_device *ndev, int new_mtu) if (netif_running(ndev)) return -EBUSY; - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); netdev_update_features(ndev); return 0; diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c index 755db89db909..e097ce3e69ea 100644 --- a/drivers/net/ethernet/rocker/rocker_main.c +++ b/drivers/net/ethernet/rocker/rocker_main.c @@ -1967,7 +1967,7 @@ static int rocker_port_change_mtu(struct net_device *dev, int new_mtu) rocker_port_stop(dev); netdev_info(dev, "MTU change from %d to %d\n", dev->mtu, new_mtu); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); err = rocker_cmd_set_port_settings_mtu(rocker_port, new_mtu); if (err) diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c index ecbe3994f2b1..12c8396b6942 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c @@ -1804,7 +1804,7 @@ static int sxgbe_set_features(struct net_device *dev, */ static int sxgbe_change_mtu(struct net_device *dev, int new_mtu) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (!netif_running(dev)) return 0; diff --git a/drivers/net/ethernet/sfc/efx_common.c b/drivers/net/ethernet/sfc/efx_common.c index 551f890db90a..4ebd5ae23eca 100644 --- a/drivers/net/ethernet/sfc/efx_common.c +++ b/drivers/net/ethernet/sfc/efx_common.c @@ -302,7 +302,7 @@ int efx_change_mtu(struct net_device *net_dev, int new_mtu) efx_stop_all(efx); mutex_lock(&efx->mac_lock); - net_dev->mtu = new_mtu; + WRITE_ONCE(net_dev->mtu, new_mtu); efx_mac_reconfigure(efx, true); mutex_unlock(&efx->mac_lock); diff --git a/drivers/net/ethernet/sfc/falcon/efx.c b/drivers/net/ethernet/sfc/falcon/efx.c index 1cb32aedd89c..8925745f1c17 100644 --- a/drivers/net/ethernet/sfc/falcon/efx.c +++ b/drivers/net/ethernet/sfc/falcon/efx.c @@ -2125,7 +2125,7 @@ static int ef4_change_mtu(struct net_device *net_dev, int new_mtu) ef4_stop_all(efx); mutex_lock(&efx->mac_lock); - net_dev->mtu = new_mtu; + WRITE_ONCE(net_dev->mtu, new_mtu); ef4_mac_reconfigure(efx); mutex_unlock(&efx->mac_lock); diff --git a/drivers/net/ethernet/sfc/siena/efx_common.c b/drivers/net/ethernet/sfc/siena/efx_common.c index 88e5bc347a44..cf195162e270 100644 --- a/drivers/net/ethernet/sfc/siena/efx_common.c +++ b/drivers/net/ethernet/sfc/siena/efx_common.c @@ -306,7 +306,7 @@ int efx_siena_change_mtu(struct net_device *net_dev, int new_mtu) efx_siena_stop_all(efx); mutex_lock(&efx->mac_lock); - net_dev->mtu = new_mtu; + WRITE_ONCE(net_dev->mtu, new_mtu); efx_siena_mac_reconfigure(efx, true); mutex_unlock(&efx->mac_lock); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 59bf83904b62..3d828904db0d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -5909,7 +5909,7 @@ static int stmmac_change_mtu(struct net_device *dev, int new_mtu) stmmac_set_rx_mode(dev); } - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); netdev_update_features(dev); return 0; diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index 8f1f43dbb76d..b8948d5b779a 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -3804,7 +3804,7 @@ static int cas_change_mtu(struct net_device *dev, int new_mtu) { struct cas *cp = netdev_priv(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (!netif_running(dev) || !netif_device_present(dev)) return 0; diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c index f68aa813d4fb..41a27ae58ced 100644 --- a/drivers/net/ethernet/sun/niu.c +++ b/drivers/net/ethernet/sun/niu.c @@ -6751,7 +6751,7 @@ static int niu_change_mtu(struct net_device *dev, int new_mtu) orig_jumbo = (dev->mtu > ETH_DATA_LEN); new_jumbo = (new_mtu > ETH_DATA_LEN); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (!netif_running(dev) || (orig_jumbo == new_jumbo)) diff --git a/drivers/net/ethernet/sun/sungem.c b/drivers/net/ethernet/sun/sungem.c index 9bd1df8308d2..4aa356e2f299 100644 --- a/drivers/net/ethernet/sun/sungem.c +++ b/drivers/net/ethernet/sun/sungem.c @@ -2499,7 +2499,7 @@ static int gem_change_mtu(struct net_device *dev, int new_mtu) { struct gem *gp = netdev_priv(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); /* We'll just catch it later when the device is up'd or resumed */ if (!netif_running(dev) || !netif_device_present(dev)) diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c index 36b948820c1e..d1793b6154c7 100644 --- a/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c +++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c @@ -823,7 +823,7 @@ static int xlgmac_change_mtu(struct net_device *netdev, int mtu) return ret; pdata->rx_buf_size = ret; - netdev->mtu = mtu; + WRITE_ONCE(netdev->mtu, mtu); xlgmac_restart_dev(pdata); diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c index ca409515ead5..ede5f7890fb4 100644 --- a/drivers/net/ethernet/tehuti/tehuti.c +++ b/drivers/net/ethernet/tehuti/tehuti.c @@ -756,7 +756,7 @@ static int bdx_change_mtu(struct net_device *ndev, int new_mtu) { ENTER; - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); if (netif_running(ndev)) { bdx_close(ndev); bdx_open(ndev); diff --git a/drivers/net/ethernet/via/via-velocity.c b/drivers/net/ethernet/via/via-velocity.c index 1c6b2a9bba08..55fff4d0d380 100644 --- a/drivers/net/ethernet/via/via-velocity.c +++ b/drivers/net/ethernet/via/via-velocity.c @@ -2294,7 +2294,7 @@ static int velocity_change_mtu(struct net_device *dev, int new_mtu) int ret = 0; if (!netif_running(dev)) { - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); goto out_0; } @@ -2336,7 +2336,7 @@ static int velocity_change_mtu(struct net_device *dev, int new_mtu) tmp_vptr->rx = rx; tmp_vptr->tx = tx; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); velocity_init_registers(vptr, VELOCITY_INIT_COLD); diff --git a/drivers/net/ethernet/wangxun/libwx/wx_hw.c b/drivers/net/ethernet/wangxun/libwx/wx_hw.c index 945c13d1a982..3662483bfe2e 100644 --- a/drivers/net/ethernet/wangxun/libwx/wx_hw.c +++ b/drivers/net/ethernet/wangxun/libwx/wx_hw.c @@ -1408,7 +1408,7 @@ int wx_change_mtu(struct net_device *netdev, int new_mtu) { struct wx *wx = netdev_priv(netdev); - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); wx_set_rx_buffer_len(wx); return 0; diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index aaf780fd4f5e..c29809cd9201 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -1641,7 +1641,7 @@ static int axienet_change_mtu(struct net_device *ndev, int new_mtu) XAE_TRL_SIZE) > lp->rxmem) return -EINVAL; - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c index e0d26148dfd9..8aff6a73ca0a 100644 --- a/drivers/net/ethernet/xscale/ixp4xx_eth.c +++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c @@ -1233,7 +1233,7 @@ static int ixp4xx_eth_change_mtu(struct net_device *dev, int new_mtu) return ret; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c index 5fbe33a09bb0..6411da359e5a 100644 --- a/drivers/net/fjes/fjes_main.c +++ b/drivers/net/fjes/fjes_main.c @@ -811,7 +811,7 @@ static int fjes_change_mtu(struct net_device *netdev, int new_mtu) netif_tx_stop_all_queues(netdev); } - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); if (running) { for (epidx = 0; epidx < hw->max_epid; epidx++) { diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index f918ca6146c8..51495cb4b9be 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -1059,7 +1059,7 @@ static int geneve_change_mtu(struct net_device *dev, int new_mtu) else if (new_mtu < dev->min_mtu) new_mtu = dev->min_mtu; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 11831a1c9762..44142245343d 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1233,14 +1233,14 @@ static int netvsc_change_mtu(struct net_device *ndev, int mtu) if (ret) goto rollback_vf; - ndev->mtu = mtu; + WRITE_ONCE(ndev->mtu, mtu); ret = netvsc_attach(ndev, device_info); if (!ret) goto out; /* Attempt rollback to original MTU */ - ndev->mtu = orig_mtu; + WRITE_ONCE(ndev->mtu, orig_mtu); if (netvsc_attach(ndev, device_info)) netdev_err(ndev, "restoring mtu failed\n"); diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index ff016c11b4a0..2da70bc3dd86 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3753,7 +3753,7 @@ static int macsec_change_mtu(struct net_device *dev, int new_mtu) if (macsec->real_dev->mtu - extra < new_mtu) return -ERANGE; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 0cec2783a3e7..67b7ef2d463f 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -865,7 +865,7 @@ static int macvlan_change_mtu(struct net_device *dev, int new_mtu) if (vlan->lowerdev->mtu < new_mtu) return -EINVAL; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/net_failover.c b/drivers/net/net_failover.c index d0c916a53d7c..963d8b4af28d 100644 --- a/drivers/net/net_failover.c +++ b/drivers/net/net_failover.c @@ -231,7 +231,7 @@ static int net_failover_change_mtu(struct net_device *dev, int new_mtu) } } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c index d127856f8f36..e3a385a32269 100644 --- a/drivers/net/netdevsim/netdev.c +++ b/drivers/net/netdevsim/netdev.c @@ -74,7 +74,7 @@ static int nsim_change_mtu(struct net_device *dev, int new_mtu) if (ns->xdp.prog && new_mtu > NSIM_XDP_MAX_MTU) return -EBUSY; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/ntb_netdev.c b/drivers/net/ntb_netdev.c index 536bd6564f8b..ffe7f463e16e 100644 --- a/drivers/net/ntb_netdev.c +++ b/drivers/net/ntb_netdev.c @@ -306,7 +306,7 @@ static int ntb_netdev_change_mtu(struct net_device *ndev, int new_mtu) return -EINVAL; if (!netif_running(ndev)) { - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); return 0; } @@ -335,7 +335,7 @@ static int ntb_netdev_change_mtu(struct net_device *ndev, int new_mtu) } } - ndev->mtu = new_mtu; + WRITE_ONCE(ndev->mtu, new_mtu); ntb_transport_link_up(dev->qp); diff --git a/drivers/net/slip/slip.c b/drivers/net/slip/slip.c index 0aba3569ccc0..fb362ee248ff 100644 --- a/drivers/net/slip/slip.c +++ b/drivers/net/slip/slip.c @@ -286,7 +286,7 @@ static int sl_realloc_bufs(struct slip *sl, int mtu) } } sl->mtu = mtu; - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); sl->buffsize = len; err = 0; diff --git a/drivers/net/team/team_core.c b/drivers/net/team/team_core.c index 8c7dbaf7c22e..ab1935a4aa2c 100644 --- a/drivers/net/team/team_core.c +++ b/drivers/net/team/team_core.c @@ -1831,7 +1831,7 @@ static int team_change_mtu(struct net_device *dev, int new_mtu) team->port_mtu_change_allowed = false; mutex_unlock(&team->lock); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; diff --git a/drivers/net/usb/aqc111.c b/drivers/net/usb/aqc111.c index 7b8afa589a53..74571c8e7691 100644 --- a/drivers/net/usb/aqc111.c +++ b/drivers/net/usb/aqc111.c @@ -424,7 +424,7 @@ static int aqc111_change_mtu(struct net_device *net, int new_mtu) u16 reg16 = 0; u8 buf[5]; - net->mtu = new_mtu; + WRITE_ONCE(net->mtu, new_mtu); dev->hard_mtu = net->mtu + net->hard_header_len; aqc111_read16_cmd(dev, AQ_ACCESS_MAC, SFR_MEDIUM_STATUS_MODE, diff --git a/drivers/net/usb/asix_devices.c b/drivers/net/usb/asix_devices.c index f7cff58fe044..57d6e5abc30e 100644 --- a/drivers/net/usb/asix_devices.c +++ b/drivers/net/usb/asix_devices.c @@ -1230,7 +1230,7 @@ static int ax88178_change_mtu(struct net_device *net, int new_mtu) if ((ll_mtu % dev->maxpacket) == 0) return -EDOM; - net->mtu = new_mtu; + WRITE_ONCE(net->mtu, new_mtu); dev->hard_mtu = net->mtu + net->hard_header_len; ax88178_set_mfb(dev); diff --git a/drivers/net/usb/ax88179_178a.c b/drivers/net/usb/ax88179_178a.c index b25535aa15ae..c5196b5313c0 100644 --- a/drivers/net/usb/ax88179_178a.c +++ b/drivers/net/usb/ax88179_178a.c @@ -943,7 +943,7 @@ static int ax88179_change_mtu(struct net_device *net, int new_mtu) struct usbnet *dev = netdev_priv(net); u16 tmp16; - net->mtu = new_mtu; + WRITE_ONCE(net->mtu, new_mtu); dev->hard_mtu = net->mtu + net->hard_header_len; if (net->mtu > 1500) { diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index db05622f1f70..bf76ecccc2e6 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -798,7 +798,7 @@ int cdc_ncm_change_mtu(struct net_device *net, int new_mtu) { struct usbnet *dev = netdev_priv(net); - net->mtu = new_mtu; + WRITE_ONCE(net->mtu, new_mtu); cdc_ncm_set_dgram_size(dev, new_mtu + cdc_ncm_eth_hlen(dev)); return 0; diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index 0030be502daa..5a2c38b63012 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -2528,7 +2528,7 @@ static int lan78xx_change_mtu(struct net_device *netdev, int new_mtu) ret = lan78xx_set_rx_max_frame_length(dev, max_frame_len); if (!ret) - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); usb_autopm_put_interface(dev->intf); diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 51e9f5b2dccf..19df1cd9f072 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -9365,7 +9365,7 @@ static int rtl8152_change_mtu(struct net_device *dev, int new_mtu) case RTL_VER_01: case RTL_VER_02: case RTL_VER_07: - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; default: break; @@ -9377,7 +9377,7 @@ static int rtl8152_change_mtu(struct net_device *dev, int new_mtu) mutex_lock(&tp->control); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); if (netif_running(dev)) { if (tp->rtl_ops.change_mtu) diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c index f3f7f686fe9c..9fd516e8bb10 100644 --- a/drivers/net/usb/usbnet.c +++ b/drivers/net/usb/usbnet.c @@ -398,7 +398,7 @@ int usbnet_change_mtu (struct net_device *net, int new_mtu) // no second zero-length packet read wanted after mtu-sized packets if ((ll_mtu % dev->maxpacket) == 0) return -EDOM; - net->mtu = new_mtu; + WRITE_ONCE(net->mtu, new_mtu); dev->hard_mtu = net->mtu + net->hard_header_len; if (dev->rx_urb_size == old_hard_mtu) { diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index 0578864792b6..89ca6e75fcc6 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -3457,7 +3457,7 @@ vmxnet3_change_mtu(struct net_device *netdev, int new_mtu) struct vmxnet3_adapter *adapter = netdev_priv(netdev); int err = 0; - netdev->mtu = new_mtu; + WRITE_ONCE(netdev->mtu, new_mtu); /* * Reset_work may be in the middle of resetting the device, wait for its diff --git a/drivers/net/vsockmon.c b/drivers/net/vsockmon.c index a1ba5169ed5d..4c260074c091 100644 --- a/drivers/net/vsockmon.c +++ b/drivers/net/vsockmon.c @@ -58,7 +58,7 @@ static int vsockmon_change_mtu(struct net_device *dev, int new_mtu) if (!vsockmon_is_valid_mtu(new_mtu)) return -EINVAL; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index 7e3a7d1f2018..f78dd0438843 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -3181,7 +3181,7 @@ static int vxlan_change_mtu(struct net_device *dev, int new_mtu) return -EINVAL; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index 65db5f14465f..325fcb3d1075 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -358,7 +358,7 @@ static int xenvif_change_mtu(struct net_device *dev, int mtu) if (mtu > max) return -EINVAL; - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); return 0; } diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 8d2aee88526c..4265c1cd0ff7 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -1376,7 +1376,7 @@ static int xennet_change_mtu(struct net_device *dev, int mtu) if (mtu > max) return -EINVAL; - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); return 0; } diff --git a/drivers/s390/net/ctcm_main.c b/drivers/s390/net/ctcm_main.c index 878fe3ce53ad..b93c2eb45916 100644 --- a/drivers/s390/net/ctcm_main.c +++ b/drivers/s390/net/ctcm_main.c @@ -996,7 +996,7 @@ static int ctcm_change_mtu(struct net_device *dev, int new_mtu) return -EINVAL; dev->hard_header_len = LL_HEADER_LENGTH + 2; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 39876eff51d2..3efba4f857ac 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -149,7 +149,7 @@ static int vlan_dev_change_mtu(struct net_device *dev, int new_mtu) if (max_mtu < new_mtu) return -ERANGE; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 89c51b3cf430..30ecbc2ef1fd 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -159,7 +159,7 @@ static int batadv_interface_change_mtu(struct net_device *dev, int new_mtu) if (new_mtu < ETH_MIN_MTU || new_mtu > batadv_hardif_min_mtu(dev)) return -EINVAL; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); bat_priv->mtu_set_by_user = new_mtu; return 0; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index ab4d33e02014..71b50001ca58 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -197,7 +197,7 @@ static int br_change_mtu(struct net_device *dev, int new_mtu) { struct net_bridge *br = netdev_priv(dev); - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); /* this flag will be cleared if the MTU was automatically adjusted */ br_opt_toggle(br, BROPT_MTU_SET_BY_USER, true); diff --git a/net/dsa/user.c b/net/dsa/user.c index c94b868855aa..1f7b8f21db9e 100644 --- a/net/dsa/user.c +++ b/net/dsa/user.c @@ -2120,7 +2120,7 @@ int dsa_user_change_mtu(struct net_device *dev, int new_mtu) if (err) goto out_port_failed; - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); dsa_bridge_mtu_normalization(dp); diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index 86127300b102..26d5282a6496 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -123,7 +123,7 @@ static int hsr_dev_change_mtu(struct net_device *dev, int new_mtu) return -EINVAL; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } diff --git a/net/hsr/hsr_main.c b/net/hsr/hsr_main.c index 9756e657bab9..d7ae32473c41 100644 --- a/net/hsr/hsr_main.c +++ b/net/hsr/hsr_main.c @@ -96,7 +96,7 @@ static int hsr_netdev_notify(struct notifier_block *nb, unsigned long event, break; /* Handled in ndo_change_mtu() */ mtu_max = hsr_get_max_mtu(port->hsr); master = hsr_port_get_hsr(port->hsr, HSR_PT_MASTER); - master->dev->mtu = mtu_max; + WRITE_ONCE(master->dev->mtu, mtu_max); break; case NETDEV_UNREGISTER: if (!is_hsr_master(dev)) { diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index c3af965dc407..ba205473522e 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -793,7 +793,7 @@ static void ipgre_link_update(struct net_device *dev, bool set_mtu) dev->needed_headroom += len; if (set_mtu) - dev->mtu = max_t(int, dev->mtu - len, 68); + WRITE_ONCE(dev->mtu, max_t(int, dev->mtu - len, 68)); if (test_bit(IP_TUNNEL_SEQ_BIT, tunnel->parms.o_flags) || (test_bit(IP_TUNNEL_CSUM_BIT, tunnel->parms.o_flags) && diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index f1c5f6c3f2f8..bccef2fcf620 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -897,7 +897,7 @@ static void ip_tunnel_update(struct ip_tunnel_net *itn, t->fwmark = fwmark; mtu = ip_tunnel_bind_dev(dev); if (set_mtu) - dev->mtu = mtu; + WRITE_ONCE(dev->mtu, mtu); } dst_cache_reset(&t->dst_cache); netdev_state_change(dev); @@ -1082,7 +1082,7 @@ int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict) new_mtu = max_mtu; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } EXPORT_SYMBOL_GPL(__ip_tunnel_change_mtu); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 5aec79c2af1a..9dee0c127955 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1746,7 +1746,7 @@ int ip6_tnl_change_mtu(struct net_device *dev, int new_mtu) if (new_mtu > IP_MAX_MTU - dev->hard_header_len) return -EINVAL; } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } EXPORT_SYMBOL(ip6_tnl_change_mtu); diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index 78344cf3867e..590737c27537 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -666,7 +666,8 @@ static void vti6_link_config(struct ip6_tnl *t, bool keep_mtu) dev->flags &= ~IFF_POINTOPOINT; if (keep_mtu && dev->mtu) { - dev->mtu = clamp(dev->mtu, dev->min_mtu, dev->max_mtu); + WRITE_ONCE(dev->mtu, + clamp(dev->mtu, dev->min_mtu, dev->max_mtu)); return; } diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 29850d0f0733..8badec6d82a2 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -424,7 +424,7 @@ static int teql_master_mtu(struct net_device *dev, int new_mtu) } while ((q = NEXT_SLAVE(q)) != m->slaves); } - dev->mtu = new_mtu; + WRITE_ONCE(dev->mtu, new_mtu); return 0; } -- cgit v1.2.3 From 445c0b69c72903528fdf0b34015b5b9142062c4b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 May 2024 12:30:32 +0000 Subject: mptcp: fix possible NULL dereferences subflow_add_reset_reason(skb, ...) can fail. We can not assume mptcp_get_ext(skb) always return a non NULL pointer. syzbot reported: general protection fault, probably for non-canonical address 0xdffffc0000000003: 0000 [#1] PREEMPT SMP KASAN PTI KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f] CPU: 0 PID: 5098 Comm: syz-executor132 Not tainted 6.9.0-rc6-syzkaller-01478-gcdc74c9d06e7 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 RIP: 0010:subflow_v6_route_req+0x2c7/0x490 net/mptcp/subflow.c:388 Code: 8d 7b 07 48 89 f8 48 c1 e8 03 42 0f b6 04 20 84 c0 0f 85 c0 01 00 00 0f b6 43 07 48 8d 1c c3 48 83 c3 18 48 89 d8 48 c1 e8 03 <42> 0f b6 04 20 84 c0 0f 85 84 01 00 00 0f b6 5b 01 83 e3 0f 48 89 RSP: 0018:ffffc9000362eb68 EFLAGS: 00010206 RAX: 0000000000000003 RBX: 0000000000000018 RCX: ffff888022039e00 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: ffff88807d961140 R08: ffffffff8b6cb76b R09: 1ffff1100fb2c230 R10: dffffc0000000000 R11: ffffed100fb2c231 R12: dffffc0000000000 R13: ffff888022bfe273 R14: ffff88802cf9cc80 R15: ffff88802ad5a700 FS: 0000555587ad2380(0000) GS:ffff8880b9400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f420c3f9720 CR3: 0000000022bfc000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: tcp_conn_request+0xf07/0x32c0 net/ipv4/tcp_input.c:7180 tcp_rcv_state_process+0x183c/0x4500 net/ipv4/tcp_input.c:6663 tcp_v6_do_rcv+0x8b2/0x1310 net/ipv6/tcp_ipv6.c:1673 tcp_v6_rcv+0x22b4/0x30b0 net/ipv6/tcp_ipv6.c:1910 ip6_protocol_deliver_rcu+0xc76/0x1570 net/ipv6/ip6_input.c:438 ip6_input_finish+0x186/0x2d0 net/ipv6/ip6_input.c:483 NF_HOOK+0x3a4/0x450 include/linux/netfilter.h:314 NF_HOOK+0x3a4/0x450 include/linux/netfilter.h:314 __netif_receive_skb_one_core net/core/dev.c:5625 [inline] __netif_receive_skb+0x1ea/0x650 net/core/dev.c:5739 netif_receive_skb_internal net/core/dev.c:5825 [inline] netif_receive_skb+0x1e8/0x890 net/core/dev.c:5885 tun_rx_batched+0x1b7/0x8f0 drivers/net/tun.c:1549 tun_get_user+0x2f35/0x4560 drivers/net/tun.c:2002 tun_chr_write_iter+0x113/0x1f0 drivers/net/tun.c:2048 call_write_iter include/linux/fs.h:2110 [inline] new_sync_write fs/read_write.c:497 [inline] vfs_write+0xa84/0xcb0 fs/read_write.c:590 ksys_write+0x1a0/0x2c0 fs/read_write.c:643 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 3e140491dd80 ("mptcp: support rstreason for passive reset") Reported-by: syzbot Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Reviewed-by: Jason Xing Link: https://lore.kernel.org/r/20240506123032.3351895-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index 0ac8d4850751..c1d13e555d10 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -287,6 +287,16 @@ int mptcp_subflow_init_cookie_req(struct request_sock *req, } EXPORT_SYMBOL_GPL(mptcp_subflow_init_cookie_req); +static enum sk_rst_reason mptcp_get_rst_reason(const struct sk_buff *skb) +{ + const struct mptcp_ext *mpext = mptcp_get_ext(skb); + + if (!mpext) + return SK_RST_REASON_NOT_SPECIFIED; + + return sk_rst_convert_mptcp_reason(mpext->reset_reason); +} + static struct dst_entry *subflow_v4_route_req(const struct sock *sk, struct sk_buff *skb, struct flowi *fl, @@ -308,13 +318,9 @@ static struct dst_entry *subflow_v4_route_req(const struct sock *sk, return dst; dst_release(dst); - if (!req->syncookie) { - struct mptcp_ext *mpext = mptcp_get_ext(skb); - enum sk_rst_reason reason; - - reason = sk_rst_convert_mptcp_reason(mpext->reset_reason); - tcp_request_sock_ops.send_reset(sk, skb, reason); - } + if (!req->syncookie) + tcp_request_sock_ops.send_reset(sk, skb, + mptcp_get_rst_reason(skb)); return NULL; } @@ -381,13 +387,9 @@ static struct dst_entry *subflow_v6_route_req(const struct sock *sk, return dst; dst_release(dst); - if (!req->syncookie) { - struct mptcp_ext *mpext = mptcp_get_ext(skb); - enum sk_rst_reason reason; - - reason = sk_rst_convert_mptcp_reason(mpext->reset_reason); - tcp6_request_sock_ops.send_reset(sk, skb, reason); - } + if (!req->syncookie) + tcp6_request_sock_ops.send_reset(sk, skb, + mptcp_get_rst_reason(skb)); return NULL; } #endif @@ -923,7 +925,7 @@ dispose_child: tcp_rsk(req)->drop_req = true; inet_csk_prepare_for_destroy_sock(child); tcp_done(child); - reason = sk_rst_convert_mptcp_reason(mptcp_get_ext(skb)->reset_reason); + reason = mptcp_get_rst_reason(skb); req->rsk_ops->send_reset(sk, skb, reason); /* The last child reference will be released by the caller */ -- cgit v1.2.3 From 19e35f24750ddf860c51e51c68cf07ea181b4881 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Sun, 5 May 2024 19:36:49 +0900 Subject: nfc: nci: Fix kcov check in nci_rx_work() Commit 7e8cdc97148c ("nfc: Add KCOV annotations") added kcov_remote_start_common()/kcov_remote_stop() pair into nci_rx_work(), with an assumption that kcov_remote_stop() is called upon continue of the for loop. But commit d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet") forgot to call kcov_remote_stop() before break of the for loop. Reported-by: syzbot Closes: https://syzkaller.appspot.com/bug?extid=0438378d6f157baae1a2 Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet") Suggested-by: Andrey Konovalov Signed-off-by: Tetsuo Handa Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/6d10f829-5a0c-405a-b39a-d7266f3a1a0b@I-love.SAKURA.ne.jp Signed-off-by: Jakub Kicinski --- net/nfc/nci/core.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c index 0d26c8ec9993..b133dc55304c 100644 --- a/net/nfc/nci/core.c +++ b/net/nfc/nci/core.c @@ -1518,6 +1518,7 @@ static void nci_rx_work(struct work_struct *work) if (!nci_plen(skb->data)) { kfree_skb(skb); + kcov_remote_stop(); break; } -- cgit v1.2.3 From 6963c508fd7ab66ae0b7ae3db9a62ca6267f1ae8 Mon Sep 17 00:00:00 2001 From: Gregory Detal Date: Mon, 6 May 2024 17:35:28 +0200 Subject: mptcp: only allow set existing scheduler for net.mptcp.scheduler The current behavior is to accept any strings as inputs, this results in an inconsistent result where an unexisting scheduler can be set: # sysctl -w net.mptcp.scheduler=notdefault net.mptcp.scheduler = notdefault This patch changes this behavior by checking for existing scheduler before accepting the input. Fixes: e3b2870b6d22 ("mptcp: add a new sysctl scheduler") Cc: stable@vger.kernel.org Signed-off-by: Gregory Detal Reviewed-by: Matthieu Baerts (NGI0) Tested-by: Geliang Tang Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240506-upstream-net-20240506-mptcp-sched-exist-v1-1-2ed1529e521e@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/ctrl.c | 39 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 38 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c index 13fe0748dde8..2963ba84e2ee 100644 --- a/net/mptcp/ctrl.c +++ b/net/mptcp/ctrl.c @@ -96,6 +96,43 @@ static void mptcp_pernet_set_defaults(struct mptcp_pernet *pernet) } #ifdef CONFIG_SYSCTL +static int mptcp_set_scheduler(const struct net *net, const char *name) +{ + struct mptcp_pernet *pernet = mptcp_get_pernet(net); + struct mptcp_sched_ops *sched; + int ret = 0; + + rcu_read_lock(); + sched = mptcp_sched_find(name); + if (sched) + strscpy(pernet->scheduler, name, MPTCP_SCHED_NAME_MAX); + else + ret = -ENOENT; + rcu_read_unlock(); + + return ret; +} + +static int proc_scheduler(struct ctl_table *ctl, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + const struct net *net = current->nsproxy->net_ns; + char val[MPTCP_SCHED_NAME_MAX]; + struct ctl_table tbl = { + .data = val, + .maxlen = MPTCP_SCHED_NAME_MAX, + }; + int ret; + + strscpy(val, mptcp_get_scheduler(net), MPTCP_SCHED_NAME_MAX); + + ret = proc_dostring(&tbl, write, buffer, lenp, ppos); + if (write && ret == 0) + ret = mptcp_set_scheduler(net, val); + + return ret; +} + static struct ctl_table mptcp_sysctl_table[] = { { .procname = "enabled", @@ -148,7 +185,7 @@ static struct ctl_table mptcp_sysctl_table[] = { .procname = "scheduler", .maxlen = MPTCP_SCHED_NAME_MAX, .mode = 0644, - .proc_handler = proc_dostring, + .proc_handler = proc_scheduler, }, { .procname = "close_timeout", -- cgit v1.2.3 From 4321de4497b24fbf22389331f4ecd4039a451aa9 Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Tue, 7 May 2024 13:20:25 +0200 Subject: page_pool: check for DMA sync shortcut earlier We can save a couple more function calls in the Page Pool code if we check for dma_need_sync() earlier, just when we test pp->p.dma_sync. Move both these checks into an inline wrapper and call the PP wrapper over the generic DMA sync function only when both are true. You can't cache the result of dma_need_sync() in &page_pool, as it may change anytime if an SWIOTLB buffer is allocated or mapped. Signed-off-by: Alexander Lobakin Signed-off-by: Christoph Hellwig --- net/core/page_pool.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/core/page_pool.c b/net/core/page_pool.c index c2819ff03dd2..4f9d1bd7f4d1 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -398,16 +398,26 @@ static struct page *__page_pool_get_cached(struct page_pool *pool) return page; } -static void page_pool_dma_sync_for_device(struct page_pool *pool, - struct page *page, - unsigned int dma_sync_size) +static void __page_pool_dma_sync_for_device(const struct page_pool *pool, + struct page *page, + u32 dma_sync_size) { +#if defined(CONFIG_HAS_DMA) && defined(CONFIG_DMA_NEED_SYNC) dma_addr_t dma_addr = page_pool_get_dma_addr(page); dma_sync_size = min(dma_sync_size, pool->p.max_len); - dma_sync_single_range_for_device(pool->p.dev, dma_addr, - pool->p.offset, dma_sync_size, - pool->p.dma_dir); + __dma_sync_single_for_device(pool->p.dev, dma_addr + pool->p.offset, + dma_sync_size, pool->p.dma_dir); +#endif +} + +static __always_inline void +page_pool_dma_sync_for_device(const struct page_pool *pool, + struct page *page, + u32 dma_sync_size) +{ + if (pool->dma_sync && dma_dev_need_sync(pool->p.dev)) + __page_pool_dma_sync_for_device(pool, page, dma_sync_size); } static bool page_pool_dma_map(struct page_pool *pool, struct page *page) @@ -429,8 +439,7 @@ static bool page_pool_dma_map(struct page_pool *pool, struct page *page) if (page_pool_set_dma_addr(page, dma)) goto unmap_failed; - if (pool->dma_sync) - page_pool_dma_sync_for_device(pool, page, pool->p.max_len); + page_pool_dma_sync_for_device(pool, page, pool->p.max_len); return true; @@ -699,9 +708,7 @@ __page_pool_put_page(struct page_pool *pool, struct page *page, if (likely(__page_pool_page_can_be_recycled(page))) { /* Read barrier done in page_ref_count / READ_ONCE */ - if (pool->dma_sync) - page_pool_dma_sync_for_device(pool, page, - dma_sync_size); + page_pool_dma_sync_for_device(pool, page, dma_sync_size); if (allow_direct && in_softirq() && page_pool_recycle_in_cache(page, pool)) @@ -812,9 +819,7 @@ static struct page *page_pool_drain_frag(struct page_pool *pool, return NULL; if (__page_pool_page_can_be_recycled(page)) { - if (pool->dma_sync) - page_pool_dma_sync_for_device(pool, page, -1); - + page_pool_dma_sync_for_device(pool, page, -1); return page; } -- cgit v1.2.3 From 163943ac00cb31ac1a88ce5f78a7e2ead37329ec Mon Sep 17 00:00:00 2001 From: Alexander Lobakin Date: Tue, 7 May 2024 13:20:26 +0200 Subject: xsk: use generic DMA sync shortcut instead of a custom one XSk infra's been using its own DMA sync shortcut to try avoiding redundant function calls. Now that there is a generic one, remove the custom implementation and rely on the generic helpers. xsk_buff_dma_sync_for_cpu() doesn't need the second argument anymore, remove it. Signed-off-by: Alexander Lobakin Signed-off-by: Christoph Hellwig --- drivers/net/ethernet/engleder/tsnep_main.c | 2 +- drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c | 2 +- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 2 +- drivers/net/ethernet/intel/ice/ice_xsk.c | 2 +- drivers/net/ethernet/intel/igc/igc_main.c | 2 +- drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 2 +- .../net/ethernet/mellanox/mlx5/core/en/xsk/rx.c | 4 +-- drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 2 +- drivers/net/ethernet/netronome/nfp/nfd3/xsk.c | 2 +- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- include/net/xdp_sock_drv.h | 7 ++---- include/net/xsk_buff_pool.h | 14 +++-------- net/xdp/xsk_buff_pool.c | 29 +++------------------- 13 files changed, 21 insertions(+), 51 deletions(-) (limited to 'net') diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c index 4b15af6b7122..44da335d66bd 100644 --- a/drivers/net/ethernet/engleder/tsnep_main.c +++ b/drivers/net/ethernet/engleder/tsnep_main.c @@ -1587,7 +1587,7 @@ static int tsnep_rx_poll_zc(struct tsnep_rx *rx, struct napi_struct *napi, length = __le32_to_cpu(entry->desc_wb->properties) & TSNEP_DESC_LENGTH_MASK; xsk_buff_set_size(entry->xdp, length - ETH_FCS_LEN); - xsk_buff_dma_sync_for_cpu(entry->xdp, rx->xsk_pool); + xsk_buff_dma_sync_for_cpu(entry->xdp); /* RX metadata with timestamps is in front of actual data, * subtract metadata size to get length of actual data and diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c index 051748b997f3..a466c2379146 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-xsk.c @@ -55,7 +55,7 @@ static u32 dpaa2_xsk_run_xdp(struct dpaa2_eth_priv *priv, xdp_set_data_meta_invalid(xdp_buff); xdp_buff->rxq = &ch->xdp_rxq; - xsk_buff_dma_sync_for_cpu(xdp_buff, ch->xsk_pool); + xsk_buff_dma_sync_for_cpu(xdp_buff); xdp_act = bpf_prog_run_xdp(xdp_prog, xdp_buff); /* xdp.data pointer may have changed */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c index 11500003af0d..d20ce517426e 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c @@ -483,7 +483,7 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) bi = *i40e_rx_bi(rx_ring, next_to_process); xsk_buff_set_size(bi, size); - xsk_buff_dma_sync_for_cpu(bi, rx_ring->xsk_pool); + xsk_buff_dma_sync_for_cpu(bi); if (!first) first = bi; diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 1857220d27fe..cecd5b1e0757 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -879,7 +879,7 @@ int ice_clean_rx_irq_zc(struct ice_rx_ring *rx_ring, int budget) ICE_RX_FLX_DESC_PKT_LEN_M; xsk_buff_set_size(xdp, size); - xsk_buff_dma_sync_for_cpu(xdp, xsk_pool); + xsk_buff_dma_sync_for_cpu(xdp); if (!first) { first = xdp; diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 4d975d620a8e..07692e2a7c64 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -2813,7 +2813,7 @@ static int igc_clean_rx_irq_zc(struct igc_q_vector *q_vector, const int budget) } bi->xdp->data_end = bi->xdp->data + size; - xsk_buff_dma_sync_for_cpu(bi->xdp, ring->xsk_pool); + xsk_buff_dma_sync_for_cpu(bi->xdp); res = __igc_xdp_run_prog(adapter, prog, bi->xdp); switch (res) { diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c index d34d715c59eb..ee2d0ec12b2d 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c @@ -304,7 +304,7 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, } bi->xdp->data_end = bi->xdp->data + size; - xsk_buff_dma_sync_for_cpu(bi->xdp, rx_ring->xsk_pool); + xsk_buff_dma_sync_for_cpu(bi->xdp); xdp_res = ixgbe_run_xdp_zc(adapter, rx_ring, bi->xdp); if (likely(xdp_res & (IXGBE_XDP_TX | IXGBE_XDP_REDIR))) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c index b8dd74453655..1b7132fa70de 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c @@ -270,7 +270,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, /* mxbuf->rq is set on allocation, but cqe is per-packet so set it here */ mxbuf->cqe = cqe; xsk_buff_set_size(&mxbuf->xdp, cqe_bcnt); - xsk_buff_dma_sync_for_cpu(&mxbuf->xdp, rq->xsk_pool); + xsk_buff_dma_sync_for_cpu(&mxbuf->xdp); net_prefetch(mxbuf->xdp.data); /* Possible flows: @@ -319,7 +319,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, /* mxbuf->rq is set on allocation, but cqe is per-packet so set it here */ mxbuf->cqe = cqe; xsk_buff_set_size(&mxbuf->xdp, cqe_bcnt); - xsk_buff_dma_sync_for_cpu(&mxbuf->xdp, rq->xsk_pool); + xsk_buff_dma_sync_for_cpu(&mxbuf->xdp); net_prefetch(mxbuf->xdp.data); prog = rcu_dereference(rq->xdp_prog); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index d601b5faaed5..b5333da20e8a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -917,7 +917,7 @@ INDIRECT_CALLABLE_SCOPE bool mlx5e_post_rx_wqes(struct mlx5e_rq *rq) if (!rq->xsk_pool) { count = mlx5e_refill_rx_wqes(rq, head, wqe_bulk); - } else if (likely(!rq->xsk_pool->dma_need_sync)) { + } else if (likely(!dma_dev_need_sync(rq->pdev))) { mlx5e_xsk_free_rx_wqes(rq, head, wqe_bulk); count = mlx5e_xsk_alloc_rx_wqes_batched(rq, head, wqe_bulk); } else { diff --git a/drivers/net/ethernet/netronome/nfp/nfd3/xsk.c b/drivers/net/ethernet/netronome/nfp/nfd3/xsk.c index 45be6954d5aa..01cfa9cc1b5e 100644 --- a/drivers/net/ethernet/netronome/nfp/nfd3/xsk.c +++ b/drivers/net/ethernet/netronome/nfp/nfd3/xsk.c @@ -184,7 +184,7 @@ nfp_nfd3_xsk_rx(struct nfp_net_rx_ring *rx_ring, int budget, xrxbuf->xdp->data += meta_len; xrxbuf->xdp->data_end = xrxbuf->xdp->data + pkt_len; xdp_set_data_meta_invalid(xrxbuf->xdp); - xsk_buff_dma_sync_for_cpu(xrxbuf->xdp, r_vec->xsk_pool); + xsk_buff_dma_sync_for_cpu(xrxbuf->xdp); net_prefetch(xrxbuf->xdp->data); if (meta_len) { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 7c6fb14b5555..206cba44dd30 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -5352,7 +5352,7 @@ read_again: /* RX buffer is good and fit into a XSK pool buffer */ buf->xdp->data_end = buf->xdp->data + buf1_len; - xsk_buff_dma_sync_for_cpu(buf->xdp, rx_q->xsk_pool); + xsk_buff_dma_sync_for_cpu(buf->xdp); prog = READ_ONCE(priv->xdp_prog); res = __stmmac_xdp_run_prog(priv, prog, buf->xdp); diff --git a/include/net/xdp_sock_drv.h b/include/net/xdp_sock_drv.h index c9aec9ab6191..0a5dca2b2b3f 100644 --- a/include/net/xdp_sock_drv.h +++ b/include/net/xdp_sock_drv.h @@ -219,13 +219,10 @@ static inline struct xsk_tx_metadata *xsk_buff_get_metadata(struct xsk_buff_pool return meta; } -static inline void xsk_buff_dma_sync_for_cpu(struct xdp_buff *xdp, struct xsk_buff_pool *pool) +static inline void xsk_buff_dma_sync_for_cpu(struct xdp_buff *xdp) { struct xdp_buff_xsk *xskb = container_of(xdp, struct xdp_buff_xsk, xdp); - if (!pool->dma_need_sync) - return; - xp_dma_sync_for_cpu(xskb); } @@ -402,7 +399,7 @@ static inline struct xsk_tx_metadata *xsk_buff_get_metadata(struct xsk_buff_pool return NULL; } -static inline void xsk_buff_dma_sync_for_cpu(struct xdp_buff *xdp, struct xsk_buff_pool *pool) +static inline void xsk_buff_dma_sync_for_cpu(struct xdp_buff *xdp) { } diff --git a/include/net/xsk_buff_pool.h b/include/net/xsk_buff_pool.h index 99dd7376df6a..bacb33f1e3e5 100644 --- a/include/net/xsk_buff_pool.h +++ b/include/net/xsk_buff_pool.h @@ -43,7 +43,6 @@ struct xsk_dma_map { refcount_t users; struct list_head list; /* Protected by the RTNL_LOCK */ u32 dma_pages_cnt; - bool dma_need_sync; }; struct xsk_buff_pool { @@ -82,7 +81,6 @@ struct xsk_buff_pool { u8 tx_metadata_len; /* inherited from umem */ u8 cached_need_wakeup; bool uses_need_wakeup; - bool dma_need_sync; bool unaligned; bool tx_sw_csum; void *addrs; @@ -155,21 +153,17 @@ static inline dma_addr_t xp_get_frame_dma(struct xdp_buff_xsk *xskb) return xskb->frame_dma; } -void xp_dma_sync_for_cpu_slow(struct xdp_buff_xsk *xskb); static inline void xp_dma_sync_for_cpu(struct xdp_buff_xsk *xskb) { - xp_dma_sync_for_cpu_slow(xskb); + dma_sync_single_for_cpu(xskb->pool->dev, xskb->dma, + xskb->pool->frame_len, + DMA_BIDIRECTIONAL); } -void xp_dma_sync_for_device_slow(struct xsk_buff_pool *pool, dma_addr_t dma, - size_t size); static inline void xp_dma_sync_for_device(struct xsk_buff_pool *pool, dma_addr_t dma, size_t size) { - if (!pool->dma_need_sync) - return; - - xp_dma_sync_for_device_slow(pool, dma, size); + dma_sync_single_for_device(pool->dev, dma, size, DMA_BIDIRECTIONAL); } /* Masks for xdp_umem_page flags. diff --git a/net/xdp/xsk_buff_pool.c b/net/xdp/xsk_buff_pool.c index ce60ecd48a4d..c0e0204b9630 100644 --- a/net/xdp/xsk_buff_pool.c +++ b/net/xdp/xsk_buff_pool.c @@ -338,7 +338,6 @@ static struct xsk_dma_map *xp_create_dma_map(struct device *dev, struct net_devi dma_map->netdev = netdev; dma_map->dev = dev; - dma_map->dma_need_sync = false; dma_map->dma_pages_cnt = nr_pages; refcount_set(&dma_map->users, 1); list_add(&dma_map->list, &umem->xsk_dma_list); @@ -424,7 +423,6 @@ static int xp_init_dma_info(struct xsk_buff_pool *pool, struct xsk_dma_map *dma_ pool->dev = dma_map->dev; pool->dma_pages_cnt = dma_map->dma_pages_cnt; - pool->dma_need_sync = dma_map->dma_need_sync; memcpy(pool->dma_pages, dma_map->dma_pages, pool->dma_pages_cnt * sizeof(*pool->dma_pages)); @@ -460,8 +458,6 @@ int xp_dma_map(struct xsk_buff_pool *pool, struct device *dev, __xp_dma_unmap(dma_map, attrs); return -ENOMEM; } - if (dma_need_sync(dev, dma)) - dma_map->dma_need_sync = true; dma_map->dma_pages[i] = dma; } @@ -557,11 +553,9 @@ struct xdp_buff *xp_alloc(struct xsk_buff_pool *pool) xskb->xdp.data_meta = xskb->xdp.data; xskb->xdp.flags = 0; - if (pool->dma_need_sync) { - dma_sync_single_range_for_device(pool->dev, xskb->dma, 0, - pool->frame_len, - DMA_BIDIRECTIONAL); - } + if (pool->dev) + xp_dma_sync_for_device(pool, xskb->dma, pool->frame_len); + return &xskb->xdp; } EXPORT_SYMBOL(xp_alloc); @@ -633,7 +627,7 @@ u32 xp_alloc_batch(struct xsk_buff_pool *pool, struct xdp_buff **xdp, u32 max) { u32 nb_entries1 = 0, nb_entries2; - if (unlikely(pool->dma_need_sync)) { + if (unlikely(pool->dev && dma_dev_need_sync(pool->dev))) { struct xdp_buff *buff; /* Slow path */ @@ -693,18 +687,3 @@ dma_addr_t xp_raw_get_dma(struct xsk_buff_pool *pool, u64 addr) (addr & ~PAGE_MASK); } EXPORT_SYMBOL(xp_raw_get_dma); - -void xp_dma_sync_for_cpu_slow(struct xdp_buff_xsk *xskb) -{ - dma_sync_single_range_for_cpu(xskb->pool->dev, xskb->dma, 0, - xskb->pool->frame_len, DMA_BIDIRECTIONAL); -} -EXPORT_SYMBOL(xp_dma_sync_for_cpu_slow); - -void xp_dma_sync_for_device_slow(struct xsk_buff_pool *pool, dma_addr_t dma, - size_t size) -{ - dma_sync_single_range_for_device(pool->dev, dma, 0, - size, DMA_BIDIRECTIONAL); -} -EXPORT_SYMBOL(xp_dma_sync_for_device_slow); -- cgit v1.2.3 From 96c6f337951a03ab40e13a44cf5ea59b14725721 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Fri, 3 May 2024 15:13:40 +0200 Subject: net: dsa: add support for DCB get/set apptrust configuration Add DCB support to get/set trust configuration for different packet priority information sources. Some switch allow to chose different source of packet priority classification. For example on KSZ switches it is possible to configure VLAN PCP and/or DSCP sources. Signed-off-by: Oleksij Rempel Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 4 ++++ net/dsa/user.c | 28 ++++++++++++++++++++++++++++ 2 files changed, 32 insertions(+) (limited to 'net') diff --git a/include/net/dsa.h b/include/net/dsa.h index eef702dbea78..40c127a30187 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -955,6 +955,10 @@ struct dsa_switch_ops { u8 prio); int (*port_del_dscp_prio)(struct dsa_switch *ds, int port, u8 dscp, u8 prio); + int (*port_set_apptrust)(struct dsa_switch *ds, int port, + const u8 *sel, int nsel); + int (*port_get_apptrust)(struct dsa_switch *ds, int port, u8 *sel, + int *nsel); /* * Suspend and resume diff --git a/net/dsa/user.c b/net/dsa/user.c index 1f7b8f21db9e..89282c086776 100644 --- a/net/dsa/user.c +++ b/net/dsa/user.c @@ -2136,6 +2136,32 @@ out_conduit_failed: return err; } +static int __maybe_unused +dsa_user_dcbnl_set_apptrust(struct net_device *dev, u8 *sel, int nsel) +{ + struct dsa_port *dp = dsa_user_to_port(dev); + struct dsa_switch *ds = dp->ds; + int port = dp->index; + + if (!ds->ops->port_set_apptrust) + return -EOPNOTSUPP; + + return ds->ops->port_set_apptrust(ds, port, sel, nsel); +} + +static int __maybe_unused +dsa_user_dcbnl_get_apptrust(struct net_device *dev, u8 *sel, int *nsel) +{ + struct dsa_port *dp = dsa_user_to_port(dev); + struct dsa_switch *ds = dp->ds; + int port = dp->index; + + if (!ds->ops->port_get_apptrust) + return -EOPNOTSUPP; + + return ds->ops->port_get_apptrust(ds, port, sel, nsel); +} + static int __maybe_unused dsa_user_dcbnl_set_default_prio(struct net_device *dev, struct dcb_app *app) { @@ -2376,6 +2402,8 @@ static const struct ethtool_ops dsa_user_ethtool_ops = { static const struct dcbnl_rtnl_ops __maybe_unused dsa_user_dcbnl_ops = { .ieee_setapp = dsa_user_dcbnl_ieee_setapp, .ieee_delapp = dsa_user_dcbnl_ieee_delapp, + .dcbnl_setapptrust = dsa_user_dcbnl_set_apptrust, + .dcbnl_getapptrust = dsa_user_dcbnl_get_apptrust, }; static void dsa_user_get_stats64(struct net_device *dev, -- cgit v1.2.3 From 768cf8413883892c6ff0db783a0295a60886ec33 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Fri, 3 May 2024 15:13:42 +0200 Subject: net: add IEEE 802.1q specific helpers IEEE 802.1q specification provides recommendation and examples which can be used as good default values for different drivers. This patch implements mapping examples documented in IEEE 802.1Q-2022 in Annex I "I.3 Traffic type to traffic class mapping" and IETF DSCP naming and mapping DSCP to Traffic Type inspired by RFC8325. This helpers will be used in followup patches for dsa/microchip DCB implementation. Signed-off-by: Oleksij Rempel Signed-off-by: David S. Miller --- include/net/dscp.h | 76 ++++++++++++++ include/net/ieee8021q.h | 57 ++++++++++ net/Kconfig | 3 + net/core/Makefile | 1 + net/core/ieee8021q_helpers.c | 242 +++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 379 insertions(+) create mode 100644 include/net/dscp.h create mode 100644 include/net/ieee8021q.h create mode 100644 net/core/ieee8021q_helpers.c (limited to 'net') diff --git a/include/net/dscp.h b/include/net/dscp.h new file mode 100644 index 000000000000..ba40540868c9 --- /dev/null +++ b/include/net/dscp.h @@ -0,0 +1,76 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2024 Pengutronix, Oleksij Rempel */ + +#ifndef __DSCP_H__ +#define __DSCP_H__ + +/* + * DSCP Pools and Codepoint Space Division: + * + * The Differentiated Services (Diffserv) architecture defines a method for + * classifying and managing network traffic using the DS field in IPv4 and IPv6 + * packet headers. This field can carry one of 64 distinct DSCP (Differentiated + * Services Code Point) values, which are divided into three pools based on + * their Least Significant Bits (LSB) patterns and intended usage. Each pool has + * a specific registration procedure for assigning DSCP values: + * + * Pool 1 (Standards Action Pool): + * - Codepoint Space: xxxxx0 + * This pool includes DSCP values ending in '0' (binary), allocated via + * Standards Action. It is intended for globally recognized traffic classes, + * ensuring interoperability across the internet. This pool encompasses + * well-known DSCP values such as CS0-CS7, AFxx, EF, and VOICE-ADMIT. + * + * Pool 2 (Experimental/Local Use Pool): + * - Codepoint Space: xxxx11 + * Reserved for DSCP values ending in '11' (binary), this pool is designated + * for Experimental or Local Use. It allows for private or temporary traffic + * marking schemes not intended for standardized global use, facilitating + * testing and network-specific configurations without impacting + * interoperability. + * + * Pool 3 (Preferential Standardization Pool): + * - Codepoint Space: xxxx01 + * Initially reserved for experimental or local use, this pool now serves as + * a secondary standardization resource should Pool 1 become exhausted. DSCP + * values ending in '01' (binary) are assigned via Standards Action, with a + * focus on adopting new, standardized traffic classes as the need arises. + * + * For pool updates see: + * https://www.iana.org/assignments/dscp-registry/dscp-registry.xhtml + */ + +/* Pool 1: Standardized DSCP values as per [RFC8126] */ +#define DSCP_CS0 0 /* 000000, [RFC2474] */ +/* CS0 is some times called default (DF) */ +#define DSCP_DF 0 /* 000000, [RFC2474] */ +#define DSCP_CS1 8 /* 001000, [RFC2474] */ +#define DSCP_CS2 16 /* 010000, [RFC2474] */ +#define DSCP_CS3 24 /* 011000, [RFC2474] */ +#define DSCP_CS4 32 /* 100000, [RFC2474] */ +#define DSCP_CS5 40 /* 101000, [RFC2474] */ +#define DSCP_CS6 48 /* 110000, [RFC2474] */ +#define DSCP_CS7 56 /* 111000, [RFC2474] */ +#define DSCP_AF11 10 /* 001010, [RFC2597] */ +#define DSCP_AF12 12 /* 001100, [RFC2597] */ +#define DSCP_AF13 14 /* 001110, [RFC2597] */ +#define DSCP_AF21 18 /* 010010, [RFC2597] */ +#define DSCP_AF22 20 /* 010100, [RFC2597] */ +#define DSCP_AF23 22 /* 010110, [RFC2597] */ +#define DSCP_AF31 26 /* 011010, [RFC2597] */ +#define DSCP_AF32 28 /* 011100, [RFC2597] */ +#define DSCP_AF33 30 /* 011110, [RFC2597] */ +#define DSCP_AF41 34 /* 100010, [RFC2597] */ +#define DSCP_AF42 36 /* 100100, [RFC2597] */ +#define DSCP_AF43 38 /* 100110, [RFC2597] */ +#define DSCP_EF 46 /* 101110, [RFC3246] */ +#define DSCP_VOICE_ADMIT 44 /* 101100, [RFC5865] */ + +/* Pool 3: Standardized assignments, previously available for experimental/local + * use + */ +#define DSCP_LE 1 /* 000001, [RFC8622] */ + +#define DSCP_MAX 64 + +#endif /* __DSCP_H__ */ diff --git a/include/net/ieee8021q.h b/include/net/ieee8021q.h new file mode 100644 index 000000000000..8bfe903dd3d0 --- /dev/null +++ b/include/net/ieee8021q.h @@ -0,0 +1,57 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2024 Pengutronix, Oleksij Rempel */ + +#ifndef _NET_IEEE8021Q_H +#define _NET_IEEE8021Q_H + +#include + +/** + * enum ieee8021q_traffic_type - 802.1Q traffic type priority values (802.1Q-2022) + * + * @IEEE8021Q_TT_BK: Background + * @IEEE8021Q_TT_BE: Best Effort (default). According to 802.1Q-2022, BE is 0 + * but has higher priority than BK which is 1. + * @IEEE8021Q_TT_EE: Excellent Effort + * @IEEE8021Q_TT_CA: Critical Applications + * @IEEE8021Q_TT_VI: Video, < 100 ms latency and jitter + * @IEEE8021Q_TT_VO: Voice, < 10 ms latency and jitter + * @IEEE8021Q_TT_IC: Internetwork Control + * @IEEE8021Q_TT_NC: Network Control + */ +enum ieee8021q_traffic_type { + IEEE8021Q_TT_BK = 0, + IEEE8021Q_TT_BE = 1, + IEEE8021Q_TT_EE = 2, + IEEE8021Q_TT_CA = 3, + IEEE8021Q_TT_VI = 4, + IEEE8021Q_TT_VO = 5, + IEEE8021Q_TT_IC = 6, + IEEE8021Q_TT_NC = 7, + + /* private: */ + IEEE8021Q_TT_MAX, +}; + +#define SIMPLE_IETF_DSCP_TO_IEEE8021Q_TT(dscp) ((dscp >> 3) & 0x7) + +#if IS_ENABLED(CONFIG_NET_IEEE8021Q_HELPERS) + +int ietf_dscp_to_ieee8021q_tt(u8 dscp); +int ieee8021q_tt_to_tc(enum ieee8021q_traffic_type tt, unsigned int num_queues); + +#else + +static inline int ietf_dscp_to_ieee8021q_tt(u8 dscp) +{ + return -EOPNOTSUPP; +} + +static inline int ieee8021q_tt_to_tc(enum ieee8021q_traffic_type tt, + unsigned int num_queues) +{ + return -EOPNOTSUPP; +} + +#endif +#endif /* _NET_IEEE8021Q_H */ diff --git a/net/Kconfig b/net/Kconfig index d5ab791f7afa..f0a8692496ff 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -452,6 +452,9 @@ config GRO_CELLS config SOCK_VALIDATE_XMIT bool +config NET_IEEE8021Q_HELPERS + bool + config NET_SELFTESTS def_tristate PHYLIB depends on PHYLIB && INET diff --git a/net/core/Makefile b/net/core/Makefile index 21d6fbc7e884..62be9aef2528 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -26,6 +26,7 @@ obj-$(CONFIG_NETPOLL) += netpoll.o obj-$(CONFIG_FIB_RULES) += fib_rules.o obj-$(CONFIG_TRACEPOINTS) += net-traces.o obj-$(CONFIG_NET_DROP_MONITOR) += drop_monitor.o +obj-$(CONFIG_NET_IEEE8021Q_HELPERS) += ieee8021q_helpers.o obj-$(CONFIG_NET_SELFTESTS) += selftests.o obj-$(CONFIG_NETWORK_PHY_TIMESTAMPING) += timestamping.o obj-$(CONFIG_NET_PTP_CLASSIFY) += ptp_classifier.o diff --git a/net/core/ieee8021q_helpers.c b/net/core/ieee8021q_helpers.c new file mode 100644 index 000000000000..759a9b9f3f89 --- /dev/null +++ b/net/core/ieee8021q_helpers.c @@ -0,0 +1,242 @@ +// SPDX-License-Identifier: GPL-2.0 +// Copyright (c) 2024 Pengutronix, Oleksij Rempel + +#include +#include +#include +#include +#include + +/* The following arrays map Traffic Types (TT) to traffic classes (TC) for + * different number of queues as shown in the example provided by + * IEEE 802.1Q-2022 in Annex I "I.3 Traffic type to traffic class mapping" and + * Table I-1 "Traffic type to traffic class mapping". + */ +static const u8 ieee8021q_8queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, + [IEEE8021Q_TT_BE] = 1, + [IEEE8021Q_TT_EE] = 2, + [IEEE8021Q_TT_CA] = 3, + [IEEE8021Q_TT_VI] = 4, + [IEEE8021Q_TT_VO] = 5, + [IEEE8021Q_TT_IC] = 6, + [IEEE8021Q_TT_NC] = 7, +}; + +static const u8 ieee8021q_7queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, + [IEEE8021Q_TT_BE] = 1, + [IEEE8021Q_TT_EE] = 2, + [IEEE8021Q_TT_CA] = 3, + [IEEE8021Q_TT_VI] = 4, [IEEE8021Q_TT_VO] = 4, + [IEEE8021Q_TT_IC] = 5, + [IEEE8021Q_TT_NC] = 6, +}; + +static const u8 ieee8021q_6queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, + [IEEE8021Q_TT_BE] = 1, + [IEEE8021Q_TT_EE] = 2, [IEEE8021Q_TT_CA] = 2, + [IEEE8021Q_TT_VI] = 3, [IEEE8021Q_TT_VO] = 3, + [IEEE8021Q_TT_IC] = 4, + [IEEE8021Q_TT_NC] = 5, +}; + +static const u8 ieee8021q_5queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, [IEEE8021Q_TT_BE] = 0, + [IEEE8021Q_TT_EE] = 1, [IEEE8021Q_TT_CA] = 1, + [IEEE8021Q_TT_VI] = 2, [IEEE8021Q_TT_VO] = 2, + [IEEE8021Q_TT_IC] = 3, + [IEEE8021Q_TT_NC] = 4, +}; + +static const u8 ieee8021q_4queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, [IEEE8021Q_TT_BE] = 0, + [IEEE8021Q_TT_EE] = 1, [IEEE8021Q_TT_CA] = 1, + [IEEE8021Q_TT_VI] = 2, [IEEE8021Q_TT_VO] = 2, + [IEEE8021Q_TT_IC] = 3, [IEEE8021Q_TT_NC] = 3, +}; + +static const u8 ieee8021q_3queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, [IEEE8021Q_TT_BE] = 0, + [IEEE8021Q_TT_EE] = 0, [IEEE8021Q_TT_CA] = 0, + [IEEE8021Q_TT_VI] = 1, [IEEE8021Q_TT_VO] = 1, + [IEEE8021Q_TT_IC] = 2, [IEEE8021Q_TT_NC] = 2, +}; + +static const u8 ieee8021q_2queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, [IEEE8021Q_TT_BE] = 0, + [IEEE8021Q_TT_EE] = 0, [IEEE8021Q_TT_CA] = 0, + [IEEE8021Q_TT_VI] = 1, [IEEE8021Q_TT_VO] = 1, + [IEEE8021Q_TT_IC] = 1, [IEEE8021Q_TT_NC] = 1, +}; + +static const u8 ieee8021q_1queue_tt_tc_map[] = { + [IEEE8021Q_TT_BK] = 0, [IEEE8021Q_TT_BE] = 0, + [IEEE8021Q_TT_EE] = 0, [IEEE8021Q_TT_CA] = 0, + [IEEE8021Q_TT_VI] = 0, [IEEE8021Q_TT_VO] = 0, + [IEEE8021Q_TT_IC] = 0, [IEEE8021Q_TT_NC] = 0, +}; + +/** + * ieee8021q_tt_to_tc - Map IEEE 802.1Q Traffic Type to Traffic Class + * @tt: IEEE 802.1Q Traffic Type + * @num_queues: Number of queues + * + * This function maps an IEEE 802.1Q Traffic Type to a Traffic Class (TC) based + * on the number of queues configured on the NIC. The mapping is based on the + * example provided by IEEE 802.1Q-2022 in Annex I "I.3 Traffic type to traffic + * class mapping" and Table I-1 "Traffic type to traffic class mapping". + * + * Return: Traffic Class corresponding to the given Traffic Type or negative + * value in case of error. + */ +int ieee8021q_tt_to_tc(enum ieee8021q_traffic_type tt, unsigned int num_queues) +{ + if (tt < 0 || tt >= IEEE8021Q_TT_MAX) { + pr_err("Requested Traffic Type (%d) is out of range (%d)\n", tt, + IEEE8021Q_TT_MAX); + return -EINVAL; + } + + switch (num_queues) { + case 8: + compiletime_assert(ARRAY_SIZE(ieee8021q_8queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_8queue_tt_tc_map != max - 1"); + return ieee8021q_8queue_tt_tc_map[tt]; + case 7: + compiletime_assert(ARRAY_SIZE(ieee8021q_7queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_7queue_tt_tc_map != max - 1"); + + return ieee8021q_7queue_tt_tc_map[tt]; + case 6: + compiletime_assert(ARRAY_SIZE(ieee8021q_6queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_6queue_tt_tc_map != max - 1"); + + return ieee8021q_6queue_tt_tc_map[tt]; + case 5: + compiletime_assert(ARRAY_SIZE(ieee8021q_5queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_5queue_tt_tc_map != max - 1"); + + return ieee8021q_5queue_tt_tc_map[tt]; + case 4: + compiletime_assert(ARRAY_SIZE(ieee8021q_4queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_4queue_tt_tc_map != max - 1"); + + return ieee8021q_4queue_tt_tc_map[tt]; + case 3: + compiletime_assert(ARRAY_SIZE(ieee8021q_3queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_3queue_tt_tc_map != max - 1"); + + return ieee8021q_3queue_tt_tc_map[tt]; + case 2: + compiletime_assert(ARRAY_SIZE(ieee8021q_2queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_2queue_tt_tc_map != max - 1"); + + return ieee8021q_2queue_tt_tc_map[tt]; + case 1: + compiletime_assert(ARRAY_SIZE(ieee8021q_1queue_tt_tc_map) != + IEEE8021Q_TT_MAX - 1, + "ieee8021q_1queue_tt_tc_map != max - 1"); + + return ieee8021q_1queue_tt_tc_map[tt]; + } + + pr_err("Invalid number of queues %d\n", num_queues); + + return -EINVAL; +} +EXPORT_SYMBOL_GPL(ieee8021q_tt_to_tc); + +/** + * ietf_dscp_to_ieee8021q_tt - Map IETF DSCP to IEEE 802.1Q Traffic Type + * @dscp: IETF DSCP value + * + * This function maps an IETF DSCP value to an IEEE 802.1Q Traffic Type (TT). + * Since there is no corresponding mapping between DSCP and IEEE 802.1Q Traffic + * Type, this function is inspired by the RFC8325 documentation which describe + * the mapping between DSCP and 802.11 User Priority (UP) values. + * + * Return: IEEE 802.1Q Traffic Type corresponding to the given DSCP value + */ +int ietf_dscp_to_ieee8021q_tt(u8 dscp) +{ + switch (dscp) { + case DSCP_CS0: + /* Comment from RFC8325: + * [RFC4594], Section 4.8, recommends High-Throughput Data be marked + * AF1x (that is, AF11, AF12, and AF13, according to the rules defined + * in [RFC2475]). + * + * By default (as described in Section 2.3), High-Throughput Data will + * map to UP 1 and, thus, to the Background Access Category (AC_BK), + * which is contrary to the intent expressed in [RFC4594]. + + * Unfortunately, there really is no corresponding fit for the High- + * Throughput Data service class within the constrained 4 Access + * Category [IEEE.802.11-2016] model. If the High-Throughput Data + * service class is assigned to the Best Effort Access Category (AC_BE), + * then it would contend with Low-Latency Data (while [RFC4594] + * recommends a distinction in servicing between these service classes) + * as well as with the default service class; alternatively, if it is + * assigned to the Background Access Category (AC_BK), then it would + * receive a less-then-best-effort service and contend with Low-Priority + * Data (as discussed in Section 4.2.10). + * + * As such, since there is no directly corresponding fit for the High- + * Throughout Data service class within the [IEEE.802.11-2016] model, it + * is generally RECOMMENDED to map High-Throughput Data to UP 0, thereby + * admitting it to the Best Effort Access Category (AC_BE). + * + * Note: The above text is from RFC8325 which is describing the mapping + * between DSCP and 802.11 User Priority (UP) values. The mapping + * between UP and IEEE 802.1Q Traffic Type is not defined in the RFC but + * the 802.11 AC_BK and AC_BE are closely related to the IEEE 802.1Q + * Traffic Types BE and BK. + */ + case DSCP_AF11: + case DSCP_AF12: + case DSCP_AF13: + return IEEE8021Q_TT_BE; + /* Comment from RFC8325: + * RFC3662 and RFC4594 both recommend Low-Priority Data be marked + * with DSCP CS1. The Low-Priority Data service class loosely + * corresponds to the [IEEE.802.11-2016] Background Access Category + */ + case DSCP_CS1: + return IEEE8021Q_TT_BK; + case DSCP_CS2: + case DSCP_AF21: + case DSCP_AF22: + case DSCP_AF23: + return IEEE8021Q_TT_EE; + case DSCP_CS3: + case DSCP_AF31: + case DSCP_AF32: + case DSCP_AF33: + return IEEE8021Q_TT_CA; + case DSCP_CS4: + case DSCP_AF41: + case DSCP_AF42: + case DSCP_AF43: + return IEEE8021Q_TT_VI; + case DSCP_CS5: + case DSCP_EF: + case DSCP_VOICE_ADMIT: + return IEEE8021Q_TT_VO; + case DSCP_CS6: + return IEEE8021Q_TT_IC; + case DSCP_CS7: + return IEEE8021Q_TT_NC; + } + + return SIMPLE_IETF_DSCP_TO_IEEE8021Q_TT(dscp); +} +EXPORT_SYMBOL_GPL(ietf_dscp_to_ieee8021q_tt); -- cgit v1.2.3 From 5f5109af47535bc613c12a089dded425a373a12f Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Fri, 3 May 2024 15:13:49 +0200 Subject: net: dsa: add support switches global DSCP priority mapping Some switches like Microchip KSZ variants do not support per port DSCP priority configuration. Instead there is a global DSCP mapping table. To handle it, we will accept set/del request to any of user ports to make global configuration and update dcb app entries for all other ports. Signed-off-by: Oleksij Rempel Signed-off-by: David S. Miller --- include/net/dsa.h | 9 +++++++ net/dsa/user.c | 75 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 84 insertions(+) (limited to 'net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 40c127a30187..b60e7e410aba 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -433,6 +433,11 @@ struct dsa_switch { */ u32 fdb_isolation:1; + /* Drivers that have global DSCP mapping settings must set this to + * true to automatically apply the settings to all ports. + */ + u32 dscp_prio_mapping_is_global:1; + /* Listener for switch fabric events */ struct notifier_block nb; @@ -586,6 +591,10 @@ static inline bool dsa_is_user_port(struct dsa_switch *ds, int p) dsa_switch_for_each_port((_dp), (_ds)) \ if (dsa_port_is_user((_dp))) +#define dsa_switch_for_each_user_port_continue_reverse(_dp, _ds) \ + dsa_switch_for_each_port_continue_reverse((_dp), (_ds)) \ + if (dsa_port_is_user((_dp))) + #define dsa_switch_for_each_cpu_port(_dp, _ds) \ dsa_switch_for_each_port((_dp), (_ds)) \ if (dsa_port_is_cpu((_dp))) diff --git a/net/dsa/user.c b/net/dsa/user.c index 89282c086776..867c5fe9a4da 100644 --- a/net/dsa/user.c +++ b/net/dsa/user.c @@ -2189,6 +2189,58 @@ dsa_user_dcbnl_set_default_prio(struct net_device *dev, struct dcb_app *app) return 0; } +/* Update the DSCP prio entries on all user ports of the switch in case + * the switch supports global DSCP prio instead of per port DSCP prios. + */ +static int dsa_user_dcbnl_ieee_global_dscp_setdel(struct net_device *dev, + struct dcb_app *app, bool del) +{ + int (*setdel)(struct net_device *dev, struct dcb_app *app); + struct dsa_port *dp = dsa_user_to_port(dev); + struct dsa_switch *ds = dp->ds; + struct dsa_port *other_dp; + int err, restore_err; + + if (del) + setdel = dcb_ieee_delapp; + else + setdel = dcb_ieee_setapp; + + dsa_switch_for_each_user_port(other_dp, ds) { + struct net_device *user = other_dp->user; + + if (!user || user == dev) + continue; + + err = setdel(user, app); + if (err) + goto err_try_to_restore; + } + + return 0; + +err_try_to_restore: + + /* Revert logic to restore previous state of app entries */ + if (!del) + setdel = dcb_ieee_delapp; + else + setdel = dcb_ieee_setapp; + + dsa_switch_for_each_user_port_continue_reverse(other_dp, ds) { + struct net_device *user = other_dp->user; + + if (!user || user == dev) + continue; + + restore_err = setdel(user, app); + if (restore_err) + netdev_err(user, "Failed to restore DSCP prio entry configuration\n"); + } + + return err; +} + static int __maybe_unused dsa_user_dcbnl_add_dscp_prio(struct net_device *dev, struct dcb_app *app) { @@ -2220,6 +2272,17 @@ dsa_user_dcbnl_add_dscp_prio(struct net_device *dev, struct dcb_app *app) return err; } + if (!ds->dscp_prio_mapping_is_global) + return 0; + + err = dsa_user_dcbnl_ieee_global_dscp_setdel(dev, app, false); + if (err) { + if (ds->ops->port_del_dscp_prio) + ds->ops->port_del_dscp_prio(ds, port, dscp, new_prio); + dcb_ieee_delapp(dev, app); + return err; + } + return 0; } @@ -2290,6 +2353,18 @@ dsa_user_dcbnl_del_dscp_prio(struct net_device *dev, struct dcb_app *app) return err; } + if (!ds->dscp_prio_mapping_is_global) + return 0; + + err = dsa_user_dcbnl_ieee_global_dscp_setdel(dev, app, true); + if (err) { + if (ds->ops->port_add_dscp_prio) + ds->ops->port_add_dscp_prio(ds, port, dscp, + app->priority); + dcb_ieee_setapp(dev, app); + return err; + } + return 0; } -- cgit v1.2.3 From 86b29d830ad69eecff25b22dc96c14c6573718e6 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Sun, 5 May 2024 20:42:38 +0200 Subject: net: bridge: fix corrupted ethernet header on multicast-to-unicast The change from skb_copy to pskb_copy unfortunately changed the data copying to omit the ethernet header, since it was pulled before reaching this point. Fix this by calling __skb_push/pull around pskb_copy. Fixes: 59c878cbcdd8 ("net: bridge: fix multicast-to-unicast with fraglist GSO") Signed-off-by: Felix Fietkau Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_forward.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index d7c35f55bd69..d97064d460dc 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -258,6 +258,7 @@ static void maybe_deliver_addr(struct net_bridge_port *p, struct sk_buff *skb, { struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; const unsigned char *src = eth_hdr(skb)->h_source; + struct sk_buff *nskb; if (!should_deliver(p, skb)) return; @@ -266,12 +267,16 @@ static void maybe_deliver_addr(struct net_bridge_port *p, struct sk_buff *skb, if (skb->dev == p->dev && ether_addr_equal(src, addr)) return; - skb = pskb_copy(skb, GFP_ATOMIC); - if (!skb) { + __skb_push(skb, ETH_HLEN); + nskb = pskb_copy(skb, GFP_ATOMIC); + __skb_pull(skb, ETH_HLEN); + if (!nskb) { DEV_STATS_INC(dev, tx_dropped); return; } + skb = nskb; + __skb_pull(skb, ETH_HLEN); if (!is_broadcast_ether_addr(addr)) memcpy(eth_hdr(skb)->h_dest, addr, ETH_ALEN); -- cgit v1.2.3 From db3efdcf70c752e8a8deb16071d8e693c3ef8746 Mon Sep 17 00:00:00 2001 From: Peilin He Date: Tue, 7 May 2024 15:41:03 +0800 Subject: net/ipv4: add tracepoint for icmp_send Introduce a tracepoint for icmp_send, which can help users to get more detail information conveniently when icmp abnormal events happen. 1. Giving an usecase example: ============================= When an application experiences packet loss due to an unreachable UDP destination port, the kernel will send an exception message through the icmp_send function. By adding a trace point for icmp_send, developers or system administrators can obtain detailed information about the UDP packet loss, including the type, code, source address, destination address, source port, and destination port. This facilitates the trouble-shooting of UDP packet loss issues especially for those network-service applications. 2. Operation Instructions: ========================== Switch to the tracing directory. cd /sys/kernel/tracing Filter for destination port unreachable. echo "type==3 && code==3" > events/icmp/icmp_send/filter Enable trace event. echo 1 > events/icmp/icmp_send/enable 3. Result View: ================ udp_client_erro-11370 [002] ...s.12 124.728002: icmp_send: icmp_send: type=3, code=3. From 127.0.0.1:41895 to 127.0.0.1:6666 ulen=23 skbaddr=00000000589b167a Signed-off-by: Peilin He Signed-off-by: xu xin Reviewed-by: Yunkai Zhang Cc: Yang Yang Cc: Liu Chun Cc: Xuexin Jiang Reviewed-by: Steven Rostedt (Google) Signed-off-by: David S. Miller --- include/trace/events/icmp.h | 67 +++++++++++++++++++++++++++++++++++++++++++++ net/ipv4/icmp.c | 4 +++ 2 files changed, 71 insertions(+) create mode 100644 include/trace/events/icmp.h (limited to 'net') diff --git a/include/trace/events/icmp.h b/include/trace/events/icmp.h new file mode 100644 index 000000000000..31559796949a --- /dev/null +++ b/include/trace/events/icmp.h @@ -0,0 +1,67 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM icmp + +#if !defined(_TRACE_ICMP_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_ICMP_H + +#include +#include + +TRACE_EVENT(icmp_send, + + TP_PROTO(const struct sk_buff *skb, int type, int code), + + TP_ARGS(skb, type, code), + + TP_STRUCT__entry( + __field(const void *, skbaddr) + __field(int, type) + __field(int, code) + __array(__u8, saddr, 4) + __array(__u8, daddr, 4) + __field(__u16, sport) + __field(__u16, dport) + __field(unsigned short, ulen) + ), + + TP_fast_assign( + struct iphdr *iph = ip_hdr(skb); + struct udphdr *uh = udp_hdr(skb); + int proto_4 = iph->protocol; + __be32 *p32; + + __entry->skbaddr = skb; + __entry->type = type; + __entry->code = code; + + if (proto_4 != IPPROTO_UDP || (u8 *)uh < skb->head || + (u8 *)uh + sizeof(struct udphdr) + > skb_tail_pointer(skb)) { + __entry->sport = 0; + __entry->dport = 0; + __entry->ulen = 0; + } else { + __entry->sport = ntohs(uh->source); + __entry->dport = ntohs(uh->dest); + __entry->ulen = ntohs(uh->len); + } + + p32 = (__be32 *) __entry->saddr; + *p32 = iph->saddr; + + p32 = (__be32 *) __entry->daddr; + *p32 = iph->daddr; + ), + + TP_printk("icmp_send: type=%d, code=%d. From %pI4:%u to %pI4:%u ulen=%d skbaddr=%p", + __entry->type, __entry->code, + __entry->saddr, __entry->sport, __entry->daddr, + __entry->dport, __entry->ulen, __entry->skbaddr) +); + +#endif /* _TRACE_ICMP_H */ + +/* This part must be outside protection */ +#include + diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 207482d30dc7..ab6d0d98dbc3 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -93,6 +93,8 @@ #include #include #include +#define CREATE_TRACE_POINTS +#include /* * Build xmit assembly blocks @@ -770,6 +772,8 @@ void __icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info, if (!fl4.saddr) fl4.saddr = htonl(INADDR_DUMMY); + trace_icmp_send(skb_in, type, code); + icmp_push_reply(sk, &icmp_param, &fl4, &ipc, &rt); ende: ip_rt_put(rt); -- cgit v1.2.3 From 2e82a58d6c0797092eabe7ba66a532c11548047f Mon Sep 17 00:00:00 2001 From: Vincent Duvert Date: Sun, 5 May 2024 11:54:57 -0700 Subject: appletalk: Improve handling of broadcast packets When a broadcast AppleTalk packet is received, prefer queuing it on the socket whose address matches the address of the interface that received the packet (and is listening on the correct port). Userspace applications that handle such packets will usually send a response on the same socket that received the packet; this fix allows the response to be sent on the correct interface. If a socket matching the interface's address is not found, an arbitrary socket listening on the correct port will be used, if any. This matches the implementation's previous behavior. Fixes atalkd's responses to network information requests when multiple network interfaces are configured to use AppleTalk. Link: https://lore.kernel.org/netdev/20200722113752.1218-2-vincent.ldev@duvert.net/ Link: https://gist.github.com/VinDuv/4db433b6dce39d51a5b7847ee749b2a4 Signed-off-by: Vincent Duvert Signed-off-by: Doug Brown Signed-off-by: David S. Miller --- net/appletalk/ddp.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 198f5ba2feae..b068651984fe 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -88,6 +88,7 @@ static inline void atalk_remove_socket(struct sock *sk) static struct sock *atalk_search_socket(struct sockaddr_at *to, struct atalk_iface *atif) { + struct sock *def_socket = NULL; struct sock *s; read_lock_bh(&atalk_sockets_lock); @@ -98,8 +99,20 @@ static struct sock *atalk_search_socket(struct sockaddr_at *to, continue; if (to->sat_addr.s_net == ATADDR_ANYNET && - to->sat_addr.s_node == ATADDR_BCAST) - goto found; + to->sat_addr.s_node == ATADDR_BCAST) { + if (atif->address.s_node == at->src_node && + atif->address.s_net == at->src_net) { + /* This socket's address matches the address of the interface + * that received the packet -- use it + */ + goto found; + } + + /* Continue searching for a socket matching the interface address, + * but use this socket by default if no other one is found + */ + def_socket = s; + } if (to->sat_addr.s_net == at->src_net && (to->sat_addr.s_node == at->src_node || @@ -116,7 +129,7 @@ static struct sock *atalk_search_socket(struct sockaddr_at *to, goto found; } } - s = NULL; + s = def_socket; found: read_unlock_bh(&atalk_sockets_lock); return s; -- cgit v1.2.3 From b7ffab29a8e4705e7b9f05293669dc8a30272bc6 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Mon, 6 May 2024 12:32:05 +0200 Subject: net: bridge: switchdev: Improve error message for port_obj_add/del functions Enhance the error reporting mechanism in the switchdev framework to provide more informative and user-friendly error messages. Following feedback from users struggling to understand the implications of error messages like "failed (err=-28) to add object (id=2)", this update aims to clarify what operation failed and how this might impact the system or network. With this change, error messages now include a description of the failed operation, the specific object involved, and a brief explanation of the potential impact on the system. This approach helps administrators and developers better understand the context and severity of errors, facilitating quicker and more effective troubleshooting. Example of the improved logging: [ 70.516446] ksz-switch spi0.0 uplink: Failed to add Port Multicast Database entry (object id=2) with error: -ENOSPC (-28). [ 70.516446] Failure in updating the port's Multicast Database could lead to multicast forwarding issues. [ 70.516446] Current HW/SW setup lacks sufficient resources. This comprehensive update includes handling for a range of switchdev object IDs, ensuring that most operations within the switchdev framework benefit from clearer error reporting. Reviewed-by: Simon Horman Signed-off-by: Oleksij Rempel Signed-off-by: David S. Miller --- net/switchdev/switchdev.c | 99 +++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 95 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index c9189a970eec..6488ead9e464 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -244,6 +244,99 @@ static int switchdev_port_obj_notify(enum switchdev_notifier_type nt, return 0; } +static void switchdev_obj_id_to_helpful_msg(struct net_device *dev, + enum switchdev_obj_id obj_id, + int err, bool add) +{ + const char *action = add ? "add" : "del"; + const char *reason = ""; + const char *problem; + const char *obj_str; + + switch (obj_id) { + case SWITCHDEV_OBJ_ID_UNDEFINED: + obj_str = "Undefined object"; + problem = "Attempted operation is undefined, indicating a possible programming\n" + "error.\n"; + break; + case SWITCHDEV_OBJ_ID_PORT_VLAN: + obj_str = "VLAN entry"; + problem = "Failure in VLAN settings on this port might disrupt network\n" + "segmentation or traffic isolation, affecting network partitioning.\n"; + break; + case SWITCHDEV_OBJ_ID_PORT_MDB: + obj_str = "Port Multicast Database entry"; + problem = "Failure in updating the port's Multicast Database could lead to\n" + "multicast forwarding issues.\n"; + break; + case SWITCHDEV_OBJ_ID_HOST_MDB: + obj_str = "Host Multicast Database entry"; + problem = "Failure in updating the host's Multicast Database may impact multicast\n" + "group memberships or traffic delivery, affecting multicast\n" + "communication.\n"; + break; + case SWITCHDEV_OBJ_ID_MRP: + obj_str = "Media Redundancy Protocol configuration for port"; + problem = "Failure to set MRP ring ID on this port prevents communication with\n" + "the specified redundancy ring, resulting in an inability to engage\n" + "in MRP-based network operations.\n"; + break; + case SWITCHDEV_OBJ_ID_RING_TEST_MRP: + obj_str = "MRP Test Frame Operations for port"; + problem = "Failure to generate/monitor MRP test frames may lead to inability to\n" + "assess the ring's operational integrity and fault response, hindering\n" + "proactive network management.\n"; + break; + case SWITCHDEV_OBJ_ID_RING_ROLE_MRP: + obj_str = "MRP Ring Role Configuration"; + problem = "Improper MRP ring role configuration may create conflicts in the ring,\n" + "disrupting communication for all participants, or isolate the local\n" + "system from the ring, hindering its ability to communicate with other\n" + "participants.\n"; + break; + case SWITCHDEV_OBJ_ID_RING_STATE_MRP: + obj_str = "MRP Ring State Configuration"; + problem = "Failure to correctly set the MRP ring state can result in network\n" + "loops or leave segments without communication. In a Closed state,\n" + "it maintains loop prevention by blocking one MRM port, while an Open\n" + "state activates in response to failures, changing port states to\n" + "preserve network connectivity.\n"; + break; + case SWITCHDEV_OBJ_ID_IN_TEST_MRP: + obj_str = "MRP_InTest Frame Generation Configuration"; + problem = "Failure in managing MRP_InTest frame generation can misjudge the\n" + "interconnection ring's state, leading to incorrect blocking or\n" + "unblocking of the I/C port. This misconfiguration might result\n" + "in unintended network loops or isolate critical network segments,\n" + "compromising network integrity and reliability.\n"; + break; + case SWITCHDEV_OBJ_ID_IN_ROLE_MRP: + obj_str = "Interconnection Ring Role Configuration"; + problem = "Failure in incorrect assignment of interconnection ring roles\n" + "(MIM/MIC) can impair the formation of the interconnection rings.\n"; + break; + case SWITCHDEV_OBJ_ID_IN_STATE_MRP: + obj_str = "Interconnection Ring State Configuration"; + problem = "Failure in updating the interconnection ring state can lead in\n" + "case of Open state to incorrect blocking or unblocking of the\n" + "I/C port, resulting in unintended network loops or isolation\n" + "of critical network\n"; + break; + default: + obj_str = "Unknown object"; + problem = "Indicating a possible programming error.\n"; + } + + switch (err) { + case -ENOSPC: + reason = "Current HW/SW setup lacks sufficient resources.\n"; + break; + } + + netdev_err(dev, "Failed to %s %s (object id=%d) with error: %pe (%d).\n%s%s\n", + action, obj_str, obj_id, ERR_PTR(err), err, problem, reason); +} + static void switchdev_port_obj_add_deferred(struct net_device *dev, const void *data) { @@ -254,8 +347,7 @@ static void switchdev_port_obj_add_deferred(struct net_device *dev, err = switchdev_port_obj_notify(SWITCHDEV_PORT_OBJ_ADD, dev, obj, NULL); if (err && err != -EOPNOTSUPP) - netdev_err(dev, "failed (err=%d) to add object (id=%d)\n", - err, obj->id); + switchdev_obj_id_to_helpful_msg(dev, obj->id, err, true); if (obj->complete) obj->complete(dev, err, obj->complete_priv); } @@ -304,8 +396,7 @@ static void switchdev_port_obj_del_deferred(struct net_device *dev, err = switchdev_port_obj_del_now(dev, obj); if (err && err != -EOPNOTSUPP) - netdev_err(dev, "failed (err=%d) to del object (id=%d)\n", - err, obj->id); + switchdev_obj_id_to_helpful_msg(dev, obj->id, err, false); if (obj->complete) obj->complete(dev, err, obj->complete_priv); } -- cgit v1.2.3 From 4e13d3a9c25b7080f8a619f961e943fe08c2672c Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Mon, 6 May 2024 23:11:29 +0900 Subject: ipv6: Fix potential uninit-value access in __ip6_make_skb() As it was done in commit fc1092f51567 ("ipv4: Fix uninit-value access in __ip_make_skb()") for IPv4, check FLOWI_FLAG_KNOWN_NH on fl6->flowi6_flags instead of testing HDRINCL on the socket to avoid a race condition which causes uninit-value access. Fixes: ea30388baebc ("ipv6: Fix an uninit variable access bug in __ip6_make_skb()") Signed-off-by: Shigeru Yoshida Signed-off-by: David S. Miller --- net/ipv6/ip6_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index b9dd3a66e423..fa2937732665 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1933,7 +1933,7 @@ struct sk_buff *__ip6_make_skb(struct sock *sk, u8 icmp6_type; if (sk->sk_socket->type == SOCK_RAW && - !inet_test_bit(HDRINCL, sk)) + !(fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH)) icmp6_type = fl6->fl6_icmp_type; else icmp6_type = icmp6_hdr(skb)->icmp6_type; -- cgit v1.2.3 From 643bb5dbaef7d01f9bcf2a495bef855e81fd5714 Mon Sep 17 00:00:00 2001 From: Alexander Mikhalitsyn Date: Mon, 6 May 2024 16:14:43 +0200 Subject: ipvs: add READ_ONCE barrier for ipvs->sysctl_amemthresh Cc: Julian Anastasov Cc: Simon Horman Cc: Pablo Neira Ayuso Cc: Jozsef Kadlecsik Cc: Florian Westphal Suggested-by: Julian Anastasov Signed-off-by: Alexander Mikhalitsyn Acked-by: Julian Anastasov Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_ctl.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 50b5dbe40eb8..e122fa367b81 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -94,6 +94,7 @@ static void update_defense_level(struct netns_ipvs *ipvs) { struct sysinfo i; int availmem; + int amemthresh; int nomem; int to_change = -1; @@ -105,7 +106,8 @@ static void update_defense_level(struct netns_ipvs *ipvs) /* si_swapinfo(&i); */ /* availmem = availmem - (i.totalswap - i.freeswap); */ - nomem = (availmem < ipvs->sysctl_amemthresh); + amemthresh = max(READ_ONCE(ipvs->sysctl_amemthresh), 0); + nomem = (availmem < amemthresh); local_bh_disable(); @@ -145,9 +147,8 @@ static void update_defense_level(struct netns_ipvs *ipvs) break; case 1: if (nomem) { - ipvs->drop_rate = ipvs->drop_counter - = ipvs->sysctl_amemthresh / - (ipvs->sysctl_amemthresh-availmem); + ipvs->drop_counter = amemthresh / (amemthresh - availmem); + ipvs->drop_rate = ipvs->drop_counter; ipvs->sysctl_drop_packet = 2; } else { ipvs->drop_rate = 0; @@ -155,9 +156,8 @@ static void update_defense_level(struct netns_ipvs *ipvs) break; case 2: if (nomem) { - ipvs->drop_rate = ipvs->drop_counter - = ipvs->sysctl_amemthresh / - (ipvs->sysctl_amemthresh-availmem); + ipvs->drop_counter = amemthresh / (amemthresh - availmem); + ipvs->drop_rate = ipvs->drop_counter; } else { ipvs->drop_rate = 0; ipvs->sysctl_drop_packet = 1; -- cgit v1.2.3 From 2b696a2a101d36530227056b43e5c4ec05bcefc2 Mon Sep 17 00:00:00 2001 From: Alexander Mikhalitsyn Date: Mon, 6 May 2024 16:14:44 +0200 Subject: ipvs: allow some sysctls in non-init user namespaces Let's make all IPVS sysctls writtable even when network namespace is owned by non-initial user namespace. Let's make a few sysctls to be read-only for non-privileged users: - sync_qlen_max - sync_sock_size - run_estimation - est_cpulist - est_nice I'm trying to be conservative with this to prevent introducing any security issues in there. Maybe, we can allow more sysctls to be writable, but let's do this on-demand and when we see real use-case. This patch is motivated by user request in the LXC project [1]. Having this can help with running some Kubernetes [2] or Docker Swarm [3] workloads inside the system containers. Link: https://github.com/lxc/lxc/issues/4278 [1] Link: https://github.com/kubernetes/kubernetes/blob/b722d017a34b300a2284b890448e5a605f21d01e/pkg/proxy/ipvs/proxier.go#L103 [2] Link: https://github.com/moby/libnetwork/blob/3797618f9a38372e8107d8c06f6ae199e1133ae8/osl/namespace_linux.go#L682 [3] Cc: Julian Anastasov Cc: Simon Horman Cc: Pablo Neira Ayuso Cc: Jozsef Kadlecsik Cc: Florian Westphal Signed-off-by: Alexander Mikhalitsyn Acked-by: Julian Anastasov Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_ctl.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index e122fa367b81..b6d0dcf3a5c3 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -4269,6 +4269,7 @@ static int __net_init ip_vs_control_net_init_sysctl(struct netns_ipvs *ipvs) struct ctl_table *tbl; int idx, ret; size_t ctl_table_size = ARRAY_SIZE(vs_vars); + bool unpriv = net->user_ns != &init_user_ns; atomic_set(&ipvs->dropentry, 0); spin_lock_init(&ipvs->dropentry_lock); @@ -4283,10 +4284,6 @@ static int __net_init ip_vs_control_net_init_sysctl(struct netns_ipvs *ipvs) tbl = kmemdup(vs_vars, sizeof(vs_vars), GFP_KERNEL); if (tbl == NULL) return -ENOMEM; - - /* Don't export sysctls to unprivileged users */ - if (net->user_ns != &init_user_ns) - ctl_table_size = 0; } else tbl = vs_vars; /* Initialize sysctl defaults */ @@ -4312,10 +4309,17 @@ static int __net_init ip_vs_control_net_init_sysctl(struct netns_ipvs *ipvs) ipvs->sysctl_sync_ports = 1; tbl[idx++].data = &ipvs->sysctl_sync_ports; tbl[idx++].data = &ipvs->sysctl_sync_persist_mode; + ipvs->sysctl_sync_qlen_max = nr_free_buffer_pages() / 32; + if (unpriv) + tbl[idx].mode = 0444; tbl[idx++].data = &ipvs->sysctl_sync_qlen_max; + ipvs->sysctl_sync_sock_size = 0; + if (unpriv) + tbl[idx].mode = 0444; tbl[idx++].data = &ipvs->sysctl_sync_sock_size; + tbl[idx++].data = &ipvs->sysctl_cache_bypass; tbl[idx++].data = &ipvs->sysctl_expire_nodest_conn; tbl[idx++].data = &ipvs->sysctl_sloppy_tcp; @@ -4338,15 +4342,22 @@ static int __net_init ip_vs_control_net_init_sysctl(struct netns_ipvs *ipvs) tbl[idx++].data = &ipvs->sysctl_conn_reuse_mode; tbl[idx++].data = &ipvs->sysctl_schedule_icmp; tbl[idx++].data = &ipvs->sysctl_ignore_tunneled; + ipvs->sysctl_run_estimation = 1; + if (unpriv) + tbl[idx].mode = 0444; tbl[idx].extra2 = ipvs; tbl[idx++].data = &ipvs->sysctl_run_estimation; ipvs->est_cpulist_valid = 0; + if (unpriv) + tbl[idx].mode = 0444; tbl[idx].extra2 = ipvs; tbl[idx++].data = &ipvs->sysctl_est_cpulist; ipvs->sysctl_est_nice = IPVS_EST_NICE; + if (unpriv) + tbl[idx].mode = 0444; tbl[idx].extra2 = ipvs; tbl[idx++].data = &ipvs->sysctl_est_nice; -- cgit v1.2.3 From 1d3985ed0dd3de7ee152ede633fe859806c38595 Mon Sep 17 00:00:00 2001 From: Joel Granados Date: Tue, 7 May 2024 10:40:39 +0200 Subject: ax25: Remove superfuous "return" from ax25_ds_set_timer Remove the explicit call to "return" in the void ax25_ds_set_timer function that was introduced in 78a7b5dbc060 ("ax.25: x.25: Remove the now superfluous sentinel elements from ctl_table array"). Signed-off-by: Joel Granados Signed-off-by: David S. Miller --- net/ax25/ax25_ds_timer.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/ax25/ax25_ds_timer.c b/net/ax25/ax25_ds_timer.c index c50a58d9e368..c4f8adbf8144 100644 --- a/net/ax25/ax25_ds_timer.c +++ b/net/ax25/ax25_ds_timer.c @@ -55,7 +55,6 @@ void ax25_ds_set_timer(ax25_dev *ax25_dev) ax25_dev->dama.slave_timeout = msecs_to_jiffies(ax25_dev->values[AX25_VALUES_DS_TIMEOUT]) / 10; mod_timer(&ax25_dev->dama.slave_timer, jiffies + HZ); - return; } /* -- cgit v1.2.3 From ba4e103848d3a2a28a0445e39f4a9564187efe54 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 3 May 2024 16:07:39 +0100 Subject: rxrpc: Fix congestion control algorithm Make the following fixes to the congestion control algorithm: (1) Don't vary the cwnd starting value by the size of RXRPC_TX_SMSS since that's currently held constant - set to the size of a jumbo subpacket payload so that we can create jumbo packets on the fly. The current code invariably picks 3 as the starting value. Further, the starting cwnd needs to be an even number because we ack every other packet, so set it to 4. (2) Don't cut ssthresh when we see an ACK come from the peer with a receive window (rwind) less than ssthresh. ssthresh keeps track of characteristics of the connection whereas rwind may be reduced by the peer for any reason - and may be reduced to 0. Fixes: 1fc4fa2ac93d ("rxrpc: Fix congestion management") Fixes: 0851115090a3 ("rxrpc: Reduce ssthresh to peer's receive window") Signed-off-by: David Howells Suggested-by: Simon Wilkinson cc: Marc Dionne cc: linux-afs@lists.infradead.org Reviewed-by: Jeffrey Altman > Link: https://lore.kernel.org/r/20240503150749.1001323-2-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/ar-internal.h | 2 +- net/rxrpc/call_object.c | 7 +------ net/rxrpc/input.c | 3 --- 3 files changed, 2 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 08c0a32db8c7..08de24658f4f 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -697,7 +697,7 @@ struct rxrpc_call { * packets) rather than bytes. */ #define RXRPC_TX_SMSS RXRPC_JUMBO_DATALEN -#define RXRPC_MIN_CWND (RXRPC_TX_SMSS > 2190 ? 2 : RXRPC_TX_SMSS > 1095 ? 3 : 4) +#define RXRPC_MIN_CWND 4 u8 cong_cwnd; /* Congestion window size */ u8 cong_extra; /* Extra to send for congestion management */ u8 cong_ssthresh; /* Slow-start threshold */ diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 01fa71e8b1f7..f9e983a12c14 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -174,12 +174,7 @@ struct rxrpc_call *rxrpc_alloc_call(struct rxrpc_sock *rx, gfp_t gfp, call->rx_winsize = rxrpc_rx_window_size; call->tx_winsize = 16; - if (RXRPC_TX_SMSS > 2190) - call->cong_cwnd = 2; - else if (RXRPC_TX_SMSS > 1095) - call->cong_cwnd = 3; - else - call->cong_cwnd = 4; + call->cong_cwnd = RXRPC_MIN_CWND; call->cong_ssthresh = RXRPC_TX_MAX_WINDOW; call->rxnet = rxnet; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 3dedb8c0618c..860075f1290b 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -685,9 +685,6 @@ static void rxrpc_input_ack_trailer(struct rxrpc_call *call, struct sk_buff *skb call->tx_winsize = rwind; } - if (call->cong_ssthresh > rwind) - call->cong_ssthresh = rwind; - mtu = min(ntohl(trailer->maxMTU), ntohl(trailer->ifMTU)); peer = call->peer; -- cgit v1.2.3 From 012b7206918dcc5a4dcf1432b3e643114c95957e Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 3 May 2024 16:07:40 +0100 Subject: rxrpc: Only transmit one ACK per jumbo packet received Only generate one ACK packet for all the subpackets in a jumbo packet. If we would like to generate more than one ACK, we prioritise them base on their reason code, in the order, highest first: OutOfSeq > NoSpace > ExceedsWin > Duplicate > Requested > Delay > Idle For the first four, we reference the lowest offending subpacket; for the last three, the highest. This reduces the number of ACKs we end up transmitting to one per UDP packet transmitted to reduce network loading and packet parsing. Fixes: 5d7edbc9231e ("rxrpc: Get rid of the Rx ring") Signed-off-by: David Howells cc: Marc Dionne cc: linux-afs@lists.infradead.org Reviewed-by: Jeffrey Altman > Link: https://lore.kernel.org/r/20240503150749.1001323-3-dhowells@redhat.com Signed-off-by: Jakub Kicinski --- net/rxrpc/input.c | 46 +++++++++++++++++++++++++++++++++++----------- 1 file changed, 35 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 860075f1290b..16d49a861dbb 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -9,6 +9,17 @@ #include "ar-internal.h" +/* Override priority when generating ACKs for received DATA */ +static const u8 rxrpc_ack_priority[RXRPC_ACK__INVALID] = { + [RXRPC_ACK_IDLE] = 1, + [RXRPC_ACK_DELAY] = 2, + [RXRPC_ACK_REQUESTED] = 3, + [RXRPC_ACK_DUPLICATE] = 4, + [RXRPC_ACK_EXCEEDS_WINDOW] = 5, + [RXRPC_ACK_NOSPACE] = 6, + [RXRPC_ACK_OUT_OF_SEQUENCE] = 7, +}; + static void rxrpc_proto_abort(struct rxrpc_call *call, rxrpc_seq_t seq, enum rxrpc_abort_reason why) { @@ -365,7 +376,7 @@ static void rxrpc_input_queue_data(struct rxrpc_call *call, struct sk_buff *skb, * Process a DATA packet. */ static void rxrpc_input_data_one(struct rxrpc_call *call, struct sk_buff *skb, - bool *_notify) + bool *_notify, rxrpc_serial_t *_ack_serial, int *_ack_reason) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct sk_buff *oos; @@ -418,8 +429,6 @@ static void rxrpc_input_data_one(struct rxrpc_call *call, struct sk_buff *skb, /* Send an immediate ACK if we fill in a hole */ else if (!skb_queue_empty(&call->rx_oos_queue)) ack_reason = RXRPC_ACK_DELAY; - else - call->ackr_nr_unacked++; window++; if (after(window, wtop)) { @@ -497,12 +506,16 @@ static void rxrpc_input_data_one(struct rxrpc_call *call, struct sk_buff *skb, } send_ack: - if (ack_reason >= 0) - rxrpc_send_ACK(call, ack_reason, serial, - rxrpc_propose_ack_input_data); - else - rxrpc_propose_delay_ACK(call, serial, - rxrpc_propose_ack_input_data); + if (ack_reason >= 0) { + if (rxrpc_ack_priority[ack_reason] > rxrpc_ack_priority[*_ack_reason]) { + *_ack_serial = serial; + *_ack_reason = ack_reason; + } else if (rxrpc_ack_priority[ack_reason] == rxrpc_ack_priority[*_ack_reason] && + ack_reason == RXRPC_ACK_REQUESTED) { + *_ack_serial = serial; + *_ack_reason = ack_reason; + } + } } /* @@ -513,9 +526,11 @@ static bool rxrpc_input_split_jumbo(struct rxrpc_call *call, struct sk_buff *skb struct rxrpc_jumbo_header jhdr; struct rxrpc_skb_priv *sp = rxrpc_skb(skb), *jsp; struct sk_buff *jskb; + rxrpc_serial_t ack_serial = 0; unsigned int offset = sizeof(struct rxrpc_wire_header); unsigned int len = skb->len - offset; bool notify = false; + int ack_reason = 0; while (sp->hdr.flags & RXRPC_JUMBO_PACKET) { if (len < RXRPC_JUMBO_SUBPKTLEN) @@ -535,7 +550,7 @@ static bool rxrpc_input_split_jumbo(struct rxrpc_call *call, struct sk_buff *skb jsp = rxrpc_skb(jskb); jsp->offset = offset; jsp->len = RXRPC_JUMBO_DATALEN; - rxrpc_input_data_one(call, jskb, ¬ify); + rxrpc_input_data_one(call, jskb, ¬ify, &ack_serial, &ack_reason); rxrpc_free_skb(jskb, rxrpc_skb_put_jumbo_subpacket); sp->hdr.flags = jhdr.flags; @@ -548,7 +563,16 @@ static bool rxrpc_input_split_jumbo(struct rxrpc_call *call, struct sk_buff *skb sp->offset = offset; sp->len = len; - rxrpc_input_data_one(call, skb, ¬ify); + rxrpc_input_data_one(call, skb, ¬ify, &ack_serial, &ack_reason); + + if (ack_reason > 0) { + rxrpc_send_ACK(call, ack_reason, ack_serial, + rxrpc_propose_ack_input_data); + } else { + call->ackr_nr_unacked++; + rxrpc_propose_delay_ACK(call, sp->hdr.serial, + rxrpc_propose_ack_input_data); + } if (notify) { trace_rxrpc_notify_socket(call->debug_id, sp->hdr.serial); rxrpc_notify_socket(call); -- cgit v1.2.3 From 3b09b2bd0d62de1f359b0c130570711c99c5e80b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 May 2024 13:20:00 +0000 Subject: net: dst_cache: annotate data-races around dst_cache->reset_ts dst_cache->reset_ts is read or written locklessly, add READ_ONCE() and WRITE_ONCE() annotations. Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240507132000.614591-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/dst_cache.h | 2 +- net/core/dst_cache.c | 5 +++-- 2 files changed, 4 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/dst_cache.h b/include/net/dst_cache.h index df6622a5fe98..b4a55d2d5e71 100644 --- a/include/net/dst_cache.h +++ b/include/net/dst_cache.h @@ -76,7 +76,7 @@ struct dst_entry *dst_cache_get_ip6(struct dst_cache *dst_cache, */ static inline void dst_cache_reset(struct dst_cache *dst_cache) { - dst_cache->reset_ts = jiffies; + WRITE_ONCE(dst_cache->reset_ts, jiffies); } /** diff --git a/net/core/dst_cache.c b/net/core/dst_cache.c index 0c0bdb058c5b..f9df84a6c4b2 100644 --- a/net/core/dst_cache.c +++ b/net/core/dst_cache.c @@ -47,7 +47,8 @@ static struct dst_entry *dst_cache_per_cpu_get(struct dst_cache *dst_cache, /* the cache already hold a dst reference; it can't go away */ dst_hold(dst); - if (unlikely(!time_after(idst->refresh_ts, dst_cache->reset_ts) || + if (unlikely(!time_after(idst->refresh_ts, + READ_ONCE(dst_cache->reset_ts)) || (dst->obsolete && !dst->ops->check(dst, idst->cookie)))) { dst_cache_per_cpu_dst_set(idst, NULL, 0); dst_release(dst); @@ -170,7 +171,7 @@ void dst_cache_reset_now(struct dst_cache *dst_cache) if (!dst_cache->cache) return; - dst_cache->reset_ts = jiffies; + dst_cache_reset(dst_cache); for_each_possible_cpu(i) { struct dst_cache_pcpu *idst = per_cpu_ptr(dst_cache->cache, i); struct dst_entry *dst = idst->dst; -- cgit v1.2.3 From e2d09e5a1e8fb17a807dc4dce8dbc39d9fea3788 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 May 2024 13:27:17 +0000 Subject: net: dst_cache: minor optimization in dst_cache_set_ip6() There is no need to use this_cpu_ptr(dst_cache->cache) twice. Compiler is unable to optimize the second call, because of per-cpu constraints. Signed-off-by: Eric Dumazet Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240507132717.627518-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/core/dst_cache.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/core/dst_cache.c b/net/core/dst_cache.c index f9df84a6c4b2..6a0482e676d3 100644 --- a/net/core/dst_cache.c +++ b/net/core/dst_cache.c @@ -112,7 +112,7 @@ void dst_cache_set_ip6(struct dst_cache *dst_cache, struct dst_entry *dst, return; idst = this_cpu_ptr(dst_cache->cache); - dst_cache_per_cpu_dst_set(this_cpu_ptr(dst_cache->cache), dst, + dst_cache_per_cpu_dst_set(idst, dst, rt6_get_cookie(dst_rt6_info(dst))); idst->in6_saddr = *saddr; } -- cgit v1.2.3 From d101291b2681e5ab938554e3e323f7a7ee33e3aa Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 May 2024 16:31:45 +0000 Subject: ipv6: fib6_rules: avoid possible NULL dereference in fib6_rule_action() syzbot is able to trigger the following crash [1], caused by unsafe ip6_dst_idev() use. Indeed ip6_dst_idev() can return NULL, and must always be checked. [1] Oops: general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN PTI KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] CPU: 0 PID: 31648 Comm: syz-executor.0 Not tainted 6.9.0-rc4-next-20240417-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 RIP: 0010:__fib6_rule_action net/ipv6/fib6_rules.c:237 [inline] RIP: 0010:fib6_rule_action+0x241/0x7b0 net/ipv6/fib6_rules.c:267 Code: 02 00 00 49 8d 9f d8 00 00 00 48 89 d8 48 c1 e8 03 42 80 3c 20 00 74 08 48 89 df e8 f9 32 bf f7 48 8b 1b 48 89 d8 48 c1 e8 03 <42> 80 3c 20 00 74 08 48 89 df e8 e0 32 bf f7 4c 8b 03 48 89 ef 4c RSP: 0018:ffffc9000fc1f2f0 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 1a772f98c8186700 RDX: 0000000000000003 RSI: ffffffff8bcac4e0 RDI: ffffffff8c1f9760 RBP: ffff8880673fb980 R08: ffffffff8fac15ef R09: 1ffffffff1f582bd R10: dffffc0000000000 R11: fffffbfff1f582be R12: dffffc0000000000 R13: 0000000000000080 R14: ffff888076509000 R15: ffff88807a029a00 FS: 00007f55e82ca6c0(0000) GS:ffff8880b9400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000001b31d23000 CR3: 0000000022b66000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: fib_rules_lookup+0x62c/0xdb0 net/core/fib_rules.c:317 fib6_rule_lookup+0x1fd/0x790 net/ipv6/fib6_rules.c:108 ip6_route_output_flags_noref net/ipv6/route.c:2637 [inline] ip6_route_output_flags+0x38e/0x610 net/ipv6/route.c:2649 ip6_route_output include/net/ip6_route.h:93 [inline] ip6_dst_lookup_tail+0x189/0x11a0 net/ipv6/ip6_output.c:1120 ip6_dst_lookup_flow+0xb9/0x180 net/ipv6/ip6_output.c:1250 sctp_v6_get_dst+0x792/0x1e20 net/sctp/ipv6.c:326 sctp_transport_route+0x12c/0x2e0 net/sctp/transport.c:455 sctp_assoc_add_peer+0x614/0x15c0 net/sctp/associola.c:662 sctp_connect_new_asoc+0x31d/0x6c0 net/sctp/socket.c:1099 __sctp_connect+0x66d/0xe30 net/sctp/socket.c:1197 sctp_connect net/sctp/socket.c:4819 [inline] sctp_inet_connect+0x149/0x1f0 net/sctp/socket.c:4834 __sys_connect_file net/socket.c:2048 [inline] __sys_connect+0x2df/0x310 net/socket.c:2065 __do_sys_connect net/socket.c:2075 [inline] __se_sys_connect net/socket.c:2072 [inline] __x64_sys_connect+0x7a/0x90 net/socket.c:2072 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 5e5f3f0f8013 ("[IPV6] ADDRCONF: Convert ipv6_get_saddr() to ipv6_dev_get_saddr().") Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240507163145.835254-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/fib6_rules.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 52c04f0ac498..9e254de7462f 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -233,8 +233,12 @@ static int __fib6_rule_action(struct fib_rule *rule, struct flowi *flp, rt = pol_lookup_func(lookup, net, table, flp6, arg->lookup_data, flags); if (rt != net->ipv6.ip6_null_entry) { + struct inet6_dev *idev = ip6_dst_idev(&rt->dst); + + if (!idev) + goto again; err = fib6_rule_saddr(net, rule, flags, flp6, - ip6_dst_idev(&rt->dst)->dev); + idev->dev); if (err == -EAGAIN) goto again; -- cgit v1.2.3 From 58a4ff5d77b187086eb12d41d613749420947f19 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 May 2024 12:17:48 +0000 Subject: phonet: no longer hold RTNL in route_dumpit() route_dumpit() already relies on RCU, RTNL is not needed. Also change return value at the end of a dump. This allows NLMSG_DONE to be appended to the current skb at the end of a dump, saving a couple of recvmsg() system calls. Signed-off-by: Eric Dumazet Cc: Remi Denis-Courmont Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240507121748.416287-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/phonet/pn_netlink.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c index 59aebe296890..92245fdfa846 100644 --- a/net/phonet/pn_netlink.c +++ b/net/phonet/pn_netlink.c @@ -178,7 +178,7 @@ static int fill_route(struct sk_buff *skb, struct net_device *dev, u8 dst, rtm->rtm_type = RTN_UNICAST; rtm->rtm_flags = 0; if (nla_put_u8(skb, RTA_DST, dst) || - nla_put_u32(skb, RTA_OIF, dev->ifindex)) + nla_put_u32(skb, RTA_OIF, READ_ONCE(dev->ifindex))) goto nla_put_failure; nlmsg_end(skb, nlh); return 0; @@ -263,6 +263,7 @@ static int route_doit(struct sk_buff *skb, struct nlmsghdr *nlh, static int route_dumpit(struct sk_buff *skb, struct netlink_callback *cb) { struct net *net = sock_net(skb->sk); + int err = 0; u8 addr; rcu_read_lock(); @@ -272,16 +273,16 @@ static int route_dumpit(struct sk_buff *skb, struct netlink_callback *cb) if (!dev) continue; - if (fill_route(skb, dev, addr << 2, NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, RTM_NEWROUTE) < 0) - goto out; + err = fill_route(skb, dev, addr << 2, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, RTM_NEWROUTE); + if (err < 0) + break; } - -out: rcu_read_unlock(); cb->args[0] = addr; - return skb->len; + return err; } int __init phonet_netlink_register(void) @@ -301,6 +302,6 @@ int __init phonet_netlink_register(void) rtnl_register_module(THIS_MODULE, PF_PHONET, RTM_DELROUTE, route_doit, NULL, 0); rtnl_register_module(THIS_MODULE, PF_PHONET, RTM_GETROUTE, - NULL, route_dumpit, 0); + NULL, route_dumpit, RTNL_FLAG_DUMP_UNLOCKED); return 0; } -- cgit v1.2.3 From 4893b8b3ef8db2b182d1a1bebf6c7acf91405000 Mon Sep 17 00:00:00 2001 From: Lukasz Majewski Date: Tue, 7 May 2024 13:12:14 +0200 Subject: hsr: Simplify code for announcing HSR nodes timer setup Up till now the code to start HSR announce timer, which triggers sending supervisory frames, was assuming that hsr_netdev_notify() would be called at least twice for hsrX interface. This was required to have different values for old and current values of network device's operstate. This is problematic for a case where hsrX interface is already in the operational state when hsr_netdev_notify() is called, so timer is not configured to trigger and as a result the hsrX is not sending supervisory frames to HSR ring. This error has been discovered when hsr_ping.sh script was run. To be more specific - for the hsr1 and hsr2 the hsr_netdev_notify() was called at least twice with different IF_OPER_{LOWERDOWN|DOWN|UP} states assigned in hsr_check_carrier_and_operstate(hsr). As a result there was no issue with sending supervisory frames. However, with hsr3, the notify function was called only once with operstate set to IF_OPER_UP and timer responsible for triggering supervisory frames was not fired. The solution is to use netif_oper_up() and netif_running() helper functions to assess if network hsrX device is up. Only then, when the timer is not already pending, it is started. Otherwise it is deactivated. Fixes: f421436a591d ("net/hsr: Add support for the High-availability Seamless Redundancy protocol (HSRv0)") Signed-off-by: Lukasz Majewski Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240507111214.3519800-1-lukma@denx.de Signed-off-by: Jakub Kicinski --- net/hsr/hsr_device.c | 27 ++++++++++++--------------- 1 file changed, 12 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/hsr/hsr_device.c b/net/hsr/hsr_device.c index e9d45133d641..5afc450d0855 100644 --- a/net/hsr/hsr_device.c +++ b/net/hsr/hsr_device.c @@ -61,39 +61,36 @@ static bool hsr_check_carrier(struct hsr_port *master) return false; } -static void hsr_check_announce(struct net_device *hsr_dev, - unsigned char old_operstate) +static void hsr_check_announce(struct net_device *hsr_dev) { struct hsr_priv *hsr; hsr = netdev_priv(hsr_dev); - - if (READ_ONCE(hsr_dev->operstate) == IF_OPER_UP && old_operstate != IF_OPER_UP) { - /* Went up */ - hsr->announce_count = 0; - mod_timer(&hsr->announce_timer, - jiffies + msecs_to_jiffies(HSR_ANNOUNCE_INTERVAL)); + if (netif_running(hsr_dev) && netif_oper_up(hsr_dev)) { + /* Enable announce timer and start sending supervisory frames */ + if (!timer_pending(&hsr->announce_timer)) { + hsr->announce_count = 0; + mod_timer(&hsr->announce_timer, jiffies + + msecs_to_jiffies(HSR_ANNOUNCE_INTERVAL)); + } + } else { + /* Deactivate the announce timer */ + timer_delete(&hsr->announce_timer); } - - if (READ_ONCE(hsr_dev->operstate) != IF_OPER_UP && old_operstate == IF_OPER_UP) - /* Went down */ - del_timer(&hsr->announce_timer); } void hsr_check_carrier_and_operstate(struct hsr_priv *hsr) { struct hsr_port *master; - unsigned char old_operstate; bool has_carrier; master = hsr_port_get_hsr(hsr, HSR_PT_MASTER); /* netif_stacked_transfer_operstate() cannot be used here since * it doesn't set IF_OPER_LOWERLAYERDOWN (?) */ - old_operstate = READ_ONCE(master->dev->operstate); has_carrier = hsr_check_carrier(master); hsr_set_operstate(master, has_carrier); - hsr_check_announce(master->dev, old_operstate); + hsr_check_announce(master->dev); } int hsr_get_max_mtu(struct hsr_priv *hsr) -- cgit v1.2.3 From 4db783d68b9b39a411a96096c10828ff5dfada7a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 May 2024 16:18:42 +0000 Subject: ipv6: prevent NULL dereference in ip6_output() According to syzbot, there is a chance that ip6_dst_idev() returns NULL in ip6_output(). Most places in IPv6 stack deal with a NULL idev just fine, but not here. syzbot reported: general protection fault, probably for non-canonical address 0xdffffc00000000bc: 0000 [#1] PREEMPT SMP KASAN PTI KASAN: null-ptr-deref in range [0x00000000000005e0-0x00000000000005e7] CPU: 0 PID: 9775 Comm: syz-executor.4 Not tainted 6.9.0-rc5-syzkaller-00157-g6a30653b604a #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 RIP: 0010:ip6_output+0x231/0x3f0 net/ipv6/ip6_output.c:237 Code: 3c 1e 00 49 89 df 74 08 4c 89 ef e8 19 58 db f7 48 8b 44 24 20 49 89 45 00 49 89 c5 48 8d 9d e0 05 00 00 48 89 d8 48 c1 e8 03 <42> 0f b6 04 38 84 c0 4c 8b 74 24 28 0f 85 61 01 00 00 8b 1b 31 ff RSP: 0018:ffffc9000927f0d8 EFLAGS: 00010202 RAX: 00000000000000bc RBX: 00000000000005e0 RCX: 0000000000040000 RDX: ffffc900131f9000 RSI: 0000000000004f47 RDI: 0000000000004f48 RBP: 0000000000000000 R08: ffffffff8a1f0b9a R09: 1ffffffff1f51fad R10: dffffc0000000000 R11: fffffbfff1f51fae R12: ffff8880293ec8c0 R13: ffff88805d7fc000 R14: 1ffff1100527d91a R15: dffffc0000000000 FS: 00007f135c6856c0(0000) GS:ffff8880b9400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020000080 CR3: 0000000064096000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: NF_HOOK include/linux/netfilter.h:314 [inline] ip6_xmit+0xefe/0x17f0 net/ipv6/ip6_output.c:358 sctp_v6_xmit+0x9f2/0x13f0 net/sctp/ipv6.c:248 sctp_packet_transmit+0x26ad/0x2ca0 net/sctp/output.c:653 sctp_packet_singleton+0x22c/0x320 net/sctp/outqueue.c:783 sctp_outq_flush_ctrl net/sctp/outqueue.c:914 [inline] sctp_outq_flush+0x6d5/0x3e20 net/sctp/outqueue.c:1212 sctp_side_effects net/sctp/sm_sideeffect.c:1198 [inline] sctp_do_sm+0x59cc/0x60c0 net/sctp/sm_sideeffect.c:1169 sctp_primitive_ASSOCIATE+0x95/0xc0 net/sctp/primitive.c:73 __sctp_connect+0x9cd/0xe30 net/sctp/socket.c:1234 sctp_connect net/sctp/socket.c:4819 [inline] sctp_inet_connect+0x149/0x1f0 net/sctp/socket.c:4834 __sys_connect_file net/socket.c:2048 [inline] __sys_connect+0x2df/0x310 net/socket.c:2065 __do_sys_connect net/socket.c:2075 [inline] __se_sys_connect net/socket.c:2072 [inline] __x64_sys_connect+0x7a/0x90 net/socket.c:2072 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 778d80be5269 ("ipv6: Add disable_ipv6 sysctl to disable IPv6 operaion on specific interface.") Reported-by: syzbot Signed-off-by: Eric Dumazet Reviewed-by: Larysa Zaremba Link: https://lore.kernel.org/r/20240507161842.773961-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index fa2937732665..97b0788b31ba 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -234,7 +234,7 @@ int ip6_output(struct net *net, struct sock *sk, struct sk_buff *skb) skb->protocol = htons(ETH_P_IPV6); skb->dev = dev; - if (unlikely(READ_ONCE(idev->cnf.disable_ipv6))) { + if (unlikely(!idev || READ_ONCE(idev->cnf.disable_ipv6))) { IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); kfree_skb_reason(skb, SKB_DROP_REASON_IPV6DISABLED); return 0; -- cgit v1.2.3 From 2ddc0dd7fec86ee53b8928a5cca5fbddd4fc7c06 Mon Sep 17 00:00:00 2001 From: Wen Gu Date: Tue, 7 May 2024 20:53:31 +0800 Subject: net/smc: fix neighbour and rtable leak in smc_ib_find_route() In smc_ib_find_route(), the neighbour found by neigh_lookup() and rtable resolved by ip_route_output_flow() are not released or put before return. It may cause the refcount leak, so fix it. Link: https://lore.kernel.org/r/20240506015439.108739-1-guwen@linux.alibaba.com Fixes: e5c4744cfb59 ("net/smc: add SMC-Rv2 connection establishment") Signed-off-by: Wen Gu Link: https://lore.kernel.org/r/20240507125331.2808-1-guwen@linux.alibaba.com Signed-off-by: Paolo Abeni --- net/smc/smc_ib.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/smc/smc_ib.c b/net/smc/smc_ib.c index 97704a9e84c7..9297dc20bfe2 100644 --- a/net/smc/smc_ib.c +++ b/net/smc/smc_ib.c @@ -209,13 +209,18 @@ int smc_ib_find_route(struct net *net, __be32 saddr, __be32 daddr, if (IS_ERR(rt)) goto out; if (rt->rt_uses_gateway && rt->rt_gw_family != AF_INET) - goto out; - neigh = rt->dst.ops->neigh_lookup(&rt->dst, NULL, &fl4.daddr); - if (neigh) { - memcpy(nexthop_mac, neigh->ha, ETH_ALEN); - *uses_gateway = rt->rt_uses_gateway; - return 0; - } + goto out_rt; + neigh = dst_neigh_lookup(&rt->dst, &fl4.daddr); + if (!neigh) + goto out_rt; + memcpy(nexthop_mac, neigh->ha, ETH_ALEN); + *uses_gateway = rt->rt_uses_gateway; + neigh_release(neigh); + ip_rt_put(rt); + return 0; + +out_rt: + ip_rt_put(rt); out: return -ENOENT; } -- cgit v1.2.3 From bafa6b4d95d97877baa61883ff90f7e374427fae Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 7 May 2024 09:10:41 -0400 Subject: SUNRPC: Fix gss_free_in_token_pages() Dan Carpenter says: > Commit 5866efa8cbfb ("SUNRPC: Fix svcauth_gss_proxy_init()") from Oct > 24, 2019 (linux-next), leads to the following Smatch static checker > warning: > > net/sunrpc/auth_gss/svcauth_gss.c:1039 gss_free_in_token_pages() > warn: iterator 'i' not incremented > > net/sunrpc/auth_gss/svcauth_gss.c > 1034 static void gss_free_in_token_pages(struct gssp_in_token *in_token) > 1035 { > 1036 u32 inlen; > 1037 int i; > 1038 > --> 1039 i = 0; > 1040 inlen = in_token->page_len; > 1041 while (inlen) { > 1042 if (in_token->pages[i]) > 1043 put_page(in_token->pages[i]); > ^ > This puts page zero over and over. > > 1044 inlen -= inlen > PAGE_SIZE ? PAGE_SIZE : inlen; > 1045 } > 1046 > 1047 kfree(in_token->pages); > 1048 in_token->pages = NULL; > 1049 } Based on the way that the ->pages[] array is constructed in gss_read_proxy_verf(), we know that once the loop encounters a NULL page pointer, the remaining array elements must also be NULL. Reported-by: Dan Carpenter Suggested-by: Trond Myklebust Fixes: 5866efa8cbfb ("SUNRPC: Fix svcauth_gss_proxy_init()") Signed-off-by: Chuck Lever --- net/sunrpc/auth_gss/svcauth_gss.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index 24de94184700..96ab50eda9c2 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -1033,17 +1033,11 @@ null_verifier: static void gss_free_in_token_pages(struct gssp_in_token *in_token) { - u32 inlen; int i; i = 0; - inlen = in_token->page_len; - while (inlen) { - if (in_token->pages[i]) - put_page(in_token->pages[i]); - inlen -= inlen > PAGE_SIZE ? PAGE_SIZE : inlen; - } - + while (in_token->pages[i]) + put_page(in_token->pages[i++]); kfree(in_token->pages); in_token->pages = NULL; } -- cgit v1.2.3 From 628bc3e5a1beae395b5b515998396c60559ed3a9 Mon Sep 17 00:00:00 2001 From: Samuel Thibault Date: Mon, 6 May 2024 23:53:35 +0200 Subject: l2tp: Support several sockets with same IP/port quadruple Some l2tp providers will use 1701 as origin port and open several tunnels for the same origin and target. On the Linux side, this may mean opening several sockets, but then trafic will go to only one of them, losing the trafic for the tunnel of the other socket (or leaving it up to userland, consuming a lot of cpu%). This can also happen when the l2tp provider uses a cluster, and load-balancing happens to migrate from one origin IP to another one, for which a socket was already established. Managing reassigning tunnels from one socket to another would be very hairy for userland. Lastly, as documented in l2tpconfig(1), as client it may be necessary to use 1701 as origin port for odd firewalls reasons, which could prevent from establishing several tunnels to a l2tp server, for the same reason: trafic would get only on one of the two sockets. With the V2 protocol it is however easy to route trafic to the proper tunnel, by looking up the tunnel number in the network namespace. This fixes the three cases altogether. Signed-off-by: Samuel Thibault Link: https://lore.kernel.org/r/20240506215336.1470009-1-samuel.thibault@ens-lyon.org Signed-off-by: Paolo Abeni --- net/l2tp/l2tp_core.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 8d21ff25f160..2ab45e3f48bf 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -794,6 +794,7 @@ static void l2tp_session_queue_purge(struct l2tp_session *session) static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) { struct l2tp_session *session = NULL; + struct l2tp_tunnel *orig_tunnel = tunnel; unsigned char *ptr, *optr; u16 hdrflags; u32 tunnel_id, session_id; @@ -845,6 +846,20 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) /* Extract tunnel and session ID */ tunnel_id = ntohs(*(__be16 *)ptr); ptr += 2; + + if (tunnel_id != tunnel->tunnel_id) { + /* We are receiving trafic for another tunnel, probably + * because we have several tunnels between the same + * IP/port quadruple, look it up. + */ + struct l2tp_tunnel *alt_tunnel; + + alt_tunnel = l2tp_tunnel_get(tunnel->l2tp_net, tunnel_id); + if (!alt_tunnel || alt_tunnel->version != L2TP_HDR_VER_2) + goto pass; + tunnel = alt_tunnel; + } + session_id = ntohs(*(__be16 *)ptr); ptr += 2; } else { @@ -875,6 +890,9 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) l2tp_recv_common(session, skb, ptr, optr, hdrflags, length); l2tp_session_dec_refcount(session); + if (tunnel != orig_tunnel) + l2tp_tunnel_dec_refcount(tunnel); + return 0; invalid: @@ -884,6 +902,9 @@ pass: /* Put UDP header back */ __skb_push(skb, sizeof(struct udphdr)); + if (tunnel != orig_tunnel) + l2tp_tunnel_dec_refcount(tunnel); + return 1; } -- cgit v1.2.3 From b1992c3772e69a6fd0e3fc81cd4d2820c8b6eca0 Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Sat, 27 Apr 2024 23:55:02 +0900 Subject: kbuild: use $(src) instead of $(srctree)/$(src) for source directory Kbuild conventionally uses $(obj)/ for generated files, and $(src)/ for checked-in source files. It is merely a convention without any functional difference. In fact, $(obj) and $(src) are exactly the same, as defined in scripts/Makefile.build: src := $(obj) When the kernel is built in a separate output directory, $(src) does not accurately reflect the source directory location. While Kbuild resolves this discrepancy by specifying VPATH=$(srctree) to search for source files, it does not cover all cases. For example, when adding a header search path for local headers, -I$(srctree)/$(src) is typically passed to the compiler. This introduces inconsistency between upstream and downstream Makefiles because $(src) is used instead of $(srctree)/$(src) for the latter. To address this inconsistency, this commit changes the semantics of $(src) so that it always points to the directory in the source tree. Going forward, the variables used in Makefiles will have the following meanings: $(obj) - directory in the object tree $(src) - directory in the source tree (changed by this commit) $(objtree) - the top of the kernel object tree $(srctree) - the top of the kernel source tree Consequently, $(srctree)/$(src) in upstream Makefiles need to be replaced with $(src). Signed-off-by: Masahiro Yamada Reviewed-by: Nicolas Schier --- Documentation/Makefile | 8 ++++---- Documentation/devicetree/bindings/Makefile | 6 +++--- Documentation/kbuild/makefiles.rst | 12 ++++++------ Makefile | 7 +++++++ arch/arc/boot/dts/Makefile | 3 +-- arch/arm/Kbuild | 2 +- arch/arm/boot/Makefile | 3 +-- arch/arm/mach-s3c/Makefile | 2 +- arch/arm/plat-orion/Makefile | 2 +- arch/arm/tools/Makefile | 2 +- arch/arm64/kernel/vdso/Makefile | 2 +- arch/arm64/kvm/Makefile | 4 ++-- arch/arm64/kvm/hyp/Makefile | 2 +- arch/csky/boot/dts/Makefile | 4 +--- arch/csky/kernel/vdso/Makefile | 2 +- arch/loongarch/kvm/Makefile | 2 +- arch/loongarch/vdso/Makefile | 2 +- arch/mips/kernel/syscalls/Makefile | 2 +- arch/mips/vdso/Makefile | 4 ++-- arch/nios2/boot/dts/Makefile | 3 +-- arch/parisc/kernel/vdso32/Makefile | 2 +- arch/parisc/kernel/vdso64/Makefile | 2 +- arch/powerpc/boot/Makefile | 6 +++--- arch/powerpc/boot/dts/Makefile | 3 +-- arch/powerpc/boot/dts/fsl/Makefile | 3 +-- arch/powerpc/kernel/vdso/Makefile | 4 ++-- arch/riscv/kernel/compat_vdso/Makefile | 2 +- arch/riscv/kernel/vdso/Makefile | 2 +- arch/riscv/kvm/Makefile | 2 +- arch/s390/kernel/syscalls/Makefile | 4 ++-- arch/s390/kernel/vdso32/Makefile | 2 +- arch/s390/kernel/vdso64/Makefile | 2 +- arch/sparc/vdso/Makefile | 2 +- arch/um/kernel/Makefile | 2 +- arch/x86/boot/Makefile | 2 +- arch/x86/entry/vdso/Makefile | 2 +- arch/x86/kernel/Makefile | 2 +- arch/x86/kernel/cpu/Makefile | 2 +- arch/x86/mm/Makefile | 2 +- arch/x86/um/vdso/Makefile | 2 +- arch/xtensa/boot/dts/Makefile | 3 +-- certs/Makefile | 4 ++-- drivers/Makefile | 5 ----- drivers/crypto/intel/qat/qat_420xx/Makefile | 2 +- drivers/crypto/intel/qat/qat_4xxx/Makefile | 2 +- drivers/crypto/intel/qat/qat_c3xxx/Makefile | 2 +- drivers/crypto/intel/qat/qat_c3xxxvf/Makefile | 2 +- drivers/crypto/intel/qat/qat_c62x/Makefile | 2 +- drivers/crypto/intel/qat/qat_c62xvf/Makefile | 2 +- drivers/crypto/intel/qat/qat_dh895xcc/Makefile | 2 +- drivers/crypto/intel/qat/qat_dh895xccvf/Makefile | 2 +- drivers/gpu/drm/amd/amdgpu/Makefile | 2 +- drivers/gpu/drm/arm/display/komeda/Makefile | 4 ++-- drivers/gpu/drm/i915/Makefile | 4 ++-- drivers/gpu/drm/imagination/Makefile | 2 +- drivers/gpu/drm/msm/Makefile | 8 ++++---- drivers/gpu/drm/nouveau/Kbuild | 10 ++++------ drivers/gpu/drm/xe/Makefile | 10 +++++----- drivers/hid/amd-sfh-hid/Makefile | 2 +- drivers/hid/intel-ish-hid/Makefile | 2 +- drivers/md/dm-vdo/Makefile | 2 +- drivers/net/ethernet/aquantia/atlantic/Makefile | 2 +- drivers/net/ethernet/chelsio/libcxgb/Makefile | 2 +- drivers/net/ethernet/fungible/funeth/Makefile | 2 +- drivers/net/ethernet/hisilicon/hns3/Makefile | 2 +- drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile | 4 ++-- .../net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile | 6 +++--- .../net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile | 6 +++--- .../net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile | 6 +++--- drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile | 6 +++--- drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/dvm/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/mei/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/mvm/Makefile | 2 +- drivers/net/wireless/intel/iwlwifi/tests/Makefile | 2 +- drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile | 2 +- drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile | 2 +- drivers/scsi/aic7xxx/Makefile | 12 ++++++------ drivers/staging/rtl8723bs/Makefile | 2 +- fs/iomap/Makefile | 2 +- fs/unicode/Makefile | 14 +++++++------- fs/xfs/Makefile | 4 ++-- init/Makefile | 2 +- lib/Makefile | 6 +++--- lib/raid6/Makefile | 2 +- net/wireless/Makefile | 2 +- rust/Makefile | 6 +++--- samples/bpf/Makefile | 2 +- samples/hid/Makefile | 2 +- scripts/Kbuild.include | 3 +-- scripts/Makefile.asm-generic | 6 +++--- scripts/Makefile.build | 2 +- scripts/Makefile.clean | 2 +- scripts/Makefile.lib | 10 +++++----- scripts/Makefile.modpost | 2 +- scripts/dtc/Makefile | 6 +++--- scripts/gdb/linux/Makefile | 2 +- scripts/genksyms/Makefile | 4 ++-- scripts/kconfig/Makefile | 8 ++++---- security/tomoyo/Makefile | 2 +- usr/Makefile | 2 +- usr/include/Makefile | 2 +- 102 files changed, 173 insertions(+), 182 deletions(-) (limited to 'net') diff --git a/Documentation/Makefile b/Documentation/Makefile index b68f8c816897..66d15fff2089 100644 --- a/Documentation/Makefile +++ b/Documentation/Makefile @@ -76,22 +76,22 @@ loop_cmd = $(echo-cmd) $(cmd_$(1)) || exit; # * dest folder relative to $(BUILDDIR) and # * cache folder relative to $(BUILDDIR)/.doctrees # $4 dest subfolder e.g. "man" for man pages at userspace-api/media/man -# $5 reST source folder relative to $(srctree)/$(src), +# $5 reST source folder relative to $(src), # e.g. "userspace-api/media" for the linux-tv book-set at ./Documentation/userspace-api/media quiet_cmd_sphinx = SPHINX $@ --> file://$(abspath $(BUILDDIR)/$3/$4) cmd_sphinx = $(MAKE) BUILDDIR=$(abspath $(BUILDDIR)) $(build)=Documentation/userspace-api/media $2 && \ PYTHONDONTWRITEBYTECODE=1 \ - BUILDDIR=$(abspath $(BUILDDIR)) SPHINX_CONF=$(abspath $(srctree)/$(src)/$5/$(SPHINX_CONF)) \ + BUILDDIR=$(abspath $(BUILDDIR)) SPHINX_CONF=$(abspath $(src)/$5/$(SPHINX_CONF)) \ $(PYTHON3) $(srctree)/scripts/jobserver-exec \ $(CONFIG_SHELL) $(srctree)/Documentation/sphinx/parallel-wrapper.sh \ $(SPHINXBUILD) \ -b $2 \ - -c $(abspath $(srctree)/$(src)) \ + -c $(abspath $(src)) \ -d $(abspath $(BUILDDIR)/.doctrees/$3) \ -D version=$(KERNELVERSION) -D release=$(KERNELRELEASE) \ $(ALLSPHINXOPTS) \ - $(abspath $(srctree)/$(src)/$5) \ + $(abspath $(src)/$5) \ $(abspath $(BUILDDIR)/$3/$4) && \ if [ "x$(DOCS_CSS)" != "x" ]; then \ cp $(if $(patsubst /%,,$(DOCS_CSS)),$(abspath $(srctree)/$(DOCS_CSS)),$(DOCS_CSS)) $(BUILDDIR)/$3/_static/; \ diff --git a/Documentation/devicetree/bindings/Makefile b/Documentation/devicetree/bindings/Makefile index 8cdda477987f..bf7d64632e20 100644 --- a/Documentation/devicetree/bindings/Makefile +++ b/Documentation/devicetree/bindings/Makefile @@ -25,7 +25,7 @@ quiet_cmd_extract_ex = DTEX $@ $(obj)/%.example.dts: $(src)/%.yaml check_dtschema_version FORCE $(call if_changed,extract_ex) -find_all_cmd = find $(srctree)/$(src) \( -name '*.yaml' ! \ +find_all_cmd = find $(src) \( -name '*.yaml' ! \ -name 'processed-schema*' \) find_cmd = $(find_all_cmd) | \ @@ -37,12 +37,12 @@ CHK_DT_EXAMPLES := $(patsubst $(srctree)/%.yaml,%.example.dtb, $(shell $(find_cm quiet_cmd_yamllint = LINT $(src) cmd_yamllint = ($(find_cmd) | \ xargs -n200 -P$$(nproc) \ - $(DT_SCHEMA_LINT) -f parsable -c $(srctree)/$(src)/.yamllint >&2) \ + $(DT_SCHEMA_LINT) -f parsable -c $(src)/.yamllint >&2) \ && touch $@ || true quiet_cmd_chk_bindings = CHKDT $(src) cmd_chk_bindings = ($(find_cmd) | \ - xargs -n200 -P$$(nproc) $(DT_DOC_CHECKER) -u $(srctree)/$(src)) \ + xargs -n200 -P$$(nproc) $(DT_DOC_CHECKER) -u $(src)) \ && touch $@ || true quiet_cmd_mk_schema = SCHEMA $@ diff --git a/Documentation/kbuild/makefiles.rst b/Documentation/kbuild/makefiles.rst index ad118b7a1806..991ce6081e35 100644 --- a/Documentation/kbuild/makefiles.rst +++ b/Documentation/kbuild/makefiles.rst @@ -346,7 +346,7 @@ ccflags-y, asflags-y and ldflags-y Example:: #arch/cris/boot/compressed/Makefile - ldflags-y += -T $(srctree)/$(src)/decompress_$(arch-y).lds + ldflags-y += -T $(src)/decompress_$(arch-y).lds subdir-ccflags-y, subdir-asflags-y The two flags listed above are similar to ccflags-y and asflags-y. @@ -426,14 +426,14 @@ path to prerequisite files and target files. Two variables are used when defining custom rules: $(src) - $(src) is a relative path which points to the directory - where the Makefile is located. Always use $(src) when + $(src) is the directory where the Makefile is located. Always use $(src) when referring to files located in the src tree. $(obj) - $(obj) is a relative path which points to the directory - where the target is saved. Always use $(obj) when - referring to generated files. + $(obj) is the directory where the target is saved. Always use $(obj) when + referring to generated files. Use $(obj) for pattern rules that need to work + for both generated files and real sources (VPATH will help to find the + prerequisites not only in the object tree but also in the source tree). Example:: diff --git a/Makefile b/Makefile index 12e1a792b8de..c97c8d96061c 100644 --- a/Makefile +++ b/Makefile @@ -263,7 +263,14 @@ srctree := $(abs_srctree) endif objtree := . + +VPATH := + +ifeq ($(KBUILD_EXTMOD),) +ifdef building_out_of_srctree VPATH := $(srctree) +endif +endif export building_out_of_srctree srctree objtree VPATH diff --git a/arch/arc/boot/dts/Makefile b/arch/arc/boot/dts/Makefile index 4237aa5de3a3..48704dfdf75c 100644 --- a/arch/arc/boot/dts/Makefile +++ b/arch/arc/boot/dts/Makefile @@ -10,8 +10,7 @@ obj-y += $(builtindtb-y).dtb.o dtb-y := $(builtindtb-y).dtb # for CONFIG_OF_ALL_DTBS test -dtstree := $(srctree)/$(src) -dtb- := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb- := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) # board-specific dtc flags DTC_FLAGS_hsdk += --pad 20 diff --git a/arch/arm/Kbuild b/arch/arm/Kbuild index b506622e7e23..69de6b6243c7 100644 --- a/arch/arm/Kbuild +++ b/arch/arm/Kbuild @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only obj-$(CONFIG_FPE_NWFPE) += nwfpe/ # Put arch/arm/fastfpe/ to use this. -obj-$(CONFIG_FPE_FASTFPE) += $(patsubst $(srctree)/$(src)/%,%,$(wildcard $(srctree)/$(src)/fastfpe/)) +obj-$(CONFIG_FPE_FASTFPE) += $(patsubst $(src)/%,%,$(wildcard $(src)/fastfpe/)) obj-$(CONFIG_VFP) += vfp/ obj-$(CONFIG_XEN) += xen/ obj-$(CONFIG_VDSO) += vdso/ diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index abd6a2889fd0..ba9b9a802469 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -25,8 +25,7 @@ targets := Image zImage xipImage bootpImage uImage ifeq ($(CONFIG_XIP_KERNEL),y) -cmd_deflate_xip_data = $(CONFIG_SHELL) -c \ - '$(srctree)/$(src)/deflate_xip_data.sh $< $@' +cmd_deflate_xip_data = $(CONFIG_SHELL) -c '$(src)/deflate_xip_data.sh $< $@' ifeq ($(CONFIG_XIP_DEFLATED_DATA),y) quiet_cmd_mkxip = XIPZ $@ diff --git a/arch/arm/mach-s3c/Makefile b/arch/arm/mach-s3c/Makefile index 713827bef831..988c49672715 100644 --- a/arch/arm/mach-s3c/Makefile +++ b/arch/arm/mach-s3c/Makefile @@ -2,7 +2,7 @@ # # Copyright 2009 Simtec Electronics -include $(srctree)/$(src)/Makefile.s3c64xx +include $(src)/Makefile.s3c64xx # Objects we always build independent of SoC choice diff --git a/arch/arm/plat-orion/Makefile b/arch/arm/plat-orion/Makefile index 830b0be038c6..e8c7580df8ca 100644 --- a/arch/arm/plat-orion/Makefile +++ b/arch/arm/plat-orion/Makefile @@ -2,7 +2,7 @@ # # Makefile for the linux kernel. # -ccflags-y := -I$(srctree)/$(src)/include +ccflags-y := -I$(src)/include orion-gpio-$(CONFIG_GPIOLIB) += gpio.o obj-$(CONFIG_PLAT_ORION_LEGACY) += irq.o pcie.o time.o common.o mpp.o diff --git a/arch/arm/tools/Makefile b/arch/arm/tools/Makefile index 81f13bdf32f2..28b6da8ac5f6 100644 --- a/arch/arm/tools/Makefile +++ b/arch/arm/tools/Makefile @@ -9,7 +9,7 @@ gen := arch/$(ARCH)/include/generated kapi := $(gen)/asm uapi := $(gen)/uapi/asm syshdr := $(srctree)/scripts/syscallhdr.sh -sysnr := $(srctree)/$(src)/syscallnr.sh +sysnr := $(src)/syscallnr.sh systbl := $(srctree)/scripts/syscalltbl.sh syscall := $(src)/syscall.tbl diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile index 8818287f1095..53e86d3bc159 100644 --- a/arch/arm64/kernel/vdso/Makefile +++ b/arch/arm64/kernel/vdso/Makefile @@ -68,7 +68,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile index c0c050e53157..ab474d444cef 100644 --- a/arch/arm64/kvm/Makefile +++ b/arch/arm64/kvm/Makefile @@ -3,7 +3,7 @@ # Makefile for Kernel-based Virtual Machine module # -ccflags-y += -I $(srctree)/$(src) +ccflags-y += -I $(src) include $(srctree)/virt/kvm/Makefile.kvm @@ -30,7 +30,7 @@ define rule_gen_hyp_constants $(call filechk,offsets,__HYP_CONSTANTS_H__) endef -CFLAGS_hyp-constants.o = -I $(srctree)/$(src)/hyp/include +CFLAGS_hyp-constants.o = -I $(src)/hyp/include $(obj)/hyp-constants.s: $(src)/hyp/hyp-constants.c FORCE $(call if_changed_dep,cc_s_c) diff --git a/arch/arm64/kvm/hyp/Makefile b/arch/arm64/kvm/hyp/Makefile index a38dea6186c9..d61e44642f98 100644 --- a/arch/arm64/kvm/hyp/Makefile +++ b/arch/arm64/kvm/hyp/Makefile @@ -3,7 +3,7 @@ # Makefile for Kernel-based Virtual Machine module, HYP part # -incdir := $(srctree)/$(src)/include +incdir := $(src)/include subdir-asflags-y := -I$(incdir) subdir-ccflags-y := -I$(incdir) diff --git a/arch/csky/boot/dts/Makefile b/arch/csky/boot/dts/Makefile index 5f1f55e911ad..aabffc1912e8 100644 --- a/arch/csky/boot/dts/Makefile +++ b/arch/csky/boot/dts/Makefile @@ -1,4 +1,2 @@ # SPDX-License-Identifier: GPL-2.0-only -dtstree := $(srctree)/$(src) - -dtb-y := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-y := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/csky/kernel/vdso/Makefile b/arch/csky/kernel/vdso/Makefile index 24a01c22293e..e79a725f5075 100644 --- a/arch/csky/kernel/vdso/Makefile +++ b/arch/csky/kernel/vdso/Makefile @@ -57,4 +57,4 @@ quiet_cmd_vdsold = VDSOLD $@ # Extracts symbol offsets from the VDSO, converting them into an assembly file # that contains the same symbols at the same offsets. quiet_cmd_so2s = SO2S $@ - cmd_so2s = $(NM) -D $< | $(srctree)/$(src)/so2s.sh > $@ + cmd_so2s = $(NM) -D $< | $(src)/so2s.sh > $@ diff --git a/arch/loongarch/kvm/Makefile b/arch/loongarch/kvm/Makefile index 244467d7792a..b2f4cbe01ae8 100644 --- a/arch/loongarch/kvm/Makefile +++ b/arch/loongarch/kvm/Makefile @@ -3,7 +3,7 @@ # Makefile for LoongArch KVM support # -ccflags-y += -I $(srctree)/$(src) +ccflags-y += -I $(src) include $(srctree)/virt/kvm/Makefile.kvm diff --git a/arch/loongarch/vdso/Makefile b/arch/loongarch/vdso/Makefile index 75c6726382c3..cdfc4c793e2c 100644 --- a/arch/loongarch/vdso/Makefile +++ b/arch/loongarch/vdso/Makefile @@ -52,7 +52,7 @@ quiet_cmd_vdsoas_o_S = AS $@ cmd_vdsoas_o_S = $(CC) $(a_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/mips/kernel/syscalls/Makefile b/arch/mips/kernel/syscalls/Makefile index e6b21de65cca..56f6f093bb88 100644 --- a/arch/mips/kernel/syscalls/Makefile +++ b/arch/mips/kernel/syscalls/Makefile @@ -5,7 +5,7 @@ uapi := arch/$(SRCARCH)/include/generated/uapi/asm $(shell mkdir -p $(uapi) $(kapi)) syshdr := $(srctree)/scripts/syscallhdr.sh -sysnr := $(srctree)/$(src)/syscallnr.sh +sysnr := $(src)/syscallnr.sh systbl := $(srctree)/scripts/syscalltbl.sh quiet_cmd_syshdr = SYSHDR $@ diff --git a/arch/mips/vdso/Makefile b/arch/mips/vdso/Makefile index eb56581f6d73..40b839e91806 100644 --- a/arch/mips/vdso/Makefile +++ b/arch/mips/vdso/Makefile @@ -43,8 +43,8 @@ CFLAGS_vgettimeofday.o = -include $(c-gettimeofday-y) # config-n32-o32-env.c prepares the environment to build a 32bit vDSO # library on a 64bit kernel. # Note: Needs to be included before than the generic library. -CFLAGS_vgettimeofday-o32.o = -include $(srctree)/$(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) -CFLAGS_vgettimeofday-n32.o = -include $(srctree)/$(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) +CFLAGS_vgettimeofday-o32.o = -include $(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) +CFLAGS_vgettimeofday-n32.o = -include $(src)/config-n32-o32-env.c -include $(c-gettimeofday-y) endif CFLAGS_REMOVE_vgettimeofday.o = $(CC_FLAGS_FTRACE) diff --git a/arch/nios2/boot/dts/Makefile b/arch/nios2/boot/dts/Makefile index e9e31bb40df8..1a2e8996bec7 100644 --- a/arch/nios2/boot/dts/Makefile +++ b/arch/nios2/boot/dts/Makefile @@ -2,5 +2,4 @@ obj-y := $(patsubst %.dts,%.dtb.o,$(CONFIG_NIOS2_DTB_SOURCE)) -dtstree := $(srctree)/$(src) -dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/parisc/kernel/vdso32/Makefile b/arch/parisc/kernel/vdso32/Makefile index 60dc708a0f80..1350d50c6306 100644 --- a/arch/parisc/kernel/vdso32/Makefile +++ b/arch/parisc/kernel/vdso32/Makefile @@ -40,7 +40,7 @@ quiet_cmd_vdso32as = VDSO32A $@ cmd_vdso32as = $(CROSS32CC) $(a_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/parisc/kernel/vdso64/Makefile b/arch/parisc/kernel/vdso64/Makefile index 0982f3099ae2..0b1c1cc4c2c7 100644 --- a/arch/parisc/kernel/vdso64/Makefile +++ b/arch/parisc/kernel/vdso64/Makefile @@ -40,7 +40,7 @@ quiet_cmd_vdso64as = VDSO64A $@ cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile index 968aee2025b8..a0b3b5cc14cd 100644 --- a/arch/powerpc/boot/Makefile +++ b/arch/powerpc/boot/Makefile @@ -218,7 +218,7 @@ $(addprefix $(obj)/,$(libfdt) $(libfdtheader)): $(obj)/%: $(srctree)/scripts/dtc $(obj)/empty.c: $(Q)touch $@ -$(obj)/zImage.coff.lds $(obj)/zImage.ps3.lds : $(obj)/%: $(srctree)/$(src)/%.S +$(obj)/zImage.coff.lds $(obj)/zImage.ps3.lds : $(obj)/%: $(src)/%.S $(Q)cp $< $@ clean-files := $(zlib-) $(zlibheader-) $(zliblinuxheader-) \ @@ -252,9 +252,9 @@ targets += $(patsubst $(obj)/%,%,$(obj-boot) wrapper.a) zImage.lds extra-y := $(obj)/wrapper.a $(obj-plat) $(obj)/empty.o \ $(obj)/zImage.lds $(obj)/zImage.coff.lds $(obj)/zImage.ps3.lds -dtstree := $(srctree)/$(src)/dts +dtstree := $(src)/dts -wrapper :=$(srctree)/$(src)/wrapper +wrapper := $(src)/wrapper wrapperbits := $(extra-y) $(addprefix $(obj)/,addnote hack-coff mktree) \ $(wrapper) FORCE diff --git a/arch/powerpc/boot/dts/Makefile b/arch/powerpc/boot/dts/Makefile index fb335d05aae8..0cd0d8558b47 100644 --- a/arch/powerpc/boot/dts/Makefile +++ b/arch/powerpc/boot/dts/Makefile @@ -2,5 +2,4 @@ subdir-y += fsl -dtstree := $(srctree)/$(src) -dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/powerpc/boot/dts/fsl/Makefile b/arch/powerpc/boot/dts/fsl/Makefile index 3bae982641e9..d3ecdf14bc42 100644 --- a/arch/powerpc/boot/dts/fsl/Makefile +++ b/arch/powerpc/boot/dts/fsl/Makefile @@ -1,4 +1,3 @@ # SPDX-License-Identifier: GPL-2.0 -dtstree := $(srctree)/$(src) -dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/arch/powerpc/kernel/vdso/Makefile b/arch/powerpc/kernel/vdso/Makefile index a14eab366993..7d66b6e07993 100644 --- a/arch/powerpc/kernel/vdso/Makefile +++ b/arch/powerpc/kernel/vdso/Makefile @@ -90,10 +90,10 @@ $(obj)/vgettimeofday-64.o: %-64.o: %.c FORCE $(call if_changed_dep,cc_o_c) # Generate VDSO offsets using helper script -gen-vdso32sym := $(srctree)/$(src)/gen_vdso32_offsets.sh +gen-vdso32sym := $(src)/gen_vdso32_offsets.sh quiet_cmd_vdso32sym = VDSO32SYM $@ cmd_vdso32sym = $(NM) $< | $(gen-vdso32sym) | LC_ALL=C sort > $@ -gen-vdso64sym := $(srctree)/$(src)/gen_vdso64_offsets.sh +gen-vdso64sym := $(src)/gen_vdso64_offsets.sh quiet_cmd_vdso64sym = VDSO64SYM $@ cmd_vdso64sym = $(NM) $< | $(gen-vdso64sym) | LC_ALL=C sort > $@ diff --git a/arch/riscv/kernel/compat_vdso/Makefile b/arch/riscv/kernel/compat_vdso/Makefile index 3df4cb788c1f..362208dfa7ee 100644 --- a/arch/riscv/kernel/compat_vdso/Makefile +++ b/arch/riscv/kernel/compat_vdso/Makefile @@ -58,7 +58,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # Generate VDSO offsets using helper script -gen-compat_vdsosym := $(srctree)/$(src)/gen_compat_vdso_offsets.sh +gen-compat_vdsosym := $(src)/gen_compat_vdso_offsets.sh quiet_cmd_compat_vdsosym = VDSOSYM $@ cmd_compat_vdsosym = $(NM) $< | $(gen-compat_vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile index 272c431ac5b9..25f0ee629971 100644 --- a/arch/riscv/kernel/vdso/Makefile +++ b/arch/riscv/kernel/vdso/Makefile @@ -60,7 +60,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE $(call if_changed,objcopy) # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/riscv/kvm/Makefile b/arch/riscv/kvm/Makefile index c9646521f113..c2cacfbc06a0 100644 --- a/arch/riscv/kvm/Makefile +++ b/arch/riscv/kvm/Makefile @@ -3,7 +3,7 @@ # Makefile for RISC-V KVM support # -ccflags-y += -I $(srctree)/$(src) +ccflags-y += -I $(src) include $(srctree)/virt/kvm/Makefile.kvm diff --git a/arch/s390/kernel/syscalls/Makefile b/arch/s390/kernel/syscalls/Makefile index fb85e797946d..1bb78b9468e8 100644 --- a/arch/s390/kernel/syscalls/Makefile +++ b/arch/s390/kernel/syscalls/Makefile @@ -4,8 +4,8 @@ gen := arch/$(ARCH)/include/generated kapi := $(gen)/asm uapi := $(gen)/uapi/asm -syscall := $(srctree)/$(src)/syscall.tbl -systbl := $(srctree)/$(src)/syscalltbl +syscall := $(src)/syscall.tbl +systbl := $(src)/syscalltbl gen-y := $(kapi)/syscall_table.h kapi-hdrs-y := $(kapi)/unistd_nr.h diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile index 70e9949c2612..c263b91adfb1 100644 --- a/arch/s390/kernel/vdso32/Makefile +++ b/arch/s390/kernel/vdso32/Makefile @@ -62,7 +62,7 @@ quiet_cmd_vdso32cc = VDSO32C $@ cmd_vdso32cc = $(CC) $(c_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile index 2b3617b6d162..9566bed7d5b2 100644 --- a/arch/s390/kernel/vdso64/Makefile +++ b/arch/s390/kernel/vdso64/Makefile @@ -72,7 +72,7 @@ quiet_cmd_vdso64cc = VDSO64C $@ cmd_vdso64cc = $(CC) $(c_flags) -c -o $@ $< # Generate VDSO offsets using helper script -gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh +gen-vdsosym := $(src)/gen_vdso_offsets.sh quiet_cmd_vdsosym = VDSOSYM $@ cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ diff --git a/arch/sparc/vdso/Makefile b/arch/sparc/vdso/Makefile index e8aef2c8ae99..0fe134abbcf1 100644 --- a/arch/sparc/vdso/Makefile +++ b/arch/sparc/vdso/Makefile @@ -103,7 +103,7 @@ quiet_cmd_vdso = VDSO $@ cmd_vdso = $(LD) -nostdlib -o $@ \ $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -T $(filter %.lds,$^) $(filter %.o,$^) && \ - sh $(srctree)/$(src)/checkundef.sh '$(OBJDUMP)' '$@' + sh $(src)/checkundef.sh '$(OBJDUMP)' '$@' VDSO_LDFLAGS = -shared --hash-style=both --build-id=sha1 -Bsymbolic GCOV_PROFILE := n diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 811188be954c..f8567b933ffa 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -47,7 +47,7 @@ $(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE $(call if_changed,quote2) quiet_cmd_mkcapflags = MKCAP $@ - cmd_mkcapflags = $(CONFIG_SHELL) $(srctree)/$(src)/../../x86/kernel/cpu/mkcapflags.sh $@ $^ + cmd_mkcapflags = $(CONFIG_SHELL) $(src)/../../x86/kernel/cpu/mkcapflags.sh $@ $^ cpufeature = $(src)/../../x86/include/asm/cpufeatures.h vmxfeature = $(src)/../../x86/include/asm/vmxfeatures.h diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index 3cece19b7473..29cda98c65f8 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -129,7 +129,7 @@ targets += mtools.conf # genimage.sh requires bash, but it also has a bunch of other # external dependencies. quiet_cmd_genimage = GENIMAGE $3 -cmd_genimage = $(BASH) $(srctree)/$(src)/genimage.sh $2 $3 $(obj)/bzImage \ + cmd_genimage = $(BASH) $(src)/genimage.sh $2 $3 $(obj)/bzImage \ $(obj)/mtools.conf '$(FDARGS)' $(FDINITRD) PHONY += bzdisk fdimage fdimage144 fdimage288 hdimage isoimage diff --git a/arch/x86/entry/vdso/Makefile b/arch/x86/entry/vdso/Makefile index 3d64bcc403cf..c003452dba8c 100644 --- a/arch/x86/entry/vdso/Makefile +++ b/arch/x86/entry/vdso/Makefile @@ -176,7 +176,7 @@ quiet_cmd_vdso = VDSO $@ cmd_vdso = $(LD) -o $@ \ $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -T $(filter %.lds,$^) $(filter %.o,$^) && \ - sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@' + sh $(src)/checkundef.sh '$(NM)' '$@' VDSO_LDFLAGS = -shared --hash-style=both --build-id=sha1 \ $(call ld-option, --eh-frame-hdr) -Bsymbolic -z noexecstack diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 74077694da7d..8a9fc9ec69d3 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -40,7 +40,7 @@ KMSAN_SANITIZE_sev.o := n KCOV_INSTRUMENT_head$(BITS).o := n KCOV_INSTRUMENT_sev.o := n -CFLAGS_irq.o := -I $(srctree)/$(src)/../include/asm/trace +CFLAGS_irq.o := -I $(src)/../include/asm/trace obj-y += head_$(BITS).o obj-y += head$(BITS).o diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index eb4dbcdf41f1..a02bba0ed6b9 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -60,7 +60,7 @@ obj-$(CONFIG_ACRN_GUEST) += acrn.o obj-$(CONFIG_DEBUG_FS) += debugfs.o quiet_cmd_mkcapflags = MKCAP $@ - cmd_mkcapflags = $(CONFIG_SHELL) $(srctree)/$(src)/mkcapflags.sh $@ $^ + cmd_mkcapflags = $(CONFIG_SHELL) $(src)/mkcapflags.sh $@ $^ cpufeature = $(src)/../../include/asm/cpufeatures.h vmxfeature = $(src)/../../include/asm/vmxfeatures.h diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 428048e73bd2..8d3a00e5c528 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -34,7 +34,7 @@ obj-y += pat/ CFLAGS_physaddr.o := -fno-stack-protector CFLAGS_mem_encrypt_identity.o := -fno-stack-protector -CFLAGS_fault.o := -I $(srctree)/$(src)/../include/asm/trace +CFLAGS_fault.o := -I $(src)/../include/asm/trace obj-$(CONFIG_X86_32) += pgtable_32.o iomap_32.o diff --git a/arch/x86/um/vdso/Makefile b/arch/x86/um/vdso/Makefile index b86d634730b2..2303fa59971c 100644 --- a/arch/x86/um/vdso/Makefile +++ b/arch/x86/um/vdso/Makefile @@ -63,7 +63,7 @@ quiet_cmd_vdso = VDSO $@ cmd_vdso = $(CC) -nostdlib -o $@ \ $(CC_FLAGS_LTO) $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) && \ - sh $(srctree)/$(src)/checkundef.sh '$(NM)' '$@' + sh $(src)/checkundef.sh '$(NM)' '$@' VDSO_LDFLAGS = -fPIC -shared -Wl,--hash-style=sysv -z noexecstack GCOV_PROFILE := n diff --git a/arch/xtensa/boot/dts/Makefile b/arch/xtensa/boot/dts/Makefile index 720628c0d8b9..d6408c16d74e 100644 --- a/arch/xtensa/boot/dts/Makefile +++ b/arch/xtensa/boot/dts/Makefile @@ -10,5 +10,4 @@ obj-$(CONFIG_OF) += $(addsuffix .dtb.o, $(CONFIG_BUILTIN_DTB_SOURCE)) # for CONFIG_OF_ALL_DTBS test -dtstree := $(srctree)/$(src) -dtb- := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts)) +dtb- := $(patsubst $(src)/%.dts,%.dtb, $(wildcard $(src)/*.dts)) diff --git a/certs/Makefile b/certs/Makefile index 799ad7b9e68a..1094e3860c2a 100644 --- a/certs/Makefile +++ b/certs/Makefile @@ -13,7 +13,7 @@ CFLAGS_blacklist_hashes.o := -I $(obj) quiet_cmd_check_and_copy_blacklist_hash_list = GEN $@ cmd_check_and_copy_blacklist_hash_list = \ $(if $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST), \ - $(AWK) -f $(srctree)/$(src)/check-blacklist-hashes.awk $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST) >&2; \ + $(AWK) -f $(src)/check-blacklist-hashes.awk $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST) >&2; \ { cat $(CONFIG_SYSTEM_BLACKLIST_HASH_LIST); echo $(comma) NULL; } > $@, \ echo NULL > $@) @@ -55,7 +55,7 @@ $(obj)/signing_key.pem: $(obj)/x509.genkey FORCE targets += signing_key.pem quiet_cmd_copy_x509_config = COPY $@ - cmd_copy_x509_config = cat $(srctree)/$(src)/default_x509.genkey > $@ + cmd_copy_x509_config = cat $(src)/default_x509.genkey > $@ # You can provide your own config file. If not present, copy the default one. $(obj)/x509.genkey: diff --git a/drivers/Makefile b/drivers/Makefile index 3bf5cab4b451..fe9ceb0d2288 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -6,11 +6,6 @@ # Rewritten to use lists instead of if-statements. # -# Some driver Makefiles miss $(srctree)/ for include directive. -ifdef building_out_of_srctree -MAKEFLAGS += --include-dir=$(srctree) -endif - obj-y += cache/ obj-y += irqchip/ obj-y += bus/ diff --git a/drivers/crypto/intel/qat/qat_420xx/Makefile b/drivers/crypto/intel/qat/qat_420xx/Makefile index a90fbe00b3c8..45728659fbc4 100644 --- a/drivers/crypto/intel/qat/qat_420xx/Makefile +++ b/drivers/crypto/intel/qat/qat_420xx/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_420XX) += qat_420xx.o qat_420xx-objs := adf_drv.o adf_420xx_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_4xxx/Makefile b/drivers/crypto/intel/qat/qat_4xxx/Makefile index ff9c8b5897ea..9ba202079a22 100644 --- a/drivers/crypto/intel/qat/qat_4xxx/Makefile +++ b/drivers/crypto/intel/qat/qat_4xxx/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: (BSD-3-Clause OR GPL-2.0-only) -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_4XXX) += qat_4xxx.o qat_4xxx-objs := adf_drv.o adf_4xxx_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c3xxx/Makefile b/drivers/crypto/intel/qat/qat_c3xxx/Makefile index 92ef416ccc78..7a06ad519bfc 100644 --- a/drivers/crypto/intel/qat/qat_c3xxx/Makefile +++ b/drivers/crypto/intel/qat/qat_c3xxx/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C3XXX) += qat_c3xxx.o qat_c3xxx-objs := adf_drv.o adf_c3xxx_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile b/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile index b6d76825a92c..7ef633058c4f 100644 --- a/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile +++ b/drivers/crypto/intel/qat/qat_c3xxxvf/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C3XXXVF) += qat_c3xxxvf.o qat_c3xxxvf-objs := adf_drv.o adf_c3xxxvf_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c62x/Makefile b/drivers/crypto/intel/qat/qat_c62x/Makefile index d581f7c87d6c..cc9255b3b198 100644 --- a/drivers/crypto/intel/qat/qat_c62x/Makefile +++ b/drivers/crypto/intel/qat/qat_c62x/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C62X) += qat_c62x.o qat_c62x-objs := adf_drv.o adf_c62x_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_c62xvf/Makefile b/drivers/crypto/intel/qat/qat_c62xvf/Makefile index 446c3d638605..256786662d60 100644 --- a/drivers/crypto/intel/qat/qat_c62xvf/Makefile +++ b/drivers/crypto/intel/qat/qat_c62xvf/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_C62XVF) += qat_c62xvf.o qat_c62xvf-objs := adf_drv.o adf_c62xvf_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_dh895xcc/Makefile b/drivers/crypto/intel/qat/qat_dh895xcc/Makefile index 38d6f8e1624a..cfd3bd757715 100644 --- a/drivers/crypto/intel/qat/qat_dh895xcc/Makefile +++ b/drivers/crypto/intel/qat/qat_dh895xcc/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_DH895xCC) += qat_dh895xcc.o qat_dh895xcc-objs := adf_drv.o adf_dh895xcc_hw_data.o diff --git a/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile b/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile index 0153c85ce743..64b54e92b2b4 100644 --- a/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile +++ b/drivers/crypto/intel/qat/qat_dh895xccvf/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../qat_common +ccflags-y := -I $(src)/../qat_common obj-$(CONFIG_CRYPTO_DEV_QAT_DH895xCCVF) += qat_dh895xccvf.o qat_dh895xccvf-objs := adf_drv.o adf_dh895xccvf_hw_data.o diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 4536c8ad0e11..e568e89283c0 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -23,7 +23,7 @@ # Makefile for the drm device driver. This driver provides support for the # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. -FULL_AMD_PATH=$(srctree)/$(src)/.. +FULL_AMD_PATH=$(src)/.. DISPLAY_FOLDER_NAME=display FULL_AMD_DISPLAY_PATH = $(FULL_AMD_PATH)/$(DISPLAY_FOLDER_NAME) diff --git a/drivers/gpu/drm/arm/display/komeda/Makefile b/drivers/gpu/drm/arm/display/komeda/Makefile index 1931a7fa1a14..cf5287fcbbc2 100644 --- a/drivers/gpu/drm/arm/display/komeda/Makefile +++ b/drivers/gpu/drm/arm/display/komeda/Makefile @@ -1,8 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 ccflags-y := \ - -I $(srctree)/$(src)/../include \ - -I $(srctree)/$(src) + -I $(src)/../include \ + -I $(src) komeda-y := \ komeda_drv.o \ diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index fba73c38e235..2220448ed07f 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -41,7 +41,7 @@ CFLAGS_display/intel_fbdev.o = -Wno-override-init # drivers. Define I915 when building i915. subdir-ccflags-y += -DI915 -subdir-ccflags-y += -I$(srctree)/$(src) +subdir-ccflags-y += -I$(src) # Please keep these build lists sorted! @@ -434,7 +434,7 @@ no-header-test := \ always-$(CONFIG_DRM_I915_WERROR) += \ $(patsubst %.h,%.hdrtest, $(filter-out $(no-header-test), \ - $(shell cd $(srctree)/$(src) && find * -name '*.h'))) + $(shell cd $(src) && find * -name '*.h'))) quiet_cmd_hdrtest = HDRTEST $(patsubst %.hdrtest,%.h,$@) cmd_hdrtest = $(CC) $(filter-out $(CFLAGS_GCOV), $(c_flags)) -S -o /dev/null -x c /dev/null -include $<; \ diff --git a/drivers/gpu/drm/imagination/Makefile b/drivers/gpu/drm/imagination/Makefile index ec6db8e9b403..9bc6a3884c22 100644 --- a/drivers/gpu/drm/imagination/Makefile +++ b/drivers/gpu/drm/imagination/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only OR MIT # Copyright (c) 2023 Imagination Technologies Ltd. -subdir-ccflags-y := -I$(srctree)/$(src) +subdir-ccflags-y := -I$(src) powervr-y := \ pvr_ccb.o \ diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index b21ae2880c71..b8cc007fc1b9 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -1,8 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 -ccflags-y := -I $(srctree)/$(src) -ccflags-y += -I $(srctree)/$(src)/disp/dpu1 -ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(srctree)/$(src)/dsi -ccflags-$(CONFIG_DRM_MSM_DP) += -I $(srctree)/$(src)/dp +ccflags-y := -I $(src) +ccflags-y += -I $(src)/disp/dpu1 +ccflags-$(CONFIG_DRM_MSM_DSI) += -I $(src)/dsi +ccflags-$(CONFIG_DRM_MSM_DP) += -I $(src)/dp msm-y := \ adreno/adreno_device.o \ diff --git a/drivers/gpu/drm/nouveau/Kbuild b/drivers/gpu/drm/nouveau/Kbuild index cf6b3a80c0c8..c32c01827c1d 100644 --- a/drivers/gpu/drm/nouveau/Kbuild +++ b/drivers/gpu/drm/nouveau/Kbuild @@ -1,10 +1,8 @@ -NOUVEAU_PATH ?= $(srctree) - # SPDX-License-Identifier: MIT -ccflags-y += -I $(NOUVEAU_PATH)/$(src)/include -ccflags-y += -I $(NOUVEAU_PATH)/$(src)/include/nvkm -ccflags-y += -I $(NOUVEAU_PATH)/$(src)/nvkm -ccflags-y += -I $(NOUVEAU_PATH)/$(src) +ccflags-y += -I $(src)/include +ccflags-y += -I $(src)/include/nvkm +ccflags-y += -I $(src)/nvkm +ccflags-y += -I $(src) # NVKM - HW resource manager #- code also used by various userspace tools/tests diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index c29a850859ad..8162a522222b 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -32,7 +32,7 @@ endif # Enable -Werror in CI and development subdir-ccflags-$(CONFIG_DRM_XE_WERROR) += -Werror -subdir-ccflags-y += -I$(obj) -I$(srctree)/$(src) +subdir-ccflags-y += -I$(obj) -I$(src) # generated sources hostprogs := xe_gen_wa_oob @@ -43,7 +43,7 @@ quiet_cmd_wa_oob = GEN $(notdir $(generated_oob)) cmd_wa_oob = mkdir -p $(@D); $^ $(generated_oob) $(obj)/generated/%_wa_oob.c $(obj)/generated/%_wa_oob.h: $(obj)/xe_gen_wa_oob \ - $(srctree)/$(src)/xe_wa_oob.rules + $(src)/xe_wa_oob.rules $(call cmd,wa_oob) uses_generated_oob := \ @@ -166,8 +166,8 @@ endif # i915 Display compat #defines and #includes subdir-ccflags-$(CONFIG_DRM_XE_DISPLAY) += \ - -I$(srctree)/$(src)/display/ext \ - -I$(srctree)/$(src)/compat-i915-headers \ + -I$(src)/display/ext \ + -I$(src)/compat-i915-headers \ -I$(srctree)/drivers/gpu/drm/i915/display/ \ -Ddrm_i915_gem_object=xe_bo \ -Ddrm_i915_private=xe_device @@ -310,7 +310,7 @@ ifneq ($(CONFIG_DRM_XE_DISPLAY),y) endif always-$(CONFIG_DRM_XE_WERROR) += \ - $(patsubst %.h,%.hdrtest, $(shell cd $(srctree)/$(src) && find * -name '*.h' $(hdrtest_find_args))) + $(patsubst %.h,%.hdrtest, $(shell cd $(src) && find * -name '*.h' $(hdrtest_find_args))) quiet_cmd_hdrtest = HDRTEST $(patsubst %.hdrtest,%.h,$@) cmd_hdrtest = $(CC) -DHDRTEST $(filter-out $(CFLAGS_GCOV), $(c_flags)) -S -o /dev/null -x c /dev/null -include $<; touch $@ diff --git a/drivers/hid/amd-sfh-hid/Makefile b/drivers/hid/amd-sfh-hid/Makefile index 0222170ab7ad..106514b54d16 100644 --- a/drivers/hid/amd-sfh-hid/Makefile +++ b/drivers/hid/amd-sfh-hid/Makefile @@ -13,4 +13,4 @@ amd_sfh-objs += sfh1_1/amd_sfh_init.o amd_sfh-objs += sfh1_1/amd_sfh_interface.o amd_sfh-objs += sfh1_1/amd_sfh_desc.o -ccflags-y += -I $(srctree)/$(src)/ +ccflags-y += -I $(src)/ diff --git a/drivers/hid/intel-ish-hid/Makefile b/drivers/hid/intel-ish-hid/Makefile index f0a82b1c7cb9..db4974c43431 100644 --- a/drivers/hid/intel-ish-hid/Makefile +++ b/drivers/hid/intel-ish-hid/Makefile @@ -23,4 +23,4 @@ intel-ishtp-hid-objs += ishtp-hid-client.o obj-$(CONFIG_INTEL_ISH_FIRMWARE_DOWNLOADER) += intel-ishtp-loader.o intel-ishtp-loader-objs += ishtp-fw-loader.o -ccflags-y += -I $(srctree)/$(src)/ishtp +ccflags-y += -I $(src)/ishtp diff --git a/drivers/md/dm-vdo/Makefile b/drivers/md/dm-vdo/Makefile index 33e09abc6acd..9476957bfbf4 100644 --- a/drivers/md/dm-vdo/Makefile +++ b/drivers/md/dm-vdo/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I$(srctree)/$(src) -I$(srctree)/$(src)/indexer +ccflags-y := -I$(src) -I$(src)/indexer obj-$(CONFIG_DM_VDO) += dm-vdo.o diff --git a/drivers/net/ethernet/aquantia/atlantic/Makefile b/drivers/net/ethernet/aquantia/atlantic/Makefile index 8ebcc68e807f..f6a96931c89a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/Makefile +++ b/drivers/net/ethernet/aquantia/atlantic/Makefile @@ -8,7 +8,7 @@ obj-$(CONFIG_AQTION) += atlantic.o -ccflags-y += -I$(srctree)/$(src) +ccflags-y += -I$(src) atlantic-objs := aq_main.o \ aq_nic.o \ diff --git a/drivers/net/ethernet/chelsio/libcxgb/Makefile b/drivers/net/ethernet/chelsio/libcxgb/Makefile index aa79264e72ba..fbedc31674b3 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/Makefile +++ b/drivers/net/ethernet/chelsio/libcxgb/Makefile @@ -1,5 +1,5 @@ # SPDX-License-Identifier: GPL-2.0-only -ccflags-y := -I $(srctree)/$(src)/../cxgb4 +ccflags-y := -I $(src)/../cxgb4 obj-$(CONFIG_CHELSIO_LIB) += libcxgb.o diff --git a/drivers/net/ethernet/fungible/funeth/Makefile b/drivers/net/ethernet/fungible/funeth/Makefile index 646d69595b4f..d51e4c2b4a1a 100644 --- a/drivers/net/ethernet/fungible/funeth/Makefile +++ b/drivers/net/ethernet/fungible/funeth/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause) -ccflags-y += -I$(srctree)/$(src)/../funcore -I$(srctree)/$(src) +ccflags-y += -I$(src)/../funcore -I$(src) obj-$(CONFIG_FUN_ETH) += funeth.o diff --git a/drivers/net/ethernet/hisilicon/hns3/Makefile b/drivers/net/ethernet/hisilicon/hns3/Makefile index e214bfaece1f..8e9293e57bfd 100644 --- a/drivers/net/ethernet/hisilicon/hns3/Makefile +++ b/drivers/net/ethernet/hisilicon/hns3/Makefile @@ -3,7 +3,7 @@ # Makefile for the HISILICON network device drivers. # -ccflags-y += -I$(srctree)/$(src) +ccflags-y += -I$(src) ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3pf ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3vf ccflags-y += -I$(srctree)/drivers/net/ethernet/hisilicon/hns3/hns3_common diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile index dc6d27a36faa..e5ca0f511822 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile @@ -6,8 +6,8 @@ # ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/../include + -I $(src) \ + -I $(src)/../include obj-$(CONFIG_BRCMFMAC) += brcmfmac.o brcmfmac-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile index 46098705e236..5e37c638f966 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bca/Makefile @@ -3,9 +3,9 @@ # Copyright (c) 2022 Broadcom Corporation ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/.. \ - -I $(srctree)/$(src)/../../include + -I $(src) \ + -I $(src)/.. \ + -I $(src)/../../include obj-m += brcmfmac-bca.o brcmfmac-bca-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile index 5e1fddaff79e..33e86724ba14 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cyw/Makefile @@ -3,9 +3,9 @@ # Copyright (c) 2022 Broadcom Corporation ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/.. \ - -I $(srctree)/$(src)/../../include + -I $(src) \ + -I $(src)/.. \ + -I $(src)/../../include obj-m += brcmfmac-cyw.o brcmfmac-cyw-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile index 7f455a19a2b1..3db4cce44f42 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/wcc/Makefile @@ -3,9 +3,9 @@ # Copyright (c) 2022 Broadcom Corporation ccflags-y += \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/.. \ - -I $(srctree)/$(src)/../../include + -I $(src) \ + -I $(src)/.. \ + -I $(src)/../../include obj-m += brcmfmac-wcc.o brcmfmac-wcc-objs += \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile index 090757730ba6..ffdd17eabe1d 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile @@ -16,9 +16,9 @@ # CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. ccflags-y := \ - -I $(srctree)/$(src) \ - -I $(srctree)/$(src)/phy \ - -I $(srctree)/$(src)/../include + -I $(src) \ + -I $(src)/phy \ + -I $(src)/../include brcmsmac-y := \ mac80211_if.o \ diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile b/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile index 7a82d615ba2a..f9b40cab766c 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile +++ b/drivers/net/wireless/broadcom/brcm80211/brcmutil/Makefile @@ -4,7 +4,7 @@ # # Copyright (c) 2011 Broadcom Corporation # -ccflags-y := -I $(srctree)/$(src)/../include +ccflags-y := -I $(src)/../include obj-$(CONFIG_BRCMUTIL) += brcmutil.o brcmutil-objs = utils.o d11.o diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/Makefile b/drivers/net/wireless/intel/iwlwifi/dvm/Makefile index 0486b17d7c41..6109d64006db 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/dvm/Makefile @@ -11,4 +11,4 @@ iwldvm-objs += rxon.o devices.o iwldvm-$(CONFIG_IWLWIFI_LEDS) += led.o iwldvm-$(CONFIG_IWLWIFI_DEBUGFS) += debugfs.o -ccflags-y += -I $(srctree)/$(src)/../ +ccflags-y += -I $(src)/../ diff --git a/drivers/net/wireless/intel/iwlwifi/mei/Makefile b/drivers/net/wireless/intel/iwlwifi/mei/Makefile index 8e3ef0347db7..e05f9421be4a 100644 --- a/drivers/net/wireless/intel/iwlwifi/mei/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/mei/Makefile @@ -5,4 +5,4 @@ iwlmei-y += net.o iwlmei-$(CONFIG_IWLWIFI_DEVICE_TRACING) += trace.o CFLAGS_trace.o := -I$(src) -ccflags-y += -I $(srctree)/$(src)/../ +ccflags-y += -I $(src)/../ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile index 593fe28d89cf..764ba73cde1e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/mvm/Makefile @@ -15,4 +15,4 @@ iwlmvm-$(CONFIG_IWLWIFI_LEDS) += led.o iwlmvm-$(CONFIG_PM) += d3.o iwlmvm-$(CONFIG_IWLMEI) += vendor-cmd.o -ccflags-y += -I $(srctree)/$(src)/../ +ccflags-y += -I $(src)/../ diff --git a/drivers/net/wireless/intel/iwlwifi/tests/Makefile b/drivers/net/wireless/intel/iwlwifi/tests/Makefile index 5658471bdf0a..84491488f589 100644 --- a/drivers/net/wireless/intel/iwlwifi/tests/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/tests/Makefile @@ -2,6 +2,6 @@ iwlwifi-tests-y += module.o devinfo.o -ccflags-y += -I$(srctree)/$(src)/../ +ccflags-y += -I$(src)/../ obj-$(CONFIG_IWLWIFI_KUNIT_TESTS) += iwlwifi-tests.o diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile b/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile index 565a9a114134..1044105c2557 100644 --- a/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile +++ b/drivers/net/wireless/realtek/rtl818x/rtl8180/Makefile @@ -3,4 +3,4 @@ rtl818x_pci-objs := dev.o rtl8225.o sa2400.o max2820.o grf5101.o rtl8225se.o obj-$(CONFIG_RTL8180) += rtl818x_pci.o -ccflags-y += -I $(srctree)/$(src)/.. +ccflags-y += -I $(src)/.. diff --git a/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile b/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile index 0bf64dfb233a..3deded2c05b6 100644 --- a/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile +++ b/drivers/net/wireless/realtek/rtl818x/rtl8187/Makefile @@ -3,4 +3,4 @@ rtl8187-objs := dev.o rtl8225.o leds.o rfkill.o obj-$(CONFIG_RTL8187) += rtl8187.o -ccflags-y += -I $(srctree)/$(src)/.. +ccflags-y += -I $(src)/.. diff --git a/drivers/scsi/aic7xxx/Makefile b/drivers/scsi/aic7xxx/Makefile index e0188ecd85b2..853c72a81ae0 100644 --- a/drivers/scsi/aic7xxx/Makefile +++ b/drivers/scsi/aic7xxx/Makefile @@ -55,9 +55,9 @@ aicasm-7xxx-opts-$(CONFIG_AIC7XXX_REG_PRETTY_PRINT) := \ ifeq ($(CONFIG_AIC7XXX_BUILD_FIRMWARE),y) $(obj)/aic7xxx_seq.h: $(src)/aic7xxx.seq $(src)/aic7xxx.reg $(obj)/aicasm/aicasm - $(obj)/aicasm/aicasm -I$(srctree)/$(src) -r $(obj)/aic7xxx_reg.h \ + $(obj)/aicasm/aicasm -I $(src) -r $(obj)/aic7xxx_reg.h \ $(aicasm-7xxx-opts-y) -o $(obj)/aic7xxx_seq.h \ - $(srctree)/$(src)/aic7xxx.seq + $(src)/aic7xxx.seq $(aic7xxx-gen-y): $(objtree)/$(obj)/aic7xxx_seq.h @true @@ -73,9 +73,9 @@ aicasm-79xx-opts-$(CONFIG_AIC79XX_REG_PRETTY_PRINT) := \ ifeq ($(CONFIG_AIC79XX_BUILD_FIRMWARE),y) $(obj)/aic79xx_seq.h: $(src)/aic79xx.seq $(src)/aic79xx.reg $(obj)/aicasm/aicasm - $(obj)/aicasm/aicasm -I$(srctree)/$(src) -r $(obj)/aic79xx_reg.h \ + $(obj)/aicasm/aicasm -I $(src) -r $(obj)/aic79xx_reg.h \ $(aicasm-79xx-opts-y) -o $(obj)/aic79xx_seq.h \ - $(srctree)/$(src)/aic79xx.seq + $(src)/aic79xx.seq $(aic79xx-gen-y): $(objtree)/$(obj)/aic79xx_seq.h @true @@ -83,5 +83,5 @@ else $(obj)/aic79xx_reg_print.c: $(src)/aic79xx_reg_print.c_shipped endif -$(obj)/aicasm/aicasm: $(srctree)/$(src)/aicasm/*.[chyl] - $(MAKE) -C $(srctree)/$(src)/aicasm OUTDIR=$(shell pwd)/$(obj)/aicasm/ +$(obj)/aicasm/aicasm: $(src)/aicasm/*.[chyl] + $(MAKE) -C $(src)/aicasm OUTDIR=$(shell pwd)/$(obj)/aicasm/ diff --git a/drivers/staging/rtl8723bs/Makefile b/drivers/staging/rtl8723bs/Makefile index 590bde02058c..7f5067e89295 100644 --- a/drivers/staging/rtl8723bs/Makefile +++ b/drivers/staging/rtl8723bs/Makefile @@ -62,4 +62,4 @@ r8723bs-y = \ obj-$(CONFIG_RTL8723BS) := r8723bs.o -ccflags-y += -I$(srctree)/$(src)/include -I$(srctree)/$(src)/hal +ccflags-y += -I$(src)/include -I$(src)/hal diff --git a/fs/iomap/Makefile b/fs/iomap/Makefile index fc070184b7fa..381d76c5c232 100644 --- a/fs/iomap/Makefile +++ b/fs/iomap/Makefile @@ -4,7 +4,7 @@ # All Rights Reserved. # -ccflags-y += -I $(srctree)/$(src) # needed for trace events +ccflags-y += -I $(src) # needed for trace events obj-$(CONFIG_FS_IOMAP) += iomap.o diff --git a/fs/unicode/Makefile b/fs/unicode/Makefile index 0e51c0025a16..e309afe2b2bb 100644 --- a/fs/unicode/Makefile +++ b/fs/unicode/Makefile @@ -18,13 +18,13 @@ ifdef REGENERATE_UTF8DATA quiet_cmd_utf8data = GEN $@ cmd_utf8data = $< \ - -a $(srctree)/$(src)/DerivedAge.txt \ - -c $(srctree)/$(src)/DerivedCombiningClass.txt \ - -p $(srctree)/$(src)/DerivedCoreProperties.txt \ - -d $(srctree)/$(src)/UnicodeData.txt \ - -f $(srctree)/$(src)/CaseFolding.txt \ - -n $(srctree)/$(src)/NormalizationCorrections.txt \ - -t $(srctree)/$(src)/NormalizationTest.txt \ + -a $(src)/DerivedAge.txt \ + -c $(src)/DerivedCombiningClass.txt \ + -p $(src)/DerivedCoreProperties.txt \ + -d $(src)/UnicodeData.txt \ + -f $(src)/CaseFolding.txt \ + -n $(src)/NormalizationCorrections.txt \ + -t $(src)/NormalizationTest.txt \ -o $@ $(obj)/utf8data.c: $(obj)/mkutf8data $(filter %.txt, $(cmd_utf8data)) FORCE diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 76674ad5833e..c5a35e32adf0 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -4,8 +4,8 @@ # All Rights Reserved. # -ccflags-y += -I $(srctree)/$(src) # needed for trace events -ccflags-y += -I $(srctree)/$(src)/libxfs +ccflags-y += -I $(src) # needed for trace events +ccflags-y += -I $(src)/libxfs obj-$(CONFIG_XFS_FS) += xfs.o diff --git a/init/Makefile b/init/Makefile index cbac576c57d6..3c48d97538c1 100644 --- a/init/Makefile +++ b/init/Makefile @@ -52,7 +52,7 @@ CFLAGS_version.o := -include $(obj)/utsversion-tmp.h # Build version-timestamp.c with final UTS_VERSION # -include/generated/utsversion.h: build-version-auto = $(shell $(srctree)/$(src)/build-version) +include/generated/utsversion.h: build-version-auto = $(shell $(src)/build-version) include/generated/utsversion.h: build-timestamp-auto = $(shell LC_ALL=C date) include/generated/utsversion.h: FORCE $(call filechk,uts_version) diff --git a/lib/Makefile b/lib/Makefile index ffc6b2341b45..27baa6e6daa8 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -352,7 +352,7 @@ $(obj)/oid_registry_data.c: $(srctree)/include/linux/oid_registry.h \ $(call cmd,build_OID_registry) quiet_cmd_build_OID_registry = GEN $@ - cmd_build_OID_registry = perl $(srctree)/$(src)/build_OID_registry $< $@ + cmd_build_OID_registry = perl $(src)/build_OID_registry $< $@ clean-files += oid_registry_data.c @@ -412,8 +412,8 @@ obj-$(CONFIG_GENERIC_LIB_DEVMEM_IS_ALLOWED) += devmem_is_allowed.o obj-$(CONFIG_FIRMWARE_TABLE) += fw_table.o # FORTIFY_SOURCE compile-time behavior tests -TEST_FORTIFY_SRCS = $(wildcard $(srctree)/$(src)/test_fortify/*-*.c) -TEST_FORTIFY_LOGS = $(patsubst $(srctree)/$(src)/%.c, %.log, $(TEST_FORTIFY_SRCS)) +TEST_FORTIFY_SRCS = $(wildcard $(src)/test_fortify/*-*.c) +TEST_FORTIFY_LOGS = $(patsubst $(src)/%.c, %.log, $(TEST_FORTIFY_SRCS)) TEST_FORTIFY_LOG = test_fortify.log quiet_cmd_test_fortify = TEST $@ diff --git a/lib/raid6/Makefile b/lib/raid6/Makefile index 385a94aa0b99..8785353c6140 100644 --- a/lib/raid6/Makefile +++ b/lib/raid6/Makefile @@ -53,7 +53,7 @@ endif endif quiet_cmd_unroll = UNROLL $@ - cmd_unroll = $(AWK) -v N=$* -f $(srctree)/$(src)/unroll.awk < $< > $@ + cmd_unroll = $(AWK) -v N=$* -f $(src)/unroll.awk < $< > $@ targets += int1.c int2.c int4.c int8.c $(obj)/int%.c: $(src)/int.uc $(src)/unroll.awk FORCE diff --git a/net/wireless/Makefile b/net/wireless/Makefile index 72074fd36df4..1d49cc8b6da1 100644 --- a/net/wireless/Makefile +++ b/net/wireless/Makefile @@ -25,7 +25,7 @@ ifneq ($(CONFIG_CFG80211_EXTRA_REGDB_KEYDIR),) cfg80211-y += extra-certs.o endif -$(obj)/shipped-certs.c: $(sort $(wildcard $(srctree)/$(src)/certs/*.hex)) +$(obj)/shipped-certs.c: $(sort $(wildcard $(src)/certs/*.hex)) @$(kecho) " GEN $@" $(Q)(echo '#include "reg.h"'; \ echo 'const u8 shipped_regdb_certs[] = {'; \ diff --git a/rust/Makefile b/rust/Makefile index 846e6ab9d5a9..85e46dde2a5b 100644 --- a/rust/Makefile +++ b/rust/Makefile @@ -239,7 +239,7 @@ quiet_cmd_rustsysroot = RUSTSYSROOT rm -rf $(objtree)/$(obj)/test; \ mkdir -p $(objtree)/$(obj)/test; \ cp -a $(rustc_sysroot) $(objtree)/$(obj)/test/sysroot; \ - cp -r $(srctree)/$(src)/alloc/* \ + cp -r $(src)/alloc/* \ $(objtree)/$(obj)/test/sysroot/lib/rustlib/src/rust/library/alloc/src; \ echo '\#!/bin/sh' > $(objtree)/$(obj)/test/rustc_sysroot; \ echo "$(RUSTC) --sysroot=$(abspath $(objtree)/$(obj)/test/sysroot) \"\$$@\"" \ @@ -340,7 +340,7 @@ quiet_cmd_bindgen = BINDGEN $@ $(bindgen_target_cflags) $(bindgen_target_extra) $(obj)/bindings/bindings_generated.rs: private bindgen_target_flags = \ - $(shell grep -Ev '^#|^$$' $(srctree)/$(src)/bindgen_parameters) + $(shell grep -Ev '^#|^$$' $(src)/bindgen_parameters) $(obj)/bindings/bindings_generated.rs: private bindgen_target_extra = ; \ sed -Ei 's/pub const RUST_CONST_HELPER_([a-zA-Z0-9_]*)/pub const \1/g' $@ $(obj)/bindings/bindings_generated.rs: $(src)/bindings/bindings_helper.h \ @@ -348,7 +348,7 @@ $(obj)/bindings/bindings_generated.rs: $(src)/bindings/bindings_helper.h \ $(call if_changed_dep,bindgen) $(obj)/uapi/uapi_generated.rs: private bindgen_target_flags = \ - $(shell grep -Ev '^#|^$$' $(srctree)/$(src)/bindgen_parameters) + $(shell grep -Ev '^#|^$$' $(src)/bindgen_parameters) $(obj)/uapi/uapi_generated.rs: $(src)/uapi/uapi_helper.h \ $(src)/bindgen_parameters FORCE $(call if_changed_dep,bindgen) diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index 933f6c3fe6b0..f054e0828f13 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 -BPF_SAMPLES_PATH ?= $(abspath $(srctree)/$(src)) +BPF_SAMPLES_PATH ?= $(abspath $(src)) TOOLS_PATH := $(BPF_SAMPLES_PATH)/../../tools pound := \# diff --git a/samples/hid/Makefile b/samples/hid/Makefile index 9f7fe29dd749..c128ccd49974 100644 --- a/samples/hid/Makefile +++ b/samples/hid/Makefile @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 -HID_SAMPLES_PATH ?= $(abspath $(srctree)/$(src)) +HID_SAMPLES_PATH ?= $(abspath $(src)) TOOLS_PATH := $(HID_SAMPLES_PATH)/../../tools pound := \# diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index 2f331879816b..faf37bafa3f8 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -62,8 +62,7 @@ stringify = $(squote)$(quote)$1$(quote)$(squote) ### # The path to Kbuild or Makefile. Kbuild has precedence over Makefile. -kbuild-dir = $(if $(filter /%,$(src)),$(src),$(srctree)/$(src)) -kbuild-file = $(or $(wildcard $(kbuild-dir)/Kbuild),$(kbuild-dir)/Makefile) +kbuild-file = $(or $(wildcard $(src)/Kbuild),$(src)/Makefile) ### # Read a file, replacing newlines with spaces diff --git a/scripts/Makefile.asm-generic b/scripts/Makefile.asm-generic index 8d01b37b7677..1486abf34c7c 100644 --- a/scripts/Makefile.asm-generic +++ b/scripts/Makefile.asm-generic @@ -9,7 +9,7 @@ PHONY := all all: -src := $(subst /generated,,$(obj)) +src := $(srctree)/$(subst /generated,,$(obj)) include $(srctree)/scripts/Kbuild.include -include $(kbuild-file) @@ -20,14 +20,14 @@ include $(srctree)/$(generic)/Kbuild endif redundant := $(filter $(mandatory-y) $(generated-y), $(generic-y)) -redundant += $(foreach f, $(generic-y), $(if $(wildcard $(srctree)/$(src)/$(f)),$(f))) +redundant += $(foreach f, $(generic-y), $(if $(wildcard $(src)/$(f)),$(f))) redundant := $(sort $(redundant)) $(if $(redundant),\ $(warning redundant generic-y found in $(src)/Kbuild: $(redundant))) # If arch does not implement mandatory headers, fallback to asm-generic ones. mandatory-y := $(filter-out $(generated-y), $(mandatory-y)) -generic-y += $(foreach f, $(mandatory-y), $(if $(wildcard $(srctree)/$(src)/$(f)),,$(f))) +generic-y += $(foreach f, $(mandatory-y), $(if $(wildcard $(src)/$(f)),,$(f))) generic-y := $(addprefix $(obj)/, $(generic-y)) generated-y := $(addprefix $(obj)/, $(generated-y)) diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 41138346afb0..43b71c3d0ff6 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -3,7 +3,7 @@ # Building # ========================================================================== -src := $(obj) +src := $(if $(VPATH),$(VPATH)/)$(obj) PHONY := $(obj)/ $(obj)/: diff --git a/scripts/Makefile.clean b/scripts/Makefile.clean index f2cb4d7ffd96..4fcfab40ed61 100644 --- a/scripts/Makefile.clean +++ b/scripts/Makefile.clean @@ -3,7 +3,7 @@ # Cleaning up # ========================================================================== -src := $(obj) +src := $(if $(VPATH),$(VPATH)/)$(obj) PHONY := __clean __clean: diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index 42d3c772c8e0..08d42e93bea0 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -209,13 +209,13 @@ _c_flags += $(if $(patsubst n%,, \ -D__KCSAN_INSTRUMENT_BARRIERS__) endif -# $(srctree)/$(src) for including checkin headers from generated source files -# $(objtree)/$(obj) for including generated headers from checkin source files +# $(src) for including checkin headers from generated source files +# $(obj) for including generated headers from checkin source files ifeq ($(KBUILD_EXTMOD),) ifdef building_out_of_srctree -_c_flags += $(addprefix -I $(srctree)/,$(src)) $(addprefix -I $(objtree)/,$(obj)) -_a_flags += $(addprefix -I $(srctree)/,$(src)) $(addprefix -I $(objtree)/,$(obj)) -_cpp_flags += $(addprefix -I $(srctree)/,$(src)) $(addprefix -I $(objtree)/,$(obj)) +_c_flags += $(addprefix -I, $(src) $(obj)) +_a_flags += $(addprefix -I, $(src) $(obj)) +_cpp_flags += $(addprefix -I, $(src) $(obj)) endif endif diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index 739402f45509..68c4f9ee0c90 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -112,7 +112,7 @@ else # set src + obj - they may be used in the modules's Makefile obj := $(KBUILD_EXTMOD) -src := $(obj) +src := $(if $(VPATH),$(VPATH)/)$(obj) # Include the module's Makefile to find KBUILD_EXTRA_SYMBOLS include $(kbuild-file) diff --git a/scripts/dtc/Makefile b/scripts/dtc/Makefile index 4d32b9497da9..a18657072541 100644 --- a/scripts/dtc/Makefile +++ b/scripts/dtc/Makefile @@ -16,12 +16,12 @@ libfdt = $(addprefix libfdt/,$(libfdt-objs)) fdtoverlay-objs := $(libfdt) fdtoverlay.o util.o # Source files need to get at the userspace version of libfdt_env.h to compile -HOST_EXTRACFLAGS += -I $(srctree)/$(src)/libfdt +HOST_EXTRACFLAGS += -I $(src)/libfdt HOST_EXTRACFLAGS += -DNO_YAML # Generated files need one more search path to include headers in source tree -HOSTCFLAGS_dtc-lexer.lex.o := -I $(srctree)/$(src) -HOSTCFLAGS_dtc-parser.tab.o := -I $(srctree)/$(src) +HOSTCFLAGS_dtc-lexer.lex.o := -I $(src) +HOSTCFLAGS_dtc-parser.tab.o := -I $(src) # dependencies on generated files need to be listed explicitly $(obj)/dtc-lexer.lex.o: $(obj)/dtc-parser.tab.h diff --git a/scripts/gdb/linux/Makefile b/scripts/gdb/linux/Makefile index 48941faa6ea6..d77ad9079d0f 100644 --- a/scripts/gdb/linux/Makefile +++ b/scripts/gdb/linux/Makefile @@ -2,7 +2,7 @@ ifdef building_out_of_srctree -symlinks := $(patsubst $(srctree)/$(src)/%,%,$(wildcard $(srctree)/$(src)/*.py)) +symlinks := $(patsubst $(src)/%,%,$(wildcard $(src)/*.py)) quiet_cmd_symlink = SYMLINK $@ cmd_symlink = ln -fsn $(patsubst $(obj)/%,$(abspath $(srctree))/$(src)/%,$@) $@ diff --git a/scripts/genksyms/Makefile b/scripts/genksyms/Makefile index d6a422a63b6a..312edccda736 100644 --- a/scripts/genksyms/Makefile +++ b/scripts/genksyms/Makefile @@ -23,8 +23,8 @@ $(obj)/pars%.tab.c $(obj)/pars%.tab.h: $(src)/pars%.y FORCE endif # -I needed for generated C source to include headers in source tree -HOSTCFLAGS_parse.tab.o := -I $(srctree)/$(src) -HOSTCFLAGS_lex.lex.o := -I $(srctree)/$(src) +HOSTCFLAGS_parse.tab.o := -I $(src) +HOSTCFLAGS_lex.lex.o := -I $(src) # dependencies on generated files need to be listed explicitly $(obj)/lex.lex.o: $(obj)/parse.tab.h diff --git a/scripts/kconfig/Makefile b/scripts/kconfig/Makefile index ea1bf3b3dbde..a0a0be38cbdc 100644 --- a/scripts/kconfig/Makefile +++ b/scripts/kconfig/Makefile @@ -57,7 +57,7 @@ $(foreach c, config menuconfig nconfig gconfig xconfig, $(eval $(call config_rul PHONY += localmodconfig localyesconfig localyesconfig localmodconfig: $(obj)/conf - $(Q)$(PERL) $(srctree)/$(src)/streamline_config.pl --$@ $(srctree) $(Kconfig) > .tmp.config + $(Q)$(PERL) $(src)/streamline_config.pl --$@ $(srctree) $(Kconfig) > .tmp.config $(Q)if [ -f .config ]; then \ cmp -s .tmp.config .config || \ (mv -f .config .config.old.1; \ @@ -118,7 +118,7 @@ tinyconfig: # CHECK: -o cache_dir= working? PHONY += testconfig testconfig: $(obj)/conf - $(Q)$(PYTHON3) -B -m pytest $(srctree)/$(src)/tests \ + $(Q)$(PYTHON3) -B -m pytest $(src)/tests \ -o cache_dir=$(abspath $(obj)/tests/.cache) \ $(if $(findstring 1,$(KBUILD_VERBOSE)),--capture=no) clean-files += tests/.cache @@ -165,8 +165,8 @@ common-objs := confdata.o expr.o lexer.lex.o menu.o parser.tab.o \ preprocess.o symbol.o util.o $(obj)/lexer.lex.o: $(obj)/parser.tab.h -HOSTCFLAGS_lexer.lex.o := -I $(srctree)/$(src) -HOSTCFLAGS_parser.tab.o := -I $(srctree)/$(src) +HOSTCFLAGS_lexer.lex.o := -I $(src) +HOSTCFLAGS_parser.tab.o := -I $(src) # conf: Used for defconfig, oldconfig and related targets hostprogs += conf diff --git a/security/tomoyo/Makefile b/security/tomoyo/Makefile index 884ff155edc3..55c67b9846a9 100644 --- a/security/tomoyo/Makefile +++ b/security/tomoyo/Makefile @@ -11,7 +11,7 @@ quiet_cmd_policy = POLICY $@ printf '\t"";\n';) \ } > $@ -$(obj)/builtin-policy.h: $(wildcard $(obj)/policy/*.conf $(srctree)/$(src)/policy/*.conf.default) FORCE +$(obj)/builtin-policy.h: $(wildcard $(obj)/policy/*.conf $(src)/policy/*.conf.default) FORCE $(call if_changed,policy) ifndef CONFIG_SECURITY_TOMOYO_INSECURE_BUILTIN_SETTING diff --git a/usr/Makefile b/usr/Makefile index f8e1ad19e05c..132ef7e96e6d 100644 --- a/usr/Makefile +++ b/usr/Makefile @@ -22,7 +22,7 @@ cpio-data := # If CONFIG_INITRAMFS_SOURCE is empty, generate a small initramfs with the # default contents. ifeq ($(ramfs-input),) -ramfs-input := $(srctree)/$(src)/default_cpio_list +ramfs-input := $(src)/default_cpio_list endif ifeq ($(words $(ramfs-input)),1) diff --git a/usr/include/Makefile b/usr/include/Makefile index 338c81f1fcf3..771e32872b2a 100644 --- a/usr/include/Makefile +++ b/usr/include/Makefile @@ -78,7 +78,7 @@ quiet_cmd_hdrtest = HDRTEST $< cmd_hdrtest = \ $(CC) $(c_flags) -fsyntax-only -x c /dev/null \ $(if $(filter-out $(no-header-test), $*.h), -include $< -include $<); \ - $(PERL) $(srctree)/$(src)/headers_check.pl $(obj) $(SRCARCH) $<; \ + $(PERL) $(src)/headers_check.pl $(obj) $(SRCARCH) $<; \ touch $@ $(obj)/%.hdrtest: $(obj)/%.h FORCE -- cgit v1.2.3 From 33fb988b67050d9bb512f77f08453fa00088943c Mon Sep 17 00:00:00 2001 From: Praveen Kumar Kannoju Date: Wed, 8 May 2024 19:06:17 +0530 Subject: net/sched: adjust device watchdog timer to detect stopped queue at right time Applications are sensitive to long network latency, particularly heartbeat monitoring ones. Longer the tx timeout recovery higher the risk with such applications on a production machines. This patch remedies, yet honoring device set tx timeout. Modify watchdog next timeout to be shorter than the device specified. Compute the next timeout be equal to device watchdog timeout less the how long ago queue stop had been done. At next watchdog timeout tx timeout handler is called into if still in stopped state. Either called or not called, restore the watchdog timeout back to device specified. Signed-off-by: Praveen Kumar Kannoju Link: https://lore.kernel.org/r/20240508133617.4424-1-praveen.kannoju@oracle.com Signed-off-by: Jakub Kicinski --- net/sched/sch_generic.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index d3f6006b563c..2a637a17061b 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -506,19 +506,22 @@ static void dev_watchdog(struct timer_list *t) unsigned int timedout_ms = 0; unsigned int i; unsigned long trans_start; + unsigned long oldest_start = jiffies; for (i = 0; i < dev->num_tx_queues; i++) { struct netdev_queue *txq; txq = netdev_get_tx_queue(dev, i); trans_start = READ_ONCE(txq->trans_start); - if (netif_xmit_stopped(txq) && - time_after(jiffies, (trans_start + - dev->watchdog_timeo))) { + if (!netif_xmit_stopped(txq)) + continue; + if (time_after(jiffies, trans_start + dev->watchdog_timeo)) { timedout_ms = jiffies_to_msecs(jiffies - trans_start); atomic_long_inc(&txq->trans_timeout); break; } + if (time_after(oldest_start, trans_start)) + oldest_start = trans_start; } if (unlikely(timedout_ms)) { @@ -531,7 +534,7 @@ static void dev_watchdog(struct timer_list *t) netif_unfreeze_queues(dev); } if (!mod_timer(&dev->watchdog_timer, - round_jiffies(jiffies + + round_jiffies(oldest_start + dev->watchdog_timeo))) release = false; } -- cgit v1.2.3 From 383eed2de529287337d9153a5084d4291a7c69f2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 May 2024 16:41:40 +0000 Subject: tcp: get rid of twsk_unique() DCCP is going away soon, and had no twsk_unique() method. We can directly call tcp_twsk_unique() for TCP sockets. Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240507164140.940547-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- include/net/timewait_sock.h | 9 --------- net/ipv4/inet_hashtables.c | 3 ++- net/ipv4/tcp_ipv4.c | 1 - net/ipv6/inet6_hashtables.c | 4 +++- net/ipv6/tcp_ipv6.c | 1 - 5 files changed, 5 insertions(+), 13 deletions(-) (limited to 'net') diff --git a/include/net/timewait_sock.h b/include/net/timewait_sock.h index 74d2b463cc95..62b3e9f2aed4 100644 --- a/include/net/timewait_sock.h +++ b/include/net/timewait_sock.h @@ -15,18 +15,9 @@ struct timewait_sock_ops { struct kmem_cache *twsk_slab; char *twsk_slab_name; unsigned int twsk_obj_size; - int (*twsk_unique)(struct sock *sk, - struct sock *sktw, void *twp); void (*twsk_destructor)(struct sock *sk); }; -static inline int twsk_unique(struct sock *sk, struct sock *sktw, void *twp) -{ - if (sk->sk_prot->twsk_prot->twsk_unique != NULL) - return sk->sk_prot->twsk_prot->twsk_unique(sk, sktw, twp); - return 0; -} - static inline void twsk_destructor(struct sock *sk) { if (sk->sk_prot->twsk_prot->twsk_destructor != NULL) diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index cf88eca5f1b4..48d0d494185b 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -565,7 +565,8 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, if (likely(inet_match(net, sk2, acookie, ports, dif, sdif))) { if (sk2->sk_state == TCP_TIME_WAIT) { tw = inet_twsk(sk2); - if (twsk_unique(sk, sk2, twp)) + if (sk->sk_protocol == IPPROTO_TCP && + tcp_twsk_unique(sk, sk2, twp)) break; } goto not_unique; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 95e3d28b83b8..108a438dc247 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2437,7 +2437,6 @@ do_time_wait: static struct timewait_sock_ops tcp_timewait_sock_ops = { .twsk_obj_size = sizeof(struct tcp_timewait_sock), - .twsk_unique = tcp_twsk_unique, .twsk_destructor= tcp_twsk_destructor, }; diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 2e81383b663b..6db71bb1cd30 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -21,6 +21,7 @@ #include #include #include +#include u32 inet6_ehashfn(const struct net *net, const struct in6_addr *laddr, const u16 lport, @@ -289,7 +290,8 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, dif, sdif))) { if (sk2->sk_state == TCP_TIME_WAIT) { tw = inet_twsk(sk2); - if (twsk_unique(sk, sk2, twp)) + if (sk->sk_protocol == IPPROTO_TCP && + tcp_twsk_unique(sk, sk2, twp)) break; } goto not_unique; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 37201c4fb393..7f6693e794bd 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2049,7 +2049,6 @@ void tcp_v6_early_demux(struct sk_buff *skb) static struct timewait_sock_ops tcp6_timewait_sock_ops = { .twsk_obj_size = sizeof(struct tcp6_timewait_sock), - .twsk_unique = tcp_twsk_unique, .twsk_destructor = tcp_twsk_destructor, }; -- cgit v1.2.3 From 1cd354fe1e4864eeaff62f66ee513080ec946f20 Mon Sep 17 00:00:00 2001 From: gaoxingwang Date: Mon, 22 Apr 2024 17:19:17 +0800 Subject: net: ipv6: fix wrong start position when receive hop-by-hop fragment In IPv6, ipv6_rcv_core will parse the hop-by-hop type extension header and increase skb->transport_header by one extension header length. But if there are more other extension headers like fragment header at this time, the skb->transport_header points to the second extension header, not the transport layer header or the first extension header. This will result in the start and nexthdrp variable not pointing to the same position in ipv6frag_thdr_trunced, and ipv6_skip_exthdr returning incorrect offset and frag_off.Sometimes,the length of the last sharded packet is smaller than the calculated incorrect offset, resulting in packet loss. We can use network header to offset and calculate the correct position to solve this problem. Fixes: 9d9e937b1c8b (ipv6/netfilter: Discard first fragment not including all headers) Signed-off-by: Gao Xingwang Signed-off-by: David S. Miller --- net/ipv6/reassembly.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 439f93512b0a..327caca64257 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -369,7 +369,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb) * the source of the fragment, with the Pointer field set to zero. */ nexthdr = hdr->nexthdr; - if (ipv6frag_thdr_truncated(skb, skb_transport_offset(skb), &nexthdr)) { + if (ipv6frag_thdr_truncated(skb, skb_network_offset(skb) + sizeof(struct ipv6hdr), &nexthdr)) { __IP6_INC_STATS(net, __in6_dev_get_safely(skb->dev), IPSTATS_MIB_INHDRERRORS); icmpv6_param_prob(skb, ICMPV6_HDR_INCOMP, 0); -- cgit v1.2.3 From a238106703ab4ae1090b86eba128815b8626d8f1 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:45 +0200 Subject: netfilter: nft_set_pipapo: prepare pipapo_get helper for on-demand clone The helper uses priv->clone unconditionally which will fail once we do the clone conditionally on first insert or removal. 'nft get element' from userspace needs to use priv->match since this runs from rcu read side lock section. Prepare for this by passing the match backend data as argument. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index 7c11f568069c..6657aa34f4d7 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -504,6 +504,7 @@ out: * pipapo_get() - Get matching element reference given key data * @net: Network namespace * @set: nftables API set representation + * @m: storage containing active/existing elements * @data: Key data to be matched against existing elements * @genmask: If set, check that element is active in given genmask * @tstamp: timestamp to check for expired elements @@ -517,17 +518,15 @@ out: */ static struct nft_pipapo_elem *pipapo_get(const struct net *net, const struct nft_set *set, + const struct nft_pipapo_match *m, const u8 *data, u8 genmask, u64 tstamp, gfp_t gfp) { struct nft_pipapo_elem *ret = ERR_PTR(-ENOENT); - struct nft_pipapo *priv = nft_set_priv(set); unsigned long *res_map, *fill_map = NULL; - const struct nft_pipapo_match *m; const struct nft_pipapo_field *f; int i; - m = priv->clone; if (m->bsize_max == 0) return ret; @@ -612,9 +611,11 @@ static struct nft_elem_priv * nft_pipapo_get(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem, unsigned int flags) { + struct nft_pipapo *priv = nft_set_priv(set); + struct nft_pipapo_match *m = rcu_dereference(priv->match); struct nft_pipapo_elem *e; - e = pipapo_get(net, set, (const u8 *)elem->key.val.data, + e = pipapo_get(net, set, m, (const u8 *)elem->key.val.data, nft_genmask_cur(net), get_jiffies_64(), GFP_ATOMIC); if (IS_ERR(e)) @@ -1288,7 +1289,7 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, else end = start; - dup = pipapo_get(net, set, start, genmask, tstamp, GFP_KERNEL); + dup = pipapo_get(net, set, m, start, genmask, tstamp, GFP_KERNEL); if (!IS_ERR(dup)) { /* Check if we already have the same exact entry */ const struct nft_data *dup_key, *dup_end; @@ -1310,7 +1311,7 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, if (PTR_ERR(dup) == -ENOENT) { /* Look for partially overlapping entries */ - dup = pipapo_get(net, set, end, nft_genmask_next(net), tstamp, + dup = pipapo_get(net, set, m, end, nft_genmask_next(net), tstamp, GFP_KERNEL); } @@ -1862,9 +1863,11 @@ static struct nft_elem_priv * nft_pipapo_deactivate(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem) { + const struct nft_pipapo *priv = nft_set_priv(set); + struct nft_pipapo_match *m = priv->clone; struct nft_pipapo_elem *e; - e = pipapo_get(net, set, (const u8 *)elem->key.val.data, + e = pipapo_get(net, set, m, (const u8 *)elem->key.val.data, nft_genmask_next(net), nft_net_tstamp(net), GFP_KERNEL); if (IS_ERR(e)) return NULL; -- cgit v1.2.3 From 3f1d886cc7c3525d4dbeee24bfa9bb3fe0d48ddc Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:46 +0200 Subject: netfilter: nft_set_pipapo: move cloning of match info to insert/removal path This set type keeps two copies of the sets' content, priv->match (live version, used to match from packet path) priv->clone (work-in-progress version of the 'future' priv->match). All additions and removals are done on priv->clone. When transaction completes, priv->clone becomes priv->match and a new clone is allocated for use by next transaction. Problem is that the cloning requires GFP_KERNEL allocations but we cannot fail at either commit or abort time. This patch defers the clone until we get an insertion or removal request. This allows us to handle OOM situations correctly. This also allows to remove ->dirty in a followup change: If ->clone exists, ->dirty is always true If ->clone is NULL, ->dirty is always false, no elements were added or removed (except catchall elements which are external to the specific set backend). Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 70 +++++++++++++++++++++++++++++------------- 1 file changed, 49 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index 6657aa34f4d7..dd9696120ea4 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -1259,6 +1259,29 @@ static bool nft_pipapo_transaction_mutex_held(const struct nft_set *set) #endif } +static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old); + +/** + * pipapo_maybe_clone() - Build clone for pending data changes, if not existing + * @set: nftables API set representation + * + * Return: newly created or existing clone, if any. NULL on allocation failure + */ +static struct nft_pipapo_match *pipapo_maybe_clone(const struct nft_set *set) +{ + struct nft_pipapo *priv = nft_set_priv(set); + struct nft_pipapo_match *m; + + if (priv->clone) + return priv->clone; + + m = rcu_dereference_protected(priv->match, + nft_pipapo_transaction_mutex_held(set)); + priv->clone = pipapo_clone(m); + + return priv->clone; +} + /** * nft_pipapo_insert() - Validate and insert ranged elements * @net: Network namespace @@ -1275,8 +1298,8 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv); union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS]; const u8 *start = (const u8 *)elem->key.val.data, *end; + struct nft_pipapo_match *m = pipapo_maybe_clone(set); struct nft_pipapo *priv = nft_set_priv(set); - struct nft_pipapo_match *m = priv->clone; u8 genmask = nft_genmask_next(net); struct nft_pipapo_elem *e, *dup; u64 tstamp = nft_net_tstamp(net); @@ -1284,6 +1307,9 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, const u8 *start_p, *end_p; int i, bsize_max, err = 0; + if (!m) + return -ENOMEM; + if (nft_set_ext_exists(ext, NFT_SET_EXT_KEY_END)) end = (const u8 *)nft_set_ext_key_end(ext)->data; else @@ -1789,7 +1815,10 @@ static void pipapo_reclaim_match(struct rcu_head *rcu) static void nft_pipapo_commit(struct nft_set *set) { struct nft_pipapo *priv = nft_set_priv(set); - struct nft_pipapo_match *new_clone, *old; + struct nft_pipapo_match *old; + + if (!priv->clone) + return; if (time_after_eq(jiffies, priv->last_gc + nft_set_gc_interval(set))) pipapo_gc(set, priv->clone); @@ -1797,38 +1826,27 @@ static void nft_pipapo_commit(struct nft_set *set) if (!priv->dirty) return; - new_clone = pipapo_clone(priv->clone); - if (!new_clone) - return; - + old = rcu_replace_pointer(priv->match, priv->clone, + nft_pipapo_transaction_mutex_held(set)); + priv->clone = NULL; priv->dirty = false; - old = rcu_access_pointer(priv->match); - rcu_assign_pointer(priv->match, priv->clone); if (old) call_rcu(&old->rcu, pipapo_reclaim_match); - - priv->clone = new_clone; } static void nft_pipapo_abort(const struct nft_set *set) { struct nft_pipapo *priv = nft_set_priv(set); - struct nft_pipapo_match *new_clone, *m; if (!priv->dirty) return; - m = rcu_dereference_protected(priv->match, nft_pipapo_transaction_mutex_held(set)); - - new_clone = pipapo_clone(m); - if (!new_clone) + if (!priv->clone) return; - priv->dirty = false; - pipapo_free_match(priv->clone); - priv->clone = new_clone; + priv->clone = NULL; } /** @@ -1863,10 +1881,15 @@ static struct nft_elem_priv * nft_pipapo_deactivate(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem) { - const struct nft_pipapo *priv = nft_set_priv(set); - struct nft_pipapo_match *m = priv->clone; + struct nft_pipapo_match *m = pipapo_maybe_clone(set); struct nft_pipapo_elem *e; + /* removal must occur on priv->clone, if we are low on memory + * we have no choice and must fail the removal request. + */ + if (!m) + return NULL; + e = pipapo_get(net, set, m, (const u8 *)elem->key.val.data, nft_genmask_next(net), nft_net_tstamp(net), GFP_KERNEL); if (IS_ERR(e)) @@ -2145,7 +2168,12 @@ static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set, switch (iter->type) { case NFT_ITER_UPDATE: - m = priv->clone; + m = pipapo_maybe_clone(set); + if (!m) { + iter->err = -ENOMEM; + return; + } + nft_pipapo_do_walk(ctx, set, m, iter); break; case NFT_ITER_READ: -- cgit v1.2.3 From 532aec7e878b527fcee8877350ab5c5341789626 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 25 Apr 2024 14:06:47 +0200 Subject: netfilter: nft_set_pipapo: remove dirty flag After previous change: ->clone exists: ->dirty is always true ->clone == NULL ->dirty is always false So remove this flag. Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo.c | 25 ------------------------- net/netfilter/nft_set_pipapo.h | 2 -- 2 files changed, 27 deletions(-) (limited to 'net') diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c index dd9696120ea4..15a236bebb46 100644 --- a/net/netfilter/nft_set_pipapo.c +++ b/net/netfilter/nft_set_pipapo.c @@ -1299,7 +1299,6 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS]; const u8 *start = (const u8 *)elem->key.val.data, *end; struct nft_pipapo_match *m = pipapo_maybe_clone(set); - struct nft_pipapo *priv = nft_set_priv(set); u8 genmask = nft_genmask_next(net); struct nft_pipapo_elem *e, *dup; u64 tstamp = nft_net_tstamp(net); @@ -1370,8 +1369,6 @@ static int nft_pipapo_insert(const struct net *net, const struct nft_set *set, } /* Insert */ - priv->dirty = true; - bsize_max = m->bsize_max; nft_pipapo_for_each_field(f, i, m) { @@ -1736,8 +1733,6 @@ static void pipapo_gc(struct nft_set *set, struct nft_pipapo_match *m) * NFT_SET_ELEM_DEAD_BIT. */ if (__nft_set_elem_expired(&e->ext, tstamp)) { - priv->dirty = true; - gc = nft_trans_gc_queue_sync(gc, GFP_KERNEL); if (!gc) return; @@ -1823,13 +1818,9 @@ static void nft_pipapo_commit(struct nft_set *set) if (time_after_eq(jiffies, priv->last_gc + nft_set_gc_interval(set))) pipapo_gc(set, priv->clone); - if (!priv->dirty) - return; - old = rcu_replace_pointer(priv->match, priv->clone, nft_pipapo_transaction_mutex_held(set)); priv->clone = NULL; - priv->dirty = false; if (old) call_rcu(&old->rcu, pipapo_reclaim_match); @@ -1839,12 +1830,8 @@ static void nft_pipapo_abort(const struct nft_set *set) { struct nft_pipapo *priv = nft_set_priv(set); - if (!priv->dirty) - return; - if (!priv->clone) return; - priv->dirty = false; pipapo_free_match(priv->clone); priv->clone = NULL; } @@ -2098,7 +2085,6 @@ static void nft_pipapo_remove(const struct net *net, const struct nft_set *set, match_end += NFT_PIPAPO_GROUPS_PADDED_SIZE(f); if (last && f->mt[rulemap[i].to].e == e) { - priv->dirty = true; pipapo_drop(m, rulemap); return; } @@ -2295,21 +2281,10 @@ static int nft_pipapo_init(const struct nft_set *set, f->mt = NULL; } - /* Create an initial clone of matching data for next insertion */ - priv->clone = pipapo_clone(m); - if (!priv->clone) { - err = -ENOMEM; - goto out_free; - } - - priv->dirty = false; - rcu_assign_pointer(priv->match, m); return 0; -out_free: - free_percpu(m->scratch); out_scratch: kfree(m); diff --git a/net/netfilter/nft_set_pipapo.h b/net/netfilter/nft_set_pipapo.h index 24cd1ff73f98..0d2e40e10f7f 100644 --- a/net/netfilter/nft_set_pipapo.h +++ b/net/netfilter/nft_set_pipapo.h @@ -155,14 +155,12 @@ struct nft_pipapo_match { * @match: Currently in-use matching data * @clone: Copy where pending insertions and deletions are kept * @width: Total bytes to be matched for one packet, including padding - * @dirty: Working copy has pending insertions or deletions * @last_gc: Timestamp of last garbage collection run, jiffies */ struct nft_pipapo { struct nft_pipapo_match __rcu *match; struct nft_pipapo_match *clone; int width; - bool dirty; unsigned long last_gc; }; -- cgit v1.2.3 From fa23e0d4b756d25829e124d6b670a4c6bbd4bf7e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 8 May 2024 14:52:47 +0200 Subject: netfilter: nf_tables: allow clone callbacks to sleep Sven Auhagen reports transaction failures with following error: ./main.nft:13:1-26: Error: Could not process rule: Cannot allocate memory percpu: allocation failed, size=16 align=8 atomic=1, atomic alloc failed, no space left This points to failing pcpu allocation with GFP_ATOMIC flag. However, transactions happen from user context and are allowed to sleep. One case where we can call into percpu allocator with GFP_ATOMIC is nft_counter expression. Normally this happens from control plane, so this could use GFP_KERNEL instead. But one use case, element insertion from packet path, needs to use GFP_ATOMIC allocations (nft_dynset expression). At this time, .clone callbacks always use GFP_ATOMIC for this reason. Add gfp_t argument to the .clone function and pass GFP_KERNEL or GFP_ATOMIC flag depending on context, this allows all clone memory allocations to sleep for the normal (transaction) case. Cc: Sven Auhagen Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 4 ++-- net/netfilter/nf_tables_api.c | 8 ++++---- net/netfilter/nft_connlimit.c | 4 ++-- net/netfilter/nft_counter.c | 4 ++-- net/netfilter/nft_dynset.c | 2 +- net/netfilter/nft_last.c | 4 ++-- net/netfilter/nft_limit.c | 14 ++++++++------ net/netfilter/nft_quota.c | 4 ++-- 8 files changed, 23 insertions(+), 21 deletions(-) (limited to 'net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 3f1ed467f951..2796153b03da 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -416,7 +416,7 @@ struct nft_expr_info; int nft_expr_inner_parse(const struct nft_ctx *ctx, const struct nlattr *nla, struct nft_expr_info *info); -int nft_expr_clone(struct nft_expr *dst, struct nft_expr *src); +int nft_expr_clone(struct nft_expr *dst, struct nft_expr *src, gfp_t gfp); void nft_expr_destroy(const struct nft_ctx *ctx, struct nft_expr *expr); int nft_expr_dump(struct sk_buff *skb, unsigned int attr, const struct nft_expr *expr, bool reset); @@ -935,7 +935,7 @@ struct nft_expr_ops { struct nft_regs *regs, const struct nft_pktinfo *pkt); int (*clone)(struct nft_expr *dst, - const struct nft_expr *src); + const struct nft_expr *src, gfp_t gfp); unsigned int size; int (*init)(const struct nft_ctx *ctx, diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index a7f54eb68d9a..be3b4c90d2ed 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3333,7 +3333,7 @@ err_expr_parse: return ERR_PTR(err); } -int nft_expr_clone(struct nft_expr *dst, struct nft_expr *src) +int nft_expr_clone(struct nft_expr *dst, struct nft_expr *src, gfp_t gfp) { int err; @@ -3341,7 +3341,7 @@ int nft_expr_clone(struct nft_expr *dst, struct nft_expr *src) return -EINVAL; dst->ops = src->ops; - err = src->ops->clone(dst, src); + err = src->ops->clone(dst, src, gfp); if (err < 0) return err; @@ -6525,7 +6525,7 @@ int nft_set_elem_expr_clone(const struct nft_ctx *ctx, struct nft_set *set, if (!expr) goto err_expr; - err = nft_expr_clone(expr, set->exprs[i]); + err = nft_expr_clone(expr, set->exprs[i], GFP_KERNEL_ACCOUNT); if (err < 0) { kfree(expr); goto err_expr; @@ -6564,7 +6564,7 @@ static int nft_set_elem_expr_setup(struct nft_ctx *ctx, for (i = 0; i < num_exprs; i++) { expr = nft_setelem_expr_at(elem_expr, elem_expr->size); - err = nft_expr_clone(expr, expr_array[i]); + err = nft_expr_clone(expr, expr_array[i], GFP_KERNEL_ACCOUNT); if (err < 0) goto err_elem_expr_setup; diff --git a/net/netfilter/nft_connlimit.c b/net/netfilter/nft_connlimit.c index de9d1980df69..92b984fa8175 100644 --- a/net/netfilter/nft_connlimit.c +++ b/net/netfilter/nft_connlimit.c @@ -210,12 +210,12 @@ static void nft_connlimit_destroy(const struct nft_ctx *ctx, nft_connlimit_do_destroy(ctx, priv); } -static int nft_connlimit_clone(struct nft_expr *dst, const struct nft_expr *src) +static int nft_connlimit_clone(struct nft_expr *dst, const struct nft_expr *src, gfp_t gfp) { struct nft_connlimit *priv_dst = nft_expr_priv(dst); struct nft_connlimit *priv_src = nft_expr_priv(src); - priv_dst->list = kmalloc(sizeof(*priv_dst->list), GFP_ATOMIC); + priv_dst->list = kmalloc(sizeof(*priv_dst->list), gfp); if (!priv_dst->list) return -ENOMEM; diff --git a/net/netfilter/nft_counter.c b/net/netfilter/nft_counter.c index dccc68a5135a..291ed2026367 100644 --- a/net/netfilter/nft_counter.c +++ b/net/netfilter/nft_counter.c @@ -226,7 +226,7 @@ static void nft_counter_destroy(const struct nft_ctx *ctx, nft_counter_do_destroy(priv); } -static int nft_counter_clone(struct nft_expr *dst, const struct nft_expr *src) +static int nft_counter_clone(struct nft_expr *dst, const struct nft_expr *src, gfp_t gfp) { struct nft_counter_percpu_priv *priv = nft_expr_priv(src); struct nft_counter_percpu_priv *priv_clone = nft_expr_priv(dst); @@ -236,7 +236,7 @@ static int nft_counter_clone(struct nft_expr *dst, const struct nft_expr *src) nft_counter_fetch(priv, &total); - cpu_stats = alloc_percpu_gfp(struct nft_counter, GFP_ATOMIC); + cpu_stats = alloc_percpu_gfp(struct nft_counter, gfp); if (cpu_stats == NULL) return -ENOMEM; diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c index c09dba57354c..b4ada3ab2167 100644 --- a/net/netfilter/nft_dynset.c +++ b/net/netfilter/nft_dynset.c @@ -35,7 +35,7 @@ static int nft_dynset_expr_setup(const struct nft_dynset *priv, for (i = 0; i < priv->num_exprs; i++) { expr = nft_setelem_expr_at(elem_expr, elem_expr->size); - if (nft_expr_clone(expr, priv->expr_array[i]) < 0) + if (nft_expr_clone(expr, priv->expr_array[i], GFP_ATOMIC) < 0) return -1; elem_expr->size += priv->expr_array[i]->ops->size; diff --git a/net/netfilter/nft_last.c b/net/netfilter/nft_last.c index 8e6d7eaf9dc8..de1b6066bfa8 100644 --- a/net/netfilter/nft_last.c +++ b/net/netfilter/nft_last.c @@ -102,12 +102,12 @@ static void nft_last_destroy(const struct nft_ctx *ctx, kfree(priv->last); } -static int nft_last_clone(struct nft_expr *dst, const struct nft_expr *src) +static int nft_last_clone(struct nft_expr *dst, const struct nft_expr *src, gfp_t gfp) { struct nft_last_priv *priv_dst = nft_expr_priv(dst); struct nft_last_priv *priv_src = nft_expr_priv(src); - priv_dst->last = kzalloc(sizeof(*priv_dst->last), GFP_ATOMIC); + priv_dst->last = kzalloc(sizeof(*priv_dst->last), gfp); if (!priv_dst->last) return -ENOMEM; diff --git a/net/netfilter/nft_limit.c b/net/netfilter/nft_limit.c index cefa25e0dbb0..21d26b79b460 100644 --- a/net/netfilter/nft_limit.c +++ b/net/netfilter/nft_limit.c @@ -150,7 +150,7 @@ static void nft_limit_destroy(const struct nft_ctx *ctx, } static int nft_limit_clone(struct nft_limit_priv *priv_dst, - const struct nft_limit_priv *priv_src) + const struct nft_limit_priv *priv_src, gfp_t gfp) { priv_dst->tokens_max = priv_src->tokens_max; priv_dst->rate = priv_src->rate; @@ -158,7 +158,7 @@ static int nft_limit_clone(struct nft_limit_priv *priv_dst, priv_dst->burst = priv_src->burst; priv_dst->invert = priv_src->invert; - priv_dst->limit = kmalloc(sizeof(*priv_dst->limit), GFP_ATOMIC); + priv_dst->limit = kmalloc(sizeof(*priv_dst->limit), gfp); if (!priv_dst->limit) return -ENOMEM; @@ -223,14 +223,15 @@ static void nft_limit_pkts_destroy(const struct nft_ctx *ctx, nft_limit_destroy(ctx, &priv->limit); } -static int nft_limit_pkts_clone(struct nft_expr *dst, const struct nft_expr *src) +static int nft_limit_pkts_clone(struct nft_expr *dst, const struct nft_expr *src, + gfp_t gfp) { struct nft_limit_priv_pkts *priv_dst = nft_expr_priv(dst); struct nft_limit_priv_pkts *priv_src = nft_expr_priv(src); priv_dst->cost = priv_src->cost; - return nft_limit_clone(&priv_dst->limit, &priv_src->limit); + return nft_limit_clone(&priv_dst->limit, &priv_src->limit, gfp); } static struct nft_expr_type nft_limit_type; @@ -281,12 +282,13 @@ static void nft_limit_bytes_destroy(const struct nft_ctx *ctx, nft_limit_destroy(ctx, priv); } -static int nft_limit_bytes_clone(struct nft_expr *dst, const struct nft_expr *src) +static int nft_limit_bytes_clone(struct nft_expr *dst, const struct nft_expr *src, + gfp_t gfp) { struct nft_limit_priv *priv_dst = nft_expr_priv(dst); struct nft_limit_priv *priv_src = nft_expr_priv(src); - return nft_limit_clone(priv_dst, priv_src); + return nft_limit_clone(priv_dst, priv_src, gfp); } static const struct nft_expr_ops nft_limit_bytes_ops = { diff --git a/net/netfilter/nft_quota.c b/net/netfilter/nft_quota.c index 3ba12a7471b0..9b2d7463d3d3 100644 --- a/net/netfilter/nft_quota.c +++ b/net/netfilter/nft_quota.c @@ -233,7 +233,7 @@ static void nft_quota_destroy(const struct nft_ctx *ctx, return nft_quota_do_destroy(ctx, priv); } -static int nft_quota_clone(struct nft_expr *dst, const struct nft_expr *src) +static int nft_quota_clone(struct nft_expr *dst, const struct nft_expr *src, gfp_t gfp) { struct nft_quota *priv_dst = nft_expr_priv(dst); struct nft_quota *priv_src = nft_expr_priv(src); @@ -241,7 +241,7 @@ static int nft_quota_clone(struct nft_expr *dst, const struct nft_expr *src) priv_dst->quota = priv_src->quota; priv_dst->flags = priv_src->flags; - priv_dst->consumed = kmalloc(sizeof(*priv_dst->consumed), GFP_ATOMIC); + priv_dst->consumed = kmalloc(sizeof(*priv_dst->consumed), gfp); if (!priv_dst->consumed) return -ENOMEM; -- cgit v1.2.3 From 7172dc93d621d5dc302d007e95ddd1311ec64283 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Wed, 8 May 2024 10:11:50 -0700 Subject: af_unix: Add dead flag to struct scm_fp_list. Commit 1af2dface5d2 ("af_unix: Don't access successor in unix_del_edges() during GC.") fixed use-after-free by avoid accessing edge->successor while GC is in progress. However, there could be a small race window where another process could call unix_del_edges() while gc_in_progress is true and __skb_queue_purge() is on the way. So, we need another marker for struct scm_fp_list which indicates if the skb is garbage-collected. This patch adds dead flag in struct scm_fp_list and set it true before calling __skb_queue_purge(). Fixes: 1af2dface5d2 ("af_unix: Don't access successor in unix_del_edges() during GC.") Signed-off-by: Kuniyuki Iwashima Acked-by: Paolo Abeni Link: https://lore.kernel.org/r/20240508171150.50601-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski --- include/net/scm.h | 1 + net/core/scm.c | 1 + net/unix/garbage.c | 14 ++++++++++---- 3 files changed, 12 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/include/net/scm.h b/include/net/scm.h index bbc5527809d1..0d35c7c77a74 100644 --- a/include/net/scm.h +++ b/include/net/scm.h @@ -33,6 +33,7 @@ struct scm_fp_list { short max; #ifdef CONFIG_UNIX bool inflight; + bool dead; struct list_head vertices; struct unix_edge *edges; #endif diff --git a/net/core/scm.c b/net/core/scm.c index 5763f3320358..4f6a14babe5a 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -91,6 +91,7 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp) fpl->user = NULL; #if IS_ENABLED(CONFIG_UNIX) fpl->inflight = false; + fpl->dead = false; fpl->edges = NULL; INIT_LIST_HEAD(&fpl->vertices); #endif diff --git a/net/unix/garbage.c b/net/unix/garbage.c index d76450133e4f..1f8b8cdfcdc8 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -158,13 +158,11 @@ static void unix_add_edge(struct scm_fp_list *fpl, struct unix_edge *edge) unix_update_graph(unix_edge_successor(edge)); } -static bool gc_in_progress; - static void unix_del_edge(struct scm_fp_list *fpl, struct unix_edge *edge) { struct unix_vertex *vertex = edge->predecessor->vertex; - if (!gc_in_progress) + if (!fpl->dead) unix_update_graph(unix_edge_successor(edge)); list_del(&edge->vertex_entry); @@ -240,7 +238,7 @@ void unix_del_edges(struct scm_fp_list *fpl) unix_del_edge(fpl, edge); } while (i < fpl->count_unix); - if (!gc_in_progress) { + if (!fpl->dead) { receiver = fpl->edges[0].successor; receiver->scm_stat.nr_unix_fds -= fpl->count_unix; } @@ -559,9 +557,12 @@ static void unix_walk_scc_fast(struct sk_buff_head *hitlist) list_replace_init(&unix_visited_vertices, &unix_unvisited_vertices); } +static bool gc_in_progress; + static void __unix_gc(struct work_struct *work) { struct sk_buff_head hitlist; + struct sk_buff *skb; spin_lock(&unix_gc_lock); @@ -579,6 +580,11 @@ static void __unix_gc(struct work_struct *work) spin_unlock(&unix_gc_lock); + skb_queue_walk(&hitlist, skb) { + if (UNIXCB(skb).fp) + UNIXCB(skb).fp->dead = true; + } + __skb_queue_purge(&hitlist); skip_gc: WRITE_ONCE(gc_in_progress, false); -- cgit v1.2.3 From 540bf24fba16b88c1b3b9353927204b4f1074e25 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 9 May 2024 01:14:46 -0700 Subject: af_unix: Fix data races in unix_release_sock/unix_stream_sendmsg A data-race condition has been identified in af_unix. In one data path, the write function unix_release_sock() atomically writes to sk->sk_shutdown using WRITE_ONCE. However, on the reader side, unix_stream_sendmsg() does not read it atomically. Consequently, this issue is causing the following KCSAN splat to occur: BUG: KCSAN: data-race in unix_release_sock / unix_stream_sendmsg write (marked) to 0xffff88867256ddbb of 1 bytes by task 7270 on cpu 28: unix_release_sock (net/unix/af_unix.c:640) unix_release (net/unix/af_unix.c:1050) sock_close (net/socket.c:659 net/socket.c:1421) __fput (fs/file_table.c:422) __fput_sync (fs/file_table.c:508) __se_sys_close (fs/open.c:1559 fs/open.c:1541) __x64_sys_close (fs/open.c:1541) x64_sys_call (arch/x86/entry/syscall_64.c:33) do_syscall_64 (arch/x86/entry/common.c:?) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) read to 0xffff88867256ddbb of 1 bytes by task 989 on cpu 14: unix_stream_sendmsg (net/unix/af_unix.c:2273) __sock_sendmsg (net/socket.c:730 net/socket.c:745) ____sys_sendmsg (net/socket.c:2584) __sys_sendmmsg (net/socket.c:2638 net/socket.c:2724) __x64_sys_sendmmsg (net/socket.c:2753 net/socket.c:2750 net/socket.c:2750) x64_sys_call (arch/x86/entry/syscall_64.c:33) do_syscall_64 (arch/x86/entry/common.c:?) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) value changed: 0x01 -> 0x03 The line numbers are related to commit dd5a440a31fa ("Linux 6.9-rc7"). Commit e1d09c2c2f57 ("af_unix: Fix data races around sk->sk_shutdown.") addressed a comparable issue in the past regarding sk->sk_shutdown. However, it overlooked resolving this particular data path. This patch only offending unix_stream_sendmsg() function, since the other reads seem to be protected by unix_state_lock() as discussed in Link: https://lore.kernel.org/all/20240508173324.53565-1-kuniyu@amazon.com/ Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Breno Leitao Reviewed-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240509081459.2807828-1-leitao@debian.org Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 9a6ad5974dff..e94839d89b09 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2270,7 +2270,7 @@ static int unix_stream_sendmsg(struct socket *sock, struct msghdr *msg, goto out_err; } - if (sk->sk_shutdown & SEND_SHUTDOWN) + if (READ_ONCE(sk->sk_shutdown) & SEND_SHUTDOWN) goto pipe_err; while (sent < len) { -- cgit v1.2.3 From 7c988176b6c16c516474f6fceebe0f055af5eb56 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Thu, 9 May 2024 11:38:05 +0200 Subject: net: openvswitch: fix overwriting ct original tuple for ICMPv6 OVS_PACKET_CMD_EXECUTE has 3 main attributes: - OVS_PACKET_ATTR_KEY - Packet metadata in a netlink format. - OVS_PACKET_ATTR_PACKET - Binary packet content. - OVS_PACKET_ATTR_ACTIONS - Actions to execute on the packet. OVS_PACKET_ATTR_KEY is parsed first to populate sw_flow_key structure with the metadata like conntrack state, input port, recirculation id, etc. Then the packet itself gets parsed to populate the rest of the keys from the packet headers. Whenever the packet parsing code starts parsing the ICMPv6 header, it first zeroes out fields in the key corresponding to Neighbor Discovery information even if it is not an ND packet. It is an 'ipv6.nd' field. However, the 'ipv6' is a union that shares the space between 'nd' and 'ct_orig' that holds the original tuple conntrack metadata parsed from the OVS_PACKET_ATTR_KEY. ND packets should not normally have conntrack state, so it's fine to share the space, but normal ICMPv6 Echo packets or maybe other types of ICMPv6 can have the state attached and it should not be overwritten. The issue results in all but the last 4 bytes of the destination address being wiped from the original conntrack tuple leading to incorrect packet matching and potentially executing wrong actions in case this packet recirculates within the datapath or goes back to userspace. ND fields should not be accessed in non-ND packets, so not clearing them should be fine. Executing memset() only for actual ND packets to avoid the issue. Initializing the whole thing before parsing is needed because ND packet may not contain all the options. The issue only affects the OVS_PACKET_CMD_EXECUTE path and doesn't affect packets entering OVS datapath from network interfaces, because in this case CT metadata is populated from skb after the packet is already parsed. Fixes: 9dd7f8907c37 ("openvswitch: Add original direction conntrack tuple to sw_flow_key.") Reported-by: Antonin Bas Closes: https://github.com/openvswitch/ovs-issues/issues/327 Signed-off-by: Ilya Maximets Acked-by: Aaron Conole Acked-by: Eelco Chaudron Link: https://lore.kernel.org/r/20240509094228.1035477-1-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- net/openvswitch/flow.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 33b21a0c0548..8a848ce72e29 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -561,7 +561,6 @@ static int parse_icmpv6(struct sk_buff *skb, struct sw_flow_key *key, */ key->tp.src = htons(icmp->icmp6_type); key->tp.dst = htons(icmp->icmp6_code); - memset(&key->ipv6.nd, 0, sizeof(key->ipv6.nd)); if (icmp->icmp6_code == 0 && (icmp->icmp6_type == NDISC_NEIGHBOUR_SOLICITATION || @@ -570,6 +569,8 @@ static int parse_icmpv6(struct sk_buff *skb, struct sw_flow_key *key, struct nd_msg *nd; int offset; + memset(&key->ipv6.nd, 0, sizeof(key->ipv6.nd)); + /* In order to process neighbor discovery options, we need the * entire packet. */ -- cgit v1.2.3 From 3321687e321307629c71b664225b861ebf3e5753 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Thu, 9 May 2024 21:18:10 +0800 Subject: ipv6: sr: add missing seg6_local_exit Currently, we only call seg6_local_exit() in seg6_init() if seg6_local_init() failed. But forgot to call it in seg6_exit(). Fixes: d1df6fd8a1d2 ("ipv6: sr: define core operations for seg6local lightweight tunnel") Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240509131812.1662197-2-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/seg6.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index 35508abd76f4..5423f1f2aa62 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -564,6 +564,7 @@ void seg6_exit(void) seg6_hmac_exit(); #endif #ifdef CONFIG_IPV6_SEG6_LWTUNNEL + seg6_local_exit(); seg6_iptunnel_exit(); #endif unregister_pernet_subsys(&ip6_segments_ops); -- cgit v1.2.3 From 6e370a771d2985107e82d0f6174381c1acb49c20 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Thu, 9 May 2024 21:18:11 +0800 Subject: ipv6: sr: fix incorrect unregister order Commit 5559cea2d5aa ("ipv6: sr: fix possible use-after-free and null-ptr-deref") changed the register order in seg6_init(). But the unregister order in seg6_exit() is not updated. Fixes: 5559cea2d5aa ("ipv6: sr: fix possible use-after-free and null-ptr-deref") Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240509131812.1662197-3-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/seg6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index 5423f1f2aa62..c4ef96c8fdac 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -567,6 +567,6 @@ void seg6_exit(void) seg6_local_exit(); seg6_iptunnel_exit(); #endif - unregister_pernet_subsys(&ip6_segments_ops); genl_unregister_family(&seg6_genl_family); + unregister_pernet_subsys(&ip6_segments_ops); } -- cgit v1.2.3 From 160e9d2752181fcf18c662e74022d77d3164cd45 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Thu, 9 May 2024 21:18:12 +0800 Subject: ipv6: sr: fix invalid unregister error path The error path of seg6_init() is wrong in case CONFIG_IPV6_SEG6_LWTUNNEL is not defined. In that case if seg6_hmac_init() fails, the genl_unregister_family() isn't called. This issue exist since commit 46738b1317e1 ("ipv6: sr: add option to control lwtunnel support"), and commit 5559cea2d5aa ("ipv6: sr: fix possible use-after-free and null-ptr-deref") replaced unregister_pernet_subsys() with genl_unregister_family() in this error path. Fixes: 46738b1317e1 ("ipv6: sr: add option to control lwtunnel support") Reported-by: Guillaume Nault Signed-off-by: Hangbin Liu Reviewed-by: Sabrina Dubroca Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240509131812.1662197-4-liuhangbin@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/seg6.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index c4ef96c8fdac..a31521e270f7 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -551,6 +551,8 @@ out_unregister_iptun: #endif #ifdef CONFIG_IPV6_SEG6_LWTUNNEL out_unregister_genl: +#endif +#if IS_ENABLED(CONFIG_IPV6_SEG6_LWTUNNEL) || IS_ENABLED(CONFIG_IPV6_SEG6_HMAC) genl_unregister_family(&seg6_genl_family); #endif out_unregister_pernet: -- cgit v1.2.3 From 1af7f88af269c4e06a4dc3bc920ff6cdf7471124 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 May 2024 07:29:32 +0000 Subject: inet: fix inet_fill_ifaddr() flags truncation I missed that (struct ifaddrmsg)->ifa_flags was only 8bits, while (struct in_ifaddr)->ifa_flags is 32bits. Use a temporary 32bit variable as I did in set_ifa_lifetime() and check_lifetime(). Fixes: 3ddc2231c810 ("inet: annotate data-races around ifa->ifa_flags") Reported-by: Yu Watanabe Dianosed-by: Yu Watanabe Closes: https://github.com/systemd/systemd/pull/32666#issuecomment-2103977928 Signed-off-by: Eric Dumazet Reviewed-by: Larysa Zaremba Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240510072932.2678952-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/devinet.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 7a437f0d4190..7e45c34c8340 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1683,6 +1683,7 @@ static int inet_fill_ifaddr(struct sk_buff *skb, const struct in_ifaddr *ifa, struct nlmsghdr *nlh; unsigned long tstamp; u32 preferred, valid; + u32 flags; nlh = nlmsg_put(skb, args->portid, args->seq, args->event, sizeof(*ifm), args->flags); @@ -1692,7 +1693,13 @@ static int inet_fill_ifaddr(struct sk_buff *skb, const struct in_ifaddr *ifa, ifm = nlmsg_data(nlh); ifm->ifa_family = AF_INET; ifm->ifa_prefixlen = ifa->ifa_prefixlen; - ifm->ifa_flags = READ_ONCE(ifa->ifa_flags); + + flags = READ_ONCE(ifa->ifa_flags); + /* Warning : ifm->ifa_flags is an __u8, it holds only 8 bits. + * The 32bit value is given in IFA_FLAGS attribute. + */ + ifm->ifa_flags = (__u8)flags; + ifm->ifa_scope = ifa->ifa_scope; ifm->ifa_index = ifa->ifa_dev->dev->ifindex; @@ -1701,7 +1708,7 @@ static int inet_fill_ifaddr(struct sk_buff *skb, const struct in_ifaddr *ifa, goto nla_put_failure; tstamp = READ_ONCE(ifa->ifa_tstamp); - if (!(ifm->ifa_flags & IFA_F_PERMANENT)) { + if (!(flags & IFA_F_PERMANENT)) { preferred = READ_ONCE(ifa->ifa_preferred_lft); valid = READ_ONCE(ifa->ifa_valid_lft); if (preferred != INFINITY_LIFE_TIME) { @@ -1732,7 +1739,7 @@ static int inet_fill_ifaddr(struct sk_buff *skb, const struct in_ifaddr *ifa, nla_put_string(skb, IFA_LABEL, ifa->ifa_label)) || (ifa->ifa_proto && nla_put_u8(skb, IFA_PROTO, ifa->ifa_proto)) || - nla_put_u32(skb, IFA_FLAGS, ifm->ifa_flags) || + nla_put_u32(skb, IFA_FLAGS, flags) || (ifa->ifa_rt_priority && nla_put_u32(skb, IFA_RT_PRIORITY, ifa->ifa_rt_priority)) || put_cacheinfo(skb, READ_ONCE(ifa->ifa_cstamp), tstamp, -- cgit v1.2.3 From 186b1ea73ad8f30d1d7afdb1d07dfd5b5de8f2da Mon Sep 17 00:00:00 2001 From: Richard Gobert Date: Thu, 9 May 2024 21:08:17 +0200 Subject: net: gro: use cb instead of skb->network_header This patch converts references of skb->network_header to napi_gro_cb's network_offset and inner_network_offset. Signed-off-by: Richard Gobert Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/20240509190819.2985-2-richardbgobert@gmail.com Signed-off-by: Jakub Kicinski --- include/net/gro.h | 9 +++++++-- net/ipv4/af_inet.c | 4 ---- net/ipv4/tcp_offload.c | 3 ++- net/ipv6/ip6_offload.c | 5 ++--- net/ipv6/tcpv6_offload.c | 3 ++- 5 files changed, 13 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/net/gro.h b/include/net/gro.h index 5df8bf318197..cbc1b0aaf295 100644 --- a/include/net/gro.h +++ b/include/net/gro.h @@ -181,12 +181,17 @@ static inline void *skb_gro_header(struct sk_buff *skb, unsigned int hlen, return ptr; } +static inline int skb_gro_receive_network_offset(const struct sk_buff *skb) +{ + return NAPI_GRO_CB(skb)->network_offsets[NAPI_GRO_CB(skb)->encap_mark]; +} + static inline void *skb_gro_network_header(const struct sk_buff *skb) { if (skb_gro_may_pull(skb, skb_gro_offset(skb))) - return skb_gro_header_fast(skb, skb_network_offset(skb)); + return skb_gro_header_fast(skb, skb_gro_receive_network_offset(skb)); - return skb_network_header(skb); + return skb->data + skb_gro_receive_network_offset(skb); } static inline __wsum inet_gro_compute_pseudo(const struct sk_buff *skb, diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a7bad18bc8b5..428196e1541f 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1569,10 +1569,6 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) NAPI_GRO_CB(skb)->is_atomic = !!(iph->frag_off & htons(IP_DF)); NAPI_GRO_CB(skb)->flush |= flush; - skb_set_network_header(skb, off); - /* The above will be needed by the transport layer if there is one - * immediately following this IP hdr. - */ NAPI_GRO_CB(skb)->inner_network_offset = off; /* Note : No need to call skb_gro_postpull_rcsum() here, diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index c90704befd7b..2809667ac924 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -463,7 +463,8 @@ flush: INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff) { - const struct iphdr *iph = ip_hdr(skb); + const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation]; + const struct iphdr *iph = (struct iphdr *)(skb->data + offset); struct tcphdr *th = tcp_hdr(skb); if (unlikely(NAPI_GRO_CB(skb)->is_flist)) { diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c index c8b909a9904f..288c7c6ea50f 100644 --- a/net/ipv6/ip6_offload.c +++ b/net/ipv6/ip6_offload.c @@ -67,7 +67,7 @@ static int ipv6_gro_pull_exthdrs(struct sk_buff *skb, int off, int proto) off += len; } - skb_gro_pull(skb, off - skb_network_offset(skb)); + skb_gro_pull(skb, off - skb_gro_receive_network_offset(skb)); return proto; } @@ -236,7 +236,6 @@ INDIRECT_CALLABLE_SCOPE struct sk_buff *ipv6_gro_receive(struct list_head *head, if (unlikely(!iph)) goto out; - skb_set_network_header(skb, off); NAPI_GRO_CB(skb)->inner_network_offset = off; flush += ntohs(iph->payload_len) != skb->len - hlen; @@ -260,7 +259,7 @@ INDIRECT_CALLABLE_SCOPE struct sk_buff *ipv6_gro_receive(struct list_head *head, NAPI_GRO_CB(skb)->proto = proto; flush--; - nlen = skb_network_header_len(skb); + nlen = skb_gro_offset(skb) - off; list_for_each_entry(p, head, list) { const struct ipv6hdr *iph2; diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index d59f58cbd306..23971903e66d 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -72,7 +72,8 @@ flush: INDIRECT_CALLABLE_SCOPE int tcp6_gro_complete(struct sk_buff *skb, int thoff) { - const struct ipv6hdr *iph = ipv6_hdr(skb); + const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation]; + const struct ipv6hdr *iph = (struct ipv6hdr *)(skb->data + offset); struct tcphdr *th = tcp_hdr(skb); if (unlikely(NAPI_GRO_CB(skb)->is_flist)) { -- cgit v1.2.3 From 4b0ebbca3e1679765c06d5c466ee7f3228d4b156 Mon Sep 17 00:00:00 2001 From: Richard Gobert Date: Thu, 9 May 2024 21:08:18 +0200 Subject: net: gro: move L3 flush checks to tcp_gro_receive and udp_gro_receive_segment {inet,ipv6}_gro_receive functions perform flush checks (ttl, flags, iph->id, ...) against all packets in a loop. These flush checks are used in all merging UDP and TCP flows. These checks need to be done only once and only against the found p skb, since they only affect flush and not same_flow. This patch leverages correct network header offsets from the cb for both outer and inner network headers - allowing these checks to be done only once, in tcp_gro_receive and udp_gro_receive_segment. As a result, NAPI_GRO_CB(p)->flush is not used at all. In addition, flush_id checks are more declarative and contained in inet_gro_flush, thus removing the need for flush_id in napi_gro_cb. This results in less parsing code for non-loop flush tests for TCP and UDP flows. To make sure results are not within noise range - I've made netfilter drop all TCP packets, and measured CPU performance in GRO (in this case GRO is responsible for about 50% of the CPU utilization). perf top while replaying 64 parallel IP/TCP streams merging in GRO: (gro_receive_network_flush is compiled inline to tcp_gro_receive) net-next: 6.94% [kernel] [k] inet_gro_receive 3.02% [kernel] [k] tcp_gro_receive patch applied: 4.27% [kernel] [k] tcp_gro_receive 4.22% [kernel] [k] inet_gro_receive perf top while replaying 64 parallel IP/IP/TCP streams merging in GRO (same results for any encapsulation, in this case inet_gro_receive is top offender in net-next) net-next: 10.09% [kernel] [k] inet_gro_receive 2.08% [kernel] [k] tcp_gro_receive patch applied: 6.97% [kernel] [k] inet_gro_receive 3.68% [kernel] [k] tcp_gro_receive Signed-off-by: Richard Gobert Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/20240509190819.2985-3-richardbgobert@gmail.com Signed-off-by: Jakub Kicinski --- include/net/gro.h | 76 ++++++++++++++++++++++++++++++++++++++++++++------ net/core/gro.c | 3 -- net/ipv4/af_inet.c | 41 +-------------------------- net/ipv4/tcp_offload.c | 17 ++--------- net/ipv4/udp_offload.c | 9 +----- net/ipv6/ip6_offload.c | 11 -------- 6 files changed, 73 insertions(+), 84 deletions(-) (limited to 'net') diff --git a/include/net/gro.h b/include/net/gro.h index cbc1b0aaf295..f13634b1f4c1 100644 --- a/include/net/gro.h +++ b/include/net/gro.h @@ -36,15 +36,15 @@ struct napi_gro_cb { /* This is non-zero if the packet cannot be merged with the new skb. */ u16 flush; - /* Save the IP ID here and check when we get to the transport layer */ - u16 flush_id; - /* Number of segments aggregated. */ u16 count; /* Used in ipv6_gro_receive() and foo-over-udp and esp-in-udp */ u16 proto; + /* used to support CHECKSUM_COMPLETE for tunneling protocols */ + __wsum csum; + /* Used in napi_gro_cb::free */ #define NAPI_GRO_FREE 1 #define NAPI_GRO_FREE_STOLEN_HEAD 2 @@ -75,8 +75,8 @@ struct napi_gro_cb { /* Used in GRE, set in fou/gue_gro_receive */ u8 is_fou:1; - /* Used to determine if flush_id can be ignored */ - u8 is_atomic:1; + /* Used to determine if ipid_offset can be ignored */ + u8 ip_fixedid:1; /* Number of gro_receive callbacks this packet already went through */ u8 recursion_counter:4; @@ -85,9 +85,6 @@ struct napi_gro_cb { u8 is_flist:1; ); - /* used to support CHECKSUM_COMPLETE for tunneling protocols */ - __wsum csum; - /* L3 offsets */ union { struct { @@ -442,6 +439,69 @@ static inline __wsum ip6_gro_compute_pseudo(const struct sk_buff *skb, skb_gro_len(skb), proto, 0)); } +static inline int inet_gro_flush(const struct iphdr *iph, const struct iphdr *iph2, + struct sk_buff *p, bool outer) +{ + const u32 id = ntohl(*(__be32 *)&iph->id); + const u32 id2 = ntohl(*(__be32 *)&iph2->id); + const u16 ipid_offset = (id >> 16) - (id2 >> 16); + const u16 count = NAPI_GRO_CB(p)->count; + const u32 df = id & IP_DF; + int flush; + + /* All fields must match except length and checksum. */ + flush = (iph->ttl ^ iph2->ttl) | (iph->tos ^ iph2->tos) | (df ^ (id2 & IP_DF)); + + if (flush | (outer && df)) + return flush; + + /* When we receive our second frame we can make a decision on if we + * continue this flow as an atomic flow with a fixed ID or if we use + * an incrementing ID. + */ + if (count == 1 && df && !ipid_offset) + NAPI_GRO_CB(p)->ip_fixedid = true; + + return ipid_offset ^ (count * !NAPI_GRO_CB(p)->ip_fixedid); +} + +static inline int ipv6_gro_flush(const struct ipv6hdr *iph, const struct ipv6hdr *iph2) +{ + /* */ + __be32 first_word = *(__be32 *)iph ^ *(__be32 *)iph2; + + /* Flush if Traffic Class fields are different. */ + return !!((first_word & htonl(0x0FF00000)) | + (__force __be32)(iph->hop_limit ^ iph2->hop_limit)); +} + +static inline int __gro_receive_network_flush(const void *th, const void *th2, + struct sk_buff *p, const u16 diff, + bool outer) +{ + const void *nh = th - diff; + const void *nh2 = th2 - diff; + + if (((struct iphdr *)nh)->version == 6) + return ipv6_gro_flush(nh, nh2); + else + return inet_gro_flush(nh, nh2, p, outer); +} + +static inline int gro_receive_network_flush(const void *th, const void *th2, + struct sk_buff *p) +{ + const bool encap_mark = NAPI_GRO_CB(p)->encap_mark; + int off = skb_transport_offset(p); + int flush; + + flush = __gro_receive_network_flush(th, th2, p, off - NAPI_GRO_CB(p)->network_offset, encap_mark); + if (encap_mark) + flush |= __gro_receive_network_flush(th, th2, p, off - NAPI_GRO_CB(p)->inner_network_offset, false); + + return flush; +} + int skb_gro_receive(struct sk_buff *p, struct sk_buff *skb); int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb); diff --git a/net/core/gro.c b/net/core/gro.c index e2f84947fb74..b3b43de1a650 100644 --- a/net/core/gro.c +++ b/net/core/gro.c @@ -358,8 +358,6 @@ static void gro_list_prepare(const struct list_head *head, list_for_each_entry(p, head, list) { unsigned long diffs; - NAPI_GRO_CB(p)->flush = 0; - if (hash != skb_get_hash_raw(p)) { NAPI_GRO_CB(p)->same_flow = 0; continue; @@ -499,7 +497,6 @@ found_ptype: sizeof(u32))); /* Avoid slow unaligned acc */ *(u32 *)&NAPI_GRO_CB(skb)->zeroed = 0; NAPI_GRO_CB(skb)->flush = skb_has_frag_list(skb); - NAPI_GRO_CB(skb)->is_atomic = 1; NAPI_GRO_CB(skb)->count = 1; if (unlikely(skb_is_gso(skb))) { NAPI_GRO_CB(skb)->count = skb_shinfo(skb)->gso_segs; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 428196e1541f..44564d009e95 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1482,7 +1482,6 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) struct sk_buff *p; unsigned int hlen; unsigned int off; - unsigned int id; int flush = 1; int proto; @@ -1508,13 +1507,10 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) goto out; NAPI_GRO_CB(skb)->proto = proto; - id = ntohl(*(__be32 *)&iph->id); - flush = (u16)((ntohl(*(__be32 *)iph) ^ skb_gro_len(skb)) | (id & ~IP_DF)); - id >>= 16; + flush = (u16)((ntohl(*(__be32 *)iph) ^ skb_gro_len(skb)) | (ntohl(*(__be32 *)&iph->id) & ~IP_DF)); list_for_each_entry(p, head, list) { struct iphdr *iph2; - u16 flush_id; if (!NAPI_GRO_CB(p)->same_flow) continue; @@ -1531,43 +1527,8 @@ struct sk_buff *inet_gro_receive(struct list_head *head, struct sk_buff *skb) NAPI_GRO_CB(p)->same_flow = 0; continue; } - - /* All fields must match except length and checksum. */ - NAPI_GRO_CB(p)->flush |= - (iph->ttl ^ iph2->ttl) | - (iph->tos ^ iph2->tos) | - ((iph->frag_off ^ iph2->frag_off) & htons(IP_DF)); - - NAPI_GRO_CB(p)->flush |= flush; - - /* We need to store of the IP ID check to be included later - * when we can verify that this packet does in fact belong - * to a given flow. - */ - flush_id = (u16)(id - ntohs(iph2->id)); - - /* This bit of code makes it much easier for us to identify - * the cases where we are doing atomic vs non-atomic IP ID - * checks. Specifically an atomic check can return IP ID - * values 0 - 0xFFFF, while a non-atomic check can only - * return 0 or 0xFFFF. - */ - if (!NAPI_GRO_CB(p)->is_atomic || - !(iph->frag_off & htons(IP_DF))) { - flush_id ^= NAPI_GRO_CB(p)->count; - flush_id = flush_id ? 0xFFFF : 0; - } - - /* If the previous IP ID value was based on an atomic - * datagram we can overwrite the value and ignore it. - */ - if (NAPI_GRO_CB(skb)->is_atomic) - NAPI_GRO_CB(p)->flush_id = flush_id; - else - NAPI_GRO_CB(p)->flush_id |= flush_id; } - NAPI_GRO_CB(skb)->is_atomic = !!(iph->frag_off & htons(IP_DF)); NAPI_GRO_CB(skb)->flush |= flush; NAPI_GRO_CB(skb)->inner_network_offset = off; diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index 2809667ac924..4b791e74529e 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -313,10 +313,8 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, if (!p) goto out_check_final; - /* Include the IP ID check below from the inner most IP hdr */ th2 = tcp_hdr(p); - flush = NAPI_GRO_CB(p)->flush; - flush |= (__force int)(flags & TCP_FLAG_CWR); + flush = (__force int)(flags & TCP_FLAG_CWR); flush |= (__force int)((flags ^ tcp_flag_word(th2)) & ~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH)); flush |= (__force int)(th->ack_seq ^ th2->ack_seq); @@ -324,16 +322,7 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, flush |= *(u32 *)((u8 *)th + i) ^ *(u32 *)((u8 *)th2 + i); - /* When we receive our second frame we can made a decision on if we - * continue this flow as an atomic flow with a fixed ID or if we use - * an incrementing ID. - */ - if (NAPI_GRO_CB(p)->flush_id != 1 || - NAPI_GRO_CB(p)->count != 1 || - !NAPI_GRO_CB(p)->is_atomic) - flush |= NAPI_GRO_CB(p)->flush_id; - else - NAPI_GRO_CB(p)->is_atomic = false; + flush |= gro_receive_network_flush(th, th2, p); mss = skb_shinfo(p)->gso_size; @@ -480,7 +469,7 @@ INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff) iph->daddr, 0); skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV4 | - (NAPI_GRO_CB(skb)->is_atomic * SKB_GSO_TCP_FIXEDID); + (NAPI_GRO_CB(skb)->ip_fixedid * SKB_GSO_TCP_FIXEDID); tcp_gro_complete(skb); return 0; diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index f6d3c442e260..59448a2dbf2c 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -478,14 +478,7 @@ static struct sk_buff *udp_gro_receive_segment(struct list_head *head, return p; } - flush = NAPI_GRO_CB(p)->flush; - - if (NAPI_GRO_CB(p)->flush_id != 1 || - NAPI_GRO_CB(p)->count != 1 || - !NAPI_GRO_CB(p)->is_atomic) - flush |= NAPI_GRO_CB(p)->flush_id; - else - NAPI_GRO_CB(p)->is_atomic = false; + flush = gro_receive_network_flush(uh, uh2, p); /* Terminate the flow on len mismatch or if it grow "too much". * Under small packet flood GRO count could elsewhere grow a lot diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c index 288c7c6ea50f..bd5aff97d8b1 100644 --- a/net/ipv6/ip6_offload.c +++ b/net/ipv6/ip6_offload.c @@ -290,19 +290,8 @@ not_same_flow: nlen - sizeof(struct ipv6hdr))) goto not_same_flow; } - /* flush if Traffic Class fields are different */ - NAPI_GRO_CB(p)->flush |= !!((first_word & htonl(0x0FF00000)) | - (__force __be32)(iph->hop_limit ^ iph2->hop_limit)); - NAPI_GRO_CB(p)->flush |= flush; - - /* If the previous IP ID value was based on an atomic - * datagram we can overwrite the value and ignore it. - */ - if (NAPI_GRO_CB(skb)->is_atomic) - NAPI_GRO_CB(p)->flush_id = 0; } - NAPI_GRO_CB(skb)->is_atomic = true; NAPI_GRO_CB(skb)->flush |= flush; skb_gro_postpull_rcsum(skb, iph, nlen); -- cgit v1.2.3 From c084ebd77a00b1a16d9daa57b6ecdfdf1f43c78a Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Thu, 9 May 2024 20:10:10 +0200 Subject: tcp: socket option to check for MPTCP fallback to TCP A way for an application to know if an MPTCP connection fell back to TCP is to use getsockopt(MPTCP_INFO) and look for errors. The issue with this technique is that the same errors -- EOPNOTSUPP (IPv4) and ENOPROTOOPT (IPv6) -- are returned if there was a fallback, *or* if the kernel doesn't support this socket option. The userspace then has to look at the kernel version to understand what the errors mean. It is not clean, and it doesn't take into account older kernels where the socket option has been backported. A cleaner way would be to expose this info to the TCP socket level. In case of MPTCP socket where no fallback happened, the socket options for the TCP level will be handled in MPTCP code, in mptcp_getsockopt_sol_tcp(). If not, that will be in TCP code, in do_tcp_getsockopt(). So MPTCP simply has to set the value 1, while TCP has to set 0. If the socket option is not supported, one of these two errors will be reported: - EOPNOTSUPP (95 - Operation not supported) for MPTCP sockets - ENOPROTOOPT (92 - Protocol not available) for TCP sockets, e.g. on the socket received after an 'accept()', when the client didn't request to use MPTCP: this socket will be a TCP one, even if the listen socket was an MPTCP one. With this new option, the kernel can return a clear answer to both "Is this kernel new enough to tell me the fallback status?" and "If it is new enough, is it currently a TCP or MPTCP socket?" questions, while not breaking the previous method. Acked-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://lore.kernel.org/r/20240509-upstream-net-next-20240509-mptcp-tcp_is_mptcp-v1-1-f846df999202@kernel.org Signed-off-by: Jakub Kicinski --- include/uapi/linux/tcp.h | 2 ++ net/ipv4/tcp.c | 3 +++ net/mptcp/sockopt.c | 2 ++ 3 files changed, 7 insertions(+) (limited to 'net') diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index c07e9f90c084..dbf896f3146c 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -135,6 +135,8 @@ enum { #define TCP_AO_GET_KEYS 41 /* List MKT(s) */ #define TCP_AO_REPAIR 42 /* Get/Set SNEs and ISNs */ +#define TCP_IS_MPTCP 43 /* Is MPTCP being used? */ + #define TCP_REPAIR_ON 1 #define TCP_REPAIR_OFF 0 #define TCP_REPAIR_OFF_NO_WP -1 /* Turn off without window probes */ diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 06aab937d60a..681b54e1f3a6 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -4363,6 +4363,9 @@ zerocopy_rcv_out: return err; } + case TCP_IS_MPTCP: + val = 0; + break; default: return -ENOPROTOOPT; } diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 1fea43f5b6f3..eaa3b79651a4 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -1348,6 +1348,8 @@ static int mptcp_getsockopt_sol_tcp(struct mptcp_sock *msk, int optname, return mptcp_put_int_option(msk, optval, optlen, msk->nodelay); case TCP_NOTSENT_LOWAT: return mptcp_put_int_option(msk, optval, optlen, msk->notsent_lowat); + case TCP_IS_MPTCP: + return mptcp_put_int_option(msk, optval, optlen, 1); } return -EOPNOTSUPP; } -- cgit v1.2.3 From 8ec9897ec2e93a412b9e3119e3137583a85969e0 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 10 May 2024 19:19:12 +0200 Subject: netlabel: fix RCU annotation for IPv4 options on socket creation Xiumei reports the following splat when netlabel and TCP socket are used: ============================= WARNING: suspicious RCU usage 6.9.0-rc2+ #637 Not tainted ----------------------------- net/ipv4/cipso_ipv4.c:1880 suspicious rcu_dereference_protected() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 1 1 lock held by ncat/23333: #0: ffffffff906030c0 (rcu_read_lock){....}-{1:2}, at: netlbl_sock_setattr+0x25/0x1b0 stack backtrace: CPU: 11 PID: 23333 Comm: ncat Kdump: loaded Not tainted 6.9.0-rc2+ #637 Hardware name: Supermicro SYS-6027R-72RF/X9DRH-7TF/7F/iTF/iF, BIOS 3.0 07/26/2013 Call Trace: dump_stack_lvl+0xa9/0xc0 lockdep_rcu_suspicious+0x117/0x190 cipso_v4_sock_setattr+0x1ab/0x1b0 netlbl_sock_setattr+0x13e/0x1b0 selinux_netlbl_socket_post_create+0x3f/0x80 selinux_socket_post_create+0x1a0/0x460 security_socket_post_create+0x42/0x60 __sock_create+0x342/0x3a0 __sys_socket_create.part.22+0x42/0x70 __sys_socket+0x37/0xb0 __x64_sys_socket+0x16/0x20 do_syscall_64+0x96/0x180 ? do_user_addr_fault+0x68d/0xa30 ? exc_page_fault+0x171/0x280 ? asm_exc_page_fault+0x22/0x30 entry_SYSCALL_64_after_hwframe+0x71/0x79 RIP: 0033:0x7fbc0ca3fc1b Code: 73 01 c3 48 8b 0d 05 f2 1b 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa b8 29 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d d5 f1 1b 00 f7 d8 64 89 01 48 RSP: 002b:00007fff18635208 EFLAGS: 00000246 ORIG_RAX: 0000000000000029 RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007fbc0ca3fc1b RDX: 0000000000000006 RSI: 0000000000000001 RDI: 0000000000000002 RBP: 000055d24f80f8a0 R08: 0000000000000003 R09: 0000000000000001 R10: 0000000000020000 R11: 0000000000000246 R12: 000055d24f80f8a0 R13: 0000000000000000 R14: 000055d24f80fb88 R15: 0000000000000000 The current implementation of cipso_v4_sock_setattr() replaces IP options under the assumption that the caller holds the socket lock; however, such assumption is not true, nor needed, in selinux_socket_post_create() hook. Let all callers of cipso_v4_sock_setattr() specify the "socket lock held" condition, except selinux_socket_post_create() _ where such condition can safely be set as true even without holding the socket lock. Fixes: f6d8bd051c39 ("inet: add RCU protection to inet->opt") Reported-by: Xiumei Mu Signed-off-by: Davide Caratti Acked-by: Casey Schaufler Acked-by: Paul Moore Link: https://lore.kernel.org/r/f4260d000a3a55b9e8b6a3b4e3fffc7da9f82d41.1715359817.git.dcaratti@redhat.com Signed-off-by: Jakub Kicinski --- include/net/cipso_ipv4.h | 6 ++++-- include/net/netlabel.h | 12 ++++++++++-- net/ipv4/cipso_ipv4.c | 7 ++++--- net/netlabel/netlabel_kapi.c | 31 ++++++++++++++++++++++++++++--- security/selinux/netlabel.c | 5 ++++- security/smack/smack_lsm.c | 3 ++- 6 files changed, 52 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h index 53dd7d988a2d..c9111bb2f59b 100644 --- a/include/net/cipso_ipv4.h +++ b/include/net/cipso_ipv4.h @@ -183,7 +183,8 @@ int cipso_v4_getattr(const unsigned char *cipso, struct netlbl_lsm_secattr *secattr); int cipso_v4_sock_setattr(struct sock *sk, const struct cipso_v4_doi *doi_def, - const struct netlbl_lsm_secattr *secattr); + const struct netlbl_lsm_secattr *secattr, + bool sk_locked); void cipso_v4_sock_delattr(struct sock *sk); int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr); int cipso_v4_req_setattr(struct request_sock *req, @@ -214,7 +215,8 @@ static inline int cipso_v4_getattr(const unsigned char *cipso, static inline int cipso_v4_sock_setattr(struct sock *sk, const struct cipso_v4_doi *doi_def, - const struct netlbl_lsm_secattr *secattr) + const struct netlbl_lsm_secattr *secattr, + bool sk_locked) { return -ENOSYS; } diff --git a/include/net/netlabel.h b/include/net/netlabel.h index f3ab0b8a4b18..2133ad723fc1 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -470,7 +470,8 @@ void netlbl_bitmap_setbit(unsigned char *bitmap, u32 bit, u8 state); int netlbl_enabled(void); int netlbl_sock_setattr(struct sock *sk, u16 family, - const struct netlbl_lsm_secattr *secattr); + const struct netlbl_lsm_secattr *secattr, + bool sk_locked); void netlbl_sock_delattr(struct sock *sk); int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr); @@ -487,6 +488,7 @@ int netlbl_skbuff_getattr(const struct sk_buff *skb, u16 family, struct netlbl_lsm_secattr *secattr); void netlbl_skbuff_err(struct sk_buff *skb, u16 family, int error, int gateway); +bool netlbl_sk_lock_check(struct sock *sk); /* * LSM label mapping cache operations @@ -614,7 +616,8 @@ static inline int netlbl_enabled(void) } static inline int netlbl_sock_setattr(struct sock *sk, u16 family, - const struct netlbl_lsm_secattr *secattr) + const struct netlbl_lsm_secattr *secattr, + bool sk_locked) { return -ENOSYS; } @@ -673,6 +676,11 @@ static inline struct audit_buffer *netlbl_audit_start(int type, { return NULL; } + +static inline bool netlbl_sk_lock_check(struct sock *sk) +{ + return true; +} #endif /* CONFIG_NETLABEL */ const struct netlbl_calipso_ops * diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 8b17d83e5fde..dd6d46015058 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -1815,6 +1815,7 @@ static int cipso_v4_genopt(unsigned char *buf, u32 buf_len, * @sk: the socket * @doi_def: the CIPSO DOI to use * @secattr: the specific security attributes of the socket + * @sk_locked: true if caller holds the socket lock * * Description: * Set the CIPSO option on the given socket using the DOI definition and @@ -1826,7 +1827,8 @@ static int cipso_v4_genopt(unsigned char *buf, u32 buf_len, */ int cipso_v4_sock_setattr(struct sock *sk, const struct cipso_v4_doi *doi_def, - const struct netlbl_lsm_secattr *secattr) + const struct netlbl_lsm_secattr *secattr, + bool sk_locked) { int ret_val = -EPERM; unsigned char *buf = NULL; @@ -1876,8 +1878,7 @@ int cipso_v4_sock_setattr(struct sock *sk, sk_inet = inet_sk(sk); - old = rcu_dereference_protected(sk_inet->inet_opt, - lockdep_sock_is_held(sk)); + old = rcu_dereference_protected(sk_inet->inet_opt, sk_locked); if (inet_test_bit(IS_ICSK, sk)) { sk_conn = inet_csk(sk); if (old) diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index 1ba4f58e1d35..cd9160bbc919 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -965,6 +965,7 @@ int netlbl_enabled(void) * @sk: the socket to label * @family: protocol family * @secattr: the security attributes + * @sk_locked: true if caller holds the socket lock * * Description: * Attach the correct label to the given socket using the security attributes @@ -977,7 +978,8 @@ int netlbl_enabled(void) */ int netlbl_sock_setattr(struct sock *sk, u16 family, - const struct netlbl_lsm_secattr *secattr) + const struct netlbl_lsm_secattr *secattr, + bool sk_locked) { int ret_val; struct netlbl_dom_map *dom_entry; @@ -997,7 +999,7 @@ int netlbl_sock_setattr(struct sock *sk, case NETLBL_NLTYPE_CIPSOV4: ret_val = cipso_v4_sock_setattr(sk, dom_entry->def.cipso, - secattr); + secattr, sk_locked); break; case NETLBL_NLTYPE_UNLABELED: ret_val = 0; @@ -1090,6 +1092,28 @@ int netlbl_sock_getattr(struct sock *sk, return ret_val; } +/** + * netlbl_sk_lock_check - Check if the socket lock has been acquired. + * @sk: the socket to be checked + * + * Return: true if socket @sk is locked or if lock debugging is disabled at + * runtime or compile-time; false otherwise + * + */ +#ifdef CONFIG_LOCKDEP +bool netlbl_sk_lock_check(struct sock *sk) +{ + if (debug_locks) + return lockdep_sock_is_held(sk); + return true; +} +#else +bool netlbl_sk_lock_check(struct sock *sk) +{ + return true; +} +#endif + /** * netlbl_conn_setattr - Label a connected socket using the correct protocol * @sk: the socket to label @@ -1126,7 +1150,8 @@ int netlbl_conn_setattr(struct sock *sk, switch (entry->type) { case NETLBL_NLTYPE_CIPSOV4: ret_val = cipso_v4_sock_setattr(sk, - entry->cipso, secattr); + entry->cipso, secattr, + netlbl_sk_lock_check(sk)); break; case NETLBL_NLTYPE_UNLABELED: /* just delete the protocols we support for right now diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c index 8f182800e412..55885634e880 100644 --- a/security/selinux/netlabel.c +++ b/security/selinux/netlabel.c @@ -402,7 +402,10 @@ int selinux_netlbl_socket_post_create(struct sock *sk, u16 family) secattr = selinux_netlbl_sock_genattr(sk); if (secattr == NULL) return -ENOMEM; - rc = netlbl_sock_setattr(sk, family, secattr); + /* On socket creation, replacement of IP options is safe even if + * the caller does not hold the socket lock. + */ + rc = netlbl_sock_setattr(sk, family, secattr, true); switch (rc) { case 0: sksec->nlbl_state = NLBL_LABELED; diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 146667937811..efeac8365ad0 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -2565,7 +2565,8 @@ static int smack_netlbl_add(struct sock *sk) local_bh_disable(); bh_lock_sock_nested(sk); - rc = netlbl_sock_setattr(sk, sk->sk_family, &skp->smk_netlabel); + rc = netlbl_sock_setattr(sk, sk->sk_family, &skp->smk_netlabel, + netlbl_sk_lock_check(sk)); switch (rc) { case 0: ssp->smk_state = SMK_NETLBL_LABELED; -- cgit v1.2.3 From b56035101e1cdd9c4420ea5da17f09f87fb69285 Mon Sep 17 00:00:00 2001 From: Daniel Jurgens Date: Fri, 10 May 2024 23:19:26 +0300 Subject: netdev: Add queue stats for TX stop and wake TX queue stop and wake are counted by some drivers. Support reporting these via netdev-genl queue stats. Signed-off-by: Daniel Jurgens Reviewed-by: Jiri Pirko Reviewed-by: Jason Xing Link: https://lore.kernel.org/r/20240510201927.1821109-2-danielj@nvidia.com Signed-off-by: Jakub Kicinski --- Documentation/netlink/specs/netdev.yaml | 14 ++++++++++++++ include/net/netdev_queues.h | 3 +++ include/uapi/linux/netdev.h | 2 ++ net/core/netdev-genl.c | 4 +++- tools/include/uapi/linux/netdev.h | 2 ++ 5 files changed, 24 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml index 2be4b3714d17..11a32373365a 100644 --- a/Documentation/netlink/specs/netdev.yaml +++ b/Documentation/netlink/specs/netdev.yaml @@ -439,6 +439,20 @@ attribute-sets: Number of the packets dropped by the device due to the transmit packets bitrate exceeding the device rate limit. type: uint + - + name: tx-stop + doc: | + Number of times driver paused accepting new tx packets + from the stack to this queue, because the queue was full. + Note that if BQL is supported and enabled on the device + the networking stack will avoid queuing a lot of data at once. + type: uint + - + name: tx-wake + doc: | + Number of times driver re-started accepting send + requests to this queue from the stack. + type: uint operations: list: diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h index e7b84f018cee..a8a7e48dfa6c 100644 --- a/include/net/netdev_queues.h +++ b/include/net/netdev_queues.h @@ -41,6 +41,9 @@ struct netdev_queue_stats_tx { u64 hw_gso_wire_bytes; u64 hw_drop_ratelimits; + + u64 stop; + u64 wake; }; /** diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h index cf24f1d9adf8..a8188202413e 100644 --- a/include/uapi/linux/netdev.h +++ b/include/uapi/linux/netdev.h @@ -165,6 +165,8 @@ enum { NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, + NETDEV_A_QSTATS_TX_STOP, + NETDEV_A_QSTATS_TX_WAKE, __NETDEV_A_QSTATS_MAX, NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1) diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 4b5054087309..1f6ae6379e0f 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -517,7 +517,9 @@ netdev_nl_stats_write_tx(struct sk_buff *rsp, struct netdev_queue_stats_tx *tx) netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_BYTES, tx->hw_gso_bytes) || netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, tx->hw_gso_wire_packets) || netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, tx->hw_gso_wire_bytes) || - netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits)) + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_STOP, tx->stop) || + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_WAKE, tx->wake)) return -EMSGSIZE; return 0; } diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h index cf24f1d9adf8..a8188202413e 100644 --- a/tools/include/uapi/linux/netdev.h +++ b/tools/include/uapi/linux/netdev.h @@ -165,6 +165,8 @@ enum { NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, + NETDEV_A_QSTATS_TX_STOP, + NETDEV_A_QSTATS_TX_WAKE, __NETDEV_A_QSTATS_MAX, NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1) -- cgit v1.2.3 From 364798056f518b0bf2f17cd9eaf0dd4e856d7393 Mon Sep 17 00:00:00 2001 From: Samuel Thibault Date: Thu, 9 May 2024 22:58:12 +0200 Subject: l2tp: Support different protocol versions with same IP/port quadruple 628bc3e5a1be ("l2tp: Support several sockets with same IP/port quadruple") added support for several L2TPv2 tunnels using the same IP/port quadruple, but if an L2TPv3 socket exists it could eat all the trafic. We thus have to first use the version from the packet to get the proper tunnel, and only then check that the version matches. Signed-off-by: Samuel Thibault Reviewed-by: James Chapman Link: https://lore.kernel.org/r/20240509205812.4063198-1-samuel.thibault@ens-lyon.org Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_core.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 2ab45e3f48bf..7d519a46a844 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -820,13 +820,8 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) /* Get L2TP header flags */ hdrflags = ntohs(*(__be16 *)ptr); - /* Check protocol version */ + /* Get protocol version */ version = hdrflags & L2TP_HDR_VER_MASK; - if (version != tunnel->version) { - pr_debug_ratelimited("%s: recv protocol version mismatch: got %d expected %d\n", - tunnel->name, version, tunnel->version); - goto invalid; - } /* Get length of L2TP packet */ length = skb->len; @@ -838,7 +833,7 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) /* Skip flags */ ptr += 2; - if (tunnel->version == L2TP_HDR_VER_2) { + if (version == L2TP_HDR_VER_2) { /* If length is present, skip it */ if (hdrflags & L2TP_HDRFLAG_L) ptr += 2; @@ -855,7 +850,7 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) struct l2tp_tunnel *alt_tunnel; alt_tunnel = l2tp_tunnel_get(tunnel->l2tp_net, tunnel_id); - if (!alt_tunnel || alt_tunnel->version != L2TP_HDR_VER_2) + if (!alt_tunnel) goto pass; tunnel = alt_tunnel; } @@ -869,6 +864,13 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb) ptr += 4; } + /* Check protocol version */ + if (version != tunnel->version) { + pr_debug_ratelimited("%s: recv protocol version mismatch: got %d expected %d\n", + tunnel->name, version, tunnel->version); + goto invalid; + } + /* Find the session context */ session = l2tp_tunnel_get_session(tunnel, session_id); if (!session || !session->recv_skb) { -- cgit v1.2.3 From a7d6e36b9ad052926ba2ecba3a59d8bb67dabcb4 Mon Sep 17 00:00:00 2001 From: Duoming Zhou Date: Thu, 9 May 2024 17:36:33 +0800 Subject: ax25: Use kernel universal linked list to implement ax25_dev_list The origin ax25_dev_list implements its own single linked list, which is complicated and error-prone. For example, when deleting the node of ax25_dev_list in ax25_dev_device_down(), we have to operate on the head node and other nodes separately. This patch uses kernel universal linked list to replace original ax25_dev_list, which make the operation of ax25_dev_list easier. We should do "dev->ax25_ptr = ax25_dev;" and "dev->ax25_ptr = NULL;" while holding the spinlock, otherwise the ax25_dev_device_up() and ax25_dev_device_down() could race. Suggested-by: Dan Carpenter Signed-off-by: Duoming Zhou Reviewed-by: Dan Carpenter Link: https://lore.kernel.org/r/85bba3af651ca0e1a519da8d0d715b949891171c.1715247018.git.duoming@zju.edu.cn Signed-off-by: Jakub Kicinski --- include/net/ax25.h | 3 +-- net/ax25/ax25_dev.c | 40 +++++++++++++++------------------------- 2 files changed, 16 insertions(+), 27 deletions(-) (limited to 'net') diff --git a/include/net/ax25.h b/include/net/ax25.h index 0d939e5aee4e..c2a85fd3f5ea 100644 --- a/include/net/ax25.h +++ b/include/net/ax25.h @@ -216,7 +216,7 @@ typedef struct { struct ctl_table; typedef struct ax25_dev { - struct ax25_dev *next; + struct list_head list; struct net_device *dev; netdevice_tracker dev_tracker; @@ -330,7 +330,6 @@ int ax25_addr_size(const ax25_digi *); void ax25_digi_invert(const ax25_digi *, ax25_digi *); /* ax25_dev.c */ -extern ax25_dev *ax25_dev_list; extern spinlock_t ax25_dev_lock; #if IS_ENABLED(CONFIG_AX25) diff --git a/net/ax25/ax25_dev.c b/net/ax25/ax25_dev.c index 282ec581c072..f16ee5c09d07 100644 --- a/net/ax25/ax25_dev.c +++ b/net/ax25/ax25_dev.c @@ -22,11 +22,12 @@ #include #include #include +#include #include #include #include -ax25_dev *ax25_dev_list; +static LIST_HEAD(ax25_dev_list); DEFINE_SPINLOCK(ax25_dev_lock); ax25_dev *ax25_addr_ax25dev(ax25_address *addr) @@ -34,7 +35,7 @@ ax25_dev *ax25_addr_ax25dev(ax25_address *addr) ax25_dev *ax25_dev, *res = NULL; spin_lock_bh(&ax25_dev_lock); - for (ax25_dev = ax25_dev_list; ax25_dev != NULL; ax25_dev = ax25_dev->next) + list_for_each_entry(ax25_dev, &ax25_dev_list, list) if (ax25cmp(addr, (const ax25_address *)ax25_dev->dev->dev_addr) == 0) { res = ax25_dev; ax25_dev_hold(ax25_dev); @@ -59,7 +60,6 @@ void ax25_dev_device_up(struct net_device *dev) } refcount_set(&ax25_dev->refcount, 1); - dev->ax25_ptr = ax25_dev; ax25_dev->dev = dev; netdev_hold(dev, &ax25_dev->dev_tracker, GFP_KERNEL); ax25_dev->forward = NULL; @@ -85,8 +85,8 @@ void ax25_dev_device_up(struct net_device *dev) #endif spin_lock_bh(&ax25_dev_lock); - ax25_dev->next = ax25_dev_list; - ax25_dev_list = ax25_dev; + list_add(&ax25_dev->list, &ax25_dev_list); + dev->ax25_ptr = ax25_dev; spin_unlock_bh(&ax25_dev_lock); ax25_dev_hold(ax25_dev); @@ -111,32 +111,25 @@ void ax25_dev_device_down(struct net_device *dev) /* * Remove any packet forwarding that points to this device. */ - for (s = ax25_dev_list; s != NULL; s = s->next) + list_for_each_entry(s, &ax25_dev_list, list) if (s->forward == dev) s->forward = NULL; - if ((s = ax25_dev_list) == ax25_dev) { - ax25_dev_list = s->next; - goto unlock_put; - } - - while (s != NULL && s->next != NULL) { - if (s->next == ax25_dev) { - s->next = ax25_dev->next; + list_for_each_entry(s, &ax25_dev_list, list) { + if (s == ax25_dev) { + list_del(&s->list); goto unlock_put; } - - s = s->next; } - spin_unlock_bh(&ax25_dev_lock); dev->ax25_ptr = NULL; + spin_unlock_bh(&ax25_dev_lock); ax25_dev_put(ax25_dev); return; unlock_put: + dev->ax25_ptr = NULL; spin_unlock_bh(&ax25_dev_lock); ax25_dev_put(ax25_dev); - dev->ax25_ptr = NULL; netdev_put(dev, &ax25_dev->dev_tracker); ax25_dev_put(ax25_dev); } @@ -200,16 +193,13 @@ struct net_device *ax25_fwd_dev(struct net_device *dev) */ void __exit ax25_dev_free(void) { - ax25_dev *s, *ax25_dev; + ax25_dev *s, *n; spin_lock_bh(&ax25_dev_lock); - ax25_dev = ax25_dev_list; - while (ax25_dev != NULL) { - s = ax25_dev; - netdev_put(ax25_dev->dev, &ax25_dev->dev_tracker); - ax25_dev = ax25_dev->next; + list_for_each_entry_safe(s, n, &ax25_dev_list, list) { + netdev_put(s->dev, &s->dev_tracker); + list_del(&s->list); kfree(s); } - ax25_dev_list = NULL; spin_unlock_bh(&ax25_dev_lock); } -- cgit v1.2.3 From b505e0319852b08a3a716b64620168eab21f4ced Mon Sep 17 00:00:00 2001 From: Duoming Zhou Date: Thu, 9 May 2024 17:36:47 +0800 Subject: ax25: Fix reference count leak issues of ax25_dev The ax25_addr_ax25dev() and ax25_dev_device_down() exist a reference count leak issue of the object "ax25_dev". Memory leak issue in ax25_addr_ax25dev(): The reference count of the object "ax25_dev" can be increased multiple times in ax25_addr_ax25dev(). This will cause a memory leak. Memory leak issues in ax25_dev_device_down(): The reference count of ax25_dev is set to 1 in ax25_dev_device_up() and then increase the reference count when ax25_dev is added to ax25_dev_list. As a result, the reference count of ax25_dev is 2. But when the device is shutting down. The ax25_dev_device_down() drops the reference count once or twice depending on if we goto unlock_put or not, which will cause memory leak. As for the issue of ax25_addr_ax25dev(), it is impossible for one pointer to be on a list twice. So add a break in ax25_addr_ax25dev(). As for the issue of ax25_dev_device_down(), increase the reference count of ax25_dev once in ax25_dev_device_up() and decrease the reference count of ax25_dev after it is removed from the ax25_dev_list. Fixes: d01ffb9eee4a ("ax25: add refcount in ax25_dev to avoid UAF bugs") Suggested-by: Dan Carpenter Signed-off-by: Duoming Zhou Reviewed-by: Dan Carpenter Link: https://lore.kernel.org/r/361bbf2a4b091e120006279ec3b382d73c4a0c17.1715247018.git.duoming@zju.edu.cn Signed-off-by: Jakub Kicinski --- net/ax25/ax25_dev.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/ax25/ax25_dev.c b/net/ax25/ax25_dev.c index f16ee5c09d07..52ccc37d5687 100644 --- a/net/ax25/ax25_dev.c +++ b/net/ax25/ax25_dev.c @@ -39,6 +39,7 @@ ax25_dev *ax25_addr_ax25dev(ax25_address *addr) if (ax25cmp(addr, (const ax25_address *)ax25_dev->dev->dev_addr) == 0) { res = ax25_dev; ax25_dev_hold(ax25_dev); + break; } spin_unlock_bh(&ax25_dev_lock); @@ -88,7 +89,6 @@ void ax25_dev_device_up(struct net_device *dev) list_add(&ax25_dev->list, &ax25_dev_list); dev->ax25_ptr = ax25_dev; spin_unlock_bh(&ax25_dev_lock); - ax25_dev_hold(ax25_dev); ax25_register_dev_sysctl(ax25_dev); } @@ -129,7 +129,6 @@ void ax25_dev_device_down(struct net_device *dev) unlock_put: dev->ax25_ptr = NULL; spin_unlock_bh(&ax25_dev_lock); - ax25_dev_put(ax25_dev); netdev_put(dev, &ax25_dev->dev_tracker); ax25_dev_put(ax25_dev); } -- cgit v1.2.3 From 36e56b1b002bb26440403053f19f9e1a8bc075b2 Mon Sep 17 00:00:00 2001 From: Duoming Zhou Date: Thu, 9 May 2024 17:37:02 +0800 Subject: ax25: Fix reference count leak issue of net_device There is a reference count leak issue of the object "net_device" in ax25_dev_device_down(). When the ax25 device is shutting down, the ax25_dev_device_down() drops the reference count of net_device one or zero times depending on if we goto unlock_put or not, which will cause memory leak. In order to solve the above issue, decrease the reference count of net_device after dev->ax25_ptr is set to null. Fixes: d01ffb9eee4a ("ax25: add refcount in ax25_dev to avoid UAF bugs") Suggested-by: Dan Carpenter Signed-off-by: Duoming Zhou Reviewed-by: Dan Carpenter Link: https://lore.kernel.org/r/7ce3b23a40d9084657ba1125432f0ecc380cbc80.1715247018.git.duoming@zju.edu.cn Signed-off-by: Jakub Kicinski --- net/ax25/ax25_dev.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'net') diff --git a/net/ax25/ax25_dev.c b/net/ax25/ax25_dev.c index 52ccc37d5687..c9d55b99a7a5 100644 --- a/net/ax25/ax25_dev.c +++ b/net/ax25/ax25_dev.c @@ -118,15 +118,10 @@ void ax25_dev_device_down(struct net_device *dev) list_for_each_entry(s, &ax25_dev_list, list) { if (s == ax25_dev) { list_del(&s->list); - goto unlock_put; + break; } } - dev->ax25_ptr = NULL; - spin_unlock_bh(&ax25_dev_lock); - ax25_dev_put(ax25_dev); - return; -unlock_put: dev->ax25_ptr = NULL; spin_unlock_bh(&ax25_dev_lock); netdev_put(dev, &ax25_dev->dev_tracker); -- cgit v1.2.3 From 92ef0fd55ac80dfc2e4654edfe5d1ddfa6e070fe Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 May 2024 09:20:08 -0600 Subject: net: change proto and proto_ops accept type Rather than pass in flags, error pointer, and whether this is a kernel invocation or not, add a struct proto_accept_arg struct as the argument. This then holds all of these arguments, and prepares accept for being able to pass back more information. No functional changes in this patch. Acked-by: Jakub Kicinski Signed-off-by: Jens Axboe --- crypto/af_alg.c | 11 ++++++----- crypto/algif_hash.c | 10 +++++----- drivers/xen/pvcalls-back.c | 6 +++++- fs/ocfs2/cluster/tcp.c | 5 ++++- include/crypto/if_alg.h | 3 ++- include/linux/net.h | 4 +++- include/net/inet_common.h | 4 ++-- include/net/inet_connection_sock.h | 2 +- include/net/sock.h | 12 +++++++++--- net/atm/svc.c | 8 ++++---- net/ax25/af_ax25.c | 6 +++--- net/bluetooth/iso.c | 4 ++-- net/bluetooth/l2cap_sock.c | 4 ++-- net/bluetooth/rfcomm/sock.c | 6 +++--- net/bluetooth/sco.c | 4 ++-- net/core/sock.c | 4 ++-- net/ipv4/af_inet.c | 10 +++++----- net/ipv4/inet_connection_sock.c | 6 +++--- net/iucv/af_iucv.c | 4 ++-- net/llc/af_llc.c | 7 +++---- net/mptcp/protocol.c | 11 +++++------ net/netrom/af_netrom.c | 6 +++--- net/nfc/llcp_sock.c | 4 ++-- net/phonet/pep.c | 12 ++++++------ net/phonet/socket.c | 7 +++---- net/rds/tcp_listen.c | 6 +++++- net/rose/af_rose.c | 6 +++--- net/sctp/socket.c | 8 ++++---- net/smc/af_smc.c | 6 +++--- net/socket.c | 13 ++++++++++--- net/tipc/socket.c | 13 +++++-------- net/unix/af_unix.c | 21 ++++++++++----------- net/vmw_vsock/af_vsock.c | 6 +++--- net/x25/af_x25.c | 4 ++-- 34 files changed, 132 insertions(+), 111 deletions(-) (limited to 'net') diff --git a/crypto/af_alg.c b/crypto/af_alg.c index 5bc6d0fa7498..18cfead0081d 100644 --- a/crypto/af_alg.c +++ b/crypto/af_alg.c @@ -407,7 +407,8 @@ unlock: return err; } -int af_alg_accept(struct sock *sk, struct socket *newsock, bool kern) +int af_alg_accept(struct sock *sk, struct socket *newsock, + struct proto_accept_arg *arg) { struct alg_sock *ask = alg_sk(sk); const struct af_alg_type *type; @@ -422,7 +423,7 @@ int af_alg_accept(struct sock *sk, struct socket *newsock, bool kern) if (!type) goto unlock; - sk2 = sk_alloc(sock_net(sk), PF_ALG, GFP_KERNEL, &alg_proto, kern); + sk2 = sk_alloc(sock_net(sk), PF_ALG, GFP_KERNEL, &alg_proto, arg->kern); err = -ENOMEM; if (!sk2) goto unlock; @@ -468,10 +469,10 @@ unlock: } EXPORT_SYMBOL_GPL(af_alg_accept); -static int alg_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int alg_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { - return af_alg_accept(sock->sk, newsock, kern); + return af_alg_accept(sock->sk, newsock, arg); } static const struct proto_ops alg_proto_ops = { diff --git a/crypto/algif_hash.c b/crypto/algif_hash.c index e24c829d7a01..7c7394d46a23 100644 --- a/crypto/algif_hash.c +++ b/crypto/algif_hash.c @@ -223,8 +223,8 @@ unlock: return err ?: len; } -static int hash_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int hash_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct alg_sock *ask = alg_sk(sk); @@ -252,7 +252,7 @@ static int hash_accept(struct socket *sock, struct socket *newsock, int flags, if (err) goto out_free_state; - err = af_alg_accept(ask->parent, newsock, kern); + err = af_alg_accept(ask->parent, newsock, arg); if (err) goto out_free_state; @@ -355,7 +355,7 @@ static int hash_recvmsg_nokey(struct socket *sock, struct msghdr *msg, } static int hash_accept_nokey(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { int err; @@ -363,7 +363,7 @@ static int hash_accept_nokey(struct socket *sock, struct socket *newsock, if (err) return err; - return hash_accept(sock, newsock, flags, kern); + return hash_accept(sock, newsock, arg); } static struct proto_ops algif_hash_ops_nokey = { diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c index d52593466a79..fd7ed65e0197 100644 --- a/drivers/xen/pvcalls-back.c +++ b/drivers/xen/pvcalls-back.c @@ -517,6 +517,10 @@ static void __pvcalls_back_accept(struct work_struct *work) { struct sockpass_mapping *mappass = container_of( work, struct sockpass_mapping, register_work); + struct proto_accept_arg arg = { + .flags = O_NONBLOCK, + .kern = true, + }; struct sock_mapping *map; struct pvcalls_ioworker *iow; struct pvcalls_fedata *fedata; @@ -548,7 +552,7 @@ static void __pvcalls_back_accept(struct work_struct *work) sock->type = mappass->sock->type; sock->ops = mappass->sock->ops; - ret = inet_accept(mappass->sock, sock, O_NONBLOCK, true); + ret = inet_accept(mappass->sock, sock, &arg); if (ret == -EAGAIN) { sock_release(sock); return; diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index 960080753d3b..2b8fa3e782fb 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -1784,6 +1784,9 @@ static int o2net_accept_one(struct socket *sock, int *more) struct o2nm_node *node = NULL; struct o2nm_node *local_node = NULL; struct o2net_sock_container *sc = NULL; + struct proto_accept_arg arg = { + .flags = O_NONBLOCK, + }; struct o2net_node *nn; unsigned int nofs_flag; @@ -1802,7 +1805,7 @@ static int o2net_accept_one(struct socket *sock, int *more) new_sock->type = sock->type; new_sock->ops = sock->ops; - ret = sock->ops->accept(sock, new_sock, O_NONBLOCK, false); + ret = sock->ops->accept(sock, new_sock, &arg); if (ret < 0) goto out; diff --git a/include/crypto/if_alg.h b/include/crypto/if_alg.h index 78ecaf5db04c..f7b3b93f3a49 100644 --- a/include/crypto/if_alg.h +++ b/include/crypto/if_alg.h @@ -166,7 +166,8 @@ int af_alg_unregister_type(const struct af_alg_type *type); int af_alg_release(struct socket *sock); void af_alg_release_parent(struct sock *sk); -int af_alg_accept(struct sock *sk, struct socket *newsock, bool kern); +int af_alg_accept(struct sock *sk, struct socket *newsock, + struct proto_accept_arg *arg); void af_alg_free_sg(struct af_alg_sgl *sgl); diff --git a/include/linux/net.h b/include/linux/net.h index 15df6d5f27a7..688320b79fcc 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -153,6 +153,7 @@ struct sockaddr; struct msghdr; struct module; struct sk_buff; +struct proto_accept_arg; typedef int (*sk_read_actor_t)(read_descriptor_t *, struct sk_buff *, unsigned int, size_t); typedef int (*skb_read_actor_t)(struct sock *, struct sk_buff *); @@ -171,7 +172,8 @@ struct proto_ops { int (*socketpair)(struct socket *sock1, struct socket *sock2); int (*accept) (struct socket *sock, - struct socket *newsock, int flags, bool kern); + struct socket *newsock, + struct proto_accept_arg *arg); int (*getname) (struct socket *sock, struct sockaddr *addr, int peer); diff --git a/include/net/inet_common.h b/include/net/inet_common.h index f50a644d87a9..c17a6585d0b0 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -29,8 +29,8 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags, int is_sendmsg); int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr, int addr_len, int flags); -int inet_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern); +int inet_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg); void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *newsk); int inet_send_prepare(struct sock *sk); diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 20e7b0c0b3d1..7d6b1254c92d 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -250,7 +250,7 @@ inet_csk_rto_backoff(const struct inet_connection_sock *icsk, return (unsigned long)min_t(u64, when, max_when); } -struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern); +struct sock *inet_csk_accept(struct sock *sk, struct proto_accept_arg *arg); int inet_csk_get_port(struct sock *sk, unsigned short snum); diff --git a/include/net/sock.h b/include/net/sock.h index 0450494a1766..217079b3e3e8 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1194,6 +1194,12 @@ static inline void sk_prot_clear_nulls(struct sock *sk, int size) size - offsetof(struct sock, sk_node.pprev)); } +struct proto_accept_arg { + int flags; + int err; + bool kern; +}; + /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface */ @@ -1208,8 +1214,8 @@ struct proto { int addr_len); int (*disconnect)(struct sock *sk, int flags); - struct sock * (*accept)(struct sock *sk, int flags, int *err, - bool kern); + struct sock * (*accept)(struct sock *sk, + struct proto_accept_arg *arg); int (*ioctl)(struct sock *sk, int cmd, int *karg); @@ -1804,7 +1810,7 @@ int sock_cmsg_send(struct sock *sk, struct msghdr *msg, int sock_no_bind(struct socket *, struct sockaddr *, int); int sock_no_connect(struct socket *, struct sockaddr *, int, int); int sock_no_socketpair(struct socket *, struct socket *); -int sock_no_accept(struct socket *, struct socket *, int, bool); +int sock_no_accept(struct socket *, struct socket *, struct proto_accept_arg *); int sock_no_getname(struct socket *, struct sockaddr *, int); int sock_no_ioctl(struct socket *, unsigned int, unsigned long); int sock_no_listen(struct socket *, int); diff --git a/net/atm/svc.c b/net/atm/svc.c index 36a814f1fbd1..f8137ae693b0 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -324,8 +324,8 @@ out: return error; } -static int svc_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int svc_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sk_buff *skb; @@ -336,7 +336,7 @@ static int svc_accept(struct socket *sock, struct socket *newsock, int flags, lock_sock(sk); - error = svc_create(sock_net(sk), newsock, 0, kern); + error = svc_create(sock_net(sk), newsock, 0, arg->kern); if (error) goto out; @@ -355,7 +355,7 @@ static int svc_accept(struct socket *sock, struct socket *newsock, int flags, error = -sk->sk_err; break; } - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { error = -EAGAIN; break; } diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index 9169efb2f43a..8077cf2ee448 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -1373,8 +1373,8 @@ out_release: return err; } -static int ax25_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int ax25_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sk_buff *skb; struct sock *newsk; @@ -1409,7 +1409,7 @@ static int ax25_accept(struct socket *sock, struct socket *newsock, int flags, if (skb) break; - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { err = -EWOULDBLOCK; break; } diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index ef0cc80b4c0c..2a075119d65d 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1186,7 +1186,7 @@ done: } static int iso_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *ch; @@ -1195,7 +1195,7 @@ static int iso_sock_accept(struct socket *sock, struct socket *newsock, lock_sock(sk); - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 5cc83f906c12..125dddc77452 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -327,7 +327,7 @@ done: } static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *nsk; @@ -336,7 +336,7 @@ static int l2cap_sock_accept(struct socket *sock, struct socket *newsock, lock_sock_nested(sk, L2CAP_NESTING_PARENT); - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index 29aa07e9db9d..37d63d768afb 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -468,8 +468,8 @@ done: return err; } -static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *nsk; @@ -483,7 +483,7 @@ static int rfcomm_sock_accept(struct socket *sock, struct socket *newsock, int f goto done; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index e0ad30862ee4..94c6f2b46279 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -647,7 +647,7 @@ done: } static int sco_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DEFINE_WAIT_FUNC(wait, woken_wake_function); struct sock *sk = sock->sk, *ch; @@ -656,7 +656,7 @@ static int sco_sock_accept(struct socket *sock, struct socket *newsock, lock_sock(sk); - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); BT_DBG("sk %p timeo %ld", sk, timeo); diff --git a/net/core/sock.c b/net/core/sock.c index 8d6e638b5426..8629f9aecf91 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3241,8 +3241,8 @@ int sock_no_socketpair(struct socket *sock1, struct socket *sock2) } EXPORT_SYMBOL(sock_no_socketpair); -int sock_no_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +int sock_no_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { return -EOPNOTSUPP; } diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a7bad18bc8b5..de3449e16b89 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -771,16 +771,16 @@ void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *new * Accept a pending connection. The TCP layer now gives BSD semantics. */ -int inet_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +int inet_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk1 = sock->sk, *sk2; - int err = -EINVAL; /* IPV6_ADDRFORM can change sk->sk_prot under us. */ - sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, flags, &err, kern); + arg->err = -EINVAL; + sk2 = READ_ONCE(sk1->sk_prot)->accept(sk1, arg); if (!sk2) - return err; + return arg->err; lock_sock(sk2); __inet_accept(sock, newsock, sk2); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 3b38610958ee..7734d189c66b 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -661,7 +661,7 @@ static int inet_csk_wait_for_connect(struct sock *sk, long timeo) /* * This will accept the next outstanding connection. */ -struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern) +struct sock *inet_csk_accept(struct sock *sk, struct proto_accept_arg *arg) { struct inet_connection_sock *icsk = inet_csk(sk); struct request_sock_queue *queue = &icsk->icsk_accept_queue; @@ -680,7 +680,7 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern) /* Find already established connection */ if (reqsk_queue_empty(queue)) { - long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + long timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); /* If this is a non blocking socket don't sleep */ error = -EAGAIN; @@ -745,7 +745,7 @@ out: out_err: newsk = NULL; req = NULL; - *err = error; + arg->err = error; goto out; } EXPORT_SYMBOL(inet_csk_accept); diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index c951bb9cc2e0..c3b0b610b0aa 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -795,7 +795,7 @@ done: /* Accept a pending connection */ static int iucv_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DECLARE_WAITQUEUE(wait, current); struct sock *sk = sock->sk, *nsk; @@ -809,7 +809,7 @@ static int iucv_sock_accept(struct socket *sock, struct socket *newsock, goto done; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); /* Wait for an incoming connection */ add_wait_queue_exclusive(sk_sleep(sk), &wait); diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index fde1140d899e..4eb52add7103 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -688,14 +688,13 @@ static void llc_cmsg_rcv(struct msghdr *msg, struct sk_buff *skb) * llc_ui_accept - accept a new incoming connection. * @sock: Socket which connections arrive on. * @newsock: Socket to move incoming connection to. - * @flags: User specified operational flags. - * @kern: If the socket is kernel internal + * @arg: User specified arguments * * Accept a new incoming connection. * Returns 0 upon success, negative otherwise. */ -static int llc_ui_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int llc_ui_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk, *newsk; struct llc_sock *llc, *newllc; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index bb8f96f2b86f..815ce439183c 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3881,11 +3881,10 @@ unlock: } static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { struct mptcp_sock *msk = mptcp_sk(sock->sk); struct sock *ssk, *newsk; - int err; pr_debug("msk=%p", msk); @@ -3897,9 +3896,9 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, return -EINVAL; pr_debug("ssk=%p, listener=%p", ssk, mptcp_subflow_ctx(ssk)); - newsk = inet_csk_accept(ssk, flags, &err, kern); + newsk = inet_csk_accept(ssk, arg); if (!newsk) - return err; + return arg->err; pr_debug("newsk=%p, subflow is mptcp=%d", newsk, sk_is_mptcp(newsk)); if (sk_is_mptcp(newsk)) { @@ -3920,7 +3919,7 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, newsk = new_mptcp_sock; MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_MPCAPABLEPASSIVEACK); - newsk->sk_kern_sock = kern; + newsk->sk_kern_sock = arg->kern; lock_sock(newsk); __inet_accept(sock, newsock, newsk); @@ -3949,7 +3948,7 @@ static int mptcp_stream_accept(struct socket *sock, struct socket *newsock, } } else { tcpfallback: - newsk->sk_kern_sock = kern; + newsk->sk_kern_sock = arg->kern; lock_sock(newsk); __inet_accept(sock, newsock, newsk); /* we are being invoked after accepting a non-mp-capable diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index 104a80b75477..6ee148f0e6d0 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -772,8 +772,8 @@ out_release: return err; } -static int nr_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int nr_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sk_buff *skb; struct sock *newsk; @@ -805,7 +805,7 @@ static int nr_accept(struct socket *sock, struct socket *newsock, int flags, if (skb) break; - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { err = -EWOULDBLOCK; break; } diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c index d5344563e525..57a2f97004e1 100644 --- a/net/nfc/llcp_sock.c +++ b/net/nfc/llcp_sock.c @@ -447,7 +447,7 @@ struct sock *nfc_llcp_accept_dequeue(struct sock *parent, } static int llcp_sock_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { DECLARE_WAITQUEUE(wait, current); struct sock *sk = sock->sk, *new_sk; @@ -463,7 +463,7 @@ static int llcp_sock_accept(struct socket *sock, struct socket *newsock, goto error; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); /* Wait for an incoming connection. */ add_wait_queue_exclusive(sk_sleep(sk), &wait); diff --git a/net/phonet/pep.c b/net/phonet/pep.c index 3dd5f52bc1b5..53a858478e22 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -759,8 +759,8 @@ static void pep_sock_close(struct sock *sk, long timeout) sock_put(sk); } -static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, - bool kern) +static struct sock *pep_sock_accept(struct sock *sk, + struct proto_accept_arg *arg) { struct pep_sock *pn = pep_sk(sk), *newpn; struct sock *newsk = NULL; @@ -772,8 +772,8 @@ static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, u8 pipe_handle, enabled, n_sb; u8 aligned = 0; - skb = skb_recv_datagram(sk, (flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, - errp); + skb = skb_recv_datagram(sk, (arg->flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, + &arg->err); if (!skb) return NULL; @@ -836,7 +836,7 @@ static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, /* Create a new to-be-accepted sock */ newsk = sk_alloc(sock_net(sk), PF_PHONET, GFP_KERNEL, sk->sk_prot, - kern); + arg->kern); if (!newsk) { pep_reject_conn(sk, skb, PN_PIPE_ERR_OVERLOAD, GFP_KERNEL); err = -ENOBUFS; @@ -878,7 +878,7 @@ static struct sock *pep_sock_accept(struct sock *sk, int flags, int *errp, drop: release_sock(sk); kfree_skb(skb); - *errp = err; + arg->err = err; return newsk; } diff --git a/net/phonet/socket.c b/net/phonet/socket.c index 1018340d89a7..5ce0b3ee5def 100644 --- a/net/phonet/socket.c +++ b/net/phonet/socket.c @@ -292,18 +292,17 @@ out: } static int pn_socket_accept(struct socket *sock, struct socket *newsock, - int flags, bool kern) + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sock *newsk; - int err; if (unlikely(sk->sk_state != TCP_LISTEN)) return -EINVAL; - newsk = sk->sk_prot->accept(sk, flags, &err, kern); + newsk = sk->sk_prot->accept(sk, arg); if (!newsk) - return err; + return arg->err; lock_sock(newsk); sock_graft(newsk, newsock); diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 05008ce5c421..d89bd8d0c354 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -105,6 +105,10 @@ int rds_tcp_accept_one(struct socket *sock) int conn_state; struct rds_conn_path *cp; struct in6_addr *my_addr, *peer_addr; + struct proto_accept_arg arg = { + .flags = O_NONBLOCK, + .kern = true, + }; #if !IS_ENABLED(CONFIG_IPV6) struct in6_addr saddr, daddr; #endif @@ -119,7 +123,7 @@ int rds_tcp_accept_one(struct socket *sock) if (ret) goto out; - ret = sock->ops->accept(sock, new_sock, O_NONBLOCK, true); + ret = sock->ops->accept(sock, new_sock, &arg); if (ret < 0) goto out; diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index ef81d019b20f..59050caab65c 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -919,8 +919,8 @@ out_release: return err; } -static int rose_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int rose_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sk_buff *skb; struct sock *newsk; @@ -953,7 +953,7 @@ static int rose_accept(struct socket *sock, struct socket *newsock, int flags, if (skb) break; - if (flags & O_NONBLOCK) { + if (arg->flags & O_NONBLOCK) { err = -EWOULDBLOCK; break; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 64196b1dce1d..c009383369b2 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4847,7 +4847,7 @@ static int sctp_disconnect(struct sock *sk, int flags) * descriptor will be returned from accept() to represent the newly * formed association. */ -static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) +static struct sock *sctp_accept(struct sock *sk, struct proto_accept_arg *arg) { struct sctp_sock *sp; struct sctp_endpoint *ep; @@ -4871,7 +4871,7 @@ static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) goto out; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); error = sctp_wait_for_accept(sk, timeo); if (error) @@ -4882,7 +4882,7 @@ static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) */ asoc = list_entry(ep->asocs.next, struct sctp_association, asocs); - newsk = sp->pf->create_accept_sk(sk, asoc, kern); + newsk = sp->pf->create_accept_sk(sk, asoc, arg->kern); if (!newsk) { error = -ENOMEM; goto out; @@ -4899,7 +4899,7 @@ static struct sock *sctp_accept(struct sock *sk, int flags, int *err, bool kern) out: release_sock(sk); - *err = error; + arg->err = error; return newsk; } diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 9389f0cfa374..e50a286fd0fb 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -2689,7 +2689,7 @@ out: } static int smc_accept(struct socket *sock, struct socket *new_sock, - int flags, bool kern) + struct proto_accept_arg *arg) { struct sock *sk = sock->sk, *nsk; DECLARE_WAITQUEUE(wait, current); @@ -2708,7 +2708,7 @@ static int smc_accept(struct socket *sock, struct socket *new_sock, } /* Wait for an incoming connection */ - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); add_wait_queue_exclusive(sk_sleep(sk), &wait); while (!(nsk = smc_accept_dequeue(sk, new_sock))) { set_current_state(TASK_INTERRUPTIBLE); @@ -2735,7 +2735,7 @@ static int smc_accept(struct socket *sock, struct socket *new_sock, if (rc) goto out; - if (lsmc->sockopt_defer_accept && !(flags & O_NONBLOCK)) { + if (lsmc->sockopt_defer_accept && !(arg->flags & O_NONBLOCK)) { /* wait till data arrives on the socket */ timeo = msecs_to_jiffies(lsmc->sockopt_defer_accept * MSEC_PER_SEC); diff --git a/net/socket.c b/net/socket.c index 01a71ae10c35..6ff5f21d9633 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1898,6 +1898,9 @@ struct file *do_accept(struct file *file, unsigned file_flags, struct file *newfile; int err, len; struct sockaddr_storage address; + struct proto_accept_arg arg = { + .flags = file_flags, + }; const struct proto_ops *ops; sock = sock_from_file(file); @@ -1926,8 +1929,8 @@ struct file *do_accept(struct file *file, unsigned file_flags, if (err) goto out_fd; - err = ops->accept(sock, newsock, sock->file->f_flags | file_flags, - false); + arg.flags |= sock->file->f_flags; + err = ops->accept(sock, newsock, &arg); if (err < 0) goto out_fd; @@ -3580,6 +3583,10 @@ int kernel_accept(struct socket *sock, struct socket **newsock, int flags) { struct sock *sk = sock->sk; const struct proto_ops *ops = READ_ONCE(sock->ops); + struct proto_accept_arg arg = { + .flags = flags, + .kern = true, + }; int err; err = sock_create_lite(sk->sk_family, sk->sk_type, sk->sk_protocol, @@ -3587,7 +3594,7 @@ int kernel_accept(struct socket *sock, struct socket **newsock, int flags) if (err < 0) goto done; - err = ops->accept(sock, *newsock, flags, true); + err = ops->accept(sock, *newsock, &arg); if (err < 0) { sock_release(*newsock); *newsock = NULL; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 798397b6811e..2d58ecae4e21 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -146,8 +146,6 @@ static void tipc_data_ready(struct sock *sk); static void tipc_write_space(struct sock *sk); static void tipc_sock_destruct(struct sock *sk); static int tipc_release(struct socket *sock); -static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, - bool kern); static void tipc_sk_timeout(struct timer_list *t); static int tipc_sk_publish(struct tipc_sock *tsk, struct tipc_uaddr *ua); static int tipc_sk_withdraw(struct tipc_sock *tsk, struct tipc_uaddr *ua); @@ -2711,13 +2709,12 @@ static int tipc_wait_for_accept(struct socket *sock, long timeo) * tipc_accept - wait for connection request * @sock: listening socket * @new_sock: new socket that is to be connected - * @flags: file-related flags associated with socket - * @kern: caused by kernel or by userspace? + * @arg: arguments for accept * * Return: 0 on success, errno otherwise */ -static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, - bool kern) +static int tipc_accept(struct socket *sock, struct socket *new_sock, + struct proto_accept_arg *arg) { struct sock *new_sk, *sk = sock->sk; struct tipc_sock *new_tsock; @@ -2733,14 +2730,14 @@ static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags, res = -EINVAL; goto exit; } - timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + timeo = sock_rcvtimeo(sk, arg->flags & O_NONBLOCK); res = tipc_wait_for_accept(sock, timeo); if (res) goto exit; buf = skb_peek(&sk->sk_receive_queue); - res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, kern); + res = tipc_sk_create(sock_net(sock->sk), new_sock, 0, arg->kern); if (res) goto exit; security_sk_clone(sock->sk, new_sock->sk); diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index dc1651541723..26e3b5f1ee46 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -755,7 +755,7 @@ static int unix_bind(struct socket *, struct sockaddr *, int); static int unix_stream_connect(struct socket *, struct sockaddr *, int addr_len, int flags); static int unix_socketpair(struct socket *, struct socket *); -static int unix_accept(struct socket *, struct socket *, int, bool); +static int unix_accept(struct socket *, struct socket *, struct proto_accept_arg *arg); static int unix_getname(struct socket *, struct sockaddr *, int); static __poll_t unix_poll(struct file *, struct socket *, poll_table *); static __poll_t unix_dgram_poll(struct file *, struct socket *, @@ -1689,19 +1689,18 @@ static void unix_sock_inherit_flags(const struct socket *old, set_bit(SOCK_PASSSEC, &new->flags); } -static int unix_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int unix_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sk_buff *skb; struct sock *tsk; - int err; - err = -EOPNOTSUPP; + arg->err = -EOPNOTSUPP; if (sock->type != SOCK_STREAM && sock->type != SOCK_SEQPACKET) goto out; - err = -EINVAL; + arg->err = -EINVAL; if (sk->sk_state != TCP_LISTEN) goto out; @@ -1709,12 +1708,12 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, * so that no locks are necessary. */ - skb = skb_recv_datagram(sk, (flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, - &err); + skb = skb_recv_datagram(sk, (arg->flags & O_NONBLOCK) ? MSG_DONTWAIT : 0, + &arg->err); if (!skb) { /* This means receive shutdown. */ - if (err == 0) - err = -EINVAL; + if (arg->err == 0) + arg->err = -EINVAL; goto out; } @@ -1732,7 +1731,7 @@ static int unix_accept(struct socket *sock, struct socket *newsock, int flags, return 0; out: - return err; + return arg->err; } diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 54ba7316f808..4b040285aa78 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1500,8 +1500,8 @@ out: return err; } -static int vsock_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int vsock_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *listener; int err; @@ -1528,7 +1528,7 @@ static int vsock_accept(struct socket *sock, struct socket *newsock, int flags, /* Wait for children sockets to appear; these are the new sockets * created upon connection establishment. */ - timeout = sock_rcvtimeo(listener, flags & O_NONBLOCK); + timeout = sock_rcvtimeo(listener, arg->flags & O_NONBLOCK); prepare_to_wait(sk_sleep(listener), &wait, TASK_INTERRUPTIBLE); while ((connected = vsock_dequeue_accept(listener)) == NULL && diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index d18d51412cc0..8dda4178497c 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -871,8 +871,8 @@ static int x25_wait_for_data(struct sock *sk, long timeout) return rc; } -static int x25_accept(struct socket *sock, struct socket *newsock, int flags, - bool kern) +static int x25_accept(struct socket *sock, struct socket *newsock, + struct proto_accept_arg *arg) { struct sock *sk = sock->sk; struct sock *newsk; -- cgit v1.2.3 From 0645fbe760afcc5332c858d1cbf416bf77ef3c29 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 May 2024 09:31:05 -0600 Subject: net: have do_accept() take a struct proto_accept_arg argument In preparation for passing in more information via this API, change do_accept() to take a proto_accept_arg struct pointer rather than just the file flags separately. No functional changes in this patch. Acked-by: Jakub Kicinski Signed-off-by: Jens Axboe --- include/linux/socket.h | 3 ++- io_uring/net.c | 6 ++++-- net/socket.c | 12 +++++------- 3 files changed, 11 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/include/linux/socket.h b/include/linux/socket.h index 139c330ccf2c..89d16b90370b 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -16,6 +16,7 @@ struct cred; struct socket; struct sock; struct sk_buff; +struct proto_accept_arg; #define __sockaddr_check_size(size) \ BUILD_BUG_ON(((size) > sizeof(struct __kernel_sockaddr_storage))) @@ -433,7 +434,7 @@ extern int __sys_recvfrom(int fd, void __user *ubuf, size_t size, extern int __sys_sendto(int fd, void __user *buff, size_t len, unsigned int flags, struct sockaddr __user *addr, int addr_len); -extern struct file *do_accept(struct file *file, unsigned file_flags, +extern struct file *do_accept(struct file *file, struct proto_accept_arg *arg, struct sockaddr __user *upeer_sockaddr, int __user *upeer_addrlen, int flags); extern int __sys_accept4(int fd, struct sockaddr __user *upeer_sockaddr, diff --git a/io_uring/net.c b/io_uring/net.c index 070dea9a4eda..d4d1fc93635c 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -1528,8 +1528,10 @@ int io_accept(struct io_kiocb *req, unsigned int issue_flags) { struct io_accept *accept = io_kiocb_to_cmd(req, struct io_accept); bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; - unsigned int file_flags = force_nonblock ? O_NONBLOCK : 0; bool fixed = !!accept->file_slot; + struct proto_accept_arg arg = { + .flags = force_nonblock ? O_NONBLOCK : 0, + }; struct file *file; int ret, fd; @@ -1543,7 +1545,7 @@ retry: if (unlikely(fd < 0)) return fd; } - file = do_accept(req->file, file_flags, accept->addr, accept->addr_len, + file = do_accept(req->file, &arg, accept->addr, accept->addr_len, accept->flags); if (IS_ERR(file)) { if (!fixed) diff --git a/net/socket.c b/net/socket.c index 6ff5f21d9633..e416920e9399 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1890,7 +1890,7 @@ SYSCALL_DEFINE2(listen, int, fd, int, backlog) return __sys_listen(fd, backlog); } -struct file *do_accept(struct file *file, unsigned file_flags, +struct file *do_accept(struct file *file, struct proto_accept_arg *arg, struct sockaddr __user *upeer_sockaddr, int __user *upeer_addrlen, int flags) { @@ -1898,9 +1898,6 @@ struct file *do_accept(struct file *file, unsigned file_flags, struct file *newfile; int err, len; struct sockaddr_storage address; - struct proto_accept_arg arg = { - .flags = file_flags, - }; const struct proto_ops *ops; sock = sock_from_file(file); @@ -1929,8 +1926,8 @@ struct file *do_accept(struct file *file, unsigned file_flags, if (err) goto out_fd; - arg.flags |= sock->file->f_flags; - err = ops->accept(sock, newsock, &arg); + arg->flags |= sock->file->f_flags; + err = ops->accept(sock, newsock, arg); if (err < 0) goto out_fd; @@ -1956,6 +1953,7 @@ out_fd: static int __sys_accept4_file(struct file *file, struct sockaddr __user *upeer_sockaddr, int __user *upeer_addrlen, int flags) { + struct proto_accept_arg arg = { }; struct file *newfile; int newfd; @@ -1969,7 +1967,7 @@ static int __sys_accept4_file(struct file *file, struct sockaddr __user *upeer_s if (unlikely(newfd < 0)) return newfd; - newfile = do_accept(file, 0, upeer_sockaddr, upeer_addrlen, + newfile = do_accept(file, &arg, upeer_sockaddr, upeer_addrlen, flags); if (IS_ERR(newfile)) { put_unused_fd(newfd); -- cgit v1.2.3 From 7951e36ac620a9ba1bae0ac0ddd62d2e82285725 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 9 May 2024 09:35:01 -0600 Subject: net: pass back whether socket was empty post accept This adds an 'is_empty' argument to struct proto_accept_arg, which can be used to pass back information on whether or not the given socket has more connections to accept post the one just accepted. To utilize this information, the caller should initialize the 'is_empty' field to, eg, -1 and then check for 0/1 after the accept. If the field has been set, the caller knows whether there are more pending connections or not. If the field remains -1 after the accept call, the protocol doesn't support passing back this information. This patch wires it up for ipv4/6 TCP. Acked-by: Jakub Kicinski Signed-off-by: Jens Axboe --- include/net/sock.h | 1 + net/ipv4/inet_connection_sock.c | 1 + 2 files changed, 2 insertions(+) (limited to 'net') diff --git a/include/net/sock.h b/include/net/sock.h index 217079b3e3e8..5f4d0629348f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1197,6 +1197,7 @@ static inline void sk_prot_clear_nulls(struct sock *sk, int size) struct proto_accept_arg { int flags; int err; + int is_empty; bool kern; }; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 7734d189c66b..d81f74ce0f02 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -692,6 +692,7 @@ struct sock *inet_csk_accept(struct sock *sk, struct proto_accept_arg *arg) goto out_err; } req = reqsk_queue_remove(queue, sk); + arg->is_empty = reqsk_queue_empty(queue); newsk = req->sk; if (sk->sk_protocol == IPPROTO_TCP && -- cgit v1.2.3 From 22a32557758a7100e46dfa8f383a401125e60b16 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Fri, 10 May 2024 20:25:01 +0800 Subject: tcp: rstreason: handle timewait cases in the receive path There are two possible cases where TCP layer can send an RST. Since they happen in the same place, I think using one independent reason is enough to identify this special situation. Signed-off-by: Jason Xing Link: https://lore.kernel.org/r/20240510122502.27850-5-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- include/net/rstreason.h | 5 +++++ net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 3 files changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/rstreason.h b/include/net/rstreason.h index fc1b99702771..7ae5bb55559b 100644 --- a/include/net/rstreason.h +++ b/include/net/rstreason.h @@ -15,6 +15,7 @@ FN(TCP_FLAGS) \ FN(TCP_OLD_ACK) \ FN(TCP_ABORT_ON_DATA) \ + FN(TCP_TIMEWAIT_SOCKET) \ FN(MPTCP_RST_EUNSPEC) \ FN(MPTCP_RST_EMPTCP) \ FN(MPTCP_RST_ERESOURCE) \ @@ -72,6 +73,10 @@ enum sk_rst_reason { */ SK_RST_REASON_TCP_ABORT_ON_DATA, + /* Here start with the independent reasons */ + /** @SK_RST_REASON_TCP_TIMEWAIT_SOCKET: happen on the timewait socket */ + SK_RST_REASON_TCP_TIMEWAIT_SOCKET, + /* Copy from include/uapi/linux/mptcp.h. * These reset fields will not be changed since they adhere to * RFC 8684. So do not touch them. I'm going to list each definition diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 108a438dc247..30ef0c8f5e92 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2427,7 +2427,7 @@ do_time_wait: tcp_v4_timewait_ack(sk, skb); break; case TCP_TW_RST: - tcp_v4_send_reset(sk, skb, sk_rst_convert_drop_reason(drop_reason)); + tcp_v4_send_reset(sk, skb, SK_RST_REASON_TCP_TIMEWAIT_SOCKET); inet_twsk_deschedule_put(inet_twsk(sk)); goto discard_it; case TCP_TW_SUCCESS:; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 7f6693e794bd..4c3605485b68 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1999,7 +1999,7 @@ do_time_wait: tcp_v6_timewait_ack(sk, skb); break; case TCP_TW_RST: - tcp_v6_send_reset(sk, skb, sk_rst_convert_drop_reason(drop_reason)); + tcp_v6_send_reset(sk, skb, SK_RST_REASON_TCP_TIMEWAIT_SOCKET); inet_twsk_deschedule_put(inet_twsk(sk)); goto discard_it; case TCP_TW_SUCCESS: -- cgit v1.2.3 From 11f46ea9814d2f0a3c7a5bc749d7619e47251f75 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Fri, 10 May 2024 20:25:02 +0800 Subject: tcp: rstreason: fully support in tcp_check_req() We're going to send an RST due to invalid syn packet which is already checked whether 1) it is in sequence, 2) it is a retransmitted skb. As RFC 793 says, if the state of socket is not CLOSED/LISTEN/SYN-SENT, then we should send an RST when receiving bad syn packet: "fourth, check the SYN bit,...If the SYN is in the window it is an error, send a reset" Signed-off-by: Jason Xing Link: https://lore.kernel.org/r/20240510122502.27850-6-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- include/net/rstreason.h | 8 ++++++++ net/ipv4/tcp_minisocks.c | 2 +- 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/rstreason.h b/include/net/rstreason.h index 7ae5bb55559b..2575c85d7f7a 100644 --- a/include/net/rstreason.h +++ b/include/net/rstreason.h @@ -16,6 +16,7 @@ FN(TCP_OLD_ACK) \ FN(TCP_ABORT_ON_DATA) \ FN(TCP_TIMEWAIT_SOCKET) \ + FN(INVALID_SYN) \ FN(MPTCP_RST_EUNSPEC) \ FN(MPTCP_RST_EMPTCP) \ FN(MPTCP_RST_ERESOURCE) \ @@ -76,6 +77,13 @@ enum sk_rst_reason { /* Here start with the independent reasons */ /** @SK_RST_REASON_TCP_TIMEWAIT_SOCKET: happen on the timewait socket */ SK_RST_REASON_TCP_TIMEWAIT_SOCKET, + /** + * @SK_RST_REASON_INVALID_SYN: receive bad syn packet + * RFC 793 says if the state is not CLOSED/LISTEN/SYN-SENT then + * "fourth, check the SYN bit,...If the SYN is in the window it is + * an error, send a reset" + */ + SK_RST_REASON_INVALID_SYN, /* Copy from include/uapi/linux/mptcp.h. * These reset fields will not be changed since they adhere to diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 7d543569a180..b93619b2384b 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -879,7 +879,7 @@ embryonic_reset: * avoid becoming vulnerable to outside attack aiming at * resetting legit local connections. */ - req->rsk_ops->send_reset(sk, skb, SK_RST_REASON_NOT_SPECIFIED); + req->rsk_ops->send_reset(sk, skb, SK_RST_REASON_INVALID_SYN); } else if (fastopen) { /* received a valid RST pkt */ reqsk_fastopen_remove(sk, req, true); tcp_reset(sk, skb); -- cgit v1.2.3 From a65198136eaa15b74ee0abf73f12ef83d469a334 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:25 -0700 Subject: mptcp: SO_KEEPALIVE: fix getsockopt support SO_KEEPALIVE support has to be set on each subflow: on each TCP socket, where sk_prot->keepalive is defined. Technically, nothing has to be done on the MPTCP socket. That's why mptcp_sol_socket_sync_intval() was called instead of mptcp_sol_socket_intval(). Except that when nothing is done on the MPTCP socket, the getsockopt(SO_KEEPALIVE), handled in net/core/sock.c:sk_getsockopt(), will not know if SO_KEEPALIVE has been set on the different subflows or not. The fix is simple: simply call mptcp_sol_socket_intval() which will end up calling net/core/sock.c:sk_setsockopt() where the SOCK_KEEPOPEN flag will be set, the one used in sk_getsockopt(). So now, getsockopt(SO_KEEPALIVE) on an MPTCP socket will return the same value as the one previously set with setsockopt(SO_KEEPALIVE). Fixes: 1b3e7ede1365 ("mptcp: setsockopt: handle SO_KEEPALIVE and SO_PRIORITY") Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-2-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index eaa3b79651a4..5ab506c96609 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -181,8 +181,6 @@ static int mptcp_setsockopt_sol_socket_int(struct mptcp_sock *msk, int optname, switch (optname) { case SO_KEEPALIVE: - mptcp_sol_socket_sync_intval(msk, optname, val); - return 0; case SO_DEBUG: case SO_MARK: case SO_PRIORITY: -- cgit v1.2.3 From bd11dc4fb969ec148e50cd87f88a78246dbc4d0b Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:26 -0700 Subject: mptcp: fix full TCP keep-alive support SO_KEEPALIVE support has been added a while ago, as part of a series "adding SOL_SOCKET" support. To have a full control of this keep-alive feature, it is important to also support TCP_KEEP* socket options at the SOL_TCP level. Supporting them on the setsockopt() part is easy, it is just a matter of remembering each value in the MPTCP sock structure, and calling tcp_sock_set_keep*() helpers on each subflow. If the value is not modified (0), calling these helpers will not do anything. For the getsockopt() part, the corresponding value from the MPTCP sock structure or the default one is simply returned. All of this is very similar to other TCP_* socket options supported by MPTCP. It looks important for kernels supporting SO_KEEPALIVE, to also support TCP_KEEP* options as well: some apps seem to (wrongly) consider that if the former is supported, the latter ones will be supported as well. But also, not having this simple and isolated change is preventing MPTCP support in some apps, and libraries like GoLang [1]. This is why this patch is seen as a fix. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/383 Fixes: 1b3e7ede1365 ("mptcp: setsockopt: handle SO_KEEPALIVE and SO_PRIORITY") Link: https://github.com/golang/go/issues/56539 [1] Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-3-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.h | 3 +++ net/mptcp/sockopt.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 61 insertions(+) (limited to 'net') diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index cfc5f9c3f113..4dcce3641d1d 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -312,6 +312,9 @@ struct mptcp_sock { free_first:1, rcvspace_init:1; u32 notsent_lowat; + int keepalive_cnt; + int keepalive_idle; + int keepalive_intvl; struct work_struct work; struct sk_buff *ooo_last_skb; struct rb_root out_of_order_queue; diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 5ab506c96609..fcca9433c858 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -622,6 +622,31 @@ static int mptcp_setsockopt_sol_tcp_congestion(struct mptcp_sock *msk, sockptr_t return ret; } +static int __mptcp_setsockopt_set_val(struct mptcp_sock *msk, int max, + int (*set_val)(struct sock *, int), + int *msk_val, int val) +{ + struct mptcp_subflow_context *subflow; + int err = 0; + + mptcp_for_each_subflow(msk, subflow) { + struct sock *ssk = mptcp_subflow_tcp_sock(subflow); + int ret; + + lock_sock(ssk); + ret = set_val(ssk, val); + err = err ? : ret; + release_sock(ssk); + } + + if (!err) { + *msk_val = val; + sockopt_seq_inc(msk); + } + + return err; +} + static int __mptcp_setsockopt_sol_tcp_cork(struct mptcp_sock *msk, int val) { struct mptcp_subflow_context *subflow; @@ -818,6 +843,22 @@ static int mptcp_setsockopt_sol_tcp(struct mptcp_sock *msk, int optname, case TCP_NODELAY: ret = __mptcp_setsockopt_sol_tcp_nodelay(msk, val); break; + case TCP_KEEPIDLE: + ret = __mptcp_setsockopt_set_val(msk, MAX_TCP_KEEPIDLE, + &tcp_sock_set_keepidle_locked, + &msk->keepalive_idle, val); + break; + case TCP_KEEPINTVL: + ret = __mptcp_setsockopt_set_val(msk, MAX_TCP_KEEPINTVL, + &tcp_sock_set_keepintvl, + &msk->keepalive_intvl, val); + break; + case TCP_KEEPCNT: + ret = __mptcp_setsockopt_set_val(msk, MAX_TCP_KEEPCNT, + &tcp_sock_set_keepcnt, + &msk->keepalive_cnt, + val); + break; default: ret = -ENOPROTOOPT; } @@ -1326,6 +1367,8 @@ static int mptcp_put_int_option(struct mptcp_sock *msk, char __user *optval, static int mptcp_getsockopt_sol_tcp(struct mptcp_sock *msk, int optname, char __user *optval, int __user *optlen) { + struct sock *sk = (void *)msk; + switch (optname) { case TCP_ULP: case TCP_CONGESTION: @@ -1344,6 +1387,18 @@ static int mptcp_getsockopt_sol_tcp(struct mptcp_sock *msk, int optname, return mptcp_put_int_option(msk, optval, optlen, msk->cork); case TCP_NODELAY: return mptcp_put_int_option(msk, optval, optlen, msk->nodelay); + case TCP_KEEPIDLE: + return mptcp_put_int_option(msk, optval, optlen, + msk->keepalive_idle ? : + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_keepalive_time) / HZ); + case TCP_KEEPINTVL: + return mptcp_put_int_option(msk, optval, optlen, + msk->keepalive_intvl ? : + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_keepalive_intvl) / HZ); + case TCP_KEEPCNT: + return mptcp_put_int_option(msk, optval, optlen, + msk->keepalive_cnt ? : + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_keepalive_probes)); case TCP_NOTSENT_LOWAT: return mptcp_put_int_option(msk, optval, optlen, msk->notsent_lowat); case TCP_IS_MPTCP: @@ -1463,6 +1518,9 @@ static void sync_socket_options(struct mptcp_sock *msk, struct sock *ssk) tcp_set_congestion_control(ssk, msk->ca_name, false, true); __tcp_sock_set_cork(ssk, !!msk->cork); __tcp_sock_set_nodelay(ssk, !!msk->nodelay); + tcp_sock_set_keepidle_locked(ssk, msk->keepalive_idle); + tcp_sock_set_keepintvl(ssk, msk->keepalive_intvl); + tcp_sock_set_keepcnt(ssk, msk->keepalive_cnt); inet_assign_bit(TRANSPARENT, ssk, inet_test_bit(TRANSPARENT, sk)); inet_assign_bit(FREEBIND, ssk, inet_test_bit(FREEBIND, sk)); -- cgit v1.2.3 From ce5f6f71b029f6a96267a7d876da4a055d7d1611 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:27 -0700 Subject: mptcp: sockopt: info: stop early if no buffer Up to recently, it has been recommended to use getsockopt(MPTCP_INFO) to check if a fallback to TCP happened, or if the client requested to use MPTCP. In this case, the userspace app is only interested by the returned value of the getsocktop() call, and can then give 0 for the option length, and NULL for the buffer address. An easy optimisation is then to stop early, and avoid filling a local buffer -- which now requires two different locks -- if it is not needed. Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-4-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'net') diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index fcca9433c858..a77b33488176 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -999,6 +999,10 @@ static int mptcp_getsockopt_info(struct mptcp_sock *msk, char __user *optval, in if (get_user(len, optlen)) return -EFAULT; + /* When used only to check if a fallback to TCP happened. */ + if (len == 0) + return 0; + len = min_t(unsigned int, len, sizeof(struct mptcp_info)); mptcp_diag_fill_info(msk, &m_info); -- cgit v1.2.3 From 73c900aa3660dae2050af456d1031d098ab4cb1d Mon Sep 17 00:00:00 2001 From: Gregory Detal Date: Mon, 13 May 2024 18:13:28 -0700 Subject: mptcp: add net.mptcp.available_schedulers The sysctl lists the available schedulers that can be set using net.mptcp.scheduler similarly to net.ipv4.tcp_available_congestion_control. Signed-off-by: Gregory Detal Reviewed-by: Mat Martineau Tested-by: Geliang Tang Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-5-martineau@kernel.org Signed-off-by: Jakub Kicinski --- include/net/mptcp.h | 3 +++ net/mptcp/ctrl.c | 27 ++++++++++++++++++++++++++- net/mptcp/protocol.h | 1 + net/mptcp/sched.c | 22 ++++++++++++++++++++++ 4 files changed, 52 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/include/net/mptcp.h b/include/net/mptcp.h index fb996124b3d5..0bc4ab03f487 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -97,6 +97,9 @@ struct mptcp_out_options { }; #define MPTCP_SCHED_NAME_MAX 16 +#define MPTCP_SCHED_MAX 128 +#define MPTCP_SCHED_BUF_MAX (MPTCP_SCHED_NAME_MAX * MPTCP_SCHED_MAX) + #define MPTCP_SUBFLOWS_MAX 8 struct mptcp_sched_data { diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c index 49abf80b3fad..542555ba474c 100644 --- a/net/mptcp/ctrl.c +++ b/net/mptcp/ctrl.c @@ -133,6 +133,24 @@ static int proc_scheduler(struct ctl_table *ctl, int write, return ret; } +static int proc_available_schedulers(struct ctl_table *ctl, + int write, void *buffer, + size_t *lenp, loff_t *ppos) +{ + struct ctl_table tbl = { .maxlen = MPTCP_SCHED_BUF_MAX, }; + int ret; + + tbl.data = kmalloc(tbl.maxlen, GFP_USER); + if (!tbl.data) + return -ENOMEM; + + mptcp_get_available_schedulers(tbl.data, MPTCP_SCHED_BUF_MAX); + ret = proc_dostring(&tbl, write, buffer, lenp, ppos); + kfree(tbl.data); + + return ret; +} + static struct ctl_table mptcp_sysctl_table[] = { { .procname = "enabled", @@ -187,6 +205,12 @@ static struct ctl_table mptcp_sysctl_table[] = { .mode = 0644, .proc_handler = proc_scheduler, }, + { + .procname = "available_schedulers", + .maxlen = MPTCP_SCHED_BUF_MAX, + .mode = 0644, + .proc_handler = proc_available_schedulers, + }, { .procname = "close_timeout", .maxlen = sizeof(unsigned int), @@ -214,7 +238,8 @@ static int mptcp_pernet_new_table(struct net *net, struct mptcp_pernet *pernet) table[4].data = &pernet->stale_loss_cnt; table[5].data = &pernet->pm_type; table[6].data = &pernet->scheduler; - table[7].data = &pernet->close_timeout; + /* table[7] is for available_schedulers which is read-only info */ + table[8].data = &pernet->close_timeout; hdr = register_net_sysctl_sz(net, MPTCP_SYSCTL_PATH, table, ARRAY_SIZE(mptcp_sysctl_table)); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 4dcce3641d1d..39fc47e6b88a 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -686,6 +686,7 @@ unsigned int mptcp_stale_loss_cnt(const struct net *net); unsigned int mptcp_close_timeout(const struct sock *sk); int mptcp_get_pm_type(const struct net *net); const char *mptcp_get_scheduler(const struct net *net); +void mptcp_get_available_schedulers(char *buf, size_t maxlen); void __mptcp_subflow_fully_established(struct mptcp_sock *msk, struct mptcp_subflow_context *subflow, const struct mptcp_options_received *mp_opt); diff --git a/net/mptcp/sched.c b/net/mptcp/sched.c index 4ab0693c069c..4a7fd0508ad2 100644 --- a/net/mptcp/sched.c +++ b/net/mptcp/sched.c @@ -51,6 +51,28 @@ struct mptcp_sched_ops *mptcp_sched_find(const char *name) return ret; } +/* Build string with list of available scheduler values. + * Similar to tcp_get_available_congestion_control() + */ +void mptcp_get_available_schedulers(char *buf, size_t maxlen) +{ + struct mptcp_sched_ops *sched; + size_t offs = 0; + + rcu_read_lock(); + spin_lock(&mptcp_sched_list_lock); + list_for_each_entry_rcu(sched, &mptcp_sched_list, list) { + offs += snprintf(buf + offs, maxlen - offs, + "%s%s", + offs == 0 ? "" : " ", sched->name); + + if (WARN_ON_ONCE(offs >= maxlen)) + break; + } + spin_unlock(&mptcp_sched_list_lock); + rcu_read_unlock(); +} + int mptcp_register_scheduler(struct mptcp_sched_ops *sched) { if (!sched->get_subflow) -- cgit v1.2.3 From 5eae7a8202f368a77ff0148729a9fa73cc24033a Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:29 -0700 Subject: mptcp: prefer strscpy over strcpy strcpy() performs no bounds checking on the destination buffer. This could result in linear overflows beyond the end of the buffer, leading to all kinds of misbehaviors. The safe replacement is strscpy() [1]. This is in preparation of a possible future step where all strcpy() uses will be removed in favour of strscpy() [2]. This fixes CheckPatch warnings: WARNING: Prefer strscpy over strcpy Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy [1] Link: https://github.com/KSPP/linux/issues/88 [2] Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-6-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/ctrl.c | 2 +- net/mptcp/protocol.c | 5 +++-- net/mptcp/sockopt.c | 2 +- 3 files changed, 5 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/mptcp/ctrl.c b/net/mptcp/ctrl.c index 542555ba474c..98b1dd498ff6 100644 --- a/net/mptcp/ctrl.c +++ b/net/mptcp/ctrl.c @@ -92,7 +92,7 @@ static void mptcp_pernet_set_defaults(struct mptcp_pernet *pernet) pernet->allow_join_initial_addr_port = 1; pernet->stale_loss_cnt = 4; pernet->pm_type = MPTCP_PM_TYPE_KERNEL; - strcpy(pernet->scheduler, "default"); + strscpy(pernet->scheduler, "default", sizeof(pernet->scheduler)); } #ifdef CONFIG_SYSCTL diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index bb8f96f2b86f..a42494d3a71b 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2814,7 +2814,8 @@ static void mptcp_ca_reset(struct sock *sk) struct inet_connection_sock *icsk = inet_csk(sk); tcp_assign_congestion_control(sk); - strcpy(mptcp_sk(sk)->ca_name, icsk->icsk_ca_ops->name); + strscpy(mptcp_sk(sk)->ca_name, icsk->icsk_ca_ops->name, + sizeof(mptcp_sk(sk)->ca_name)); /* no need to keep a reference to the ops, the name will suffice */ tcp_cleanup_congestion_control(sk); @@ -4169,7 +4170,7 @@ int __init mptcp_proto_v6_init(void) int err; mptcp_v6_prot = mptcp_prot; - strcpy(mptcp_v6_prot.name, "MPTCPv6"); + strscpy(mptcp_v6_prot.name, "MPTCPv6", sizeof(mptcp_v6_prot.name)); mptcp_v6_prot.slab = NULL; mptcp_v6_prot.obj_size = sizeof(struct mptcp6_sock); mptcp_v6_prot.ipv6_pinfo_offset = offsetof(struct mptcp6_sock, np); diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index a77b33488176..f9a4fb17b5b7 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -616,7 +616,7 @@ static int mptcp_setsockopt_sol_tcp_congestion(struct mptcp_sock *msk, sockptr_t } if (ret == 0) - strcpy(msk->ca_name, name); + strscpy(msk->ca_name, name, sizeof(msk->ca_name)); release_sock(sk); return ret; -- cgit v1.2.3 From 00797af95f5eedfcec9b247c7e06efbc789e8f6e Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:30 -0700 Subject: mptcp: remove unnecessary else statements The 'else' statements are not needed here, because their previous 'if' block ends with a 'return'. This fixes CheckPatch warnings: WARNING: else is not generally useful after a break or return Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-7-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/subflow.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) (limited to 'net') diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index c1d13e555d10..612c38570a64 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1119,6 +1119,8 @@ static enum mapping_status get_mapping_status(struct sock *ssk, } if (mpext->data_fin == 1) { + u64 data_fin_seq; + if (data_len == 1) { bool updated = mptcp_update_rcv_data_fin(msk, mpext->data_seq, mpext->dsn64); @@ -1131,26 +1133,26 @@ static enum mapping_status get_mapping_status(struct sock *ssk, */ skb_ext_del(skb, SKB_EXT_MPTCP); return MAPPING_OK; - } else { - if (updated) - mptcp_schedule_work((struct sock *)msk); - - return MAPPING_DATA_FIN; } - } else { - u64 data_fin_seq = mpext->data_seq + data_len - 1; - /* If mpext->data_seq is a 32-bit value, data_fin_seq - * must also be limited to 32 bits. - */ - if (!mpext->dsn64) - data_fin_seq &= GENMASK_ULL(31, 0); + if (updated) + mptcp_schedule_work((struct sock *)msk); - mptcp_update_rcv_data_fin(msk, data_fin_seq, mpext->dsn64); - pr_debug("DATA_FIN with mapping seq=%llu dsn64=%d", - data_fin_seq, mpext->dsn64); + return MAPPING_DATA_FIN; } + data_fin_seq = mpext->data_seq + data_len - 1; + + /* If mpext->data_seq is a 32-bit value, data_fin_seq must also + * be limited to 32 bits. + */ + if (!mpext->dsn64) + data_fin_seq &= GENMASK_ULL(31, 0); + + mptcp_update_rcv_data_fin(msk, data_fin_seq, mpext->dsn64); + pr_debug("DATA_FIN with mapping seq=%llu dsn64=%d", + data_fin_seq, mpext->dsn64); + /* Adjust for DATA_FIN using 1 byte of sequence space */ data_len--; } -- cgit v1.2.3 From 76a86686e3f0ca68b555131ceefa141a57340ed0 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:31 -0700 Subject: mptcp: move mptcp_pm_gen.h's include Nothing from protocol.h depends on mptcp_pm_gen.h, only code from pm_netlink.c and pm_userspace.c depends on it. So this include can be moved where it is needed to avoid a "unused includes" warning. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-8-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_netlink.c | 1 + net/mptcp/pm_userspace.c | 1 + net/mptcp/protocol.h | 2 -- 3 files changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c index 5c17d39146ea..7f53e022e27e 100644 --- a/net/mptcp/pm_netlink.c +++ b/net/mptcp/pm_netlink.c @@ -14,6 +14,7 @@ #include "protocol.h" #include "mib.h" +#include "mptcp_pm_gen.h" static int pm_nl_pernet_id; diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c index 9f5d422d5ef6..f0a4590506c6 100644 --- a/net/mptcp/pm_userspace.c +++ b/net/mptcp/pm_userspace.c @@ -6,6 +6,7 @@ #include "protocol.h" #include "mib.h" +#include "mptcp_pm_gen.h" void mptcp_free_local_addr_list(struct mptcp_sock *msk) { diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 39fc47e6b88a..7aa47e2dd52b 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -14,8 +14,6 @@ #include #include -#include "mptcp_pm_gen.h" - #define MPTCP_SUPPORTED_VERSION 1 /* MPTCP option bits */ -- cgit v1.2.3 From 7fad5b37561175527e92a17bc6a71009b0b718e2 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Mon, 13 May 2024 18:13:32 -0700 Subject: mptcp: include inet_common in mib.h So this file is now self-contained: it can be compiled alone with analytic tools. Reviewed-by: Geliang Tang Signed-off-by: Matthieu Baerts (NGI0) Signed-off-by: Mat Martineau Link: https://lore.kernel.org/r/20240514011335.176158-9-martineau@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/mib.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/mptcp/mib.h b/net/mptcp/mib.h index dd7fd1f246b5..2704afd0dfe4 100644 --- a/net/mptcp/mib.h +++ b/net/mptcp/mib.h @@ -1,5 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0-or-later */ +#include + enum linux_mptcp_mib_field { MPTCP_MIB_NUM = 0, MPTCP_MIB_MPCAPABLEPASSIVE, /* Received SYN with MP_CAPABLE */ -- cgit v1.2.3 From 5c1672705a1a2389f5ad78e0fea6f08ed32d6f18 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 13 May 2024 08:41:55 -0700 Subject: net: revert partially applied PHY topology series The series is causing issues with PHY drivers built as modules. Since it was only partially applied and the merge window has opened let's revert and try again for v6.11. Revert 6916e461e793 ("net: phy: Introduce ethernet link topology representation") Revert 0ec5ed6c130e ("net: sfp: pass the phy_device when disconnecting an sfp module's PHY") Revert e75e4e074c44 ("net: phy: add helpers to handle sfp phy connect/disconnect") Revert fdd353965b52 ("net: sfp: Add helper to return the SFP bus name") Revert 841942bc6212 ("net: ethtool: Allow passing a phy index for some commands") Link: https://lore.kernel.org/all/171242462917.4000.9759453824684907063.git-patchwork-notify@kernel.org/ Link: https://lore.kernel.org/all/20240507102822.2023826-1-maxime.chevallier@bootlin.com/ Link: https://lore.kernel.org/r/20240513154156.104281-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/networking/ethtool-netlink.rst | 7 -- MAINTAINERS | 2 - drivers/net/phy/Makefile | 2 +- drivers/net/phy/marvell-88x2222.c | 2 - drivers/net/phy/marvell.c | 2 - drivers/net/phy/marvell10g.c | 2 - drivers/net/phy/phy_device.c | 55 -------------- drivers/net/phy/phy_link_topology.c | 105 --------------------------- drivers/net/phy/phylink.c | 3 +- drivers/net/phy/qcom/at803x.c | 2 - drivers/net/phy/qcom/qca807x.c | 2 - drivers/net/phy/sfp-bus.c | 15 +--- include/linux/netdevice.h | 4 +- include/linux/phy.h | 6 -- include/linux/phy_link_topology.h | 72 ------------------ include/linux/phy_link_topology_core.h | 25 ------- include/linux/sfp.h | 8 +- include/uapi/linux/ethtool.h | 16 ---- include/uapi/linux/ethtool_netlink.h | 1 - net/core/dev.c | 9 --- net/ethtool/netlink.c | 48 +----------- net/ethtool/netlink.h | 5 -- 22 files changed, 8 insertions(+), 385 deletions(-) delete mode 100644 drivers/net/phy/phy_link_topology.c delete mode 100644 include/linux/phy_link_topology.h delete mode 100644 include/linux/phy_link_topology_core.h (limited to 'net') diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst index 8bc71f249448..160bfb0ae8ba 100644 --- a/Documentation/networking/ethtool-netlink.rst +++ b/Documentation/networking/ethtool-netlink.rst @@ -57,7 +57,6 @@ Structure of this header is ``ETHTOOL_A_HEADER_DEV_INDEX`` u32 device ifindex ``ETHTOOL_A_HEADER_DEV_NAME`` string device name ``ETHTOOL_A_HEADER_FLAGS`` u32 flags common for all requests - ``ETHTOOL_A_HEADER_PHY_INDEX`` u32 phy device index ============================== ====== ============================= ``ETHTOOL_A_HEADER_DEV_INDEX`` and ``ETHTOOL_A_HEADER_DEV_NAME`` identify the @@ -82,12 +81,6 @@ the behaviour is backward compatible, i.e. requests from old clients not aware of the flag should be interpreted the way the client expects. A client must not set flags it does not understand. -``ETHTOOL_A_HEADER_PHY_INDEX`` identifies the Ethernet PHY the message relates to. -As there are numerous commands that are related to PHY configuration, and because -there may be more than one PHY on the link, the PHY index can be passed in the -request for the commands that needs it. It is, however, not mandatory, and if it -is not passed for commands that target a PHY, the net_device.phydev pointer -is used. Bit sets ======== diff --git a/MAINTAINERS b/MAINTAINERS index 40eb26e6c60c..b4ed6480e919 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8027,8 +8027,6 @@ F: include/linux/mii.h F: include/linux/of_net.h F: include/linux/phy.h F: include/linux/phy_fixed.h -F: include/linux/phy_link_topology.h -F: include/linux/phy_link_topology_core.h F: include/linux/phylib_stubs.h F: include/linux/platform_data/mdio-bcm-unimac.h F: include/linux/platform_data/mdio-gpio.h diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index 1d8be374915f..202ed7f450da 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -2,7 +2,7 @@ # Makefile for Linux PHY drivers libphy-y := phy.o phy-c45.o phy-core.o phy_device.o \ - linkmode.o phy_link_topology.o + linkmode.o mdio-bus-y += mdio_bus.o mdio_device.o ifdef CONFIG_MDIO_DEVICE diff --git a/drivers/net/phy/marvell-88x2222.c b/drivers/net/phy/marvell-88x2222.c index 0b777cdd7078..b88398e6872b 100644 --- a/drivers/net/phy/marvell-88x2222.c +++ b/drivers/net/phy/marvell-88x2222.c @@ -553,8 +553,6 @@ static const struct sfp_upstream_ops sfp_phy_ops = { .link_down = mv2222_sfp_link_down, .attach = phy_sfp_attach, .detach = phy_sfp_detach, - .connect_phy = phy_sfp_connect_phy, - .disconnect_phy = phy_sfp_disconnect_phy, }; static int mv2222_probe(struct phy_device *phydev) diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 9964bf3dea2f..b89fbffa6a93 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -3613,8 +3613,6 @@ static const struct sfp_upstream_ops m88e1510_sfp_ops = { .module_remove = m88e1510_sfp_remove, .attach = phy_sfp_attach, .detach = phy_sfp_detach, - .connect_phy = phy_sfp_connect_phy, - .disconnect_phy = phy_sfp_disconnect_phy, }; static int m88e1510_probe(struct phy_device *phydev) diff --git a/drivers/net/phy/marvell10g.c b/drivers/net/phy/marvell10g.c index 6642eb642d4b..ad43e280930c 100644 --- a/drivers/net/phy/marvell10g.c +++ b/drivers/net/phy/marvell10g.c @@ -503,8 +503,6 @@ static int mv3310_sfp_insert(void *upstream, const struct sfp_eeprom_id *id) static const struct sfp_upstream_ops mv3310_sfp_ops = { .attach = phy_sfp_attach, .detach = phy_sfp_detach, - .connect_phy = phy_sfp_connect_phy, - .disconnect_phy = phy_sfp_disconnect_phy, .module_insert = mv3310_sfp_insert, }; diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 616bd7ba46cb..6c6ec9475709 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -29,7 +29,6 @@ #include #include #include -#include #include #include #include @@ -277,14 +276,6 @@ static void phy_mdio_device_remove(struct mdio_device *mdiodev) static struct phy_driver genphy_driver; -static struct phy_link_topology *phy_get_link_topology(struct phy_device *phydev) -{ - if (phydev->attached_dev) - return phydev->attached_dev->link_topo; - - return NULL; -} - static LIST_HEAD(phy_fixup_list); static DEFINE_MUTEX(phy_fixup_lock); @@ -1378,46 +1369,6 @@ phy_standalone_show(struct device *dev, struct device_attribute *attr, } static DEVICE_ATTR_RO(phy_standalone); -/** - * phy_sfp_connect_phy - Connect the SFP module's PHY to the upstream PHY - * @upstream: pointer to the upstream phy device - * @phy: pointer to the SFP module's phy device - * - * This helper allows keeping track of PHY devices on the link. It adds the - * SFP module's phy to the phy namespace of the upstream phy - */ -int phy_sfp_connect_phy(void *upstream, struct phy_device *phy) -{ - struct phy_device *phydev = upstream; - struct phy_link_topology *topo = phy_get_link_topology(phydev); - - if (topo) - return phy_link_topo_add_phy(topo, phy, PHY_UPSTREAM_PHY, phydev); - - return 0; -} -EXPORT_SYMBOL(phy_sfp_connect_phy); - -/** - * phy_sfp_disconnect_phy - Disconnect the SFP module's PHY from the upstream PHY - * @upstream: pointer to the upstream phy device - * @phy: pointer to the SFP module's phy device - * - * This helper allows keeping track of PHY devices on the link. It removes the - * SFP module's phy to the phy namespace of the upstream phy. As the module phy - * will be destroyed, re-inserting the same module will add a new phy with a - * new index. - */ -void phy_sfp_disconnect_phy(void *upstream, struct phy_device *phy) -{ - struct phy_device *phydev = upstream; - struct phy_link_topology *topo = phy_get_link_topology(phydev); - - if (topo) - phy_link_topo_del_phy(topo, phy); -} -EXPORT_SYMBOL(phy_sfp_disconnect_phy); - /** * phy_sfp_attach - attach the SFP bus to the PHY upstream network device * @upstream: pointer to the phy device @@ -1560,11 +1511,6 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev, if (phydev->sfp_bus_attached) dev->sfp_bus = phydev->sfp_bus; - - err = phy_link_topo_add_phy(dev->link_topo, phydev, - PHY_UPSTREAM_MAC, dev); - if (err) - goto error; } /* Some Ethernet drivers try to connect to a PHY device before @@ -1992,7 +1938,6 @@ void phy_detach(struct phy_device *phydev) if (dev) { phydev->attached_dev->phydev = NULL; phydev->attached_dev = NULL; - phy_link_topo_del_phy(dev->link_topo, phydev); } phydev->phylink = NULL; diff --git a/drivers/net/phy/phy_link_topology.c b/drivers/net/phy/phy_link_topology.c deleted file mode 100644 index 985941c5c558..000000000000 --- a/drivers/net/phy/phy_link_topology.c +++ /dev/null @@ -1,105 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0+ -/* - * Infrastructure to handle all PHY devices connected to a given netdev, - * either directly or indirectly attached. - * - * Copyright (c) 2023 Maxime Chevallier - */ - -#include -#include -#include -#include -#include - -struct phy_link_topology *phy_link_topo_create(struct net_device *dev) -{ - struct phy_link_topology *topo; - - topo = kzalloc(sizeof(*topo), GFP_KERNEL); - if (!topo) - return ERR_PTR(-ENOMEM); - - xa_init_flags(&topo->phys, XA_FLAGS_ALLOC1); - topo->next_phy_index = 1; - - return topo; -} - -void phy_link_topo_destroy(struct phy_link_topology *topo) -{ - if (!topo) - return; - - xa_destroy(&topo->phys); - kfree(topo); -} - -int phy_link_topo_add_phy(struct phy_link_topology *topo, - struct phy_device *phy, - enum phy_upstream upt, void *upstream) -{ - struct phy_device_node *pdn; - int ret; - - pdn = kzalloc(sizeof(*pdn), GFP_KERNEL); - if (!pdn) - return -ENOMEM; - - pdn->phy = phy; - switch (upt) { - case PHY_UPSTREAM_MAC: - pdn->upstream.netdev = (struct net_device *)upstream; - if (phy_on_sfp(phy)) - pdn->parent_sfp_bus = pdn->upstream.netdev->sfp_bus; - break; - case PHY_UPSTREAM_PHY: - pdn->upstream.phydev = (struct phy_device *)upstream; - if (phy_on_sfp(phy)) - pdn->parent_sfp_bus = pdn->upstream.phydev->sfp_bus; - break; - default: - ret = -EINVAL; - goto err; - } - pdn->upstream_type = upt; - - /* Attempt to re-use a previously allocated phy_index */ - if (phy->phyindex) { - ret = xa_insert(&topo->phys, phy->phyindex, pdn, GFP_KERNEL); - - /* Errors could be either -ENOMEM or -EBUSY. If the phy has an - * index, and there's another entry at the same index, this is - * unexpected and we still error-out - */ - if (ret) - goto err; - return 0; - } - - ret = xa_alloc_cyclic(&topo->phys, &phy->phyindex, pdn, xa_limit_32b, - &topo->next_phy_index, GFP_KERNEL); - if (ret) - goto err; - - return 0; - -err: - kfree(pdn); - return ret; -} -EXPORT_SYMBOL_GPL(phy_link_topo_add_phy); - -void phy_link_topo_del_phy(struct phy_link_topology *topo, - struct phy_device *phy) -{ - struct phy_device_node *pdn = xa_erase(&topo->phys, phy->phyindex); - - /* We delete the PHY from the topology, however we don't re-set the - * phy->phyindex field. If the PHY isn't gone, we can re-assign it the - * same index next time it's added back to the topology - */ - - kfree(pdn); -} -EXPORT_SYMBOL_GPL(phy_link_topo_del_phy); diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index b7e5c669dc8e..994471fad833 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -3411,8 +3411,7 @@ static int phylink_sfp_connect_phy(void *upstream, struct phy_device *phy) return ret; } -static void phylink_sfp_disconnect_phy(void *upstream, - struct phy_device *phydev) +static void phylink_sfp_disconnect_phy(void *upstream) { phylink_disconnect_phy(upstream); } diff --git a/drivers/net/phy/qcom/at803x.c b/drivers/net/phy/qcom/at803x.c index 105602581a03..c8f83e5f78ab 100644 --- a/drivers/net/phy/qcom/at803x.c +++ b/drivers/net/phy/qcom/at803x.c @@ -770,8 +770,6 @@ static const struct sfp_upstream_ops at8031_sfp_ops = { .attach = phy_sfp_attach, .detach = phy_sfp_detach, .module_insert = at8031_sfp_insert, - .connect_phy = phy_sfp_connect_phy, - .disconnect_phy = phy_sfp_disconnect_phy, }; static int at8031_parse_dt(struct phy_device *phydev) diff --git a/drivers/net/phy/qcom/qca807x.c b/drivers/net/phy/qcom/qca807x.c index 5eb0ab1cb70e..672c6929119a 100644 --- a/drivers/net/phy/qcom/qca807x.c +++ b/drivers/net/phy/qcom/qca807x.c @@ -699,8 +699,6 @@ static const struct sfp_upstream_ops qca807x_sfp_ops = { .detach = phy_sfp_detach, .module_insert = qca807x_sfp_insert, .module_remove = qca807x_sfp_remove, - .connect_phy = phy_sfp_connect_phy, - .disconnect_phy = phy_sfp_disconnect_phy, }; static int qca807x_probe(struct phy_device *phydev) diff --git a/drivers/net/phy/sfp-bus.c b/drivers/net/phy/sfp-bus.c index 37c85f1e6534..2f44fc51848f 100644 --- a/drivers/net/phy/sfp-bus.c +++ b/drivers/net/phy/sfp-bus.c @@ -487,7 +487,7 @@ static void sfp_unregister_bus(struct sfp_bus *bus) bus->socket_ops->stop(bus->sfp); bus->socket_ops->detach(bus->sfp); if (bus->phydev && ops && ops->disconnect_phy) - ops->disconnect_phy(bus->upstream, bus->phydev); + ops->disconnect_phy(bus->upstream); } bus->registered = false; } @@ -743,7 +743,7 @@ void sfp_remove_phy(struct sfp_bus *bus) const struct sfp_upstream_ops *ops = sfp_get_upstream_ops(bus); if (ops && ops->disconnect_phy) - ops->disconnect_phy(bus->upstream, bus->phydev); + ops->disconnect_phy(bus->upstream); bus->phydev = NULL; } EXPORT_SYMBOL_GPL(sfp_remove_phy); @@ -860,14 +860,3 @@ void sfp_unregister_socket(struct sfp_bus *bus) sfp_bus_put(bus); } EXPORT_SYMBOL_GPL(sfp_unregister_socket); - -const char *sfp_get_name(struct sfp_bus *bus) -{ - ASSERT_RTNL(); - - if (bus->sfp_dev) - return dev_name(bus->sfp_dev); - - return NULL; -} -EXPORT_SYMBOL_GPL(sfp_get_name); diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index cf261fb89d73..d20c6c99eb88 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -40,6 +40,7 @@ #include #endif #include + #include #include #include @@ -51,7 +52,6 @@ #include #include #include -#include struct netpoll_info; struct device; @@ -1975,7 +1975,6 @@ enum netdev_reg_state { * @fcoe_ddp_xid: Max exchange id for FCoE LRO by ddp * * @priomap: XXX: need comments on this one - * @link_topo: Physical link topology tracking attached PHYs * @phydev: Physical device may attach itself * for hardware timestamping * @sfp_bus: attached &struct sfp_bus structure. @@ -2368,7 +2367,6 @@ struct net_device { #if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) struct netprio_map __rcu *priomap; #endif - struct phy_link_topology *link_topo; struct phy_device *phydev; struct sfp_bus *sfp_bus; struct lock_class_key *qdisc_tx_busylock; diff --git a/include/linux/phy.h b/include/linux/phy.h index 3ddfe7fe781a..e6e83304558e 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -550,9 +550,6 @@ struct macsec_ops; * @drv: Pointer to the driver for this PHY instance * @devlink: Create a link between phy dev and mac dev, if the external phy * used by current mac interface is managed by another mac interface. - * @phyindex: Unique id across the phy's parent tree of phys to address the PHY - * from userspace, similar to ifindex. A zero index means the PHY - * wasn't assigned an id yet. * @phy_id: UID for this device found during discovery * @c45_ids: 802.3-c45 Device Identifiers if is_c45. * @is_c45: Set to true if this PHY uses clause 45 addressing. @@ -653,7 +650,6 @@ struct phy_device { struct device_link *devlink; - u32 phyindex; u32 phy_id; struct phy_c45_device_ids c45_ids; @@ -1758,8 +1754,6 @@ int phy_suspend(struct phy_device *phydev); int phy_resume(struct phy_device *phydev); int __phy_resume(struct phy_device *phydev); int phy_loopback(struct phy_device *phydev, bool enable); -int phy_sfp_connect_phy(void *upstream, struct phy_device *phy); -void phy_sfp_disconnect_phy(void *upstream, struct phy_device *phy); void phy_sfp_attach(void *upstream, struct sfp_bus *bus); void phy_sfp_detach(void *upstream, struct sfp_bus *bus); int phy_sfp_probe(struct phy_device *phydev, diff --git a/include/linux/phy_link_topology.h b/include/linux/phy_link_topology.h deleted file mode 100644 index 6b79feb607e7..000000000000 --- a/include/linux/phy_link_topology.h +++ /dev/null @@ -1,72 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -/* - * PHY device list allow maintaining a list of PHY devices that are - * part of a netdevice's link topology. PHYs can for example be chained, - * as is the case when using a PHY that exposes an SFP module, on which an - * SFP transceiver that embeds a PHY is connected. - * - * This list can then be used by userspace to leverage individual PHY - * capabilities. - */ -#ifndef __PHY_LINK_TOPOLOGY_H -#define __PHY_LINK_TOPOLOGY_H - -#include -#include - -struct xarray; -struct phy_device; -struct net_device; -struct sfp_bus; - -struct phy_device_node { - enum phy_upstream upstream_type; - - union { - struct net_device *netdev; - struct phy_device *phydev; - } upstream; - - struct sfp_bus *parent_sfp_bus; - - struct phy_device *phy; -}; - -struct phy_link_topology { - struct xarray phys; - u32 next_phy_index; -}; - -static inline struct phy_device * -phy_link_topo_get_phy(struct phy_link_topology *topo, u32 phyindex) -{ - struct phy_device_node *pdn = xa_load(&topo->phys, phyindex); - - if (pdn) - return pdn->phy; - - return NULL; -} - -#if IS_REACHABLE(CONFIG_PHYLIB) -int phy_link_topo_add_phy(struct phy_link_topology *topo, - struct phy_device *phy, - enum phy_upstream upt, void *upstream); - -void phy_link_topo_del_phy(struct phy_link_topology *lt, struct phy_device *phy); - -#else -static inline int phy_link_topo_add_phy(struct phy_link_topology *topo, - struct phy_device *phy, - enum phy_upstream upt, void *upstream) -{ - return 0; -} - -static inline void phy_link_topo_del_phy(struct phy_link_topology *topo, - struct phy_device *phy) -{ -} -#endif - -#endif /* __PHY_LINK_TOPOLOGY_H */ diff --git a/include/linux/phy_link_topology_core.h b/include/linux/phy_link_topology_core.h deleted file mode 100644 index 0a6479055745..000000000000 --- a/include/linux/phy_link_topology_core.h +++ /dev/null @@ -1,25 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __PHY_LINK_TOPOLOGY_CORE_H -#define __PHY_LINK_TOPOLOGY_CORE_H - -struct phy_link_topology; - -#if IS_REACHABLE(CONFIG_PHYLIB) - -struct phy_link_topology *phy_link_topo_create(struct net_device *dev); -void phy_link_topo_destroy(struct phy_link_topology *topo); - -#else - -static inline struct phy_link_topology *phy_link_topo_create(struct net_device *dev) -{ - return NULL; -} - -static inline void phy_link_topo_destroy(struct phy_link_topology *topo) -{ -} - -#endif - -#endif /* __PHY_LINK_TOPOLOGY_CORE_H */ diff --git a/include/linux/sfp.h b/include/linux/sfp.h index 5ebc57f78c95..a45da7eef9a2 100644 --- a/include/linux/sfp.h +++ b/include/linux/sfp.h @@ -544,7 +544,7 @@ struct sfp_upstream_ops { void (*link_down)(void *priv); void (*link_up)(void *priv); int (*connect_phy)(void *priv, struct phy_device *); - void (*disconnect_phy)(void *priv, struct phy_device *); + void (*disconnect_phy)(void *priv); }; #if IS_ENABLED(CONFIG_SFP) @@ -570,7 +570,6 @@ struct sfp_bus *sfp_bus_find_fwnode(const struct fwnode_handle *fwnode); int sfp_bus_add_upstream(struct sfp_bus *bus, void *upstream, const struct sfp_upstream_ops *ops); void sfp_bus_del_upstream(struct sfp_bus *bus); -const char *sfp_get_name(struct sfp_bus *bus); #else static inline int sfp_parse_port(struct sfp_bus *bus, const struct sfp_eeprom_id *id, @@ -649,11 +648,6 @@ static inline int sfp_bus_add_upstream(struct sfp_bus *bus, void *upstream, static inline void sfp_bus_del_upstream(struct sfp_bus *bus) { } - -static inline const char *sfp_get_name(struct sfp_bus *bus) -{ - return NULL; -} #endif #endif diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index 041e09c3515d..8733a3117902 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -2323,20 +2323,4 @@ struct ethtool_link_settings { * __u32 map_lp_advertising[link_mode_masks_nwords]; */ }; - -/** - * enum phy_upstream - Represents the upstream component a given PHY device - * is connected to, as in what is on the other end of the MII bus. Most PHYs - * will be attached to an Ethernet MAC controller, but in some cases, there's - * an intermediate PHY used as a media-converter, which will driver another - * MII interface as its output. - * @PHY_UPSTREAM_MAC: Upstream component is a MAC (a switch port, - * or ethernet controller) - * @PHY_UPSTREAM_PHY: Upstream component is a PHY (likely a media converter) - */ -enum phy_upstream { - PHY_UPSTREAM_MAC, - PHY_UPSTREAM_PHY, -}; - #endif /* _UAPI_LINUX_ETHTOOL_H */ diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h index f17dbe54bf5e..b49b804b9495 100644 --- a/include/uapi/linux/ethtool_netlink.h +++ b/include/uapi/linux/ethtool_netlink.h @@ -132,7 +132,6 @@ enum { ETHTOOL_A_HEADER_DEV_INDEX, /* u32 */ ETHTOOL_A_HEADER_DEV_NAME, /* string */ ETHTOOL_A_HEADER_FLAGS, /* u32 - ETHTOOL_FLAG_* */ - ETHTOOL_A_HEADER_PHY_INDEX, /* u32 */ /* add new constants above here */ __ETHTOOL_A_HEADER_CNT, diff --git a/net/core/dev.c b/net/core/dev.c index d2ce91a334c1..e1bb6d7856d9 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -158,7 +158,6 @@ #include #include #include -#include #include "dev.h" #include "net-sysfs.h" @@ -10998,12 +10997,6 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name, #ifdef CONFIG_NET_SCHED hash_init(dev->qdisc_hash); #endif - dev->link_topo = phy_link_topo_create(dev); - if (IS_ERR(dev->link_topo)) { - dev->link_topo = NULL; - goto free_all; - } - dev->priv_flags = IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM; setup(dev); @@ -11092,8 +11085,6 @@ void free_netdev(struct net_device *dev) free_percpu(dev->xdp_bulkq); dev->xdp_bulkq = NULL; - phy_link_topo_destroy(dev->link_topo); - /* Compatibility with error handling in drivers */ if (dev->reg_state == NETREG_UNINITIALIZED || dev->reg_state == NETREG_DUMMY) { diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c index 563e94e0cbd8..bd04f28d5cf4 100644 --- a/net/ethtool/netlink.c +++ b/net/ethtool/netlink.c @@ -4,7 +4,6 @@ #include #include #include "netlink.h" -#include static struct genl_family ethtool_genl_family; @@ -31,24 +30,6 @@ const struct nla_policy ethnl_header_policy_stats[] = { ETHTOOL_FLAGS_STATS), }; -const struct nla_policy ethnl_header_policy_phy[] = { - [ETHTOOL_A_HEADER_DEV_INDEX] = { .type = NLA_U32 }, - [ETHTOOL_A_HEADER_DEV_NAME] = { .type = NLA_NUL_STRING, - .len = ALTIFNAMSIZ - 1 }, - [ETHTOOL_A_HEADER_FLAGS] = NLA_POLICY_MASK(NLA_U32, - ETHTOOL_FLAGS_BASIC), - [ETHTOOL_A_HEADER_PHY_INDEX] = NLA_POLICY_MIN(NLA_U32, 1), -}; - -const struct nla_policy ethnl_header_policy_phy_stats[] = { - [ETHTOOL_A_HEADER_DEV_INDEX] = { .type = NLA_U32 }, - [ETHTOOL_A_HEADER_DEV_NAME] = { .type = NLA_NUL_STRING, - .len = ALTIFNAMSIZ - 1 }, - [ETHTOOL_A_HEADER_FLAGS] = NLA_POLICY_MASK(NLA_U32, - ETHTOOL_FLAGS_STATS), - [ETHTOOL_A_HEADER_PHY_INDEX] = NLA_POLICY_MIN(NLA_U32, 1), -}; - int ethnl_ops_begin(struct net_device *dev) { int ret; @@ -108,9 +89,8 @@ int ethnl_parse_header_dev_get(struct ethnl_req_info *req_info, const struct nlattr *header, struct net *net, struct netlink_ext_ack *extack, bool require_dev) { - struct nlattr *tb[ARRAY_SIZE(ethnl_header_policy_phy)]; + struct nlattr *tb[ARRAY_SIZE(ethnl_header_policy)]; const struct nlattr *devname_attr; - struct phy_device *phydev = NULL; struct net_device *dev = NULL; u32 flags = 0; int ret; @@ -124,7 +104,7 @@ int ethnl_parse_header_dev_get(struct ethnl_req_info *req_info, /* No validation here, command policy should have a nested policy set * for the header, therefore validation should have already been done. */ - ret = nla_parse_nested(tb, ARRAY_SIZE(ethnl_header_policy_phy) - 1, header, + ret = nla_parse_nested(tb, ARRAY_SIZE(ethnl_header_policy) - 1, header, NULL, extack); if (ret < 0) return ret; @@ -165,30 +145,6 @@ int ethnl_parse_header_dev_get(struct ethnl_req_info *req_info, return -EINVAL; } - if (dev) { - if (tb[ETHTOOL_A_HEADER_PHY_INDEX]) { - struct nlattr *phy_id; - - phy_id = tb[ETHTOOL_A_HEADER_PHY_INDEX]; - phydev = phy_link_topo_get_phy(dev->link_topo, - nla_get_u32(phy_id)); - if (!phydev) { - NL_SET_BAD_ATTR(extack, phy_id); - return -ENODEV; - } - } else { - /* If we need a PHY but no phy index is specified, fallback - * to dev->phydev - */ - phydev = dev->phydev; - } - } else if (tb[ETHTOOL_A_HEADER_PHY_INDEX]) { - NL_SET_ERR_MSG_ATTR(extack, header, - "can't target a PHY without a netdev"); - return -EINVAL; - } - - req_info->phydev = phydev; req_info->dev = dev; req_info->flags = flags; return 0; diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h index d57a890b5d9e..9a333a8d04c1 100644 --- a/net/ethtool/netlink.h +++ b/net/ethtool/netlink.h @@ -250,7 +250,6 @@ static inline unsigned int ethnl_reply_header_size(void) * @dev: network device the request is for (may be null) * @dev_tracker: refcount tracker for @dev reference * @flags: request flags common for all request types - * @phydev: phy_device connected to @dev this request is for (may be null) * * This is a common base for request specific structures holding data from * parsed userspace request. These always embed struct ethnl_req_info at @@ -260,7 +259,6 @@ struct ethnl_req_info { struct net_device *dev; netdevice_tracker dev_tracker; u32 flags; - struct phy_device *phydev; }; static inline void ethnl_parse_header_dev_put(struct ethnl_req_info *req_info) @@ -397,12 +395,9 @@ extern const struct ethnl_request_ops ethnl_rss_request_ops; extern const struct ethnl_request_ops ethnl_plca_cfg_request_ops; extern const struct ethnl_request_ops ethnl_plca_status_request_ops; extern const struct ethnl_request_ops ethnl_mm_request_ops; -extern const struct ethnl_request_ops ethnl_phy_request_ops; extern const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_FLAGS + 1]; extern const struct nla_policy ethnl_header_policy_stats[ETHTOOL_A_HEADER_FLAGS + 1]; -extern const struct nla_policy ethnl_header_policy_phy[ETHTOOL_A_HEADER_PHY_INDEX + 1]; -extern const struct nla_policy ethnl_header_policy_phy_stats[ETHTOOL_A_HEADER_PHY_INDEX + 1]; extern const struct nla_policy ethnl_strset_get_policy[ETHTOOL_A_STRSET_COUNTS_ONLY + 1]; extern const struct nla_policy ethnl_linkinfo_get_policy[ETHTOOL_A_LINKINFO_HEADER + 1]; extern const struct nla_policy ethnl_linkinfo_set_policy[ETHTOOL_A_LINKINFO_TP_MDIX_CTRL + 1]; -- cgit v1.2.3 From 7c2cc5b1db509fb6e4a20c182cfbaf94255e4a7c Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 28 Mar 2024 15:46:01 -0400 Subject: Bluetooth: Add proper definitions for scan interval and window This adds proper definitions for scan interval and window and then make use of them instead their values. Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 18 ++++++++++++++++-- net/bluetooth/hci_core.c | 16 ++++++++-------- 2 files changed, 24 insertions(+), 10 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index e8f581f3f3ce..9fc1eccad9bb 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -2235,8 +2235,22 @@ void hci_mgmt_chan_unregister(struct hci_mgmt_chan *c); /* These LE scan and inquiry parameters were chosen according to LE General * Discovery Procedure specification. */ -#define DISCOV_LE_SCAN_WIN 0x12 -#define DISCOV_LE_SCAN_INT 0x12 +#define DISCOV_LE_SCAN_WIN 0x0012 /* 11.25 msec */ +#define DISCOV_LE_SCAN_INT 0x0012 /* 11.25 msec */ +#define DISCOV_LE_SCAN_INT_FAST 0x0060 /* 60 msec */ +#define DISCOV_LE_SCAN_WIN_FAST 0x0030 /* 30 msec */ +#define DISCOV_LE_SCAN_INT_CONN 0x0060 /* 60 msec */ +#define DISCOV_LE_SCAN_WIN_CONN 0x0060 /* 60 msec */ +#define DISCOV_LE_SCAN_INT_SLOW1 0x0800 /* 1.28 sec */ +#define DISCOV_LE_SCAN_WIN_SLOW1 0x0012 /* 11.25 msec */ +#define DISCOV_LE_SCAN_INT_SLOW2 0x1000 /* 2.56 sec */ +#define DISCOV_LE_SCAN_WIN_SLOW2 0x0024 /* 22.5 msec */ +#define DISCOV_CODED_SCAN_INT_FAST 0x0120 /* 180 msec */ +#define DISCOV_CODED_SCAN_WIN_FAST 0x0090 /* 90 msec */ +#define DISCOV_CODED_SCAN_INT_SLOW1 0x1800 /* 3.84 sec */ +#define DISCOV_CODED_SCAN_WIN_SLOW1 0x0036 /* 33.75 msec */ +#define DISCOV_CODED_SCAN_INT_SLOW2 0x3000 /* 7.68 sec */ +#define DISCOV_CODED_SCAN_WIN_SLOW2 0x006c /* 67.5 msec */ #define DISCOV_LE_TIMEOUT 10240 /* msec */ #define DISCOV_INTERLEAVED_TIMEOUT 5120 /* msec */ #define DISCOV_INTERLEAVED_INQUIRY_LEN 0x04 diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index bc5086423ab8..d35015219c26 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -2523,16 +2523,16 @@ struct hci_dev *hci_alloc_dev_priv(int sizeof_priv) hdev->le_adv_channel_map = 0x07; hdev->le_adv_min_interval = 0x0800; hdev->le_adv_max_interval = 0x0800; - hdev->le_scan_interval = 0x0060; - hdev->le_scan_window = 0x0030; - hdev->le_scan_int_suspend = 0x0400; - hdev->le_scan_window_suspend = 0x0012; + hdev->le_scan_interval = DISCOV_LE_SCAN_INT_FAST; + hdev->le_scan_window = DISCOV_LE_SCAN_WIN_FAST; + hdev->le_scan_int_suspend = DISCOV_LE_SCAN_INT_SLOW1; + hdev->le_scan_window_suspend = DISCOV_LE_SCAN_WIN_SLOW1; hdev->le_scan_int_discovery = DISCOV_LE_SCAN_INT; hdev->le_scan_window_discovery = DISCOV_LE_SCAN_WIN; - hdev->le_scan_int_adv_monitor = 0x0060; - hdev->le_scan_window_adv_monitor = 0x0030; - hdev->le_scan_int_connect = 0x0060; - hdev->le_scan_window_connect = 0x0060; + hdev->le_scan_int_adv_monitor = DISCOV_LE_SCAN_INT_FAST; + hdev->le_scan_window_adv_monitor = DISCOV_LE_SCAN_WIN_FAST; + hdev->le_scan_int_connect = DISCOV_LE_SCAN_INT_CONN; + hdev->le_scan_window_connect = DISCOV_LE_SCAN_WIN_CONN; hdev->le_conn_min_interval = 0x0018; hdev->le_conn_max_interval = 0x0028; hdev->le_conn_latency = 0x0000; -- cgit v1.2.3 From 2e2515c1ba384ae44f6bf13dd64b9a0a950798c4 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Thu, 28 Mar 2024 17:40:53 -0400 Subject: Bluetooth: hci_event: Set DISCOVERY_FINDING on SCAN_ENABLED This makes sure that discovery state is properly synchronized otherwise reports may not generate MGMT DeviceFound events as it would be assumed that it was not initiated by a discovery session. Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 11 +++++++++-- net/bluetooth/hci_event.c | 3 +-- 2 files changed, 10 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index d35015219c26..34cd05856401 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -149,8 +149,6 @@ void hci_discovery_set_state(struct hci_dev *hdev, int state) { int old_state = hdev->discovery.state; - BT_DBG("%s state %u -> %u", hdev->name, hdev->discovery.state, state); - if (old_state == state) return; @@ -166,6 +164,13 @@ void hci_discovery_set_state(struct hci_dev *hdev, int state) case DISCOVERY_STARTING: break; case DISCOVERY_FINDING: + /* If discovery was not started then it was initiated by the + * MGMT interface so no MGMT event shall be generated either + */ + if (old_state != DISCOVERY_STARTING) { + hdev->discovery.state = old_state; + return; + } mgmt_discovering(hdev, 1); break; case DISCOVERY_RESOLVING: @@ -173,6 +178,8 @@ void hci_discovery_set_state(struct hci_dev *hdev, int state) case DISCOVERY_STOPPING: break; } + + bt_dev_dbg(hdev, "state %u -> %u", old_state, state); } void hci_inquiry_cache_flush(struct hci_dev *hdev) diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index d72d238c1656..71fedb2d4b48 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1779,8 +1779,7 @@ static void le_set_scan_enable_complete(struct hci_dev *hdev, u8 enable) hci_dev_set_flag(hdev, HCI_LE_SCAN); if (hdev->le_scan_type == LE_SCAN_ACTIVE) clear_pending_adv_report(hdev); - if (hci_dev_test_flag(hdev, HCI_MESH)) - hci_discovery_set_state(hdev, DISCOVERY_FINDING); + hci_discovery_set_state(hdev, DISCOVERY_FINDING); break; case LE_SCAN_DISABLE: -- cgit v1.2.3 From 311527e9dafdcae0c5a20d62f4f84ad01b33b5f4 Mon Sep 17 00:00:00 2001 From: Iulia Tanasescu Date: Tue, 2 Apr 2024 14:39:30 +0300 Subject: Bluetooth: ISO: Make iso_get_sock_listen generic This makes iso_get_sock_listen more generic, to return matching socket in the state provided as argument. Signed-off-by: Iulia Tanasescu Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/bluetooth.h | 2 +- net/bluetooth/iso.c | 75 ++++++++++++++++++++++----------------- 2 files changed, 43 insertions(+), 34 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index eaec5d6caa29..b3228bd6cd6b 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -285,7 +285,7 @@ void bt_err_ratelimited(const char *fmt, ...); bt_err_ratelimited("%s: " fmt, bt_dev_name(hdev), ##__VA_ARGS__) /* Connection and socket states */ -enum { +enum bt_sock_state { BT_CONNECTED = 1, /* Equal to TCP_ESTABLISHED to make net code happy */ BT_OPEN, BT_BOUND, diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index ef0cc80b4c0c..89dee805c60d 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -85,8 +85,9 @@ static void iso_sock_disconn(struct sock *sk); typedef bool (*iso_sock_match_t)(struct sock *sk, void *data); -static struct sock *iso_get_sock_listen(bdaddr_t *src, bdaddr_t *dst, - iso_sock_match_t match, void *data); +static struct sock *iso_get_sock(bdaddr_t *src, bdaddr_t *dst, + enum bt_sock_state state, + iso_sock_match_t match, void *data); /* ---- ISO timers ---- */ #define ISO_CONN_TIMEOUT (HZ * 40) @@ -233,10 +234,11 @@ static void iso_conn_del(struct hci_conn *hcon, int err) * terminated are not processed anymore. */ if (test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) { - parent = iso_get_sock_listen(&hcon->src, - &hcon->dst, - iso_match_conn_sync_handle, - hcon); + parent = iso_get_sock(&hcon->src, + &hcon->dst, + BT_LISTEN, + iso_match_conn_sync_handle, + hcon); if (parent) { set_bit(BT_SK_PA_SYNC_TERM, @@ -581,22 +583,23 @@ static struct sock *__iso_get_sock_listen_by_sid(bdaddr_t *ba, bdaddr_t *bc, return NULL; } -/* Find socket listening: +/* Find socket in given state: * source bdaddr (Unicast) * destination bdaddr (Broadcast only) * match func - pass NULL to ignore * match func data - pass -1 to ignore * Returns closest match. */ -static struct sock *iso_get_sock_listen(bdaddr_t *src, bdaddr_t *dst, - iso_sock_match_t match, void *data) +static struct sock *iso_get_sock(bdaddr_t *src, bdaddr_t *dst, + enum bt_sock_state state, + iso_sock_match_t match, void *data) { struct sock *sk = NULL, *sk1 = NULL; read_lock(&iso_sk_list.lock); sk_for_each(sk, &iso_sk_list.head) { - if (sk->sk_state != BT_LISTEN) + if (sk->sk_state != state) continue; /* Match Broadcast destination */ @@ -1777,32 +1780,37 @@ static void iso_conn_ready(struct iso_conn *conn) HCI_EVT_LE_BIG_SYNC_ESTABILISHED); /* Get reference to PA sync parent socket, if it exists */ - parent = iso_get_sock_listen(&hcon->src, - &hcon->dst, - iso_match_pa_sync_flag, NULL); + parent = iso_get_sock(&hcon->src, &hcon->dst, + BT_LISTEN, + iso_match_pa_sync_flag, + NULL); if (!parent && ev) - parent = iso_get_sock_listen(&hcon->src, - &hcon->dst, - iso_match_big, ev); + parent = iso_get_sock(&hcon->src, + &hcon->dst, + BT_LISTEN, + iso_match_big, ev); } else if (test_bit(HCI_CONN_PA_SYNC_FAILED, &hcon->flags)) { ev2 = hci_recv_event_data(hcon->hdev, HCI_EV_LE_PA_SYNC_ESTABLISHED); if (ev2) - parent = iso_get_sock_listen(&hcon->src, - &hcon->dst, - iso_match_sid, ev2); + parent = iso_get_sock(&hcon->src, + &hcon->dst, + BT_LISTEN, + iso_match_sid, ev2); } else if (test_bit(HCI_CONN_PA_SYNC, &hcon->flags)) { ev3 = hci_recv_event_data(hcon->hdev, HCI_EVT_LE_BIG_INFO_ADV_REPORT); if (ev3) - parent = iso_get_sock_listen(&hcon->src, - &hcon->dst, - iso_match_sync_handle, ev3); + parent = iso_get_sock(&hcon->src, + &hcon->dst, + BT_LISTEN, + iso_match_sync_handle, + ev3); } if (!parent) - parent = iso_get_sock_listen(&hcon->src, - BDADDR_ANY, NULL, NULL); + parent = iso_get_sock(&hcon->src, BDADDR_ANY, + BT_LISTEN, NULL, NULL); if (!parent) return; @@ -1923,8 +1931,8 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) */ ev1 = hci_recv_event_data(hdev, HCI_EV_LE_PA_SYNC_ESTABLISHED); if (ev1) { - sk = iso_get_sock_listen(&hdev->bdaddr, bdaddr, iso_match_sid, - ev1); + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + iso_match_sid, ev1); if (sk && !ev1->status) iso_pi(sk)->sync_handle = le16_to_cpu(ev1->handle); @@ -1934,12 +1942,12 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) ev2 = hci_recv_event_data(hdev, HCI_EVT_LE_BIG_INFO_ADV_REPORT); if (ev2) { /* Try to get PA sync listening socket, if it exists */ - sk = iso_get_sock_listen(&hdev->bdaddr, bdaddr, - iso_match_pa_sync_flag, NULL); + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + iso_match_pa_sync_flag, NULL); if (!sk) { - sk = iso_get_sock_listen(&hdev->bdaddr, bdaddr, - iso_match_sync_handle, ev2); + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + iso_match_sync_handle, ev2); /* If PA Sync is in process of terminating, * do not handle any more BIGInfo adv reports. @@ -1979,8 +1987,8 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) u8 *base; struct hci_conn *hcon; - sk = iso_get_sock_listen(&hdev->bdaddr, bdaddr, - iso_match_sync_handle_pa_report, ev3); + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, + iso_match_sync_handle_pa_report, ev3); if (!sk) goto done; @@ -2029,7 +2037,8 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) hcon->le_per_adv_data_len = 0; } } else { - sk = iso_get_sock_listen(&hdev->bdaddr, BDADDR_ANY, NULL, NULL); + sk = iso_get_sock(&hdev->bdaddr, BDADDR_ANY, + BT_LISTEN, NULL, NULL); } done: -- cgit v1.2.3 From d356c924e7a3adbea1e3e4ff4e098bcd9b99a82d Mon Sep 17 00:00:00 2001 From: Iulia Tanasescu Date: Tue, 2 Apr 2024 14:39:31 +0300 Subject: Bluetooth: ISO: Handle PA sync when no BIGInfo reports are generated In case of a Broadcast Source that has PA enabled but no active BIG, a Broadcast Sink needs to establish PA sync and parse BASE from PA reports. This commit moves the allocation of a PA sync hcon from the BIGInfo advertising report event to the PA sync established event. After the first complete PA report, the hcon is notified to the ISO layer. A child socket is allocated and enqueued in the parent's accept queue. BIGInfo reports also need to be processed, to extract the encryption field and inform userspace. After the first BIGInfo report is received, the PA sync hcon is notified again to the ISO layer. Since a socket will be found this time, the socket state will transition to BT_CONNECTED and the userspace will be woken up using sk_state_change. Signed-off-by: Iulia Tanasescu Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 3 +- net/bluetooth/hci_event.c | 58 ++++++++++++++----------- net/bluetooth/iso.c | 94 +++++++++++++++++----------------------- 3 files changed, 74 insertions(+), 81 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 9fc1eccad9bb..2283d12ee358 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1373,8 +1373,7 @@ hci_conn_hash_lookup_pa_sync_handle(struct hci_dev *hdev, __u16 sync_handle) rcu_read_lock(); list_for_each_entry_rcu(c, &h->list, list) { - if (c->type != ISO_LINK || - !test_bit(HCI_CONN_PA_SYNC, &c->flags)) + if (c->type != ISO_LINK) continue; if (c->sync_handle == sync_handle) { diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 71fedb2d4b48..d228321605ae 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1,7 +1,7 @@ /* BlueZ - Bluetooth protocol stack for Linux Copyright (c) 2000-2001, 2010, Code Aurora Forum. All rights reserved. - Copyright 2023 NXP + Copyright 2023-2024 NXP Written 2000,2001 by Maxim Krasnyansky @@ -6492,14 +6492,16 @@ static void hci_le_pa_sync_estabilished_evt(struct hci_dev *hdev, void *data, if (!(flags & HCI_PROTO_DEFER)) goto unlock; - if (ev->status) { - /* Add connection to indicate the failed PA sync event */ - pa_sync = hci_conn_add_unset(hdev, ISO_LINK, BDADDR_ANY, - HCI_ROLE_SLAVE); + /* Add connection to indicate PA sync event */ + pa_sync = hci_conn_add_unset(hdev, ISO_LINK, BDADDR_ANY, + HCI_ROLE_SLAVE); - if (!pa_sync) - goto unlock; + if (!pa_sync) + goto unlock; + + pa_sync->sync_handle = le16_to_cpu(ev->handle); + if (ev->status) { set_bit(HCI_CONN_PA_SYNC_FAILED, &pa_sync->flags); /* Notify iso layer */ @@ -6516,6 +6518,7 @@ static void hci_le_per_adv_report_evt(struct hci_dev *hdev, void *data, struct hci_ev_le_per_adv_report *ev = data; int mask = hdev->link_mode; __u8 flags = 0; + struct hci_conn *pa_sync; bt_dev_dbg(hdev, "sync_handle 0x%4.4x", le16_to_cpu(ev->sync_handle)); @@ -6523,8 +6526,28 @@ static void hci_le_per_adv_report_evt(struct hci_dev *hdev, void *data, mask |= hci_proto_connect_ind(hdev, BDADDR_ANY, ISO_LINK, &flags); if (!(mask & HCI_LM_ACCEPT)) - hci_le_pa_term_sync(hdev, ev->sync_handle); + goto unlock; + if (!(flags & HCI_PROTO_DEFER)) + goto unlock; + + pa_sync = hci_conn_hash_lookup_pa_sync_handle + (hdev, + le16_to_cpu(ev->sync_handle)); + + if (!pa_sync) + goto unlock; + + if (ev->data_status == LE_PA_DATA_COMPLETE && + !test_and_set_bit(HCI_CONN_PA_SYNC, &pa_sync->flags)) { + /* Notify iso layer */ + hci_connect_cfm(pa_sync, 0); + + /* Notify MGMT layer */ + mgmt_device_connected(hdev, pa_sync, NULL, 0); + } + +unlock: hci_dev_unlock(hdev); } @@ -7059,10 +7082,8 @@ static void hci_le_big_info_adv_report_evt(struct hci_dev *hdev, void *data, hci_dev_lock(hdev); mask |= hci_proto_connect_ind(hdev, BDADDR_ANY, ISO_LINK, &flags); - if (!(mask & HCI_LM_ACCEPT)) { - hci_le_pa_term_sync(hdev, ev->sync_handle); + if (!(mask & HCI_LM_ACCEPT)) goto unlock; - } if (!(flags & HCI_PROTO_DEFER)) goto unlock; @@ -7071,24 +7092,11 @@ static void hci_le_big_info_adv_report_evt(struct hci_dev *hdev, void *data, (hdev, le16_to_cpu(ev->sync_handle)); - if (pa_sync) - goto unlock; - - /* Add connection to indicate the PA sync event */ - pa_sync = hci_conn_add_unset(hdev, ISO_LINK, BDADDR_ANY, - HCI_ROLE_SLAVE); - if (!pa_sync) goto unlock; - pa_sync->sync_handle = le16_to_cpu(ev->sync_handle); - set_bit(HCI_CONN_PA_SYNC, &pa_sync->flags); - /* Notify iso layer */ - hci_connect_cfm(pa_sync, 0x00); - - /* Notify MGMT layer */ - mgmt_device_connected(hdev, pa_sync, NULL, 0); + hci_connect_cfm(pa_sync, 0); unlock: hci_dev_unlock(hdev); diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 89dee805c60d..d0ef7290e5b7 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -54,7 +54,6 @@ static void iso_sock_kill(struct sock *sk); enum { BT_SK_BIG_SYNC, BT_SK_PA_SYNC, - BT_SK_PA_SYNC_TERM, }; struct iso_pinfo { @@ -81,6 +80,7 @@ static bool check_ucast_qos(struct bt_iso_qos *qos); static bool check_bcast_qos(struct bt_iso_qos *qos); static bool iso_match_sid(struct sock *sk, void *data); static bool iso_match_sync_handle(struct sock *sk, void *data); +static bool iso_match_sync_handle_pa_report(struct sock *sk, void *data); static void iso_sock_disconn(struct sock *sk); typedef bool (*iso_sock_match_t)(struct sock *sk, void *data); @@ -197,21 +197,10 @@ static void iso_chan_del(struct sock *sk, int err) sock_set_flag(sk, SOCK_ZAPPED); } -static bool iso_match_conn_sync_handle(struct sock *sk, void *data) -{ - struct hci_conn *hcon = data; - - if (test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) - return false; - - return hcon->sync_handle == iso_pi(sk)->sync_handle; -} - static void iso_conn_del(struct hci_conn *hcon, int err) { struct iso_conn *conn = hcon->iso_data; struct sock *sk; - struct sock *parent; if (!conn) return; @@ -227,26 +216,6 @@ static void iso_conn_del(struct hci_conn *hcon, int err) if (sk) { lock_sock(sk); - - /* While a PA sync hcon is in the process of closing, - * mark parent socket with a flag, so that any residual - * BIGInfo adv reports that arrive before PA sync is - * terminated are not processed anymore. - */ - if (test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) { - parent = iso_get_sock(&hcon->src, - &hcon->dst, - BT_LISTEN, - iso_match_conn_sync_handle, - hcon); - - if (parent) { - set_bit(BT_SK_PA_SYNC_TERM, - &iso_pi(parent)->flags); - sock_put(parent); - } - } - iso_sock_clear_timer(sk); iso_chan_del(sk, err); release_sock(sk); @@ -860,6 +829,7 @@ static struct sock *iso_sock_alloc(struct net *net, struct socket *sock, iso_pi(sk)->src_type = BDADDR_LE_PUBLIC; iso_pi(sk)->qos = default_qos; + iso_pi(sk)->sync_handle = -1; bt_sock_link(&iso_sk_list, sk); return sk; @@ -907,7 +877,6 @@ static int iso_sock_bind_bc(struct socket *sock, struct sockaddr *addr, return -EINVAL; iso_pi(sk)->dst_type = sa->iso_bc->bc_bdaddr_type; - iso_pi(sk)->sync_handle = -1; if (sa->iso_bc->bc_sid > 0x0f) return -EINVAL; @@ -984,7 +953,8 @@ static int iso_sock_bind(struct socket *sock, struct sockaddr *addr, /* Allow the user to bind a PA sync socket to a number * of BISes to sync to. */ - if (sk->sk_state == BT_CONNECT2 && + if ((sk->sk_state == BT_CONNECT2 || + sk->sk_state == BT_CONNECTED) && test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) { err = iso_sock_bind_pa_sk(sk, sa, addr_len); goto done; @@ -1396,6 +1366,16 @@ static int iso_sock_recvmsg(struct socket *sock, struct msghdr *msg, } release_sock(sk); return 0; + case BT_CONNECTED: + if (test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags)) { + iso_conn_big_sync(sk); + sk->sk_state = BT_LISTEN; + release_sock(sk); + return 0; + } + + release_sock(sk); + break; case BT_CONNECT: release_sock(sk); return iso_connect_cis(sk); @@ -1541,7 +1521,9 @@ static int iso_sock_setsockopt(struct socket *sock, int level, int optname, case BT_ISO_QOS: if (sk->sk_state != BT_OPEN && sk->sk_state != BT_BOUND && - sk->sk_state != BT_CONNECT2) { + sk->sk_state != BT_CONNECT2 && + (!test_bit(BT_SK_PA_SYNC, &iso_pi(sk)->flags) || + sk->sk_state != BT_CONNECTED)) { err = -EINVAL; break; } @@ -1762,7 +1744,7 @@ static void iso_conn_ready(struct iso_conn *conn) struct sock *sk = conn->sk; struct hci_ev_le_big_sync_estabilished *ev = NULL; struct hci_ev_le_pa_sync_established *ev2 = NULL; - struct hci_evt_le_big_info_adv_report *ev3 = NULL; + struct hci_ev_le_per_adv_report *ev3 = NULL; struct hci_conn *hcon; BT_DBG("conn %p", conn); @@ -1799,12 +1781,12 @@ static void iso_conn_ready(struct iso_conn *conn) iso_match_sid, ev2); } else if (test_bit(HCI_CONN_PA_SYNC, &hcon->flags)) { ev3 = hci_recv_event_data(hcon->hdev, - HCI_EVT_LE_BIG_INFO_ADV_REPORT); + HCI_EV_LE_PER_ADV_REPORT); if (ev3) parent = iso_get_sock(&hcon->src, &hcon->dst, BT_LISTEN, - iso_match_sync_handle, + iso_match_sync_handle_pa_report, ev3); } @@ -1847,7 +1829,6 @@ static void iso_conn_ready(struct iso_conn *conn) if (ev3) { iso_pi(sk)->qos = iso_pi(parent)->qos; - iso_pi(sk)->qos.bcast.encryption = ev3->encryption; hcon->iso_qos = iso_pi(sk)->qos; iso_pi(sk)->bc_num_bis = iso_pi(parent)->bc_num_bis; memcpy(iso_pi(sk)->bc_bis, iso_pi(parent)->bc_bis, ISO_MAX_NUM_BIS); @@ -1941,26 +1922,29 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) ev2 = hci_recv_event_data(hdev, HCI_EVT_LE_BIG_INFO_ADV_REPORT); if (ev2) { - /* Try to get PA sync listening socket, if it exists */ - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, - iso_match_pa_sync_flag, NULL); - - if (!sk) { - sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_LISTEN, - iso_match_sync_handle, ev2); - - /* If PA Sync is in process of terminating, - * do not handle any more BIGInfo adv reports. - */ - - if (sk && test_bit(BT_SK_PA_SYNC_TERM, - &iso_pi(sk)->flags)) - return 0; + /* Check if BIGInfo report has already been handled */ + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_CONNECTED, + iso_match_sync_handle, ev2); + if (sk) { + sock_put(sk); + sk = NULL; + goto done; } + /* Try to get PA sync socket, if it exists */ + sk = iso_get_sock(&hdev->bdaddr, bdaddr, BT_CONNECT2, + iso_match_sync_handle, ev2); + if (!sk) + sk = iso_get_sock(&hdev->bdaddr, bdaddr, + BT_LISTEN, + iso_match_sync_handle, + ev2); + if (sk) { int err; + iso_pi(sk)->qos.bcast.encryption = ev2->encryption; + if (ev2->num_bis < iso_pi(sk)->bc_num_bis) iso_pi(sk)->bc_num_bis = ev2->num_bis; @@ -1979,6 +1963,8 @@ int iso_connect_ind(struct hci_dev *hdev, bdaddr_t *bdaddr, __u8 *flags) } } } + + goto done; } ev3 = hci_recv_event_data(hdev, HCI_EV_LE_PER_ADV_REPORT); -- cgit v1.2.3 From 1c08108f3014881ad5f4c35a2abaf9c65475035d Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 27 Mar 2024 10:23:51 -0600 Subject: Bluetooth: L2CAP: Avoid -Wflex-array-member-not-at-end warnings -Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. There are currently a couple of objects (`req` and `rsp`), in a couple of structures, that contain flexible structures (`struct l2cap_ecred_conn_req` and `struct l2cap_ecred_conn_rsp`), for example: struct l2cap_ecred_rsp_data { struct { struct l2cap_ecred_conn_rsp rsp; __le16 scid[L2CAP_ECRED_MAX_CID]; } __packed pdu; int count; }; in the struct above, `struct l2cap_ecred_conn_rsp` is a flexible structure: struct l2cap_ecred_conn_rsp { __le16 mtu; __le16 mps; __le16 credits; __le16 result; __le16 dcid[]; }; So, in order to avoid ending up with a flexible-array member in the middle of another structure, we use the `struct_group_tagged()` (and `__struct_group()` when the flexible structure is `__packed`) helper to separate the flexible array from the rest of the members in the flexible structure: struct l2cap_ecred_conn_rsp { struct_group_tagged(l2cap_ecred_conn_rsp_hdr, hdr, ... the rest of members ); __le16 dcid[]; }; With the change described above, we now declare objects of the type of the tagged struct, in this example `struct l2cap_ecred_conn_rsp_hdr`, without embedding flexible arrays in the middle of other structures: struct l2cap_ecred_rsp_data { struct { struct l2cap_ecred_conn_rsp_hdr rsp; __le16 scid[L2CAP_ECRED_MAX_CID]; } __packed pdu; int count; }; Also, when the flexible-array member needs to be accessed, we use `container_of()` to retrieve a pointer to the flexible structure. We also use the `DEFINE_RAW_FLEX()` helper for a couple of on-stack definitions of a flexible structure where the size of the flexible-array member is known at compile-time. So, with these changes, fix the following warnings: net/bluetooth/l2cap_core.c:1260:45: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] net/bluetooth/l2cap_core.c:3740:45: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] net/bluetooth/l2cap_core.c:4999:45: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] net/bluetooth/l2cap_core.c:7116:47: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/l2cap.h | 22 +++++++++++++-------- net/bluetooth/l2cap_core.c | 46 ++++++++++++++++++++----------------------- 2 files changed, 35 insertions(+), 33 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index a4278aa618ab..7d4a3e766e7b 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -463,18 +463,24 @@ struct l2cap_le_credits { #define L2CAP_ECRED_MAX_CID 5 struct l2cap_ecred_conn_req { - __le16 psm; - __le16 mtu; - __le16 mps; - __le16 credits; + /* New members must be added within the struct_group() macro below. */ + __struct_group(l2cap_ecred_conn_req_hdr, hdr, __packed, + __le16 psm; + __le16 mtu; + __le16 mps; + __le16 credits; + ); __le16 scid[]; } __packed; struct l2cap_ecred_conn_rsp { - __le16 mtu; - __le16 mps; - __le16 credits; - __le16 result; + /* New members must be added within the struct_group() macro below. */ + struct_group_tagged(l2cap_ecred_conn_rsp_hdr, hdr, + __le16 mtu; + __le16 mps; + __le16 credits; + __le16 result; + ); __le16 dcid[]; }; diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 9223b1a698e3..fa714f5bd0b3 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -1260,7 +1260,7 @@ static void l2cap_le_connect(struct l2cap_chan *chan) struct l2cap_ecred_conn_data { struct { - struct l2cap_ecred_conn_req req; + struct l2cap_ecred_conn_req_hdr req; __le16 scid[5]; } __packed pdu; struct l2cap_chan *chan; @@ -3740,7 +3740,7 @@ static void l2cap_ecred_list_defer(struct l2cap_chan *chan, void *data) struct l2cap_ecred_rsp_data { struct { - struct l2cap_ecred_conn_rsp rsp; + struct l2cap_ecred_conn_rsp_hdr rsp; __le16 scid[L2CAP_ECRED_MAX_CID]; } __packed pdu; int count; @@ -3749,6 +3749,8 @@ struct l2cap_ecred_rsp_data { static void l2cap_ecred_rsp_defer(struct l2cap_chan *chan, void *data) { struct l2cap_ecred_rsp_data *rsp = data; + struct l2cap_ecred_conn_rsp *rsp_flex = + container_of(&rsp->pdu.rsp, struct l2cap_ecred_conn_rsp, hdr); if (test_bit(FLAG_ECRED_CONN_REQ_SENT, &chan->flags)) return; @@ -3758,7 +3760,7 @@ static void l2cap_ecred_rsp_defer(struct l2cap_chan *chan, void *data) /* Include all channels pending with the same ident */ if (!rsp->pdu.rsp.result) - rsp->pdu.rsp.dcid[rsp->count++] = cpu_to_le16(chan->scid); + rsp_flex->dcid[rsp->count++] = cpu_to_le16(chan->scid); else l2cap_chan_del(chan, ECONNRESET); } @@ -4996,10 +4998,7 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, u8 *data) { struct l2cap_ecred_conn_req *req = (void *) data; - struct { - struct l2cap_ecred_conn_rsp rsp; - __le16 dcid[L2CAP_ECRED_MAX_CID]; - } __packed pdu; + DEFINE_RAW_FLEX(struct l2cap_ecred_conn_rsp, pdu, dcid, L2CAP_ECRED_MAX_CID); struct l2cap_chan *chan, *pchan; u16 mtu, mps; __le16 psm; @@ -5018,7 +5017,7 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, cmd_len -= sizeof(*req); num_scid = cmd_len / sizeof(u16); - if (num_scid > ARRAY_SIZE(pdu.dcid)) { + if (num_scid > L2CAP_ECRED_MAX_CID) { result = L2CAP_CR_LE_INVALID_PARAMS; goto response; } @@ -5047,7 +5046,7 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, BT_DBG("psm 0x%2.2x mtu %u mps %u", __le16_to_cpu(psm), mtu, mps); - memset(&pdu, 0, sizeof(pdu)); + memset(pdu, 0, sizeof(*pdu)); /* Check if we have socket listening on psm */ pchan = l2cap_global_chan_by_psm(BT_LISTEN, psm, &conn->hcon->src, @@ -5073,8 +5072,8 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, BT_DBG("scid[%d] 0x%4.4x", i, scid); - pdu.dcid[i] = 0x0000; - len += sizeof(*pdu.dcid); + pdu->dcid[i] = 0x0000; + len += sizeof(*pdu->dcid); /* Check for valid dynamic CID range */ if (scid < L2CAP_CID_DYN_START || scid > L2CAP_CID_LE_DYN_END) { @@ -5108,13 +5107,13 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, l2cap_ecred_init(chan, __le16_to_cpu(req->credits)); /* Init response */ - if (!pdu.rsp.credits) { - pdu.rsp.mtu = cpu_to_le16(chan->imtu); - pdu.rsp.mps = cpu_to_le16(chan->mps); - pdu.rsp.credits = cpu_to_le16(chan->rx_credits); + if (!pdu->credits) { + pdu->mtu = cpu_to_le16(chan->imtu); + pdu->mps = cpu_to_le16(chan->mps); + pdu->credits = cpu_to_le16(chan->rx_credits); } - pdu.dcid[i] = cpu_to_le16(chan->scid); + pdu->dcid[i] = cpu_to_le16(chan->scid); __set_chan_timer(chan, chan->ops->get_sndtimeo(chan)); @@ -5136,13 +5135,13 @@ unlock: l2cap_chan_put(pchan); response: - pdu.rsp.result = cpu_to_le16(result); + pdu->result = cpu_to_le16(result); if (defer) return 0; l2cap_send_cmd(conn, cmd->ident, L2CAP_ECRED_CONN_RSP, - sizeof(pdu.rsp) + len, &pdu); + sizeof(*pdu) + len, pdu); return 0; } @@ -7113,14 +7112,11 @@ EXPORT_SYMBOL_GPL(l2cap_chan_connect); static void l2cap_ecred_reconfigure(struct l2cap_chan *chan) { struct l2cap_conn *conn = chan->conn; - struct { - struct l2cap_ecred_reconf_req req; - __le16 scid; - } pdu; + DEFINE_RAW_FLEX(struct l2cap_ecred_reconf_req, pdu, scid, 1); - pdu.req.mtu = cpu_to_le16(chan->imtu); - pdu.req.mps = cpu_to_le16(chan->mps); - pdu.scid = cpu_to_le16(chan->scid); + pdu->mtu = cpu_to_le16(chan->imtu); + pdu->mps = cpu_to_le16(chan->mps); + pdu->scid[0] = cpu_to_le16(chan->scid); chan->ident = l2cap_get_ident(conn); -- cgit v1.2.3 From d68d8a7a2c62312c81c066d38bb0662e67b58206 Mon Sep 17 00:00:00 2001 From: Zijun Hu Date: Mon, 22 Apr 2024 22:46:34 +0800 Subject: Bluetooth: hci_conn: Remove a redundant check for HFP offload Remove a redundant check !hdev->get_codec_config_data. Signed-off-by: Zijun Hu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 05346250f719..c1d6b4d713b4 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -241,13 +241,13 @@ static int configure_datapath_sync(struct hci_dev *hdev, struct bt_codec *codec) __u8 vnd_len, *vnd_data = NULL; struct hci_op_configure_data_path *cmd = NULL; + /* Do not take below 2 checks as error since the 1st means user do not + * want to use HFP offload mode and the 2nd means the vendor controller + * do not need to send below HCI command for offload mode. + */ if (!codec->data_path || !hdev->get_codec_config_data) return 0; - /* Do not take me as error */ - if (!hdev->get_codec_config_data) - return 0; - err = hdev->get_codec_config_data(hdev, ESCO_LINK, codec, &vnd_len, &vnd_data); if (err < 0) -- cgit v1.2.3 From 94c603c28e59091f698efc5caaddc93c365455ab Mon Sep 17 00:00:00 2001 From: Zijun Hu Date: Mon, 22 Apr 2024 22:51:50 +0800 Subject: Bluetooth: Remove 3 repeated macro definitions Macros HCI_REQ_DONE, HCI_REQ_PEND and HCI_REQ_CANCELED are repeatedly defined twice with hci_request.h, so remove a copy of definition. Signed-off-by: Zijun Hu Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_request.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'net') diff --git a/net/bluetooth/hci_request.h b/net/bluetooth/hci_request.h index 0be75cf0efed..c91f2838f542 100644 --- a/net/bluetooth/hci_request.h +++ b/net/bluetooth/hci_request.h @@ -29,10 +29,6 @@ #define hci_req_sync_lock(hdev) mutex_lock(&hdev->req_lock) #define hci_req_sync_unlock(hdev) mutex_unlock(&hdev->req_lock) -#define HCI_REQ_DONE 0 -#define HCI_REQ_PEND 1 -#define HCI_REQ_CANCELED 2 - struct hci_request { struct hci_dev *hdev; struct sk_buff_head cmd_q; -- cgit v1.2.3 From c4585edf708edb5277a3cc4b8581ccb833f3307d Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Fri, 26 Apr 2024 16:52:46 -0600 Subject: Bluetooth: hci_conn, hci_sync: Use __counted_by() to avoid -Wfamnae warnings Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). Also, -Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. So, use the `DEFINE_FLEX()` helper for multiple on-stack definitions of a flexible structure where the size of the flexible-array member is known at compile-time, and refactor the rest of the code, accordingly. Notice that, due to the use of `__counted_by()` in `struct hci_cp_le_create_cis`, the for loop in function `hci_cs_le_create_cis()` had to be modified. Once the index `i`, through which `cp->cis[i]` is accessed, falls in the interval [0, cp->num_cis), `cp->num_cis` cannot be decremented all the way down to zero while accessing `cp->cis[]`: net/bluetooth/hci_event.c:4310: 4310 for (i = 0; cp->num_cis; cp->num_cis--, i++) { ... 4314 handle = __le16_to_cpu(cp->cis[i].cis_handle); otherwise, only half (one iteration before `cp->num_cis == i`) or half plus one (one iteration before `cp->num_cis < i`) of the items in the array will be accessed before running into an out-of-bounds issue. So, in order to avoid this, set `cp->num_cis` to zero just after the for loop. Also, make use of `aux_num_cis` variable to update `cmd->num_cis` after a `list_for_each_entry_rcu()` loop. With these changes, fix the following warnings: net/bluetooth/hci_sync.c:1239:56: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] net/bluetooth/hci_sync.c:1415:51: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] net/bluetooth/hci_sync.c:1731:51: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] net/bluetooth/hci_sync.c:6497:45: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 8 ++--- net/bluetooth/hci_event.c | 3 +- net/bluetooth/hci_sync.c | 84 ++++++++++++++++++--------------------------- 3 files changed, 40 insertions(+), 55 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 5c12761cbc0e..0b0ac7f06570 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -2026,7 +2026,7 @@ struct hci_cp_le_set_ext_adv_data { __u8 operation; __u8 frag_pref; __u8 length; - __u8 data[]; + __u8 data[] __counted_by(length); } __packed; #define HCI_OP_LE_SET_EXT_SCAN_RSP_DATA 0x2038 @@ -2035,7 +2035,7 @@ struct hci_cp_le_set_ext_scan_rsp_data { __u8 operation; __u8 frag_pref; __u8 length; - __u8 data[]; + __u8 data[] __counted_by(length); } __packed; #define HCI_OP_LE_SET_EXT_ADV_ENABLE 0x2039 @@ -2061,7 +2061,7 @@ struct hci_cp_le_set_per_adv_data { __u8 handle; __u8 operation; __u8 length; - __u8 data[]; + __u8 data[] __counted_by(length); } __packed; #define HCI_OP_LE_SET_PER_ADV_ENABLE 0x2040 @@ -2162,7 +2162,7 @@ struct hci_cis { struct hci_cp_le_create_cis { __u8 num_cis; - struct hci_cis cis[]; + struct hci_cis cis[] __counted_by(num_cis); } __packed; #define HCI_OP_LE_REMOVE_CIG 0x2065 diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index d228321605ae..122a6a9092e0 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -4307,7 +4307,7 @@ static void hci_cs_le_create_cis(struct hci_dev *hdev, u8 status) hci_dev_lock(hdev); /* Remove connection if command failed */ - for (i = 0; cp->num_cis; cp->num_cis--, i++) { + for (i = 0; i < cp->num_cis; i++) { struct hci_conn *conn; u16 handle; @@ -4323,6 +4323,7 @@ static void hci_cs_le_create_cis(struct hci_dev *hdev, u8 status) hci_conn_del(conn); } } + cp->num_cis = 0; if (pending) hci_le_create_cis_pending(hdev); diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 4c707eb64e6f..406e36cacd79 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -1235,31 +1235,27 @@ int hci_setup_ext_adv_instance_sync(struct hci_dev *hdev, u8 instance) static int hci_set_ext_scan_rsp_data_sync(struct hci_dev *hdev, u8 instance) { - struct { - struct hci_cp_le_set_ext_scan_rsp_data cp; - u8 data[HCI_MAX_EXT_AD_LENGTH]; - } pdu; + DEFINE_FLEX(struct hci_cp_le_set_ext_scan_rsp_data, pdu, data, length, + HCI_MAX_EXT_AD_LENGTH); u8 len; struct adv_info *adv = NULL; int err; - memset(&pdu, 0, sizeof(pdu)); - if (instance) { adv = hci_find_adv_instance(hdev, instance); if (!adv || !adv->scan_rsp_changed) return 0; } - len = eir_create_scan_rsp(hdev, instance, pdu.data); + len = eir_create_scan_rsp(hdev, instance, pdu->data); - pdu.cp.handle = instance; - pdu.cp.length = len; - pdu.cp.operation = LE_SET_ADV_DATA_OP_COMPLETE; - pdu.cp.frag_pref = LE_SET_ADV_DATA_NO_FRAG; + pdu->handle = instance; + pdu->length = len; + pdu->operation = LE_SET_ADV_DATA_OP_COMPLETE; + pdu->frag_pref = LE_SET_ADV_DATA_NO_FRAG; err = __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_EXT_SCAN_RSP_DATA, - sizeof(pdu.cp) + len, &pdu.cp, + struct_size(pdu, data, len), pdu, HCI_CMD_TIMEOUT); if (err) return err; @@ -1267,7 +1263,7 @@ static int hci_set_ext_scan_rsp_data_sync(struct hci_dev *hdev, u8 instance) if (adv) { adv->scan_rsp_changed = false; } else { - memcpy(hdev->scan_rsp_data, pdu.data, len); + memcpy(hdev->scan_rsp_data, pdu->data, len); hdev->scan_rsp_data_len = len; } @@ -1411,14 +1407,10 @@ static int hci_set_per_adv_params_sync(struct hci_dev *hdev, u8 instance, static int hci_set_per_adv_data_sync(struct hci_dev *hdev, u8 instance) { - struct { - struct hci_cp_le_set_per_adv_data cp; - u8 data[HCI_MAX_PER_AD_LENGTH]; - } pdu; + DEFINE_FLEX(struct hci_cp_le_set_per_adv_data, pdu, data, length, + HCI_MAX_PER_AD_LENGTH); u8 len; - memset(&pdu, 0, sizeof(pdu)); - if (instance) { struct adv_info *adv = hci_find_adv_instance(hdev, instance); @@ -1426,14 +1418,14 @@ static int hci_set_per_adv_data_sync(struct hci_dev *hdev, u8 instance) return 0; } - len = eir_create_per_adv_data(hdev, instance, pdu.data); + len = eir_create_per_adv_data(hdev, instance, pdu->data); - pdu.cp.length = len; - pdu.cp.handle = instance; - pdu.cp.operation = LE_SET_ADV_DATA_OP_COMPLETE; + pdu->length = len; + pdu->handle = instance; + pdu->operation = LE_SET_ADV_DATA_OP_COMPLETE; return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_PER_ADV_DATA, - sizeof(pdu.cp) + len, &pdu, + struct_size(pdu, data, len), pdu, HCI_CMD_TIMEOUT); } @@ -1727,31 +1719,27 @@ int hci_le_terminate_big_sync(struct hci_dev *hdev, u8 handle, u8 reason) static int hci_set_ext_adv_data_sync(struct hci_dev *hdev, u8 instance) { - struct { - struct hci_cp_le_set_ext_adv_data cp; - u8 data[HCI_MAX_EXT_AD_LENGTH]; - } pdu; + DEFINE_FLEX(struct hci_cp_le_set_ext_adv_data, pdu, data, length, + HCI_MAX_EXT_AD_LENGTH); u8 len; struct adv_info *adv = NULL; int err; - memset(&pdu, 0, sizeof(pdu)); - if (instance) { adv = hci_find_adv_instance(hdev, instance); if (!adv || !adv->adv_data_changed) return 0; } - len = eir_create_adv_data(hdev, instance, pdu.data); + len = eir_create_adv_data(hdev, instance, pdu->data); - pdu.cp.length = len; - pdu.cp.handle = instance; - pdu.cp.operation = LE_SET_ADV_DATA_OP_COMPLETE; - pdu.cp.frag_pref = LE_SET_ADV_DATA_NO_FRAG; + pdu->length = len; + pdu->handle = instance; + pdu->operation = LE_SET_ADV_DATA_OP_COMPLETE; + pdu->frag_pref = LE_SET_ADV_DATA_NO_FRAG; err = __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_EXT_ADV_DATA, - sizeof(pdu.cp) + len, &pdu.cp, + struct_size(pdu, data, len), pdu, HCI_CMD_TIMEOUT); if (err) return err; @@ -1760,7 +1748,7 @@ static int hci_set_ext_adv_data_sync(struct hci_dev *hdev, u8 instance) if (adv) { adv->adv_data_changed = false; } else { - memcpy(hdev->adv_data, pdu.data, len); + memcpy(hdev->adv_data, pdu->data, len); hdev->adv_data_len = len; } @@ -6493,10 +6481,8 @@ done: int hci_le_create_cis_sync(struct hci_dev *hdev) { - struct { - struct hci_cp_le_create_cis cp; - struct hci_cis cis[0x1f]; - } cmd; + DEFINE_FLEX(struct hci_cp_le_create_cis, cmd, cis, num_cis, 0x1f); + size_t aux_num_cis = 0; struct hci_conn *conn; u8 cig = BT_ISO_QOS_CIG_UNSET; @@ -6523,8 +6509,6 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) * remains pending. */ - memset(&cmd, 0, sizeof(cmd)); - hci_dev_lock(hdev); rcu_read_lock(); @@ -6561,7 +6545,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) goto done; list_for_each_entry_rcu(conn, &hdev->conn_hash.list, list) { - struct hci_cis *cis = &cmd.cis[cmd.cp.num_cis]; + struct hci_cis *cis = &cmd->cis[aux_num_cis]; if (hci_conn_check_create_cis(conn) || conn->iso_qos.ucast.cig != cig) @@ -6570,25 +6554,25 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) set_bit(HCI_CONN_CREATE_CIS, &conn->flags); cis->acl_handle = cpu_to_le16(conn->parent->handle); cis->cis_handle = cpu_to_le16(conn->handle); - cmd.cp.num_cis++; + aux_num_cis++; - if (cmd.cp.num_cis >= ARRAY_SIZE(cmd.cis)) + if (aux_num_cis >= 0x1f) break; } + cmd->num_cis = aux_num_cis; done: rcu_read_unlock(); hci_dev_unlock(hdev); - if (!cmd.cp.num_cis) + if (!aux_num_cis) return 0; /* Wait for HCI_LE_CIS_Established */ return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CREATE_CIS, - sizeof(cmd.cp) + sizeof(cmd.cis[0]) * - cmd.cp.num_cis, &cmd, - HCI_EVT_LE_CIS_ESTABLISHED, + struct_size(cmd, cis, cmd->num_cis), + cmd, HCI_EVT_LE_CIS_ESTABLISHED, conn->conn_timeout, NULL); } -- cgit v1.2.3 From c90748b898cf9a0654aeadac1ba2254e3ab8e9fa Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Fri, 26 Apr 2024 10:45:17 -0600 Subject: Bluetooth: hci_conn: Use __counted_by() to avoid -Wfamnae warning Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). Also, -Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. So, use the `DEFINE_FLEX()` helper for an on-stack definition of a flexible structure where the size of the flexible-array member is known at compile-time, and refactor the rest of the code, accordingly. With these changes, fix the following warning: net/bluetooth/hci_conn.c:2116:50: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Reviewed-by: Kees Cook Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 2 +- net/bluetooth/hci_conn.c | 26 +++++++++++--------------- 2 files changed, 12 insertions(+), 16 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 0b0ac7f06570..c4c6b8810701 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -2216,7 +2216,7 @@ struct hci_cp_le_big_create_sync { __u8 mse; __le16 timeout; __u8 num_bis; - __u8 bis[]; + __u8 bis[] __counted_by(num_bis); } __packed; #define HCI_OP_LE_BIG_TERM_SYNC 0x206c diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index c1d6b4d713b4..8375f7bd58ee 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -2109,13 +2109,10 @@ int hci_le_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon, struct bt_iso_qos *qos, __u16 sync_handle, __u8 num_bis, __u8 bis[]) { - struct _packed { - struct hci_cp_le_big_create_sync cp; - __u8 bis[0x11]; - } pdu; + DEFINE_FLEX(struct hci_cp_le_big_create_sync, pdu, bis, num_bis, 0x11); int err; - if (num_bis < 0x01 || num_bis > sizeof(pdu.bis)) + if (num_bis < 0x01 || num_bis > pdu->num_bis) return -EINVAL; err = qos_set_big(hdev, qos); @@ -2125,18 +2122,17 @@ int hci_le_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon, if (hcon) hcon->iso_qos.bcast.big = qos->bcast.big; - memset(&pdu, 0, sizeof(pdu)); - pdu.cp.handle = qos->bcast.big; - pdu.cp.sync_handle = cpu_to_le16(sync_handle); - pdu.cp.encryption = qos->bcast.encryption; - memcpy(pdu.cp.bcode, qos->bcast.bcode, sizeof(pdu.cp.bcode)); - pdu.cp.mse = qos->bcast.mse; - pdu.cp.timeout = cpu_to_le16(qos->bcast.timeout); - pdu.cp.num_bis = num_bis; - memcpy(pdu.bis, bis, num_bis); + pdu->handle = qos->bcast.big; + pdu->sync_handle = cpu_to_le16(sync_handle); + pdu->encryption = qos->bcast.encryption; + memcpy(pdu->bcode, qos->bcast.bcode, sizeof(pdu->bcode)); + pdu->mse = qos->bcast.mse; + pdu->timeout = cpu_to_le16(qos->bcast.timeout); + pdu->num_bis = num_bis; + memcpy(pdu->bis, bis, num_bis); return hci_send_cmd(hdev, HCI_OP_LE_BIG_CREATE_SYNC, - sizeof(pdu.cp) + num_bis, &pdu); + sizeof(*pdu) + num_bis, pdu); } static void create_big_complete(struct hci_dev *hdev, void *data, int err) -- cgit v1.2.3 From d6bb8782b4b6ec14fb6e336f7f8c6a1ccce613b1 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 1 May 2024 12:09:30 -0600 Subject: Bluetooth: hci_conn: Use struct_size() in hci_le_big_create_sync() Use struct_size() instead of the open-coded version. Similarly to this other patch[1]. Link: https://lore.kernel.org/linux-hardening/ZiwwPmCvU25YzWek@neat/ [1] Signed-off-by: Gustavo A. R. Silva Reviewed-by: Kees Cook Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 8375f7bd58ee..fc1aeb513b08 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -2132,7 +2132,7 @@ int hci_le_big_create_sync(struct hci_dev *hdev, struct hci_conn *hcon, memcpy(pdu->bis, bis, num_bis); return hci_send_cmd(hdev, HCI_OP_LE_BIG_CREATE_SYNC, - sizeof(*pdu) + num_bis, pdu); + struct_size(pdu, bis, num_bis), pdu); } static void create_big_complete(struct hci_dev *hdev, void *data, int err) -- cgit v1.2.3 From 73b2652cbbb9993fe6a4340e6f5a2870d7ce6fb4 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 1 May 2024 11:50:02 -0600 Subject: Bluetooth: hci_sync: Use cmd->num_cis instead of magic number At the moment of the check, `cmd->num_cis` holds the value of 0x1f, which is the max number of elements in the `cmd->cis[]` array at declaration, which is 0x1f. So, avoid using 0x1f directly, and instead use `cmd->num_cis`. Similarly to this other patch[1]. Link: https://lore.kernel.org/linux-hardening/ZivaHUQyDDK9fXEk@neat/ [1] Signed-off-by: Gustavo A. R. Silva Reviewed-by: Kees Cook Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 406e36cacd79..813c11d61bca 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -6556,7 +6556,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) cis->cis_handle = cpu_to_le16(conn->handle); aux_num_cis++; - if (aux_num_cis >= 0x1f) + if (aux_num_cis >= cmd->num_cis) break; } cmd->num_cis = aux_num_cis; -- cgit v1.2.3 From ce60b9231b66710b6ee24042ded26efee120ecfc Mon Sep 17 00:00:00 2001 From: Sebastian Urban Date: Wed, 1 May 2024 12:08:58 +0200 Subject: Bluetooth: compute LE flow credits based on recvbuf space Previously LE flow credits were returned to the sender even if the socket's receive buffer was full. This meant that no back-pressure was applied to the sender, thus it continued to send data, resulting in data loss without any error being reported. Furthermore, the amount of credits was essentially fixed to a small amount, leading to reduced performance. This is fixed by computing the number of returned LE flow credits based on the estimated available space in the receive buffer of an L2CAP socket. Consequently, if the receive buffer is full, no credits are returned until the buffer is read and thus cleared by user-space. Since the computation of available receive buffer space can only be performed approximately (due to sk_buff overhead) and the receive buffer size may be changed by user-space after flow credits have been sent, superfluous received data is temporary stored within l2cap_pinfo. This is necessary because Bluetooth LE provides no retransmission mechanism once the data has been acked by the physical layer. If receive buffer space estimation is not possible at the moment, we fall back to providing credits for one full packet as before. This is currently the case during connection setup, when MPS is not yet available. Fixes: b1c325c23d75 ("Bluetooth: Implement returning of LE L2CAP credits") Signed-off-by: Sebastian Urban Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/l2cap.h | 11 +++++- net/bluetooth/l2cap_core.c | 56 ++++++++++++++++++++++---- net/bluetooth/l2cap_sock.c | 91 ++++++++++++++++++++++++++++++++++--------- 3 files changed, 132 insertions(+), 26 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 7d4a3e766e7b..5cfdc813491a 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -554,6 +554,9 @@ struct l2cap_chan { __u16 tx_credits; __u16 rx_credits; + /* estimated available receive buffer space or -1 if unknown */ + ssize_t rx_avail; + __u8 tx_state; __u8 rx_state; @@ -688,10 +691,15 @@ struct l2cap_user { /* ----- L2CAP socket info ----- */ #define l2cap_pi(sk) ((struct l2cap_pinfo *) sk) +struct l2cap_rx_busy { + struct list_head list; + struct sk_buff *skb; +}; + struct l2cap_pinfo { struct bt_sock bt; struct l2cap_chan *chan; - struct sk_buff *rx_busy_skb; + struct list_head rx_busy; }; enum { @@ -949,6 +957,7 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid, int l2cap_chan_reconfigure(struct l2cap_chan *chan, __u16 mtu); int l2cap_chan_send(struct l2cap_chan *chan, struct msghdr *msg, size_t len); void l2cap_chan_busy(struct l2cap_chan *chan, int busy); +void l2cap_chan_rx_avail(struct l2cap_chan *chan, ssize_t rx_avail); int l2cap_chan_check_security(struct l2cap_chan *chan, bool initiator); void l2cap_chan_set_defaults(struct l2cap_chan *chan); int l2cap_ertm_init(struct l2cap_chan *chan); diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index fa714f5bd0b3..443423538a4c 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -457,6 +457,9 @@ struct l2cap_chan *l2cap_chan_create(void) /* Set default lock nesting level */ atomic_set(&chan->nesting, L2CAP_NESTING_NORMAL); + /* Available receive buffer space is initially unknown */ + chan->rx_avail = -1; + write_lock(&chan_list_lock); list_add(&chan->global_l, &chan_list); write_unlock(&chan_list_lock); @@ -538,6 +541,28 @@ void l2cap_chan_set_defaults(struct l2cap_chan *chan) } EXPORT_SYMBOL_GPL(l2cap_chan_set_defaults); +static __u16 l2cap_le_rx_credits(struct l2cap_chan *chan) +{ + size_t sdu_len = chan->sdu ? chan->sdu->len : 0; + + if (chan->mps == 0) + return 0; + + /* If we don't know the available space in the receiver buffer, give + * enough credits for a full packet. + */ + if (chan->rx_avail == -1) + return (chan->imtu / chan->mps) + 1; + + /* If we know how much space is available in the receive buffer, give + * out as many credits as would fill the buffer. + */ + if (chan->rx_avail <= sdu_len) + return 0; + + return DIV_ROUND_UP(chan->rx_avail - sdu_len, chan->mps); +} + static void l2cap_le_flowctl_init(struct l2cap_chan *chan, u16 tx_credits) { chan->sdu = NULL; @@ -546,8 +571,7 @@ static void l2cap_le_flowctl_init(struct l2cap_chan *chan, u16 tx_credits) chan->tx_credits = tx_credits; /* Derive MPS from connection MTU to stop HCI fragmentation */ chan->mps = min_t(u16, chan->imtu, chan->conn->mtu - L2CAP_HDR_SIZE); - /* Give enough credits for a full packet */ - chan->rx_credits = (chan->imtu / chan->mps) + 1; + chan->rx_credits = l2cap_le_rx_credits(chan); skb_queue_head_init(&chan->tx_q); } @@ -559,7 +583,7 @@ static void l2cap_ecred_init(struct l2cap_chan *chan, u16 tx_credits) /* L2CAP implementations shall support a minimum MPS of 64 octets */ if (chan->mps < L2CAP_ECRED_MIN_MPS) { chan->mps = L2CAP_ECRED_MIN_MPS; - chan->rx_credits = (chan->imtu / chan->mps) + 1; + chan->rx_credits = l2cap_le_rx_credits(chan); } } @@ -6512,9 +6536,7 @@ static void l2cap_chan_le_send_credits(struct l2cap_chan *chan) { struct l2cap_conn *conn = chan->conn; struct l2cap_le_credits pkt; - u16 return_credits; - - return_credits = (chan->imtu / chan->mps) + 1; + u16 return_credits = l2cap_le_rx_credits(chan); if (chan->rx_credits >= return_credits) return; @@ -6533,6 +6555,19 @@ static void l2cap_chan_le_send_credits(struct l2cap_chan *chan) l2cap_send_cmd(conn, chan->ident, L2CAP_LE_CREDITS, sizeof(pkt), &pkt); } +void l2cap_chan_rx_avail(struct l2cap_chan *chan, ssize_t rx_avail) +{ + if (chan->rx_avail == rx_avail) + return; + + BT_DBG("chan %p has %zd bytes avail for rx", chan, rx_avail); + + chan->rx_avail = rx_avail; + + if (chan->state == BT_CONNECTED) + l2cap_chan_le_send_credits(chan); +} + static int l2cap_ecred_recv(struct l2cap_chan *chan, struct sk_buff *skb) { int err; @@ -6542,6 +6577,12 @@ static int l2cap_ecred_recv(struct l2cap_chan *chan, struct sk_buff *skb) /* Wait recv to confirm reception before updating the credits */ err = chan->ops->recv(chan, skb); + if (err < 0 && chan->rx_avail != -1) { + BT_ERR("Queueing received LE L2CAP data failed"); + l2cap_send_disconn_req(chan, ECONNRESET); + return err; + } + /* Update credits whenever an SDU is received */ l2cap_chan_le_send_credits(chan); @@ -6564,7 +6605,8 @@ static int l2cap_ecred_data_rcv(struct l2cap_chan *chan, struct sk_buff *skb) } chan->rx_credits--; - BT_DBG("rx_credits %u -> %u", chan->rx_credits + 1, chan->rx_credits); + BT_DBG("chan %p: rx_credits %u -> %u", + chan, chan->rx_credits + 1, chan->rx_credits); /* Update if remote had run out of credits, this should only happens * if the remote is not using the entire MPS. diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 5cc83f906c12..8645461d45e8 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1131,6 +1131,34 @@ static int l2cap_sock_sendmsg(struct socket *sock, struct msghdr *msg, return err; } +static void l2cap_publish_rx_avail(struct l2cap_chan *chan) +{ + struct sock *sk = chan->data; + ssize_t avail = sk->sk_rcvbuf - atomic_read(&sk->sk_rmem_alloc); + int expected_skbs, skb_overhead; + + if (avail <= 0) { + l2cap_chan_rx_avail(chan, 0); + return; + } + + if (!chan->mps) { + l2cap_chan_rx_avail(chan, -1); + return; + } + + /* Correct available memory by estimated sk_buff overhead. + * This is significant due to small transfer sizes. However, accept + * at least one full packet if receive space is non-zero. + */ + expected_skbs = DIV_ROUND_UP(avail, chan->mps); + skb_overhead = expected_skbs * sizeof(struct sk_buff); + if (skb_overhead < avail) + l2cap_chan_rx_avail(chan, avail - skb_overhead); + else + l2cap_chan_rx_avail(chan, -1); +} + static int l2cap_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, int flags) { @@ -1167,28 +1195,33 @@ static int l2cap_sock_recvmsg(struct socket *sock, struct msghdr *msg, else err = bt_sock_recvmsg(sock, msg, len, flags); - if (pi->chan->mode != L2CAP_MODE_ERTM) + if (pi->chan->mode != L2CAP_MODE_ERTM && + pi->chan->mode != L2CAP_MODE_LE_FLOWCTL && + pi->chan->mode != L2CAP_MODE_EXT_FLOWCTL) return err; - /* Attempt to put pending rx data in the socket buffer */ - lock_sock(sk); - if (!test_bit(CONN_LOCAL_BUSY, &pi->chan->conn_state)) - goto done; + l2cap_publish_rx_avail(pi->chan); - if (pi->rx_busy_skb) { - if (!__sock_queue_rcv_skb(sk, pi->rx_busy_skb)) - pi->rx_busy_skb = NULL; - else + /* Attempt to put pending rx data in the socket buffer */ + while (!list_empty(&pi->rx_busy)) { + struct l2cap_rx_busy *rx_busy = + list_first_entry(&pi->rx_busy, + struct l2cap_rx_busy, + list); + if (__sock_queue_rcv_skb(sk, rx_busy->skb) < 0) goto done; + list_del(&rx_busy->list); + kfree(rx_busy); } /* Restore data flow when half of the receive buffer is * available. This avoids resending large numbers of * frames. */ - if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf >> 1) + if (test_bit(CONN_LOCAL_BUSY, &pi->chan->conn_state) && + atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf >> 1) l2cap_chan_busy(pi->chan, 0); done: @@ -1449,17 +1482,20 @@ static struct l2cap_chan *l2cap_sock_new_connection_cb(struct l2cap_chan *chan) static int l2cap_sock_recv_cb(struct l2cap_chan *chan, struct sk_buff *skb) { struct sock *sk = chan->data; + struct l2cap_pinfo *pi = l2cap_pi(sk); int err; lock_sock(sk); - if (l2cap_pi(sk)->rx_busy_skb) { + if (chan->mode == L2CAP_MODE_ERTM && !list_empty(&pi->rx_busy)) { err = -ENOMEM; goto done; } if (chan->mode != L2CAP_MODE_ERTM && - chan->mode != L2CAP_MODE_STREAMING) { + chan->mode != L2CAP_MODE_STREAMING && + chan->mode != L2CAP_MODE_LE_FLOWCTL && + chan->mode != L2CAP_MODE_EXT_FLOWCTL) { /* Even if no filter is attached, we could potentially * get errors from security modules, etc. */ @@ -1470,7 +1506,9 @@ static int l2cap_sock_recv_cb(struct l2cap_chan *chan, struct sk_buff *skb) err = __sock_queue_rcv_skb(sk, skb); - /* For ERTM, handle one skb that doesn't fit into the recv + l2cap_publish_rx_avail(chan); + + /* For ERTM and LE, handle a skb that doesn't fit into the recv * buffer. This is important to do because the data frames * have already been acked, so the skb cannot be discarded. * @@ -1479,8 +1517,18 @@ static int l2cap_sock_recv_cb(struct l2cap_chan *chan, struct sk_buff *skb) * acked and reassembled until there is buffer space * available. */ - if (err < 0 && chan->mode == L2CAP_MODE_ERTM) { - l2cap_pi(sk)->rx_busy_skb = skb; + if (err < 0 && + (chan->mode == L2CAP_MODE_ERTM || + chan->mode == L2CAP_MODE_LE_FLOWCTL || + chan->mode == L2CAP_MODE_EXT_FLOWCTL)) { + struct l2cap_rx_busy *rx_busy = + kmalloc(sizeof(*rx_busy), GFP_KERNEL); + if (!rx_busy) { + err = -ENOMEM; + goto done; + } + rx_busy->skb = skb; + list_add_tail(&rx_busy->list, &pi->rx_busy); l2cap_chan_busy(chan, 1); err = 0; } @@ -1706,6 +1754,8 @@ static const struct l2cap_ops l2cap_chan_ops = { static void l2cap_sock_destruct(struct sock *sk) { + struct l2cap_rx_busy *rx_busy, *next; + BT_DBG("sk %p", sk); if (l2cap_pi(sk)->chan) { @@ -1713,9 +1763,10 @@ static void l2cap_sock_destruct(struct sock *sk) l2cap_chan_put(l2cap_pi(sk)->chan); } - if (l2cap_pi(sk)->rx_busy_skb) { - kfree_skb(l2cap_pi(sk)->rx_busy_skb); - l2cap_pi(sk)->rx_busy_skb = NULL; + list_for_each_entry_safe(rx_busy, next, &l2cap_pi(sk)->rx_busy, list) { + kfree_skb(rx_busy->skb); + list_del(&rx_busy->list); + kfree(rx_busy); } skb_queue_purge(&sk->sk_receive_queue); @@ -1799,6 +1850,8 @@ static void l2cap_sock_init(struct sock *sk, struct sock *parent) chan->data = sk; chan->ops = &l2cap_chan_ops; + + l2cap_publish_rx_avail(chan); } static struct proto l2cap_proto = { @@ -1820,6 +1873,8 @@ static struct sock *l2cap_sock_alloc(struct net *net, struct socket *sock, sk->sk_destruct = l2cap_sock_destruct; sk->sk_sndtimeo = L2CAP_CONN_TIMEOUT; + INIT_LIST_HEAD(&l2cap_pi(sk)->rx_busy); + chan = l2cap_chan_create(); if (!chan) { sk_free(sk); -- cgit v1.2.3 From 21d74b6b4e82d27eead168d2b1ce3096937c0237 Mon Sep 17 00:00:00 2001 From: Mahesh Talewad Date: Thu, 2 May 2024 12:01:23 +0300 Subject: LE Create Connection command timeout increased to 20 secs On our DUT, we can see that the host issues create connection cancel command after 4-sec if there is no connection complete event for LE create connection cmd. As per core spec v5.3 section 7.8.5, advertisement interval range is- Advertising_Interval_Min Default : 0x0800(1.28s) Time Range: 20ms to 10.24s Advertising_Interval_Max Default : 0x0800(1.28s) Time Range: 20ms to 10.24s If the remote device is using adv interval of > 4 sec, it is difficult to make a connection with the current timeout value. Also, with the default interval of 1.28 sec, we will get only 3 chances to capture the adv packets with the 4 sec window. Hence we want to increase this timeout to 20sec. Signed-off-by: Mahesh Talewad Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 1 - net/bluetooth/hci_core.c | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index c4c6b8810701..38eb7ec86a1a 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -456,7 +456,6 @@ enum { #define HCI_AUTO_OFF_TIMEOUT msecs_to_jiffies(2000) /* 2 seconds */ #define HCI_ACL_CONN_TIMEOUT msecs_to_jiffies(20000) /* 20 seconds */ #define HCI_LE_CONN_TIMEOUT msecs_to_jiffies(20000) /* 20 seconds */ -#define HCI_LE_AUTOCONN_TIMEOUT msecs_to_jiffies(4000) /* 4 seconds */ /* HCI data types */ #define HCI_COMMAND_PKT 0x01 diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 34cd05856401..b52125512985 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -2556,7 +2556,7 @@ struct hci_dev *hci_alloc_dev_priv(int sizeof_priv) hdev->le_rx_def_phys = HCI_LE_SET_PHY_1M; hdev->le_num_of_adv_sets = HCI_MAX_ADV_INSTANCES; hdev->def_multi_adv_rotation_duration = HCI_DEFAULT_ADV_DURATION; - hdev->def_le_autoconnect_timeout = HCI_LE_AUTOCONN_TIMEOUT; + hdev->def_le_autoconnect_timeout = HCI_LE_CONN_TIMEOUT; hdev->min_le_tx_power = HCI_TX_POWER_INVALID; hdev->max_le_tx_power = HCI_TX_POWER_INVALID; -- cgit v1.2.3 From ea9e148c803b24ebbc7a74171f22f42c8fd8d644 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Thu, 2 May 2024 10:22:00 -0600 Subject: Bluetooth: hci_conn: Use __counted_by() and avoid -Wfamnae warning Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). Also, -Wflex-array-member-not-at-end is coming in GCC-14, and we are getting ready to enable it globally. So, use the `DEFINE_FLEX()` helper for an on-stack definition of a flexible structure where the size of the flexible-array member is known at compile-time, and refactor the rest of the code, accordingly. With these changes, fix the following warning: net/bluetooth/hci_conn.c:669:41: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end] Link: https://github.com/KSPP/linux/issues/202 Signed-off-by: Gustavo A. R. Silva Reviewed-by: Kees Cook Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 2 +- net/bluetooth/hci_conn.c | 38 ++++++++++++++++---------------------- 2 files changed, 17 insertions(+), 23 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 38eb7ec86a1a..21ebd70f3dcc 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -2143,7 +2143,7 @@ struct hci_cp_le_set_cig_params { __le16 c_latency; __le16 p_latency; __u8 num_cis; - struct hci_cis_params cis[]; + struct hci_cis_params cis[] __counted_by(num_cis); } __packed; struct hci_rp_le_set_cig_params { diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index fc1aeb513b08..b7f8e99001dd 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -664,11 +664,6 @@ static void le_conn_timeout(struct work_struct *work) hci_abort_conn(conn, HCI_ERROR_REMOTE_USER_TERM); } -struct iso_cig_params { - struct hci_cp_le_set_cig_params cp; - struct hci_cis_params cis[0x1f]; -}; - struct iso_list_data { union { u8 cig; @@ -1722,34 +1717,33 @@ static int hci_le_create_big(struct hci_conn *conn, struct bt_iso_qos *qos) static int set_cig_params_sync(struct hci_dev *hdev, void *data) { + DEFINE_FLEX(struct hci_cp_le_set_cig_params, pdu, cis, num_cis, 0x1f); u8 cig_id = PTR_UINT(data); struct hci_conn *conn; struct bt_iso_qos *qos; - struct iso_cig_params pdu; + u8 aux_num_cis = 0; u8 cis_id; conn = hci_conn_hash_lookup_cig(hdev, cig_id); if (!conn) return 0; - memset(&pdu, 0, sizeof(pdu)); - qos = &conn->iso_qos; - pdu.cp.cig_id = cig_id; - hci_cpu_to_le24(qos->ucast.out.interval, pdu.cp.c_interval); - hci_cpu_to_le24(qos->ucast.in.interval, pdu.cp.p_interval); - pdu.cp.sca = qos->ucast.sca; - pdu.cp.packing = qos->ucast.packing; - pdu.cp.framing = qos->ucast.framing; - pdu.cp.c_latency = cpu_to_le16(qos->ucast.out.latency); - pdu.cp.p_latency = cpu_to_le16(qos->ucast.in.latency); + pdu->cig_id = cig_id; + hci_cpu_to_le24(qos->ucast.out.interval, pdu->c_interval); + hci_cpu_to_le24(qos->ucast.in.interval, pdu->p_interval); + pdu->sca = qos->ucast.sca; + pdu->packing = qos->ucast.packing; + pdu->framing = qos->ucast.framing; + pdu->c_latency = cpu_to_le16(qos->ucast.out.latency); + pdu->p_latency = cpu_to_le16(qos->ucast.in.latency); /* Reprogram all CIS(s) with the same CIG, valid range are: * num_cis: 0x00 to 0x1F * cis_id: 0x00 to 0xEF */ for (cis_id = 0x00; cis_id < 0xf0 && - pdu.cp.num_cis < ARRAY_SIZE(pdu.cis); cis_id++) { + aux_num_cis < pdu->num_cis; cis_id++) { struct hci_cis_params *cis; conn = hci_conn_hash_lookup_cis(hdev, NULL, 0, cig_id, cis_id); @@ -1758,7 +1752,7 @@ static int set_cig_params_sync(struct hci_dev *hdev, void *data) qos = &conn->iso_qos; - cis = &pdu.cis[pdu.cp.num_cis++]; + cis = &pdu->cis[aux_num_cis++]; cis->cis_id = cis_id; cis->c_sdu = cpu_to_le16(conn->iso_qos.ucast.out.sdu); cis->p_sdu = cpu_to_le16(conn->iso_qos.ucast.in.sdu); @@ -1769,14 +1763,14 @@ static int set_cig_params_sync(struct hci_dev *hdev, void *data) cis->c_rtn = qos->ucast.out.rtn; cis->p_rtn = qos->ucast.in.rtn; } + pdu->num_cis = aux_num_cis; - if (!pdu.cp.num_cis) + if (!pdu->num_cis) return 0; return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_CIG_PARAMS, - sizeof(pdu.cp) + - pdu.cp.num_cis * sizeof(pdu.cis[0]), &pdu, - HCI_CMD_TIMEOUT); + struct_size(pdu, cis, pdu->num_cis), + pdu, HCI_CMD_TIMEOUT); } static bool hci_le_set_cig_params(struct hci_conn *conn, struct bt_iso_qos *qos) -- cgit v1.2.3 From a5b862c6a221459d54e494e88965b48dcfa6cc44 Mon Sep 17 00:00:00 2001 From: Sungwoo Kim Date: Sat, 4 May 2024 15:23:29 -0400 Subject: Bluetooth: L2CAP: Fix div-by-zero in l2cap_le_flowctl_init() l2cap_le_flowctl_init() can cause both div-by-zero and an integer overflow since hdev->le_mtu may not fall in the valid range. Move MTU from hci_dev to hci_conn to validate MTU and stop the connection process earlier if MTU is invalid. Also, add a missing validation in read_buffer_size() and make it return an error value if the validation fails. Now hci_conn_add() returns ERR_PTR() as it can fail due to the both a kzalloc failure and invalid MTU value. divide error: 0000 [#1] PREEMPT SMP KASAN NOPTI CPU: 0 PID: 67 Comm: kworker/u5:0 Tainted: G W 6.9.0-rc5+ #20 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Workqueue: hci0 hci_rx_work RIP: 0010:l2cap_le_flowctl_init+0x19e/0x3f0 net/bluetooth/l2cap_core.c:547 Code: e8 17 17 0c 00 66 41 89 9f 84 00 00 00 bf 01 00 00 00 41 b8 02 00 00 00 4c 89 fe 4c 89 e2 89 d9 e8 27 17 0c 00 44 89 f0 31 d2 <66> f7 f3 89 c3 ff c3 4d 8d b7 88 00 00 00 4c 89 f0 48 c1 e8 03 42 RSP: 0018:ffff88810bc0f858 EFLAGS: 00010246 RAX: 00000000000002a0 RBX: 0000000000000000 RCX: dffffc0000000000 RDX: 0000000000000000 RSI: ffff88810bc0f7c0 RDI: ffffc90002dcb66f RBP: ffff88810bc0f880 R08: aa69db2dda70ff01 R09: 0000ffaaaaaaaaaa R10: 0084000000ffaaaa R11: 0000000000000000 R12: ffff88810d65a084 R13: dffffc0000000000 R14: 00000000000002a0 R15: ffff88810d65a000 FS: 0000000000000000(0000) GS:ffff88811ac00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020000100 CR3: 0000000103268003 CR4: 0000000000770ef0 PKRU: 55555554 Call Trace: l2cap_le_connect_req net/bluetooth/l2cap_core.c:4902 [inline] l2cap_le_sig_cmd net/bluetooth/l2cap_core.c:5420 [inline] l2cap_le_sig_channel net/bluetooth/l2cap_core.c:5486 [inline] l2cap_recv_frame+0xe59d/0x11710 net/bluetooth/l2cap_core.c:6809 l2cap_recv_acldata+0x544/0x10a0 net/bluetooth/l2cap_core.c:7506 hci_acldata_packet net/bluetooth/hci_core.c:3939 [inline] hci_rx_work+0x5e5/0xb20 net/bluetooth/hci_core.c:4176 process_one_work kernel/workqueue.c:3254 [inline] process_scheduled_works+0x90f/0x1530 kernel/workqueue.c:3335 worker_thread+0x926/0xe70 kernel/workqueue.c:3416 kthread+0x2e3/0x380 kernel/kthread.c:388 ret_from_fork+0x5c/0x90 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 Modules linked in: ---[ end trace 0000000000000000 ]--- Fixes: 6ed58ec520ad ("Bluetooth: Use LE buffers for LE traffic") Suggested-by: Luiz Augusto von Dentz Signed-off-by: Sungwoo Kim Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 9 +++++ include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_conn.c | 75 +++++++++++++++++++++++++++------------- net/bluetooth/hci_event.c | 31 +++++++++++------ net/bluetooth/iso.c | 2 +- net/bluetooth/l2cap_core.c | 17 ++------- net/bluetooth/sco.c | 6 ++-- 7 files changed, 88 insertions(+), 53 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 21ebd70f3dcc..de530262d824 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -1665,6 +1665,15 @@ struct hci_cp_le_set_event_mask { __u8 mask[8]; } __packed; +/* BLUETOOTH CORE SPECIFICATION Version 5.4 | Vol 4, Part E + * 7.8.2 LE Read Buffer Size command + * MAX_LE_MTU is 0xffff. + * 0 is also valid. It means that no dedicated LE Buffer exists. + * It should use the HCI_Read_Buffer_Size command and mtu is shared + * between BR/EDR and LE. + */ +#define HCI_MIN_LE_MTU 0x001b + #define HCI_OP_LE_READ_BUFFER_SIZE 0x2002 struct hci_rp_le_read_buffer_size { __u8 status; diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 2283d12ee358..be012f393bcb 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -706,6 +706,7 @@ struct hci_conn { __u16 handle; __u16 sync_handle; __u16 state; + __u16 mtu; __u8 mode; __u8 type; __u8 role; diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index b7f8e99001dd..f21e7c73021d 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -904,11 +904,37 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, { struct hci_conn *conn; + switch (type) { + case ACL_LINK: + if (!hdev->acl_mtu) + return ERR_PTR(-ECONNREFUSED); + break; + case ISO_LINK: + if (hdev->iso_mtu) + /* Dedicated ISO Buffer exists */ + break; + fallthrough; + case LE_LINK: + if (hdev->le_mtu && hdev->le_mtu < HCI_MIN_LE_MTU) + return ERR_PTR(-ECONNREFUSED); + if (!hdev->le_mtu && hdev->acl_mtu < HCI_MIN_LE_MTU) + return ERR_PTR(-ECONNREFUSED); + break; + case SCO_LINK: + case ESCO_LINK: + if (!hdev->sco_pkts) + /* Controller does not support SCO or eSCO over HCI */ + return ERR_PTR(-ECONNREFUSED); + break; + default: + return ERR_PTR(-ECONNREFUSED); + } + bt_dev_dbg(hdev, "dst %pMR handle 0x%4.4x", dst, handle); conn = kzalloc(sizeof(*conn), GFP_KERNEL); if (!conn) - return NULL; + return ERR_PTR(-ENOMEM); bacpy(&conn->dst, dst); bacpy(&conn->src, &hdev->bdaddr); @@ -939,10 +965,12 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, switch (type) { case ACL_LINK: conn->pkt_type = hdev->pkt_type & ACL_PTYPE_MASK; + conn->mtu = hdev->acl_mtu; break; case LE_LINK: /* conn->src should reflect the local identity address */ hci_copy_identity_address(hdev, &conn->src, &conn->src_type); + conn->mtu = hdev->le_mtu ? hdev->le_mtu : hdev->acl_mtu; break; case ISO_LINK: /* conn->src should reflect the local identity address */ @@ -954,6 +982,8 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, else if (conn->role == HCI_ROLE_MASTER) conn->cleanup = cis_cleanup; + conn->mtu = hdev->iso_mtu ? hdev->iso_mtu : + hdev->le_mtu ? hdev->le_mtu : hdev->acl_mtu; break; case SCO_LINK: if (lmp_esco_capable(hdev)) @@ -961,9 +991,12 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst, (hdev->esco_type & EDR_ESCO_MASK); else conn->pkt_type = hdev->pkt_type & SCO_PTYPE_MASK; + + conn->mtu = hdev->sco_mtu; break; case ESCO_LINK: conn->pkt_type = hdev->esco_type & ~EDR_ESCO_MASK; + conn->mtu = hdev->sco_mtu; break; } @@ -1006,7 +1039,7 @@ struct hci_conn *hci_conn_add_unset(struct hci_dev *hdev, int type, handle = hci_conn_hash_alloc_unset(hdev); if (unlikely(handle < 0)) - return NULL; + return ERR_PTR(-ECONNREFUSED); return hci_conn_add(hdev, type, dst, role, handle); } @@ -1312,8 +1345,8 @@ struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, bacpy(&conn->dst, dst); } else { conn = hci_conn_add_unset(hdev, LE_LINK, dst, role); - if (!conn) - return ERR_PTR(-ENOMEM); + if (IS_ERR(conn)) + return conn; hci_conn_hold(conn); conn->pending_sec_level = sec_level; } @@ -1489,8 +1522,8 @@ static struct hci_conn *hci_add_bis(struct hci_dev *hdev, bdaddr_t *dst, return ERR_PTR(-EADDRINUSE); conn = hci_conn_add_unset(hdev, ISO_LINK, dst, HCI_ROLE_MASTER); - if (!conn) - return ERR_PTR(-ENOMEM); + if (IS_ERR(conn)) + return conn; conn->state = BT_CONNECT; @@ -1533,8 +1566,8 @@ struct hci_conn *hci_connect_le_scan(struct hci_dev *hdev, bdaddr_t *dst, BT_DBG("requesting refresh of dst_addr"); conn = hci_conn_add_unset(hdev, LE_LINK, dst, HCI_ROLE_MASTER); - if (!conn) - return ERR_PTR(-ENOMEM); + if (IS_ERR(conn)) + return conn; if (hci_explicit_conn_params_set(hdev, dst, dst_type) < 0) { hci_conn_del(conn); @@ -1581,8 +1614,8 @@ struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst, acl = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst); if (!acl) { acl = hci_conn_add_unset(hdev, ACL_LINK, dst, HCI_ROLE_MASTER); - if (!acl) - return ERR_PTR(-ENOMEM); + if (IS_ERR(acl)) + return acl; } hci_conn_hold(acl); @@ -1650,9 +1683,9 @@ struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, bdaddr_t *dst, sco = hci_conn_hash_lookup_ba(hdev, type, dst); if (!sco) { sco = hci_conn_add_unset(hdev, type, dst, HCI_ROLE_MASTER); - if (!sco) { + if (IS_ERR(sco)) { hci_conn_drop(acl); - return ERR_PTR(-ENOMEM); + return sco; } } @@ -1841,8 +1874,8 @@ struct hci_conn *hci_bind_cis(struct hci_dev *hdev, bdaddr_t *dst, qos->ucast.cis); if (!cis) { cis = hci_conn_add_unset(hdev, ISO_LINK, dst, HCI_ROLE_MASTER); - if (!cis) - return ERR_PTR(-ENOMEM); + if (IS_ERR(cis)) + return cis; cis->cleanup = cis_cleanup; cis->dst_type = dst_type; cis->iso_qos.ucast.cig = BT_ISO_QOS_CIG_UNSET; @@ -1977,14 +2010,8 @@ static void hci_iso_qos_setup(struct hci_dev *hdev, struct hci_conn *conn, struct bt_iso_io_qos *qos, __u8 phy) { /* Only set MTU if PHY is enabled */ - if (!qos->sdu && qos->phy) { - if (hdev->iso_mtu > 0) - qos->sdu = hdev->iso_mtu; - else if (hdev->le_mtu > 0) - qos->sdu = hdev->le_mtu; - else - qos->sdu = hdev->acl_mtu; - } + if (!qos->sdu && qos->phy) + qos->sdu = conn->mtu; /* Use the same PHY as ACL if set to any */ if (qos->phy == BT_ISO_PHY_ANY) @@ -2065,8 +2092,8 @@ struct hci_conn *hci_pa_create_sync(struct hci_dev *hdev, bdaddr_t *dst, return ERR_PTR(-EBUSY); conn = hci_conn_add_unset(hdev, ISO_LINK, dst, HCI_ROLE_SLAVE); - if (!conn) - return ERR_PTR(-ENOMEM); + if (IS_ERR(conn)) + return conn; conn->iso_qos = *qos; conn->state = BT_LISTEN; diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 122a6a9092e0..97c6f34ba640 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -954,6 +954,9 @@ static u8 hci_cc_read_buffer_size(struct hci_dev *hdev, void *data, BT_DBG("%s acl mtu %d:%d sco mtu %d:%d", hdev->name, hdev->acl_mtu, hdev->acl_pkts, hdev->sco_mtu, hdev->sco_pkts); + if (!hdev->acl_mtu || !hdev->acl_pkts) + return HCI_ERROR_INVALID_PARAMETERS; + return rp->status; } @@ -1263,6 +1266,9 @@ static u8 hci_cc_le_read_buffer_size(struct hci_dev *hdev, void *data, BT_DBG("%s le mtu %d:%d", hdev->name, hdev->le_mtu, hdev->le_pkts); + if (hdev->le_mtu && hdev->le_mtu < HCI_MIN_LE_MTU) + return HCI_ERROR_INVALID_PARAMETERS; + return rp->status; } @@ -2341,8 +2347,8 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status) if (!conn) { conn = hci_conn_add_unset(hdev, ACL_LINK, &cp->bdaddr, HCI_ROLE_MASTER); - if (!conn) - bt_dev_err(hdev, "no memory for new connection"); + if (IS_ERR(conn)) + bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); } } @@ -3153,8 +3159,8 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, void *data, BDADDR_BREDR)) { conn = hci_conn_add_unset(hdev, ev->link_type, &ev->bdaddr, HCI_ROLE_SLAVE); - if (!conn) { - bt_dev_err(hdev, "no memory for new conn"); + if (IS_ERR(conn)) { + bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); goto unlock; } } else { @@ -3342,8 +3348,8 @@ static void hci_conn_request_evt(struct hci_dev *hdev, void *data, if (!conn) { conn = hci_conn_add_unset(hdev, ev->link_type, &ev->bdaddr, HCI_ROLE_SLAVE); - if (!conn) { - bt_dev_err(hdev, "no memory for new connection"); + if (IS_ERR(conn)) { + bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); goto unlock; } } @@ -3820,6 +3826,9 @@ static u8 hci_cc_le_read_buffer_size_v2(struct hci_dev *hdev, void *data, BT_DBG("%s acl mtu %d:%d iso mtu %d:%d", hdev->name, hdev->acl_mtu, hdev->acl_pkts, hdev->iso_mtu, hdev->iso_pkts); + if (hdev->le_mtu && hdev->le_mtu < HCI_MIN_LE_MTU) + return HCI_ERROR_INVALID_PARAMETERS; + return rp->status; } @@ -5768,8 +5777,8 @@ static void le_conn_complete_evt(struct hci_dev *hdev, u8 status, goto unlock; conn = hci_conn_add_unset(hdev, LE_LINK, bdaddr, role); - if (!conn) { - bt_dev_err(hdev, "no memory for new connection"); + if (IS_ERR(conn)) { + bt_dev_err(hdev, "connection err: %ld", PTR_ERR(conn)); goto unlock; } @@ -6497,7 +6506,7 @@ static void hci_le_pa_sync_estabilished_evt(struct hci_dev *hdev, void *data, pa_sync = hci_conn_add_unset(hdev, ISO_LINK, BDADDR_ANY, HCI_ROLE_SLAVE); - if (!pa_sync) + if (IS_ERR(pa_sync)) goto unlock; pa_sync->sync_handle = le16_to_cpu(ev->handle); @@ -6921,7 +6930,7 @@ static void hci_le_cis_req_evt(struct hci_dev *hdev, void *data, if (!cis) { cis = hci_conn_add(hdev, ISO_LINK, &acl->dst, HCI_ROLE_SLAVE, cis_handle); - if (!cis) { + if (IS_ERR(cis)) { hci_le_reject_cis(hdev, ev->cis_handle); goto unlock; } @@ -7030,7 +7039,7 @@ static void hci_le_big_sync_established_evt(struct hci_dev *hdev, void *data, if (!bis) { bis = hci_conn_add(hdev, ISO_LINK, BDADDR_ANY, HCI_ROLE_SLAVE, handle); - if (!bis) + if (IS_ERR(bis)) continue; } diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index d0ef7290e5b7..00c0d8413c63 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1258,7 +1258,7 @@ static int iso_sock_sendmsg(struct socket *sock, struct msghdr *msg, return -ENOTCONN; } - mtu = iso_pi(sk)->conn->hcon->hdev->iso_mtu; + mtu = iso_pi(sk)->conn->hcon->mtu; release_sock(sk); diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 443423538a4c..adc8f990f13e 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -6264,7 +6264,7 @@ static int l2cap_finish_move(struct l2cap_chan *chan) BT_DBG("chan %p", chan); chan->rx_state = L2CAP_RX_STATE_RECV; - chan->conn->mtu = chan->conn->hcon->hdev->acl_mtu; + chan->conn->mtu = chan->conn->hcon->mtu; return l2cap_resegment(chan); } @@ -6331,7 +6331,7 @@ static int l2cap_rx_state_wait_f(struct l2cap_chan *chan, */ chan->next_tx_seq = control->reqseq; chan->unacked_frames = 0; - chan->conn->mtu = chan->conn->hcon->hdev->acl_mtu; + chan->conn->mtu = chan->conn->hcon->mtu; err = l2cap_resegment(chan); @@ -6889,18 +6889,7 @@ static struct l2cap_conn *l2cap_conn_add(struct hci_conn *hcon) BT_DBG("hcon %p conn %p hchan %p", hcon, conn, hchan); - switch (hcon->type) { - case LE_LINK: - if (hcon->hdev->le_mtu) { - conn->mtu = hcon->hdev->le_mtu; - break; - } - fallthrough; - default: - conn->mtu = hcon->hdev->acl_mtu; - break; - } - + conn->mtu = hcon->mtu; conn->feat_mask = 0; conn->local_fixed_chan = L2CAP_FC_SIG_BREDR | L2CAP_FC_CONNLESS; diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index e0ad30862ee4..71d36582d4ef 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -126,7 +126,6 @@ static void sco_sock_clear_timer(struct sock *sk) /* ---- SCO connections ---- */ static struct sco_conn *sco_conn_add(struct hci_conn *hcon) { - struct hci_dev *hdev = hcon->hdev; struct sco_conn *conn = hcon->sco_data; if (conn) { @@ -144,9 +143,10 @@ static struct sco_conn *sco_conn_add(struct hci_conn *hcon) hcon->sco_data = conn; conn->hcon = hcon; + conn->mtu = hcon->mtu; - if (hdev->sco_mtu > 0) - conn->mtu = hdev->sco_mtu; + if (hcon->mtu > 0) + conn->mtu = hcon->mtu; else conn->mtu = 60; -- cgit v1.2.3 From 84a4bb6548a29326564f0e659fb8064503ecc1c7 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 6 May 2024 18:33:52 -0400 Subject: Bluetooth: HCI: Remove HCI_AMP support Since BT_HS has been remove HCI_AMP controllers no longer has any use so remove it along with the capability of creating AMP controllers. Since we no longer need to differentiate between AMP and Primary controllers, as only HCI_PRIMARY is left, this also remove hdev->dev_type altogether. Fixes: e7b02296fb40 ("Bluetooth: Remove BT_HS") Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btmrvl_main.c | 9 --- drivers/bluetooth/btrsi.c | 1 - drivers/bluetooth/btsdio.c | 8 --- drivers/bluetooth/btusb.c | 5 -- drivers/bluetooth/hci_bcm4377.c | 1 - drivers/bluetooth/hci_ldisc.c | 6 -- drivers/bluetooth/hci_serdev.c | 5 -- drivers/bluetooth/hci_uart.h | 1 - drivers/bluetooth/hci_vhci.c | 10 +-- drivers/bluetooth/virtio_bt.c | 2 - include/net/bluetooth/hci.h | 114 ------------------------------ include/net/bluetooth/hci_core.h | 46 +----------- include/uapi/linux/virtio_bt.h | 1 - net/bluetooth/hci_conn.c | 3 +- net/bluetooth/hci_core.c | 132 +++-------------------------------- net/bluetooth/hci_event.c | 147 --------------------------------------- net/bluetooth/hci_sock.c | 5 +- net/bluetooth/hci_sync.c | 112 ++--------------------------- net/bluetooth/l2cap_core.c | 21 ++---- net/bluetooth/mgmt.c | 84 +++++++--------------- 20 files changed, 49 insertions(+), 664 deletions(-) (limited to 'net') diff --git a/drivers/bluetooth/btmrvl_main.c b/drivers/bluetooth/btmrvl_main.c index 9658b33c824a..18f34998a120 100644 --- a/drivers/bluetooth/btmrvl_main.c +++ b/drivers/bluetooth/btmrvl_main.c @@ -121,13 +121,6 @@ int btmrvl_process_event(struct btmrvl_private *priv, struct sk_buff *skb) ((event->data[2] == MODULE_BROUGHT_UP) || (event->data[2] == MODULE_ALREADY_UP)) ? "Bring-up succeed" : "Bring-up failed"); - - if (event->length > 3 && event->data[3]) - priv->btmrvl_dev.dev_type = HCI_AMP; - else - priv->btmrvl_dev.dev_type = HCI_PRIMARY; - - BT_DBG("dev_type: %d", priv->btmrvl_dev.dev_type); } else if (priv->btmrvl_dev.sendcmdflag && event->data[1] == MODULE_SHUTDOWN_REQ) { BT_DBG("EVENT:%s", (event->data[2]) ? @@ -686,8 +679,6 @@ int btmrvl_register_hdev(struct btmrvl_private *priv) hdev->wakeup = btmrvl_wakeup; SET_HCIDEV_DEV(hdev, &card->func->dev); - hdev->dev_type = priv->btmrvl_dev.dev_type; - ret = hci_register_dev(hdev); if (ret < 0) { BT_ERR("Can not register HCI device"); diff --git a/drivers/bluetooth/btrsi.c b/drivers/bluetooth/btrsi.c index 634cf8f5ed2d..0c91d7635ac3 100644 --- a/drivers/bluetooth/btrsi.c +++ b/drivers/bluetooth/btrsi.c @@ -134,7 +134,6 @@ static int rsi_hci_attach(void *priv, struct rsi_proto_ops *ops) hdev->bus = HCI_USB; hci_set_drvdata(hdev, h_adapter); - hdev->dev_type = HCI_PRIMARY; hdev->open = rsi_hci_open; hdev->close = rsi_hci_close; hdev->flush = rsi_hci_flush; diff --git a/drivers/bluetooth/btsdio.c b/drivers/bluetooth/btsdio.c index f19d31ee37ea..fdcfe9c50313 100644 --- a/drivers/bluetooth/btsdio.c +++ b/drivers/bluetooth/btsdio.c @@ -32,9 +32,6 @@ static const struct sdio_device_id btsdio_table[] = { /* Generic Bluetooth Type-B SDIO device */ { SDIO_DEVICE_CLASS(SDIO_CLASS_BT_B) }, - /* Generic Bluetooth AMP controller */ - { SDIO_DEVICE_CLASS(SDIO_CLASS_BT_AMP) }, - { } /* Terminating entry */ }; @@ -319,11 +316,6 @@ static int btsdio_probe(struct sdio_func *func, hdev->bus = HCI_SDIO; hci_set_drvdata(hdev, data); - if (id->class == SDIO_CLASS_BT_AMP) - hdev->dev_type = HCI_AMP; - else - hdev->dev_type = HCI_PRIMARY; - data->hdev = hdev; SET_HCIDEV_DEV(hdev, &func->dev); diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 90d1f1697357..e384ef6ff050 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -4359,11 +4359,6 @@ static int btusb_probe(struct usb_interface *intf, hdev->bus = HCI_USB; hci_set_drvdata(hdev, data); - if (id->driver_info & BTUSB_AMP) - hdev->dev_type = HCI_AMP; - else - hdev->dev_type = HCI_PRIMARY; - data->hdev = hdev; SET_HCIDEV_DEV(hdev, &intf->dev); diff --git a/drivers/bluetooth/hci_bcm4377.c b/drivers/bluetooth/hci_bcm4377.c index 9a7243d5db71..0c2f15235b4c 100644 --- a/drivers/bluetooth/hci_bcm4377.c +++ b/drivers/bluetooth/hci_bcm4377.c @@ -2361,7 +2361,6 @@ static int bcm4377_probe(struct pci_dev *pdev, const struct pci_device_id *id) bcm4377->hdev = hdev; hdev->bus = HCI_PCI; - hdev->dev_type = HCI_PRIMARY; hdev->open = bcm4377_hci_open; hdev->close = bcm4377_hci_close; hdev->send = bcm4377_hci_send_frame; diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c index a26367e9fb19..17a2f158a0df 100644 --- a/drivers/bluetooth/hci_ldisc.c +++ b/drivers/bluetooth/hci_ldisc.c @@ -667,11 +667,6 @@ static int hci_uart_register_dev(struct hci_uart *hu) if (!test_bit(HCI_UART_RESET_ON_INIT, &hu->hdev_flags)) set_bit(HCI_QUIRK_RESET_ON_CLOSE, &hdev->quirks); - if (test_bit(HCI_UART_CREATE_AMP, &hu->hdev_flags)) - hdev->dev_type = HCI_AMP; - else - hdev->dev_type = HCI_PRIMARY; - /* Only call open() for the protocol after hdev is fully initialized as * open() (or a timer/workqueue it starts) may attempt to reference it. */ @@ -722,7 +717,6 @@ static int hci_uart_set_flags(struct hci_uart *hu, unsigned long flags) { unsigned long valid_flags = BIT(HCI_UART_RAW_DEVICE) | BIT(HCI_UART_RESET_ON_INIT) | - BIT(HCI_UART_CREATE_AMP) | BIT(HCI_UART_INIT_PENDING) | BIT(HCI_UART_EXT_CONFIG) | BIT(HCI_UART_VND_DETECT); diff --git a/drivers/bluetooth/hci_serdev.c b/drivers/bluetooth/hci_serdev.c index 85c0d9b68f5f..89a22e9b3253 100644 --- a/drivers/bluetooth/hci_serdev.c +++ b/drivers/bluetooth/hci_serdev.c @@ -366,11 +366,6 @@ int hci_uart_register_device_priv(struct hci_uart *hu, if (test_bit(HCI_UART_EXT_CONFIG, &hu->hdev_flags)) set_bit(HCI_QUIRK_EXTERNAL_CONFIG, &hdev->quirks); - if (test_bit(HCI_UART_CREATE_AMP, &hu->hdev_flags)) - hdev->dev_type = HCI_AMP; - else - hdev->dev_type = HCI_PRIMARY; - if (test_bit(HCI_UART_INIT_PENDING, &hu->hdev_flags)) return 0; diff --git a/drivers/bluetooth/hci_uart.h b/drivers/bluetooth/hci_uart.h index 68c8c7e95d64..00bf7ae82c5b 100644 --- a/drivers/bluetooth/hci_uart.h +++ b/drivers/bluetooth/hci_uart.h @@ -37,7 +37,6 @@ #define HCI_UART_RAW_DEVICE 0 #define HCI_UART_RESET_ON_INIT 1 -#define HCI_UART_CREATE_AMP 2 #define HCI_UART_INIT_PENDING 3 #define HCI_UART_EXT_CONFIG 4 #define HCI_UART_VND_DETECT 5 diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index 572d68d52965..28750a40f0ed 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -384,17 +384,10 @@ static int __vhci_create_device(struct vhci_data *data, __u8 opcode) { struct hci_dev *hdev; struct sk_buff *skb; - __u8 dev_type; if (data->hdev) return -EBADFD; - /* bits 0-1 are dev_type (Primary or AMP) */ - dev_type = opcode & 0x03; - - if (dev_type != HCI_PRIMARY && dev_type != HCI_AMP) - return -EINVAL; - /* bits 2-5 are reserved (must be zero) */ if (opcode & 0x3c) return -EINVAL; @@ -412,7 +405,6 @@ static int __vhci_create_device(struct vhci_data *data, __u8 opcode) data->hdev = hdev; hdev->bus = HCI_VIRTUAL; - hdev->dev_type = dev_type; hci_set_drvdata(hdev, data); hdev->open = vhci_open_dev; @@ -634,7 +626,7 @@ static void vhci_open_timeout(struct work_struct *work) struct vhci_data *data = container_of(work, struct vhci_data, open_timeout.work); - vhci_create_device(data, amp ? HCI_AMP : HCI_PRIMARY); + vhci_create_device(data, 0x00); } static int vhci_open(struct inode *inode, struct file *file) diff --git a/drivers/bluetooth/virtio_bt.c b/drivers/bluetooth/virtio_bt.c index 2ac70b560c46..18208e152a36 100644 --- a/drivers/bluetooth/virtio_bt.c +++ b/drivers/bluetooth/virtio_bt.c @@ -274,7 +274,6 @@ static int virtbt_probe(struct virtio_device *vdev) switch (type) { case VIRTIO_BT_CONFIG_TYPE_PRIMARY: - case VIRTIO_BT_CONFIG_TYPE_AMP: break; default: return -EINVAL; @@ -303,7 +302,6 @@ static int virtbt_probe(struct virtio_device *vdev) vbt->hdev = hdev; hdev->bus = HCI_VIRTIO; - hdev->dev_type = type; hci_set_drvdata(hdev, vbt); hdev->open = virtbt_open; diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index de530262d824..fe932ca3bc8c 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -33,9 +33,6 @@ #define HCI_MAX_FRAME_SIZE (HCI_MAX_ACL_SIZE + 4) #define HCI_LINK_KEY_SIZE 16 -#define HCI_AMP_LINK_KEY_SIZE (2 * HCI_LINK_KEY_SIZE) - -#define HCI_MAX_AMP_ASSOC_SIZE 672 #define HCI_MAX_CPB_DATA_SIZE 252 @@ -71,26 +68,6 @@ #define HCI_SMD 9 #define HCI_VIRTIO 10 -/* HCI controller types */ -#define HCI_PRIMARY 0x00 -#define HCI_AMP 0x01 - -/* First BR/EDR Controller shall have ID = 0 */ -#define AMP_ID_BREDR 0x00 - -/* AMP controller types */ -#define AMP_TYPE_BREDR 0x00 -#define AMP_TYPE_80211 0x01 - -/* AMP controller status */ -#define AMP_STATUS_POWERED_DOWN 0x00 -#define AMP_STATUS_BLUETOOTH_ONLY 0x01 -#define AMP_STATUS_NO_CAPACITY 0x02 -#define AMP_STATUS_LOW_CAPACITY 0x03 -#define AMP_STATUS_MEDIUM_CAPACITY 0x04 -#define AMP_STATUS_HIGH_CAPACITY 0x05 -#define AMP_STATUS_FULL_CAPACITY 0x06 - /* HCI device quirks */ enum { /* When this quirk is set, the HCI Reset command is send when @@ -527,7 +504,6 @@ enum { #define ESCO_LINK 0x02 /* Low Energy links do not have defined link type. Use invented one */ #define LE_LINK 0x80 -#define AMP_LINK 0x81 #define ISO_LINK 0x82 #define INVALID_LINK 0xff @@ -943,56 +919,6 @@ struct hci_cp_io_capability_neg_reply { __u8 reason; } __packed; -#define HCI_OP_CREATE_PHY_LINK 0x0435 -struct hci_cp_create_phy_link { - __u8 phy_handle; - __u8 key_len; - __u8 key_type; - __u8 key[HCI_AMP_LINK_KEY_SIZE]; -} __packed; - -#define HCI_OP_ACCEPT_PHY_LINK 0x0436 -struct hci_cp_accept_phy_link { - __u8 phy_handle; - __u8 key_len; - __u8 key_type; - __u8 key[HCI_AMP_LINK_KEY_SIZE]; -} __packed; - -#define HCI_OP_DISCONN_PHY_LINK 0x0437 -struct hci_cp_disconn_phy_link { - __u8 phy_handle; - __u8 reason; -} __packed; - -struct ext_flow_spec { - __u8 id; - __u8 stype; - __le16 msdu; - __le32 sdu_itime; - __le32 acc_lat; - __le32 flush_to; -} __packed; - -#define HCI_OP_CREATE_LOGICAL_LINK 0x0438 -#define HCI_OP_ACCEPT_LOGICAL_LINK 0x0439 -struct hci_cp_create_accept_logical_link { - __u8 phy_handle; - struct ext_flow_spec tx_flow_spec; - struct ext_flow_spec rx_flow_spec; -} __packed; - -#define HCI_OP_DISCONN_LOGICAL_LINK 0x043a -struct hci_cp_disconn_logical_link { - __le16 log_handle; -} __packed; - -#define HCI_OP_LOGICAL_LINK_CANCEL 0x043b -struct hci_cp_logical_link_cancel { - __u8 phy_handle; - __u8 flow_spec_id; -} __packed; - #define HCI_OP_ENHANCED_SETUP_SYNC_CONN 0x043d struct hci_coding_format { __u8 id; @@ -1614,46 +1540,6 @@ struct hci_rp_read_enc_key_size { __u8 key_size; } __packed; -#define HCI_OP_READ_LOCAL_AMP_INFO 0x1409 -struct hci_rp_read_local_amp_info { - __u8 status; - __u8 amp_status; - __le32 total_bw; - __le32 max_bw; - __le32 min_latency; - __le32 max_pdu; - __u8 amp_type; - __le16 pal_cap; - __le16 max_assoc_size; - __le32 max_flush_to; - __le32 be_flush_to; -} __packed; - -#define HCI_OP_READ_LOCAL_AMP_ASSOC 0x140a -struct hci_cp_read_local_amp_assoc { - __u8 phy_handle; - __le16 len_so_far; - __le16 max_len; -} __packed; -struct hci_rp_read_local_amp_assoc { - __u8 status; - __u8 phy_handle; - __le16 rem_len; - __u8 frag[]; -} __packed; - -#define HCI_OP_WRITE_REMOTE_AMP_ASSOC 0x140b -struct hci_cp_write_remote_amp_assoc { - __u8 phy_handle; - __le16 len_so_far; - __le16 rem_len; - __u8 frag[]; -} __packed; -struct hci_rp_write_remote_amp_assoc { - __u8 status; - __u8 phy_handle; -} __packed; - #define HCI_OP_GET_MWS_TRANSPORT_CONFIG 0x140c #define HCI_OP_ENABLE_DUT_MODE 0x1803 diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index be012f393bcb..431d0c8a0e27 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -126,7 +126,6 @@ enum suspended_state { struct hci_conn_hash { struct list_head list; unsigned int acl_num; - unsigned int amp_num; unsigned int sco_num; unsigned int iso_num; unsigned int le_num; @@ -341,14 +340,6 @@ struct adv_monitor { /* Default authenticated payload timeout 30s */ #define DEFAULT_AUTH_PAYLOAD_TIMEOUT 0x0bb8 -struct amp_assoc { - __u16 len; - __u16 offset; - __u16 rem_len; - __u16 len_so_far; - __u8 data[HCI_MAX_AMP_ASSOC_SIZE]; -}; - #define HCI_MAX_PAGES 3 struct hci_dev { @@ -361,7 +352,6 @@ struct hci_dev { unsigned long flags; __u16 id; __u8 bus; - __u8 dev_type; bdaddr_t bdaddr; bdaddr_t setup_addr; bdaddr_t public_addr; @@ -467,21 +457,6 @@ struct hci_dev { __u16 sniff_min_interval; __u16 sniff_max_interval; - __u8 amp_status; - __u32 amp_total_bw; - __u32 amp_max_bw; - __u32 amp_min_latency; - __u32 amp_max_pdu; - __u8 amp_type; - __u16 amp_pal_cap; - __u16 amp_assoc_size; - __u32 amp_max_flush_to; - __u32 amp_be_flush_to; - - struct amp_assoc loc_assoc; - - __u8 flow_ctl_mode; - unsigned int auto_accept_delay; unsigned long quirks; @@ -501,11 +476,6 @@ struct hci_dev { unsigned int le_pkts; unsigned int iso_pkts; - __u16 block_len; - __u16 block_mtu; - __u16 num_blocks; - __u16 block_cnt; - unsigned long acl_last_tx; unsigned long sco_last_tx; unsigned long le_last_tx; @@ -778,7 +748,6 @@ struct hci_conn { void *l2cap_data; void *sco_data; void *iso_data; - struct amp_mgr *amp_mgr; struct list_head link_list; struct hci_conn *parent; @@ -805,7 +774,6 @@ struct hci_chan { struct sk_buff_head data_q; unsigned int sent; __u8 state; - bool amp; }; struct hci_conn_params { @@ -1014,9 +982,6 @@ static inline void hci_conn_hash_add(struct hci_dev *hdev, struct hci_conn *c) case ACL_LINK: h->acl_num++; break; - case AMP_LINK: - h->amp_num++; - break; case LE_LINK: h->le_num++; if (c->role == HCI_ROLE_SLAVE) @@ -1043,9 +1008,6 @@ static inline void hci_conn_hash_del(struct hci_dev *hdev, struct hci_conn *c) case ACL_LINK: h->acl_num--; break; - case AMP_LINK: - h->amp_num--; - break; case LE_LINK: h->le_num--; if (c->role == HCI_ROLE_SLAVE) @@ -1067,8 +1029,6 @@ static inline unsigned int hci_conn_num(struct hci_dev *hdev, __u8 type) switch (type) { case ACL_LINK: return h->acl_num; - case AMP_LINK: - return h->amp_num; case LE_LINK: return h->le_num; case SCO_LINK: @@ -1085,7 +1045,7 @@ static inline unsigned int hci_conn_count(struct hci_dev *hdev) { struct hci_conn_hash *c = &hdev->conn_hash; - return c->acl_num + c->amp_num + c->sco_num + c->le_num + c->iso_num; + return c->acl_num + c->sco_num + c->le_num + c->iso_num; } static inline bool hci_conn_valid(struct hci_dev *hdev, struct hci_conn *conn) @@ -1610,10 +1570,6 @@ static inline void hci_conn_drop(struct hci_conn *conn) } break; - case AMP_LINK: - timeo = conn->disc_timeout; - break; - default: timeo = 0; break; diff --git a/include/uapi/linux/virtio_bt.h b/include/uapi/linux/virtio_bt.h index af798f4c9680..3cc7d633456b 100644 --- a/include/uapi/linux/virtio_bt.h +++ b/include/uapi/linux/virtio_bt.h @@ -13,7 +13,6 @@ enum virtio_bt_config_type { VIRTIO_BT_CONFIG_TYPE_PRIMARY = 0, - VIRTIO_BT_CONFIG_TYPE_AMP = 1, }; enum virtio_bt_config_vendor { diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index f21e7c73021d..0c76dcde5361 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -1168,8 +1168,7 @@ struct hci_dev *hci_get_route(bdaddr_t *dst, bdaddr_t *src, uint8_t src_type) list_for_each_entry(d, &hci_dev_list, list) { if (!test_bit(HCI_UP, &d->flags) || - hci_dev_test_flag(d, HCI_USER_CHANNEL) || - d->dev_type != HCI_PRIMARY) + hci_dev_test_flag(d, HCI_USER_CHANNEL)) continue; /* Simple routing: diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index b52125512985..7bc431af7b04 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -402,11 +402,6 @@ int hci_inquiry(void __user *arg) goto done; } - if (hdev->dev_type != HCI_PRIMARY) { - err = -EOPNOTSUPP; - goto done; - } - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { err = -EOPNOTSUPP; goto done; @@ -759,11 +754,6 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) goto done; } - if (hdev->dev_type != HCI_PRIMARY) { - err = -EOPNOTSUPP; - goto done; - } - if (!hci_dev_test_flag(hdev, HCI_BREDR_ENABLED)) { err = -EOPNOTSUPP; goto done; @@ -917,7 +907,7 @@ int hci_get_dev_info(void __user *arg) strscpy(di.name, hdev->name, sizeof(di.name)); di.bdaddr = hdev->bdaddr; - di.type = (hdev->bus & 0x0f) | ((hdev->dev_type & 0x03) << 4); + di.type = (hdev->bus & 0x0f); di.flags = flags; di.pkt_type = hdev->pkt_type; if (lmp_bredr_capable(hdev)) { @@ -1033,8 +1023,7 @@ static void hci_power_on(struct work_struct *work) */ if (hci_dev_test_flag(hdev, HCI_RFKILLED) || hci_dev_test_flag(hdev, HCI_UNCONFIGURED) || - (hdev->dev_type == HCI_PRIMARY && - !bacmp(&hdev->bdaddr, BDADDR_ANY) && + (!bacmp(&hdev->bdaddr, BDADDR_ANY) && !bacmp(&hdev->static_addr, BDADDR_ANY))) { hci_dev_clear_flag(hdev, HCI_AUTO_OFF); hci_dev_do_close(hdev); @@ -2642,21 +2631,7 @@ int hci_register_dev(struct hci_dev *hdev) if (!hdev->open || !hdev->close || !hdev->send) return -EINVAL; - /* Do not allow HCI_AMP devices to register at index 0, - * so the index can be used as the AMP controller ID. - */ - switch (hdev->dev_type) { - case HCI_PRIMARY: - id = ida_alloc_max(&hci_index_ida, HCI_MAX_ID - 1, GFP_KERNEL); - break; - case HCI_AMP: - id = ida_alloc_range(&hci_index_ida, 1, HCI_MAX_ID - 1, - GFP_KERNEL); - break; - default: - return -EINVAL; - } - + id = ida_alloc_max(&hci_index_ida, HCI_MAX_ID - 1, GFP_KERNEL); if (id < 0) return id; @@ -2708,12 +2683,10 @@ int hci_register_dev(struct hci_dev *hdev) hci_dev_set_flag(hdev, HCI_SETUP); hci_dev_set_flag(hdev, HCI_AUTO_OFF); - if (hdev->dev_type == HCI_PRIMARY) { - /* Assume BR/EDR support until proven otherwise (such as - * through reading supported features during init. - */ - hci_dev_set_flag(hdev, HCI_BREDR_ENABLED); - } + /* Assume BR/EDR support until proven otherwise (such as + * through reading supported features during init. + */ + hci_dev_set_flag(hdev, HCI_BREDR_ENABLED); write_lock(&hci_dev_list_lock); list_add(&hdev->list, &hci_dev_list); @@ -3249,17 +3222,7 @@ static void hci_queue_acl(struct hci_chan *chan, struct sk_buff_head *queue, hci_skb_pkt_type(skb) = HCI_ACLDATA_PKT; - switch (hdev->dev_type) { - case HCI_PRIMARY: - hci_add_acl_hdr(skb, conn->handle, flags); - break; - case HCI_AMP: - hci_add_acl_hdr(skb, chan->handle, flags); - break; - default: - bt_dev_err(hdev, "unknown dev_type %d", hdev->dev_type); - return; - } + hci_add_acl_hdr(skb, conn->handle, flags); list = skb_shinfo(skb)->frag_list; if (!list) { @@ -3419,9 +3382,6 @@ static inline void hci_quote_sent(struct hci_conn *conn, int num, int *quote) case ACL_LINK: cnt = hdev->acl_cnt; break; - case AMP_LINK: - cnt = hdev->block_cnt; - break; case SCO_LINK: case ESCO_LINK: cnt = hdev->sco_cnt; @@ -3619,12 +3579,6 @@ static void hci_prio_recalculate(struct hci_dev *hdev, __u8 type) } -static inline int __get_blocks(struct hci_dev *hdev, struct sk_buff *skb) -{ - /* Calculate count of blocks used by this packet */ - return DIV_ROUND_UP(skb->len - HCI_ACL_HDR_SIZE, hdev->block_len); -} - static void __check_timeout(struct hci_dev *hdev, unsigned int cnt, u8 type) { unsigned long last_tx; @@ -3738,81 +3692,15 @@ static void hci_sched_acl_pkt(struct hci_dev *hdev) hci_prio_recalculate(hdev, ACL_LINK); } -static void hci_sched_acl_blk(struct hci_dev *hdev) -{ - unsigned int cnt = hdev->block_cnt; - struct hci_chan *chan; - struct sk_buff *skb; - int quote; - u8 type; - - BT_DBG("%s", hdev->name); - - if (hdev->dev_type == HCI_AMP) - type = AMP_LINK; - else - type = ACL_LINK; - - __check_timeout(hdev, cnt, type); - - while (hdev->block_cnt > 0 && - (chan = hci_chan_sent(hdev, type, "e))) { - u32 priority = (skb_peek(&chan->data_q))->priority; - while (quote > 0 && (skb = skb_peek(&chan->data_q))) { - int blocks; - - BT_DBG("chan %p skb %p len %d priority %u", chan, skb, - skb->len, skb->priority); - - /* Stop if priority has changed */ - if (skb->priority < priority) - break; - - skb = skb_dequeue(&chan->data_q); - - blocks = __get_blocks(hdev, skb); - if (blocks > hdev->block_cnt) - return; - - hci_conn_enter_active_mode(chan->conn, - bt_cb(skb)->force_active); - - hci_send_frame(hdev, skb); - hdev->acl_last_tx = jiffies; - - hdev->block_cnt -= blocks; - quote -= blocks; - - chan->sent += blocks; - chan->conn->sent += blocks; - } - } - - if (cnt != hdev->block_cnt) - hci_prio_recalculate(hdev, type); -} - static void hci_sched_acl(struct hci_dev *hdev) { BT_DBG("%s", hdev->name); /* No ACL link over BR/EDR controller */ - if (!hci_conn_num(hdev, ACL_LINK) && hdev->dev_type == HCI_PRIMARY) - return; - - /* No AMP link over AMP controller */ - if (!hci_conn_num(hdev, AMP_LINK) && hdev->dev_type == HCI_AMP) + if (!hci_conn_num(hdev, ACL_LINK)) return; - switch (hdev->flow_ctl_mode) { - case HCI_FLOW_CTL_MODE_PACKET_BASED: - hci_sched_acl_pkt(hdev); - break; - - case HCI_FLOW_CTL_MODE_BLOCK_BASED: - hci_sched_acl_blk(hdev); - break; - } + hci_sched_acl_pkt(hdev); } static void hci_sched_le(struct hci_dev *hdev) diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 97c6f34ba640..a487f9df8145 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -913,21 +913,6 @@ static u8 hci_cc_read_local_ext_features(struct hci_dev *hdev, void *data, return rp->status; } -static u8 hci_cc_read_flow_control_mode(struct hci_dev *hdev, void *data, - struct sk_buff *skb) -{ - struct hci_rp_read_flow_control_mode *rp = data; - - bt_dev_dbg(hdev, "status 0x%2.2x", rp->status); - - if (rp->status) - return rp->status; - - hdev->flow_ctl_mode = rp->mode; - - return rp->status; -} - static u8 hci_cc_read_buffer_size(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -1071,28 +1056,6 @@ static u8 hci_cc_write_page_scan_type(struct hci_dev *hdev, void *data, return rp->status; } -static u8 hci_cc_read_data_block_size(struct hci_dev *hdev, void *data, - struct sk_buff *skb) -{ - struct hci_rp_read_data_block_size *rp = data; - - bt_dev_dbg(hdev, "status 0x%2.2x", rp->status); - - if (rp->status) - return rp->status; - - hdev->block_mtu = __le16_to_cpu(rp->max_acl_len); - hdev->block_len = __le16_to_cpu(rp->block_len); - hdev->num_blocks = __le16_to_cpu(rp->num_blocks); - - hdev->block_cnt = hdev->num_blocks; - - BT_DBG("%s blk mtu %d cnt %d len %d", hdev->name, hdev->block_mtu, - hdev->block_cnt, hdev->block_len); - - return rp->status; -} - static u8 hci_cc_read_clock(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -1127,30 +1090,6 @@ unlock: return rp->status; } -static u8 hci_cc_read_local_amp_info(struct hci_dev *hdev, void *data, - struct sk_buff *skb) -{ - struct hci_rp_read_local_amp_info *rp = data; - - bt_dev_dbg(hdev, "status 0x%2.2x", rp->status); - - if (rp->status) - return rp->status; - - hdev->amp_status = rp->amp_status; - hdev->amp_total_bw = __le32_to_cpu(rp->total_bw); - hdev->amp_max_bw = __le32_to_cpu(rp->max_bw); - hdev->amp_min_latency = __le32_to_cpu(rp->min_latency); - hdev->amp_max_pdu = __le32_to_cpu(rp->max_pdu); - hdev->amp_type = rp->amp_type; - hdev->amp_pal_cap = __le16_to_cpu(rp->pal_cap); - hdev->amp_assoc_size = __le16_to_cpu(rp->max_assoc_size); - hdev->amp_be_flush_to = __le32_to_cpu(rp->be_flush_to); - hdev->amp_max_flush_to = __le32_to_cpu(rp->max_flush_to); - - return rp->status; -} - static u8 hci_cc_read_inq_rsp_tx_power(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -4120,12 +4059,6 @@ static const struct hci_cc { HCI_CC(HCI_OP_READ_PAGE_SCAN_TYPE, hci_cc_read_page_scan_type, sizeof(struct hci_rp_read_page_scan_type)), HCI_CC_STATUS(HCI_OP_WRITE_PAGE_SCAN_TYPE, hci_cc_write_page_scan_type), - HCI_CC(HCI_OP_READ_DATA_BLOCK_SIZE, hci_cc_read_data_block_size, - sizeof(struct hci_rp_read_data_block_size)), - HCI_CC(HCI_OP_READ_FLOW_CONTROL_MODE, hci_cc_read_flow_control_mode, - sizeof(struct hci_rp_read_flow_control_mode)), - HCI_CC(HCI_OP_READ_LOCAL_AMP_INFO, hci_cc_read_local_amp_info, - sizeof(struct hci_rp_read_local_amp_info)), HCI_CC(HCI_OP_READ_CLOCK, hci_cc_read_clock, sizeof(struct hci_rp_read_clock)), HCI_CC(HCI_OP_READ_ENC_KEY_SIZE, hci_cc_read_enc_key_size, @@ -4461,11 +4394,6 @@ static void hci_num_comp_pkts_evt(struct hci_dev *hdev, void *data, flex_array_size(ev, handles, ev->num))) return; - if (hdev->flow_ctl_mode != HCI_FLOW_CTL_MODE_PACKET_BASED) { - bt_dev_err(hdev, "wrong event for mode %d", hdev->flow_ctl_mode); - return; - } - bt_dev_dbg(hdev, "num %d", ev->num); for (i = 0; i < ev->num; i++) { @@ -4533,78 +4461,6 @@ static void hci_num_comp_pkts_evt(struct hci_dev *hdev, void *data, queue_work(hdev->workqueue, &hdev->tx_work); } -static struct hci_conn *__hci_conn_lookup_handle(struct hci_dev *hdev, - __u16 handle) -{ - struct hci_chan *chan; - - switch (hdev->dev_type) { - case HCI_PRIMARY: - return hci_conn_hash_lookup_handle(hdev, handle); - case HCI_AMP: - chan = hci_chan_lookup_handle(hdev, handle); - if (chan) - return chan->conn; - break; - default: - bt_dev_err(hdev, "unknown dev_type %d", hdev->dev_type); - break; - } - - return NULL; -} - -static void hci_num_comp_blocks_evt(struct hci_dev *hdev, void *data, - struct sk_buff *skb) -{ - struct hci_ev_num_comp_blocks *ev = data; - int i; - - if (!hci_ev_skb_pull(hdev, skb, HCI_EV_NUM_COMP_BLOCKS, - flex_array_size(ev, handles, ev->num_hndl))) - return; - - if (hdev->flow_ctl_mode != HCI_FLOW_CTL_MODE_BLOCK_BASED) { - bt_dev_err(hdev, "wrong event for mode %d", - hdev->flow_ctl_mode); - return; - } - - bt_dev_dbg(hdev, "num_blocks %d num_hndl %d", ev->num_blocks, - ev->num_hndl); - - for (i = 0; i < ev->num_hndl; i++) { - struct hci_comp_blocks_info *info = &ev->handles[i]; - struct hci_conn *conn = NULL; - __u16 handle, block_count; - - handle = __le16_to_cpu(info->handle); - block_count = __le16_to_cpu(info->blocks); - - conn = __hci_conn_lookup_handle(hdev, handle); - if (!conn) - continue; - - conn->sent -= block_count; - - switch (conn->type) { - case ACL_LINK: - case AMP_LINK: - hdev->block_cnt += block_count; - if (hdev->block_cnt > hdev->num_blocks) - hdev->block_cnt = hdev->num_blocks; - break; - - default: - bt_dev_err(hdev, "unknown type %d conn %p", - conn->type, conn); - break; - } - } - - queue_work(hdev->workqueue, &hdev->tx_work); -} - static void hci_mode_change_evt(struct hci_dev *hdev, void *data, struct sk_buff *skb) { @@ -7520,9 +7376,6 @@ static const struct hci_ev { /* [0x3e = HCI_EV_LE_META] */ HCI_EV_REQ_VL(HCI_EV_LE_META, hci_le_meta_evt, sizeof(struct hci_ev_le_meta), HCI_MAX_EVENT_SIZE), - /* [0x48 = HCI_EV_NUM_COMP_BLOCKS] */ - HCI_EV(HCI_EV_NUM_COMP_BLOCKS, hci_num_comp_blocks_evt, - sizeof(struct hci_ev_num_comp_blocks)), /* [0xff = HCI_EV_VENDOR] */ HCI_EV_VL(HCI_EV_VENDOR, msft_vendor_evt, 0, HCI_MAX_EVENT_SIZE), }; diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 703b84bd48d5..69c2ba1e843e 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -485,7 +485,7 @@ static struct sk_buff *create_monitor_event(struct hci_dev *hdev, int event) return NULL; ni = skb_put(skb, HCI_MON_NEW_INDEX_SIZE); - ni->type = hdev->dev_type; + ni->type = 0x00; /* Old hdev->dev_type */ ni->bus = hdev->bus; bacpy(&ni->bdaddr, &hdev->bdaddr); memcpy_and_pad(ni->name, sizeof(ni->name), hdev->name, @@ -1007,9 +1007,6 @@ static int hci_sock_bound_ioctl(struct sock *sk, unsigned int cmd, if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) return -EOPNOTSUPP; - if (hdev->dev_type != HCI_PRIMARY) - return -EOPNOTSUPP; - switch (cmd) { case HCISETRAW: if (!capable(CAP_NET_ADMIN)) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 813c11d61bca..891cae8a30da 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -3511,10 +3511,6 @@ static int hci_unconf_init_sync(struct hci_dev *hdev) /* Read Local Supported Features. */ static int hci_read_local_features_sync(struct hci_dev *hdev) { - /* Not all AMP controllers support this command */ - if (hdev->dev_type == HCI_AMP && !(hdev->commands[14] & 0x20)) - return 0; - return __hci_cmd_sync_status(hdev, HCI_OP_READ_LOCAL_FEATURES, 0, NULL, HCI_CMD_TIMEOUT); } @@ -3549,51 +3545,6 @@ static int hci_read_local_cmds_sync(struct hci_dev *hdev) return 0; } -/* Read Local AMP Info */ -static int hci_read_local_amp_info_sync(struct hci_dev *hdev) -{ - return __hci_cmd_sync_status(hdev, HCI_OP_READ_LOCAL_AMP_INFO, - 0, NULL, HCI_CMD_TIMEOUT); -} - -/* Read Data Blk size */ -static int hci_read_data_block_size_sync(struct hci_dev *hdev) -{ - return __hci_cmd_sync_status(hdev, HCI_OP_READ_DATA_BLOCK_SIZE, - 0, NULL, HCI_CMD_TIMEOUT); -} - -/* Read Flow Control Mode */ -static int hci_read_flow_control_mode_sync(struct hci_dev *hdev) -{ - return __hci_cmd_sync_status(hdev, HCI_OP_READ_FLOW_CONTROL_MODE, - 0, NULL, HCI_CMD_TIMEOUT); -} - -/* Read Location Data */ -static int hci_read_location_data_sync(struct hci_dev *hdev) -{ - return __hci_cmd_sync_status(hdev, HCI_OP_READ_LOCATION_DATA, - 0, NULL, HCI_CMD_TIMEOUT); -} - -/* AMP Controller init stage 1 command sequence */ -static const struct hci_init_stage amp_init1[] = { - /* HCI_OP_READ_LOCAL_VERSION */ - HCI_INIT(hci_read_local_version_sync), - /* HCI_OP_READ_LOCAL_COMMANDS */ - HCI_INIT(hci_read_local_cmds_sync), - /* HCI_OP_READ_LOCAL_AMP_INFO */ - HCI_INIT(hci_read_local_amp_info_sync), - /* HCI_OP_READ_DATA_BLOCK_SIZE */ - HCI_INIT(hci_read_data_block_size_sync), - /* HCI_OP_READ_FLOW_CONTROL_MODE */ - HCI_INIT(hci_read_flow_control_mode_sync), - /* HCI_OP_READ_LOCATION_DATA */ - HCI_INIT(hci_read_location_data_sync), - {} -}; - static int hci_init1_sync(struct hci_dev *hdev) { int err; @@ -3607,28 +3558,9 @@ static int hci_init1_sync(struct hci_dev *hdev) return err; } - switch (hdev->dev_type) { - case HCI_PRIMARY: - hdev->flow_ctl_mode = HCI_FLOW_CTL_MODE_PACKET_BASED; - return hci_init_stage_sync(hdev, br_init1); - case HCI_AMP: - hdev->flow_ctl_mode = HCI_FLOW_CTL_MODE_BLOCK_BASED; - return hci_init_stage_sync(hdev, amp_init1); - default: - bt_dev_err(hdev, "Unknown device type %d", hdev->dev_type); - break; - } - - return 0; + return hci_init_stage_sync(hdev, br_init1); } -/* AMP Controller init stage 2 command sequence */ -static const struct hci_init_stage amp_init2[] = { - /* HCI_OP_READ_LOCAL_FEATURES */ - HCI_INIT(hci_read_local_features_sync), - {} -}; - /* Read Buffer Size (ACL mtu, max pkt, etc.) */ static int hci_read_buffer_size_sync(struct hci_dev *hdev) { @@ -3886,9 +3818,6 @@ static int hci_init2_sync(struct hci_dev *hdev) bt_dev_dbg(hdev, ""); - if (hdev->dev_type == HCI_AMP) - return hci_init_stage_sync(hdev, amp_init2); - err = hci_init_stage_sync(hdev, hci_init2); if (err) return err; @@ -4716,13 +4645,6 @@ static int hci_init_sync(struct hci_dev *hdev) if (err < 0) return err; - /* HCI_PRIMARY covers both single-mode LE, BR/EDR and dual-mode - * BR/EDR/LE type controllers. AMP controllers only need the - * first two stages of init. - */ - if (hdev->dev_type != HCI_PRIMARY) - return 0; - err = hci_init3_sync(hdev); if (err < 0) return err; @@ -4951,12 +4873,8 @@ int hci_dev_open_sync(struct hci_dev *hdev) * In case of user channel usage, it is not important * if a public address or static random address is * available. - * - * This check is only valid for BR/EDR controllers - * since AMP controllers do not have an address. */ if (!hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && - hdev->dev_type == HCI_PRIMARY && !bacmp(&hdev->bdaddr, BDADDR_ANY) && !bacmp(&hdev->static_addr, BDADDR_ANY)) { ret = -EADDRNOTAVAIL; @@ -4991,8 +4909,7 @@ int hci_dev_open_sync(struct hci_dev *hdev) !hci_dev_test_flag(hdev, HCI_CONFIG) && !hci_dev_test_flag(hdev, HCI_UNCONFIGURED) && !hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && - hci_dev_test_flag(hdev, HCI_MGMT) && - hdev->dev_type == HCI_PRIMARY) { + hci_dev_test_flag(hdev, HCI_MGMT)) { ret = hci_powered_update_sync(hdev); mgmt_power_on(hdev, ret); } @@ -5137,8 +5054,7 @@ int hci_dev_close_sync(struct hci_dev *hdev) auto_off = hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF); - if (!auto_off && hdev->dev_type == HCI_PRIMARY && - !hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && + if (!auto_off && !hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && hci_dev_test_flag(hdev, HCI_MGMT)) __mgmt_power_off(hdev); @@ -5200,9 +5116,6 @@ int hci_dev_close_sync(struct hci_dev *hdev) hdev->flags &= BIT(HCI_RAW); hci_dev_clear_volatile_flags(hdev); - /* Controller radio is available but is currently powered down */ - hdev->amp_status = AMP_STATUS_POWERED_DOWN; - memset(hdev->eir, 0, sizeof(hdev->eir)); memset(hdev->dev_class, 0, sizeof(hdev->dev_class)); bacpy(&hdev->random_addr, BDADDR_ANY); @@ -5239,8 +5152,7 @@ static int hci_power_on_sync(struct hci_dev *hdev) */ if (hci_dev_test_flag(hdev, HCI_RFKILLED) || hci_dev_test_flag(hdev, HCI_UNCONFIGURED) || - (hdev->dev_type == HCI_PRIMARY && - !bacmp(&hdev->bdaddr, BDADDR_ANY) && + (!bacmp(&hdev->bdaddr, BDADDR_ANY) && !bacmp(&hdev->static_addr, BDADDR_ANY))) { hci_dev_clear_flag(hdev, HCI_AUTO_OFF); hci_dev_close_sync(hdev); @@ -5342,27 +5254,11 @@ int hci_stop_discovery_sync(struct hci_dev *hdev) return 0; } -static int hci_disconnect_phy_link_sync(struct hci_dev *hdev, u16 handle, - u8 reason) -{ - struct hci_cp_disconn_phy_link cp; - - memset(&cp, 0, sizeof(cp)); - cp.phy_handle = HCI_PHY_HANDLE(handle); - cp.reason = reason; - - return __hci_cmd_sync_status(hdev, HCI_OP_DISCONN_PHY_LINK, - sizeof(cp), &cp, HCI_CMD_TIMEOUT); -} - static int hci_disconnect_sync(struct hci_dev *hdev, struct hci_conn *conn, u8 reason) { struct hci_cp_disconnect cp; - if (conn->type == AMP_LINK) - return hci_disconnect_phy_link_sync(hdev, conn->handle, reason); - if (test_bit(HCI_CONN_BIG_CREATED, &conn->flags)) { /* This is a BIS connection, hci_conn_del will * do the necessary cleanup. diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index adc8f990f13e..5b509b767557 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -3932,7 +3932,7 @@ static inline int l2cap_command_rej(struct l2cap_conn *conn, } static void l2cap_connect(struct l2cap_conn *conn, struct l2cap_cmd_hdr *cmd, - u8 *data, u8 rsp_code, u8 amp_id) + u8 *data, u8 rsp_code) { struct l2cap_conn_req *req = (struct l2cap_conn_req *) data; struct l2cap_conn_rsp rsp; @@ -4011,17 +4011,8 @@ static void l2cap_connect(struct l2cap_conn *conn, struct l2cap_cmd_hdr *cmd, status = L2CAP_CS_AUTHOR_PEND; chan->ops->defer(chan); } else { - /* Force pending result for AMP controllers. - * The connection will succeed after the - * physical link is up. - */ - if (amp_id == AMP_ID_BREDR) { - l2cap_state_change(chan, BT_CONFIG); - result = L2CAP_CR_SUCCESS; - } else { - l2cap_state_change(chan, BT_CONNECT2); - result = L2CAP_CR_PEND; - } + l2cap_state_change(chan, BT_CONNECT2); + result = L2CAP_CR_PEND; status = L2CAP_CS_NO_INFO; } } else { @@ -4086,7 +4077,7 @@ static int l2cap_connect_req(struct l2cap_conn *conn, mgmt_device_connected(hdev, hcon, NULL, 0); hci_dev_unlock(hdev); - l2cap_connect(conn, cmd, data, L2CAP_CONN_RSP, 0); + l2cap_connect(conn, cmd, data, L2CAP_CONN_RSP); return 0; } @@ -7491,10 +7482,6 @@ void l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 flags) struct l2cap_conn *conn = hcon->l2cap_data; int len; - /* For AMP controller do not create l2cap conn */ - if (!conn && hcon->hdev->dev_type != HCI_PRIMARY) - goto drop; - if (!conn) conn = l2cap_conn_add(hcon); diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 965f621ef865..80f220b7e19d 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -443,8 +443,7 @@ static int read_index_list(struct sock *sk, struct hci_dev *hdev, void *data, count = 0; list_for_each_entry(d, &hci_dev_list, list) { - if (d->dev_type == HCI_PRIMARY && - !hci_dev_test_flag(d, HCI_UNCONFIGURED)) + if (!hci_dev_test_flag(d, HCI_UNCONFIGURED)) count++; } @@ -468,8 +467,7 @@ static int read_index_list(struct sock *sk, struct hci_dev *hdev, void *data, if (test_bit(HCI_QUIRK_RAW_DEVICE, &d->quirks)) continue; - if (d->dev_type == HCI_PRIMARY && - !hci_dev_test_flag(d, HCI_UNCONFIGURED)) { + if (!hci_dev_test_flag(d, HCI_UNCONFIGURED)) { rp->index[count++] = cpu_to_le16(d->id); bt_dev_dbg(hdev, "Added hci%u", d->id); } @@ -503,8 +501,7 @@ static int read_unconf_index_list(struct sock *sk, struct hci_dev *hdev, count = 0; list_for_each_entry(d, &hci_dev_list, list) { - if (d->dev_type == HCI_PRIMARY && - hci_dev_test_flag(d, HCI_UNCONFIGURED)) + if (hci_dev_test_flag(d, HCI_UNCONFIGURED)) count++; } @@ -528,8 +525,7 @@ static int read_unconf_index_list(struct sock *sk, struct hci_dev *hdev, if (test_bit(HCI_QUIRK_RAW_DEVICE, &d->quirks)) continue; - if (d->dev_type == HCI_PRIMARY && - hci_dev_test_flag(d, HCI_UNCONFIGURED)) { + if (hci_dev_test_flag(d, HCI_UNCONFIGURED)) { rp->index[count++] = cpu_to_le16(d->id); bt_dev_dbg(hdev, "Added hci%u", d->id); } @@ -561,10 +557,8 @@ static int read_ext_index_list(struct sock *sk, struct hci_dev *hdev, read_lock(&hci_dev_list_lock); count = 0; - list_for_each_entry(d, &hci_dev_list, list) { - if (d->dev_type == HCI_PRIMARY || d->dev_type == HCI_AMP) - count++; - } + list_for_each_entry(d, &hci_dev_list, list) + count++; rp = kmalloc(struct_size(rp, entry, count), GFP_ATOMIC); if (!rp) { @@ -585,16 +579,10 @@ static int read_ext_index_list(struct sock *sk, struct hci_dev *hdev, if (test_bit(HCI_QUIRK_RAW_DEVICE, &d->quirks)) continue; - if (d->dev_type == HCI_PRIMARY) { - if (hci_dev_test_flag(d, HCI_UNCONFIGURED)) - rp->entry[count].type = 0x01; - else - rp->entry[count].type = 0x00; - } else if (d->dev_type == HCI_AMP) { - rp->entry[count].type = 0x02; - } else { - continue; - } + if (hci_dev_test_flag(d, HCI_UNCONFIGURED)) + rp->entry[count].type = 0x01; + else + rp->entry[count].type = 0x00; rp->entry[count].bus = d->bus; rp->entry[count++].index = cpu_to_le16(d->id); @@ -9331,23 +9319,14 @@ void mgmt_index_added(struct hci_dev *hdev) if (test_bit(HCI_QUIRK_RAW_DEVICE, &hdev->quirks)) return; - switch (hdev->dev_type) { - case HCI_PRIMARY: - if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { - mgmt_index_event(MGMT_EV_UNCONF_INDEX_ADDED, hdev, - NULL, 0, HCI_MGMT_UNCONF_INDEX_EVENTS); - ev.type = 0x01; - } else { - mgmt_index_event(MGMT_EV_INDEX_ADDED, hdev, NULL, 0, - HCI_MGMT_INDEX_EVENTS); - ev.type = 0x00; - } - break; - case HCI_AMP: - ev.type = 0x02; - break; - default: - return; + if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { + mgmt_index_event(MGMT_EV_UNCONF_INDEX_ADDED, hdev, NULL, 0, + HCI_MGMT_UNCONF_INDEX_EVENTS); + ev.type = 0x01; + } else { + mgmt_index_event(MGMT_EV_INDEX_ADDED, hdev, NULL, 0, + HCI_MGMT_INDEX_EVENTS); + ev.type = 0x00; } ev.bus = hdev->bus; @@ -9364,25 +9343,16 @@ void mgmt_index_removed(struct hci_dev *hdev) if (test_bit(HCI_QUIRK_RAW_DEVICE, &hdev->quirks)) return; - switch (hdev->dev_type) { - case HCI_PRIMARY: - mgmt_pending_foreach(0, hdev, cmd_complete_rsp, &status); + mgmt_pending_foreach(0, hdev, cmd_complete_rsp, &status); - if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { - mgmt_index_event(MGMT_EV_UNCONF_INDEX_REMOVED, hdev, - NULL, 0, HCI_MGMT_UNCONF_INDEX_EVENTS); - ev.type = 0x01; - } else { - mgmt_index_event(MGMT_EV_INDEX_REMOVED, hdev, NULL, 0, - HCI_MGMT_INDEX_EVENTS); - ev.type = 0x00; - } - break; - case HCI_AMP: - ev.type = 0x02; - break; - default: - return; + if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { + mgmt_index_event(MGMT_EV_UNCONF_INDEX_REMOVED, hdev, NULL, 0, + HCI_MGMT_UNCONF_INDEX_EVENTS); + ev.type = 0x01; + } else { + mgmt_index_event(MGMT_EV_INDEX_REMOVED, hdev, NULL, 0, + HCI_MGMT_INDEX_EVENTS); + ev.type = 0x00; } ev.bus = hdev->bus; -- cgit v1.2.3 From e77f43d531af41e9ce299eab10dcae8fa5dbc293 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 13 May 2024 16:07:55 -0400 Subject: Bluetooth: hci_core: Fix not handling hdev->le_num_of_adv_sets=1 If hdev->le_num_of_adv_sets is set to 1 it means that only handle 0x00 can be used, but since the MGMT interface instances start from 1 (instance 0 means all instances in case of MGMT_OP_REMOVE_ADVERTISING) the code needs to map the instance to handle otherwise users will not be able to advertise as instance 1 would attempt to use handle 0x01. Fixes: 1d0fac2c38ed ("Bluetooth: Use controller sets when available") Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_core.c | 9 +++++++++ net/bluetooth/hci_sync.c | 17 ++++++++--------- 3 files changed, 18 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 431d0c8a0e27..9231396fe96f 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -246,6 +246,7 @@ struct adv_info { bool periodic; __u8 mesh; __u8 instance; + __u8 handle; __u32 flags; __u16 timeout; __u16 remaining_time; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 7bc431af7b04..dd3b0f501018 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -1765,6 +1765,15 @@ struct adv_info *hci_add_adv_instance(struct hci_dev *hdev, u8 instance, adv->pending = true; adv->instance = instance; + + /* If controller support only one set and the instance is set to + * 1 then there is no option other than using handle 0x00. + */ + if (hdev->le_num_of_adv_sets == 1 && instance == 1) + adv->handle = 0x00; + else + adv->handle = instance; + list_add(&adv->list, &hdev->adv_instances); hdev->adv_instance_cnt++; } diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 891cae8a30da..16daa79b7981 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -1043,11 +1043,10 @@ static int hci_disable_ext_adv_instance_sync(struct hci_dev *hdev, u8 instance) struct hci_cp_ext_adv_set *set; u8 data[sizeof(*cp) + sizeof(*set) * 1]; u8 size; + struct adv_info *adv = NULL; /* If request specifies an instance that doesn't exist, fail */ if (instance > 0) { - struct adv_info *adv; - adv = hci_find_adv_instance(hdev, instance); if (!adv) return -EINVAL; @@ -1066,7 +1065,7 @@ static int hci_disable_ext_adv_instance_sync(struct hci_dev *hdev, u8 instance) cp->num_of_sets = !!instance; cp->enable = 0x00; - set->handle = instance; + set->handle = adv ? adv->handle : instance; size = sizeof(*cp) + sizeof(*set) * cp->num_of_sets; @@ -1249,7 +1248,7 @@ static int hci_set_ext_scan_rsp_data_sync(struct hci_dev *hdev, u8 instance) len = eir_create_scan_rsp(hdev, instance, pdu->data); - pdu->handle = instance; + pdu->handle = adv ? adv->handle : instance; pdu->length = len; pdu->operation = LE_SET_ADV_DATA_OP_COMPLETE; pdu->frag_pref = LE_SET_ADV_DATA_NO_FRAG; @@ -1331,7 +1330,7 @@ int hci_enable_ext_advertising_sync(struct hci_dev *hdev, u8 instance) memset(set, 0, sizeof(*set)); - set->handle = instance; + set->handle = adv ? adv->handle : instance; /* Set duration per instance since controller is responsible for * scheduling it. @@ -1410,10 +1409,10 @@ static int hci_set_per_adv_data_sync(struct hci_dev *hdev, u8 instance) DEFINE_FLEX(struct hci_cp_le_set_per_adv_data, pdu, data, length, HCI_MAX_PER_AD_LENGTH); u8 len; + struct adv_info *adv = NULL; if (instance) { - struct adv_info *adv = hci_find_adv_instance(hdev, instance); - + adv = hci_find_adv_instance(hdev, instance); if (!adv || !adv->periodic) return 0; } @@ -1421,7 +1420,7 @@ static int hci_set_per_adv_data_sync(struct hci_dev *hdev, u8 instance) len = eir_create_per_adv_data(hdev, instance, pdu->data); pdu->length = len; - pdu->handle = instance; + pdu->handle = adv ? adv->handle : instance; pdu->operation = LE_SET_ADV_DATA_OP_COMPLETE; return __hci_cmd_sync_status(hdev, HCI_OP_LE_SET_PER_ADV_DATA, @@ -1734,7 +1733,7 @@ static int hci_set_ext_adv_data_sync(struct hci_dev *hdev, u8 instance) len = eir_create_adv_data(hdev, instance, pdu->data); pdu->length = len; - pdu->handle = instance; + pdu->handle = adv ? adv->handle : instance; pdu->operation = LE_SET_ADV_DATA_OP_COMPLETE; pdu->frag_pref = LE_SET_ADV_DATA_NO_FRAG; -- cgit v1.2.3 From 4452e8ef8c364113495f414d7e6846d74d7eff81 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 6 May 2024 21:44:49 +0200 Subject: s390/iucv: Provide iucv_alloc_device() / iucv_release_device() Provide iucv_alloc_device() and iucv_release_device() helper functions, which can be used to deduplicate more or less identical IUCV device allocation and release code in four different drivers. Suggested-by: Arnd Bergmann Acked-by: Alexandra Winter Link: https://lore.kernel.org/r/20240506194454.1160315-2-hca@linux.ibm.com Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev --- include/net/iucv/iucv.h | 6 ++++++ net/iucv/iucv.c | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 41 insertions(+) (limited to 'net') diff --git a/include/net/iucv/iucv.h b/include/net/iucv/iucv.h index 5cd7871127c9..b3736e66fe1a 100644 --- a/include/net/iucv/iucv.h +++ b/include/net/iucv/iucv.h @@ -84,6 +84,12 @@ struct iucv_array { extern const struct bus_type iucv_bus; extern struct device *iucv_root; +struct device_driver; + +struct device *iucv_alloc_device(const struct attribute_group **attrs, + struct device_driver *driver, void *priv, + const char *fmt, ...) __printf(4, 5); + /* * struct iucv_path * pathid: 16 bit path identification diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index a4ab615ca3e3..9db7c2c0ae72 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -76,6 +76,41 @@ EXPORT_SYMBOL(iucv_bus); struct device *iucv_root; EXPORT_SYMBOL(iucv_root); +static void iucv_release_device(struct device *device) +{ + kfree(device); +} + +struct device *iucv_alloc_device(const struct attribute_group **attrs, + struct device_driver *driver, + void *priv, const char *fmt, ...) +{ + struct device *dev; + va_list vargs; + int rc; + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) + goto out_error; + va_start(vargs, fmt); + rc = dev_set_name(dev, fmt, vargs); + va_end(vargs); + if (rc) + goto out_error; + dev->bus = &iucv_bus; + dev->parent = iucv_root; + dev->driver = driver; + dev->groups = attrs; + dev->release = iucv_release_device; + dev_set_drvdata(dev, priv); + return dev; + +out_error: + kfree(dev); + return NULL; +} +EXPORT_SYMBOL(iucv_alloc_device); + static int iucv_available; /* General IUCV interrupt structure */ -- cgit v1.2.3 From effb83572685eaa70d05a8dd6307ca574a11fcf3 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 6 May 2024 21:44:54 +0200 Subject: s390/iucv: Unexport iucv_root There is no user of iucv_root outside of the core IUCV code left. Therefore remove the EXPORT_SYMBOL. Acked-by: Alexandra Winter Link: https://lore.kernel.org/r/20240506194454.1160315-7-hca@linux.ibm.com Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev --- include/net/iucv/iucv.h | 1 - net/iucv/iucv.c | 3 +-- 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'net') diff --git a/include/net/iucv/iucv.h b/include/net/iucv/iucv.h index b3736e66fe1a..4d114e6d6d23 100644 --- a/include/net/iucv/iucv.h +++ b/include/net/iucv/iucv.h @@ -82,7 +82,6 @@ struct iucv_array { } __attribute__ ((aligned (8))); extern const struct bus_type iucv_bus; -extern struct device *iucv_root; struct device_driver; diff --git a/net/iucv/iucv.c b/net/iucv/iucv.c index 9db7c2c0ae72..2e61f19621ee 100644 --- a/net/iucv/iucv.c +++ b/net/iucv/iucv.c @@ -73,8 +73,7 @@ const struct bus_type iucv_bus = { }; EXPORT_SYMBOL(iucv_bus); -struct device *iucv_root; -EXPORT_SYMBOL(iucv_root); +static struct device *iucv_root; static void iucv_release_device(struct device *device) { -- cgit v1.2.3 From 8bd67ebb50c0145fd2ca8681ab65eb7e8cde1afc Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Mon, 13 May 2024 13:34:19 +0300 Subject: net: bridge: xmit: make sure we have at least eth header len bytes syzbot triggered an uninit value[1] error in bridge device's xmit path by sending a short (less than ETH_HLEN bytes) skb. To fix it check if we can actually pull that amount instead of assuming. Tested with dropwatch: drop at: br_dev_xmit+0xb93/0x12d0 [bridge] (0xffffffffc06739b3) origin: software timestamp: Mon May 13 11:31:53 2024 778214037 nsec protocol: 0x88a8 length: 2 original length: 2 drop reason: PKT_TOO_SMALL [1] BUG: KMSAN: uninit-value in br_dev_xmit+0x61d/0x1cb0 net/bridge/br_device.c:65 br_dev_xmit+0x61d/0x1cb0 net/bridge/br_device.c:65 __netdev_start_xmit include/linux/netdevice.h:4903 [inline] netdev_start_xmit include/linux/netdevice.h:4917 [inline] xmit_one net/core/dev.c:3531 [inline] dev_hard_start_xmit+0x247/0xa20 net/core/dev.c:3547 __dev_queue_xmit+0x34db/0x5350 net/core/dev.c:4341 dev_queue_xmit include/linux/netdevice.h:3091 [inline] __bpf_tx_skb net/core/filter.c:2136 [inline] __bpf_redirect_common net/core/filter.c:2180 [inline] __bpf_redirect+0x14a6/0x1620 net/core/filter.c:2187 ____bpf_clone_redirect net/core/filter.c:2460 [inline] bpf_clone_redirect+0x328/0x470 net/core/filter.c:2432 ___bpf_prog_run+0x13fe/0xe0f0 kernel/bpf/core.c:1997 __bpf_prog_run512+0xb5/0xe0 kernel/bpf/core.c:2238 bpf_dispatcher_nop_func include/linux/bpf.h:1234 [inline] __bpf_prog_run include/linux/filter.h:657 [inline] bpf_prog_run include/linux/filter.h:664 [inline] bpf_test_run+0x499/0xc30 net/bpf/test_run.c:425 bpf_prog_test_run_skb+0x14ea/0x1f20 net/bpf/test_run.c:1058 bpf_prog_test_run+0x6b7/0xad0 kernel/bpf/syscall.c:4269 __sys_bpf+0x6aa/0xd90 kernel/bpf/syscall.c:5678 __do_sys_bpf kernel/bpf/syscall.c:5767 [inline] __se_sys_bpf kernel/bpf/syscall.c:5765 [inline] __x64_sys_bpf+0xa0/0xe0 kernel/bpf/syscall.c:5765 x64_sys_call+0x96b/0x3b50 arch/x86/include/generated/asm/syscalls_64.h:322 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot+a63a1f6a062033cf0f40@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=a63a1f6a062033cf0f40 Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- net/bridge/br_device.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index c366ccc8b3db..ecac7886988b 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -27,6 +27,7 @@ EXPORT_SYMBOL_GPL(nf_br_ops); /* net device transmit always called with BH disabled */ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) { + enum skb_drop_reason reason = pskb_may_pull_reason(skb, ETH_HLEN); struct net_bridge_mcast_port *pmctx_null = NULL; struct net_bridge *br = netdev_priv(dev); struct net_bridge_mcast *brmctx = &br->multicast_ctx; @@ -38,6 +39,11 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) const unsigned char *dest; u16 vid = 0; + if (unlikely(reason != SKB_NOT_DROPPED_YET)) { + kfree_skb_reason(skb, reason); + return NETDEV_TX_OK; + } + memset(skb->cb, 0, sizeof(struct br_input_skb_cb)); br_tc_skb_miss_set(skb, false); -- cgit v1.2.3 From 3a7c1661ae1383364cd6092d851f5e5da64d476b Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Mon, 13 May 2024 14:06:27 +0300 Subject: net: bridge: mst: fix vlan use-after-free syzbot reported a suspicious rcu usage[1] in bridge's mst code. While fixing it I noticed that nothing prevents a vlan to be freed while walking the list from the same path (br forward delay timer). Fix the rcu usage and also make sure we are not accessing freed memory by making br_mst_vlan_set_state use rcu read lock. [1] WARNING: suspicious RCU usage 6.9.0-rc6-syzkaller #0 Not tainted ----------------------------- net/bridge/br_private.h:1599 suspicious rcu_dereference_protected() usage! ... stack backtrace: CPU: 1 PID: 8017 Comm: syz-executor.1 Not tainted 6.9.0-rc6-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 lockdep_rcu_suspicious+0x221/0x340 kernel/locking/lockdep.c:6712 nbp_vlan_group net/bridge/br_private.h:1599 [inline] br_mst_set_state+0x1ea/0x650 net/bridge/br_mst.c:105 br_set_state+0x28a/0x7b0 net/bridge/br_stp.c:47 br_forward_delay_timer_expired+0x176/0x440 net/bridge/br_stp_timer.c:88 call_timer_fn+0x18e/0x650 kernel/time/timer.c:1793 expire_timers kernel/time/timer.c:1844 [inline] __run_timers kernel/time/timer.c:2418 [inline] __run_timer_base+0x66a/0x8e0 kernel/time/timer.c:2429 run_timer_base kernel/time/timer.c:2438 [inline] run_timer_softirq+0xb7/0x170 kernel/time/timer.c:2448 __do_softirq+0x2c6/0x980 kernel/softirq.c:554 invoke_softirq kernel/softirq.c:428 [inline] __irq_exit_rcu+0xf2/0x1c0 kernel/softirq.c:633 irq_exit_rcu+0x9/0x30 kernel/softirq.c:645 instr_sysvec_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1043 [inline] sysvec_apic_timer_interrupt+0xa6/0xc0 arch/x86/kernel/apic/apic.c:1043 asm_sysvec_apic_timer_interrupt+0x1a/0x20 arch/x86/include/asm/idtentry.h:702 RIP: 0010:lock_acquire+0x264/0x550 kernel/locking/lockdep.c:5758 Code: 2b 00 74 08 4c 89 f7 e8 ba d1 84 00 f6 44 24 61 02 0f 85 85 01 00 00 41 f7 c7 00 02 00 00 74 01 fb 48 c7 44 24 40 0e 36 e0 45 <4b> c7 44 25 00 00 00 00 00 43 c7 44 25 09 00 00 00 00 43 c7 44 25 RSP: 0018:ffffc90013657100 EFLAGS: 00000206 RAX: 0000000000000001 RBX: 1ffff920026cae2c RCX: 0000000000000001 RDX: dffffc0000000000 RSI: ffffffff8bcaca00 RDI: ffffffff8c1eaa60 RBP: ffffc90013657260 R08: ffffffff92efe507 R09: 1ffffffff25dfca0 R10: dffffc0000000000 R11: fffffbfff25dfca1 R12: 1ffff920026cae28 R13: dffffc0000000000 R14: ffffc90013657160 R15: 0000000000000246 Fixes: ec7328b59176 ("net: bridge: mst: Multiple Spanning Tree (MST) mode") Reported-by: syzbot+fa04eb8a56fd923fc5d8@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=fa04eb8a56fd923fc5d8 Signed-off-by: Nikolay Aleksandrov Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/bridge/br_mst.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/bridge/br_mst.c b/net/bridge/br_mst.c index ee680adcee17..3c66141d34d6 100644 --- a/net/bridge/br_mst.c +++ b/net/bridge/br_mst.c @@ -78,7 +78,7 @@ static void br_mst_vlan_set_state(struct net_bridge_port *p, struct net_bridge_v { struct net_bridge_vlan_group *vg = nbp_vlan_group(p); - if (v->state == state) + if (br_vlan_get_state(v) == state) return; br_vlan_set_state(v, state); @@ -100,11 +100,12 @@ int br_mst_set_state(struct net_bridge_port *p, u16 msti, u8 state, }; struct net_bridge_vlan_group *vg; struct net_bridge_vlan *v; - int err; + int err = 0; + rcu_read_lock(); vg = nbp_vlan_group(p); if (!vg) - return 0; + goto out; /* MSTI 0 (CST) state changes are notified via the regular * SWITCHDEV_ATTR_ID_PORT_STP_STATE. @@ -112,17 +113,20 @@ int br_mst_set_state(struct net_bridge_port *p, u16 msti, u8 state, if (msti) { err = switchdev_port_attr_set(p->dev, &attr, extack); if (err && err != -EOPNOTSUPP) - return err; + goto out; } - list_for_each_entry(v, &vg->vlan_list, vlist) { + err = 0; + list_for_each_entry_rcu(v, &vg->vlan_list, vlist) { if (v->brvlan->msti != msti) continue; br_mst_vlan_set_state(p, v, state); } - return 0; +out: + rcu_read_unlock(); + return err; } static void br_mst_vlan_sync_state(struct net_bridge_vlan *pv, u16 msti) -- cgit v1.2.3 From fd76e5ccc48f9f54eb44909dd7c0b924005f1582 Mon Sep 17 00:00:00 2001 From: Chris Lew Date: Mon, 13 May 2024 10:31:46 -0700 Subject: net: qrtr: ns: Fix module refcnt The qrtr protocol core logic and the qrtr nameservice are combined into a single module. Neither the core logic or nameservice provide much functionality by themselves; combining the two into a single module also prevents any possible issues that may stem from client modules loading inbetween qrtr and the ns. Creating a socket takes two references to the module that owns the socket protocol. Since the ns needs to create the control socket, this creates a scenario where there are always two references to the qrtr module. This prevents the execution of 'rmmod' for qrtr. To resolve this, forcefully put the module refcount for the socket opened by the nameservice. Fixes: a365023a76f2 ("net: qrtr: combine nameservice into main module") Reported-by: Jeffrey Hugo Tested-by: Jeffrey Hugo Signed-off-by: Chris Lew Reviewed-by: Manivannan Sadhasivam Reviewed-by: Jeffrey Hugo Signed-off-by: David S. Miller --- net/qrtr/ns.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) (limited to 'net') diff --git a/net/qrtr/ns.c b/net/qrtr/ns.c index abb0c70ffc8b..654a3cc0d347 100644 --- a/net/qrtr/ns.c +++ b/net/qrtr/ns.c @@ -725,6 +725,24 @@ int qrtr_ns_init(void) if (ret < 0) goto err_wq; + /* As the qrtr ns socket owner and creator is the same module, we have + * to decrease the qrtr module reference count to guarantee that it + * remains zero after the ns socket is created, otherwise, executing + * "rmmod" command is unable to make the qrtr module deleted after the + * qrtr module is inserted successfully. + * + * However, the reference count is increased twice in + * sock_create_kern(): one is to increase the reference count of owner + * of qrtr socket's proto_ops struct; another is to increment the + * reference count of owner of qrtr proto struct. Therefore, we must + * decrement the module reference count twice to ensure that it keeps + * zero after server's listening socket is created. Of course, we + * must bump the module reference count twice as well before the socket + * is closed. + */ + module_put(qrtr_ns.sock->ops->owner); + module_put(qrtr_ns.sock->sk->sk_prot_creator->owner); + return 0; err_wq: @@ -739,6 +757,15 @@ void qrtr_ns_remove(void) { cancel_work_sync(&qrtr_ns.work); destroy_workqueue(qrtr_ns.workqueue); + + /* sock_release() expects the two references that were put during + * qrtr_ns_init(). This function is only called during module remove, + * so try_stop_module() has already set the refcnt to 0. Use + * __module_get() instead of try_module_get() to successfully take two + * references. + */ + __module_get(qrtr_ns.sock->ops->owner); + __module_get(qrtr_ns.sock->sk->sk_prot_creator->owner); sock_release(qrtr_ns.sock); } EXPORT_SYMBOL_GPL(qrtr_ns_remove); -- cgit v1.2.3 From bb487272380d120295e955ad8acfcbb281b57642 Mon Sep 17 00:00:00 2001 From: xu xin Date: Tue, 14 May 2024 20:11:02 +0800 Subject: net/ipv6: Fix route deleting failure when metric equals 0 Problem ========= After commit 67f695134703 ("ipv6: Move setting default metric for routes"), we noticed that the logic of assigning the default value of fc_metirc changed in the ioctl process. That is, when users use ioctl(fd, SIOCADDRT, rt) with a non-zero metric to add a route, then they may fail to delete a route with passing in a metric value of 0 to the kernel by ioctl(fd, SIOCDELRT, rt). But iproute can succeed in deleting it. As a reference, when using iproute tools by netlink to delete routes with a metric parameter equals 0, like the command as follows: ip -6 route del fe80::/64 via fe81::5054:ff:fe11:3451 dev eth0 metric 0 the user can still succeed in deleting the route entry with the smallest metric. Root Reason =========== After commit 67f695134703 ("ipv6: Move setting default metric for routes"), When ioctl() pass in SIOCDELRT with a zero metric, rtmsg_to_fib6_config() will set a defalut value (1024) to cfg->fc_metric in kernel, and in ip6_route_del() and the line 4074 at net/ipv3/route.c, it will check by if (cfg->fc_metric && cfg->fc_metric != rt->fib6_metric) continue; and the condition is true and skip the later procedure (deleting route) because cfg->fc_metric != rt->fib6_metric. But before that commit, cfg->fc_metric is still zero there, so the condition is false and it will do the following procedure (deleting). Solution ======== In order to keep a consistent behaviour across netlink() and ioctl(), we should allow to delete a route with a metric value of 0. So we only do the default setting of fc_metric in route adding. CC: stable@vger.kernel.org # 5.4+ Fixes: 67f695134703 ("ipv6: Move setting default metric for routes") Co-developed-by: Fan Yu Signed-off-by: Fan Yu Signed-off-by: xu xin Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240514201102055dD2Ba45qKbLlUMxu_DTHP@zte.com.cn Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index c43b0616742e..bbc2a0dd9314 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -4445,7 +4445,7 @@ static void rtmsg_to_fib6_config(struct net *net, .fc_table = l3mdev_fib_table_by_index(net, rtmsg->rtmsg_ifindex) ? : RT6_TABLE_MAIN, .fc_ifindex = rtmsg->rtmsg_ifindex, - .fc_metric = rtmsg->rtmsg_metric ? : IP6_RT_PRIO_USER, + .fc_metric = rtmsg->rtmsg_metric, .fc_expires = rtmsg->rtmsg_info, .fc_dst_len = rtmsg->rtmsg_dst_len, .fc_src_len = rtmsg->rtmsg_src_len, @@ -4475,6 +4475,9 @@ int ipv6_route_ioctl(struct net *net, unsigned int cmd, struct in6_rtmsg *rtmsg) rtnl_lock(); switch (cmd) { case SIOCADDRT: + /* Only do the default setting of fc_metric in route adding */ + if (cfg.fc_metric == 0) + cfg.fc_metric = IP6_RT_PRIO_USER; err = ip6_route_add(&cfg, GFP_KERNEL, NULL); break; case SIOCDELRT: -- cgit v1.2.3 From e03e7f20ebf7e1611d40d1fdc1bde900fd3335f6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 15 May 2024 14:29:34 +0000 Subject: netrom: fix possible dead-lock in nr_rt_ioctl() syzbot loves netrom, and found a possible deadlock in nr_rt_ioctl [1] Make sure we always acquire nr_node_list_lock before nr_node_lock(nr_node) [1] WARNING: possible circular locking dependency detected 6.9.0-rc7-syzkaller-02147-g654de42f3fc6 #0 Not tainted ------------------------------------------------------ syz-executor350/5129 is trying to acquire lock: ffff8880186e2070 (&nr_node->node_lock){+...}-{2:2}, at: spin_lock_bh include/linux/spinlock.h:356 [inline] ffff8880186e2070 (&nr_node->node_lock){+...}-{2:2}, at: nr_node_lock include/net/netrom.h:152 [inline] ffff8880186e2070 (&nr_node->node_lock){+...}-{2:2}, at: nr_dec_obs net/netrom/nr_route.c:464 [inline] ffff8880186e2070 (&nr_node->node_lock){+...}-{2:2}, at: nr_rt_ioctl+0x1bb/0x1090 net/netrom/nr_route.c:697 but task is already holding lock: ffffffff8f7053b8 (nr_node_list_lock){+...}-{2:2}, at: spin_lock_bh include/linux/spinlock.h:356 [inline] ffffffff8f7053b8 (nr_node_list_lock){+...}-{2:2}, at: nr_dec_obs net/netrom/nr_route.c:462 [inline] ffffffff8f7053b8 (nr_node_list_lock){+...}-{2:2}, at: nr_rt_ioctl+0x10a/0x1090 net/netrom/nr_route.c:697 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (nr_node_list_lock){+...}-{2:2}: lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754 __raw_spin_lock_bh include/linux/spinlock_api_smp.h:126 [inline] _raw_spin_lock_bh+0x35/0x50 kernel/locking/spinlock.c:178 spin_lock_bh include/linux/spinlock.h:356 [inline] nr_remove_node net/netrom/nr_route.c:299 [inline] nr_del_node+0x4b4/0x820 net/netrom/nr_route.c:355 nr_rt_ioctl+0xa95/0x1090 net/netrom/nr_route.c:683 sock_do_ioctl+0x158/0x460 net/socket.c:1222 sock_ioctl+0x629/0x8e0 net/socket.c:1341 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:904 [inline] __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:890 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f -> #0 (&nr_node->node_lock){+...}-{2:2}: check_prev_add kernel/locking/lockdep.c:3134 [inline] check_prevs_add kernel/locking/lockdep.c:3253 [inline] validate_chain+0x18cb/0x58e0 kernel/locking/lockdep.c:3869 __lock_acquire+0x1346/0x1fd0 kernel/locking/lockdep.c:5137 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754 __raw_spin_lock_bh include/linux/spinlock_api_smp.h:126 [inline] _raw_spin_lock_bh+0x35/0x50 kernel/locking/spinlock.c:178 spin_lock_bh include/linux/spinlock.h:356 [inline] nr_node_lock include/net/netrom.h:152 [inline] nr_dec_obs net/netrom/nr_route.c:464 [inline] nr_rt_ioctl+0x1bb/0x1090 net/netrom/nr_route.c:697 sock_do_ioctl+0x158/0x460 net/socket.c:1222 sock_ioctl+0x629/0x8e0 net/socket.c:1341 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:904 [inline] __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:890 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(nr_node_list_lock); lock(&nr_node->node_lock); lock(nr_node_list_lock); lock(&nr_node->node_lock); *** DEADLOCK *** 1 lock held by syz-executor350/5129: #0: ffffffff8f7053b8 (nr_node_list_lock){+...}-{2:2}, at: spin_lock_bh include/linux/spinlock.h:356 [inline] #0: ffffffff8f7053b8 (nr_node_list_lock){+...}-{2:2}, at: nr_dec_obs net/netrom/nr_route.c:462 [inline] #0: ffffffff8f7053b8 (nr_node_list_lock){+...}-{2:2}, at: nr_rt_ioctl+0x10a/0x1090 net/netrom/nr_route.c:697 stack backtrace: CPU: 0 PID: 5129 Comm: syz-executor350 Not tainted 6.9.0-rc7-syzkaller-02147-g654de42f3fc6 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/02/2024 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114 check_noncircular+0x36a/0x4a0 kernel/locking/lockdep.c:2187 check_prev_add kernel/locking/lockdep.c:3134 [inline] check_prevs_add kernel/locking/lockdep.c:3253 [inline] validate_chain+0x18cb/0x58e0 kernel/locking/lockdep.c:3869 __lock_acquire+0x1346/0x1fd0 kernel/locking/lockdep.c:5137 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754 __raw_spin_lock_bh include/linux/spinlock_api_smp.h:126 [inline] _raw_spin_lock_bh+0x35/0x50 kernel/locking/spinlock.c:178 spin_lock_bh include/linux/spinlock.h:356 [inline] nr_node_lock include/net/netrom.h:152 [inline] nr_dec_obs net/netrom/nr_route.c:464 [inline] nr_rt_ioctl+0x1bb/0x1090 net/netrom/nr_route.c:697 sock_do_ioctl+0x158/0x460 net/socket.c:1222 sock_ioctl+0x629/0x8e0 net/socket.c:1341 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:904 [inline] __se_sys_ioctl+0xfc/0x170 fs/ioctl.c:890 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot Signed-off-by: Eric Dumazet Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240515142934.3708038-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/netrom/nr_route.c | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) (limited to 'net') diff --git a/net/netrom/nr_route.c b/net/netrom/nr_route.c index 70480869ad1c..bd2b17b219ae 100644 --- a/net/netrom/nr_route.c +++ b/net/netrom/nr_route.c @@ -285,22 +285,14 @@ static int __must_check nr_add_node(ax25_address *nr, const char *mnemonic, return 0; } -static inline void __nr_remove_node(struct nr_node *nr_node) +static void nr_remove_node_locked(struct nr_node *nr_node) { + lockdep_assert_held(&nr_node_list_lock); + hlist_del_init(&nr_node->node_node); nr_node_put(nr_node); } -#define nr_remove_node_locked(__node) \ - __nr_remove_node(__node) - -static void nr_remove_node(struct nr_node *nr_node) -{ - spin_lock_bh(&nr_node_list_lock); - __nr_remove_node(nr_node); - spin_unlock_bh(&nr_node_list_lock); -} - static inline void __nr_remove_neigh(struct nr_neigh *nr_neigh) { hlist_del_init(&nr_neigh->neigh_node); @@ -339,6 +331,7 @@ static int nr_del_node(ax25_address *callsign, ax25_address *neighbour, struct n return -EINVAL; } + spin_lock_bh(&nr_node_list_lock); nr_node_lock(nr_node); for (i = 0; i < nr_node->count; i++) { if (nr_node->routes[i].neighbour == nr_neigh) { @@ -352,7 +345,7 @@ static int nr_del_node(ax25_address *callsign, ax25_address *neighbour, struct n nr_node->count--; if (nr_node->count == 0) { - nr_remove_node(nr_node); + nr_remove_node_locked(nr_node); } else { switch (i) { case 0: @@ -367,12 +360,14 @@ static int nr_del_node(ax25_address *callsign, ax25_address *neighbour, struct n nr_node_put(nr_node); } nr_node_unlock(nr_node); + spin_unlock_bh(&nr_node_list_lock); return 0; } } nr_neigh_put(nr_neigh); nr_node_unlock(nr_node); + spin_unlock_bh(&nr_node_list_lock); nr_node_put(nr_node); return -EINVAL; -- cgit v1.2.3 From 581073f626e387d3e7eed55c48c8495584ead7ba Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 15 May 2024 16:33:58 +0000 Subject: af_packet: do not call packet_read_pending() from tpacket_destruct_skb() trafgen performance considerably sank on hosts with many cores after the blamed commit. packet_read_pending() is very expensive, and calling it in af_packet fast path defeats Daniel intent in commit b013840810c2 ("packet: use percpu mmap tx frame pending refcount") tpacket_destruct_skb() makes room for one packet, we can immediately wakeup a producer, no need to completely drain the tx ring. Fixes: 89ed5b519004 ("af_packet: Block execution of tasks waiting for transmit to complete in AF_PACKET") Signed-off-by: Eric Dumazet Cc: Neil Horman Cc: Daniel Borkmann Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/20240515163358.4105915-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/packet/af_packet.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 8c6d3fbb4ed8..ea3ebc160e25 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -2522,8 +2522,7 @@ static void tpacket_destruct_skb(struct sk_buff *skb) ts = __packet_set_timestamp(po, ph, skb); __packet_set_status(po, ph, TP_STATUS_AVAILABLE | ts); - if (!packet_read_pending(&po->tx_ring)) - complete(&po->skb_completion); + complete(&po->skb_completion); } sock_wfree(skb); -- cgit v1.2.3 From 6e828dc60e509b79ef09882264952f341cb58425 Mon Sep 17 00:00:00 2001 From: Tom Parkin Date: Mon, 13 May 2024 18:22:47 +0100 Subject: l2tp: fix ICMP error handling for UDP-encap sockets Since commit a36e185e8c85 ("udp: Handle ICMP errors for tunnels with same destination port on both endpoints") UDP's handling of ICMP errors has allowed for UDP-encap tunnels to determine socket associations in scenarios where the UDP hash lookup could not. Subsequently, commit d26796ae58940 ("udp: check udp sock encap_type in __udp_lib_err") subtly tweaked the approach such that UDP ICMP error handling would be skipped for any UDP socket which has encapsulation enabled. In the case of L2TP tunnel sockets using UDP-encap, this latter modification effectively broke ICMP error reporting for the L2TP control plane. To a degree this isn't catastrophic inasmuch as the L2TP control protocol defines a reliable transport on top of the underlying packet switching network which will eventually detect errors and time out. However, paying attention to the ICMP error reporting allows for more timely detection of errors in L2TP userspace, and aids in debugging connectivity issues. Reinstate ICMP error handling for UDP encap L2TP tunnels: * implement struct udp_tunnel_sock_cfg .encap_err_rcv in order to allow the L2TP code to handle ICMP errors; * only implement error-handling for tunnels which have a managed socket: unmanaged tunnels using a kernel socket have no userspace to report errors back to; * flag the error on the socket, which allows for userspace to get an error such as -ECONNREFUSED back from sendmsg/recvmsg; * pass the error into ip[v6]_icmp_error() which allows for userspace to get extended error information via. MSG_ERRQUEUE. Fixes: d26796ae5894 ("udp: check udp sock encap_type in __udp_lib_err") Signed-off-by: Tom Parkin Link: https://lore.kernel.org/r/20240513172248.623261-1-tparkin@katalix.com Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_core.c | 44 +++++++++++++++++++++++++++++++++----------- 1 file changed, 33 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 7d519a46a844..88a34db265d8 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -910,22 +910,20 @@ pass: return 1; } -/* UDP encapsulation receive handler. See net/ipv4/udp.c. - * Return codes: - * 0 : success. - * <0: error - * >0: skb should be passed up to userspace as UDP. +/* UDP encapsulation receive and error receive handlers. + * See net/ipv4/udp.c for details. + * + * Note that these functions are called from inside an + * RCU-protected region, but without the socket being locked. + * + * Hence we use rcu_dereference_sk_user_data to access the + * tunnel data structure rather the usual l2tp_sk_to_tunnel + * accessor function. */ int l2tp_udp_encap_recv(struct sock *sk, struct sk_buff *skb) { struct l2tp_tunnel *tunnel; - /* Note that this is called from the encap_rcv hook inside an - * RCU-protected region, but without the socket being locked. - * Hence we use rcu_dereference_sk_user_data to access the - * tunnel data structure rather the usual l2tp_sk_to_tunnel - * accessor function. - */ tunnel = rcu_dereference_sk_user_data(sk); if (!tunnel) goto pass_up; @@ -942,6 +940,29 @@ pass_up: } EXPORT_SYMBOL_GPL(l2tp_udp_encap_recv); +static void l2tp_udp_encap_err_recv(struct sock *sk, struct sk_buff *skb, int err, + __be16 port, u32 info, u8 *payload) +{ + struct l2tp_tunnel *tunnel; + + tunnel = rcu_dereference_sk_user_data(sk); + if (!tunnel || tunnel->fd < 0) + return; + + sk->sk_err = err; + sk_error_report(sk); + + if (ip_hdr(skb)->version == IPVERSION) { + if (inet_test_bit(RECVERR, sk)) + return ip_icmp_error(sk, skb, err, port, info, payload); +#if IS_ENABLED(CONFIG_IPV6) + } else { + if (inet6_test_bit(RECVERR6, sk)) + return ipv6_icmp_error(sk, skb, err, port, info, payload); +#endif + } +} + /************************************************************************ * Transmit handling ***********************************************************************/ @@ -1516,6 +1537,7 @@ int l2tp_tunnel_register(struct l2tp_tunnel *tunnel, struct net *net, .sk_user_data = tunnel, .encap_type = UDP_ENCAP_L2TPINUDP, .encap_rcv = l2tp_udp_encap_recv, + .encap_err_rcv = l2tp_udp_encap_err_recv, .encap_destroy = l2tp_udp_encap_destroy, }; -- cgit v1.2.3 From 5447f9708d9e4c17a647b16a9cb29e9e02820bd9 Mon Sep 17 00:00:00 2001 From: Andrea Mayer Date: Fri, 17 May 2024 18:45:41 +0200 Subject: ipv6: sr: fix missing sk_buff release in seg6_input_core The seg6_input() function is responsible for adding the SRH into a packet, delegating the operation to the seg6_input_core(). This function uses the skb_cow_head() to ensure that there is sufficient headroom in the sk_buff for accommodating the link-layer header. In the event that the skb_cow_header() function fails, the seg6_input_core() catches the error but it does not release the sk_buff, which will result in a memory leak. This issue was introduced in commit af3b5158b89d ("ipv6: sr: fix BUG due to headroom too small after SRH push") and persists even after commit 7a3f5b0de364 ("netfilter: add netfilter hooks to SRv6 data plane"), where the entire seg6_input() code was refactored to deal with netfilter hooks. The proposed patch addresses the identified memory leak by requiring the seg6_input_core() function to release the sk_buff in the event that skb_cow_head() fails. Fixes: af3b5158b89d ("ipv6: sr: fix BUG due to headroom too small after SRH push") Signed-off-by: Andrea Mayer Reviewed-by: Simon Horman Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/seg6_iptunnel.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'net') diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c index 03b877ff4558..a75df2ec8db0 100644 --- a/net/ipv6/seg6_iptunnel.c +++ b/net/ipv6/seg6_iptunnel.c @@ -459,10 +459,8 @@ static int seg6_input_core(struct net *net, struct sock *sk, int err; err = seg6_do_srh(skb); - if (unlikely(err)) { - kfree_skb(skb); - return err; - } + if (unlikely(err)) + goto drop; slwt = seg6_lwt_lwtunnel(orig_dst->lwtstate); @@ -486,7 +484,7 @@ static int seg6_input_core(struct net *net, struct sock *sk, err = skb_cow_head(skb, LL_RESERVED_SPACE(dst->dev)); if (unlikely(err)) - return err; + goto drop; if (static_branch_unlikely(&nf_hooks_lwtunnel_enabled)) return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, @@ -494,6 +492,9 @@ static int seg6_input_core(struct net *net, struct sock *sk, skb_dst(skb)->dev, seg6_input_finish); return seg6_input_finish(dev_net(skb->dev), NULL, skb); +drop: + kfree_skb(skb); + return err; } static int seg6_input_nf(struct sk_buff *skb) -- cgit v1.2.3 From e4a87abf588536d1cdfb128595e6e680af5cf3ed Mon Sep 17 00:00:00 2001 From: Ryosuke Yasuoka Date: Sun, 19 May 2024 18:43:03 +0900 Subject: nfc: nci: Fix uninit-value in nci_rx_work syzbot reported the following uninit-value access issue [1] nci_rx_work() parses received packet from ndev->rx_q. It should be validated header size, payload size and total packet size before processing the packet. If an invalid packet is detected, it should be silently discarded. Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet") Reported-and-tested-by: syzbot+d7b4dc6cd50410152534@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=d7b4dc6cd50410152534 [1] Signed-off-by: Ryosuke Yasuoka Reviewed-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- net/nfc/nci/core.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c index b133dc55304c..7a9897fbf4f4 100644 --- a/net/nfc/nci/core.c +++ b/net/nfc/nci/core.c @@ -1463,6 +1463,19 @@ int nci_core_ntf_packet(struct nci_dev *ndev, __u16 opcode, ndev->ops->n_core_ops); } +static bool nci_valid_size(struct sk_buff *skb) +{ + BUILD_BUG_ON(NCI_CTRL_HDR_SIZE != NCI_DATA_HDR_SIZE); + unsigned int hdr_size = NCI_CTRL_HDR_SIZE; + + if (skb->len < hdr_size || + !nci_plen(skb->data) || + skb->len < hdr_size + nci_plen(skb->data)) { + return false; + } + return true; +} + /* ---- NCI TX Data worker thread ---- */ static void nci_tx_work(struct work_struct *work) @@ -1516,7 +1529,7 @@ static void nci_rx_work(struct work_struct *work) nfc_send_to_raw_sock(ndev->nfc_dev, skb, RAW_PAYLOAD_NCI, NFC_DIRECTION_RX); - if (!nci_plen(skb->data)) { + if (!nci_valid_size(skb)) { kfree_skb(skb); kcov_remote_stop(); break; -- cgit v1.2.3 From 9b62ef6d239ecccccb0c0262ab39cfcb6d8e40ae Mon Sep 17 00:00:00 2001 From: Olga Kornievskaia Date: Thu, 18 Apr 2024 10:11:04 -0400 Subject: SUNRPC: fix handling expired GSS context MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In the case where we have received a successful reply to an RPC request, but while processing the reply the client in rpc_decode_header() finds an expired context, the code ends up propagating the error to the caller instead of getting a new context and retrying the request. To give more details, in rpc_decode_header() we call rpcauth_checkverf() will call into the gss and internally will at some point call gss_validate() which has a check if the current’s context lifetime expired, and it would fail. The reason for the failure gets ‘scrubbed’ and translated to EACCES so when we get back to rpc_decode_header() we just go to “out_verifier” which for that error would get converted to “out_garbage” (ie it’s treated as garballed reply) and the next action is call_encode. Which (1) doesn’t reencode or re-send (not to mention no upcall happens because context expires as that reason just not known) and it again fails in the same decoding process. After re-trying it 3 times the error is propagated back to the caller (ie nfs4_write_done_cb() in the case a failing write). To fix this, instead we need to look to the case where the server decides that context has expired and replies with an RPC auth error. In that case, the rpc_decode_header() goes to "out_msg_denied" in that we return EKEYREJECTED which in call_decode() is sent to “call_reserve” which triggers an upcalls and a re-try of the operation. The proposed fix is in case of a failed rpc_decode_header() to check if credentials were set to be invalid and use that as a proxy for deciding that context has expired and then treat is same way as receiving an auth error. Signed-off-by: Olga Kornievskaia Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 28f3749f6dc6..f19ad55017c9 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -2698,8 +2698,19 @@ rpc_decode_header(struct rpc_task *task, struct xdr_stream *xdr) goto out_msg_denied; error = rpcauth_checkverf(task, xdr); - if (error) + if (error) { + struct rpc_cred *cred = task->tk_rqstp->rq_cred; + + if (!test_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags)) { + rpcauth_invalcred(task); + if (!task->tk_cred_retry) + goto out_err; + task->tk_cred_retry--; + trace_rpc__stale_creds(task); + return -EKEYREJECTED; + } goto out_verifier; + } p = xdr_inline_decode(xdr, sizeof(*p)); if (!p) -- cgit v1.2.3 From 0dc9f430027b8bd9073fdafdfcdeb1a073ab5594 Mon Sep 17 00:00:00 2001 From: Dan Aloni Date: Thu, 25 Apr 2024 13:49:38 +0300 Subject: sunrpc: fix NFSACL RPC retry on soft mount It used to be quite awhile ago since 1b63a75180c6 ('SUNRPC: Refactor rpc_clone_client()'), in 2012, that `cl_timeout` was copied in so that all mount parameters propagate to NFSACL clients. However since that change, if mount options as follows are given: soft,timeo=50,retrans=16,vers=3 The resultant NFSACL client receives: cl_softrtry: 1 cl_timeout: to_initval=60000, to_maxval=60000, to_increment=0, to_retries=2, to_exponential=0 These values lead to NFSACL operations not being retried under the condition of transient network outages with soft mount. Instead, getacl call fails after 60 seconds with EIO. The simple fix is to pass the existing client's `cl_timeout` as the new client timeout. Cc: Chuck Lever Cc: Benjamin Coddington Link: https://lore.kernel.org/all/20231105154857.ryakhmgaptq3hb6b@gmail.com/T/ Fixes: 1b63a75180c6 ('SUNRPC: Refactor rpc_clone_client()') Signed-off-by: Dan Aloni Reviewed-by: Benjamin Coddington Signed-off-by: Trond Myklebust --- net/sunrpc/clnt.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net') diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index f19ad55017c9..cfd1b1bf7e35 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1071,6 +1071,7 @@ struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *old, .authflavor = old->cl_auth->au_flavor, .cred = old->cl_cred, .stats = old->cl_stats, + .timeout = old->cl_timeout, }; struct rpc_clnt *clnt; int err; -- cgit v1.2.3 From 4836da219781ec510c4c0303df901aa643507a7a Mon Sep 17 00:00:00 2001 From: Dan Aloni Date: Mon, 6 May 2024 12:37:59 +0300 Subject: rpcrdma: fix handling for RDMA_CM_EVENT_DEVICE_REMOVAL Under the scenario of IB device bonding, when bringing down one of the ports, or all ports, we saw xprtrdma entering a non-recoverable state where it is not even possible to complete the disconnect and shut it down the mount, requiring a reboot. Following debug, we saw that transport connect never ended after receiving the RDMA_CM_EVENT_DEVICE_REMOVAL callback. The DEVICE_REMOVAL callback is irrespective of whether the CM_ID is connected, and ESTABLISHED may not have happened. So need to work with each of these states accordingly. Fixes: 2acc5cae2923 ('xprtrdma: Prevent dereferencing r_xprt->rx_ep after it is freed') Cc: Sagi Grimberg Signed-off-by: Dan Aloni Reviewed-by: Sagi Grimberg Reviewed-by: Chuck Lever Signed-off-by: Trond Myklebust --- net/sunrpc/xprtrdma/verbs.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 4f8d7efa469f..432557a553e7 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -244,7 +244,11 @@ rpcrdma_cm_event_handler(struct rdma_cm_id *id, struct rdma_cm_event *event) case RDMA_CM_EVENT_DEVICE_REMOVAL: pr_info("rpcrdma: removing device %s for %pISpc\n", ep->re_id->device->name, sap); - fallthrough; + switch (xchg(&ep->re_connect_status, -ENODEV)) { + case 0: goto wake_connect_worker; + case 1: goto disconnected; + } + return 0; case RDMA_CM_EVENT_ADDR_CHANGE: ep->re_connect_status = -ENODEV; goto disconnected; -- cgit v1.2.3 From 9841991a446c87f90f66f4b9fee6fe934c1336a2 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 16 May 2024 22:48:35 +0900 Subject: af_unix: Update unix_sk(sk)->oob_skb under sk_receive_queue lock. Billy Jheng Bing-Jhong reported a race between __unix_gc() and queue_oob(). __unix_gc() tries to garbage-collect close()d inflight sockets, and then if the socket has MSG_OOB in unix_sk(sk)->oob_skb, GC will drop the reference and set NULL to it locklessly. However, the peer socket still can send MSG_OOB message and queue_oob() can update unix_sk(sk)->oob_skb concurrently, leading NULL pointer dereference. [0] To fix the issue, let's update unix_sk(sk)->oob_skb under the sk_receive_queue's lock and take it everywhere we touch oob_skb. Note that we defer kfree_skb() in manage_oob() to silence lockdep false-positive (See [1]). [0]: BUG: kernel NULL pointer dereference, address: 0000000000000008 PF: supervisor write access in kernel mode PF: error_code(0x0002) - not-present page PGD 8000000009f5e067 P4D 8000000009f5e067 PUD 9f5d067 PMD 0 Oops: 0002 [#1] PREEMPT SMP PTI CPU: 3 PID: 50 Comm: kworker/3:1 Not tainted 6.9.0-rc5-00191-gd091e579b864 #110 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 Workqueue: events delayed_fput RIP: 0010:skb_dequeue (./include/linux/skbuff.h:2386 ./include/linux/skbuff.h:2402 net/core/skbuff.c:3847) Code: 39 e3 74 3e 8b 43 10 48 89 ef 83 e8 01 89 43 10 49 8b 44 24 08 49 c7 44 24 08 00 00 00 00 49 8b 14 24 49 c7 04 24 00 00 00 00 <48> 89 42 08 48 89 10 e8 e7 c5 42 00 4c 89 e0 5b 5d 41 5c c3 cc cc RSP: 0018:ffffc900001bfd48 EFLAGS: 00000002 RAX: 0000000000000000 RBX: ffff8880088f5ae8 RCX: 00000000361289f9 RDX: 0000000000000000 RSI: 0000000000000206 RDI: ffff8880088f5b00 RBP: ffff8880088f5b00 R08: 0000000000080000 R09: 0000000000000001 R10: 0000000000000003 R11: 0000000000000001 R12: ffff8880056b6a00 R13: ffff8880088f5280 R14: 0000000000000001 R15: ffff8880088f5a80 FS: 0000000000000000(0000) GS:ffff88807dd80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000008 CR3: 0000000006314000 CR4: 00000000007506f0 PKRU: 55555554 Call Trace: unix_release_sock (net/unix/af_unix.c:654) unix_release (net/unix/af_unix.c:1050) __sock_release (net/socket.c:660) sock_close (net/socket.c:1423) __fput (fs/file_table.c:423) delayed_fput (fs/file_table.c:444 (discriminator 3)) process_one_work (kernel/workqueue.c:3259) worker_thread (kernel/workqueue.c:3329 kernel/workqueue.c:3416) kthread (kernel/kthread.c:388) ret_from_fork (arch/x86/kernel/process.c:153) ret_from_fork_asm (arch/x86/entry/entry_64.S:257) Modules linked in: CR2: 0000000000000008 Link: https://lore.kernel.org/netdev/a00d3993-c461-43f2-be6d-07259c98509a@rbox.co/ [1] Fixes: 1279f9d9dec2 ("af_unix: Call kfree_skb() for dead unix_(sk)->oob_skb in GC.") Reported-by: Billy Jheng Bing-Jhong Signed-off-by: Kuniyuki Iwashima Link: https://lore.kernel.org/r/20240516134835.8332-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni --- net/unix/af_unix.c | 28 ++++++++++++++++++++++------ 1 file changed, 22 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index fa906ec5e657..ca101690e740 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2171,13 +2171,15 @@ static int queue_oob(struct socket *sock, struct msghdr *msg, struct sock *other maybe_add_creds(skb, sock, other); skb_get(skb); + scm_stat_add(other, skb); + + spin_lock(&other->sk_receive_queue.lock); if (ousk->oob_skb) consume_skb(ousk->oob_skb); - WRITE_ONCE(ousk->oob_skb, skb); + __skb_queue_tail(&other->sk_receive_queue, skb); + spin_unlock(&other->sk_receive_queue.lock); - scm_stat_add(other, skb); - skb_queue_tail(&other->sk_receive_queue, skb); sk_send_sigurg(other); unix_state_unlock(other); other->sk_data_ready(other); @@ -2568,8 +2570,10 @@ static int unix_stream_recv_urg(struct unix_stream_read_state *state) mutex_lock(&u->iolock); unix_state_lock(sk); + spin_lock(&sk->sk_receive_queue.lock); if (sock_flag(sk, SOCK_URGINLINE) || !u->oob_skb) { + spin_unlock(&sk->sk_receive_queue.lock); unix_state_unlock(sk); mutex_unlock(&u->iolock); return -EINVAL; @@ -2581,6 +2585,8 @@ static int unix_stream_recv_urg(struct unix_stream_read_state *state) WRITE_ONCE(u->oob_skb, NULL); else skb_get(oob_skb); + + spin_unlock(&sk->sk_receive_queue.lock); unix_state_unlock(sk); chunk = state->recv_actor(oob_skb, 0, chunk, state); @@ -2609,6 +2615,10 @@ static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk, consume_skb(skb); skb = NULL; } else { + struct sk_buff *unlinked_skb = NULL; + + spin_lock(&sk->sk_receive_queue.lock); + if (skb == u->oob_skb) { if (copied) { skb = NULL; @@ -2620,13 +2630,19 @@ static struct sk_buff *manage_oob(struct sk_buff *skb, struct sock *sk, } else if (flags & MSG_PEEK) { skb = NULL; } else { - skb_unlink(skb, &sk->sk_receive_queue); + __skb_unlink(skb, &sk->sk_receive_queue); WRITE_ONCE(u->oob_skb, NULL); - if (!WARN_ON_ONCE(skb_unref(skb))) - kfree_skb(skb); + unlinked_skb = skb; skb = skb_peek(&sk->sk_receive_queue); } } + + spin_unlock(&sk->sk_receive_queue.lock); + + if (unlinked_skb) { + WARN_ON_ONCE(skb_unref(unlinked_skb)); + kfree_skb(unlinked_skb); + } } return skb; } -- cgit v1.2.3 From efb9f4f19f8e37fde43dfecebc80292d179f56c6 Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Fri, 17 May 2024 08:54:35 +0800 Subject: ipv6: sr: fix memleak in seg6_hmac_init_algo seg6_hmac_init_algo returns without cleaning up the previous allocations if one fails, so it's going to leak all that memory and the crypto tfms. Update seg6_hmac_exit to only free the memory when allocated, so we can reuse the code directly. Fixes: bf355b8d2c30 ("ipv6: sr: add core files for SR HMAC support") Reported-by: Sabrina Dubroca Closes: https://lore.kernel.org/netdev/Zj3bh-gE7eT6V6aH@hog/ Signed-off-by: Hangbin Liu Reviewed-by: Simon Horman Reviewed-by: Sabrina Dubroca Link: https://lore.kernel.org/r/20240517005435.2600277-1-liuhangbin@gmail.com Signed-off-by: Paolo Abeni --- net/ipv6/seg6_hmac.c | 42 ++++++++++++++++++++++++++++-------------- 1 file changed, 28 insertions(+), 14 deletions(-) (limited to 'net') diff --git a/net/ipv6/seg6_hmac.c b/net/ipv6/seg6_hmac.c index 861e0366f549..bbf5b84a70fc 100644 --- a/net/ipv6/seg6_hmac.c +++ b/net/ipv6/seg6_hmac.c @@ -356,6 +356,7 @@ static int seg6_hmac_init_algo(void) struct crypto_shash *tfm; struct shash_desc *shash; int i, alg_count, cpu; + int ret = -ENOMEM; alg_count = ARRAY_SIZE(hmac_algos); @@ -366,12 +367,14 @@ static int seg6_hmac_init_algo(void) algo = &hmac_algos[i]; algo->tfms = alloc_percpu(struct crypto_shash *); if (!algo->tfms) - return -ENOMEM; + goto error_out; for_each_possible_cpu(cpu) { tfm = crypto_alloc_shash(algo->name, 0, 0); - if (IS_ERR(tfm)) - return PTR_ERR(tfm); + if (IS_ERR(tfm)) { + ret = PTR_ERR(tfm); + goto error_out; + } p_tfm = per_cpu_ptr(algo->tfms, cpu); *p_tfm = tfm; } @@ -383,18 +386,22 @@ static int seg6_hmac_init_algo(void) algo->shashs = alloc_percpu(struct shash_desc *); if (!algo->shashs) - return -ENOMEM; + goto error_out; for_each_possible_cpu(cpu) { shash = kzalloc_node(shsize, GFP_KERNEL, cpu_to_node(cpu)); if (!shash) - return -ENOMEM; + goto error_out; *per_cpu_ptr(algo->shashs, cpu) = shash; } } return 0; + +error_out: + seg6_hmac_exit(); + return ret; } int __init seg6_hmac_init(void) @@ -412,22 +419,29 @@ int __net_init seg6_hmac_net_init(struct net *net) void seg6_hmac_exit(void) { struct seg6_hmac_algo *algo = NULL; + struct crypto_shash *tfm; + struct shash_desc *shash; int i, alg_count, cpu; alg_count = ARRAY_SIZE(hmac_algos); for (i = 0; i < alg_count; i++) { algo = &hmac_algos[i]; - for_each_possible_cpu(cpu) { - struct crypto_shash *tfm; - struct shash_desc *shash; - shash = *per_cpu_ptr(algo->shashs, cpu); - kfree(shash); - tfm = *per_cpu_ptr(algo->tfms, cpu); - crypto_free_shash(tfm); + if (algo->shashs) { + for_each_possible_cpu(cpu) { + shash = *per_cpu_ptr(algo->shashs, cpu); + kfree(shash); + } + free_percpu(algo->shashs); + } + + if (algo->tfms) { + for_each_possible_cpu(cpu) { + tfm = *per_cpu_ptr(algo->tfms, cpu); + crypto_free_shash(tfm); + } + free_percpu(algo->tfms); } - free_percpu(algo->tfms); - free_percpu(algo->shashs); } } EXPORT_SYMBOL(seg6_hmac_exit); -- cgit v1.2.3 From 3ebc46ca8675de6378e3f8f40768e180bb8afa66 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 17 May 2024 18:16:26 +0900 Subject: tcp: Fix shift-out-of-bounds in dctcp_update_alpha(). In dctcp_update_alpha(), we use a module parameter dctcp_shift_g as follows: alpha -= min_not_zero(alpha, alpha >> dctcp_shift_g); ... delivered_ce <<= (10 - dctcp_shift_g); It seems syzkaller started fuzzing module parameters and triggered shift-out-of-bounds [0] by setting 100 to dctcp_shift_g: memcpy((void*)0x20000080, "/sys/module/tcp_dctcp/parameters/dctcp_shift_g\000", 47); res = syscall(__NR_openat, /*fd=*/0xffffffffffffff9cul, /*file=*/0x20000080ul, /*flags=*/2ul, /*mode=*/0ul); memcpy((void*)0x20000000, "100\000", 4); syscall(__NR_write, /*fd=*/r[0], /*val=*/0x20000000ul, /*len=*/4ul); Let's limit the max value of dctcp_shift_g by param_set_uint_minmax(). With this patch: # echo 10 > /sys/module/tcp_dctcp/parameters/dctcp_shift_g # cat /sys/module/tcp_dctcp/parameters/dctcp_shift_g 10 # echo 11 > /sys/module/tcp_dctcp/parameters/dctcp_shift_g -bash: echo: write error: Invalid argument [0]: UBSAN: shift-out-of-bounds in net/ipv4/tcp_dctcp.c:143:12 shift exponent 100 is too large for 32-bit type 'u32' (aka 'unsigned int') CPU: 0 PID: 8083 Comm: syz-executor345 Not tainted 6.9.0-05151-g1b294a1f3561 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:88 [inline] dump_stack_lvl+0x201/0x300 lib/dump_stack.c:114 ubsan_epilogue lib/ubsan.c:231 [inline] __ubsan_handle_shift_out_of_bounds+0x346/0x3a0 lib/ubsan.c:468 dctcp_update_alpha+0x540/0x570 net/ipv4/tcp_dctcp.c:143 tcp_in_ack_event net/ipv4/tcp_input.c:3802 [inline] tcp_ack+0x17b1/0x3bc0 net/ipv4/tcp_input.c:3948 tcp_rcv_state_process+0x57a/0x2290 net/ipv4/tcp_input.c:6711 tcp_v4_do_rcv+0x764/0xc40 net/ipv4/tcp_ipv4.c:1937 sk_backlog_rcv include/net/sock.h:1106 [inline] __release_sock+0x20f/0x350 net/core/sock.c:2983 release_sock+0x61/0x1f0 net/core/sock.c:3549 mptcp_subflow_shutdown+0x3d0/0x620 net/mptcp/protocol.c:2907 mptcp_check_send_data_fin+0x225/0x410 net/mptcp/protocol.c:2976 __mptcp_close+0x238/0xad0 net/mptcp/protocol.c:3072 mptcp_close+0x2a/0x1a0 net/mptcp/protocol.c:3127 inet_release+0x190/0x1f0 net/ipv4/af_inet.c:437 __sock_release net/socket.c:659 [inline] sock_close+0xc0/0x240 net/socket.c:1421 __fput+0x41b/0x890 fs/file_table.c:422 task_work_run+0x23b/0x300 kernel/task_work.c:180 exit_task_work include/linux/task_work.h:38 [inline] do_exit+0x9c8/0x2540 kernel/exit.c:878 do_group_exit+0x201/0x2b0 kernel/exit.c:1027 __do_sys_exit_group kernel/exit.c:1038 [inline] __se_sys_exit_group kernel/exit.c:1036 [inline] __x64_sys_exit_group+0x3f/0x40 kernel/exit.c:1036 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xe4/0x240 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x67/0x6f RIP: 0033:0x7f6c2b5005b6 Code: Unable to access opcode bytes at 0x7f6c2b50058c. RSP: 002b:00007ffe883eb948 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7 RAX: ffffffffffffffda RBX: 00007f6c2b5862f0 RCX: 00007f6c2b5005b6 RDX: 0000000000000001 RSI: 000000000000003c RDI: 0000000000000001 RBP: 0000000000000001 R08: 00000000000000e7 R09: ffffffffffffffc0 R10: 0000000000000006 R11: 0000000000000246 R12: 00007f6c2b5862f0 R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000001 Reported-by: syzkaller Reported-by: Yue Sun Reported-by: xingwei lee Closes: https://lore.kernel.org/netdev/CAEkJfYNJM=cw-8x7_Vmj1J6uYVCWMbbvD=EFmDPVBGpTsqOxEA@mail.gmail.com/ Fixes: e3118e8359bb ("net: tcp: add DCTCP congestion control algorithm") Signed-off-by: Kuniyuki Iwashima Reviewed-by: Simon Horman Link: https://lore.kernel.org/r/20240517091626.32772-1-kuniyu@amazon.com Signed-off-by: Paolo Abeni --- net/ipv4/tcp_dctcp.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c index 6b712a33d49f..8a45a4aea933 100644 --- a/net/ipv4/tcp_dctcp.c +++ b/net/ipv4/tcp_dctcp.c @@ -58,7 +58,18 @@ struct dctcp { }; static unsigned int dctcp_shift_g __read_mostly = 4; /* g = 1/2^4 */ -module_param(dctcp_shift_g, uint, 0644); + +static int dctcp_shift_g_set(const char *val, const struct kernel_param *kp) +{ + return param_set_uint_minmax(val, kp, 0, 10); +} + +static const struct kernel_param_ops dctcp_shift_g_ops = { + .set = dctcp_shift_g_set, + .get = param_get_uint, +}; + +module_param_cb(dctcp_shift_g, &dctcp_shift_g_ops, &dctcp_shift_g, 0644); MODULE_PARM_DESC(dctcp_shift_g, "parameter g for updating dctcp_alpha"); static unsigned int dctcp_alpha_on_init __read_mostly = DCTCP_MAX_ALPHA; -- cgit v1.2.3 From 041933a1ec7b4173a8e638cae4f8e394331d7e54 Mon Sep 17 00:00:00 2001 From: Michal Luczaj Date: Fri, 17 May 2024 11:27:01 +0200 Subject: af_unix: Fix garbage collection of embryos carrying OOB with SCM_RIGHTS GC attempts to explicitly drop oob_skb's reference before purging the hit list. The problem is with embryos: kfree_skb(u->oob_skb) is never called on an embryo socket. The python script below [0] sends a listener's fd to its embryo as OOB data. While GC does collect the embryo's queue, it fails to drop the OOB skb's refcount. The skb which was in embryo's receive queue stays as unix_sk(sk)->oob_skb and keeps the listener's refcount [1]. Tell GC to dispose embryo's oob_skb. [0]: from array import array from socket import * addr = '\x00unix-oob' lis = socket(AF_UNIX, SOCK_STREAM) lis.bind(addr) lis.listen(1) s = socket(AF_UNIX, SOCK_STREAM) s.connect(addr) scm = (SOL_SOCKET, SCM_RIGHTS, array('i', [lis.fileno()])) s.sendmsg([b'x'], [scm], MSG_OOB) lis.close() [1] $ grep unix-oob /proc/net/unix $ ./unix-oob.py $ grep unix-oob /proc/net/unix 0000000000000000: 00000002 00000000 00000000 0001 02 0 @unix-oob 0000000000000000: 00000002 00000000 00010000 0001 01 6072 @unix-oob Fixes: 4090fa373f0e ("af_unix: Replace garbage collection algorithm.") Signed-off-by: Michal Luczaj Reviewed-by: Kuniyuki Iwashima Signed-off-by: Paolo Abeni --- net/unix/garbage.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) (limited to 'net') diff --git a/net/unix/garbage.c b/net/unix/garbage.c index 1f8b8cdfcdc8..dfe94a90ece4 100644 --- a/net/unix/garbage.c +++ b/net/unix/garbage.c @@ -342,6 +342,18 @@ enum unix_recv_queue_lock_class { U_RECVQ_LOCK_EMBRYO, }; +static void unix_collect_queue(struct unix_sock *u, struct sk_buff_head *hitlist) +{ + skb_queue_splice_init(&u->sk.sk_receive_queue, hitlist); + +#if IS_ENABLED(CONFIG_AF_UNIX_OOB) + if (u->oob_skb) { + WARN_ON_ONCE(skb_unref(u->oob_skb)); + u->oob_skb = NULL; + } +#endif +} + static void unix_collect_skb(struct list_head *scc, struct sk_buff_head *hitlist) { struct unix_vertex *vertex; @@ -365,18 +377,11 @@ static void unix_collect_skb(struct list_head *scc, struct sk_buff_head *hitlist /* listener -> embryo order, the inversion never happens. */ spin_lock_nested(&embryo_queue->lock, U_RECVQ_LOCK_EMBRYO); - skb_queue_splice_init(embryo_queue, hitlist); + unix_collect_queue(unix_sk(skb->sk), hitlist); spin_unlock(&embryo_queue->lock); } } else { - skb_queue_splice_init(queue, hitlist); - -#if IS_ENABLED(CONFIG_AF_UNIX_OOB) - if (u->oob_skb) { - kfree_skb(u->oob_skb); - u->oob_skb = NULL; - } -#endif + unix_collect_queue(u, hitlist); } spin_unlock(&queue->lock); -- cgit v1.2.3 From 30a92c9e3d6b073932762bef2ac66f4ee784c657 Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Thu, 16 May 2024 16:09:41 -0400 Subject: openvswitch: Set the skbuff pkt_type for proper pmtud support. Open vSwitch is originally intended to switch at layer 2, only dealing with Ethernet frames. With the introduction of l3 tunnels support, it crossed into the realm of needing to care a bit about some routing details when making forwarding decisions. If an oversized packet would need to be fragmented during this forwarding decision, there is a chance for pmtu to get involved and generate a routing exception. This is gated by the skbuff->pkt_type field. When a flow is already loaded into the openvswitch module this field is set up and transitioned properly as a packet moves from one port to another. In the case that a packet execute is invoked after a flow is newly installed this field is not properly initialized. This causes the pmtud mechanism to omit sending the required exception messages across the tunnel boundary and a second attempt needs to be made to make sure that the routing exception is properly setup. To fix this, we set the outgoing packet's pkt_type to PACKET_OUTGOING, since it can only get to the openvswitch module via a port device or packet command. Even for bridge ports as users, the pkt_type needs to be reset when doing the transmit as the packet is truly outgoing and routing needs to get involved post packet transformations, in the case of VXLAN/GENEVE/udp-tunnel packets. In general, the pkt_type on output gets ignored, since we go straight to the driver, but in the case of tunnel ports they go through IP routing layer. This issue is periodically encountered in complex setups, such as large openshift deployments, where multiple sets of tunnel traversal occurs. A way to recreate this is with the ovn-heater project that can setup a networking environment which mimics such large deployments. We need larger environments for this because we need to ensure that flow misses occur. In these environment, without this patch, we can see: ./ovn_cluster.sh start podman exec ovn-chassis-1 ip r a 170.168.0.5/32 dev eth1 mtu 1200 podman exec ovn-chassis-1 ip netns exec sw01p1 ip r flush cache podman exec ovn-chassis-1 ip netns exec sw01p1 \ ping 21.0.0.3 -M do -s 1300 -c2 PING 21.0.0.3 (21.0.0.3) 1300(1328) bytes of data. From 21.0.0.3 icmp_seq=2 Frag needed and DF set (mtu = 1142) --- 21.0.0.3 ping statistics --- ... Using tcpdump, we can also see the expected ICMP FRAG_NEEDED message is not sent into the server. With this patch, setting the pkt_type, we see the following: podman exec ovn-chassis-1 ip netns exec sw01p1 \ ping 21.0.0.3 -M do -s 1300 -c2 PING 21.0.0.3 (21.0.0.3) 1300(1328) bytes of data. From 21.0.0.3 icmp_seq=1 Frag needed and DF set (mtu = 1222) ping: local error: message too long, mtu=1222 --- 21.0.0.3 ping statistics --- ... In this case, the first ping request receives the FRAG_NEEDED message and a local routing exception is created. Tested-by: Jaime Caamano Reported-at: https://issues.redhat.com/browse/FDP-164 Fixes: 58264848a5a7 ("openvswitch: Add vxlan tunneling support.") Signed-off-by: Aaron Conole Acked-by: Eelco Chaudron Link: https://lore.kernel.org/r/20240516200941.16152-1-aconole@redhat.com Signed-off-by: Paolo Abeni --- net/openvswitch/actions.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net') diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 6fcd7e2ca81f..964225580824 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -936,6 +936,12 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, pskb_trim(skb, ovs_mac_header_len(key)); } + /* Need to set the pkt_type to involve the routing layer. The + * packet movement through the OVS datapath doesn't generally + * use routing, but this is needed for tunnel cases. + */ + skb->pkt_type = PACKET_OUTGOING; + if (likely(!mru || (skb->len <= mru + vport->dev->hard_header_len))) { ovs_vport_send(vport, skb, ovs_key_mac_proto(key)); -- cgit v1.2.3 From d26dd255ce44e325e163ff3ce53423d26911ad11 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 31 Mar 2024 10:44:05 +0200 Subject: net: 9p: virtio: drop owner assignment virtio core already sets the .owner, so driver does not need to. Signed-off-by: Krzysztof Kozlowski Message-Id: <20240331-module-owner-virtio-v2-18-98f04bfaf46a@linaro.org> Signed-off-by: Michael S. Tsirkin --- net/9p/trans_virtio.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index e305071eb7b8..0b8086f58ad5 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -781,7 +781,6 @@ static struct virtio_driver p9_virtio_drv = { .feature_table = features, .feature_table_size = ARRAY_SIZE(features), .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, .id_table = id_table, .probe = p9_virtio_probe, .remove = p9_virtio_remove, -- cgit v1.2.3 From b1c16d4a33a3625f2049400d97453f8a57f80639 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Sun, 31 Mar 2024 10:44:06 +0200 Subject: vsock/virtio: drop owner assignment virtio core already sets the .owner, so driver does not need to. Acked-by: Stefano Garzarella Signed-off-by: Krzysztof Kozlowski Message-Id: <20240331-module-owner-virtio-v2-19-98f04bfaf46a@linaro.org> Signed-off-by: Michael S. Tsirkin Reviewed-by: Stefan Hajnoczi --- net/vmw_vsock/virtio_transport.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net') diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c index ee5d306a96d0..43d405298857 100644 --- a/net/vmw_vsock/virtio_transport.c +++ b/net/vmw_vsock/virtio_transport.c @@ -859,7 +859,6 @@ static struct virtio_driver virtio_vsock_driver = { .feature_table = features, .feature_table_size = ARRAY_SIZE(features), .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, .id_table = id_table, .probe = virtio_vsock_probe, .remove = virtio_vsock_remove, -- cgit v1.2.3 From 2c92ca849fcc6ee7d0c358e9959abc9f58661aea Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Thu, 16 May 2024 13:34:54 -0400 Subject: tracing/treewide: Remove second parameter of __assign_str() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With the rework of how the __string() handles dynamic strings where it saves off the source string in field in the helper structure[1], the assignment of that value to the trace event field is stored in the helper value and does not need to be passed in again. This means that with: __string(field, mystring) Which use to be assigned with __assign_str(field, mystring), no longer needs the second parameter and it is unused. With this, __assign_str() will now only get a single parameter. There's over 700 users of __assign_str() and because coccinelle does not handle the TRACE_EVENT() macro I ended up using the following sed script: git grep -l __assign_str | while read a ; do sed -e 's/\(__assign_str([^,]*[^ ,]\) *,[^;]*/\1)/' $a > /tmp/test-file; mv /tmp/test-file $a; done I then searched for __assign_str() that did not end with ';' as those were multi line assignments that the sed script above would fail to catch. Note, the same updates will need to be done for: __assign_str_len() __assign_rel_str() __assign_rel_str_len() I tested this with both an allmodconfig and an allyesconfig (build only for both). [1] https://lore.kernel.org/linux-trace-kernel/20240222211442.634192653@goodmis.org/ Link: https://lore.kernel.org/linux-trace-kernel/20240516133454.681ba6a0@rorschach.local.home Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Cc: Linus Torvalds Cc: Julia Lawall Signed-off-by: Steven Rostedt (Google) Acked-by: Jani Nikula Acked-by: Christian König for the amdgpu parts. Acked-by: Thomas Hellström #for Acked-by: Rafael J. Wysocki # for thermal Acked-by: Takashi Iwai Acked-by: Darrick J. Wong # xfs Tested-by: Guenter Roeck --- arch/arm64/kernel/trace-events-emulation.h | 2 +- arch/powerpc/include/asm/trace.h | 4 +- arch/x86/kvm/trace.h | 2 +- drivers/base/regmap/trace.h | 18 ++-- drivers/base/trace.h | 2 +- drivers/block/rnbd/rnbd-srv-trace.h | 12 +-- drivers/bus/mhi/host/trace.h | 12 +-- drivers/cxl/core/trace.h | 32 +++--- drivers/dma-buf/sync_trace.h | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h | 16 +-- .../drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h | 2 +- drivers/gpu/drm/i915/display/intel_display_trace.h | 56 +++++----- drivers/gpu/drm/lima/lima_trace.h | 2 +- drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h | 12 +-- drivers/gpu/drm/scheduler/gpu_scheduler_trace.h | 4 +- drivers/gpu/drm/virtio/virtgpu_trace.h | 2 +- drivers/infiniband/core/cma_trace.h | 4 +- drivers/infiniband/hw/hfi1/hfi.h | 2 +- drivers/infiniband/hw/hfi1/trace_dbg.h | 2 +- drivers/infiniband/hw/hfi1/trace_rx.h | 2 +- drivers/infiniband/hw/hfi1/trace_tid.h | 4 +- drivers/infiniband/hw/hfi1/trace_tx.h | 4 +- drivers/infiniband/sw/rdmavt/trace.h | 2 +- drivers/infiniband/sw/rdmavt/trace_rvt.h | 2 +- drivers/interconnect/trace.h | 10 +- drivers/iommu/intel/trace.h | 14 +-- drivers/media/platform/nvidia/tegra-vde/trace.h | 2 +- drivers/misc/mei/mei-trace.h | 6 +- drivers/net/dsa/mv88e6xxx/trace.h | 4 +- .../net/ethernet/freescale/dpaa/dpaa_eth_trace.h | 2 +- .../net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h | 4 +- .../net/ethernet/fungible/funeth/funeth_trace.h | 6 +- drivers/net/ethernet/hisilicon/hns3/hns3_trace.h | 4 +- .../ethernet/hisilicon/hns3/hns3pf/hclge_trace.h | 12 +-- .../ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h | 10 +- drivers/net/ethernet/intel/i40e/i40e_trace.h | 10 +- drivers/net/ethernet/intel/iavf/iavf_trace.h | 6 +- drivers/net/ethernet/intel/ice/ice_trace.h | 12 +-- .../net/ethernet/marvell/octeontx2/af/rvu_trace.h | 12 +-- .../mellanox/mlx5/core/diag/cmd_tracepoint.h | 4 +- .../mellanox/mlx5/core/diag/en_rep_tracepoint.h | 2 +- .../mellanox/mlx5/core/diag/en_tc_tracepoint.h | 2 +- .../mellanox/mlx5/core/diag/fw_tracer_tracepoint.h | 5 +- .../mellanox/mlx5/core/esw/diag/qos_tracepoint.h | 8 +- .../mlx5/core/sf/dev/diag/dev_tracepoint.h | 2 +- .../mellanox/mlx5/core/sf/diag/sf_tracepoint.h | 14 +-- .../mellanox/mlx5/core/sf/diag/vhca_tracepoint.h | 2 +- drivers/net/fjes/fjes_trace.h | 10 +- drivers/net/hyperv/netvsc_trace.h | 8 +- drivers/net/wireless/ath/ath10k/trace.h | 64 +++++------ drivers/net/wireless/ath/ath11k/trace.h | 44 ++++---- drivers/net/wireless/ath/ath12k/trace.h | 16 +-- drivers/net/wireless/ath/ath6kl/trace.h | 4 +- drivers/net/wireless/ath/trace.h | 4 +- .../broadcom/brcm80211/brcmfmac/tracepoint.h | 4 +- .../brcm80211/brcmsmac/brcms_trace_brcmsmac.h | 2 +- .../brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h | 2 +- .../brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h | 6 +- .../net/wireless/intel/iwlwifi/iwl-devtrace-msg.h | 2 +- drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h | 2 +- drivers/soc/qcom/pmic_pdcharger_ulog.h | 2 +- drivers/soc/qcom/trace-aoss.h | 4 +- drivers/soc/qcom/trace-rpmh.h | 4 +- drivers/thermal/thermal_trace.h | 10 +- drivers/usb/cdns3/cdns3-trace.h | 26 ++--- drivers/usb/cdns3/cdnsp-trace.h | 10 +- drivers/usb/chipidea/trace.h | 4 +- drivers/usb/dwc3/trace.h | 8 +- drivers/usb/gadget/udc/cdns2/cdns2-trace.h | 22 ++-- drivers/usb/gadget/udc/trace.h | 4 +- drivers/usb/mtu3/mtu3_trace.h | 8 +- drivers/usb/musb/musb_trace.h | 12 +-- fs/bcachefs/trace.h | 6 +- fs/nfs/nfs4trace.h | 42 ++++---- fs/nfs/nfstrace.h | 41 ++++--- fs/nfsd/trace.h | 40 +++---- fs/ocfs2/ocfs2_trace.h | 60 +++++------ fs/smb/client/trace.h | 18 ++-- fs/xfs/scrub/trace.h | 10 +- fs/xfs/xfs_trace.h | 28 ++--- include/ras/ras_event.h | 12 +-- include/trace/events/asoc.h | 22 ++-- include/trace/events/avc.h | 6 +- include/trace/events/bridge.h | 16 +-- include/trace/events/btrfs.h | 6 +- include/trace/events/cgroup.h | 10 +- include/trace/events/clk.h | 18 ++-- include/trace/events/cma.h | 8 +- include/trace/events/devfreq.h | 4 +- include/trace/events/devlink.h | 50 ++++----- include/trace/events/dma_fence.h | 4 +- include/trace/events/erofs.h | 2 +- include/trace/events/f2fs.h | 20 ++-- include/trace/events/habanalabs.h | 10 +- include/trace/events/huge_memory.h | 4 +- include/trace/events/hwmon.h | 6 +- include/trace/events/initcall.h | 2 +- include/trace/events/intel_ish.h | 2 +- include/trace/events/io_uring.h | 14 +-- include/trace/events/iocost.h | 14 +-- include/trace/events/iommu.h | 8 +- include/trace/events/irq.h | 2 +- include/trace/events/iscsi.h | 2 +- include/trace/events/kmem.h | 2 +- include/trace/events/lock.h | 4 +- include/trace/events/mmap_lock.h | 4 +- include/trace/events/mmc.h | 4 +- include/trace/events/module.h | 8 +- include/trace/events/napi.h | 2 +- include/trace/events/neigh.h | 6 +- include/trace/events/net.h | 12 +-- include/trace/events/netlink.h | 2 +- include/trace/events/oom.h | 2 +- include/trace/events/osnoise.h | 2 +- include/trace/events/power.h | 23 ++-- include/trace/events/pwc.h | 4 +- include/trace/events/qdisc.h | 12 +-- include/trace/events/qla.h | 2 +- include/trace/events/qrtr.h | 2 +- include/trace/events/regulator.h | 6 +- include/trace/events/rpcgss.h | 20 ++-- include/trace/events/rpcrdma.h | 52 ++++----- include/trace/events/rpm.h | 6 +- include/trace/events/sched.h | 8 +- include/trace/events/sof.h | 12 +-- include/trace/events/sof_intel.h | 16 +-- include/trace/events/sunrpc.h | 118 ++++++++++----------- include/trace/events/swiotlb.h | 2 +- include/trace/events/target.h | 4 +- include/trace/events/tegra_apb_dma.h | 6 +- include/trace/events/ufs.h | 24 ++--- include/trace/events/workqueue.h | 2 +- include/trace/events/xdp.h | 2 +- include/trace/stages/stage6_event_callback.h | 4 +- kernel/trace/bpf_trace.h | 2 +- net/batman-adv/trace.h | 4 +- net/dsa/trace.h | 34 +++--- net/ieee802154/trace.h | 2 +- net/mac80211/trace.h | 2 +- net/openvswitch/openvswitch_trace.h | 8 +- net/smc/smc_tracepoint.h | 4 +- net/tipc/trace.h | 16 +-- net/wireless/trace.h | 2 +- samples/trace_events/trace-events-sample.h | 19 ++-- sound/core/pcm_trace.h | 2 +- sound/hda/trace.h | 6 +- sound/soc/intel/avs/trace.h | 4 +- 147 files changed, 794 insertions(+), 808 deletions(-) (limited to 'net') diff --git a/arch/arm64/kernel/trace-events-emulation.h b/arch/arm64/kernel/trace-events-emulation.h index 6c40f58b844a..c51b547b583e 100644 --- a/arch/arm64/kernel/trace-events-emulation.h +++ b/arch/arm64/kernel/trace-events-emulation.h @@ -18,7 +18,7 @@ TRACE_EVENT(instruction_emulation, ), TP_fast_assign( - __assign_str(instr, instr); + __assign_str(instr); __entry->addr = addr; ), diff --git a/arch/powerpc/include/asm/trace.h b/arch/powerpc/include/asm/trace.h index d9ac3a4f46e1..a7b69b25296b 100644 --- a/arch/powerpc/include/asm/trace.h +++ b/arch/powerpc/include/asm/trace.h @@ -137,7 +137,7 @@ TRACE_EVENT(rtas_input, TP_fast_assign( __entry->nargs = be32_to_cpu(rtas_args->nargs); - __assign_str(name, name); + __assign_str(name); be32_to_cpu_array(__get_dynamic_array(inputs), rtas_args->args, __entry->nargs); ), @@ -162,7 +162,7 @@ TRACE_EVENT(rtas_output, TP_fast_assign( __entry->nr_other = be32_to_cpu(rtas_args->nret) - 1; __entry->status = be32_to_cpu(rtas_args->rets[0]); - __assign_str(name, name); + __assign_str(name); be32_to_cpu_array(__get_dynamic_array(other_outputs), &rtas_args->rets[1], __entry->nr_other); ), diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h index 9d0b02ef307e..e19fed438a67 100644 --- a/arch/x86/kvm/trace.h +++ b/arch/x86/kvm/trace.h @@ -1678,7 +1678,7 @@ TRACE_EVENT(kvm_nested_vmenter_failed, ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); __entry->err = err; ), diff --git a/drivers/base/regmap/trace.h b/drivers/base/regmap/trace.h index 704e106e5dbd..bcc5a8b226a6 100644 --- a/drivers/base/regmap/trace.h +++ b/drivers/base/regmap/trace.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(regmap_reg, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->val = val; ), @@ -74,7 +74,7 @@ DECLARE_EVENT_CLASS(regmap_bulk, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->val_len = val_len; memcpy(__get_dynamic_array(buf), val, val_len); @@ -113,7 +113,7 @@ DECLARE_EVENT_CLASS(regmap_block, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->reg = reg; __entry->count = count; ), @@ -163,9 +163,9 @@ TRACE_EVENT(regcache_sync, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); - __assign_str(status, status); - __assign_str(type, type); + __assign_str(name); + __assign_str(status); + __assign_str(type); ), TP_printk("%s type=%s status=%s", __get_str(name), @@ -184,7 +184,7 @@ DECLARE_EVENT_CLASS(regmap_bool, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->flag = flag; ), @@ -216,7 +216,7 @@ DECLARE_EVENT_CLASS(regmap_async, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -264,7 +264,7 @@ TRACE_EVENT(regcache_drop_region, ), TP_fast_assign( - __assign_str(name, regmap_name(map)); + __assign_str(name); __entry->from = from; __entry->to = to; ), diff --git a/drivers/base/trace.h b/drivers/base/trace.h index 3192e18f877e..e52b6eae060d 100644 --- a/drivers/base/trace.h +++ b/drivers/base/trace.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(devres, __field(size_t, size) ), TP_fast_assign( - __assign_str(devname, dev_name(dev)); + __assign_str(devname); __entry->op = op; __entry->node = node; __entry->name = name; diff --git a/drivers/block/rnbd/rnbd-srv-trace.h b/drivers/block/rnbd/rnbd-srv-trace.h index 8dedf73bdd28..89d0bcb17195 100644 --- a/drivers/block/rnbd/rnbd-srv-trace.h +++ b/drivers/block/rnbd/rnbd-srv-trace.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(rnbd_srv_link_class, TP_fast_assign( __entry->qdepth = srv->queue_depth; - __assign_str(sessname, srv->sessname); + __assign_str(sessname); ), TP_printk("sessname: %s qdepth: %d", @@ -85,7 +85,7 @@ TRACE_EVENT(process_rdma, ), TP_fast_assign( - __assign_str(sessname, srv->sessname); + __assign_str(sessname); __entry->dir = id->dir; __entry->ver = srv->ver; __entry->device_id = le32_to_cpu(msg->device_id); @@ -130,7 +130,7 @@ TRACE_EVENT(process_msg_sess_info, __entry->proto_ver = srv->ver; __entry->clt_ver = msg->ver; __entry->srv_ver = RNBD_PROTO_VER_MAJOR; - __assign_str(sessname, srv->sessname); + __assign_str(sessname); ), TP_printk("Session %s using proto-ver %d (clt-ver: %d, srv-ver: %d)", @@ -165,8 +165,8 @@ TRACE_EVENT(process_msg_open, TP_fast_assign( __entry->access_mode = msg->access_mode; - __assign_str(sessname, srv->sessname); - __assign_str(dev_name, msg->dev_name); + __assign_str(sessname); + __assign_str(dev_name); ), TP_printk("Open message received: session='%s' path='%s' access_mode=%s", @@ -189,7 +189,7 @@ TRACE_EVENT(process_msg_close, TP_fast_assign( __entry->device_id = le32_to_cpu(msg->device_id); - __assign_str(sessname, srv->sessname); + __assign_str(sessname); ), TP_printk("Close message received: session='%s' device id='%d'", diff --git a/drivers/bus/mhi/host/trace.h b/drivers/bus/mhi/host/trace.h index 368515dcb22d..95613c8ebe06 100644 --- a/drivers/bus/mhi/host/trace.h +++ b/drivers/bus/mhi/host/trace.h @@ -103,7 +103,7 @@ TRACE_EVENT(mhi_gen_tre, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->ch_num = mhi_chan->chan; __entry->wp = mhi_tre; __entry->tre_ptr = mhi_tre->ptr; @@ -131,7 +131,7 @@ TRACE_EVENT(mhi_intvec_states, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->local_ee = mhi_cntrl->ee; __entry->state = mhi_cntrl->dev_state; __entry->dev_ee = dev_ee; @@ -158,7 +158,7 @@ TRACE_EVENT(mhi_tryset_pm_state, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); if (pm_state) pm_state = __fls(pm_state); __entry->pm_state = pm_state; @@ -184,7 +184,7 @@ DECLARE_EVENT_CLASS(mhi_process_event_ring, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->rp = rp; __entry->ptr = rp->ptr; __entry->dword0 = rp->dword[0]; @@ -226,7 +226,7 @@ DECLARE_EVENT_CLASS(mhi_update_channel_state, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->ch_num = mhi_chan->chan; __entry->state = state; __entry->reason = reason; @@ -265,7 +265,7 @@ TRACE_EVENT(mhi_pm_st_transition, ), TP_fast_assign( - __assign_str(name, mhi_cntrl->mhi_dev->name); + __assign_str(name); __entry->state = state; ), diff --git a/drivers/cxl/core/trace.h b/drivers/cxl/core/trace.h index 07a0394b1d99..ee5cd4eb2f16 100644 --- a/drivers/cxl/core/trace.h +++ b/drivers/cxl/core/trace.h @@ -60,8 +60,8 @@ TRACE_EVENT(cxl_aer_uncorrectable_error, __array(u32, header_log, CXL_HEADERLOG_SIZE_U32) ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->status = status; __entry->first_error = fe; @@ -106,8 +106,8 @@ TRACE_EVENT(cxl_aer_correctable_error, __field(u32, status) ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->status = status; ), @@ -142,8 +142,8 @@ TRACE_EVENT(cxl_overflow, ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->log = log; __entry->count = le16_to_cpu(payload->overflow_err_count); @@ -200,8 +200,8 @@ TRACE_EVENT(cxl_overflow, __field(u8, hdr_maint_op_class) #define CXL_EVT_TP_fast_assign(cxlmd, l, hdr) \ - __assign_str(memdev, dev_name(&(cxlmd)->dev)); \ - __assign_str(host, dev_name((cxlmd)->dev.parent)); \ + __assign_str(memdev); \ + __assign_str(host); \ __entry->log = (l); \ __entry->serial = (cxlmd)->cxlds->serial; \ __entry->hdr_length = (hdr).length; \ @@ -359,10 +359,10 @@ TRACE_EVENT(cxl_general_media, __entry->validity_flags = get_unaligned_le16(&rec->validity_flags); __entry->hpa = hpa; if (cxlr) { - __assign_str(region_name, dev_name(&cxlr->dev)); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &cxlr->params.uuid); } else { - __assign_str(region_name, ""); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &uuid_null); } ), @@ -462,10 +462,10 @@ TRACE_EVENT(cxl_dram, CXL_EVENT_DER_CORRECTION_MASK_SIZE); __entry->hpa = hpa; if (cxlr) { - __assign_str(region_name, dev_name(&cxlr->dev)); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &cxlr->params.uuid); } else { - __assign_str(region_name, ""); + __assign_str(region_name); uuid_copy(&__entry->region_uuid, &uuid_null); } ), @@ -692,8 +692,8 @@ TRACE_EVENT(cxl_poison, ), TP_fast_assign( - __assign_str(memdev, dev_name(&cxlmd->dev)); - __assign_str(host, dev_name(cxlmd->dev.parent)); + __assign_str(memdev); + __assign_str(host); __entry->serial = cxlmd->cxlds->serial; __entry->overflow_ts = cxl_poison_overflow(flags, overflow_ts); __entry->dpa = cxl_poison_record_dpa(record); @@ -702,12 +702,12 @@ TRACE_EVENT(cxl_poison, __entry->trace_type = trace_type; __entry->flags = flags; if (cxlr) { - __assign_str(region, dev_name(&cxlr->dev)); + __assign_str(region); memcpy(__entry->uuid, &cxlr->params.uuid, 16); __entry->hpa = cxl_trace_hpa(cxlr, cxlmd, __entry->dpa); } else { - __assign_str(region, ""); + __assign_str(region); memset(__entry->uuid, 0, 16); __entry->hpa = ULLONG_MAX; } diff --git a/drivers/dma-buf/sync_trace.h b/drivers/dma-buf/sync_trace.h index 06e468a218ff..d71dcf954b8d 100644 --- a/drivers/dma-buf/sync_trace.h +++ b/drivers/dma-buf/sync_trace.h @@ -20,7 +20,7 @@ TRACE_EVENT(sync_timeline, ), TP_fast_assign( - __assign_str(name, timeline->name); + __assign_str(name); __entry->value = timeline->value; ), diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h index f539b1d00234..7aafeb763e5d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h @@ -178,10 +178,10 @@ TRACE_EVENT(amdgpu_cs_ioctl, TP_fast_assign( __entry->sched_job_id = job->base.id; - __assign_str(timeline, AMDGPU_JOB_GET_TIMELINE_NAME(job)); + __assign_str(timeline); __entry->context = job->base.s_fence->finished.context; __entry->seqno = job->base.s_fence->finished.seqno; - __assign_str(ring, to_amdgpu_ring(job->base.sched)->name); + __assign_str(ring); __entry->num_ibs = job->num_ibs; ), TP_printk("sched_job=%llu, timeline=%s, context=%u, seqno=%u, ring_name=%s, num_ibs=%u", @@ -203,10 +203,10 @@ TRACE_EVENT(amdgpu_sched_run_job, TP_fast_assign( __entry->sched_job_id = job->base.id; - __assign_str(timeline, AMDGPU_JOB_GET_TIMELINE_NAME(job)); + __assign_str(timeline); __entry->context = job->base.s_fence->finished.context; __entry->seqno = job->base.s_fence->finished.seqno; - __assign_str(ring, to_amdgpu_ring(job->base.sched)->name); + __assign_str(ring); __entry->num_ibs = job->num_ibs; ), TP_printk("sched_job=%llu, timeline=%s, context=%u, seqno=%u, ring_name=%s, num_ibs=%u", @@ -231,7 +231,7 @@ TRACE_EVENT(amdgpu_vm_grab_id, TP_fast_assign( __entry->pasid = vm->pasid; - __assign_str(ring, ring->name); + __assign_str(ring); __entry->vmid = job->vmid; __entry->vm_hub = ring->vm_hub, __entry->pd_addr = job->vm_pd_addr; @@ -425,7 +425,7 @@ TRACE_EVENT(amdgpu_vm_flush, ), TP_fast_assign( - __assign_str(ring, ring->name); + __assign_str(ring); __entry->vmid = vmid; __entry->vm_hub = ring->vm_hub; __entry->pd_addr = pd_addr; @@ -526,7 +526,7 @@ TRACE_EVENT(amdgpu_ib_pipe_sync, ), TP_fast_assign( - __assign_str(ring, sched_job->base.sched->name); + __assign_str(ring); __entry->id = sched_job->base.id; __entry->fence = fence; __entry->ctx = fence->context; @@ -563,7 +563,7 @@ TRACE_EVENT(amdgpu_runpm_reference_dumps, ), TP_fast_assign( __entry->index = index; - __assign_str(func, func); + __assign_str(func); ), TP_printk("amdgpu runpm reference dump 0x%x: 0x%s\n", __entry->index, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h index 133af994a08c..4686d4b0cbad 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h @@ -87,7 +87,7 @@ TRACE_EVENT(amdgpu_dc_performance, __entry->writes = write_count; __entry->read_delta = read_count - *last_read; __entry->write_delta = write_count - *last_write; - __assign_str(func, func); + __assign_str(func); __entry->line = line; *last_read = read_count; *last_write = write_count; diff --git a/drivers/gpu/drm/i915/display/intel_display_trace.h b/drivers/gpu/drm/i915/display/intel_display_trace.h index 7862e7cefe02..49a5e6d9dc0d 100644 --- a/drivers/gpu/drm/i915/display/intel_display_trace.h +++ b/drivers/gpu/drm/i915/display/intel_display_trace.h @@ -34,7 +34,7 @@ TRACE_EVENT(intel_pipe_enable, TP_fast_assign( struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_crtc *it__; - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); for_each_intel_crtc(&dev_priv->drm, it__) { __entry->frame[it__->pipe] = intel_crtc_get_vblank_counter(it__); __entry->scanline[it__->pipe] = intel_get_crtc_scanline(it__); @@ -63,7 +63,7 @@ TRACE_EVENT(intel_pipe_disable, TP_fast_assign( struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_crtc *it__; - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); for_each_intel_crtc(&dev_priv->drm, it__) { __entry->frame[it__->pipe] = intel_crtc_get_vblank_counter(it__); __entry->scanline[it__->pipe] = intel_get_crtc_scanline(it__); @@ -91,7 +91,7 @@ TRACE_EVENT(intel_pipe_crc, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -119,7 +119,7 @@ TRACE_EVENT(intel_cpu_fifo_underrun, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -144,7 +144,7 @@ TRACE_EVENT(intel_pch_fifo_underrun, TP_fast_assign( enum pipe pipe = pch_transcoder; struct intel_crtc *crtc = intel_crtc_for_pipe(dev_priv, pipe); - __assign_str(dev, __dev_name_i915(dev_priv)); + __assign_str(dev); __entry->pipe = pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -169,7 +169,7 @@ TRACE_EVENT(intel_memory_cxsr, TP_fast_assign( struct intel_crtc *crtc; - __assign_str(dev, __dev_name_i915(dev_priv)); + __assign_str(dev); for_each_intel_crtc(&dev_priv->drm, crtc) { __entry->frame[crtc->pipe] = intel_crtc_get_vblank_counter(crtc); __entry->scanline[crtc->pipe] = intel_get_crtc_scanline(crtc); @@ -209,7 +209,7 @@ TRACE_EVENT(g4x_wm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -256,7 +256,7 @@ TRACE_EVENT(vlv_wm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -293,7 +293,7 @@ TRACE_EVENT(vlv_fifo_size, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -323,8 +323,8 @@ TRACE_EVENT(intel_plane_update_noarm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -354,8 +354,8 @@ TRACE_EVENT(intel_plane_update_arm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -383,8 +383,8 @@ TRACE_EVENT(intel_plane_disable_arm, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -410,8 +410,8 @@ TRACE_EVENT(intel_fbc_activate, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(to_i915(plane->base.dev), plane->pipe); - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -437,8 +437,8 @@ TRACE_EVENT(intel_fbc_deactivate, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(to_i915(plane->base.dev), plane->pipe); - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -464,8 +464,8 @@ TRACE_EVENT(intel_fbc_nuke, TP_fast_assign( struct intel_crtc *crtc = intel_crtc_for_pipe(to_i915(plane->base.dev), plane->pipe); - __assign_str(dev, __dev_name_kms(plane)); - __assign_str(name, plane->base.name); + __assign_str(dev); + __assign_str(name); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -488,7 +488,7 @@ TRACE_EVENT(intel_crtc_vblank_work_start, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -511,7 +511,7 @@ TRACE_EVENT(intel_crtc_vblank_work_end, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -536,7 +536,7 @@ TRACE_EVENT(intel_pipe_update_start, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = intel_crtc_get_vblank_counter(crtc); __entry->scanline = intel_get_crtc_scanline(crtc); @@ -564,7 +564,7 @@ TRACE_EVENT(intel_pipe_update_vblank_evaded, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = crtc->debug.start_vbl_count; __entry->scanline = crtc->debug.scanline_start; @@ -590,7 +590,7 @@ TRACE_EVENT(intel_pipe_update_end, ), TP_fast_assign( - __assign_str(dev, __dev_name_kms(crtc)); + __assign_str(dev); __entry->pipe = crtc->pipe; __entry->frame = frame; __entry->scanline = scanline_end; @@ -613,7 +613,7 @@ TRACE_EVENT(intel_frontbuffer_invalidate, ), TP_fast_assign( - __assign_str(dev, __dev_name_i915(i915)); + __assign_str(dev); __entry->frontbuffer_bits = frontbuffer_bits; __entry->origin = origin; ), @@ -634,7 +634,7 @@ TRACE_EVENT(intel_frontbuffer_flush, ), TP_fast_assign( - __assign_str(dev, __dev_name_i915(i915)); + __assign_str(dev); __entry->frontbuffer_bits = frontbuffer_bits; __entry->origin = origin; ), diff --git a/drivers/gpu/drm/lima/lima_trace.h b/drivers/gpu/drm/lima/lima_trace.h index 494b9790b1da..3a349d10304e 100644 --- a/drivers/gpu/drm/lima/lima_trace.h +++ b/drivers/gpu/drm/lima/lima_trace.h @@ -24,7 +24,7 @@ DECLARE_EVENT_CLASS(lima_task, __entry->task_id = task->base.id; __entry->context = task->base.s_fence->finished.context; __entry->seqno = task->base.s_fence->finished.seqno; - __assign_str(pipe, task->base.sched->name); + __assign_str(pipe); ), TP_printk("task=%llu, context=%u seqno=%u pipe=%s", diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h index bd92fb2979aa..0fdd41162e4b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h @@ -113,7 +113,7 @@ TRACE_EVENT(tracing_mark_write, ), TP_fast_assign( __entry->pid = pid; - __assign_str(trace_name, name); + __assign_str(trace_name); __entry->trace_begin = trace_begin; ), TP_printk("%s|%d|%s", __entry->trace_begin ? "B" : "E", @@ -130,7 +130,7 @@ TRACE_EVENT(dpu_trace_counter, ), TP_fast_assign( __entry->pid = current->tgid; - __assign_str(counter_name, name); + __assign_str(counter_name); __entry->value = value; ), TP_printk("%d|%s|%d", __entry->pid, @@ -379,7 +379,7 @@ TRACE_EVENT(dpu_enc_rc, __entry->sw_event = sw_event; __entry->idle_pc_supported = idle_pc_supported; __entry->rc_state = rc_state; - __assign_str(stage_str, stage); + __assign_str(stage_str); ), TP_printk("%s: id:%u, sw_event:%d, idle_pc_supported:%s, rc_state:%d", __get_str(stage_str), __entry->drm_id, __entry->sw_event, @@ -401,7 +401,7 @@ TRACE_EVENT(dpu_enc_frame_done_cb_not_busy, TP_fast_assign( __entry->drm_id = drm_id; __entry->event = event; - __assign_str(intf_mode_str, intf_mode); + __assign_str(intf_mode_str); __entry->intf_idx = intf_idx; __entry->wb_idx = wb_idx; ), @@ -446,7 +446,7 @@ TRACE_EVENT(dpu_enc_trigger_flush, ), TP_fast_assign( __entry->drm_id = drm_id; - __assign_str(intf_mode_str, intf_mode); + __assign_str(intf_mode_str); __entry->intf_idx = intf_idx; __entry->wb_idx = wb_idx; __entry->pending_kickoff_cnt = pending_kickoff_cnt; @@ -946,7 +946,7 @@ TRACE_EVENT(dpu_core_perf_update_clk, __field( u64, clk_rate ) ), TP_fast_assign( - __assign_str(dev_name, dev->unique); + __assign_str(dev_name); __entry->stop_req = stop_req; __entry->clk_rate = clk_rate; ), diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h index f8ed093b7356..c75302ca3427 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h @@ -48,7 +48,7 @@ DECLARE_EVENT_CLASS(drm_sched_job, __entry->entity = entity; __entry->id = sched_job->id; __entry->fence = &sched_job->s_fence->finished; - __assign_str(name, sched_job->sched->name); + __assign_str(name); __entry->job_count = spsc_queue_count(&entity->job_queue); __entry->hw_job_count = atomic_read( &sched_job->sched->credit_count); @@ -94,7 +94,7 @@ TRACE_EVENT(drm_sched_job_wait_dep, ), TP_fast_assign( - __assign_str(name, sched_job->sched->name); + __assign_str(name); __entry->id = sched_job->id; __entry->fence = fence; __entry->ctx = fence->context; diff --git a/drivers/gpu/drm/virtio/virtgpu_trace.h b/drivers/gpu/drm/virtio/virtgpu_trace.h index 031bc77689d5..227bf0ae7ed5 100644 --- a/drivers/gpu/drm/virtio/virtgpu_trace.h +++ b/drivers/gpu/drm/virtio/virtgpu_trace.h @@ -25,7 +25,7 @@ DECLARE_EVENT_CLASS(virtio_gpu_cmd, TP_fast_assign( __entry->dev = vq->vdev->index; __entry->vq = vq->index; - __assign_str(name, vq->name); + __assign_str(name); __entry->type = le32_to_cpu(hdr->type); __entry->flags = le32_to_cpu(hdr->flags); __entry->fence_id = le64_to_cpu(hdr->fence_id); diff --git a/drivers/infiniband/core/cma_trace.h b/drivers/infiniband/core/cma_trace.h index 47f3c6e4be89..dc622f3778be 100644 --- a/drivers/infiniband/core/cma_trace.h +++ b/drivers/infiniband/core/cma_trace.h @@ -84,7 +84,7 @@ TRACE_EVENT(cm_id_attach, sizeof(struct sockaddr_in6)); memcpy(__entry->dstaddr, &id_priv->id.route.addr.dst_addr, sizeof(struct sockaddr_in6)); - __assign_str(devname, device->name); + __assign_str(devname); ), TP_printk("cm.id=%u src=%pISpc dst=%pISpc device=%s", @@ -334,7 +334,7 @@ DECLARE_EVENT_CLASS(cma_client_class, ), TP_fast_assign( - __assign_str(name, device->name); + __assign_str(name); ), TP_printk("device name=%s", diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index 4b3f1cb125fc..eb38f81aeeb1 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -2425,7 +2425,7 @@ static inline bool hfi1_need_drop(struct hfi1_devdata *dd) int hfi1_tempsense_rd(struct hfi1_devdata *dd, struct hfi1_temp *temp); #define DD_DEV_ENTRY(dd) __string(dev, dev_name(&(dd)->pcidev->dev)) -#define DD_DEV_ASSIGN(dd) __assign_str(dev, dev_name(&(dd)->pcidev->dev)) +#define DD_DEV_ASSIGN(dd) __assign_str(dev) static inline void hfi1_update_ah_attr(struct ib_device *ibdev, struct rdma_ah_attr *attr) diff --git a/drivers/infiniband/hw/hfi1/trace_dbg.h b/drivers/infiniband/hw/hfi1/trace_dbg.h index 75599d5168db..58304b91380f 100644 --- a/drivers/infiniband/hw/hfi1/trace_dbg.h +++ b/drivers/infiniband/hw/hfi1/trace_dbg.h @@ -33,7 +33,7 @@ DECLARE_EVENT_CLASS(hfi1_trace_template, TP_STRUCT__entry(__string(function, function) __vstring(msg, vaf->fmt, vaf->va) ), - TP_fast_assign(__assign_str(function, function); + TP_fast_assign(__assign_str(function); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("(%s) %s", diff --git a/drivers/infiniband/hw/hfi1/trace_rx.h b/drivers/infiniband/hw/hfi1/trace_rx.h index e6904aa80c00..8d5e12fe88a5 100644 --- a/drivers/infiniband/hw/hfi1/trace_rx.h +++ b/drivers/infiniband/hw/hfi1/trace_rx.h @@ -90,7 +90,7 @@ TRACE_EVENT(hfi1_mmu_invalidate, TP_fast_assign( __entry->ctxt = ctxt; __entry->subctxt = subctxt; - __assign_str(type, type); + __assign_str(type); __entry->start = start; __entry->end = end; ), diff --git a/drivers/infiniband/hw/hfi1/trace_tid.h b/drivers/infiniband/hw/hfi1/trace_tid.h index d129b8195959..e358f5b885fa 100644 --- a/drivers/infiniband/hw/hfi1/trace_tid.h +++ b/drivers/infiniband/hw/hfi1/trace_tid.h @@ -358,7 +358,7 @@ DECLARE_EVENT_CLASS(/* msg */ ), TP_fast_assign(/* assign */ __entry->qpn = qp ? qp->ibqp.qp_num : 0; - __assign_str(msg, msg); + __assign_str(msg); __entry->more = more; ), TP_printk(/* print */ @@ -651,7 +651,7 @@ DECLARE_EVENT_CLASS(/* tid_node */ TP_fast_assign(/* assign */ DD_DEV_ASSIGN(dd_from_ibdev(qp->ibqp.device)); __entry->qpn = qp->ibqp.qp_num; - __assign_str(msg, msg); + __assign_str(msg); __entry->index = index; __entry->base = base; __entry->map = map; diff --git a/drivers/infiniband/hw/hfi1/trace_tx.h b/drivers/infiniband/hw/hfi1/trace_tx.h index c79856d4fdfb..c0ba6b0a2c4e 100644 --- a/drivers/infiniband/hw/hfi1/trace_tx.h +++ b/drivers/infiniband/hw/hfi1/trace_tx.h @@ -740,8 +740,8 @@ TRACE_EVENT(hfi1_sdma_state, __string(newstate, nstate) ), TP_fast_assign(DD_DEV_ASSIGN(sde->dd); - __assign_str(curstate, cstate); - __assign_str(newstate, nstate); + __assign_str(curstate); + __assign_str(newstate); ), TP_printk("[%s] current state %s new state %s", __get_str(dev), diff --git a/drivers/infiniband/sw/rdmavt/trace.h b/drivers/infiniband/sw/rdmavt/trace.h index 4341965a5ea7..bdb6b9326b64 100644 --- a/drivers/infiniband/sw/rdmavt/trace.h +++ b/drivers/infiniband/sw/rdmavt/trace.h @@ -4,7 +4,7 @@ */ #define RDI_DEV_ENTRY(rdi) __string(dev, rvt_get_ibdev_name(rdi)) -#define RDI_DEV_ASSIGN(rdi) __assign_str(dev, rvt_get_ibdev_name(rdi)) +#define RDI_DEV_ASSIGN(rdi) __assign_str(dev) #include "trace_rvt.h" #include "trace_qp.h" diff --git a/drivers/infiniband/sw/rdmavt/trace_rvt.h b/drivers/infiniband/sw/rdmavt/trace_rvt.h index df33c2ca9710..a00489e66ddf 100644 --- a/drivers/infiniband/sw/rdmavt/trace_rvt.h +++ b/drivers/infiniband/sw/rdmavt/trace_rvt.h @@ -24,7 +24,7 @@ TRACE_EVENT(rvt_dbg, ), TP_fast_assign( RDI_DEV_ASSIGN(rdi); - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("[%s]: %s", __get_str(dev), __get_str(msg)) ); diff --git a/drivers/interconnect/trace.h b/drivers/interconnect/trace.h index 3d668ff566bf..206373546528 100644 --- a/drivers/interconnect/trace.h +++ b/drivers/interconnect/trace.h @@ -32,9 +32,9 @@ TRACE_EVENT(icc_set_bw, ), TP_fast_assign( - __assign_str(path_name, p->name); - __assign_str(dev, dev_name(p->reqs[i].dev)); - __assign_str(node_name, n->name); + __assign_str(path_name); + __assign_str(dev); + __assign_str(node_name); __entry->avg_bw = avg_bw; __entry->peak_bw = peak_bw; __entry->node_avg_bw = n->avg_bw; @@ -64,8 +64,8 @@ TRACE_EVENT(icc_set_bw_end, ), TP_fast_assign( - __assign_str(path_name, p->name); - __assign_str(dev, dev_name(p->reqs[0].dev)); + __assign_str(path_name); + __assign_str(dev); __entry->ret = ret; ), diff --git a/drivers/iommu/intel/trace.h b/drivers/iommu/intel/trace.h index 961ac1c1bc21..9defdae6ebae 100644 --- a/drivers/iommu/intel/trace.h +++ b/drivers/iommu/intel/trace.h @@ -32,7 +32,7 @@ TRACE_EVENT(qi_submit, ), TP_fast_assign( - __assign_str(iommu, iommu->name); + __assign_str(iommu); __entry->qw0 = qw0; __entry->qw1 = qw1; __entry->qw2 = qw2; @@ -79,8 +79,8 @@ TRACE_EVENT(prq_report, __entry->dw2 = dw2; __entry->dw3 = dw3; __entry->seq = seq; - __assign_str(iommu, iommu->name); - __assign_str(dev, dev_name(dev)); + __assign_str(iommu); + __assign_str(dev); ), TP_printk("%s/%s seq# %ld: %s", @@ -102,8 +102,8 @@ DECLARE_EVENT_CLASS(cache_tag_log, __field(u32, users) ), TP_fast_assign( - __assign_str(iommu, tag->iommu->name); - __assign_str(dev, dev_name(tag->dev)); + __assign_str(iommu); + __assign_str(dev); __entry->type = tag->type; __entry->domain_id = tag->domain_id; __entry->pasid = tag->pasid; @@ -152,8 +152,8 @@ DECLARE_EVENT_CLASS(cache_tag_flush, __field(unsigned long, mask) ), TP_fast_assign( - __assign_str(iommu, tag->iommu->name); - __assign_str(dev, dev_name(tag->dev)); + __assign_str(iommu); + __assign_str(dev); __entry->type = tag->type; __entry->domain_id = tag->domain_id; __entry->pasid = tag->pasid; diff --git a/drivers/media/platform/nvidia/tegra-vde/trace.h b/drivers/media/platform/nvidia/tegra-vde/trace.h index 7853ab095ca4..e8a75a7bd05d 100644 --- a/drivers/media/platform/nvidia/tegra-vde/trace.h +++ b/drivers/media/platform/nvidia/tegra-vde/trace.h @@ -20,7 +20,7 @@ DECLARE_EVENT_CLASS(register_access, __field(u32, value) ), TP_fast_assign( - __assign_str(hw_name, tegra_vde_reg_base_name(vde, base)); + __assign_str(hw_name); __entry->offset = offset; __entry->value = value; ), diff --git a/drivers/misc/mei/mei-trace.h b/drivers/misc/mei/mei-trace.h index fe46ff2b9d69..5312edbf5190 100644 --- a/drivers/misc/mei/mei-trace.h +++ b/drivers/misc/mei/mei-trace.h @@ -26,7 +26,7 @@ TRACE_EVENT(mei_reg_read, __field(u32, val) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->reg = reg; __entry->offs = offs; __entry->val = val; @@ -45,7 +45,7 @@ TRACE_EVENT(mei_reg_write, __field(u32, val) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->reg = reg; __entry->offs = offs; __entry->val = val; @@ -64,7 +64,7 @@ TRACE_EVENT(mei_pci_cfg_read, __field(u32, val) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->reg = reg; __entry->offs = offs; __entry->val = val; diff --git a/drivers/net/dsa/mv88e6xxx/trace.h b/drivers/net/dsa/mv88e6xxx/trace.h index f59ca04768e7..5bd015b2b97a 100644 --- a/drivers/net/dsa/mv88e6xxx/trace.h +++ b/drivers/net/dsa/mv88e6xxx/trace.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(mv88e6xxx_atu_violation, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->spid = spid; __entry->portvec = portvec; memcpy(__entry->addr, addr, ETH_ALEN); @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(mv88e6xxx_vtu_violation, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->spid = spid; __entry->vid = vid; ), diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h index 889f89df9930..6f0e58a2a58a 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth_trace.h @@ -57,7 +57,7 @@ DECLARE_EVENT_CLASS(dpaa_eth_fd, __entry->fd_offset = qm_fd_get_offset(fd); __entry->fd_length = qm_fd_get_length(fd); __entry->fd_status = fd->status; - __assign_str(name, netdev->name); + __assign_str(name); ), /* This is what gets printed when the trace event is triggered */ diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h index 9b43fadb9b11..956767e0869c 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-trace.h @@ -48,7 +48,7 @@ DECLARE_EVENT_CLASS(dpaa2_eth_fd, __entry->fd_addr = dpaa2_fd_get_addr(fd); __entry->fd_len = dpaa2_fd_get_len(fd); __entry->fd_offset = dpaa2_fd_get_offset(fd); - __assign_str(name, netdev->name); + __assign_str(name); ), /* This is what gets printed when the trace event is @@ -144,7 +144,7 @@ DECLARE_EVENT_CLASS(dpaa2_eth_buf, __entry->dma_addr = dma_addr; __entry->map_size = map_size; __entry->bpid = bpid; - __assign_str(name, netdev->name); + __assign_str(name); ), /* This is what gets printed when the trace event is diff --git a/drivers/net/ethernet/fungible/funeth/funeth_trace.h b/drivers/net/ethernet/fungible/funeth/funeth_trace.h index 9e58dfec19d5..b9985900f30b 100644 --- a/drivers/net/ethernet/fungible/funeth/funeth_trace.h +++ b/drivers/net/ethernet/fungible/funeth/funeth_trace.h @@ -32,7 +32,7 @@ TRACE_EVENT(funeth_tx, __entry->len = len; __entry->sqe_idx = sqe_idx; __entry->ngle = ngle; - __assign_str(devname, txq->netdev->name); + __assign_str(devname); ), TP_printk("%s: Txq %u, SQE idx %u, len %u, num GLEs %u", @@ -62,7 +62,7 @@ TRACE_EVENT(funeth_tx_free, __entry->sqe_idx = sqe_idx; __entry->num_sqes = num_sqes; __entry->hw_head = hw_head; - __assign_str(devname, txq->netdev->name); + __assign_str(devname); ), TP_printk("%s: Txq %u, SQE idx %u, SQEs %u, HW head %u", @@ -97,7 +97,7 @@ TRACE_EVENT(funeth_rx, __entry->len = pkt_len; __entry->hash = hash; __entry->cls_vec = cls_vec; - __assign_str(devname, rxq->netdev->name); + __assign_str(devname); ), TP_printk("%s: Rxq %u, CQ head %u, RQEs %u, len %u, hash %u, CV %#x", diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h index b8a1ecb4b8fb..3362b8d14d4f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_trace.h @@ -84,7 +84,7 @@ TRACE_EVENT(hns3_tx_desc, __entry->desc_dma = ring->desc_dma_addr, memcpy(__entry->desc, &ring->desc[cur_ntu], sizeof(struct hns3_desc)); - __assign_str(devname, ring->tqp->handle->kinfo.netdev->name); + __assign_str(devname); ), TP_printk( @@ -117,7 +117,7 @@ TRACE_EVENT(hns3_rx_desc, __entry->buf_dma = ring->desc_cb[ring->next_to_clean].dma; memcpy(__entry->desc, &ring->desc[ring->next_to_clean], sizeof(struct hns3_desc)); - __assign_str(devname, ring->tqp->handle->kinfo.netdev->name); + __assign_str(devname); ), TP_printk( diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h index 7e47f0c21d88..7103cf04bffc 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h @@ -33,8 +33,8 @@ TRACE_EVENT(hclge_pf_mbx_get, __entry->vfid = req->mbx_src_vfid; __entry->code = req->msg.code; __entry->subcode = req->msg.subcode; - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->vport[0].nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_vf_to_pf_cmd)); ), @@ -64,8 +64,8 @@ TRACE_EVENT(hclge_pf_mbx_send, TP_fast_assign( __entry->vfid = req->dest_vfid; __entry->code = le16_to_cpu(req->msg.code); - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->vport[0].nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_pf_to_vf_cmd)); ), @@ -101,7 +101,7 @@ DECLARE_EVENT_CLASS(hclge_pf_cmd_template, __entry->rsv = le16_to_cpu(desc->rsv); __entry->index = index; __entry->num = num; - __assign_str(pciname, pci_name(hw->cmq.csq.pdev)); + __assign_str(pciname); for (i = 0; i < HCLGE_DESC_DATA_LEN; i++) __entry->data[i] = le32_to_cpu(desc->data[i]);), @@ -144,7 +144,7 @@ DECLARE_EVENT_CLASS(hclge_pf_special_cmd_template, TP_fast_assign(int i; __entry->index = index; __entry->num = num; - __assign_str(pciname, pci_name(hw->cmq.csq.pdev)); + __assign_str(pciname); for (i = 0; i < PF_DESC_LEN; i++) __entry->data[i] = le32_to_cpu(data[i]); ), diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h index e2e3a2602b6a..66b084309c91 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h @@ -30,8 +30,8 @@ TRACE_EVENT(hclge_vf_mbx_get, TP_fast_assign( __entry->vfid = req->dest_vfid; __entry->code = le16_to_cpu(req->msg.code); - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_pf_to_vf_cmd)); ), @@ -63,8 +63,8 @@ TRACE_EVENT(hclge_vf_mbx_send, __entry->vfid = req->mbx_src_vfid; __entry->code = req->msg.code; __entry->subcode = req->msg.subcode; - __assign_str(pciname, pci_name(hdev->pdev)); - __assign_str(devname, hdev->nic.kinfo.netdev->name); + __assign_str(pciname); + __assign_str(devname); memcpy(__entry->mbx_data, req, sizeof(struct hclge_mbx_vf_to_pf_cmd)); ), @@ -101,7 +101,7 @@ DECLARE_EVENT_CLASS(hclge_vf_cmd_template, __entry->rsv = le16_to_cpu(desc->rsv); __entry->index = index; __entry->num = num; - __assign_str(pciname, pci_name(hw->cmq.csq.pdev)); + __assign_str(pciname); for (i = 0; i < HCLGE_DESC_DATA_LEN; i++) __entry->data[i] = le32_to_cpu(desc->data[i]);), diff --git a/drivers/net/ethernet/intel/i40e/i40e_trace.h b/drivers/net/ethernet/intel/i40e/i40e_trace.h index 33b4e30f5e00..759f3d1c4c8f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_trace.h +++ b/drivers/net/ethernet/intel/i40e/i40e_trace.h @@ -89,8 +89,8 @@ TRACE_EVENT(i40e_napi_poll, __entry->tx_clean_complete = tx_clean_complete; __entry->irq_num = q->irq_num; __entry->curr_cpu = get_cpu(); - __assign_str(qname, q->name); - __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); + __assign_str(qname); + __assign_str(dev_name); __assign_bitmask(irq_affinity, cpumask_bits(&q->affinity_mask), nr_cpumask_bits); ), @@ -132,7 +132,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->buf = buf; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -177,7 +177,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->xdp = xdp; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -219,7 +219,7 @@ DECLARE_EVENT_CLASS( TP_fast_assign( __entry->skb = skb; __entry->ring = ring; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( diff --git a/drivers/net/ethernet/intel/iavf/iavf_trace.h b/drivers/net/ethernet/intel/iavf/iavf_trace.h index 82fda6f5abf0..62212011c807 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_trace.h +++ b/drivers/net/ethernet/intel/iavf/iavf_trace.h @@ -83,7 +83,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->buf = buf; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -128,7 +128,7 @@ DECLARE_EVENT_CLASS( __entry->ring = ring; __entry->desc = desc; __entry->skb = skb; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( @@ -170,7 +170,7 @@ DECLARE_EVENT_CLASS( TP_fast_assign( __entry->skb = skb; __entry->ring = ring; - __assign_str(devname, ring->netdev->name); + __assign_str(devname); ), TP_printk( diff --git a/drivers/net/ethernet/intel/ice/ice_trace.h b/drivers/net/ethernet/intel/ice/ice_trace.h index b2f5c9fe0149..244cddd2a9ea 100644 --- a/drivers/net/ethernet/intel/ice/ice_trace.h +++ b/drivers/net/ethernet/intel/ice/ice_trace.h @@ -69,7 +69,7 @@ DECLARE_EVENT_CLASS(ice_rx_dim_template, TP_fast_assign(__entry->q_vector = q_vector; __entry->dim = dim; - __assign_str(devname, q_vector->rx.rx_ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s Rx-Q: %d dim-state: %d dim-profile: %d dim-tune: %d dim-st-right: %d dim-st-left: %d dim-tired: %d", __get_str(devname), @@ -96,7 +96,7 @@ DECLARE_EVENT_CLASS(ice_tx_dim_template, TP_fast_assign(__entry->q_vector = q_vector; __entry->dim = dim; - __assign_str(devname, q_vector->tx.tx_ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s Tx-Q: %d dim-state: %d dim-profile: %d dim-tune: %d dim-st-right: %d dim-st-left: %d dim-tired: %d", __get_str(devname), @@ -128,7 +128,7 @@ DECLARE_EVENT_CLASS(ice_tx_template, TP_fast_assign(__entry->ring = ring; __entry->desc = desc; __entry->buf = buf; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s ring: %pK desc: %pK buf %pK", __get_str(devname), __entry->ring, __entry->desc, __entry->buf) @@ -156,7 +156,7 @@ DECLARE_EVENT_CLASS(ice_rx_template, TP_fast_assign(__entry->ring = ring; __entry->desc = desc; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s ring: %pK desc: %pK", __get_str(devname), __entry->ring, __entry->desc) @@ -180,7 +180,7 @@ DECLARE_EVENT_CLASS(ice_rx_indicate_template, TP_fast_assign(__entry->ring = ring; __entry->desc = desc; __entry->skb = skb; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s ring: %pK desc: %pK skb %pK", __get_str(devname), __entry->ring, __entry->desc, __entry->skb) @@ -203,7 +203,7 @@ DECLARE_EVENT_CLASS(ice_xmit_template, TP_fast_assign(__entry->ring = ring; __entry->skb = skb; - __assign_str(devname, ring->netdev->name);), + __assign_str(devname);), TP_printk("netdev: %s skb: %pK ring: %pK", __get_str(devname), __entry->skb, __entry->ring) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h index 28984d0e848a..5704520f9b02 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_trace.h @@ -24,7 +24,7 @@ TRACE_EVENT(otx2_msg_alloc, __field(u16, id) __field(u64, size) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->id = id; __entry->size = size; ), @@ -39,7 +39,7 @@ TRACE_EVENT(otx2_msg_send, __field(u16, num_msgs) __field(u64, msg_size) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->num_msgs = num_msgs; __entry->msg_size = msg_size; ), @@ -55,7 +55,7 @@ TRACE_EVENT(otx2_msg_check, __field(u16, rspid) __field(int, rc) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->reqid = reqid; __entry->rspid = rspid; __entry->rc = rc; @@ -72,8 +72,8 @@ TRACE_EVENT(otx2_msg_interrupt, __string(str, msg) __field(u64, intr) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); - __assign_str(str, msg); + TP_fast_assign(__assign_str(dev); + __assign_str(str); __entry->intr = intr; ), TP_printk("[%s] mbox interrupt %s (0x%llx)\n", __get_str(dev), @@ -87,7 +87,7 @@ TRACE_EVENT(otx2_msg_process, __field(u16, id) __field(int, err) ), - TP_fast_assign(__assign_str(dev, pci_name(pdev)); + TP_fast_assign(__assign_str(dev); __entry->id = id; __entry->err = err; ), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h index 406ebe17405f..b4b3a43e56a0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/cmd_tracepoint.h @@ -22,10 +22,10 @@ TRACE_EVENT(mlx5_cmd, __field(u32, syndrome) __field(int, err) ), - TP_fast_assign(__assign_str(command_str, command_str); + TP_fast_assign(__assign_str(command_str); __entry->opcode = opcode; __entry->op_mod = op_mod; - __assign_str(status_str, status_str); + __assign_str(status_str); __entry->status = status; __entry->syndrome = syndrome; __entry->err = err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h index f15718db5d0e..78e481b2c015 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_rep_tracepoint.h @@ -25,7 +25,7 @@ TRACE_EVENT(mlx5e_rep_neigh_update, struct in6_addr *pin6; __be32 *p32; - __assign_str(devname, nhe->neigh_dev->name); + __assign_str(devname); __entry->neigh_connected = neigh_connected; memcpy(__entry->ha, ha, ETH_ALEN); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h index ac52ef37f38a..4b1ca228012b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/en_tc_tracepoint.h @@ -86,7 +86,7 @@ TRACE_EVENT(mlx5e_tc_update_neigh_used_value, struct in6_addr *pin6; __be32 *p32; - __assign_str(devname, nhe->neigh_dev->name); + __assign_str(devname); __entry->neigh_used = neigh_used; p32 = (__be32 *)__entry->v4; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h index 3038be575923..50f8a7630f86 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h @@ -55,12 +55,11 @@ TRACE_EVENT(mlx5_fw, ), TP_fast_assign( - __assign_str(dev_name, - dev_name(tracer->dev->device)); + __assign_str(dev_name); __entry->trace_timestamp = trace_timestamp; __entry->lost = lost; __entry->event_id = event_id; - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("%s [0x%llx] %d [0x%x] %s", diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h index 458baf0c6415..1ce332f21ebe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/diag/qos_tracepoint.h @@ -17,7 +17,7 @@ TRACE_EVENT(mlx5_esw_vport_qos_destroy, __field(unsigned short, vport_id) __field(unsigned int, tsar_ix) ), - TP_fast_assign(__assign_str(devname, dev_name(vport->dev->device)); + TP_fast_assign(__assign_str(devname); __entry->vport_id = vport->vport; __entry->tsar_ix = vport->qos.esw_tsar_ix; ), @@ -36,7 +36,7 @@ DECLARE_EVENT_CLASS(mlx5_esw_vport_qos_template, __field(unsigned int, max_rate) __field(void *, group) ), - TP_fast_assign(__assign_str(devname, dev_name(vport->dev->device)); + TP_fast_assign(__assign_str(devname); __entry->vport_id = vport->vport; __entry->tsar_ix = vport->qos.esw_tsar_ix; __entry->bw_share = bw_share; @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(mlx5_esw_group_qos_template, __field(const void *, group) __field(unsigned int, tsar_ix) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->group = group; __entry->tsar_ix = tsar_ix; ), @@ -102,7 +102,7 @@ TRACE_EVENT(mlx5_esw_group_qos_config, __field(unsigned int, bw_share) __field(unsigned int, max_rate) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->group = group; __entry->tsar_ix = tsar_ix; __entry->bw_share = bw_share; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h index 7f7c9af5deed..0537de86f981 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/diag/dev_tracepoint.h @@ -22,7 +22,7 @@ DECLARE_EVENT_CLASS(mlx5_sf_dev_template, __field(u16, hw_fn_id) __field(u32, sfnum) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->sfdev = sfdev; __entry->aux_id = aux_id; __entry->hw_fn_id = sfdev->fn_id; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h index 8bf1cd90930d..302ce00da5a9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/sf_tracepoint.h @@ -24,7 +24,7 @@ TRACE_EVENT(mlx5_sf_add, __field(u16, hw_fn_id) __field(u32, sfnum) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; @@ -46,7 +46,7 @@ TRACE_EVENT(mlx5_sf_free, __field(u32, controller) __field(u16, hw_fn_id) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; @@ -67,7 +67,7 @@ TRACE_EVENT(mlx5_sf_hwc_alloc, __field(u16, hw_fn_id) __field(u32, sfnum) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; __entry->sfnum = sfnum; @@ -84,7 +84,7 @@ TRACE_EVENT(mlx5_sf_hwc_free, TP_STRUCT__entry(__string(devname, dev_name(dev->device)) __field(u16, hw_fn_id) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->hw_fn_id = hw_fn_id; ), TP_printk("(%s) hw_id=0x%x\n", __get_str(devname), __entry->hw_fn_id) @@ -97,7 +97,7 @@ TRACE_EVENT(mlx5_sf_hwc_deferred_free, TP_STRUCT__entry(__string(devname, dev_name(dev->device)) __field(u16, hw_fn_id) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->hw_fn_id = hw_fn_id; ), TP_printk("(%s) hw_id=0x%x\n", __get_str(devname), __entry->hw_fn_id) @@ -113,7 +113,7 @@ DECLARE_EVENT_CLASS(mlx5_sf_state_template, __field(unsigned int, port_index) __field(u32, controller) __field(u16, hw_fn_id)), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; @@ -152,7 +152,7 @@ TRACE_EVENT(mlx5_sf_update_state, __field(u16, hw_fn_id) __field(u8, state) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->port_index = port_index; __entry->controller = controller; __entry->hw_fn_id = hw_fn_id; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h index fd814a190b8b..6352cb004a18 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/diag/vhca_tracepoint.h @@ -20,7 +20,7 @@ TRACE_EVENT(mlx5_sf_vhca_event, __field(u32, sfnum) __field(u8, vhca_state) ), - TP_fast_assign(__assign_str(devname, dev_name(dev->device)); + TP_fast_assign(__assign_str(devname); __entry->hw_fn_id = event->function_id; __entry->sfnum = event->sw_function_id; __entry->vhca_state = event->new_vhca_state; diff --git a/drivers/net/fjes/fjes_trace.h b/drivers/net/fjes/fjes_trace.h index 6437ddbd7842..166ef015262b 100644 --- a/drivers/net/fjes/fjes_trace.h +++ b/drivers/net/fjes/fjes_trace.h @@ -85,7 +85,7 @@ TRACE_EVENT(fjes_hw_request_info_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -145,7 +145,7 @@ TRACE_EVENT(fjes_hw_register_buff_addr_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -189,7 +189,7 @@ TRACE_EVENT(fjes_hw_unregister_buff_addr_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -232,7 +232,7 @@ TRACE_EVENT(fjes_hw_start_debug_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); @@ -258,7 +258,7 @@ TRACE_EVENT(fjes_hw_stop_debug_err, __string(err, err) ), TP_fast_assign( - __assign_str(err, err); + __assign_str(err); ), TP_printk("%s", __get_str(err)) ); diff --git a/drivers/net/hyperv/netvsc_trace.h b/drivers/net/hyperv/netvsc_trace.h index f7585563dea5..05e620cbdd29 100644 --- a/drivers/net/hyperv/netvsc_trace.h +++ b/drivers/net/hyperv/netvsc_trace.h @@ -51,7 +51,7 @@ DECLARE_EVENT_CLASS(rndis_msg_class, __field( u32, msg_len ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->queue = q; __entry->req_id = msg->msg.init_req.req_id; __entry->msg_type = msg->ndis_msg_type; @@ -121,7 +121,7 @@ TRACE_EVENT(nvsp_send, __field( u32, msg_type ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->msg_type = msg->hdr.msg_type; ), TP_printk("dev=%s type=%s", @@ -142,7 +142,7 @@ TRACE_EVENT(nvsp_send_pkt, __field( u32, section_size ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->qid = chan->offermsg.offer.sub_channel_index; __entry->channel_type = rpkt->channel_type; __entry->section_index = rpkt->send_buf_section_index; @@ -165,7 +165,7 @@ TRACE_EVENT(nvsp_recv, __field( u32, msg_type ) ), TP_fast_assign( - __assign_str(name, ndev->name); + __assign_str(name); __entry->qid = chan->offermsg.offer.sub_channel_index; __entry->msg_type = msg->hdr.msg_type; ), diff --git a/drivers/net/wireless/ath/ath10k/trace.h b/drivers/net/wireless/ath/ath10k/trace.h index 64e7a767d963..68b78ca17eaa 100644 --- a/drivers/net/wireless/ath/ath10k/trace.h +++ b/drivers/net/wireless/ath/ath10k/trace.h @@ -55,8 +55,8 @@ DECLARE_EVENT_CLASS(ath10k_log_event, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk( @@ -92,8 +92,8 @@ TRACE_EVENT(ath10k_log_dbg, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->level = level; __assign_vstr(msg, vaf->fmt, vaf->va); ), @@ -121,10 +121,10 @@ TRACE_EVENT(ath10k_log_dbg_dump, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); - __assign_str(msg, msg); - __assign_str(prefix, prefix); + __assign_str(device); + __assign_str(driver); + __assign_str(msg); + __assign_str(prefix); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), @@ -152,8 +152,8 @@ TRACE_EVENT(ath10k_wmi_cmd, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -182,8 +182,8 @@ TRACE_EVENT(ath10k_wmi_event, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -211,8 +211,8 @@ TRACE_EVENT(ath10k_htt_stats, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), @@ -239,8 +239,8 @@ TRACE_EVENT(ath10k_wmi_dbglog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->hw_type = ar->hw_rev; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -269,8 +269,8 @@ TRACE_EVENT(ath10k_htt_pktlog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->hw_type = ar->hw_rev; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(pktlog), buf, buf_len); @@ -301,8 +301,8 @@ TRACE_EVENT(ath10k_htt_tx, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->msdu_id = msdu_id; __entry->msdu_len = msdu_len; __entry->vdev_id = vdev_id; @@ -332,8 +332,8 @@ TRACE_EVENT(ath10k_txrx_tx_unref, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->msdu_id = msdu_id; ), @@ -358,8 +358,8 @@ DECLARE_EVENT_CLASS(ath10k_hdr_event, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = ath10k_frm_hdr_len(data, len); memcpy(__get_dynamic_array(data), data, __entry->len); ), @@ -386,8 +386,8 @@ DECLARE_EVENT_CLASS(ath10k_payload_event, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len - ath10k_frm_hdr_len(data, len); memcpy(__get_dynamic_array(payload), data + ath10k_frm_hdr_len(data, len), __entry->len); @@ -435,8 +435,8 @@ TRACE_EVENT(ath10k_htt_rx_desc, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->hw_type = ar->hw_rev; __entry->len = len; memcpy(__get_dynamic_array(rxdesc), data, len); @@ -472,8 +472,8 @@ TRACE_EVENT(ath10k_wmi_diag_container, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->type = type; __entry->timestamp = timestamp; __entry->code = code; @@ -505,8 +505,8 @@ TRACE_EVENT(ath10k_wmi_diag, ), TP_fast_assign( - __assign_str(device, dev_name(ar->dev)); - __assign_str(driver, dev_driver_string(ar->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(data), data, len); ), diff --git a/drivers/net/wireless/ath/ath11k/trace.h b/drivers/net/wireless/ath/ath11k/trace.h index 235ab8ea715f..75246b0a82e3 100644 --- a/drivers/net/wireless/ath/ath11k/trace.h +++ b/drivers/net/wireless/ath/ath11k/trace.h @@ -48,8 +48,8 @@ TRACE_EVENT(ath11k_htt_pktlog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->buf_len = buf_len; __entry->pktlog_checksum = pktlog_checksum; memcpy(__get_dynamic_array(pktlog), buf, buf_len); @@ -77,8 +77,8 @@ TRACE_EVENT(ath11k_htt_ppdu_stats, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(ppdu), data, len); ), @@ -105,8 +105,8 @@ TRACE_EVENT(ath11k_htt_rxdesc, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; __entry->log_type = log_type; memcpy(__get_dynamic_array(rxdesc), data, len); @@ -130,8 +130,8 @@ DECLARE_EVENT_CLASS(ath11k_log_event, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk( @@ -171,8 +171,8 @@ TRACE_EVENT(ath11k_wmi_cmd, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -201,8 +201,8 @@ TRACE_EVENT(ath11k_wmi_event, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->id = id; __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); @@ -230,8 +230,8 @@ TRACE_EVENT(ath11k_log_dbg, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->level = level; WARN_ON_ONCE(vsnprintf(__get_dynamic_array(msg), ATH11K_MSG_MAX, vaf->fmt, @@ -262,10 +262,10 @@ TRACE_EVENT(ath11k_log_dbg_dump, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); - __assign_str(msg, msg); - __assign_str(prefix, prefix); + __assign_str(device); + __assign_str(driver); + __assign_str(msg); + __assign_str(prefix); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), @@ -292,8 +292,8 @@ TRACE_EVENT(ath11k_wmi_diag, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(data), data, len); ), @@ -318,8 +318,8 @@ TRACE_EVENT(ath11k_ps_timekeeper, __field(u32, peer_ps_timestamp) ), - TP_fast_assign(__assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + TP_fast_assign(__assign_str(device); + __assign_str(driver); memcpy(__get_dynamic_array(peer_addr), peer_addr, ETH_ALEN); __entry->peer_ps_state = peer_ps_state; diff --git a/drivers/net/wireless/ath/ath12k/trace.h b/drivers/net/wireless/ath/ath12k/trace.h index 240737e1542d..253c67accb0e 100644 --- a/drivers/net/wireless/ath/ath12k/trace.h +++ b/drivers/net/wireless/ath/ath12k/trace.h @@ -36,8 +36,8 @@ TRACE_EVENT(ath12k_htt_pktlog, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->buf_len = buf_len; __entry->pktlog_checksum = pktlog_checksum; memcpy(__get_dynamic_array(pktlog), buf, buf_len); @@ -73,8 +73,8 @@ TRACE_EVENT(ath12k_htt_ppdu_stats, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; __entry->info = ar->pdev->timestamp.info; __entry->sync_tstmp_lo_us = ar->pdev->timestamp.sync_timestamp_hi_us; @@ -117,8 +117,8 @@ TRACE_EVENT(ath12k_htt_rxdesc, ), TP_fast_assign( - __assign_str(device, dev_name(ar->ab->dev)); - __assign_str(driver, dev_driver_string(ar->ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; __entry->type = type; __entry->info = ar->pdev->timestamp.info; @@ -153,8 +153,8 @@ TRACE_EVENT(ath12k_wmi_diag, ), TP_fast_assign( - __assign_str(device, dev_name(ab->dev)); - __assign_str(driver, dev_driver_string(ab->dev)); + __assign_str(device); + __assign_str(driver); __entry->len = len; memcpy(__get_dynamic_array(data), data, len); ), diff --git a/drivers/net/wireless/ath/ath6kl/trace.h b/drivers/net/wireless/ath/ath6kl/trace.h index 231a94769ddb..8577aa459c58 100644 --- a/drivers/net/wireless/ath/ath6kl/trace.h +++ b/drivers/net/wireless/ath/ath6kl/trace.h @@ -304,8 +304,8 @@ TRACE_EVENT(ath6kl_log_dbg_dump, ), TP_fast_assign( - __assign_str(msg, msg); - __assign_str(prefix, prefix); + __assign_str(msg); + __assign_str(prefix); __entry->buf_len = buf_len; memcpy(__get_dynamic_array(buf), buf, buf_len); ), diff --git a/drivers/net/wireless/ath/trace.h b/drivers/net/wireless/ath/trace.h index 9935cf475b6d..82aac0a4baff 100644 --- a/drivers/net/wireless/ath/trace.h +++ b/drivers/net/wireless/ath/trace.h @@ -44,8 +44,8 @@ TRACE_EVENT(ath_log, ), TP_fast_assign( - __assign_str(device, wiphy_name(wiphy)); - __assign_str(driver, KBUILD_MODNAME); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h index 5d66e94c806d..96032322b165 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/tracepoint.h @@ -41,7 +41,7 @@ TRACE_EVENT(brcmf_err, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(func, func); + __assign_str(func); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(func), __get_str(msg)) @@ -57,7 +57,7 @@ TRACE_EVENT(brcmf_dbg, ), TP_fast_assign( __entry->level = level; - __assign_str(func, func); + __assign_str(func); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(func), __get_str(msg)) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h index a0da3248b942..53b3dba50737 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac.h @@ -81,7 +81,7 @@ TRACE_EVENT(brcms_macintstatus, __field(u32, mask) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->in_isr = in_isr; __entry->macintstatus = macintstatus; __entry->mask = mask; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h index 42b0a91656c4..908ce3c864fe 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_msg.h @@ -71,7 +71,7 @@ TRACE_EVENT(brcms_dbg, ), TP_fast_assign( __entry->level = level; - __assign_str(func, func); + __assign_str(func); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(func), __get_str(msg)) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h index cf2cc070f1e5..24ac34fa0207 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/brcms_trace_brcmsmac_tx.h @@ -31,7 +31,7 @@ TRACE_EVENT(brcms_txdesc, __dynamic_array(u8, txh, txh_len) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); memcpy(__get_dynamic_array(txh), txh, txh_len); ), TP_printk("[%s] txdesc", __get_str(dev)) @@ -54,7 +54,7 @@ TRACE_EVENT(brcms_txstatus, __field(u16, ackphyrxsh) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->framelen = framelen; __entry->frameid = frameid; __entry->status = status; @@ -85,7 +85,7 @@ TRACE_EVENT(brcms_ampdu_session, __field(u16, dma_len) ), TP_fast_assign( - __assign_str(dev, dev_name(dev)); + __assign_str(dev); __entry->max_ampdu_len = max_ampdu_len; __entry->max_ampdu_frames = max_ampdu_frames; __entry->ampdu_len = ampdu_len; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h index 1d6c292cf545..0db1fa5477af 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-msg.h @@ -57,7 +57,7 @@ TRACE_EVENT(iwlwifi_dbg, ), TP_fast_assign( __entry->level = level; - __assign_str(function, function); + __assign_str(function); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s", __get_str(msg)) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h index c3e09f4fefeb..76166e1b10e5 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h @@ -87,7 +87,7 @@ static inline void trace_ ## name(proto) {} #endif #define DEV_ENTRY __string(dev, dev_name(dev)) -#define DEV_ASSIGN __assign_str(dev, dev_name(dev)) +#define DEV_ASSIGN __assign_str(dev) #include "iwl-devtrace-io.h" #include "iwl-devtrace-ucode.h" diff --git a/drivers/soc/qcom/pmic_pdcharger_ulog.h b/drivers/soc/qcom/pmic_pdcharger_ulog.h index 152e3a6b5480..1cfa58f0e34c 100644 --- a/drivers/soc/qcom/pmic_pdcharger_ulog.h +++ b/drivers/soc/qcom/pmic_pdcharger_ulog.h @@ -18,7 +18,7 @@ TRACE_EVENT(pmic_pdcharger_ulog_msg, __string(msg, msg) ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("%s", __get_str(msg)) ); diff --git a/drivers/soc/qcom/trace-aoss.h b/drivers/soc/qcom/trace-aoss.h index 554029b33b44..fb5b0470c40d 100644 --- a/drivers/soc/qcom/trace-aoss.h +++ b/drivers/soc/qcom/trace-aoss.h @@ -18,7 +18,7 @@ TRACE_EVENT(aoss_send, __string(msg, msg) ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("%s", __get_str(msg)) ); @@ -31,7 +31,7 @@ TRACE_EVENT(aoss_send_done, __field(int, ret) ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); __entry->ret = ret; ), TP_printk("%s: %d", __get_str(msg), __entry->ret) diff --git a/drivers/soc/qcom/trace-rpmh.h b/drivers/soc/qcom/trace-rpmh.h index be6b42ecc1f8..593ec1d4e010 100644 --- a/drivers/soc/qcom/trace-rpmh.h +++ b/drivers/soc/qcom/trace-rpmh.h @@ -26,7 +26,7 @@ TRACE_EVENT(rpmh_tx_done, ), TP_fast_assign( - __assign_str(name, d->name); + __assign_str(name); __entry->m = m; __entry->addr = r->cmds[0].addr; __entry->data = r->cmds[0].data; @@ -55,7 +55,7 @@ TRACE_EVENT(rpmh_send_msg, ), TP_fast_assign( - __assign_str(name, d->name); + __assign_str(name); __entry->m = m; __entry->state = state; __entry->n = n; diff --git a/drivers/thermal/thermal_trace.h b/drivers/thermal/thermal_trace.h index 88a962f560f2..df8f4edd6068 100644 --- a/drivers/thermal/thermal_trace.h +++ b/drivers/thermal/thermal_trace.h @@ -37,7 +37,7 @@ TRACE_EVENT(thermal_temperature, ), TP_fast_assign( - __assign_str(thermal_zone, tz->type); + __assign_str(thermal_zone); __entry->id = tz->id; __entry->temp_prev = tz->last_temperature; __entry->temp = tz->temperature; @@ -60,7 +60,7 @@ TRACE_EVENT(cdev_update, ), TP_fast_assign( - __assign_str(type, cdev->type); + __assign_str(type); __entry->target = target; ), @@ -82,7 +82,7 @@ TRACE_EVENT(thermal_zone_trip, ), TP_fast_assign( - __assign_str(thermal_zone, tz->type); + __assign_str(thermal_zone); __entry->id = tz->id; __entry->trip = trip; __entry->trip_type = trip_type; @@ -156,7 +156,7 @@ TRACE_EVENT(thermal_power_devfreq_get_power, ), TP_fast_assign( - __assign_str(type, cdev->type); + __assign_str(type); __entry->freq = freq; __entry->busy_time = status->busy_time; __entry->total_time = status->total_time; @@ -184,7 +184,7 @@ TRACE_EVENT(thermal_power_devfreq_limit, ), TP_fast_assign( - __assign_str(type, cdev->type); + __assign_str(type); __entry->freq = freq; __entry->cdev_state = cdev_state; __entry->power = power; diff --git a/drivers/usb/cdns3/cdns3-trace.h b/drivers/usb/cdns3/cdns3-trace.h index 40db89e3333c..c4e542f1b9b7 100644 --- a/drivers/usb/cdns3/cdns3-trace.h +++ b/drivers/usb/cdns3/cdns3-trace.h @@ -33,7 +33,7 @@ TRACE_EVENT(cdns3_halt, __field(u8, flush) ), TP_fast_assign( - __assign_str(name, ep_priv->name); + __assign_str(name); __entry->halt = halt; __entry->flush = flush; ), @@ -49,8 +49,8 @@ TRACE_EVENT(cdns3_wa1, __string(msg, msg) ), TP_fast_assign( - __assign_str(ep_name, ep_priv->name); - __assign_str(msg, msg); + __assign_str(ep_name); + __assign_str(msg); ), TP_printk("WA1: %s %s", __get_str(ep_name), __get_str(msg)) ); @@ -63,8 +63,8 @@ TRACE_EVENT(cdns3_wa2, __string(msg, msg) ), TP_fast_assign( - __assign_str(ep_name, ep_priv->name); - __assign_str(msg, msg); + __assign_str(ep_name); + __assign_str(msg); ), TP_printk("WA2: %s %s", __get_str(ep_name), __get_str(msg)) ); @@ -77,7 +77,7 @@ DECLARE_EVENT_CLASS(cdns3_log_doorbell, __field(u32, ep_trbaddr) ), TP_fast_assign( - __assign_str(name, ep_name); + __assign_str(name); __entry->ep_trbaddr = ep_trbaddr; ), TP_printk("%s, ep_trbaddr %08x", __get_str(name), @@ -125,7 +125,7 @@ DECLARE_EVENT_CLASS(cdns3_log_epx_irq, __field(u32, use_streams) ), TP_fast_assign( - __assign_str(ep_name, priv_ep->name); + __assign_str(ep_name); __entry->ep_sts = readl(&priv_dev->regs->ep_sts); __entry->ep_traddr = readl(&priv_dev->regs->ep_traddr); __entry->ep_last_sid = priv_ep->last_stream_id; @@ -214,7 +214,7 @@ DECLARE_EVENT_CLASS(cdns3_log_request, __field(unsigned int, stream_id) ), TP_fast_assign( - __assign_str(name, req->priv_ep->name); + __assign_str(name); __entry->req = req; __entry->buf = req->request.buf; __entry->actual = req->request.actual; @@ -294,7 +294,7 @@ DECLARE_EVENT_CLASS(cdns3_stream_split_transfer_len, __field(unsigned int, stream_id) ), TP_fast_assign( - __assign_str(name, req->priv_ep->name); + __assign_str(name); __entry->req = req; __entry->actual = req->request.length; __entry->length = req->request.actual; @@ -329,7 +329,7 @@ DECLARE_EVENT_CLASS(cdns3_log_aligned_request, __field(u32, aligned_buf_size) ), TP_fast_assign( - __assign_str(name, priv_req->priv_ep->name); + __assign_str(name); __entry->req = &priv_req->request; __entry->buf = priv_req->request.buf; __entry->dma = priv_req->request.dma; @@ -364,7 +364,7 @@ DECLARE_EVENT_CLASS(cdns3_log_map_request, __field(dma_addr_t, dma) ), TP_fast_assign( - __assign_str(name, priv_req->priv_ep->name); + __assign_str(name); __entry->req = &priv_req->request; __entry->buf = priv_req->request.buf; __entry->dma = priv_req->request.dma; @@ -395,7 +395,7 @@ DECLARE_EVENT_CLASS(cdns3_log_trb, __field(unsigned int, last_stream_id) ), TP_fast_assign( - __assign_str(name, priv_ep->name); + __assign_str(name); __entry->trb = trb; __entry->buffer = le32_to_cpu(trb->buffer); __entry->length = le32_to_cpu(trb->length); @@ -467,7 +467,7 @@ DECLARE_EVENT_CLASS(cdns3_log_ep, __field(u8, dequeue) ), TP_fast_assign( - __assign_str(name, priv_ep->name); + __assign_str(name); __entry->maxpacket = priv_ep->endpoint.maxpacket; __entry->maxpacket_limit = priv_ep->endpoint.maxpacket_limit; __entry->max_streams = priv_ep->endpoint.max_streams; diff --git a/drivers/usb/cdns3/cdnsp-trace.h b/drivers/usb/cdns3/cdnsp-trace.h index 4b51011eb00b..f2bcf77a5d0a 100644 --- a/drivers/usb/cdns3/cdnsp-trace.h +++ b/drivers/usb/cdns3/cdnsp-trace.h @@ -48,7 +48,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_ep, __field(u8, drbls_count) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->state = pep->ep_state; __entry->stream_id = stream_id; __entry->enabled = pep->ep_state & EP_HAS_STREAMS; @@ -138,7 +138,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_simple, __string(text, msg) ), TP_fast_assign( - __assign_str(text, msg); + __assign_str(text); ), TP_printk("%s", __get_str(text)) ); @@ -303,7 +303,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_bounce, __field(unsigned int, unalign) ), TP_fast_assign( - __assign_str(name, preq->pep->name); + __assign_str(name); __entry->new_buf_len = new_buf_len; __entry->offset = offset; __entry->dma = dma; @@ -470,7 +470,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_request, ), TP_fast_assign( - __assign_str(name, req->pep->name); + __assign_str(name); __entry->request = &req->request; __entry->preq = req; __entry->buf = req->request.buf; @@ -674,7 +674,7 @@ DECLARE_EVENT_CLASS(cdnsp_log_td_info, __field(dma_addr_t, trb_dma) ), TP_fast_assign( - __assign_str(name, preq->pep->name); + __assign_str(name); __entry->request = &preq->request; __entry->preq = preq; __entry->first_trb = preq->td.first_trb; diff --git a/drivers/usb/chipidea/trace.h b/drivers/usb/chipidea/trace.h index ca0e65b48f0a..1875419cd17f 100644 --- a/drivers/usb/chipidea/trace.h +++ b/drivers/usb/chipidea/trace.h @@ -31,7 +31,7 @@ TRACE_EVENT(ci_log, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(name, dev_name(ci->dev)); + __assign_str(name); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(name), __get_str(msg)) @@ -51,7 +51,7 @@ DECLARE_EVENT_CLASS(ci_log_trb, __field(u32, type) ), TP_fast_assign( - __assign_str(name, hwep->name); + __assign_str(name); __entry->req = &hwreq->req; __entry->td = td; __entry->dma = td->dma; diff --git a/drivers/usb/dwc3/trace.h b/drivers/usb/dwc3/trace.h index d2997d17cfbe..bdeb1aaf65d8 100644 --- a/drivers/usb/dwc3/trace.h +++ b/drivers/usb/dwc3/trace.h @@ -112,7 +112,7 @@ DECLARE_EVENT_CLASS(dwc3_log_request, __field(int, no_interrupt) ), TP_fast_assign( - __assign_str(name, req->dep->name); + __assign_str(name); __entry->req = req; __entry->actual = req->request.actual; __entry->length = req->request.length; @@ -193,7 +193,7 @@ DECLARE_EVENT_CLASS(dwc3_log_gadget_ep_cmd, __field(int, cmd_status) ), TP_fast_assign( - __assign_str(name, dep->name); + __assign_str(name); __entry->cmd = cmd; __entry->param0 = params->param0; __entry->param1 = params->param1; @@ -229,7 +229,7 @@ DECLARE_EVENT_CLASS(dwc3_log_trb, __field(u32, dequeue) ), TP_fast_assign( - __assign_str(name, dep->name); + __assign_str(name); __entry->trb = trb; __entry->bpl = trb->bpl; __entry->bph = trb->bph; @@ -301,7 +301,7 @@ DECLARE_EVENT_CLASS(dwc3_log_ep, __field(u8, trb_dequeue) ), TP_fast_assign( - __assign_str(name, dep->name); + __assign_str(name); __entry->maxpacket = dep->endpoint.maxpacket; __entry->maxpacket_limit = dep->endpoint.maxpacket_limit; __entry->max_streams = dep->endpoint.max_streams; diff --git a/drivers/usb/gadget/udc/cdns2/cdns2-trace.h b/drivers/usb/gadget/udc/cdns2/cdns2-trace.h index 61f241634ea5..ade1752956b1 100644 --- a/drivers/usb/gadget/udc/cdns2/cdns2-trace.h +++ b/drivers/usb/gadget/udc/cdns2/cdns2-trace.h @@ -64,7 +64,7 @@ DECLARE_EVENT_CLASS(cdns2_log_simple, __string(text, msg) ), TP_fast_assign( - __assign_str(text, msg); + __assign_str(text); ), TP_printk("%s", __get_str(text)) ); @@ -103,7 +103,7 @@ TRACE_EVENT(cdns2_ep_halt, __field(u8, flush) ), TP_fast_assign( - __assign_str(name, ep_priv->name); + __assign_str(name); __entry->halt = halt; __entry->flush = flush; ), @@ -119,8 +119,8 @@ TRACE_EVENT(cdns2_wa1, __string(msg, msg) ), TP_fast_assign( - __assign_str(ep_name, ep_priv->name); - __assign_str(msg, msg); + __assign_str(ep_name); + __assign_str(msg); ), TP_printk("WA1: %s %s", __get_str(ep_name), __get_str(msg)) ); @@ -134,7 +134,7 @@ DECLARE_EVENT_CLASS(cdns2_log_doorbell, __field(u32, ep_trbaddr) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->ep_trbaddr = ep_trbaddr; ), TP_printk("%s, ep_trbaddr %08x", __get_str(name), @@ -196,7 +196,7 @@ DECLARE_EVENT_CLASS(cdns2_log_epx_irq, __field(u32, ep_traddr) ), TP_fast_assign( - __assign_str(ep_name, pep->name); + __assign_str(ep_name); __entry->ep_sts = readl(&pdev->adma_regs->ep_sts); __entry->ep_ists = readl(&pdev->adma_regs->ep_ists); __entry->ep_traddr = readl(&pdev->adma_regs->ep_traddr); @@ -288,7 +288,7 @@ DECLARE_EVENT_CLASS(cdns2_log_request, __field(int, end_trb) ), TP_fast_assign( - __assign_str(name, preq->pep->name); + __assign_str(name); __entry->request = &preq->request; __entry->preq = preq; __entry->buf = preq->request.buf; @@ -380,7 +380,7 @@ DECLARE_EVENT_CLASS(cdns2_log_map_request, __field(dma_addr_t, dma) ), TP_fast_assign( - __assign_str(name, priv_req->pep->name); + __assign_str(name); __entry->req = &priv_req->request; __entry->buf = priv_req->request.buf; __entry->dma = priv_req->request.dma; @@ -411,7 +411,7 @@ DECLARE_EVENT_CLASS(cdns2_log_trb, __field(u32, type) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->trb = trb; __entry->buffer = le32_to_cpu(trb->buffer); __entry->length = le32_to_cpu(trb->length); @@ -476,7 +476,7 @@ DECLARE_EVENT_CLASS(cdns2_log_ep, __field(u8, dequeue) ), TP_fast_assign( - __assign_str(name, pep->name); + __assign_str(name); __entry->maxpacket = pep->endpoint.maxpacket; __entry->maxpacket_limit = pep->endpoint.maxpacket_limit; __entry->flags = pep->ep_state; @@ -568,7 +568,7 @@ DECLARE_EVENT_CLASS(cdns2_log_epx_reg_config, __field(u32, ep_cfg_reg) ), TP_fast_assign( - __assign_str(ep_name, pep->name); + __assign_str(ep_name); __entry->burst_size = pep->trb_burst_size; __entry->maxpack_reg = pep->dir ? readw(&pdev->epx_regs->txmaxpack[pep->num - 1]) : readw(&pdev->epx_regs->rxmaxpack[pep->num - 1]); diff --git a/drivers/usb/gadget/udc/trace.h b/drivers/usb/gadget/udc/trace.h index a5ed26fbc2da..4e334298b0e8 100644 --- a/drivers/usb/gadget/udc/trace.h +++ b/drivers/usb/gadget/udc/trace.h @@ -157,7 +157,7 @@ DECLARE_EVENT_CLASS(udc_log_ep, __field(int, ret) ), TP_fast_assign( - __assign_str(name, ep->name); + __assign_str(name); __entry->maxpacket = ep->maxpacket; __entry->maxpacket_limit = ep->maxpacket_limit; __entry->max_streams = ep->max_streams; @@ -233,7 +233,7 @@ DECLARE_EVENT_CLASS(udc_log_req, __field(struct usb_request *, req) ), TP_fast_assign( - __assign_str(name, ep->name); + __assign_str(name); __entry->length = req->length; __entry->actual = req->actual; __entry->num_sgs = req->num_sgs; diff --git a/drivers/usb/mtu3/mtu3_trace.h b/drivers/usb/mtu3/mtu3_trace.h index 03d2a9bac27e..89870175d635 100644 --- a/drivers/usb/mtu3/mtu3_trace.h +++ b/drivers/usb/mtu3/mtu3_trace.h @@ -26,7 +26,7 @@ TRACE_EVENT(mtu3_log, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(name), __get_str(msg)) @@ -127,7 +127,7 @@ DECLARE_EVENT_CLASS(mtu3_log_request, __field(int, no_interrupt) ), TP_fast_assign( - __assign_str(name, mreq->mep->name); + __assign_str(name); __entry->mreq = mreq; __entry->gpd = mreq->gpd; __entry->actual = mreq->request.actual; @@ -182,7 +182,7 @@ DECLARE_EVENT_CLASS(mtu3_log_gpd, __field(u32, dw3) ), TP_fast_assign( - __assign_str(name, mep->name); + __assign_str(name); __entry->gpd = gpd; __entry->dw0 = le32_to_cpu(gpd->dw0_info); __entry->dw1 = le32_to_cpu(gpd->next_gpd); @@ -226,7 +226,7 @@ DECLARE_EVENT_CLASS(mtu3_log_ep, __field(struct mtu3_gpd_ring *, gpd_ring) ), TP_fast_assign( - __assign_str(name, mep->name); + __assign_str(name); __entry->type = mep->type; __entry->slot = mep->slot; __entry->maxp = mep->ep.maxpacket; diff --git a/drivers/usb/musb/musb_trace.h b/drivers/usb/musb/musb_trace.h index f246b14394c4..726e6697d475 100644 --- a/drivers/usb/musb/musb_trace.h +++ b/drivers/usb/musb/musb_trace.h @@ -31,7 +31,7 @@ TRACE_EVENT(musb_log, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __assign_vstr(msg, vaf->fmt, vaf->va); ), TP_printk("%s: %s", __get_str(name), __get_str(msg)) @@ -46,9 +46,9 @@ TRACE_EVENT(musb_state, __string(desc, desc) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __entry->devctl = devctl; - __assign_str(desc, desc); + __assign_str(desc); ), TP_printk("%s: devctl: %02x %s", __get_str(name), __entry->devctl, __get_str(desc)) @@ -160,7 +160,7 @@ TRACE_EVENT(musb_isr, __field(u16, int_rx) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __entry->int_usb = musb->int_usb; __entry->int_tx = musb->int_tx; __entry->int_rx = musb->int_rx; @@ -184,7 +184,7 @@ DECLARE_EVENT_CLASS(musb_urb, __field(u32, actual_len) ), TP_fast_assign( - __assign_str(name, dev_name(musb->controller)); + __assign_str(name); __entry->urb = urb; __entry->pipe = urb->pipe; __entry->status = urb->status; @@ -325,7 +325,7 @@ DECLARE_EVENT_CLASS(musb_cppi41, ), TP_fast_assign( __entry->ch = ch; - __assign_str(name, dev_name(ch->hw_ep->musb->controller)); + __assign_str(name); __entry->hwep = ch->hw_ep->epnum; __entry->port = ch->port_num; __entry->is_tx = ch->is_tx; diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h index 362e1fc7ef6a..84fcf26e306e 100644 --- a/fs/bcachefs/trace.h +++ b/fs/bcachefs/trace.h @@ -43,7 +43,7 @@ DECLARE_EVENT_CLASS(fs_str, TP_fast_assign( __entry->dev = c->dev; - __assign_str(str, str); + __assign_str(str); ), TP_printk("%d,%d\n%s", MAJOR(__entry->dev), MINOR(__entry->dev), __get_str(str)) @@ -64,7 +64,7 @@ DECLARE_EVENT_CLASS(trans_str, __entry->dev = trans->c->dev; strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); __entry->caller_ip = caller_ip; - __assign_str(str, str); + __assign_str(str); ), TP_printk("%d,%d %s %pS %s", @@ -85,7 +85,7 @@ DECLARE_EVENT_CLASS(trans_str_nocaller, TP_fast_assign( __entry->dev = trans->c->dev; strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn)); - __assign_str(str, str); + __assign_str(str); ), TP_printk("%d,%d %s %s", diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 10985a4b8259..4de8780a7c48 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -47,7 +47,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event, TP_fast_assign( __entry->error = error < 0 ? -error : 0; - __assign_str(dstaddr, clp->cl_hostname); + __assign_str(dstaddr); ), TP_printk( @@ -94,8 +94,8 @@ TRACE_EVENT(nfs4_trunked_exchange_id, TP_fast_assign( __entry->error = error < 0 ? -error : 0; - __assign_str(main_addr, clp->cl_hostname); - __assign_str(trunk_addr, addr); + __assign_str(main_addr); + __assign_str(trunk_addr); ), TP_printk( @@ -365,7 +365,7 @@ TRACE_EVENT(nfs4_state_mgr, TP_fast_assign( __entry->state = clp->cl_state; - __assign_str(hostname, clp->cl_hostname); + __assign_str(hostname); ), TP_printk( @@ -393,8 +393,8 @@ TRACE_EVENT(nfs4_state_mgr_failed, TP_fast_assign( __entry->error = status < 0 ? -status : 0; __entry->state = clp->cl_state; - __assign_str(hostname, clp->cl_hostname); - __assign_str(section, section); + __assign_str(hostname); + __assign_str(section); ), TP_printk( @@ -578,7 +578,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, __entry->fhandle = 0; } __entry->dir = NFS_FILEID(d_inode(ctx->dentry->d_parent)); - __assign_str(name, ctx->dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -1072,7 +1072,7 @@ DECLARE_EVENT_CLASS(nfs4_lookup_event, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->error = -error; - __assign_str(name, name->name); + __assign_str(name); ), TP_printk( @@ -1156,8 +1156,8 @@ TRACE_EVENT(nfs4_rename, __entry->olddir = NFS_FILEID(olddir); __entry->newdir = NFS_FILEID(newdir); __entry->error = error < 0 ? -error : 0; - __assign_str(oldname, oldname->name); - __assign_str(newname, newname->name); + __assign_str(oldname); + __assign_str(newname); ), TP_printk( @@ -1359,7 +1359,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event, __entry->fileid = 0; __entry->dev = 0; } - __assign_str(dstaddr, clp ? clp->cl_hostname : "unknown"); + __assign_str(dstaddr); ), TP_printk( @@ -1416,7 +1416,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, __entry->fileid = 0; __entry->dev = 0; } - __assign_str(dstaddr, clp ? clp->cl_hostname : "unknown"); + __assign_str(dstaddr); __entry->stateid_seq = be32_to_cpu(stateid->seqid); __entry->stateid_hash = @@ -1960,7 +1960,7 @@ DECLARE_EVENT_CLASS(nfs4_deviceid_event, ), TP_fast_assign( - __assign_str(dstaddr, clp->cl_hostname); + __assign_str(dstaddr); memcpy(__entry->deviceid, deviceid->data, NFS4_DEVICEID4_SIZE); ), @@ -1998,7 +1998,7 @@ DECLARE_EVENT_CLASS(nfs4_deviceid_status, TP_fast_assign( __entry->dev = server->s_dev; __entry->status = status; - __assign_str(dstaddr, server->nfs_client->cl_hostname); + __assign_str(dstaddr); memcpy(__entry->deviceid, deviceid->data, NFS4_DEVICEID4_SIZE); ), @@ -2036,8 +2036,8 @@ TRACE_EVENT(fl_getdevinfo, ), TP_fast_assign( - __assign_str(mds_addr, server->nfs_client->cl_hostname); - __assign_str(ds_ips, ds_remotestr); + __assign_str(mds_addr); + __assign_str(ds_ips); memcpy(__entry->deviceid, deviceid->data, NFS4_DEVICEID4_SIZE); ), @@ -2083,9 +2083,7 @@ DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, be32_to_cpu(hdr->args.stateid.seqid); __entry->stateid_hash = nfs_stateid_hash(&hdr->args.stateid); - __assign_str(dstaddr, hdr->ds_clp ? - rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient, - RPC_DISPLAY_ADDR) : "unknown"); + __assign_str(dstaddr); ), TP_printk( @@ -2139,9 +2137,7 @@ TRACE_EVENT(ff_layout_commit_error, __entry->dev = inode->i_sb->s_dev; __entry->offset = data->args.offset; __entry->count = data->args.count; - __assign_str(dstaddr, data->ds_clp ? - rpc_peeraddr2str(data->ds_clp->cl_rpcclient, - RPC_DISPLAY_ADDR) : "unknown"); + __assign_str(dstaddr); ), TP_printk( @@ -2579,7 +2575,7 @@ DECLARE_EVENT_CLASS(nfs4_xattr_event, __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); - __assign_str(name, name); + __assign_str(name); ), TP_printk( diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index afedb449b54f..1e710654af11 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -409,7 +409,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event, __entry->dir = NFS_FILEID(dir); __entry->flags = flags; __entry->fileid = d_is_negative(dentry) ? 0 : NFS_FILEID(d_inode(dentry)); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -457,7 +457,7 @@ DECLARE_EVENT_CLASS(nfs_lookup_event_done, __entry->error = error < 0 ? -error : 0; __entry->flags = flags; __entry->fileid = d_is_negative(dentry) ? 0 : NFS_FILEID(d_inode(dentry)); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -512,7 +512,7 @@ TRACE_EVENT(nfs_atomic_open_enter, __entry->dir = NFS_FILEID(dir); __entry->flags = flags; __entry->fmode = (__force unsigned long)ctx->mode; - __assign_str(name, ctx->dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -551,7 +551,7 @@ TRACE_EVENT(nfs_atomic_open_exit, __entry->dir = NFS_FILEID(dir); __entry->flags = flags; __entry->fmode = (__force unsigned long)ctx->mode; - __assign_str(name, ctx->dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -587,7 +587,7 @@ TRACE_EVENT(nfs_create_enter, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -623,7 +623,7 @@ TRACE_EVENT(nfs_create_exit, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->flags = flags; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -654,7 +654,7 @@ DECLARE_EVENT_CLASS(nfs_directory_event, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -693,7 +693,7 @@ DECLARE_EVENT_CLASS(nfs_directory_event_done, __entry->dev = dir->i_sb->s_dev; __entry->dir = NFS_FILEID(dir); __entry->error = error < 0 ? -error : 0; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -747,7 +747,7 @@ TRACE_EVENT(nfs_link_enter, __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->dir = NFS_FILEID(dir); - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -783,7 +783,7 @@ TRACE_EVENT(nfs_link_exit, __entry->fileid = NFS_FILEID(inode); __entry->dir = NFS_FILEID(dir); __entry->error = error < 0 ? -error : 0; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk( @@ -819,8 +819,8 @@ DECLARE_EVENT_CLASS(nfs_rename_event, __entry->dev = old_dir->i_sb->s_dev; __entry->old_dir = NFS_FILEID(old_dir); __entry->new_dir = NFS_FILEID(new_dir); - __assign_str(old_name, old_dentry->d_name.name); - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(old_name); + __assign_str(new_name); ), TP_printk( @@ -868,8 +868,8 @@ DECLARE_EVENT_CLASS(nfs_rename_event_done, __entry->error = -error; __entry->old_dir = NFS_FILEID(old_dir); __entry->new_dir = NFS_FILEID(new_dir); - __assign_str(old_name, old_dentry->d_name.name); - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(old_name); + __assign_str(new_name); ), TP_printk( @@ -1636,8 +1636,8 @@ TRACE_EVENT(nfs_mount_assign, ), TP_fast_assign( - __assign_str(option, option); - __assign_str(value, value); + __assign_str(option); + __assign_str(value); ), TP_printk("option %s=%s", @@ -1657,7 +1657,7 @@ TRACE_EVENT(nfs_mount_option, ), TP_fast_assign( - __assign_str(option, param->key); + __assign_str(option); ), TP_printk("option %s", __get_str(option)) @@ -1675,7 +1675,7 @@ TRACE_EVENT(nfs_mount_path, ), TP_fast_assign( - __assign_str(path, path); + __assign_str(path); ), TP_printk("path='%s'", __get_str(path)) @@ -1710,9 +1710,8 @@ DECLARE_EVENT_CLASS(nfs_xdr_event, __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->version = task->tk_client->cl_vers; __entry->error = error; - __assign_str(program, - task->tk_client->cl_program->name); - __assign_str(procedure, task->tk_msg.rpc_proc->p_name); + __assign_str(program); + __assign_str(procedure); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index b5e48d504062..77bbd23aa150 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -104,7 +104,7 @@ TRACE_EVENT(nfsd_compound, TP_fast_assign( __entry->xid = be32_to_cpu(rqst->rq_xid); __entry->opcnt = opcnt; - __assign_str(tag, tag); + __assign_str(tag); ), TP_printk("xid=0x%08x opcnt=%u tag=%s", __entry->xid, __entry->opcnt, __get_str(tag) @@ -127,7 +127,7 @@ TRACE_EVENT(nfsd_compound_status, __entry->args_opcnt = args_opcnt; __entry->resp_opcnt = resp_opcnt; __entry->status = be32_to_cpu(status); - __assign_str(name, name); + __assign_str(name); ), TP_printk("op=%u/%u %s status=%d", __entry->resp_opcnt, __entry->args_opcnt, @@ -318,7 +318,7 @@ TRACE_EVENT(nfsd_exp_find_key, TP_fast_assign( __entry->fsidtype = key->ek_fsidtype; memcpy(__entry->fsid, key->ek_fsid, 4*6); - __assign_str(auth_domain, key->ek_client->name); + __assign_str(auth_domain); __entry->status = status; ), TP_printk("fsid=%x::%s domain=%s status=%d", @@ -342,8 +342,8 @@ TRACE_EVENT(nfsd_expkey_update, TP_fast_assign( __entry->fsidtype = key->ek_fsidtype; memcpy(__entry->fsid, key->ek_fsid, 4*6); - __assign_str(auth_domain, key->ek_client->name); - __assign_str(path, exp_path); + __assign_str(auth_domain); + __assign_str(path); __entry->cache = !test_bit(CACHE_NEGATIVE, &key->h.flags); ), TP_printk("fsid=%x::%s domain=%s path=%s cache=%s", @@ -365,8 +365,8 @@ TRACE_EVENT(nfsd_exp_get_by_name, __field(int, status) ), TP_fast_assign( - __assign_str(path, key->ex_path.dentry->d_name.name); - __assign_str(auth_domain, key->ex_client->name); + __assign_str(path); + __assign_str(auth_domain); __entry->status = status; ), TP_printk("path=%s domain=%s status=%d", @@ -385,8 +385,8 @@ TRACE_EVENT(nfsd_export_update, __field(bool, cache) ), TP_fast_assign( - __assign_str(path, key->ex_path.dentry->d_name.name); - __assign_str(auth_domain, key->ex_client->name); + __assign_str(path); + __assign_str(auth_domain); __entry->cache = !test_bit(CACHE_NEGATIVE, &key->h.flags); ), TP_printk("path=%s domain=%s cache=%s", @@ -485,7 +485,7 @@ TRACE_EVENT(nfsd_dirent, TP_fast_assign( __entry->fh_hash = fhp ? knfsd_fh_hash(&fhp->fh_handle) : 0; __entry->ino = ino; - __assign_str(name, name); + __assign_str(name); ), TP_printk("fh_hash=0x%08x ino=%llu name=%s", __entry->fh_hash, __entry->ino, __get_str(name) @@ -1000,7 +1000,7 @@ DECLARE_EVENT_CLASS(nfsd_clid_class, __entry->flavor = clp->cl_cred.cr_flavor; memcpy(__entry->verifier, (void *)&clp->cl_verifier, NFS4_VERIFIER_SIZE); - __assign_str(name, clp->cl_name.data); + __assign_str(name); ), TP_printk("addr=%pISpc name='%s' verifier=0x%s flavor=%s client=%08x:%08x", __entry->addr, __get_str(name), @@ -1519,7 +1519,7 @@ TRACE_EVENT(nfsd_cb_setup, TP_fast_assign( __entry->cl_boot = clp->cl_clientid.cl_boot; __entry->cl_id = clp->cl_clientid.cl_id; - __assign_str(netid, netid); + __assign_str(netid); __entry->authflavor = authflavor; __assign_sockaddr(addr, &clp->cl_cb_conn.cb_addr, clp->cl_cb_conn.cb_addrlen) @@ -1864,7 +1864,7 @@ TRACE_EVENT(nfsd_ctl_unlock_ip, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(address, address); + __assign_str(address); ), TP_printk("address=%s", __get_str(address) @@ -1883,7 +1883,7 @@ TRACE_EVENT(nfsd_ctl_unlock_fs, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(path, path); + __assign_str(path); ), TP_printk("path=%s", __get_str(path) @@ -1907,8 +1907,8 @@ TRACE_EVENT(nfsd_ctl_filehandle, TP_fast_assign( __entry->netns_ino = net->ns.inum; __entry->maxsize = maxsize; - __assign_str(domain, domain); - __assign_str(path, path); + __assign_str(domain); + __assign_str(path); ), TP_printk("domain=%s path=%s maxsize=%d", __get_str(domain), __get_str(path), __entry->maxsize @@ -1968,7 +1968,7 @@ TRACE_EVENT(nfsd_ctl_version, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(mesg, mesg); + __assign_str(mesg); ), TP_printk("%s", __get_str(mesg) @@ -2009,7 +2009,7 @@ TRACE_EVENT(nfsd_ctl_ports_addxprt, TP_fast_assign( __entry->netns_ino = net->ns.inum; __entry->port = port; - __assign_str(transport, transport); + __assign_str(transport); ), TP_printk("transport=%s port=%d", __get_str(transport), __entry->port @@ -2070,7 +2070,7 @@ TRACE_EVENT(nfsd_ctl_time, TP_fast_assign( __entry->netns_ino = net->ns.inum; __entry->time = time; - __assign_str(name, name); + __assign_str(name); ), TP_printk("file=%s time=%d", __get_str(name), __entry->time @@ -2089,7 +2089,7 @@ TRACE_EVENT(nfsd_ctl_recoverydir, ), TP_fast_assign( __entry->netns_ino = net->ns.inum; - __assign_str(recdir, recdir); + __assign_str(recdir); ), TP_printk("recdir=%s", __get_str(recdir) diff --git a/fs/ocfs2/ocfs2_trace.h b/fs/ocfs2/ocfs2_trace.h index 9898c11bdfa1..60e208b01c8d 100644 --- a/fs/ocfs2/ocfs2_trace.h +++ b/fs/ocfs2/ocfs2_trace.h @@ -82,7 +82,7 @@ DECLARE_EVENT_CLASS(ocfs2__string, __string(name,name) ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) ); @@ -1289,7 +1289,7 @@ DECLARE_EVENT_CLASS(ocfs2__file_ops, __entry->dentry = dentry; __entry->ino = ino; __entry->d_len = d_len; - __assign_str(d_name, d_name); + __assign_str(d_name); __entry->para = para; ), TP_printk("%p %p %p %llu %llu %.*s", __entry->inode, __entry->file, @@ -1425,7 +1425,7 @@ TRACE_EVENT(ocfs2_setattr, __entry->dentry = dentry; __entry->ino = ino; __entry->d_len = d_len; - __assign_str(d_name, d_name); + __assign_str(d_name); __entry->ia_valid = ia_valid; __entry->ia_mode = ia_mode; __entry->ia_uid = ia_uid; @@ -1683,7 +1683,7 @@ TRACE_EVENT(ocfs2_parse_options, ), TP_fast_assign( __entry->is_remount = is_remount; - __assign_str(options, options); + __assign_str(options); ), TP_printk("%d %s", __entry->is_remount, __get_str(options)) ); @@ -1718,8 +1718,8 @@ TRACE_EVENT(ocfs2_initialize_super, __field(int, cluster_bits) ), TP_fast_assign( - __assign_str(label, label); - __assign_str(uuid_str, uuid_str); + __assign_str(label); + __assign_str(uuid_str); __entry->root_dir = root_dir; __entry->system_dir = system_dir; __entry->cluster_bits = cluster_bits; @@ -1746,7 +1746,7 @@ TRACE_EVENT(ocfs2_init_xattr_set_ctxt, __field(int, credits) ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->meta = meta; __entry->clusters = clusters; __entry->credits = credits; @@ -1770,7 +1770,7 @@ DECLARE_EVENT_CLASS(ocfs2__xattr_find, ), TP_fast_assign( __entry->ino = ino; - __assign_str(name, name); + __assign_str(name); __entry->name_index = name_index; __entry->hash = hash; __entry->location = location; @@ -2019,7 +2019,7 @@ TRACE_EVENT(ocfs2_sync_dquot_helper, __entry->dq_id = dq_id; __entry->dq_type = dq_type; __entry->type = type; - __assign_str(s_id, s_id); + __assign_str(s_id); ), TP_printk("%u %u %lu %s", __entry->dq_id, __entry->dq_type, __entry->type, __get_str(s_id)) @@ -2060,7 +2060,7 @@ TRACE_EVENT(ocfs2_dx_dir_search, TP_fast_assign( __entry->ino = ino; __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); __entry->major_hash = major_hash; __entry->minor_hash = minor_hash; __entry->blkno = blkno; @@ -2088,7 +2088,7 @@ TRACE_EVENT(ocfs2_find_files_on_disk, ), TP_fast_assign( __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); __entry->blkno = blkno; __entry->dir = dir; ), @@ -2107,7 +2107,7 @@ TRACE_EVENT(ocfs2_check_dir_for_entry, TP_fast_assign( __entry->dir = dir; __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%llu %.*s", __entry->dir, __entry->namelen, __get_str(name)) @@ -2135,7 +2135,7 @@ TRACE_EVENT(ocfs2_dx_dir_index_root_block, __entry->major_hash = major_hash; __entry->minor_hash = minor_hash; __entry->namelen = namelen; - __assign_str(name, name); + __assign_str(name); __entry->num_used = num_used; ), TP_printk("%llu %x %x %.*s %u", __entry->dir, @@ -2171,7 +2171,7 @@ DECLARE_EVENT_CLASS(ocfs2__dentry_ops, __entry->dir = dir; __entry->dentry = dentry; __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); __entry->dir_blkno = dir_blkno; __entry->extra = extra; ), @@ -2217,7 +2217,7 @@ TRACE_EVENT(ocfs2_mknod, __entry->dir = dir; __entry->dentry = dentry; __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); __entry->dir_blkno = dir_blkno; __entry->dev = dev; __entry->mode = mode; @@ -2241,9 +2241,9 @@ TRACE_EVENT(ocfs2_link, TP_fast_assign( __entry->ino = ino; __entry->old_len = old_len; - __assign_str(old_name, old_name); + __assign_str(old_name); __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%llu %.*s %.*s", __entry->ino, __entry->old_len, __get_str(old_name), @@ -2279,9 +2279,9 @@ TRACE_EVENT(ocfs2_rename, __entry->new_dir = new_dir; __entry->new_dentry = new_dentry; __entry->old_len = old_len; - __assign_str(old_name, old_name); + __assign_str(old_name); __entry->new_len = new_len; - __assign_str(new_name, new_name); + __assign_str(new_name); ), TP_printk("%p %p %p %p %.*s %.*s", __entry->old_dir, __entry->old_dentry, @@ -2301,7 +2301,7 @@ TRACE_EVENT(ocfs2_rename_target_exists, ), TP_fast_assign( __entry->new_len = new_len; - __assign_str(new_name, new_name); + __assign_str(new_name); ), TP_printk("%.*s", __entry->new_len, __get_str(new_name)) ); @@ -2344,7 +2344,7 @@ TRACE_EVENT(ocfs2_symlink_begin, __entry->dentry = dentry; __entry->symname = symname; __entry->len = len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%p %p %s %.*s", __entry->dir, __entry->dentry, __entry->symname, __entry->len, __get_str(name)) @@ -2360,7 +2360,7 @@ TRACE_EVENT(ocfs2_blkno_stringify, ), TP_fast_assign( __entry->blkno = blkno; - __assign_str(name, name); + __assign_str(name); __entry->namelen = namelen; ), TP_printk("%llu %s %d", __entry->blkno, __get_str(name), @@ -2381,7 +2381,7 @@ TRACE_EVENT(ocfs2_orphan_del, ), TP_fast_assign( __entry->dir = dir; - __assign_str(name, name); + __assign_str(name); __entry->namelen = namelen; ), TP_printk("%llu %s %d", __entry->dir, __get_str(name), @@ -2403,7 +2403,7 @@ TRACE_EVENT(ocfs2_dentry_revalidate, TP_fast_assign( __entry->dentry = dentry; __entry->len = len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%p %.*s", __entry->dentry, __entry->len, __get_str(name)) ); @@ -2420,7 +2420,7 @@ TRACE_EVENT(ocfs2_dentry_revalidate_negative, ), TP_fast_assign( __entry->len = len; - __assign_str(name, name); + __assign_str(name); __entry->pgen = pgen; __entry->gen = gen; ), @@ -2445,7 +2445,7 @@ TRACE_EVENT(ocfs2_find_local_alias, ), TP_fast_assign( __entry->len = len; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%.*s", __entry->len, __get_str(name)) ); @@ -2462,7 +2462,7 @@ TRACE_EVENT(ocfs2_dentry_attach_lock, ), TP_fast_assign( __entry->len = len; - __assign_str(name, name); + __assign_str(name); __entry->parent = parent; __entry->fsdata = fsdata; ), @@ -2480,7 +2480,7 @@ TRACE_EVENT(ocfs2_dentry_attach_lock_found, __field(unsigned long long, ino) ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->parent = parent; __entry->ino = ino; ), @@ -2527,7 +2527,7 @@ TRACE_EVENT(ocfs2_get_parent, TP_fast_assign( __entry->child = child; __entry->len = len; - __assign_str(name, name); + __assign_str(name); __entry->ino = ino; ), TP_printk("%p %.*s %llu", __entry->child, __entry->len, @@ -2551,7 +2551,7 @@ TRACE_EVENT(ocfs2_encode_fh_begin, TP_fast_assign( __entry->dentry = dentry; __entry->name_len = name_len; - __assign_str(name, name); + __assign_str(name); __entry->fh = fh; __entry->len = len; __entry->connectable = connectable; diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index af97389e983e..36d47ce59631 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -518,7 +518,7 @@ DECLARE_EVENT_CLASS(smb3_inf_compound_enter_class, __entry->xid = xid; __entry->tid = tid; __entry->sesid = sesid; - __assign_str(path, full_path); + __assign_str(path); ), TP_printk("xid=%u sid=0x%llx tid=0x%x path=%s", __entry->xid, __entry->sesid, __entry->tid, @@ -762,7 +762,7 @@ DECLARE_EVENT_CLASS(smb3_exit_err_class, ), TP_fast_assign( __entry->xid = xid; - __assign_str(func_name, func_name); + __assign_str(func_name); __entry->rc = rc; ), TP_printk("\t%s: xid=%u rc=%d", @@ -815,7 +815,7 @@ DECLARE_EVENT_CLASS(smb3_enter_exit_class, ), TP_fast_assign( __entry->xid = xid; - __assign_str(func_name, func_name); + __assign_str(func_name); ), TP_printk("\t%s: xid=%u", __get_str(func_name), __entry->xid) @@ -852,7 +852,7 @@ DECLARE_EVENT_CLASS(smb3_tcon_class, __entry->xid = xid; __entry->tid = tid; __entry->sesid = sesid; - __assign_str(name, unc_name); + __assign_str(name); __entry->rc = rc; ), TP_printk("xid=%u sid=0x%llx tid=0x%x unc_name=%s rc=%d", @@ -896,7 +896,7 @@ DECLARE_EVENT_CLASS(smb3_open_enter_class, __entry->xid = xid; __entry->tid = tid; __entry->sesid = sesid; - __assign_str(path, full_path); + __assign_str(path); __entry->create_options = create_options; __entry->desired_access = desired_access; ), @@ -1098,7 +1098,7 @@ DECLARE_EVENT_CLASS(smb3_connect_class, __entry->conn_id = conn_id; pss = (struct sockaddr_storage *)__entry->dst_addr; *pss = *dst_addr; - __assign_str(hostname, hostname); + __assign_str(hostname); ), TP_printk("conn_id=0x%llx server=%s addr=%pISpsfc", __entry->conn_id, @@ -1134,7 +1134,7 @@ DECLARE_EVENT_CLASS(smb3_connect_err_class, __entry->rc = rc; pss = (struct sockaddr_storage *)__entry->dst_addr; *pss = *dst_addr; - __assign_str(hostname, hostname); + __assign_str(hostname); ), TP_printk("rc=%d conn_id=0x%llx server=%s addr=%pISpsfc", __entry->rc, @@ -1166,7 +1166,7 @@ DECLARE_EVENT_CLASS(smb3_reconnect_class, TP_fast_assign( __entry->currmid = currmid; __entry->conn_id = conn_id; - __assign_str(hostname, hostname); + __assign_str(hostname); ), TP_printk("conn_id=0x%llx server=%s current_mid=%llu", __entry->conn_id, @@ -1255,7 +1255,7 @@ DECLARE_EVENT_CLASS(smb3_credit_class, TP_fast_assign( __entry->currmid = currmid; __entry->conn_id = conn_id; - __assign_str(hostname, hostname); + __assign_str(hostname); __entry->credits = credits; __entry->credits_to_add = credits_to_add; __entry->in_flight = in_flight; diff --git a/fs/xfs/scrub/trace.h b/fs/xfs/scrub/trace.h index e27daa51cab6..92ef4cdc486e 100644 --- a/fs/xfs/scrub/trace.h +++ b/fs/xfs/scrub/trace.h @@ -561,7 +561,7 @@ TRACE_EVENT(xchk_btree_op_error, __entry->dev = sc->mp->m_super->s_dev; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); @@ -604,7 +604,7 @@ TRACE_EVENT(xchk_ifork_btree_op_error, __entry->ino = sc->ip->i_ino; __entry->whichfork = cur->bc_ino.whichfork; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->ptr = cur->bc_levels[level].ptr; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); @@ -644,7 +644,7 @@ TRACE_EVENT(xchk_btree_error, xfs_fsblock_t fsbno = xchk_btree_cur_fsbno(cur, level); __entry->dev = sc->mp->m_super->s_dev; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); @@ -684,7 +684,7 @@ TRACE_EVENT(xchk_ifork_btree_error, __entry->ino = sc->ip->i_ino; __entry->whichfork = cur->bc_ino.whichfork; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); @@ -723,7 +723,7 @@ DECLARE_EVENT_CLASS(xchk_sbtree_class, __entry->dev = sc->mp->m_super->s_dev; __entry->type = sc->sm->sm_type; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno); __entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno); __entry->level = level; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 05cb59bd0b80..25ff6fe1eb6c 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -168,7 +168,7 @@ TRACE_EVENT(xlog_intent_recovery_failed, ), TP_fast_assign( __entry->dev = mp->m_super->s_dev; - __assign_str(name, ops->name); + __assign_str(name); __entry->error = error; ), TP_printk("dev %d:%d optype %s error %d", @@ -1913,7 +1913,7 @@ TRACE_EVENT(xfs_alloc_cur_check, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->bno = bno; __entry->len = len; __entry->diff = diff; @@ -2473,7 +2473,7 @@ DECLARE_EVENT_CLASS(xfs_btree_cur_class, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->nlevels = cur->bc_nlevels; __entry->ptr = cur->bc_levels[level].ptr; @@ -2523,7 +2523,7 @@ TRACE_EVENT(xfs_btree_alloc_block, __entry->ino = 0; break; } - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->error = error; if (!error && stat) { if (cur->bc_ops->ptr_len == XFS_BTREE_LONG_PTR_LEN) { @@ -2567,7 +2567,7 @@ TRACE_EVENT(xfs_btree_free_block, __entry->ino = cur->bc_ino.ip->i_ino; else __entry->ino = 0; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agbno = xfs_daddr_to_agbno(cur->bc_mp, xfs_buf_daddr(bp)); ), @@ -2643,7 +2643,7 @@ DECLARE_EVENT_CLASS(xfs_defer_pending_class, ), TP_fast_assign( __entry->dev = mp ? mp->m_super->s_dev : 0; - __assign_str(name, dfp->dfp_ops->name); + __assign_str(name); __entry->intent = dfp->dfp_intent; __entry->flags = dfp->dfp_flags; __entry->committed = dfp->dfp_done != NULL; @@ -2732,7 +2732,7 @@ DECLARE_EVENT_CLASS(xfs_defer_pending_item_class, ), TP_fast_assign( __entry->dev = mp ? mp->m_super->s_dev : 0; - __assign_str(name, dfp->dfp_ops->name); + __assign_str(name); __entry->intent = dfp->dfp_intent; __entry->item = item; __entry->committed = dfp->dfp_done != NULL; @@ -4244,7 +4244,7 @@ TRACE_EVENT(xfs_btree_commit_afakeroot, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agno = cur->bc_ag.pag->pag_agno; __entry->agbno = cur->bc_ag.afake->af_root; __entry->levels = cur->bc_ag.afake->af_levels; @@ -4273,7 +4273,7 @@ TRACE_EVENT(xfs_btree_commit_ifakeroot, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->agno = XFS_INO_TO_AGNO(cur->bc_mp, cur->bc_ino.ip->i_ino); __entry->agino = XFS_INO_TO_AGINO(cur->bc_mp, @@ -4312,7 +4312,7 @@ TRACE_EVENT(xfs_btree_bload_level_geometry, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->nlevels = cur->bc_nlevels; __entry->nr_this_level = nr_this_level; @@ -4350,7 +4350,7 @@ TRACE_EVENT(xfs_btree_bload_block, ), TP_fast_assign( __entry->dev = cur->bc_mp->m_super->s_dev; - __assign_str(name, cur->bc_ops->name); + __assign_str(name); __entry->level = level; __entry->block_idx = block_idx; __entry->nr_blocks = nr_blocks; @@ -4573,7 +4573,7 @@ TRACE_EVENT(xfs_force_shutdown, __entry->dev = mp->m_super->s_dev; __entry->ptag = ptag; __entry->flags = flags; - __assign_str(fname, fname); + __assign_str(fname); __entry->line_num = line_num; ), TP_printk("dev %d:%d tag %s flags %s file %s line_num %d", @@ -4755,7 +4755,7 @@ DECLARE_EVENT_CLASS(xfbtree_freesp_class, ), TP_fast_assign( __entry->xfino = file_inode(xfbt->target->bt_file)->i_ino; - __assign_str(btname, cur->bc_ops->name); + __assign_str(btname); __entry->nlevels = cur->bc_nlevels; __entry->fileoff = fileoff; ), @@ -5122,7 +5122,7 @@ DECLARE_EVENT_CLASS(xfs_getparents_rec_class, __entry->bufsize = ppi->gp_bufsize; __entry->parent_ino = pptr->gpr_parent.ha_fid.fid_ino; __entry->parent_gen = pptr->gpr_parent.ha_fid.fid_gen; - __assign_str(name, pptr->gpr_name); + __assign_str(name); ), TP_printk("dev %d:%d ino 0x%llx firstu %u reclen %u bufsize %u parent_ino 0x%llx parent_gen 0x%x name '%s'", MAJOR(__entry->dev), MINOR(__entry->dev), diff --git a/include/ras/ras_event.h b/include/ras/ras_event.h index c011ea236e9b..7c47151d5c72 100644 --- a/include/ras/ras_event.h +++ b/include/ras/ras_event.h @@ -61,7 +61,7 @@ TRACE_EVENT(extlog_mem_event, else __entry->pa_mask_lsb = ~0; __entry->fru_id = *fru_id; - __assign_str(fru_text, fru_text); + __assign_str(fru_text); cper_mem_err_pack(mem, &__entry->data); ), @@ -131,8 +131,8 @@ TRACE_EVENT(mc_event, TP_fast_assign( __entry->error_type = err_type; - __assign_str(msg, error_msg); - __assign_str(label, label); + __assign_str(msg); + __assign_str(label); __entry->error_count = error_count; __entry->mc_index = mc_index; __entry->top_layer = top_layer; @@ -141,7 +141,7 @@ TRACE_EVENT(mc_event, __entry->address = address; __entry->grain_bits = grain_bits; __entry->syndrome = syndrome; - __assign_str(driver_detail, driver_detail); + __assign_str(driver_detail); ), TP_printk("%d %s error%s:%s%s on %s (mc:%d location:%d:%d:%d address:0x%08lx grain:%d syndrome:0x%08lx%s%s)", @@ -239,7 +239,7 @@ TRACE_EVENT(non_standard_event, TP_fast_assign( memcpy(__entry->sec_type, sec_type, UUID_SIZE); memcpy(__entry->fru_id, fru_id, UUID_SIZE); - __assign_str(fru_text, fru_text); + __assign_str(fru_text); __entry->sev = sev; __entry->len = len; memcpy(__get_dynamic_array(buf), err, len); @@ -313,7 +313,7 @@ TRACE_EVENT(aer_event, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->status = status; __entry->severity = severity; __entry->tlp_header_valid = tlp_header_valid; diff --git a/include/trace/events/asoc.h b/include/trace/events/asoc.h index 517015ef36a8..202fc3680c36 100644 --- a/include/trace/events/asoc.h +++ b/include/trace/events/asoc.h @@ -32,8 +32,8 @@ DECLARE_EVENT_CLASS(snd_soc_dapm, ), TP_fast_assign( - __assign_str(card_name, dapm->card->name); - __assign_str(comp_name, dapm->component ? dapm->component->name : "(none)"); + __assign_str(card_name); + __assign_str(comp_name); __entry->val = val; ), @@ -69,7 +69,7 @@ DECLARE_EVENT_CLASS(snd_soc_dapm_basic, ), TP_fast_assign( - __assign_str(name, card->name); + __assign_str(name); __entry->event = event; ), @@ -104,7 +104,7 @@ DECLARE_EVENT_CLASS(snd_soc_dapm_widget, ), TP_fast_assign( - __assign_str(name, w->name); + __assign_str(name); __entry->val = val; ), @@ -150,7 +150,7 @@ TRACE_EVENT(snd_soc_dapm_walk_done, ), TP_fast_assign( - __assign_str(name, card->name); + __assign_str(name); __entry->power_checks = card->dapm_stats.power_checks; __entry->path_checks = card->dapm_stats.path_checks; __entry->neighbour_checks = card->dapm_stats.neighbour_checks; @@ -179,9 +179,9 @@ TRACE_EVENT(snd_soc_dapm_path, ), TP_fast_assign( - __assign_str(wname, widget->name); - __assign_str(pname, path->name ? path->name : DAPM_DIRECT); - __assign_str(pnname, path->node[dir]->name); + __assign_str(wname); + __assign_str(pname); + __assign_str(pnname); __entry->path_connect = path->connect; __entry->path_node = (long)path->node[dir]; __entry->path_dir = dir; @@ -226,7 +226,7 @@ TRACE_EVENT(snd_soc_jack_irq, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -245,7 +245,7 @@ TRACE_EVENT(snd_soc_jack_report, ), TP_fast_assign( - __assign_str(name, jack->jack->id); + __assign_str(name); __entry->mask = mask; __entry->val = val; ), @@ -266,7 +266,7 @@ TRACE_EVENT(snd_soc_jack_notify, ), TP_fast_assign( - __assign_str(name, jack->jack->id); + __assign_str(name); __entry->val = val; ), diff --git a/include/trace/events/avc.h b/include/trace/events/avc.h index b55fda2e0773..fed0f141d5f6 100644 --- a/include/trace/events/avc.h +++ b/include/trace/events/avc.h @@ -36,9 +36,9 @@ TRACE_EVENT(selinux_audited, __entry->denied = sad->denied; __entry->audited = sad->audited; __entry->result = sad->result; - __assign_str(tcontext, tcontext); - __assign_str(scontext, scontext); - __assign_str(tclass, tclass); + __assign_str(tcontext); + __assign_str(scontext); + __assign_str(tclass); ), TP_printk("requested=0x%x denied=0x%x audited=0x%x result=%d scontext=%s tcontext=%s tclass=%s", diff --git a/include/trace/events/bridge.h b/include/trace/events/bridge.h index a6b3a4e409f0..3fe4725c83ff 100644 --- a/include/trace/events/bridge.h +++ b/include/trace/events/bridge.h @@ -25,7 +25,7 @@ TRACE_EVENT(br_fdb_add, ), TP_fast_assign( - __assign_str(dev, dev->name); + __assign_str(dev); memcpy(__entry->addr, addr, ETH_ALEN); __entry->vid = vid; __entry->nlh_flags = nlh_flags; @@ -54,8 +54,8 @@ TRACE_EVENT(br_fdb_external_learn_add, ), TP_fast_assign( - __assign_str(br_dev, br->dev->name); - __assign_str(dev, p ? p->dev->name : "null"); + __assign_str(br_dev); + __assign_str(dev); memcpy(__entry->addr, addr, ETH_ALEN); __entry->vid = vid; ), @@ -80,8 +80,8 @@ TRACE_EVENT(fdb_delete, ), TP_fast_assign( - __assign_str(br_dev, br->dev->name); - __assign_str(dev, f->dst ? f->dst->dev->name : "null"); + __assign_str(br_dev); + __assign_str(dev); memcpy(__entry->addr, f->key.addr.addr, ETH_ALEN); __entry->vid = f->key.vlan_id; ), @@ -108,8 +108,8 @@ TRACE_EVENT(br_fdb_update, ), TP_fast_assign( - __assign_str(br_dev, br->dev->name); - __assign_str(dev, source->dev->name); + __assign_str(br_dev); + __assign_str(dev); memcpy(__entry->addr, addr, ETH_ALEN); __entry->vid = vid; __entry->flags = flags; @@ -141,7 +141,7 @@ TRACE_EVENT(br_mdb_full, TP_fast_assign( struct in6_addr *in6; - __assign_str(dev, dev->name); + __assign_str(dev); __entry->vid = group->vid; if (!group->proto) { diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h index d2d94d7c3fb5..fadf406b5260 100644 --- a/include/trace/events/btrfs.h +++ b/include/trace/events/btrfs.h @@ -1140,7 +1140,7 @@ TRACE_EVENT(btrfs_space_reservation, ), TP_fast_assign_btrfs(fs_info, - __assign_str(type, type); + __assign_str(type); __entry->val = val; __entry->bytes = bytes; __entry->reserve = reserve; @@ -1169,7 +1169,7 @@ TRACE_EVENT(btrfs_trigger_flush, __entry->flags = flags; __entry->bytes = bytes; __entry->flush = flush; - __assign_str(reason, reason); + __assign_str(reason); ), TP_printk_btrfs("%s: flush=%d(%s) flags=%llu(%s) bytes=%llu", @@ -1622,7 +1622,7 @@ DECLARE_EVENT_CLASS(btrfs_workqueue, TP_fast_assign_btrfs(btrfs_workqueue_owner(wq), __entry->wq = wq; - __assign_str(name, name); + __assign_str(name); ), TP_printk_btrfs("name=%s wq=%p", __get_str(name), diff --git a/include/trace/events/cgroup.h b/include/trace/events/cgroup.h index 0b95865a90f3..af2755bda6eb 100644 --- a/include/trace/events/cgroup.h +++ b/include/trace/events/cgroup.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(cgroup_root, TP_fast_assign( __entry->root = root->hierarchy_id; __entry->ss_mask = root->subsys_mask; - __assign_str(name, root->name); + __assign_str(name); ), TP_printk("root=%d ss_mask=%#x name=%s", @@ -68,7 +68,7 @@ DECLARE_EVENT_CLASS(cgroup, __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); __entry->level = cgrp->level; - __assign_str(path, path); + __assign_str(path); ), TP_printk("root=%d id=%llu level=%d path=%s", @@ -137,9 +137,9 @@ DECLARE_EVENT_CLASS(cgroup_migrate, __entry->dst_root = dst_cgrp->root->hierarchy_id; __entry->dst_id = cgroup_id(dst_cgrp); __entry->dst_level = dst_cgrp->level; - __assign_str(dst_path, path); + __assign_str(dst_path); __entry->pid = task->pid; - __assign_str(comm, task->comm); + __assign_str(comm); ), TP_printk("dst_root=%d dst_id=%llu dst_level=%d dst_path=%s pid=%d comm=%s", @@ -181,7 +181,7 @@ DECLARE_EVENT_CLASS(cgroup_event, __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); __entry->level = cgrp->level; - __assign_str(path, path); + __assign_str(path); __entry->val = val; ), diff --git a/include/trace/events/clk.h b/include/trace/events/clk.h index daed3c7a48c1..759f7371a6dc 100644 --- a/include/trace/events/clk.h +++ b/include/trace/events/clk.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(clk, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -97,7 +97,7 @@ DECLARE_EVENT_CLASS(clk_rate, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->rate = rate; ), @@ -145,7 +145,7 @@ DECLARE_EVENT_CLASS(clk_rate_range, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->min = min; __entry->max = max; ), @@ -174,8 +174,8 @@ DECLARE_EVENT_CLASS(clk_parent, ), TP_fast_assign( - __assign_str(name, core->name); - __assign_str(pname, parent ? parent->name : "none"); + __assign_str(name); + __assign_str(pname); ), TP_printk("%s %s", __get_str(name), __get_str(pname)) @@ -207,7 +207,7 @@ DECLARE_EVENT_CLASS(clk_phase, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->phase = phase; ), @@ -241,7 +241,7 @@ DECLARE_EVENT_CLASS(clk_duty_cycle, ), TP_fast_assign( - __assign_str(name, core->name); + __assign_str(name); __entry->num = duty->num; __entry->den = duty->den; ), @@ -279,8 +279,8 @@ DECLARE_EVENT_CLASS(clk_rate_request, ), TP_fast_assign( - __assign_str(name, req->core ? req->core->name : "none"); - __assign_str(pname, req->best_parent_hw ? clk_hw_get_name(req->best_parent_hw) : "none"); + __assign_str(name); + __assign_str(pname); __entry->min = req->min_rate; __entry->max = req->max_rate; __entry->prate = req->best_parent_rate; diff --git a/include/trace/events/cma.h b/include/trace/events/cma.h index 25103e67737c..383c09f583ac 100644 --- a/include/trace/events/cma.h +++ b/include/trace/events/cma.h @@ -23,7 +23,7 @@ TRACE_EVENT(cma_release, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->pfn = pfn; __entry->page = page; __entry->count = count; @@ -49,7 +49,7 @@ TRACE_EVENT(cma_alloc_start, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->count = count; __entry->align = align; ), @@ -77,7 +77,7 @@ TRACE_EVENT(cma_alloc_finish, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->pfn = pfn; __entry->page = page; __entry->count = count; @@ -110,7 +110,7 @@ TRACE_EVENT(cma_alloc_busy_retry, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->pfn = pfn; __entry->page = page; __entry->count = count; diff --git a/include/trace/events/devfreq.h b/include/trace/events/devfreq.h index 7627c620bbda..6cbc4d59fd96 100644 --- a/include/trace/events/devfreq.h +++ b/include/trace/events/devfreq.h @@ -23,7 +23,7 @@ TRACE_EVENT(devfreq_frequency, ), TP_fast_assign( - __assign_str(dev_name, dev_name(&devfreq->dev)); + __assign_str(dev_name); __entry->freq = freq; __entry->prev_freq = prev_freq; __entry->busy_time = devfreq->last_status.busy_time; @@ -54,7 +54,7 @@ TRACE_EVENT(devfreq_monitor, __entry->busy_time = devfreq->last_status.busy_time; __entry->total_time = devfreq->last_status.total_time; __entry->polling_ms = devfreq->profile->polling_ms; - __assign_str(dev_name, dev_name(&devfreq->dev)); + __assign_str(dev_name); ), TP_printk("dev_name=%-30s freq=%-12lu polling_ms=%-3u load=%-2lu", diff --git a/include/trace/events/devlink.h b/include/trace/events/devlink.h index 77ff7cfc6049..f241e204fe6b 100644 --- a/include/trace/events/devlink.h +++ b/include/trace/events/devlink.h @@ -31,9 +31,9 @@ TRACE_EVENT(devlink_hwmsg, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); __entry->incoming = incoming; __entry->type = type; memcpy(__get_dynamic_array(buf), buf, len); @@ -63,11 +63,11 @@ TRACE_EVENT(devlink_hwerr, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); __entry->err = err; - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("bus_name=%s dev_name=%s driver_name=%s err=%d %s", @@ -93,11 +93,11 @@ TRACE_EVENT(devlink_health_report, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(reporter_name, reporter_name); - __assign_str(msg, msg); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(reporter_name); + __assign_str(msg); ), TP_printk("bus_name=%s dev_name=%s driver_name=%s reporter_name=%s: %s", @@ -125,10 +125,10 @@ TRACE_EVENT(devlink_health_recover_aborted, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(reporter_name, reporter_name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(reporter_name); __entry->health_state = health_state; __entry->time_since_last_recover = time_since_last_recover; ), @@ -158,10 +158,10 @@ TRACE_EVENT(devlink_health_reporter_state_update, ), TP_fast_assign( - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(reporter_name, reporter_name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(reporter_name); __entry->new_state = new_state; ), @@ -192,11 +192,11 @@ TRACE_EVENT(devlink_trap_report, TP_fast_assign( struct net_device *input_dev = metadata->input_dev; - __assign_str(bus_name, devlink_to_dev(devlink)->bus->name); - __assign_str(dev_name, dev_name(devlink_to_dev(devlink))); - __assign_str(driver_name, devlink_to_dev(devlink)->driver->name); - __assign_str(trap_name, metadata->trap_name); - __assign_str(trap_group_name, metadata->trap_group_name); + __assign_str(bus_name); + __assign_str(dev_name); + __assign_str(driver_name); + __assign_str(trap_name); + __assign_str(trap_group_name); strscpy(__entry->input_dev_name, input_dev ? input_dev->name : "NULL", IFNAMSIZ); ), diff --git a/include/trace/events/dma_fence.h b/include/trace/events/dma_fence.h index 3963e79ca7b4..a4de3df8500b 100644 --- a/include/trace/events/dma_fence.h +++ b/include/trace/events/dma_fence.h @@ -23,8 +23,8 @@ DECLARE_EVENT_CLASS(dma_fence, ), TP_fast_assign( - __assign_str(driver, fence->ops->get_driver_name(fence)); - __assign_str(timeline, fence->ops->get_timeline_name(fence)); + __assign_str(driver); + __assign_str(timeline); __entry->context = fence->context; __entry->seqno = fence->seqno; ), diff --git a/include/trace/events/erofs.h b/include/trace/events/erofs.h index e18684b02c3d..b9bbfd855f2a 100644 --- a/include/trace/events/erofs.h +++ b/include/trace/events/erofs.h @@ -47,7 +47,7 @@ TRACE_EVENT(erofs_lookup, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->nid = EROFS_I(dir)->nid; - __assign_str(name, dentry->d_name.name); + __assign_str(name); __entry->flags = flags; ), diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h index 371ba28415f5..ed794b5fefbe 100644 --- a/include/trace/events/f2fs.h +++ b/include/trace/events/f2fs.h @@ -354,7 +354,7 @@ TRACE_EVENT(f2fs_unlink_enter, __entry->ino = dir->i_ino; __entry->size = dir->i_size; __entry->blocks = dir->i_blocks; - __assign_str(name, dentry->d_name.name); + __assign_str(name); ), TP_printk("dev = (%d,%d), dir ino = %lu, i_size = %lld, " @@ -843,7 +843,7 @@ TRACE_EVENT(f2fs_lookup_start, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->ino = dir->i_ino; - __assign_str(name, dentry->d_name.name); + __assign_str(name); __entry->flags = flags; ), @@ -871,7 +871,7 @@ TRACE_EVENT(f2fs_lookup_end, TP_fast_assign( __entry->dev = dir->i_sb->s_dev; __entry->ino = dir->i_ino; - __assign_str(name, dentry->d_name.name); + __assign_str(name); __entry->cino = ino; __entry->err = err; ), @@ -903,9 +903,9 @@ TRACE_EVENT(f2fs_rename_start, TP_fast_assign( __entry->dev = old_dir->i_sb->s_dev; __entry->ino = old_dir->i_ino; - __assign_str(old_name, old_dentry->d_name.name); + __assign_str(old_name); __entry->new_pino = new_dir->i_ino; - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(new_name); __entry->flags = flags; ), @@ -937,8 +937,8 @@ TRACE_EVENT(f2fs_rename_end, TP_fast_assign( __entry->dev = old_dentry->d_sb->s_dev; __entry->ino = old_dentry->d_inode->i_ino; - __assign_str(old_name, old_dentry->d_name.name); - __assign_str(new_name, new_dentry->d_name.name); + __assign_str(old_name); + __assign_str(new_name); __entry->flags = flags; __entry->ret = ret; ), @@ -1557,7 +1557,7 @@ TRACE_EVENT(f2fs_write_checkpoint, TP_fast_assign( __entry->dev = sb->s_dev; __entry->reason = reason; - __assign_str(dest_msg, msg); + __assign_str(dest_msg); ), TP_printk("dev = (%d,%d), checkpoint for %s, state = %s", @@ -2333,12 +2333,12 @@ DECLARE_EVENT_CLASS(f2fs__rw_start, * because this screws up the tooling that parses * the traces. */ - __assign_str(pathbuf, pathname); + __assign_str(pathbuf); (void)strreplace(__get_str(pathbuf), ' ', '_'); __entry->offset = offset; __entry->bytes = bytes; __entry->i_size = i_size_read(inode); - __assign_str(cmdline, command); + __assign_str(cmdline); (void)strreplace(__get_str(cmdline), ' ', '_'); __entry->pid = pid; __entry->ino = inode->i_ino; diff --git a/include/trace/events/habanalabs.h b/include/trace/events/habanalabs.h index a78d21fa9f29..4a2bb2c896d1 100644 --- a/include/trace/events/habanalabs.h +++ b/include/trace/events/habanalabs.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(habanalabs_mmu_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->virt_addr = virt_addr; __entry->phys_addr = phys_addr; __entry->page_size = page_size; @@ -64,7 +64,7 @@ DECLARE_EVENT_CLASS(habanalabs_dma_alloc_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->cpu_addr = cpu_addr; __entry->dma_addr = dma_addr; __entry->size = size; @@ -103,7 +103,7 @@ DECLARE_EVENT_CLASS(habanalabs_dma_map_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->phys_addr = phys_addr; __entry->dma_addr = dma_addr; __entry->len = len; @@ -141,7 +141,7 @@ DECLARE_EVENT_CLASS(habanalabs_comms_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->op_str = op_str; ), @@ -178,7 +178,7 @@ DECLARE_EVENT_CLASS(habanalabs_reg_access_template, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __entry->addr = addr; __entry->val = val; ), diff --git a/include/trace/events/huge_memory.h b/include/trace/events/huge_memory.h index ab576898a126..b5f5369b6300 100644 --- a/include/trace/events/huge_memory.h +++ b/include/trace/events/huge_memory.h @@ -191,7 +191,7 @@ TRACE_EVENT(mm_khugepaged_scan_file, TP_fast_assign( __entry->mm = mm; __entry->pfn = folio ? folio_pfn(folio) : -1; - __assign_str(filename, file->f_path.dentry->d_iname); + __assign_str(filename); __entry->present = present; __entry->swap = swap; __entry->result = result; @@ -228,7 +228,7 @@ TRACE_EVENT(mm_khugepaged_collapse_file, __entry->index = index; __entry->addr = addr; __entry->is_shmem = is_shmem; - __assign_str(filename, file->f_path.dentry->d_iname); + __assign_str(filename); __entry->nr = nr; __entry->result = result; ), diff --git a/include/trace/events/hwmon.h b/include/trace/events/hwmon.h index d7a1d0ffb679..d1ff560cd9b5 100644 --- a/include/trace/events/hwmon.h +++ b/include/trace/events/hwmon.h @@ -21,7 +21,7 @@ DECLARE_EVENT_CLASS(hwmon_attr_class, TP_fast_assign( __entry->index = index; - __assign_str(attr_name, attr_name); + __assign_str(attr_name); __entry->val = val; ), @@ -57,8 +57,8 @@ TRACE_EVENT(hwmon_attr_show_string, TP_fast_assign( __entry->index = index; - __assign_str(attr_name, attr_name); - __assign_str(label, s); + __assign_str(attr_name); + __assign_str(label); ), TP_printk("index=%d, attr_name=%s, val=%s", diff --git a/include/trace/events/initcall.h b/include/trace/events/initcall.h index eb903c3f195f..5282afdf3ddf 100644 --- a/include/trace/events/initcall.h +++ b/include/trace/events/initcall.h @@ -18,7 +18,7 @@ TRACE_EVENT(initcall_level, ), TP_fast_assign( - __assign_str(level, level); + __assign_str(level); ), TP_printk("level=%s", __get_str(level)) diff --git a/include/trace/events/intel_ish.h b/include/trace/events/intel_ish.h index e6d7ff55ee8c..64b6612c41bc 100644 --- a/include/trace/events/intel_ish.h +++ b/include/trace/events/intel_ish.h @@ -18,7 +18,7 @@ TRACE_EVENT(ishtp_dump, ), TP_fast_assign( - __assign_str(message, message); + __assign_str(message); ), TP_printk("%s", __get_str(message)) diff --git a/include/trace/events/io_uring.h b/include/trace/events/io_uring.h index e948df7ce625..412c9c210a32 100644 --- a/include/trace/events/io_uring.h +++ b/include/trace/events/io_uring.h @@ -164,7 +164,7 @@ TRACE_EVENT(io_uring_queue_async_work, __entry->work = &req->work; __entry->rw = rw; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, request %p, user_data 0x%llx, opcode %s, flags 0x%llx, %s queue, work %p", @@ -202,7 +202,7 @@ TRACE_EVENT(io_uring_defer, __entry->data = req->cqe.user_data; __entry->opcode = req->opcode; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, request %p, user_data 0x%llx, opcode %s", @@ -303,7 +303,7 @@ TRACE_EVENT(io_uring_fail_link, __entry->opcode = req->opcode; __entry->link = link; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, request %p, user_data 0x%llx, opcode %s, link %p", @@ -392,7 +392,7 @@ TRACE_EVENT(io_uring_submit_req, __entry->flags = (__force unsigned long long) req->flags; __entry->sq_thread = req->ctx->flags & IORING_SETUP_SQPOLL; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, flags 0x%llx, " @@ -436,7 +436,7 @@ TRACE_EVENT(io_uring_poll_arm, __entry->mask = mask; __entry->events = events; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, mask 0x%x, events 0x%x", @@ -475,7 +475,7 @@ TRACE_EVENT(io_uring_task_add, __entry->opcode = req->opcode; __entry->mask = mask; - __assign_str(op_str, io_uring_get_opcode(req->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, opcode %s, mask %x", @@ -538,7 +538,7 @@ TRACE_EVENT(io_uring_req_failed, __entry->addr3 = sqe->addr3; __entry->error = error; - __assign_str(op_str, io_uring_get_opcode(sqe->opcode)); + __assign_str(op_str); ), TP_printk("ring %p, req %p, user_data 0x%llx, " diff --git a/include/trace/events/iocost.h b/include/trace/events/iocost.h index af8bfed528fc..e772b1bc60d6 100644 --- a/include/trace/events/iocost.h +++ b/include/trace/events/iocost.h @@ -34,8 +34,8 @@ DECLARE_EVENT_CLASS(iocost_iocg_state, ), TP_fast_assign( - __assign_str(devname, ioc_name(iocg->ioc)); - __assign_str(cgroup, path); + __assign_str(devname); + __assign_str(cgroup); __entry->now = now->now; __entry->vnow = now->vnow; __entry->vrate = iocg->ioc->vtime_base_rate; @@ -93,8 +93,8 @@ DECLARE_EVENT_CLASS(iocg_inuse_update, ), TP_fast_assign( - __assign_str(devname, ioc_name(iocg->ioc)); - __assign_str(cgroup, path); + __assign_str(devname); + __assign_str(cgroup); __entry->now = now->now; __entry->old_inuse = old_inuse; __entry->new_inuse = new_inuse; @@ -159,7 +159,7 @@ TRACE_EVENT(iocost_ioc_vrate_adj, ), TP_fast_assign( - __assign_str(devname, ioc_name(ioc)); + __assign_str(devname); __entry->old_vrate = ioc->vtime_base_rate; __entry->new_vrate = new_vrate; __entry->busy_level = ioc->busy_level; @@ -200,8 +200,8 @@ TRACE_EVENT(iocost_iocg_forgive_debt, ), TP_fast_assign( - __assign_str(devname, ioc_name(iocg->ioc)); - __assign_str(cgroup, path); + __assign_str(devname); + __assign_str(cgroup); __entry->now = now->now; __entry->vnow = now->vnow; __entry->usage_pct = usage_pct; diff --git a/include/trace/events/iommu.h b/include/trace/events/iommu.h index 70743db1fb75..373007e567cb 100644 --- a/include/trace/events/iommu.h +++ b/include/trace/events/iommu.h @@ -28,7 +28,7 @@ DECLARE_EVENT_CLASS(iommu_group_event, TP_fast_assign( __entry->gid = group_id; - __assign_str(device, dev_name(dev)); + __assign_str(device); ), TP_printk("IOMMU: groupID=%d device=%s", @@ -62,7 +62,7 @@ DECLARE_EVENT_CLASS(iommu_device_event, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); + __assign_str(device); ), TP_printk("IOMMU: device=%s", __get_str(device) @@ -138,8 +138,8 @@ DECLARE_EVENT_CLASS(iommu_error, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); - __assign_str(driver, dev_driver_string(dev)); + __assign_str(device); + __assign_str(driver); __entry->iova = iova; __entry->flags = flags; ), diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h index a07b4607b663..837c1740d0d0 100644 --- a/include/trace/events/irq.h +++ b/include/trace/events/irq.h @@ -63,7 +63,7 @@ TRACE_EVENT(irq_handler_entry, TP_fast_assign( __entry->irq = irq; - __assign_str(name, action->name); + __assign_str(name); ), TP_printk("irq=%d name=%s", __entry->irq, __get_str(name)) diff --git a/include/trace/events/iscsi.h b/include/trace/events/iscsi.h index 8ff2a3ca5d75..990fd154f586 100644 --- a/include/trace/events/iscsi.h +++ b/include/trace/events/iscsi.h @@ -30,7 +30,7 @@ DECLARE_EVENT_CLASS(iscsi_log_msg, ), TP_fast_assign( - __assign_str(dname, dev_name(dev)); + __assign_str(dname); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/include/trace/events/kmem.h b/include/trace/events/kmem.h index 6e62cc64cd92..8a829e0f6e55 100644 --- a/include/trace/events/kmem.h +++ b/include/trace/events/kmem.h @@ -126,7 +126,7 @@ TRACE_EVENT(kmem_cache_free, TP_fast_assign( __entry->call_site = call_site; __entry->ptr = ptr; - __assign_str(name, s->name); + __assign_str(name); ), TP_printk("call_site=%pS ptr=%p name=%s", diff --git a/include/trace/events/lock.h b/include/trace/events/lock.h index 9ebd081e057e..8e89baa3775f 100644 --- a/include/trace/events/lock.h +++ b/include/trace/events/lock.h @@ -37,7 +37,7 @@ TRACE_EVENT(lock_acquire, TP_fast_assign( __entry->flags = (trylock ? 1 : 0) | (read ? 2 : 0); - __assign_str(name, lock->name); + __assign_str(name); __entry->lockdep_addr = lock; ), @@ -59,7 +59,7 @@ DECLARE_EVENT_CLASS(lock, ), TP_fast_assign( - __assign_str(name, lock->name); + __assign_str(name); __entry->lockdep_addr = lock; ), diff --git a/include/trace/events/mmap_lock.h b/include/trace/events/mmap_lock.h index 14db8044c1ff..f2827f98a44f 100644 --- a/include/trace/events/mmap_lock.h +++ b/include/trace/events/mmap_lock.h @@ -27,7 +27,7 @@ DECLARE_EVENT_CLASS(mmap_lock, TP_fast_assign( __entry->mm = mm; - __assign_str(memcg_path, memcg_path); + __assign_str(memcg_path); __entry->write = write; ), @@ -65,7 +65,7 @@ TRACE_EVENT_FN(mmap_lock_acquire_returned, TP_fast_assign( __entry->mm = mm; - __assign_str(memcg_path, memcg_path); + __assign_str(memcg_path); __entry->write = write; __entry->success = success; ), diff --git a/include/trace/events/mmc.h b/include/trace/events/mmc.h index 7b706ff21335..f1c2e94f7f68 100644 --- a/include/trace/events/mmc.h +++ b/include/trace/events/mmc.h @@ -68,7 +68,7 @@ TRACE_EVENT(mmc_request_start, __entry->need_retune = host->need_retune; __entry->hold_retune = host->hold_retune; __entry->retune_period = host->retune_period; - __assign_str(name, mmc_hostname(host)); + __assign_str(name); __entry->mrq = mrq; ), @@ -156,7 +156,7 @@ TRACE_EVENT(mmc_request_done, __entry->need_retune = host->need_retune; __entry->hold_retune = host->hold_retune; __entry->retune_period = host->retune_period; - __assign_str(name, mmc_hostname(host)); + __assign_str(name); __entry->mrq = mrq; ), diff --git a/include/trace/events/module.h b/include/trace/events/module.h index 097485c73c01..e5a006be9dc6 100644 --- a/include/trace/events/module.h +++ b/include/trace/events/module.h @@ -41,7 +41,7 @@ TRACE_EVENT(module_load, TP_fast_assign( __entry->taints = mod->taints; - __assign_str(name, mod->name); + __assign_str(name); ), TP_printk("%s %s", __get_str(name), show_module_flags(__entry->taints)) @@ -58,7 +58,7 @@ TRACE_EVENT(module_free, ), TP_fast_assign( - __assign_str(name, mod->name); + __assign_str(name); ), TP_printk("%s", __get_str(name)) @@ -82,7 +82,7 @@ DECLARE_EVENT_CLASS(module_refcnt, TP_fast_assign( __entry->ip = ip; __entry->refcnt = atomic_read(&mod->refcnt); - __assign_str(name, mod->name); + __assign_str(name); ), TP_printk("%s call_site=%ps refcnt=%d", @@ -119,7 +119,7 @@ TRACE_EVENT(module_request, TP_fast_assign( __entry->ip = ip; __entry->wait = wait; - __assign_str(name, name); + __assign_str(name); ), TP_printk("%s wait=%d call_site=%ps", diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h index dc03cf8e0369..b567b9ffedc1 100644 --- a/include/trace/events/napi.h +++ b/include/trace/events/napi.h @@ -26,7 +26,7 @@ TRACE_EVENT(napi_poll, TP_fast_assign( __entry->napi = napi; - __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); + __assign_str(dev_name); __entry->work = work; __entry->budget = budget; ), diff --git a/include/trace/events/neigh.h b/include/trace/events/neigh.h index 833143d0992e..12362c35dbc0 100644 --- a/include/trace/events/neigh.h +++ b/include/trace/events/neigh.h @@ -42,7 +42,7 @@ TRACE_EVENT(neigh_create, __be32 *p32; __entry->family = tbl->family; - __assign_str(dev, (dev ? dev->name : "NULL")); + __assign_str(dev); __entry->entries = atomic_read(&tbl->gc_entries); __entry->created = n != NULL; __entry->gc_exempt = exempt_from_gc; @@ -103,7 +103,7 @@ TRACE_EVENT(neigh_update, __be32 *p32; __entry->family = n->tbl->family; - __assign_str(dev, (n->dev ? n->dev->name : "NULL")); + __assign_str(dev); __entry->lladdr_len = lladdr_len; memcpy(__entry->lladdr, n->ha, lladdr_len); __entry->flags = n->flags; @@ -180,7 +180,7 @@ DECLARE_EVENT_CLASS(neigh__update, __be32 *p32; __entry->family = n->tbl->family; - __assign_str(dev, (n->dev ? n->dev->name : "NULL")); + __assign_str(dev); __entry->lladdr_len = lladdr_len; memcpy(__entry->lladdr, n->ha, lladdr_len); __entry->flags = n->flags; diff --git a/include/trace/events/net.h b/include/trace/events/net.h index f667c76a3b02..d55162c12f90 100644 --- a/include/trace/events/net.h +++ b/include/trace/events/net.h @@ -38,7 +38,7 @@ TRACE_EVENT(net_dev_start_xmit, ), TP_fast_assign( - __assign_str(name, dev->name); + __assign_str(name); __entry->queue_mapping = skb->queue_mapping; __entry->skbaddr = skb; __entry->vlan_tagged = skb_vlan_tag_present(skb); @@ -89,7 +89,7 @@ TRACE_EVENT(net_dev_xmit, __entry->skbaddr = skb; __entry->len = skb_len; __entry->rc = rc; - __assign_str(name, dev->name); + __assign_str(name); ), TP_printk("dev=%s skbaddr=%p len=%u rc=%d", @@ -110,8 +110,8 @@ TRACE_EVENT(net_dev_xmit_timeout, ), TP_fast_assign( - __assign_str(name, dev->name); - __assign_str(driver, netdev_drivername(dev)); + __assign_str(name); + __assign_str(driver); __entry->queue_index = queue_index; ), @@ -134,7 +134,7 @@ DECLARE_EVENT_CLASS(net_dev_template, TP_fast_assign( __entry->skbaddr = skb; __entry->len = skb->len; - __assign_str(name, skb->dev->name); + __assign_str(name); ), TP_printk("dev=%s skbaddr=%p len=%u", @@ -191,7 +191,7 @@ DECLARE_EVENT_CLASS(net_dev_rx_verbose_template, ), TP_fast_assign( - __assign_str(name, skb->dev->name); + __assign_str(name); #ifdef CONFIG_NET_RX_BUSY_POLL __entry->napi_id = skb->napi_id; #else diff --git a/include/trace/events/netlink.h b/include/trace/events/netlink.h index 3b7be3b386a4..f036b8a20505 100644 --- a/include/trace/events/netlink.h +++ b/include/trace/events/netlink.h @@ -17,7 +17,7 @@ TRACE_EVENT(netlink_extack, ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("msg=%s", __get_str(msg)) diff --git a/include/trace/events/oom.h b/include/trace/events/oom.h index b799f3bcba82..a42be4c8563b 100644 --- a/include/trace/events/oom.h +++ b/include/trace/events/oom.h @@ -92,7 +92,7 @@ TRACE_EVENT(mark_victim, TP_fast_assign( __entry->pid = task->pid; - __assign_str(comm, task->comm); + __assign_str(comm); __entry->total_vm = PG_COUNT_TO_KB(task->mm->total_vm); __entry->anon_rss = PG_COUNT_TO_KB(get_mm_counter(task->mm, MM_ANONPAGES)); __entry->file_rss = PG_COUNT_TO_KB(get_mm_counter(task->mm, MM_FILEPAGES)); diff --git a/include/trace/events/osnoise.h b/include/trace/events/osnoise.h index 82f741ec0f57..a2379a4f0684 100644 --- a/include/trace/events/osnoise.h +++ b/include/trace/events/osnoise.h @@ -75,7 +75,7 @@ TRACE_EVENT(irq_noise, ), TP_fast_assign( - __assign_str(desc, desc); + __assign_str(desc); __entry->vector = vector; __entry->start = start; __entry->duration = duration; diff --git a/include/trace/events/power.h b/include/trace/events/power.h index 77f14f7a11d4..d2349b6b531a 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -76,7 +76,7 @@ TRACE_EVENT(powernv_throttle, TP_fast_assign( __entry->chip_id = chip_id; - __assign_str(reason, reason); + __assign_str(reason); __entry->pmax = pmax; ), @@ -210,11 +210,10 @@ TRACE_EVENT(device_pm_callback_start, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); - __assign_str(driver, dev_driver_string(dev)); - __assign_str(parent, - dev->parent ? dev_name(dev->parent) : "none"); - __assign_str(pm_ops, pm_ops ? pm_ops : "none "); + __assign_str(device); + __assign_str(driver); + __assign_str(parent); + __assign_str(pm_ops); __entry->event = event; ), @@ -236,8 +235,8 @@ TRACE_EVENT(device_pm_callback_end, ), TP_fast_assign( - __assign_str(device, dev_name(dev)); - __assign_str(driver, dev_driver_string(dev)); + __assign_str(device); + __assign_str(driver); __entry->error = error; ), @@ -279,7 +278,7 @@ DECLARE_EVENT_CLASS(wakeup_source, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->state = state; ), @@ -318,7 +317,7 @@ DECLARE_EVENT_CLASS(clock, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->state = state; __entry->cpu_id = cpu_id; ), @@ -364,7 +363,7 @@ DECLARE_EVENT_CLASS(power_domain, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->state = state; __entry->cpu_id = cpu_id; ), @@ -486,7 +485,7 @@ DECLARE_EVENT_CLASS(dev_pm_qos_request, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->type = type; __entry->new_value = new_value; ), diff --git a/include/trace/events/pwc.h b/include/trace/events/pwc.h index a2da764a3b41..0543702542d9 100644 --- a/include/trace/events/pwc.h +++ b/include/trace/events/pwc.h @@ -26,7 +26,7 @@ TRACE_EVENT(pwc_handler_enter, __entry->urb__actual_length = urb->actual_length; __entry->fbuf__filled = (pdev->fill_buf ? pdev->fill_buf->filled : 0); - __assign_str(name, pdev->v4l2_dev.name); + __assign_str(name); ), TP_printk("dev=%s (fbuf=%p filled=%d) urb=%p (status=%d actual_length=%u)", __get_str(name), @@ -50,7 +50,7 @@ TRACE_EVENT(pwc_handler_exit, __entry->urb = urb; __entry->fbuf = pdev->fill_buf; __entry->fbuf__filled = pdev->fill_buf->filled; - __assign_str(name, pdev->v4l2_dev.name); + __assign_str(name); ), TP_printk(" dev=%s (fbuf=%p filled=%d) urb=%p", __get_str(name), diff --git a/include/trace/events/qdisc.h b/include/trace/events/qdisc.h index 1f4258308b96..f1b5e816e7e5 100644 --- a/include/trace/events/qdisc.h +++ b/include/trace/events/qdisc.h @@ -88,8 +88,8 @@ TRACE_EVENT(qdisc_reset, ), TP_fast_assign( - __assign_str(dev, qdisc_dev(q)->name); - __assign_str(kind, q->ops->id); + __assign_str(dev); + __assign_str(kind); __entry->parent = q->parent; __entry->handle = q->handle; ), @@ -113,8 +113,8 @@ TRACE_EVENT(qdisc_destroy, ), TP_fast_assign( - __assign_str(dev, qdisc_dev(q)->name); - __assign_str(kind, q->ops->id); + __assign_str(dev); + __assign_str(kind); __entry->parent = q->parent; __entry->handle = q->handle; ), @@ -137,8 +137,8 @@ TRACE_EVENT(qdisc_create, ), TP_fast_assign( - __assign_str(dev, dev->name); - __assign_str(kind, ops->id); + __assign_str(dev); + __assign_str(kind); __entry->parent = parent; ), diff --git a/include/trace/events/qla.h b/include/trace/events/qla.h index e7fd55e7dc3d..8800c35525a1 100644 --- a/include/trace/events/qla.h +++ b/include/trace/events/qla.h @@ -25,7 +25,7 @@ DECLARE_EVENT_CLASS(qla_log_event, __vstring(msg, vaf->fmt, vaf->va) ), TP_fast_assign( - __assign_str(buf, buf); + __assign_str(buf); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/include/trace/events/qrtr.h b/include/trace/events/qrtr.h index 441132c67133..14f822983741 100644 --- a/include/trace/events/qrtr.h +++ b/include/trace/events/qrtr.h @@ -102,7 +102,7 @@ TRACE_EVENT(qrtr_ns_message, ), TP_fast_assign( - __assign_str(ctrl_pkt_str, ctrl_pkt_str); + __assign_str(ctrl_pkt_str); __entry->sq_node = sq_node; __entry->sq_port = sq_port; ), diff --git a/include/trace/events/regulator.h b/include/trace/events/regulator.h index 72b3ba93b0a5..c58481a5d955 100644 --- a/include/trace/events/regulator.h +++ b/include/trace/events/regulator.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(regulator_basic, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); ), TP_printk("name=%s", __get_str(name)) @@ -119,7 +119,7 @@ DECLARE_EVENT_CLASS(regulator_range, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->min = min; __entry->max = max; ), @@ -152,7 +152,7 @@ DECLARE_EVENT_CLASS(regulator_value, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->val = val; ), diff --git a/include/trace/events/rpcgss.h b/include/trace/events/rpcgss.h index f50fcafc69de..7f0c1ceae726 100644 --- a/include/trace/events/rpcgss.h +++ b/include/trace/events/rpcgss.h @@ -154,7 +154,7 @@ DECLARE_EVENT_CLASS(rpcgss_ctx_class, TP_fast_assign( __entry->cred = gc; __entry->service = gc->gc_service; - __assign_str(principal, gc->gc_principal); + __assign_str(principal); ), TP_printk("cred=%p service=%s principal='%s'", @@ -189,7 +189,7 @@ DECLARE_EVENT_CLASS(rpcgss_svc_gssapi_class, TP_fast_assign( __entry->xid = __be32_to_cpu(rqstp->rq_xid); __entry->maj_stat = maj_stat; - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x maj_stat=%s", @@ -225,7 +225,7 @@ TRACE_EVENT(rpcgss_svc_wrap_failed, TP_fast_assign( __entry->xid = be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x", __get_str(addr), __entry->xid) @@ -245,7 +245,7 @@ TRACE_EVENT(rpcgss_svc_unwrap_failed, TP_fast_assign( __entry->xid = be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x", __get_str(addr), __entry->xid) @@ -271,7 +271,7 @@ TRACE_EVENT(rpcgss_svc_seqno_bad, __entry->expected = expected; __entry->received = received; __entry->xid = __be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x expected seqno %u, received seqno %u", @@ -299,7 +299,7 @@ TRACE_EVENT(rpcgss_svc_accept_upcall, __entry->minor_status = minor_status; __entry->major_status = major_status; __entry->xid = be32_to_cpu(rqstp->rq_xid); - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x major_status=%s (0x%08lx) minor_status=%u", @@ -327,7 +327,7 @@ TRACE_EVENT(rpcgss_svc_authenticate, TP_fast_assign( __entry->xid = be32_to_cpu(rqstp->rq_xid); __entry->seqno = gc->gc_seq; - __assign_str(addr, rqstp->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x seqno=%u", __get_str(addr), @@ -563,7 +563,7 @@ TRACE_EVENT(rpcgss_upcall_msg, ), TP_fast_assign( - __assign_str(msg, buf); + __assign_str(msg); ), TP_printk("msg='%s'", __get_str(msg)) @@ -618,7 +618,7 @@ TRACE_EVENT(rpcgss_context, __entry->timeout = timeout; __entry->window_size = window_size; __entry->len = len; - __assign_str(acceptor, data); + __assign_str(acceptor); ), TP_printk("win_size=%u expiry=%lu now=%lu timeout=%u acceptor=%.*s", @@ -677,7 +677,7 @@ TRACE_EVENT(rpcgss_oid_to_mech, ), TP_fast_assign( - __assign_str(oid, oid); + __assign_str(oid); ), TP_printk("mech for oid %s was not found", __get_str(oid)) diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index 027ac3ab457d..14392652273a 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -304,8 +304,8 @@ DECLARE_EVENT_CLASS(xprtrdma_reply_class, __entry->xid = be32_to_cpu(rep->rr_xid); __entry->version = be32_to_cpu(rep->rr_vers); __entry->proc = be32_to_cpu(rep->rr_proc); - __assign_str(addr, rpcrdma_addrstr(rep->rr_rxprt)); - __assign_str(port, rpcrdma_portstr(rep->rr_rxprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s xid=0x%08x version=%u proc=%u", @@ -335,8 +335,8 @@ DECLARE_EVENT_CLASS(xprtrdma_rxprt, ), TP_fast_assign( - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s", @@ -369,8 +369,8 @@ DECLARE_EVENT_CLASS(xprtrdma_connect_class, TP_fast_assign( __entry->rc = rc; __entry->connect_status = r_xprt->rx_ep->re_connect_status; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s rc=%d connection status=%d", @@ -608,8 +608,8 @@ DECLARE_EVENT_CLASS(xprtrdma_callback_class, TP_fast_assign( __entry->xid = be32_to_cpu(rqst->rq_xid); - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s xid=0x%08x", @@ -687,8 +687,8 @@ TRACE_EVENT(xprtrdma_op_connect, TP_fast_assign( __entry->delay = delay; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s delay=%lu", @@ -716,8 +716,8 @@ TRACE_EVENT(xprtrdma_op_set_cto, TP_fast_assign( __entry->connect = connect; __entry->reconnect = reconnect; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s connect=%lu reconnect=%lu", @@ -746,8 +746,8 @@ TRACE_EVENT(xprtrdma_createmrs, TP_fast_assign( __entry->count = count; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s created %u MRs", @@ -775,8 +775,8 @@ TRACE_EVENT(xprtrdma_nomrs_err, __entry->task_id = rqst->rq_task->tk_pid; __entry->client_id = rqst->rq_task->tk_client->cl_clid; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " peer=[%s]:%s", @@ -1001,8 +1001,8 @@ TRACE_EVENT(xprtrdma_post_recvs, __entry->cq_id = ep->re_attr.recv_cq->res.id; __entry->count = count; __entry->posted = ep->re_receive_count; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s cq.id=%d %u new recvs, %d active", @@ -1031,8 +1031,8 @@ TRACE_EVENT(xprtrdma_post_recvs_err, __entry->cq_id = ep->re_attr.recv_cq->res.id; __entry->status = status; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s cq.id=%d rc=%d", @@ -1445,8 +1445,8 @@ TRACE_EVENT(xprtrdma_cb_setup, TP_fast_assign( __entry->reqs = reqs; - __assign_str(addr, rpcrdma_addrstr(r_xprt)); - __assign_str(port, rpcrdma_portstr(r_xprt)); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s %u reqs", @@ -1476,7 +1476,7 @@ DECLARE_EVENT_CLASS(svcrdma_accept_class, TP_fast_assign( __entry->status = status; - __assign_str(addr, rdma->sc_xprt.xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s status=%ld", @@ -1962,7 +1962,7 @@ TRACE_EVENT(svcrdma_send_err, TP_fast_assign( __entry->status = status; __entry->xid = __be32_to_cpu(rqst->rq_xid); - __assign_str(addr, rqst->rq_xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x status=%d", __get_str(addr), @@ -2025,7 +2025,7 @@ TRACE_EVENT(svcrdma_rq_post_err, TP_fast_assign( __entry->status = status; - __assign_str(addr, rdma->sc_xprt.xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s status=%d", @@ -2138,7 +2138,7 @@ TRACE_EVENT(svcrdma_qp_error, TP_fast_assign( __entry->event = event->event; - __assign_str(device, event->device->name); + __assign_str(device); snprintf(__entry->addr, sizeof(__entry->addr) - 1, "%pISpc", sap); ), diff --git a/include/trace/events/rpm.h b/include/trace/events/rpm.h index bd120e23ce12..2b0b4b6ef862 100644 --- a/include/trace/events/rpm.h +++ b/include/trace/events/rpm.h @@ -33,7 +33,7 @@ DECLARE_EVENT_CLASS(rpm_internal, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->flags = flags; __entry->usage_count = atomic_read( &dev->power.usage_count); @@ -92,7 +92,7 @@ TRACE_EVENT(rpm_return_int, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->ip = ip; __entry->ret = ret; ), @@ -135,7 +135,7 @@ TRACE_EVENT(rpm_status, ), TP_fast_assign( - __assign_str(name, dev_name(dev)); + __assign_str(name); __entry->status = status; ), diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index 68973f650c26..6df2b4685b08 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -411,7 +411,7 @@ TRACE_EVENT(sched_process_exec, ), TP_fast_assign( - __assign_str(filename, bprm->filename); + __assign_str(filename); __entry->pid = p->pid; __entry->old_pid = old_pid; ), @@ -445,10 +445,10 @@ TRACE_EVENT(sched_prepare_exec, ), TP_fast_assign( - __assign_str(interp, bprm->interp); - __assign_str(filename, bprm->filename); + __assign_str(interp); + __assign_str(filename); __entry->pid = task->pid; - __assign_str(comm, task->comm); + __assign_str(comm); ), TP_printk("interp=%s filename=%s pid=%d comm=%s", diff --git a/include/trace/events/sof.h b/include/trace/events/sof.h index cd8e6844cca2..3061423c0667 100644 --- a/include/trace/events/sof.h +++ b/include/trace/events/sof.h @@ -23,7 +23,7 @@ DECLARE_EVENT_CLASS(sof_widget_template, __field(int, use_count) ), TP_fast_assign( - __assign_str(name, swidget->widget->name); + __assign_str(name); __entry->use_count = swidget->use_count; ), TP_printk("name=%s use_count=%d", __get_str(name), __entry->use_count) @@ -49,7 +49,7 @@ TRACE_EVENT(sof_ipc3_period_elapsed_position, __field(u64, wallclock) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->host_posn = posn->host_posn; __entry->dai_posn = posn->dai_posn; __entry->wallclock = posn->wallclock; @@ -75,7 +75,7 @@ TRACE_EVENT(sof_pcm_pointer_position, __field(unsigned long, dai_posn) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->pcm_id = le32_to_cpu(spcm->pcm.pcm_id); __entry->stream = substream->stream; __entry->dma_posn = dma_posn; @@ -93,7 +93,7 @@ TRACE_EVENT(sof_stream_position_ipc_rx, __string(device_name, dev_name(dev)) ), TP_fast_assign( - __assign_str(device_name, dev_name(dev)); + __assign_str(device_name); ), TP_printk("device_name=%s", __get_str(device_name)) ); @@ -107,8 +107,8 @@ TRACE_EVENT(sof_ipc4_fw_config, __field(u32, value) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); - __assign_str(key, key); + __assign_str(device_name); + __assign_str(key); __entry->value = value; ), TP_printk("device_name=%s key=%s value=%d", diff --git a/include/trace/events/sof_intel.h b/include/trace/events/sof_intel.h index 4cac5fef99a2..9e579e57b15c 100644 --- a/include/trace/events/sof_intel.h +++ b/include/trace/events/sof_intel.h @@ -22,8 +22,8 @@ TRACE_EVENT(sof_intel_hda_irq, __string(source, source) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); - __assign_str(source, source); + __assign_str(device_name); + __assign_str(source); ), TP_printk("device_name=%s source=%s", __get_str(device_name), __get_str(source)) @@ -38,7 +38,7 @@ DECLARE_EVENT_CLASS(sof_intel_ipc_firmware_template, __field(u32, msg_ext) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->msg = msg; __entry->msg_ext = msg_ext; ), @@ -64,7 +64,7 @@ TRACE_EVENT(sof_intel_D0I3C_updated, __field(u8, reg) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->reg = reg; ), TP_printk("device_name=%s register=%#x", @@ -79,7 +79,7 @@ TRACE_EVENT(sof_intel_hda_irq_ipc_check, __field(u32, irq_status) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->irq_status = irq_status; ), TP_printk("device_name=%s irq_status=%#x", @@ -100,7 +100,7 @@ TRACE_EVENT(sof_intel_hda_dsp_pcm, __field(unsigned long, pos) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->hstream_index = hstream->index; __entry->substream = substream->stream; __entry->pos = pos; @@ -119,7 +119,7 @@ TRACE_EVENT(sof_intel_hda_dsp_stream_status, __field(u32, status) ), TP_fast_assign( - __assign_str(device_name, dev_name(dev)); + __assign_str(device_name); __entry->stream = s->index; __entry->status = status; ), @@ -135,7 +135,7 @@ TRACE_EVENT(sof_intel_hda_dsp_check_stream_irq, __field(u32, status) ), TP_fast_assign( - __assign_str(device_name, dev_name(sdev->dev)); + __assign_str(device_name); __entry->status = status; ), TP_printk("device_name=%s status=%#x", diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index ac05ed06a071..5e8495216689 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -188,10 +188,10 @@ TRACE_EVENT(rpc_clnt_new, __entry->client_id = clnt->cl_clid; __entry->xprtsec = args->xprtsec.policy; __entry->flags = args->flags; - __assign_str(program, clnt->cl_program->name); - __assign_str(server, xprt->servername); - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(program); + __assign_str(server); + __assign_str(addr); + __assign_str(port); ), TP_printk("client=" SUNRPC_TRACE_CLID_SPECIFIER " peer=[%s]:%s" @@ -220,8 +220,8 @@ TRACE_EVENT(rpc_clnt_new_err, TP_fast_assign( __entry->error = error; - __assign_str(program, program); - __assign_str(server, server); + __assign_str(program); + __assign_str(server); ), TP_printk("program=%s server=%s error=%d", @@ -325,8 +325,8 @@ TRACE_EVENT(rpc_request, __entry->client_id = task->tk_client->cl_clid; __entry->version = task->tk_client->cl_vers; __entry->async = RPC_IS_ASYNC(task); - __assign_str(progname, task->tk_client->cl_program->name); - __assign_str(procname, rpc_proc_name(task)); + __assign_str(progname); + __assign_str(procname); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER " %sv%d %s (%ssync)", @@ -439,7 +439,7 @@ DECLARE_EVENT_CLASS(rpc_task_queued, __entry->runstate = task->tk_runstate; __entry->status = task->tk_status; __entry->flags = task->tk_flags; - __assign_str(q_name, rpc_qname(q)); + __assign_str(q_name); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -515,10 +515,10 @@ DECLARE_EVENT_CLASS(rpc_reply_event, __entry->task_id = task->tk_pid; __entry->client_id = task->tk_client->cl_clid; __entry->xid = be32_to_cpu(task->tk_rqstp->rq_xid); - __assign_str(progname, task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procname, rpc_proc_name(task)); - __assign_str(servername, task->tk_xprt->servername); + __assign_str(procname); + __assign_str(servername); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -647,8 +647,8 @@ TRACE_EVENT(rpc_stats_latency, __entry->task_id = task->tk_pid; __entry->xid = be32_to_cpu(task->tk_rqstp->rq_xid); __entry->version = task->tk_client->cl_vers; - __assign_str(progname, task->tk_client->cl_program->name); - __assign_str(procname, rpc_proc_name(task)); + __assign_str(progname); + __assign_str(procname); __entry->backlog = ktime_to_us(backlog); __entry->rtt = ktime_to_us(rtt); __entry->execute = ktime_to_us(execute); @@ -697,16 +697,15 @@ TRACE_EVENT(rpc_xdr_overflow, __entry->task_id = task->tk_pid; __entry->client_id = task->tk_client->cl_clid; - __assign_str(progname, - task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procedure, task->tk_msg.rpc_proc->p_name); + __assign_str(procedure); } else { __entry->task_id = -1; __entry->client_id = -1; - __assign_str(progname, "unknown"); + __assign_str(progname); __entry->version = 0; - __assign_str(procedure, "unknown"); + __assign_str(procedure); } __entry->requested = requested; __entry->end = xdr->end; @@ -763,10 +762,9 @@ TRACE_EVENT(rpc_xdr_alignment, __entry->task_id = task->tk_pid; __entry->client_id = task->tk_client->cl_clid; - __assign_str(progname, - task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procedure, task->tk_msg.rpc_proc->p_name); + __assign_str(procedure); __entry->offset = offset; __entry->copied = copied; @@ -1018,8 +1016,8 @@ DECLARE_EVENT_CLASS(rpc_xprt_lifetime_class, TP_fast_assign( __entry->state = xprt->state; - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s state=%s", @@ -1061,8 +1059,8 @@ DECLARE_EVENT_CLASS(rpc_xprt_event, TP_fast_assign( __entry->xid = be32_to_cpu(xid); __entry->status = status; - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s xid=0x%08x status=%d", __get_str(addr), @@ -1140,10 +1138,9 @@ TRACE_EVENT(xprt_retransmit, __entry->xid = be32_to_cpu(rqst->rq_xid); __entry->ntrans = rqst->rq_ntrans; __entry->timeout = task->tk_timeout; - __assign_str(progname, - task->tk_client->cl_program->name); + __assign_str(progname); __entry->version = task->tk_client->cl_vers; - __assign_str(procname, rpc_proc_name(task)); + __assign_str(procname); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -1167,8 +1164,8 @@ TRACE_EVENT(xprt_ping, TP_fast_assign( __entry->status = status; - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s status=%d", @@ -1315,8 +1312,8 @@ TRACE_EVENT(xs_data_ready, ), TP_fast_assign( - __assign_str(addr, xprt->address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xprt->address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s", __get_str(addr), __get_str(port)) @@ -1339,10 +1336,8 @@ TRACE_EVENT(xs_stream_read_data, TP_fast_assign( __entry->err = err; __entry->total = total; - __assign_str(addr, xprt ? - xprt->address_strings[RPC_DISPLAY_ADDR] : EVENT_NULL_STR); - __assign_str(port, xprt ? - xprt->address_strings[RPC_DISPLAY_PORT] : EVENT_NULL_STR); + __assign_str(addr); + __assign_str(port); ), TP_printk("peer=[%s]:%s err=%zd total=%zu", __get_str(addr), @@ -1364,8 +1359,8 @@ TRACE_EVENT(xs_stream_read_request, ), TP_fast_assign( - __assign_str(addr, xs->xprt.address_strings[RPC_DISPLAY_ADDR]); - __assign_str(port, xs->xprt.address_strings[RPC_DISPLAY_PORT]); + __assign_str(addr); + __assign_str(port); __entry->xid = be32_to_cpu(xs->recv.xid); __entry->copied = xs->recv.copied; __entry->reclen = xs->recv.len; @@ -1403,7 +1398,7 @@ TRACE_EVENT(rpcb_getport, __entry->version = clnt->cl_vers; __entry->protocol = task->tk_xprt->prot; __entry->bind_version = bind_version; - __assign_str(servername, task->tk_xprt->servername); + __assign_str(servername); ), TP_printk(SUNRPC_TRACE_TASK_SPECIFIER @@ -1493,8 +1488,8 @@ TRACE_EVENT(rpcb_register, TP_fast_assign( __entry->program = program; __entry->version = version; - __assign_str(addr, addr); - __assign_str(netid, netid); + __assign_str(addr); + __assign_str(netid); ), TP_printk("program=%u version=%u addr=%s netid=%s", @@ -1521,7 +1516,7 @@ TRACE_EVENT(rpcb_unregister, TP_fast_assign( __entry->program = program; __entry->version = version; - __assign_str(netid, netid); + __assign_str(netid); ), TP_printk("program=%u version=%u netid=%s", @@ -1551,8 +1546,8 @@ DECLARE_EVENT_CLASS(rpc_tls_class, TP_fast_assign( __entry->requested_policy = clnt->cl_xprtsec.policy; __entry->version = clnt->cl_vers; - __assign_str(servername, xprt->servername); - __assign_str(progname, clnt->cl_program->name) + __assign_str(servername); + __assign_str(progname); ), TP_printk("server=%s %sv%u requested_policy=%s", @@ -1794,10 +1789,9 @@ TRACE_EVENT(svc_process, __entry->xid = be32_to_cpu(rqst->rq_xid); __entry->vers = rqst->rq_vers; __entry->proc = rqst->rq_proc; - __assign_str(service, name); - __assign_str(procedure, svc_proc_name(rqst)); - __assign_str(addr, rqst->rq_xprt ? - rqst->rq_xprt->xpt_remotebuf : EVENT_NULL_STR); + __assign_str(service); + __assign_str(procedure); + __assign_str(addr); ), TP_printk("addr=%s xid=0x%08x service=%s vers=%u proc=%s", @@ -1915,7 +1909,7 @@ TRACE_EVENT(svc_stats_latency, __entry->execute = ktime_to_us(ktime_sub(ktime_get(), rqst->rq_stime)); - __assign_str(procedure, svc_proc_name(rqst)); + __assign_str(procedure); ), TP_printk(SVC_RQST_ENDPOINT_FORMAT " proc=%s execute-us=%lu", @@ -1980,8 +1974,8 @@ TRACE_EVENT(svc_xprt_create_err, TP_fast_assign( __entry->error = PTR_ERR(xprt); - __assign_str(program, program); - __assign_str(protocol, protocol); + __assign_str(program); + __assign_str(protocol); __assign_sockaddr(addr, sap, salen); ), @@ -2120,8 +2114,8 @@ TRACE_EVENT(svc_xprt_accept, TP_fast_assign( SVC_XPRT_ENDPOINT_ASSIGNMENTS(xprt); - __assign_str(protocol, xprt->xpt_class->xcl_name); - __assign_str(service, service); + __assign_str(protocol); + __assign_str(service); ), TP_printk(SVC_XPRT_ENDPOINT_FORMAT " protocol=%s service=%s", @@ -2260,7 +2254,7 @@ TRACE_EVENT(svcsock_marker, TP_fast_assign( __entry->length = be32_to_cpu(marker) & RPC_FRAGMENT_SIZE_MASK; __entry->last = be32_to_cpu(marker) & RPC_LAST_STREAM_FRAGMENT; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s length=%u%s", __get_str(addr), @@ -2284,7 +2278,7 @@ DECLARE_EVENT_CLASS(svcsock_class, TP_fast_assign( __entry->result = result; __entry->flags = xprt->xpt_flags; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s result=%zd flags=%s", __get_str(addr), @@ -2330,7 +2324,7 @@ TRACE_EVENT(svcsock_tcp_recv_short, __entry->expected = expected; __entry->received = received; __entry->flags = xprt->xpt_flags; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s flags=%s expected=%u received=%u", @@ -2358,7 +2352,7 @@ TRACE_EVENT(svcsock_tcp_state, __entry->socket_state = socket->state; __entry->sock_state = socket->sk->sk_state; __entry->flags = xprt->xpt_flags; - __assign_str(addr, xprt->xpt_remotebuf); + __assign_str(addr); ), TP_printk("addr=%s state=%s sk_state=%s flags=%s", __get_str(addr), @@ -2385,7 +2379,7 @@ DECLARE_EVENT_CLASS(svcsock_accept_class, TP_fast_assign( __entry->status = status; - __assign_str(service, service); + __assign_str(service); __entry->netns_ino = xprt->xpt_net->ns.inum; ), @@ -2421,7 +2415,7 @@ DECLARE_EVENT_CLASS(cache_event, TP_fast_assign( __entry->h = h; - __assign_str(name, cd->name); + __assign_str(name); ), TP_printk("cache=%s entry=%p", __get_str(name), __entry->h) @@ -2466,7 +2460,7 @@ DECLARE_EVENT_CLASS(register_class, __entry->protocol = protocol; __entry->port = port; __entry->error = error; - __assign_str(program, program); + __assign_str(program); ), TP_printk("program=%sv%u proto=%s port=%u family=%s error=%d", @@ -2511,7 +2505,7 @@ TRACE_EVENT(svc_unregister, TP_fast_assign( __entry->version = version; __entry->error = error; - __assign_str(program, program); + __assign_str(program); ), TP_printk("program=%sv%u error=%d", diff --git a/include/trace/events/swiotlb.h b/include/trace/events/swiotlb.h index da05c9ebd224..3b6ddb136e4e 100644 --- a/include/trace/events/swiotlb.h +++ b/include/trace/events/swiotlb.h @@ -20,7 +20,7 @@ TRACE_EVENT(swiotlb_bounced, ), TP_fast_assign( - __assign_str(dev_name, dev_name(dev)); + __assign_str(dev_name); __entry->dma_mask = (dev->dma_mask ? *dev->dma_mask : 0); __entry->dev_addr = dev_addr; __entry->size = size; diff --git a/include/trace/events/target.h b/include/trace/events/target.h index 67fad2677ed5..a13cbf2b3405 100644 --- a/include/trace/events/target.h +++ b/include/trace/events/target.h @@ -154,7 +154,7 @@ TRACE_EVENT(target_sequencer_start, __entry->task_attribute = cmd->sam_task_attr; __entry->control = scsi_command_control(cmd->t_task_cdb); memcpy(__entry->cdb, cmd->t_task_cdb, TCM_MAX_COMMAND_SIZE); - __assign_str(initiator, cmd->se_sess->se_node_acl->initiatorname); + __assign_str(initiator); ), TP_printk("%s -> LUN %03u tag %#llx %s data_length %6u CDB %s (TA:%s C:%02x)", @@ -198,7 +198,7 @@ TRACE_EVENT(target_cmd_complete, min(18, ((u8 *) cmd->sense_buffer)[SPC_ADD_SENSE_LEN_OFFSET] + 8) : 0; memcpy(__entry->cdb, cmd->t_task_cdb, TCM_MAX_COMMAND_SIZE); memcpy(__entry->sense_data, cmd->sense_buffer, __entry->sense_length); - __assign_str(initiator, cmd->se_sess->se_node_acl->initiatorname); + __assign_str(initiator); ), TP_printk("%s <- LUN %03u tag %#llx status %s (sense len %d%s%s) %s data_length %6u CDB %s (TA:%s C:%02x)", diff --git a/include/trace/events/tegra_apb_dma.h b/include/trace/events/tegra_apb_dma.h index 971cd02d2daf..6d9f5075baa3 100644 --- a/include/trace/events/tegra_apb_dma.h +++ b/include/trace/events/tegra_apb_dma.h @@ -16,7 +16,7 @@ TRACE_EVENT(tegra_dma_tx_status, __field(__u32, residue) ), TP_fast_assign( - __assign_str(chan, dev_name(&dc->dev->device)); + __assign_str(chan); __entry->cookie = cookie; __entry->residue = state ? state->residue : (u32)-1; ), @@ -33,7 +33,7 @@ TRACE_EVENT(tegra_dma_complete_cb, __field(void *, ptr) ), TP_fast_assign( - __assign_str(chan, dev_name(&dc->dev->device)); + __assign_str(chan); __entry->count = count; __entry->ptr = ptr; ), @@ -49,7 +49,7 @@ TRACE_EVENT(tegra_dma_isr, __field(int, irq) ), TP_fast_assign( - __assign_str(chan, dev_name(&dc->dev->device)); + __assign_str(chan); __entry->irq = irq; ), TP_printk("%s: irq %d\n", __get_str(chan), __entry->irq) diff --git a/include/trace/events/ufs.h b/include/trace/events/ufs.h index b930669bd1f0..c4e209fbdfbb 100644 --- a/include/trace/events/ufs.h +++ b/include/trace/events/ufs.h @@ -92,7 +92,7 @@ TRACE_EVENT(ufshcd_clk_gating, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->state = state; ), @@ -117,9 +117,9 @@ TRACE_EVENT(ufshcd_clk_scaling, ), TP_fast_assign( - __assign_str(dev_name, dev_name); - __assign_str(state, state); - __assign_str(clk, clk); + __assign_str(dev_name); + __assign_str(state); + __assign_str(clk); __entry->prev_state = prev_state; __entry->curr_state = curr_state; ), @@ -141,8 +141,8 @@ TRACE_EVENT(ufshcd_auto_bkops_state, ), TP_fast_assign( - __assign_str(dev_name, dev_name); - __assign_str(state, state); + __assign_str(dev_name); + __assign_str(state); ), TP_printk("%s: auto bkops - %s", @@ -163,8 +163,8 @@ DECLARE_EVENT_CLASS(ufshcd_profiling_template, ), TP_fast_assign( - __assign_str(dev_name, dev_name); - __assign_str(profile_info, profile_info); + __assign_str(dev_name); + __assign_str(profile_info); __entry->time_us = time_us; __entry->err = err; ), @@ -206,7 +206,7 @@ DECLARE_EVENT_CLASS(ufshcd_template, TP_fast_assign( __entry->usecs = usecs; __entry->err = err; - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->dev_state = dev_state; __entry->link_state = link_state; ), @@ -326,7 +326,7 @@ TRACE_EVENT(ufshcd_uic_command, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->str_t = str_t; __entry->cmd = cmd; __entry->arg1 = arg1; @@ -356,7 +356,7 @@ TRACE_EVENT(ufshcd_upiu, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->str_t = str_t; memcpy(__entry->hdr, hdr, sizeof(__entry->hdr)); memcpy(__entry->tsf, tsf, sizeof(__entry->tsf)); @@ -384,7 +384,7 @@ TRACE_EVENT(ufshcd_exception_event, ), TP_fast_assign( - __assign_str(dev_name, dev_name); + __assign_str(dev_name); __entry->status = status; ), diff --git a/include/trace/events/workqueue.h b/include/trace/events/workqueue.h index 6ef5b7254070..b0de2bc9ed52 100644 --- a/include/trace/events/workqueue.h +++ b/include/trace/events/workqueue.h @@ -38,7 +38,7 @@ TRACE_EVENT(workqueue_queue_work, TP_fast_assign( __entry->work = work; __entry->function = work->func; - __assign_str(workqueue, pwq->wq->name); + __assign_str(workqueue); __entry->req_cpu = req_cpu; __entry->cpu = pwq->pool->cpu; ), diff --git a/include/trace/events/xdp.h b/include/trace/events/xdp.h index 9adc2bdf2f94..a7e5452b5d21 100644 --- a/include/trace/events/xdp.h +++ b/include/trace/events/xdp.h @@ -416,7 +416,7 @@ TRACE_EVENT(bpf_xdp_link_attach_failed, ), TP_fast_assign( - __assign_str(msg, msg); + __assign_str(msg); ), TP_printk("errmsg=%s", __get_str(msg)) diff --git a/include/trace/stages/stage6_event_callback.h b/include/trace/stages/stage6_event_callback.h index 3690e677263f..1691676fd858 100644 --- a/include/trace/stages/stage6_event_callback.h +++ b/include/trace/stages/stage6_event_callback.h @@ -31,12 +31,10 @@ #define __vstring(item, fmt, ap) __dynamic_array(char, item, -1) #undef __assign_str -#define __assign_str(dst, src) \ +#define __assign_str(dst) \ do { \ char *__str__ = __get_str(dst); \ int __len__ = __get_dynamic_array_len(dst) - 1; \ - WARN_ON_ONCE(!(void *)(src) != !(void *)__data_offsets.dst##_ptr_); \ - WARN_ON_ONCE((src) && strcmp((src), __data_offsets.dst##_ptr_)); \ memcpy(__str__, __data_offsets.dst##_ptr_ ? : \ EVENT_NULL_STR, __len__); \ __str__[__len__] = '\0'; \ diff --git a/kernel/trace/bpf_trace.h b/kernel/trace/bpf_trace.h index 9acbc11ac7bb..c4075b56becc 100644 --- a/kernel/trace/bpf_trace.h +++ b/kernel/trace/bpf_trace.h @@ -19,7 +19,7 @@ TRACE_EVENT(bpf_trace_printk, ), TP_fast_assign( - __assign_str(bpf_string, bpf_string); + __assign_str(bpf_string); ), TP_printk("%s", __get_str(bpf_string)) diff --git a/net/batman-adv/trace.h b/net/batman-adv/trace.h index 5dd52bc5cabb..6b816cf1a953 100644 --- a/net/batman-adv/trace.h +++ b/net/batman-adv/trace.h @@ -40,8 +40,8 @@ TRACE_EVENT(batadv_dbg, ), TP_fast_assign( - __assign_str(device, bat_priv->soft_iface->name); - __assign_str(driver, KBUILD_MODNAME); + __assign_str(device); + __assign_str(driver); __assign_vstr(msg, vaf->fmt, vaf->va); ), diff --git a/net/dsa/trace.h b/net/dsa/trace.h index 567f29a39707..83f3e5f78491 100644 --- a/net/dsa/trace.h +++ b/net/dsa/trace.h @@ -39,8 +39,8 @@ DECLARE_EVENT_CLASS(dsa_port_addr_op_hw, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; ether_addr_copy(__entry->addr, addr); __entry->vid = vid; @@ -98,8 +98,8 @@ DECLARE_EVENT_CLASS(dsa_port_addr_op_refcount, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; ether_addr_copy(__entry->addr, addr); __entry->vid = vid; @@ -157,8 +157,8 @@ DECLARE_EVENT_CLASS(dsa_port_addr_del_not_found, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; ether_addr_copy(__entry->addr, addr); __entry->vid = vid; @@ -199,7 +199,7 @@ TRACE_EVENT(dsa_lag_fdb_add_hw, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -227,7 +227,7 @@ TRACE_EVENT(dsa_lag_fdb_add_bump, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -255,7 +255,7 @@ TRACE_EVENT(dsa_lag_fdb_del_hw, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -283,7 +283,7 @@ TRACE_EVENT(dsa_lag_fdb_del_drop, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -310,7 +310,7 @@ TRACE_EVENT(dsa_lag_fdb_del_not_found, ), TP_fast_assign( - __assign_str(dev, lag_dev->name); + __assign_str(dev); ether_addr_copy(__entry->addr, addr); __entry->vid = vid; dsa_db_print(db, __entry->db_buf); @@ -338,8 +338,8 @@ DECLARE_EVENT_CLASS(dsa_vlan_op_hw, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; __entry->vid = vlan->vid; __entry->flags = vlan->flags; @@ -383,8 +383,8 @@ DECLARE_EVENT_CLASS(dsa_vlan_op_refcount, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; __entry->vid = vlan->vid; __entry->flags = vlan->flags; @@ -426,8 +426,8 @@ TRACE_EVENT(dsa_vlan_del_not_found, ), TP_fast_assign( - __assign_str(dev, dev_name(dp->ds->dev)); - __assign_str(kind, dsa_port_kind(dp)); + __assign_str(dev); + __assign_str(kind); __entry->port = dp->index; __entry->vid = vlan->vid; ), diff --git a/net/ieee802154/trace.h b/net/ieee802154/trace.h index 62aa6465253a..591ce0a16fc0 100644 --- a/net/ieee802154/trace.h +++ b/net/ieee802154/trace.h @@ -75,7 +75,7 @@ TRACE_EVENT(802154_rdev_add_virtual_intf, ), TP_fast_assign( WPAN_PHY_ASSIGN; - __assign_str(vir_intf_name, name ? name : ""); + __assign_str(vir_intf_name); __entry->type = type; __entry->extended_addr = extended_addr; ), diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h index 8e758b5074bd..b26aacfbc622 100644 --- a/net/mac80211/trace.h +++ b/net/mac80211/trace.h @@ -33,7 +33,7 @@ __string(vif_name, sdata->name) #define VIF_ASSIGN __entry->vif_type = sdata->vif.type; __entry->sdata = sdata; \ __entry->p2p = sdata->vif.p2p; \ - __assign_str(vif_name, sdata->name) + __assign_str(vif_name) #define VIF_PR_FMT " vif:%s(%d%s)" #define VIF_PR_ARG __get_str(vif_name), __entry->vif_type, __entry->p2p ? "/p2p" : "" diff --git a/net/openvswitch/openvswitch_trace.h b/net/openvswitch/openvswitch_trace.h index 3eb35d9eb700..74d75aaebef4 100644 --- a/net/openvswitch/openvswitch_trace.h +++ b/net/openvswitch/openvswitch_trace.h @@ -43,8 +43,8 @@ TRACE_EVENT(ovs_do_execute_action, TP_fast_assign( __entry->dpaddr = dp; - __assign_str(dp_name, ovs_dp_name(dp)); - __assign_str(dev_name, skb->dev->name); + __assign_str(dp_name); + __assign_str(dev_name); __entry->skbaddr = skb; __entry->len = skb->len; __entry->data_len = skb->data_len; @@ -113,8 +113,8 @@ TRACE_EVENT(ovs_dp_upcall, TP_fast_assign( __entry->dpaddr = dp; - __assign_str(dp_name, ovs_dp_name(dp)); - __assign_str(dev_name, skb->dev->name); + __assign_str(dp_name); + __assign_str(dev_name); __entry->skbaddr = skb; __entry->len = skb->len; __entry->data_len = skb->data_len; diff --git a/net/smc/smc_tracepoint.h b/net/smc/smc_tracepoint.h index 9fc5e586d24a..a9a6e3c1113a 100644 --- a/net/smc/smc_tracepoint.h +++ b/net/smc/smc_tracepoint.h @@ -60,7 +60,7 @@ DECLARE_EVENT_CLASS(smc_msg_event, __entry->smc = smc; __entry->net_cookie = sock_net(sk)->net_cookie; __entry->len = len; - __assign_str(name, smc->conn.lnk->ibname); + __assign_str(name); ), TP_printk("smc=%p net=%llu len=%zu dev=%s", @@ -104,7 +104,7 @@ TRACE_EVENT(smcr_link_down, __entry->lgr = lgr; __entry->net_cookie = lgr->net->net_cookie; __entry->state = lnk->state; - __assign_str(name, lnk->ibname); + __assign_str(name); __entry->location = location; ), diff --git a/net/tipc/trace.h b/net/tipc/trace.h index 04af83f0500c..865142ed0ab4 100644 --- a/net/tipc/trace.h +++ b/net/tipc/trace.h @@ -145,7 +145,7 @@ DECLARE_EVENT_CLASS(tipc_skb_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); tipc_skb_dump(skb, more, __get_str(buf)); ), @@ -172,7 +172,7 @@ DECLARE_EVENT_CLASS(tipc_list_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); tipc_list_dump(list, more, __get_str(buf)); ), @@ -200,7 +200,7 @@ DECLARE_EVENT_CLASS(tipc_sk_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); __entry->portid = tipc_sock_get_portid(sk); tipc_sk_dump(sk, dqueues, __get_str(buf)); if (skb) @@ -254,7 +254,7 @@ DECLARE_EVENT_CLASS(tipc_link_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); memcpy(__entry->name, tipc_link_name(l), TIPC_MAX_LINK_NAME); tipc_link_dump(l, dqueues, __get_str(buf)); ), @@ -337,7 +337,7 @@ DECLARE_EVENT_CLASS(tipc_node_class, ), TP_fast_assign( - __assign_str(header, header); + __assign_str(header); __entry->addr = tipc_node_get_addr(n); tipc_node_dump(n, more, __get_str(buf)); ), @@ -374,7 +374,7 @@ DECLARE_EVENT_CLASS(tipc_fsm_class, ), TP_fast_assign( - __assign_str(name, name); + __assign_str(name); __entry->os = os; __entry->ns = ns; __entry->evt = evt; @@ -409,8 +409,8 @@ TRACE_EVENT(tipc_l2_device_event, ), TP_fast_assign( - __assign_str(dev_name, dev->name); - __assign_str(b_name, b->name); + __assign_str(dev_name); + __assign_str(b_name); __entry->evt = evt; __entry->b_up = test_bit(0, &b->up); __entry->carrier = netif_carrier_ok(dev); diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 9bf987519811..87986170d1b1 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -372,7 +372,7 @@ TRACE_EVENT(rdev_add_virtual_intf, ), TP_fast_assign( WIPHY_ASSIGN; - __assign_str(vir_intf_name, name ? name : ""); + __assign_str(vir_intf_name); __entry->type = type; ), TP_printk(WIPHY_PR_FMT ", virtual intf name: %s, type: %d", diff --git a/samples/trace_events/trace-events-sample.h b/samples/trace_events/trace-events-sample.h index 500981eca74d..55f9a3da92d5 100644 --- a/samples/trace_events/trace-events-sample.h +++ b/samples/trace_events/trace-events-sample.h @@ -136,10 +136,11 @@ * * To assign a string, use the helper macro __assign_str(). * - * __assign_str(foo, bar); + * __assign_str(foo); * - * In most cases, the __assign_str() macro will take the same - * parameters as the __string() macro had to declare the string. + * The __string() macro saves off the string that is passed into + * the second parameter, and the __assign_str() will store than + * saved string into the "foo" field. * * __vstring: This is similar to __string() but instead of taking a * dynamic length, it takes a variable list va_list 'va' variable. @@ -177,7 +178,7 @@ * The length is saved via the __string_len() and is retrieved in * __assign_str(). * - * __assign_str(foo, bar); + * __assign_str(foo); * * Then len + 1 is allocated to the ring buffer, and a nul terminating * byte is added. This is similar to: @@ -311,8 +312,8 @@ TRACE_EVENT(foo_bar, __entry->bar = bar; memcpy(__get_dynamic_array(list), lst, __length_of(lst) * sizeof(int)); - __assign_str(str, string); - __assign_str(lstr, foo); + __assign_str(str); + __assign_str(lstr); __assign_vstr(vstr, fmt, va); __assign_bitmask(cpus, cpumask_bits(mask), num_possible_cpus()); __assign_cpumask(cpum, cpumask_bits(mask)); @@ -418,7 +419,7 @@ TRACE_EVENT_CONDITION(foo_bar_with_cond, ), TP_fast_assign( - __assign_str(foo, foo); + __assign_str(foo); __entry->bar = bar; ), @@ -459,7 +460,7 @@ TRACE_EVENT_FN(foo_bar_with_fn, ), TP_fast_assign( - __assign_str(foo, foo); + __assign_str(foo); __entry->bar = bar; ), @@ -506,7 +507,7 @@ DECLARE_EVENT_CLASS(foo_template, ), TP_fast_assign( - __assign_str(foo, foo); + __assign_str(foo); __entry->bar = bar; ), diff --git a/sound/core/pcm_trace.h b/sound/core/pcm_trace.h index 350b40b906ca..adb9b1f3bbfa 100644 --- a/sound/core/pcm_trace.h +++ b/sound/core/pcm_trace.h @@ -95,7 +95,7 @@ TRACE_EVENT(hw_ptr_error, __entry->device = (substream)->pcm->device; __entry->number = (substream)->number; __entry->stream = (substream)->stream; - __assign_str(reason, why); + __assign_str(reason); ), TP_printk("pcmC%dD%d%s/sub%d: ERROR: %s", __entry->card, __entry->device, diff --git a/sound/hda/trace.h b/sound/hda/trace.h index 2cc493434a8f..280c42f3eb75 100644 --- a/sound/hda/trace.h +++ b/sound/hda/trace.h @@ -24,7 +24,7 @@ TRACE_EVENT(hda_send_cmd, __field(u32, cmd) ), TP_fast_assign( - __assign_str(name, dev_name((bus)->dev)); + __assign_str(name); __entry->cmd = cmd; ), TP_printk("[%s:%d] val=0x%08x", __get_str(name), __entry->cmd >> 28, __entry->cmd) @@ -39,7 +39,7 @@ TRACE_EVENT(hda_get_response, __field(u32, res) ), TP_fast_assign( - __assign_str(name, dev_name((bus)->dev)); + __assign_str(name); __entry->addr = addr; __entry->res = res; ), @@ -55,7 +55,7 @@ TRACE_EVENT(hda_unsol_event, __field(u32, res_ex) ), TP_fast_assign( - __assign_str(name, dev_name((bus)->dev)); + __assign_str(name); __entry->res = res; __entry->res_ex = res_ex; ), diff --git a/sound/soc/intel/avs/trace.h b/sound/soc/intel/avs/trace.h index 855b06bb14b0..c9eaa5a60ed3 100644 --- a/sound/soc/intel/avs/trace.h +++ b/sound/soc/intel/avs/trace.h @@ -24,7 +24,7 @@ TRACE_EVENT(avs_dsp_core_op, TP_fast_assign( __entry->reg = reg; __entry->mask = mask; - __assign_str(op, op); + __assign_str(op); __entry->flag = flag; ), @@ -135,7 +135,7 @@ TRACE_EVENT(avs_d0ix, ), TP_fast_assign( - __assign_str(op, op); + __assign_str(op); __entry->proceed = proceed; __entry->header = header; ), -- cgit v1.2.3 From 91e61dd7a0af660408e87372d8330ceb218be302 Mon Sep 17 00:00:00 2001 From: "Dae R. Jeong" Date: Tue, 21 May 2024 19:34:38 +0900 Subject: tls: fix missing memory barrier in tls_init In tls_init(), a write memory barrier is missing, and store-store reordering may cause NULL dereference in tls_{setsockopt,getsockopt}. CPU0 CPU1 ----- ----- // In tls_init() // In tls_ctx_create() ctx = kzalloc() ctx->sk_proto = READ_ONCE(sk->sk_prot) -(1) // In update_sk_prot() WRITE_ONCE(sk->sk_prot, tls_prots) -(2) // In sock_common_setsockopt() READ_ONCE(sk->sk_prot)->setsockopt() // In tls_{setsockopt,getsockopt}() ctx->sk_proto->setsockopt() -(3) In the above scenario, when (1) and (2) are reordered, (3) can observe the NULL value of ctx->sk_proto, causing NULL dereference. To fix it, we rely on rcu_assign_pointer() which implies the release barrier semantic. By moving rcu_assign_pointer() after ctx->sk_proto is initialized, we can ensure that ctx->sk_proto are visible when changing sk->sk_prot. Fixes: d5bee7374b68 ("net/tls: Annotate access to sk_prot with READ_ONCE/WRITE_ONCE") Signed-off-by: Yewon Choi Signed-off-by: Dae R. Jeong Link: https://lore.kernel.org/netdev/ZU4OJG56g2V9z_H7@dragonet/T/ Link: https://lore.kernel.org/r/Zkx4vjSFp0mfpjQ2@libra05 Signed-off-by: Paolo Abeni --- net/tls/tls_main.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index b4674f03d71a..90b7f253d363 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -816,9 +816,17 @@ struct tls_context *tls_ctx_create(struct sock *sk) return NULL; mutex_init(&ctx->tx_lock); - rcu_assign_pointer(icsk->icsk_ulp_data, ctx); ctx->sk_proto = READ_ONCE(sk->sk_prot); ctx->sk = sk; + /* Release semantic of rcu_assign_pointer() ensures that + * ctx->sk_proto is visible before changing sk->sk_prot in + * update_sk_prot(), and prevents reading uninitialized value in + * tls_{getsockopt, setsockopt}. Note that we do not need a + * read barrier in tls_{getsockopt,setsockopt} as there is an + * address dependency between sk->sk_proto->{getsockopt,setsockopt} + * and ctx->sk_proto. + */ + rcu_assign_pointer(icsk->icsk_ulp_data, ctx); return ctx; } -- cgit v1.2.3 From 378979e94e953c2070acb4f0e0c98d29260bd09d Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Tue, 21 May 2024 21:42:20 +0800 Subject: tcp: remove 64 KByte limit for initial tp->rcv_wnd value Recently, we had some servers upgraded to the latest kernel and noticed the indicator from the user side showed worse results than before. It is caused by the limitation of tp->rcv_wnd. In 2018 commit a337531b942b ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB") limited the initial value of tp->rcv_wnd to 65535, most CDN teams would not benefit from this change because they cannot have a large window to receive a big packet, which will be slowed down especially in long RTT. Small rcv_wnd means slow transfer speed, to some extent. It's the side effect for the latency/time-sensitive users. To avoid future confusion, current change doesn't affect the initial receive window on the wire in a SYN or SYN+ACK packet which are set within 65535 bytes according to RFC 7323 also due to the limit in __tcp_transmit_skb(): th->window = htons(min(tp->rcv_wnd, 65535U)); In one word, __tcp_transmit_skb() already ensures that constraint is respected, no matter how large tp->rcv_wnd is. The change doesn't violate RFC. Let me provide one example if with or without the patch: Before: client --- SYN: rwindow=65535 ---> server client <--- SYN+ACK: rwindow=65535 ---- server client --- ACK: rwindow=65536 ---> server Note: for the last ACK, the calculation is 512 << 7. After: client --- SYN: rwindow=65535 ---> server client <--- SYN+ACK: rwindow=65535 ---- server client --- ACK: rwindow=175232 ---> server Note: I use the following command to make it work: ip route change default via [ip] dev eth0 metric 100 initrwnd 120 For the last ACK, the calculation is 1369 << 7. When we apply such a patch, having a large rcv_wnd if the user tweak this knob can help transfer data more rapidly and save some rtts. Fixes: a337531b942b ("tcp: up initial rmem to 128KB and SYN rwin to around 64KB") Signed-off-by: Jason Xing Reviewed-by: Eric Dumazet Acked-by: Neal Cardwell Link: https://lore.kernel.org/r/20240521134220.12510-1-kerneljasonxing@gmail.com Signed-off-by: Paolo Abeni --- net/ipv4/tcp_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 95caf8aaa8be..95618d0e78e4 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -232,7 +232,7 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows)) (*rcv_wnd) = min(space, MAX_TCP_WINDOW); else - (*rcv_wnd) = min_t(u32, space, U16_MAX); + (*rcv_wnd) = space; if (init_rcv_wnd) *rcv_wnd = min(*rcv_wnd, init_rcv_wnd * mss); -- cgit v1.2.3 From 26afda78cda3da974fd4c287962c169e9462c495 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 21 May 2024 16:01:00 +0200 Subject: net: relax socket state check at accept time. Christoph reported the following splat: WARNING: CPU: 1 PID: 772 at net/ipv4/af_inet.c:761 __inet_accept+0x1f4/0x4a0 Modules linked in: CPU: 1 PID: 772 Comm: syz-executor510 Not tainted 6.9.0-rc7-g7da7119fe22b #56 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.0-2.el7 04/01/2014 RIP: 0010:__inet_accept+0x1f4/0x4a0 net/ipv4/af_inet.c:759 Code: 04 38 84 c0 0f 85 87 00 00 00 41 c7 04 24 03 00 00 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc e8 ec b7 da fd <0f> 0b e9 7f fe ff ff e8 e0 b7 da fd 0f 0b e9 fe fe ff ff 89 d9 80 RSP: 0018:ffffc90000c2fc58 EFLAGS: 00010293 RAX: ffffffff836bdd14 RBX: 0000000000000000 RCX: ffff888104668000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: dffffc0000000000 R08: ffffffff836bdb89 R09: fffff52000185f64 R10: dffffc0000000000 R11: fffff52000185f64 R12: dffffc0000000000 R13: 1ffff92000185f98 R14: ffff88810754d880 R15: ffff8881007b7800 FS: 000000001c772880(0000) GS:ffff88811b280000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fb9fcf2e178 CR3: 00000001045d2002 CR4: 0000000000770ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: inet_accept+0x138/0x1d0 net/ipv4/af_inet.c:786 do_accept+0x435/0x620 net/socket.c:1929 __sys_accept4_file net/socket.c:1969 [inline] __sys_accept4+0x9b/0x110 net/socket.c:1999 __do_sys_accept net/socket.c:2016 [inline] __se_sys_accept net/socket.c:2013 [inline] __x64_sys_accept+0x7d/0x90 net/socket.c:2013 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0x58/0x100 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x4315f9 Code: fd ff 48 81 c4 80 00 00 00 e9 f1 fe ff ff 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 ab b4 fd ff c3 66 2e 0f 1f 84 00 00 00 00 RSP: 002b:00007ffdb26d9c78 EFLAGS: 00000246 ORIG_RAX: 000000000000002b RAX: ffffffffffffffda RBX: 0000000000400300 RCX: 00000000004315f9 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000004 RBP: 00000000006e1018 R08: 0000000000400300 R09: 0000000000400300 R10: 0000000000400300 R11: 0000000000000246 R12: 0000000000000000 R13: 000000000040cdf0 R14: 000000000040ce80 R15: 0000000000000055 The reproducer invokes shutdown() before entering the listener status. After commit 94062790aedb ("tcp: defer shutdown(SEND_SHUTDOWN) for TCP_SYN_RECV sockets"), the above causes the child to reach the accept syscall in FIN_WAIT1 status. Eric noted we can relax the existing assertion in __inet_accept() Reported-by: Christoph Paasch Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/490 Suggested-by: Eric Dumazet Fixes: 94062790aedb ("tcp: defer shutdown(SEND_SHUTDOWN) for TCP_SYN_RECV sockets") Reviewed-by: Eric Dumazet Link: https://lore.kernel.org/r/23ab880a44d8cfd967e84de8b93dbf48848e3d8c.1716299669.git.pabeni@redhat.com Signed-off-by: Paolo Abeni --- net/ipv4/af_inet.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'net') diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 44564d009e95..86cce7e9c2d1 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -758,7 +758,9 @@ void __inet_accept(struct socket *sock, struct socket *newsock, struct sock *new sock_rps_record_flow(newsk); WARN_ON(!((1 << newsk->sk_state) & (TCPF_ESTABLISHED | TCPF_SYN_RECV | - TCPF_CLOSE_WAIT | TCPF_CLOSE))); + TCPF_FIN_WAIT1 | TCPF_FIN_WAIT2 | + TCPF_CLOSING | TCPF_CLOSE_WAIT | + TCPF_CLOSE))); if (test_bit(SOCK_SUPPORT_ZC, &sock->flags)) set_bit(SOCK_SUPPORT_ZC, &newsock->flags); -- cgit v1.2.3 From 6671e352497ca4bb07a96c48e03907065ff77d8a Mon Sep 17 00:00:00 2001 From: Ryosuke Yasuoka Date: Wed, 22 May 2024 00:34:42 +0900 Subject: nfc: nci: Fix handling of zero-length payload packets in nci_rx_work() When nci_rx_work() receives a zero-length payload packet, it should not discard the packet and exit the loop. Instead, it should continue processing subsequent packets. Fixes: d24b03535e5e ("nfc: nci: Fix uninit-value in nci_dev_up and nci_ntf_packet") Signed-off-by: Ryosuke Yasuoka Reviewed-by: Simon Horman Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20240521153444.535399-1-ryasuoka@redhat.com Signed-off-by: Paolo Abeni --- net/nfc/nci/core.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c index 7a9897fbf4f4..f456a5911e7d 100644 --- a/net/nfc/nci/core.c +++ b/net/nfc/nci/core.c @@ -1531,8 +1531,7 @@ static void nci_rx_work(struct work_struct *work) if (!nci_valid_size(skb)) { kfree_skb(skb); - kcov_remote_stop(); - break; + continue; } /* Process frame */ -- cgit v1.2.3