From d728c8ef8bea6e81f44933c0237531cda499577e Mon Sep 17 00:00:00 2001 From: Brad Volkin Date: Tue, 18 Feb 2014 10:15:56 -0800 Subject: drm/i915: Add a CMD_PARSER_VERSION getparam So userspace can query the kernel for command parser support. v2: Add i915_cmd_parser_get_version(), history log, and kerneldoc OTC-Tracker: AXIA-4631 Change-Id: I58af650db9f6753c2dcac9c54ab432fd31db302f Signed-off-by: Brad Volkin Reviewed-by: Jani Nikula Signed-off-by: Daniel Vetter --- include/uapi/drm/i915_drm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi') diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 126bfaa8bb6b..8a3e4ef00c3d 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -337,6 +337,7 @@ typedef struct drm_i915_irq_wait { #define I915_PARAM_HAS_EXEC_NO_RELOC 25 #define I915_PARAM_HAS_EXEC_HANDLE_LUT 26 #define I915_PARAM_HAS_WT 27 +#define I915_PARAM_CMD_PARSER_VERSION 28 typedef struct drm_i915_getparam { int param; -- cgit v1.2.3 From c50b960ccc5981627628302701e93e6aceccdb1c Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 28 Mar 2014 10:19:47 +0000 Subject: netfilter: nf_tables: implement proper set selection The current set selection simply choses the first set type that provides the requested features, which always results in the rbtree being chosen by virtue of being the first set in the list. What we actually want to do is choose the implementation that can provide the requested features and is optimal from either a performance or memory perspective depending on the characteristics of the elements and the preferences specified by the user. The elements are not known when creating a set. Even if we would provide them for anonymous (literal) sets, we'd still have standalone sets where the elements are not known in advance. We therefore need an abstract description of the data charcteristics. The kernel already knows the size of the key, this patch starts by introducing a nested set description which so far contains only the maximum amount of elements. Based on this the set implementations are changed to provide an estimate of the required amount of memory and the lookup complexity class. The set ops have a new callback ->estimate() that is invoked during set selection. It receives a structure containing the attributes known to the kernel and is supposed to populate a struct nft_set_estimate with the complexity class and, in case the size is known, the complete amount of memory required, or the amount of memory required per element otherwise. Based on the policy specified by the user (performance/memory, defaulting to performance) the kernel will then select the best suited implementation. Even if the set implementation would allow to add more than the specified maximum amount of elements, they are enforced since new implementations might not be able to add more than maximum based on which they were selected. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 46 ++++++++++++ include/uapi/linux/netfilter/nf_tables.h | 27 +++++++ net/netfilter/nf_tables_api.c | 121 +++++++++++++++++++++++++++---- net/netfilter/nft_hash.c | 45 +++++++++++- net/netfilter/nft_rbtree.c | 21 ++++++ 5 files changed, 242 insertions(+), 18 deletions(-) (limited to 'include/uapi') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index e6bc14d8fa9a..29ff1dc41ef3 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -145,6 +145,44 @@ struct nft_set_iter { const struct nft_set_elem *elem); }; +/** + * struct nft_set_desc - description of set elements + * + * @klen: key length + * @dlen: data length + * @size: number of set elements + */ +struct nft_set_desc { + unsigned int klen; + unsigned int dlen; + unsigned int size; +}; + +/** + * enum nft_set_class - performance class + * + * @NFT_LOOKUP_O_1: constant, O(1) + * @NFT_LOOKUP_O_LOG_N: logarithmic, O(log N) + * @NFT_LOOKUP_O_N: linear, O(N) + */ +enum nft_set_class { + NFT_SET_CLASS_O_1, + NFT_SET_CLASS_O_LOG_N, + NFT_SET_CLASS_O_N, +}; + +/** + * struct nft_set_estimate - estimation of memory and performance + * characteristics + * + * @size: required memory + * @class: lookup performance class + */ +struct nft_set_estimate { + unsigned int size; + enum nft_set_class class; +}; + /** * struct nft_set_ops - nf_tables set operations * @@ -174,7 +212,11 @@ struct nft_set_ops { struct nft_set_iter *iter); unsigned int (*privsize)(const struct nlattr * const nla[]); + bool (*estimate)(const struct nft_set_desc *desc, + u32 features, + struct nft_set_estimate *est); int (*init)(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const nla[]); void (*destroy)(const struct nft_set *set); @@ -194,6 +236,8 @@ void nft_unregister_set(struct nft_set_ops *ops); * @name: name of the set * @ktype: key type (numeric type defined by userspace, not used in the kernel) * @dtype: data type (verdict or numeric type defined by userspace) + * @size: maximum set size + * @nelems: number of elements * @ops: set ops * @flags: set flags * @klen: key length @@ -206,6 +250,8 @@ struct nft_set { char name[IFNAMSIZ]; u32 ktype; u32 dtype; + u32 size; + u32 nelems; /* runtime data below here */ const struct nft_set_ops *ops ____cacheline_aligned; u16 flags; diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index c88ccbfda5f1..160159274cab 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -211,6 +211,29 @@ enum nft_set_flags { NFT_SET_MAP = 0x8, }; +/** + * enum nft_set_policies - set selection policy + * + * @NFT_SET_POL_PERFORMANCE: prefer high performance over low memory use + * @NFT_SET_POL_MEMORY: prefer low memory use over high performance + */ +enum nft_set_policies { + NFT_SET_POL_PERFORMANCE, + NFT_SET_POL_MEMORY, +}; + +/** + * enum nft_set_desc_attributes - set element description + * + * @NFTA_SET_DESC_SIZE: number of elements in set (NLA_U32) + */ +enum nft_set_desc_attributes { + NFTA_SET_DESC_UNSPEC, + NFTA_SET_DESC_SIZE, + __NFTA_SET_DESC_MAX +}; +#define NFTA_SET_DESC_MAX (__NFTA_SET_DESC_MAX - 1) + /** * enum nft_set_attributes - nf_tables set netlink attributes * @@ -221,6 +244,8 @@ enum nft_set_flags { * @NFTA_SET_KEY_LEN: key data length (NLA_U32) * @NFTA_SET_DATA_TYPE: mapping data type (NLA_U32) * @NFTA_SET_DATA_LEN: mapping data length (NLA_U32) + * @NFTA_SET_POLICY: selection policy (NLA_U32) + * @NFTA_SET_DESC: set description (NLA_NESTED) */ enum nft_set_attributes { NFTA_SET_UNSPEC, @@ -231,6 +256,8 @@ enum nft_set_attributes { NFTA_SET_KEY_LEN, NFTA_SET_DATA_TYPE, NFTA_SET_DATA_LEN, + NFTA_SET_POLICY, + NFTA_SET_DESC, __NFTA_SET_MAX }; #define NFTA_SET_MAX (__NFTA_SET_MAX - 1) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 33045a562297..bd3381e16084 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1912,9 +1912,18 @@ void nft_unregister_set(struct nft_set_ops *ops) } EXPORT_SYMBOL_GPL(nft_unregister_set); -static const struct nft_set_ops *nft_select_set_ops(const struct nlattr * const nla[]) +/* + * Select a set implementation based on the data characteristics and the + * given policy. The total memory use might not be known if no size is + * given, in that case the amount of memory per element is used. + */ +static const struct nft_set_ops * +nft_select_set_ops(const struct nlattr * const nla[], + const struct nft_set_desc *desc, + enum nft_set_policies policy) { - const struct nft_set_ops *ops; + const struct nft_set_ops *ops, *bops; + struct nft_set_estimate est, best; u32 features; #ifdef CONFIG_MODULES @@ -1932,15 +1941,45 @@ static const struct nft_set_ops *nft_select_set_ops(const struct nlattr * const features &= NFT_SET_INTERVAL | NFT_SET_MAP; } - // FIXME: implement selection properly + bops = NULL; + best.size = ~0; + best.class = ~0; + list_for_each_entry(ops, &nf_tables_set_ops, list) { if ((ops->features & features) != features) continue; + if (!ops->estimate(desc, features, &est)) + continue; + + switch (policy) { + case NFT_SET_POL_PERFORMANCE: + if (est.class < best.class) + break; + if (est.class == best.class && est.size < best.size) + break; + continue; + case NFT_SET_POL_MEMORY: + if (est.size < best.size) + break; + if (est.size == best.size && est.class < best.class) + break; + continue; + default: + break; + } + if (!try_module_get(ops->owner)) continue; - return ops; + if (bops != NULL) + module_put(bops->owner); + + bops = ops; + best = est; } + if (bops != NULL) + return bops; + return ERR_PTR(-EOPNOTSUPP); } @@ -1952,6 +1991,12 @@ static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = { [NFTA_SET_KEY_LEN] = { .type = NLA_U32 }, [NFTA_SET_DATA_TYPE] = { .type = NLA_U32 }, [NFTA_SET_DATA_LEN] = { .type = NLA_U32 }, + [NFTA_SET_POLICY] = { .type = NLA_U32 }, + [NFTA_SET_DESC] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = { + [NFTA_SET_DESC_SIZE] = { .type = NLA_U32 }, }; static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, @@ -2043,6 +2088,7 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, { struct nfgenmsg *nfmsg; struct nlmsghdr *nlh; + struct nlattr *desc; u32 portid = NETLINK_CB(ctx->skb).portid; u32 seq = ctx->nlh->nlmsg_seq; @@ -2076,6 +2122,14 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, goto nla_put_failure; } + desc = nla_nest_start(skb, NFTA_SET_DESC); + if (desc == NULL) + goto nla_put_failure; + if (set->size && + nla_put_be32(skb, NFTA_SET_DESC_SIZE, htonl(set->size))) + goto nla_put_failure; + nla_nest_end(skb, desc); + return nlmsg_end(skb, nlh); nla_put_failure: @@ -2304,6 +2358,23 @@ err: return err; } +static int nf_tables_set_desc_parse(const struct nft_ctx *ctx, + struct nft_set_desc *desc, + const struct nlattr *nla) +{ + struct nlattr *da[NFTA_SET_DESC_MAX + 1]; + int err; + + err = nla_parse_nested(da, NFTA_SET_DESC_MAX, nla, nft_set_desc_policy); + if (err < 0) + return err; + + if (da[NFTA_SET_DESC_SIZE] != NULL) + desc->size = ntohl(nla_get_be32(da[NFTA_SET_DESC_SIZE])); + + return 0; +} + static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2318,7 +2389,8 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, char name[IFNAMSIZ]; unsigned int size; bool create; - u32 ktype, klen, dlen, dtype, flags; + u32 ktype, dtype, flags, policy; + struct nft_set_desc desc; int err; if (nla[NFTA_SET_TABLE] == NULL || @@ -2326,6 +2398,8 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, nla[NFTA_SET_KEY_LEN] == NULL) return -EINVAL; + memset(&desc, 0, sizeof(desc)); + ktype = NFT_DATA_VALUE; if (nla[NFTA_SET_KEY_TYPE] != NULL) { ktype = ntohl(nla_get_be32(nla[NFTA_SET_KEY_TYPE])); @@ -2333,8 +2407,8 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, return -EINVAL; } - klen = ntohl(nla_get_be32(nla[NFTA_SET_KEY_LEN])); - if (klen == 0 || klen > FIELD_SIZEOF(struct nft_data, data)) + desc.klen = ntohl(nla_get_be32(nla[NFTA_SET_KEY_LEN])); + if (desc.klen == 0 || desc.klen > FIELD_SIZEOF(struct nft_data, data)) return -EINVAL; flags = 0; @@ -2346,7 +2420,6 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, } dtype = 0; - dlen = 0; if (nla[NFTA_SET_DATA_TYPE] != NULL) { if (!(flags & NFT_SET_MAP)) return -EINVAL; @@ -2359,15 +2432,25 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (dtype != NFT_DATA_VERDICT) { if (nla[NFTA_SET_DATA_LEN] == NULL) return -EINVAL; - dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); - if (dlen == 0 || - dlen > FIELD_SIZEOF(struct nft_data, data)) + desc.dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); + if (desc.dlen == 0 || + desc.dlen > FIELD_SIZEOF(struct nft_data, data)) return -EINVAL; } else - dlen = sizeof(struct nft_data); + desc.dlen = sizeof(struct nft_data); } else if (flags & NFT_SET_MAP) return -EINVAL; + policy = NFT_SET_POL_PERFORMANCE; + if (nla[NFTA_SET_POLICY] != NULL) + policy = ntohl(nla_get_be32(nla[NFTA_SET_POLICY])); + + if (nla[NFTA_SET_DESC] != NULL) { + err = nf_tables_set_desc_parse(&ctx, &desc, nla[NFTA_SET_DESC]); + if (err < 0) + return err; + } + create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, create); @@ -2398,7 +2481,7 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (!(nlh->nlmsg_flags & NLM_F_CREATE)) return -ENOENT; - ops = nft_select_set_ops(nla); + ops = nft_select_set_ops(nla, &desc, policy); if (IS_ERR(ops)) return PTR_ERR(ops); @@ -2419,12 +2502,13 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, INIT_LIST_HEAD(&set->bindings); set->ops = ops; set->ktype = ktype; - set->klen = klen; + set->klen = desc.klen; set->dtype = dtype; - set->dlen = dlen; + set->dlen = desc.dlen; set->flags = flags; + set->size = desc.size; - err = ops->init(set, nla); + err = ops->init(set, &desc, nla); if (err < 0) goto err2; @@ -2733,6 +2817,9 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, enum nft_registers dreg; int err; + if (set->size && set->nelems == set->size) + return -ENFILE; + err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr, nft_set_elem_policy); if (err < 0) @@ -2798,6 +2885,7 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, err = set->ops->insert(set, &elem); if (err < 0) goto err3; + set->nelems++; return 0; @@ -2867,6 +2955,7 @@ static int nft_del_setelem(const struct nft_ctx *ctx, struct nft_set *set, goto err2; set->ops->remove(set, &elem); + set->nelems--; nft_data_uninit(&elem.key, NFT_DATA_VALUE); if (set->flags & NFT_SET_MAP) diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c index 3b1ad876d6b0..01884b315c4a 100644 --- a/net/netfilter/nft_hash.c +++ b/net/netfilter/nft_hash.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -19,7 +20,7 @@ #include #include -#define NFT_HASH_MIN_SIZE 4 +#define NFT_HASH_MIN_SIZE 4UL struct nft_hash { struct nft_hash_table __rcu *tbl; @@ -82,6 +83,11 @@ static void nft_hash_tbl_free(const struct nft_hash_table *tbl) kfree(tbl); } +static unsigned int nft_hash_tbl_size(unsigned int nelem) +{ + return max(roundup_pow_of_two(nelem * 4 / 3), NFT_HASH_MIN_SIZE); +} + static struct nft_hash_table *nft_hash_tbl_alloc(unsigned int nbuckets) { struct nft_hash_table *tbl; @@ -335,17 +341,23 @@ static unsigned int nft_hash_privsize(const struct nlattr * const nla[]) } static int nft_hash_init(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const tb[]) { struct nft_hash *priv = nft_set_priv(set); struct nft_hash_table *tbl; + unsigned int size; if (unlikely(!nft_hash_rnd_initted)) { get_random_bytes(&nft_hash_rnd, 4); nft_hash_rnd_initted = true; } - tbl = nft_hash_tbl_alloc(NFT_HASH_MIN_SIZE); + size = NFT_HASH_MIN_SIZE; + if (desc->size) + size = nft_hash_tbl_size(desc->size); + + tbl = nft_hash_tbl_alloc(size); if (tbl == NULL) return -ENOMEM; RCU_INIT_POINTER(priv->tbl, tbl); @@ -369,8 +381,37 @@ static void nft_hash_destroy(const struct nft_set *set) kfree(tbl); } +static bool nft_hash_estimate(const struct nft_set_desc *desc, u32 features, + struct nft_set_estimate *est) +{ + unsigned int esize; + + esize = sizeof(struct nft_hash_elem); + if (features & NFT_SET_MAP) + esize += FIELD_SIZEOF(struct nft_hash_elem, data[0]); + + if (desc->size) { + est->size = sizeof(struct nft_hash) + + nft_hash_tbl_size(desc->size) * + sizeof(struct nft_hash_elem *) + + desc->size * esize; + } else { + /* Resizing happens when the load drops below 30% or goes + * above 75%. The average of 52.5% load (approximated by 50%) + * is used for the size estimation of the hash buckets, + * meaning we calculate two buckets per element. + */ + est->size = esize + 2 * sizeof(struct nft_hash_elem *); + } + + est->class = NFT_SET_CLASS_O_1; + + return true; +} + static struct nft_set_ops nft_hash_ops __read_mostly = { .privsize = nft_hash_privsize, + .estimate = nft_hash_estimate, .init = nft_hash_init, .destroy = nft_hash_destroy, .get = nft_hash_get, diff --git a/net/netfilter/nft_rbtree.c b/net/netfilter/nft_rbtree.c index e21d69d13506..072e611e9f71 100644 --- a/net/netfilter/nft_rbtree.c +++ b/net/netfilter/nft_rbtree.c @@ -201,6 +201,7 @@ static unsigned int nft_rbtree_privsize(const struct nlattr * const nla[]) } static int nft_rbtree_init(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const nla[]) { struct nft_rbtree *priv = nft_set_priv(set); @@ -222,8 +223,28 @@ static void nft_rbtree_destroy(const struct nft_set *set) } } +static bool nft_rbtree_estimate(const struct nft_set_desc *desc, u32 features, + struct nft_set_estimate *est) +{ + unsigned int nsize; + + nsize = sizeof(struct nft_rbtree_elem); + if (features & NFT_SET_MAP) + nsize += FIELD_SIZEOF(struct nft_rbtree_elem, data[0]); + + if (desc->size) + est->size = sizeof(struct nft_rbtree) + desc->size * nsize; + else + est->size = nsize; + + est->class = NFT_SET_CLASS_O_LOG_N; + + return true; +} + static struct nft_set_ops nft_rbtree_ops __read_mostly = { .privsize = nft_rbtree_privsize, + .estimate = nft_rbtree_estimate, .init = nft_rbtree_init, .destroy = nft_rbtree_destroy, .insert = nft_rbtree_insert, -- cgit v1.2.3 From 78f22b6a3a9254460d23060530b48ae02a9394e3 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 24 Mar 2014 17:57:27 +0100 Subject: cfg80211: allow userspace to take ownership of interfaces When dynamically creating interfaces from userspace, e.g. for P2P usage, such interfaces are usually owned by the process that created them, i.e. wpa_supplicant. Should wpa_supplicant crash, such interfaces will often cease operating properly and cause problems on restarting the process. To avoid this problem, introduce an ownership concept for interfaces. If an interface is owned by a netlink socket, then it will be destroyed if the netlink socket is closed for any reason, including if the process it belongs to crashed. This gives us a race-free way to get rid of any such interfaces. Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 3 +++ include/uapi/linux/nl80211.h | 6 ++++++ net/wireless/core.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ net/wireless/core.h | 11 +++++++++++ net/wireless/nl80211.c | 28 +++++++++++++++++++++++++++- 5 files changed, 91 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index f3539a15c411..6510ccf53a54 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -3194,6 +3194,7 @@ struct cfg80211_cached_keys; * @ibss_dfs_possible: (private) IBSS may change to a DFS channel * @event_list: (private) list for internal event processing * @event_lock: (private) lock for event list + * @owner_nlportid: (private) owner socket port ID */ struct wireless_dev { struct wiphy *wiphy; @@ -3241,6 +3242,8 @@ struct wireless_dev { unsigned long cac_start_time; unsigned int cac_time_ms; + u32 owner_nlportid; + #ifdef CONFIG_CFG80211_WEXT /* wext data */ struct { diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 1ba9d626aa83..5e405fd55a71 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1579,6 +1579,10 @@ enum nl80211_commands { * @NL80211_ATTR_TDLS_PEER_CAPABILITY: flags for TDLS peer capabilities, u32. * As specified in the &enum nl80211_tdls_peer_capability. * + * @NL80211_ATTR_IFACE_SOCKET_OWNER: flag attribute, if set during interface + * creation then the new interface will be owned by the netlink socket + * that created it and will be destroyed when the socket is closed + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -1914,6 +1918,8 @@ enum nl80211_attrs { NL80211_ATTR_TDLS_PEER_CAPABILITY, + NL80211_ATTR_IFACE_SOCKET_OWNER, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/wireless/core.c b/net/wireless/core.c index 086cddd03ba6..5a63c3cbda2e 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -260,6 +260,45 @@ static void cfg80211_event_work(struct work_struct *work) rtnl_unlock(); } +void cfg80211_destroy_ifaces(struct cfg80211_registered_device *rdev) +{ + struct cfg80211_iface_destroy *item; + + ASSERT_RTNL(); + + spin_lock_irq(&rdev->destroy_list_lock); + while ((item = list_first_entry_or_null(&rdev->destroy_list, + struct cfg80211_iface_destroy, + list))) { + struct wireless_dev *wdev, *tmp; + u32 nlportid = item->nlportid; + + list_del(&item->list); + kfree(item); + spin_unlock_irq(&rdev->destroy_list_lock); + + list_for_each_entry_safe(wdev, tmp, &rdev->wdev_list, list) { + if (nlportid == wdev->owner_nlportid) + rdev_del_virtual_intf(rdev, wdev); + } + + spin_lock_irq(&rdev->destroy_list_lock); + } + spin_unlock_irq(&rdev->destroy_list_lock); +} + +static void cfg80211_destroy_iface_wk(struct work_struct *work) +{ + struct cfg80211_registered_device *rdev; + + rdev = container_of(work, struct cfg80211_registered_device, + destroy_work); + + rtnl_lock(); + cfg80211_destroy_ifaces(rdev); + rtnl_unlock(); +} + /* exported functions */ struct wiphy *wiphy_new(const struct cfg80211_ops *ops, int sizeof_priv) @@ -318,6 +357,10 @@ struct wiphy *wiphy_new(const struct cfg80211_ops *ops, int sizeof_priv) rdev->wiphy.dev.class = &ieee80211_class; rdev->wiphy.dev.platform_data = rdev; + INIT_LIST_HEAD(&rdev->destroy_list); + spin_lock_init(&rdev->destroy_list_lock); + INIT_WORK(&rdev->destroy_work, cfg80211_destroy_iface_wk); + #ifdef CONFIG_CFG80211_DEFAULT_PS rdev->wiphy.flags |= WIPHY_FLAG_PS_ON_BY_DEFAULT; #endif @@ -675,6 +718,7 @@ void wiphy_unregister(struct wiphy *wiphy) cancel_work_sync(&rdev->conn_work); flush_work(&rdev->event_work); cancel_delayed_work_sync(&rdev->dfs_update_channels_wk); + flush_work(&rdev->destroy_work); #ifdef CONFIG_PM if (rdev->wiphy.wowlan_config && rdev->ops->set_wakeup) diff --git a/net/wireless/core.h b/net/wireless/core.h index 5b1fdcadd469..6f6f75609852 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -80,6 +80,10 @@ struct cfg80211_registered_device { struct cfg80211_coalesce *coalesce; + spinlock_t destroy_list_lock; + struct list_head destroy_list; + struct work_struct destroy_work; + /* must be last because of the way we do wiphy_priv(), * and it should at least be aligned to NETDEV_ALIGN */ struct wiphy wiphy __aligned(NETDEV_ALIGN); @@ -232,6 +236,13 @@ struct cfg80211_beacon_registration { u32 nlportid; }; +struct cfg80211_iface_destroy { + struct list_head list; + u32 nlportid; +}; + +void cfg80211_destroy_ifaces(struct cfg80211_registered_device *rdev); + /* free object */ void cfg80211_dev_free(struct cfg80211_registered_device *rdev); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 052c1bf8ffac..b25b5ce4076d 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -385,6 +385,7 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_MAC_HINT] = { .len = ETH_ALEN }, [NL80211_ATTR_WIPHY_FREQ_HINT] = { .type = NLA_U32 }, [NL80211_ATTR_TDLS_PEER_CAPABILITY] = { .type = NLA_U32 }, + [NL80211_ATTR_IFACE_SOCKET_OWNER] = { .type = NLA_FLAG }, }; /* policy for the key attributes */ @@ -2514,6 +2515,9 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED; u32 flags; + /* to avoid failing a new interface creation due to pending removal */ + cfg80211_destroy_ifaces(rdev); + memset(¶ms, 0, sizeof(params)); if (!info->attrs[NL80211_ATTR_IFNAME]) @@ -2563,6 +2567,9 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) return PTR_ERR(wdev); } + if (info->attrs[NL80211_ATTR_IFACE_SOCKET_OWNER]) + wdev->owner_nlportid = info->snd_portid; + switch (type) { case NL80211_IFTYPE_MESH_POINT: if (!info->attrs[NL80211_ATTR_MESH_ID]) @@ -11649,9 +11656,15 @@ static int nl80211_netlink_notify(struct notifier_block * nb, rcu_read_lock(); list_for_each_entry_rcu(rdev, &cfg80211_rdev_list, list) { - list_for_each_entry_rcu(wdev, &rdev->wdev_list, list) + bool schedule_destroy_work = false; + + list_for_each_entry_rcu(wdev, &rdev->wdev_list, list) { cfg80211_mlme_unregister_socket(wdev, notify->portid); + if (wdev->owner_nlportid == notify->portid) + schedule_destroy_work = true; + } + spin_lock_bh(&rdev->beacon_registrations_lock); list_for_each_entry_safe(reg, tmp, &rdev->beacon_registrations, list) { @@ -11662,6 +11675,19 @@ static int nl80211_netlink_notify(struct notifier_block * nb, } } spin_unlock_bh(&rdev->beacon_registrations_lock); + + if (schedule_destroy_work) { + struct cfg80211_iface_destroy *destroy; + + destroy = kzalloc(sizeof(*destroy), GFP_ATOMIC); + if (destroy) { + destroy->nlportid = notify->portid; + spin_lock(&rdev->destroy_list_lock); + list_add(&destroy->list, &rdev->destroy_list); + spin_unlock(&rdev->destroy_list_lock); + schedule_work(&rdev->destroy_work); + } + } } rcu_read_unlock(); -- cgit v1.2.3 From 570dbde137d4604e4e682a5855b4425233344c19 Mon Sep 17 00:00:00 2001 From: David Spinadel Date: Sun, 23 Feb 2014 09:12:59 +0200 Subject: cfg80211: Add indoor only and GO concurrent channel attributes The FCC are clarifying some soft configuration requirements, which among other include the following: 1. Indoor operation, where a device can use channels requiring indoor operation, subject to that it can guarantee indoor operation, i.e., the device is connected to AC Power or the device is under the control of a local master that is acting as an AP and is connected to AC Power. 2. Concurrent GO operation, where devices may instantiate a P2P GO while they are under the guidance of an authorized master. For example, on a channel on which a BSS is connected to an authorized master, i.e., with DFS and radar detection capability in the UNII band. See https://apps.fcc.gov/eas/comments/GetPublishedDocument.html?id=327&tn=528122 Add support for advertising Indoor-only and GO-Concurrent channel properties. Signed-off-by: David Spinadel Signed-off-by: Ilan Peer Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 5 +++++ include/uapi/linux/nl80211.h | 23 +++++++++++++++++++++++ net/wireless/nl80211.c | 6 ++++++ net/wireless/reg.c | 2 ++ 4 files changed, 36 insertions(+) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 6510ccf53a54..14d8d3417735 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -109,6 +109,9 @@ enum ieee80211_band { * channel as the control or any of the secondary channels. * This may be due to the driver or due to regulatory bandwidth * restrictions. + * @IEEE80211_CHAN_INDOOR_ONLY: see %NL80211_FREQUENCY_ATTR_INDOOR_ONLY + * @IEEE80211_CHAN_GO_CONCURRENT: see %NL80211_FREQUENCY_ATTR_GO_CONCURRENT + * */ enum ieee80211_channel_flags { IEEE80211_CHAN_DISABLED = 1<<0, @@ -120,6 +123,8 @@ enum ieee80211_channel_flags { IEEE80211_CHAN_NO_OFDM = 1<<6, IEEE80211_CHAN_NO_80MHZ = 1<<7, IEEE80211_CHAN_NO_160MHZ = 1<<8, + IEEE80211_CHAN_INDOOR_ONLY = 1<<9, + IEEE80211_CHAN_GO_CONCURRENT = 1<<10, }; #define IEEE80211_CHAN_NO_HT40 \ diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 5e405fd55a71..ac5b2d25f0fc 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2342,9 +2342,30 @@ enum nl80211_band_attr { * using this channel as the primary or any of the secondary channels * isn't possible * @NL80211_FREQUENCY_ATTR_DFS_CAC_TIME: DFS CAC time in milliseconds. + * @NL80211_FREQUENCY_ATTR_INDOOR_ONLY: Only indoor use is permitted on this + * channel. A channel that has the INDOOR_ONLY attribute can only be + * used when there is a clear assessment that the device is operating in + * an indoor surroundings, i.e., it is connected to AC power (and not + * through portable DC inverters) or is under the control of a master + * that is acting as an AP and is connected to AC power. + * @NL80211_FREQUENCY_ATTR_GO_CONCURRENT: GO operation is allowed on this + * channel if it's connected concurrently to a BSS on the same channel on + * the 2 GHz band or to a channel in the same UNII band (on the 5 GHz + * band), and IEEE80211_CHAN_RADAR is not set. Instantiating a GO on a + * channel that has the GO_CONCURRENT attribute set can be done when there + * is a clear assessment that the device is operating under the guidance of + * an authorized master, i.e., setting up a GO while the device is also + * connected to an AP with DFS and radar detection on the UNII band (it is + * up to user-space, i.e., wpa_supplicant to perform the required + * verifications) * @NL80211_FREQUENCY_ATTR_MAX: highest frequency attribute number * currently defined * @__NL80211_FREQUENCY_ATTR_AFTER_LAST: internal use + * + * See https://apps.fcc.gov/eas/comments/GetPublishedDocument.html?id=327&tn=528122 + * for more information on the FCC description of the relaxations allowed + * by NL80211_FREQUENCY_ATTR_INDOOR_ONLY and + * NL80211_FREQUENCY_ATTR_GO_CONCURRENT. */ enum nl80211_frequency_attr { __NL80211_FREQUENCY_ATTR_INVALID, @@ -2361,6 +2382,8 @@ enum nl80211_frequency_attr { NL80211_FREQUENCY_ATTR_NO_80MHZ, NL80211_FREQUENCY_ATTR_NO_160MHZ, NL80211_FREQUENCY_ATTR_DFS_CAC_TIME, + NL80211_FREQUENCY_ATTR_INDOOR_ONLY, + NL80211_FREQUENCY_ATTR_GO_CONCURRENT, /* keep last */ __NL80211_FREQUENCY_ATTR_AFTER_LAST, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index b25b5ce4076d..c5ead18ad3ab 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -614,6 +614,12 @@ static int nl80211_msg_put_channel(struct sk_buff *msg, if ((chan->flags & IEEE80211_CHAN_NO_160MHZ) && nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_160MHZ)) goto nla_put_failure; + if ((chan->flags & IEEE80211_CHAN_INDOOR_ONLY) && + nla_put_flag(msg, NL80211_FREQUENCY_ATTR_INDOOR_ONLY)) + goto nla_put_failure; + if ((chan->flags & IEEE80211_CHAN_GO_CONCURRENT) && + nla_put_flag(msg, NL80211_FREQUENCY_ATTR_GO_CONCURRENT)) + goto nla_put_failure; } if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_MAX_TX_POWER, diff --git a/net/wireless/reg.c b/net/wireless/reg.c index e0a746d19061..1e9dc1cba335 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -873,6 +873,8 @@ static u32 map_regdom_flags(u32 rd_flags) channel_flags |= IEEE80211_CHAN_RADAR; if (rd_flags & NL80211_RRF_NO_OFDM) channel_flags |= IEEE80211_CHAN_NO_OFDM; + if (rd_flags & NL80211_RRF_NO_OUTDOOR) + channel_flags |= IEEE80211_CHAN_INDOOR_ONLY; return channel_flags; } -- cgit v1.2.3 From 52616f2b446eaad8eb2cd78bbd052f0066069757 Mon Sep 17 00:00:00 2001 From: Ilan Peer Date: Tue, 25 Feb 2014 16:26:00 +0200 Subject: cfg80211: Add an option to hint indoor operation Add the option to hint the wireless core that it is operating in an indoor environment. Signed-off-by: Ilan Peer Signed-off-by: Johannes Berg --- include/uapi/linux/nl80211.h | 3 +++ net/wireless/nl80211.c | 18 ++++++-------- net/wireless/reg.c | 58 +++++++++++++++++++++++++++++++++++++++++++- net/wireless/reg.h | 1 + 4 files changed, 68 insertions(+), 12 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index ac5b2d25f0fc..513bfd7b2e5f 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2602,10 +2602,13 @@ enum nl80211_dfs_regions { * present has been registered with the wireless core that * has listed NL80211_FEATURE_CELL_BASE_REG_HINTS as a * supported feature. + * @NL80211_USER_REG_HINT_INDOOR: a user sent an hint indicating that the + * platform is operating in an indoor environment. */ enum nl80211_user_reg_hint_type { NL80211_USER_REG_HINT_USER = 0, NL80211_USER_REG_HINT_CELL_BASE = 1, + NL80211_USER_REG_HINT_INDOOR = 2, }; /** diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index b8d81e41b0f7..85bc830fd7e3 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -4677,7 +4677,6 @@ static int parse_reg_rule(struct nlattr *tb[], static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info) { - int r; char *data = NULL; enum nl80211_user_reg_hint_type user_reg_hint_type; @@ -4690,11 +4689,6 @@ static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info) if (unlikely(!rcu_access_pointer(cfg80211_regdomain))) return -EINPROGRESS; - if (!info->attrs[NL80211_ATTR_REG_ALPHA2]) - return -EINVAL; - - data = nla_data(info->attrs[NL80211_ATTR_REG_ALPHA2]); - if (info->attrs[NL80211_ATTR_USER_REG_HINT_TYPE]) user_reg_hint_type = nla_get_u32(info->attrs[NL80211_ATTR_USER_REG_HINT_TYPE]); @@ -4704,14 +4698,16 @@ static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info) switch (user_reg_hint_type) { case NL80211_USER_REG_HINT_USER: case NL80211_USER_REG_HINT_CELL_BASE: - break; + if (!info->attrs[NL80211_ATTR_REG_ALPHA2]) + return -EINVAL; + + data = nla_data(info->attrs[NL80211_ATTR_REG_ALPHA2]); + return regulatory_hint_user(data, user_reg_hint_type); + case NL80211_USER_REG_HINT_INDOOR: + return regulatory_hint_indoor_user(); default: return -EINVAL; } - - r = regulatory_hint_user(data, user_reg_hint_type); - - return r; } static int nl80211_get_mesh_config(struct sk_buff *skb, diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 58f48b8f42ae..55d68c31ad72 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -65,11 +65,26 @@ #define REG_DBG_PRINT(args...) #endif +/** + * enum reg_request_treatment - regulatory request treatment + * + * @REG_REQ_OK: continue processing the regulatory request + * @REG_REQ_IGNORE: ignore the regulatory request + * @REG_REQ_INTERSECT: the regulatory domain resulting from this request should + * be intersected with the current one. + * @REG_REQ_ALREADY_SET: the regulatory request will not change the current + * regulatory settings, and no further processing is required. + * @REG_REQ_USER_HINT_HANDLED: a non alpha2 user hint was handled and no + * further processing is required, i.e., not need to update last_request + * etc. This should be used for user hints that do not provide an alpha2 + * but some other type of regulatory hint, i.e., indoor operation. + */ enum reg_request_treatment { REG_REQ_OK, REG_REQ_IGNORE, REG_REQ_INTERSECT, REG_REQ_ALREADY_SET, + REG_REQ_USER_HINT_HANDLED, }; static struct regulatory_request core_request_world = { @@ -106,6 +121,14 @@ const struct ieee80211_regdomain __rcu *cfg80211_regdomain; */ static int reg_num_devs_support_basehint; +/* + * State variable indicating if the platform on which the devices + * are attached is operating in an indoor environment. The state variable + * is relevant for all registered devices. + * (protected by RTNL) + */ +static bool reg_is_indoor; + static const struct ieee80211_regdomain *get_cfg80211_regdom(void) { return rtnl_dereference(cfg80211_regdomain); @@ -1128,6 +1151,13 @@ static bool reg_request_cell_base(struct regulatory_request *request) return request->user_reg_hint_type == NL80211_USER_REG_HINT_CELL_BASE; } +static bool reg_request_indoor(struct regulatory_request *request) +{ + if (request->initiator != NL80211_REGDOM_SET_BY_USER) + return false; + return request->user_reg_hint_type == NL80211_USER_REG_HINT_INDOOR; +} + bool reg_last_request_cell_base(void) { return reg_request_cell_base(get_last_request()); @@ -1570,6 +1600,11 @@ __reg_process_hint_user(struct regulatory_request *user_request) { struct regulatory_request *lr = get_last_request(); + if (reg_request_indoor(user_request)) { + reg_is_indoor = true; + return REG_REQ_USER_HINT_HANDLED; + } + if (reg_request_cell_base(user_request)) return reg_ignore_cell_hint(user_request); @@ -1617,7 +1652,8 @@ reg_process_hint_user(struct regulatory_request *user_request) treatment = __reg_process_hint_user(user_request); if (treatment == REG_REQ_IGNORE || - treatment == REG_REQ_ALREADY_SET) { + treatment == REG_REQ_ALREADY_SET || + treatment == REG_REQ_USER_HINT_HANDLED) { kfree(user_request); return treatment; } @@ -1678,6 +1714,7 @@ reg_process_hint_driver(struct wiphy *wiphy, case REG_REQ_OK: break; case REG_REQ_IGNORE: + case REG_REQ_USER_HINT_HANDLED: kfree(driver_request); return treatment; case REG_REQ_INTERSECT: @@ -1777,6 +1814,7 @@ reg_process_hint_country_ie(struct wiphy *wiphy, case REG_REQ_OK: break; case REG_REQ_IGNORE: + case REG_REQ_USER_HINT_HANDLED: /* fall through */ case REG_REQ_ALREADY_SET: kfree(country_ie_request); @@ -1969,6 +2007,22 @@ int regulatory_hint_user(const char *alpha2, return 0; } +int regulatory_hint_indoor_user(void) +{ + struct regulatory_request *request; + + request = kzalloc(sizeof(struct regulatory_request), GFP_KERNEL); + if (!request) + return -ENOMEM; + + request->wiphy_idx = WIPHY_IDX_INVALID; + request->initiator = NL80211_REGDOM_SET_BY_USER; + request->user_reg_hint_type = NL80211_USER_REG_HINT_INDOOR; + queue_regulatory_request(request); + + return 0; +} + /* Driver hints */ int regulatory_hint(struct wiphy *wiphy, const char *alpha2) { @@ -2136,6 +2190,8 @@ static void restore_regulatory_settings(bool reset_user) ASSERT_RTNL(); + reg_is_indoor = false; + reset_regdomains(true, &world_regdom); restore_alpha2(alpha2, reset_user); diff --git a/net/wireless/reg.h b/net/wireless/reg.h index 334a53af0fc8..2a3842828f6d 100644 --- a/net/wireless/reg.h +++ b/net/wireless/reg.h @@ -25,6 +25,7 @@ enum nl80211_dfs_regions reg_get_dfs_region(struct wiphy *wiphy); int regulatory_hint_user(const char *alpha2, enum nl80211_user_reg_hint_type user_reg_hint_type); +int regulatory_hint_indoor_user(void); void wiphy_regulatory_register(struct wiphy *wiphy); void wiphy_regulatory_deregister(struct wiphy *wiphy); -- cgit v1.2.3 From 3de0b592394d17b2c41a261a6a493a521213f299 Mon Sep 17 00:00:00 2001 From: Venkata Duvvuru Date: Mon, 21 Apr 2014 15:37:59 +0530 Subject: ethtool: Support for configurable RSS hash key This ethtool patch primarily copies the ioctl command data structures from/to the User space and invokes the driver hook. Signed-off-by: Venkat Duvvuru Signed-off-by: David S. Miller --- include/linux/ethtool.h | 13 +++ include/uapi/linux/ethtool.h | 32 +++++++ net/core/ethtool.c | 221 ++++++++++++++++++++++++++++++++++++++++--- 3 files changed, 252 insertions(+), 14 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 0a114d05f68d..212f537fc686 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -154,13 +154,23 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings) * @reset: Reset (part of) the device, as specified by a bitmask of * flags from &enum ethtool_reset_flags. Returns a negative * error code or zero. + * @get_rxfh_key_size: Get the size of the RX flow hash key. + * Returns zero if not supported for this specific device. * @get_rxfh_indir_size: Get the size of the RX flow hash indirection table. * Returns zero if not supported for this specific device. * @get_rxfh_indir: Get the contents of the RX flow hash indirection table. * Will not be called if @get_rxfh_indir_size returns zero. + * @get_rxfh: Get the contents of the RX flow hash indirection table and hash + * key. + * Will not be called if @get_rxfh_indir_size and @get_rxfh_key_size + * returns zero. * Returns a negative error code or zero. * @set_rxfh_indir: Set the contents of the RX flow hash indirection table. * Will not be called if @get_rxfh_indir_size returns zero. + * @set_rxfh: Set the contents of the RX flow hash indirection table and + * hash key. + * Will not be called if @get_rxfh_indir_size and @get_rxfh_key_size + * returns zero. * Returns a negative error code or zero. * @get_channels: Get number of channels. * @set_channels: Set number of channels. Returns a negative error code or @@ -232,7 +242,10 @@ struct ethtool_ops { int (*set_rxnfc)(struct net_device *, struct ethtool_rxnfc *); int (*flash_device)(struct net_device *, struct ethtool_flash *); int (*reset)(struct net_device *, u32 *); + u32 (*get_rxfh_key_size)(struct net_device *); u32 (*get_rxfh_indir_size)(struct net_device *); + int (*get_rxfh)(struct net_device *, u32 *, u8 *); + int (*set_rxfh)(struct net_device *, u32 *, u8 *); int (*get_rxfh_indir)(struct net_device *, u32 *); int (*set_rxfh_indir)(struct net_device *, const u32 *); void (*get_channels)(struct net_device *, struct ethtool_channels *); diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index fd161e91b6d7..d47d31d6fa0e 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -846,6 +846,35 @@ struct ethtool_rxfh_indir { __u32 ring_index[0]; }; +/** + * struct ethtool_rxfh - command to get/set RX flow hash indir or/and hash key. + * @cmd: Specific command number - %ETHTOOL_GRSSH or %ETHTOOL_SRSSH + * @rss_context: RSS context identifier. + * @indir_size: On entry, the array size of the user buffer, which may be zero. + * On return from %ETHTOOL_GRSSH, the array size of the hardware + * indirection table. + * @key_size: On entry, the array size of the user buffer in bytes, + * which may be zero. + * On return from %ETHTOOL_GRSSH, the size of the RSS hash key. + * @rsvd: Reserved for future extensions. + * @rss_config: RX ring/queue index for each hash value i.e., indirection table + * of size @indir_size followed by hash key of size @key_size. + * + * For %ETHTOOL_GRSSH, a @indir_size and key_size of zero means that only the + * size should be returned. For %ETHTOOL_SRSSH, a @indir_size of 0xDEADBEEF + * means that indir table setting is not requested and a @indir_size of zero + * means the indir table should be reset to default values. This last feature + * is not supported by the original implementations. + */ +struct ethtool_rxfh { + __u32 cmd; + __u32 rss_context; + __u32 indir_size; + __u32 key_size; + __u32 rsvd[2]; + __u32 rss_config[0]; +}; + /** * struct ethtool_rx_ntuple_flow_spec - specification for RX flow filter * @flow_type: Type of match to perform, e.g. %TCP_V4_FLOW @@ -1118,6 +1147,9 @@ enum ethtool_sfeatures_retval_bits { #define ETHTOOL_GEEE 0x00000044 /* Get EEE settings */ #define ETHTOOL_SEEE 0x00000045 /* Set EEE settings */ +#define ETHTOOL_GRSSH 0x00000046 /* Get RX flow hash configuration */ +#define ETHTOOL_SRSSH 0x00000047 /* Set RX flow hash configuration */ + /* compatibility with older code */ #define SPARC_ETH_GSET ETHTOOL_GSET #define SPARC_ETH_SSET ETHTOOL_SSET diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 640ba0e5831c..1d72786ef866 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -557,6 +557,23 @@ err_out: return ret; } +static int ethtool_copy_validate_indir(u32 *indir, void __user *useraddr, + struct ethtool_rxnfc *rx_rings, + u32 size) +{ + int ret = 0, i; + + if (copy_from_user(indir, useraddr, size * sizeof(indir[0]))) + ret = -EFAULT; + + /* Validate ring indices */ + for (i = 0; i < size; i++) { + if (indir[i] >= rx_rings->data) + ret = -EINVAL; + } + return ret; +} + static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev, void __user *useraddr) { @@ -613,6 +630,7 @@ static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev, u32 *indir; const struct ethtool_ops *ops = dev->ethtool_ops; int ret; + u32 ringidx_offset = offsetof(struct ethtool_rxfh_indir, ring_index[0]); if (!ops->get_rxfh_indir_size || !ops->set_rxfh_indir || !ops->get_rxnfc) @@ -643,28 +661,196 @@ static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev, for (i = 0; i < dev_size; i++) indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data); } else { - if (copy_from_user(indir, - useraddr + - offsetof(struct ethtool_rxfh_indir, - ring_index[0]), - dev_size * sizeof(indir[0]))) { + ret = ethtool_copy_validate_indir(indir, + useraddr + ringidx_offset, + &rx_rings, + dev_size); + if (ret) + goto out; + } + + ret = ops->set_rxfh_indir(dev, indir); + +out: + kfree(indir); + return ret; +} + +static noinline_for_stack int ethtool_get_rxfh(struct net_device *dev, + void __user *useraddr) +{ + int ret; + const struct ethtool_ops *ops = dev->ethtool_ops; + u32 user_indir_size = 0, user_key_size = 0; + u32 dev_indir_size = 0, dev_key_size = 0; + u32 total_size; + u32 indir_offset, indir_bytes; + u32 key_offset; + u32 *indir = NULL; + u8 *hkey = NULL; + u8 *rss_config; + + if (!(dev->ethtool_ops->get_rxfh_indir_size || + dev->ethtool_ops->get_rxfh_key_size) || + !dev->ethtool_ops->get_rxfh) + return -EOPNOTSUPP; + + if (ops->get_rxfh_indir_size) + dev_indir_size = ops->get_rxfh_indir_size(dev); + + indir_offset = offsetof(struct ethtool_rxfh, indir_size); + + if (copy_from_user(&user_indir_size, + useraddr + indir_offset, + sizeof(user_indir_size))) + return -EFAULT; + + if (copy_to_user(useraddr + indir_offset, + &dev_indir_size, sizeof(dev_indir_size))) + return -EFAULT; + + if (ops->get_rxfh_key_size) + dev_key_size = ops->get_rxfh_key_size(dev); + + if ((dev_key_size + dev_indir_size) == 0) + return -EOPNOTSUPP; + + key_offset = offsetof(struct ethtool_rxfh, key_size); + + if (copy_from_user(&user_key_size, + useraddr + key_offset, + sizeof(user_key_size))) + return -EFAULT; + + if (copy_to_user(useraddr + key_offset, + &dev_key_size, sizeof(dev_key_size))) + return -EFAULT; + + /* If the user buffer size is 0, this is just a query for the + * device table size and key size. Otherwise, if the User size is + * not equal to device table size or key size it's an error. + */ + if (!user_indir_size && !user_key_size) + return 0; + + if ((user_indir_size && (user_indir_size != dev_indir_size)) || + (user_key_size && (user_key_size != dev_key_size))) + return -EINVAL; + + indir_bytes = user_indir_size * sizeof(indir[0]); + total_size = indir_bytes + user_key_size; + rss_config = kzalloc(total_size, GFP_USER); + if (!rss_config) + return -ENOMEM; + + if (user_indir_size) + indir = (u32 *)rss_config; + + if (user_key_size) + hkey = rss_config + indir_bytes; + + ret = dev->ethtool_ops->get_rxfh(dev, indir, hkey); + if (!ret) { + if (copy_to_user(useraddr + + offsetof(struct ethtool_rxfh, rss_config[0]), + rss_config, total_size)) ret = -EFAULT; + } + + kfree(rss_config); + + return ret; +} + +static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, + void __user *useraddr) +{ + int ret; + const struct ethtool_ops *ops = dev->ethtool_ops; + struct ethtool_rxnfc rx_rings; + u32 user_indir_size = 0, dev_indir_size = 0, i; + u32 user_key_size = 0, dev_key_size = 0; + u32 *indir = NULL, indir_bytes = 0; + u8 *hkey = NULL; + u8 *rss_config; + u32 indir_offset, key_offset; + u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]); + + if (!(ops->get_rxfh_indir_size || ops->get_rxfh_key_size) || + !ops->get_rxnfc || !ops->set_rxfh) + return -EOPNOTSUPP; + + if (ops->get_rxfh_indir_size) + dev_indir_size = ops->get_rxfh_indir_size(dev); + + indir_offset = offsetof(struct ethtool_rxfh, indir_size); + if (copy_from_user(&user_indir_size, + useraddr + indir_offset, + sizeof(user_indir_size))) + return -EFAULT; + + if (ops->get_rxfh_key_size) + dev_key_size = dev->ethtool_ops->get_rxfh_key_size(dev); + + if ((dev_key_size + dev_indir_size) == 0) + return -EOPNOTSUPP; + + key_offset = offsetof(struct ethtool_rxfh, key_size); + if (copy_from_user(&user_key_size, + useraddr + key_offset, + sizeof(user_key_size))) + return -EFAULT; + + /* If either indir or hash key is valid, proceed further. + */ + if ((user_indir_size && ((user_indir_size != 0xDEADBEEF) && + user_indir_size != dev_indir_size)) || + (user_key_size && (user_key_size != dev_key_size))) + return -EINVAL; + + if (user_indir_size != 0xDEADBEEF) + indir_bytes = dev_indir_size * sizeof(indir[0]); + + rss_config = kzalloc(indir_bytes + user_key_size, GFP_USER); + if (!rss_config) + return -ENOMEM; + + rx_rings.cmd = ETHTOOL_GRXRINGS; + ret = ops->get_rxnfc(dev, &rx_rings, NULL); + if (ret) + goto out; + + /* user_indir_size == 0 means reset the indir table to default. + * user_indir_size == 0xDEADBEEF means indir setting is not requested. + */ + if (user_indir_size && user_indir_size != 0xDEADBEEF) { + indir = (u32 *)rss_config; + ret = ethtool_copy_validate_indir(indir, + useraddr + rss_cfg_offset, + &rx_rings, + user_indir_size); + if (ret) goto out; - } + } else if (user_indir_size == 0) { + indir = (u32 *)rss_config; + for (i = 0; i < dev_indir_size; i++) + indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data); + } - /* Validate ring indices */ - for (i = 0; i < dev_size; i++) { - if (indir[i] >= rx_rings.data) { - ret = -EINVAL; - goto out; - } + if (user_key_size) { + hkey = rss_config + indir_bytes; + if (copy_from_user(hkey, + useraddr + rss_cfg_offset + indir_bytes, + user_key_size)) { + ret = -EFAULT; + goto out; } } - ret = ops->set_rxfh_indir(dev, indir); + ret = ops->set_rxfh(dev, indir, hkey); out: - kfree(indir); + kfree(rss_config); return ret; } @@ -1491,6 +1677,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_GRXCLSRULE: case ETHTOOL_GRXCLSRLALL: case ETHTOOL_GRXFHINDIR: + case ETHTOOL_GRSSH: case ETHTOOL_GFEATURES: case ETHTOOL_GCHANNELS: case ETHTOOL_GET_TS_INFO: @@ -1628,6 +1815,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_SRXFHINDIR: rc = ethtool_set_rxfh_indir(dev, useraddr); break; + case ETHTOOL_GRSSH: + rc = ethtool_get_rxfh(dev, useraddr); + break; + case ETHTOOL_SRSSH: + rc = ethtool_set_rxfh(dev, useraddr); + break; case ETHTOOL_GFEATURES: rc = ethtool_get_features(dev, useraddr); break; -- cgit v1.2.3 From 4cd3675ebf74d7f559038ded6aa8088e4099a83d Mon Sep 17 00:00:00 2001 From: Chema Gonzalez Date: Mon, 21 Apr 2014 09:21:24 -0700 Subject: filter: added BPF random opcode Added a new ancillary load (bpf call in eBPF parlance) that produces a 32-bit random number. We are implementing it as an ancillary load (instead of an ISA opcode) because (a) it is simpler, (b) allows easy JITing, and (c) seems more in line with generic ISAs that do not have "get a random number" as a instruction, but as an OS call. The main use for this ancillary load is to perform random packet sampling. Signed-off-by: Chema Gonzalez Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- Documentation/networking/filter.txt | 13 +++++++++++++ include/linux/filter.h | 1 + include/uapi/linux/filter.h | 3 ++- net/core/filter.c | 12 ++++++++++++ tools/net/bpf_exp.l | 1 + tools/net/bpf_exp.y | 11 ++++++++++- 6 files changed, 39 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/Documentation/networking/filter.txt b/Documentation/networking/filter.txt index 81f940f4e884..82e1cb0b3da8 100644 --- a/Documentation/networking/filter.txt +++ b/Documentation/networking/filter.txt @@ -281,6 +281,7 @@ Possible BPF extensions are shown in the following table: cpu raw_smp_processor_id() vlan_tci vlan_tx_tag_get(skb) vlan_pr vlan_tx_tag_present(skb) + rand prandom_u32() These extensions can also be prefixed with '#'. Examples for low-level BPF: @@ -308,6 +309,18 @@ Examples for low-level BPF: ret #-1 drop: ret #0 +** icmp random packet sampling, 1 in 4 + ldh [12] + jne #0x800, drop + ldb [23] + jneq #1, drop + # get a random uint32 number + ld rand + mod #4 + jneq #1, drop + ret #-1 + drop: ret #0 + ** SECCOMP filter example: ld [4] /* offsetof(struct seccomp_data, arch) */ diff --git a/include/linux/filter.h b/include/linux/filter.h index 024fd03e5d18..759abf78dd61 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -223,6 +223,7 @@ enum { BPF_S_ANC_VLAN_TAG, BPF_S_ANC_VLAN_TAG_PRESENT, BPF_S_ANC_PAY_OFFSET, + BPF_S_ANC_RANDOM, }; #endif /* __LINUX_FILTER_H__ */ diff --git a/include/uapi/linux/filter.h b/include/uapi/linux/filter.h index 8eb9ccaa5b48..253b4d42cf2b 100644 --- a/include/uapi/linux/filter.h +++ b/include/uapi/linux/filter.h @@ -130,7 +130,8 @@ struct sock_fprog { /* Required for SO_ATTACH_FILTER. */ #define SKF_AD_VLAN_TAG 44 #define SKF_AD_VLAN_TAG_PRESENT 48 #define SKF_AD_PAY_OFFSET 52 -#define SKF_AD_MAX 56 +#define SKF_AD_RANDOM 56 +#define SKF_AD_MAX 60 #define SKF_NET_OFF (-0x100000) #define SKF_LL_OFF (-0x200000) diff --git a/net/core/filter.c b/net/core/filter.c index cd58614660cf..78a636e60a0b 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -643,6 +643,12 @@ static u64 __get_raw_cpu_id(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) return raw_smp_processor_id(); } +/* note that this only generates 32-bit random numbers */ +static u64 __get_random_u32(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) +{ + return (u64)prandom_u32(); +} + /* Register mappings for user programs. */ #define A_REG 0 #define X_REG 7 @@ -779,6 +785,7 @@ static bool convert_bpf_extensions(struct sock_filter *fp, case SKF_AD_OFF + SKF_AD_NLATTR: case SKF_AD_OFF + SKF_AD_NLATTR_NEST: case SKF_AD_OFF + SKF_AD_CPU: + case SKF_AD_OFF + SKF_AD_RANDOM: /* arg1 = ctx */ insn->code = BPF_ALU64 | BPF_MOV | BPF_X; insn->a_reg = ARG1_REG; @@ -812,6 +819,9 @@ static bool convert_bpf_extensions(struct sock_filter *fp, case SKF_AD_OFF + SKF_AD_CPU: insn->imm = __get_raw_cpu_id - __bpf_call_base; break; + case SKF_AD_OFF + SKF_AD_RANDOM: + insn->imm = __get_random_u32 - __bpf_call_base; + break; } break; @@ -1362,6 +1372,7 @@ int sk_chk_filter(struct sock_filter *filter, unsigned int flen) ANCILLARY(VLAN_TAG); ANCILLARY(VLAN_TAG_PRESENT); ANCILLARY(PAY_OFFSET); + ANCILLARY(RANDOM); } /* ancillary operation unknown or unsupported */ @@ -1746,6 +1757,7 @@ void sk_decode_filter(struct sock_filter *filt, struct sock_filter *to) [BPF_S_ANC_VLAN_TAG] = BPF_LD|BPF_B|BPF_ABS, [BPF_S_ANC_VLAN_TAG_PRESENT] = BPF_LD|BPF_B|BPF_ABS, [BPF_S_ANC_PAY_OFFSET] = BPF_LD|BPF_B|BPF_ABS, + [BPF_S_ANC_RANDOM] = BPF_LD|BPF_B|BPF_ABS, [BPF_S_LD_W_LEN] = BPF_LD|BPF_W|BPF_LEN, [BPF_S_LD_W_IND] = BPF_LD|BPF_W|BPF_IND, [BPF_S_LD_H_IND] = BPF_LD|BPF_H|BPF_IND, diff --git a/tools/net/bpf_exp.l b/tools/net/bpf_exp.l index bf7be77ddd62..833a96611da6 100644 --- a/tools/net/bpf_exp.l +++ b/tools/net/bpf_exp.l @@ -92,6 +92,7 @@ extern void yyerror(const char *str); "#"?("cpu") { return K_CPU; } "#"?("vlan_tci") { return K_VLANT; } "#"?("vlan_pr") { return K_VLANP; } +"#"?("rand") { return K_RAND; } ":" { return ':'; } "," { return ','; } diff --git a/tools/net/bpf_exp.y b/tools/net/bpf_exp.y index d15efc989ef5..e6306c51c26f 100644 --- a/tools/net/bpf_exp.y +++ b/tools/net/bpf_exp.y @@ -56,7 +56,7 @@ static void bpf_set_jmp_label(char *label, enum jmp_type type); %token OP_LDXI %token K_PKT_LEN K_PROTO K_TYPE K_NLATTR K_NLATTR_NEST K_MARK K_QUEUE K_HATYPE -%token K_RXHASH K_CPU K_IFIDX K_VLANT K_VLANP K_POFF +%token K_RXHASH K_CPU K_IFIDX K_VLANT K_VLANP K_POFF K_RAND %token ':' ',' '[' ']' '(' ')' 'x' 'a' '+' 'M' '*' '&' '#' '%' @@ -164,6 +164,9 @@ ldb | OP_LDB K_POFF { bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0, SKF_AD_OFF + SKF_AD_PAY_OFFSET); } + | OP_LDB K_RAND { + bpf_set_curr_instr(BPF_LD | BPF_B | BPF_ABS, 0, 0, + SKF_AD_OFF + SKF_AD_RANDOM); } ; ldh @@ -212,6 +215,9 @@ ldh | OP_LDH K_POFF { bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0, SKF_AD_OFF + SKF_AD_PAY_OFFSET); } + | OP_LDH K_RAND { + bpf_set_curr_instr(BPF_LD | BPF_H | BPF_ABS, 0, 0, + SKF_AD_OFF + SKF_AD_RANDOM); } ; ldi @@ -265,6 +271,9 @@ ld | OP_LD K_POFF { bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0, SKF_AD_OFF + SKF_AD_PAY_OFFSET); } + | OP_LD K_RAND { + bpf_set_curr_instr(BPF_LD | BPF_W | BPF_ABS, 0, 0, + SKF_AD_OFF + SKF_AD_RANDOM); } | OP_LD 'M' '[' number ']' { bpf_set_curr_instr(BPF_LD | BPF_MEM, 0, 0, $4); } | OP_LD '[' 'x' '+' number ']' { -- cgit v1.2.3 From 3a101b8de0d39403b2c7e5c23fd0b005668acf48 Mon Sep 17 00:00:00 2001 From: Richard Guy Briggs Date: Tue, 22 Apr 2014 21:31:56 -0400 Subject: audit: add netlink audit protocol bind to check capabilities on multicast join Register a netlink per-protocol bind fuction for audit to check userspace process capabilities before allowing a multicast group connection. Signed-off-by: Richard Guy Briggs Signed-off-by: David S. Miller --- include/uapi/linux/capability.h | 7 ++++++- kernel/audit.c | 10 ++++++++++ security/selinux/include/classmap.h | 2 +- 3 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/capability.h b/include/uapi/linux/capability.h index 154dd6d3c8fe..12c37a197d24 100644 --- a/include/uapi/linux/capability.h +++ b/include/uapi/linux/capability.h @@ -347,7 +347,12 @@ struct vfs_cap_data { #define CAP_BLOCK_SUSPEND 36 -#define CAP_LAST_CAP CAP_BLOCK_SUSPEND +/* Allow reading the audit log via multicast netlink socket */ + +#define CAP_AUDIT_READ 37 + + +#define CAP_LAST_CAP CAP_AUDIT_READ #define cap_valid(x) ((x) >= 0 && (x) <= CAP_LAST_CAP) diff --git a/kernel/audit.c b/kernel/audit.c index 7c2893602d06..223cb746f141 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -1076,10 +1076,20 @@ static void audit_receive(struct sk_buff *skb) mutex_unlock(&audit_cmd_mutex); } +/* Run custom bind function on netlink socket group connect or bind requests. */ +static int audit_bind(int group) +{ + if (!capable(CAP_AUDIT_READ)) + return -EPERM; + + return 0; +} + static int __net_init audit_net_init(struct net *net) { struct netlink_kernel_cfg cfg = { .input = audit_receive, + .bind = audit_bind, }; struct audit_net *aunet = net_generic(net, audit_net_id); diff --git a/security/selinux/include/classmap.h b/security/selinux/include/classmap.h index 14d04e63b1f0..be491a74c1ed 100644 --- a/security/selinux/include/classmap.h +++ b/security/selinux/include/classmap.h @@ -147,7 +147,7 @@ struct security_class_mapping secclass_map[] = { { "peer", { "recv", NULL } }, { "capability2", { "mac_override", "mac_admin", "syslog", "wake_alarm", "block_suspend", - NULL } }, + "audit_read", NULL } }, { "kernel_service", { "use_as_override", "create_files_as", NULL } }, { "tun_socket", { COMMON_SOCK_PERMS, "attach_queue", NULL } }, -- cgit v1.2.3 From 451f921639fea4600dfb9ab2889332bdcc7b48d3 Mon Sep 17 00:00:00 2001 From: Richard Guy Briggs Date: Tue, 22 Apr 2014 21:31:57 -0400 Subject: audit: add netlink multicast group for log read Add a netlink multicast socket with one group to kaudit for "best-effort" delivery to read-only userspace clients such as systemd, in addition to the existing bidirectional unicast auditd userspace client. Currently, auditd is intended to use the CAP_AUDIT_CONTROL and CAP_AUDIT_WRITE capabilities, but actually uses CAP_NET_ADMIN. The CAP_AUDIT_READ capability is added for use by read-only AUDIT_NLGRP_READLOG netlink multicast group clients to the kaudit subsystem. This will safely give access to services such as systemd to consume audit logs while ensuring write access remains restricted for integrity. Signed-off-by: Richard Guy Briggs Signed-off-by: David S. Miller --- include/uapi/linux/audit.h | 8 ++++++++ kernel/audit.c | 51 ++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 55 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h index 11917f747cb4..dfa4c860ccef 100644 --- a/include/uapi/linux/audit.h +++ b/include/uapi/linux/audit.h @@ -373,6 +373,14 @@ enum { */ #define AUDIT_MESSAGE_TEXT_MAX 8560 +/* Multicast Netlink socket groups (default up to 32) */ +enum audit_nlgrps { + AUDIT_NLGRP_NONE, /* Group 0 not used */ + AUDIT_NLGRP_READLOG, /* "best effort" read only socket */ + __AUDIT_NLGRP_MAX +}; +#define AUDIT_NLGRP_MAX (__AUDIT_NLGRP_MAX - 1) + struct audit_status { __u32 mask; /* Bit mask for valid entries */ __u32 enabled; /* 1 = enabled, 0 = disabled */ diff --git a/kernel/audit.c b/kernel/audit.c index 223cb746f141..d272cc11dff4 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -423,6 +423,35 @@ static void kauditd_send_skb(struct sk_buff *skb) consume_skb(skb); } +/* + * kauditd_send_multicast_skb - send the skb to multicast userspace listeners + * + * This function doesn't consume an skb as might be expected since it has to + * copy it anyways. + */ +static void kauditd_send_multicast_skb(struct sk_buff *skb) +{ + struct sk_buff *copy; + struct audit_net *aunet = net_generic(&init_net, audit_net_id); + struct sock *sock = aunet->nlsk; + + /* + * The seemingly wasteful skb_copy() rather than bumping the refcount + * using skb_get() is necessary because non-standard mods are made to + * the skb by the original kaudit unicast socket send routine. The + * existing auditd daemon assumes this breakage. Fixing this would + * require co-ordinating a change in the established protocol between + * the kaudit kernel subsystem and the auditd userspace code. There is + * no reason for new multicast clients to continue with this + * non-compliance. + */ + copy = skb_copy(skb, GFP_KERNEL); + if (!copy) + return; + + nlmsg_multicast(sock, copy, 0, AUDIT_NLGRP_READLOG, GFP_KERNEL); +} + /* * flush_hold_queue - empty the hold queue if auditd appears * @@ -1090,6 +1119,8 @@ static int __net_init audit_net_init(struct net *net) struct netlink_kernel_cfg cfg = { .input = audit_receive, .bind = audit_bind, + .flags = NL_CFG_F_NONROOT_RECV, + .groups = AUDIT_NLGRP_MAX, }; struct audit_net *aunet = net_generic(net, audit_net_id); @@ -1911,10 +1942,10 @@ out: * audit_log_end - end one audit record * @ab: the audit_buffer * - * The netlink_* functions cannot be called inside an irq context, so - * the audit buffer is placed on a queue and a tasklet is scheduled to - * remove them from the queue outside the irq context. May be called in - * any context. + * netlink_unicast() cannot be called inside an irq context because it blocks + * (last arg, flags, is not set to MSG_DONTWAIT), so the audit buffer is placed + * on a queue and a tasklet is scheduled to remove them from the queue outside + * the irq context. May be called in any context. */ void audit_log_end(struct audit_buffer *ab) { @@ -1924,6 +1955,18 @@ void audit_log_end(struct audit_buffer *ab) audit_log_lost("rate limit exceeded"); } else { struct nlmsghdr *nlh = nlmsg_hdr(ab->skb); + + kauditd_send_multicast_skb(ab->skb); + + /* + * The original kaudit unicast socket sends up messages with + * nlmsg_len set to the payload length rather than the entire + * message length. This breaks the standard set by netlink. + * The existing auditd daemon assumes this breakage. Fixing + * this would require co-ordinating a change in the established + * protocol between the kaudit kernel subsystem and the auditd + * userspace code. + */ nlh->nlmsg_len = ab->skb->len - NLMSG_HDRLEN; if (audit_pid) { -- cgit v1.2.3 From f5efc696cc711021cc73e7543cc3038e58459707 Mon Sep 17 00:00:00 2001 From: Tomasz Bursztyka Date: Mon, 14 Apr 2014 15:41:28 +0300 Subject: netfilter: nf_tables: Add meta expression key for bridge interface name NFT_META_BRI_IIFNAME to get packet input bridge interface name NFT_META_BRI_OIFNAME to get packet output bridge interface name Such meta key are accessible only through NFPROTO_BRIDGE family, on a dedicated nft meta module: nft_meta_bridge. Suggested-by: Pablo Neira Ayuso Signed-off-by: Tomasz Bursztyka Signed-off-by: Pablo Neira Ayuso --- include/uapi/linux/netfilter/nf_tables.h | 4 + net/bridge/Makefile | 2 +- net/bridge/netfilter/Kconfig | 14 +++- net/bridge/netfilter/Makefile | 1 + net/bridge/netfilter/nft_meta_bridge.c | 139 +++++++++++++++++++++++++++++++ 5 files changed, 158 insertions(+), 2 deletions(-) create mode 100644 net/bridge/netfilter/nft_meta_bridge.c (limited to 'include/uapi') diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 160159274cab..7d6433f66bf8 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -563,6 +563,8 @@ enum nft_exthdr_attributes { * @NFT_META_SECMARK: packet secmark (skb->secmark) * @NFT_META_NFPROTO: netfilter protocol * @NFT_META_L4PROTO: layer 4 protocol number + * @NFT_META_BRI_IIFNAME: packet input bridge interface name + * @NFT_META_BRI_OIFNAME: packet output bridge interface name */ enum nft_meta_keys { NFT_META_LEN, @@ -582,6 +584,8 @@ enum nft_meta_keys { NFT_META_SECMARK, NFT_META_NFPROTO, NFT_META_L4PROTO, + NFT_META_BRI_IIFNAME, + NFT_META_BRI_OIFNAME, }; /** diff --git a/net/bridge/Makefile b/net/bridge/Makefile index e85498b2f166..906a18b4e74a 100644 --- a/net/bridge/Makefile +++ b/net/bridge/Makefile @@ -16,4 +16,4 @@ bridge-$(CONFIG_BRIDGE_IGMP_SNOOPING) += br_multicast.o br_mdb.o bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o -obj-$(CONFIG_BRIDGE_NF_EBTABLES) += netfilter/ +obj-$(CONFIG_BRIDGE_NETFILTER) += netfilter/ diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig index 5ca74a0e595f..3baf29d34e62 100644 --- a/net/bridge/netfilter/Kconfig +++ b/net/bridge/netfilter/Kconfig @@ -2,13 +2,25 @@ # Bridge netfilter configuration # # -config NF_TABLES_BRIDGE +menuconfig NF_TABLES_BRIDGE depends on NF_TABLES + select BRIDGE_NETFILTER tristate "Ethernet Bridge nf_tables support" +if NF_TABLES_BRIDGE + +config NFT_BRIDGE_META + tristate "Netfilter nf_table bridge meta support" + depends on NFT_META + help + Add support for bridge dedicated meta key. + +endif # NF_TABLES_BRIDGE + menuconfig BRIDGE_NF_EBTABLES tristate "Ethernet Bridge tables (ebtables) support" depends on BRIDGE && NETFILTER + select BRIDGE_NETFILTER select NETFILTER_XTABLES help ebtables is a general, extensible frame/packet identification diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile index ea7629f58b3d..6f2f3943d66f 100644 --- a/net/bridge/netfilter/Makefile +++ b/net/bridge/netfilter/Makefile @@ -3,6 +3,7 @@ # obj-$(CONFIG_NF_TABLES_BRIDGE) += nf_tables_bridge.o +obj-$(CONFIG_NFT_BRIDGE_META) += nft_meta_bridge.o obj-$(CONFIG_BRIDGE_NF_EBTABLES) += ebtables.o diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c new file mode 100644 index 000000000000..4f02109d708f --- /dev/null +++ b/net/bridge/netfilter/nft_meta_bridge.c @@ -0,0 +1,139 @@ +/* + * Copyright (c) 2014 Intel Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../br_private.h" + +static void nft_meta_bridge_get_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + const struct nft_meta *priv = nft_expr_priv(expr); + const struct net_device *in = pkt->in, *out = pkt->out; + struct nft_data *dest = &data[priv->dreg]; + const struct net_bridge_port *p; + + switch (priv->key) { + case NFT_META_BRI_IIFNAME: + if (in == NULL || (p = br_port_get_rcu(in)) == NULL) + goto err; + break; + case NFT_META_BRI_OIFNAME: + if (out == NULL || (p = br_port_get_rcu(out)) == NULL) + goto err; + break; + default: + goto out; + } + + strncpy((char *)dest->data, p->br->dev->name, sizeof(dest->data)); + return; +out: + return nft_meta_get_eval(expr, data, pkt); +err: + data[NFT_REG_VERDICT].verdict = NFT_BREAK; +} + +static int nft_meta_bridge_get_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_meta *priv = nft_expr_priv(expr); + int err; + + priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); + switch (priv->key) { + case NFT_META_BRI_IIFNAME: + case NFT_META_BRI_OIFNAME: + break; + default: + return nft_meta_get_init(ctx, expr, tb); + } + + priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + err = nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); + if (err < 0) + return err; + + return 0; +} + +static struct nft_expr_type nft_meta_bridge_type; +static const struct nft_expr_ops nft_meta_bridge_get_ops = { + .type = &nft_meta_bridge_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), + .eval = nft_meta_bridge_get_eval, + .init = nft_meta_bridge_get_init, + .dump = nft_meta_get_dump, +}; + +static const struct nft_expr_ops nft_meta_bridge_set_ops = { + .type = &nft_meta_bridge_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), + .eval = nft_meta_set_eval, + .init = nft_meta_set_init, + .dump = nft_meta_set_dump, +}; + +static const struct nft_expr_ops * +nft_meta_bridge_select_ops(const struct nft_ctx *ctx, + const struct nlattr * const tb[]) +{ + if (tb[NFTA_META_KEY] == NULL) + return ERR_PTR(-EINVAL); + + if (tb[NFTA_META_DREG] && tb[NFTA_META_SREG]) + return ERR_PTR(-EINVAL); + + if (tb[NFTA_META_DREG]) + return &nft_meta_bridge_get_ops; + + if (tb[NFTA_META_SREG]) + return &nft_meta_bridge_set_ops; + + return ERR_PTR(-EINVAL); +} + +static struct nft_expr_type nft_meta_bridge_type __read_mostly = { + .family = NFPROTO_BRIDGE, + .name = "meta", + .select_ops = &nft_meta_bridge_select_ops, + .policy = nft_meta_policy, + .maxattr = NFTA_META_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_meta_bridge_module_init(void) +{ + return nft_register_expr(&nft_meta_bridge_type); +} + +static void __exit nft_meta_bridge_module_exit(void) +{ + nft_unregister_expr(&nft_meta_bridge_type); +} + +module_init(nft_meta_bridge_module_init); +module_exit(nft_meta_bridge_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Tomasz Bursztyka "); +MODULE_ALIAS_NFT_AF_EXPR(AF_BRIDGE, "meta"); -- cgit v1.2.3 From ea077c1cea36a6b5ded1256dcd56c72ff2a22c62 Mon Sep 17 00:00:00 2001 From: Rostislav Lisovy Date: Tue, 15 Apr 2014 14:37:55 +0200 Subject: cfg80211: Add attributes describing prohibited channel bandwidth Since there are frequency bands (e.g. 5.9GHz) allowing channels with only 10 or 5 MHz bandwidth, this patch adds attributes that allow keeping track about this information. When channel attributes are reported to user-space, make sure to not break old tools, i.e. if the 'split wiphy dump' is enabled, report the extra attributes (if present) describing the bandwidth restrictions. If the 'split wiphy dump' is not enabled, completely omit those channels that have flags set to either IEEE80211_CHAN_NO_10MHZ or IEEE80211_CHAN_NO_20MHZ. Add the check for new bandwidth restriction flags in cfg80211_chandef_usable() to comply with the restrictions. Signed-off-by: Rostislav Lisovy Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 6 ++++++ include/uapi/linux/nl80211.h | 6 ++++++ net/wireless/chan.c | 2 ++ net/wireless/nl80211.c | 13 +++++++++++++ 4 files changed, 27 insertions(+) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 3dd2cb465540..c98cf08538b9 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -111,6 +111,10 @@ enum ieee80211_band { * restrictions. * @IEEE80211_CHAN_INDOOR_ONLY: see %NL80211_FREQUENCY_ATTR_INDOOR_ONLY * @IEEE80211_CHAN_GO_CONCURRENT: see %NL80211_FREQUENCY_ATTR_GO_CONCURRENT + * @IEEE80211_CHAN_NO_20MHZ: 20 MHz bandwidth is not permitted + * on this channel. + * @IEEE80211_CHAN_NO_10MHZ: 10 MHz bandwidth is not permitted + * on this channel. * */ enum ieee80211_channel_flags { @@ -125,6 +129,8 @@ enum ieee80211_channel_flags { IEEE80211_CHAN_NO_160MHZ = 1<<8, IEEE80211_CHAN_INDOOR_ONLY = 1<<9, IEEE80211_CHAN_GO_CONCURRENT = 1<<10, + IEEE80211_CHAN_NO_20MHZ = 1<<11, + IEEE80211_CHAN_NO_10MHZ = 1<<12, }; #define IEEE80211_CHAN_NO_HT40 \ diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 513bfd7b2e5f..0592032ff160 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2358,6 +2358,10 @@ enum nl80211_band_attr { * connected to an AP with DFS and radar detection on the UNII band (it is * up to user-space, i.e., wpa_supplicant to perform the required * verifications) + * @NL80211_FREQUENCY_ATTR_NO_20MHZ: 20 MHz operation is not allowed + * on this channel in current regulatory domain. + * @NL80211_FREQUENCY_ATTR_NO_10MHZ: 10 MHz operation is not allowed + * on this channel in current regulatory domain. * @NL80211_FREQUENCY_ATTR_MAX: highest frequency attribute number * currently defined * @__NL80211_FREQUENCY_ATTR_AFTER_LAST: internal use @@ -2384,6 +2388,8 @@ enum nl80211_frequency_attr { NL80211_FREQUENCY_ATTR_DFS_CAC_TIME, NL80211_FREQUENCY_ATTR_INDOOR_ONLY, NL80211_FREQUENCY_ATTR_GO_CONCURRENT, + NL80211_FREQUENCY_ATTR_NO_20MHZ, + NL80211_FREQUENCY_ATTR_NO_10MHZ, /* keep last */ __NL80211_FREQUENCY_ATTR_AFTER_LAST, diff --git a/net/wireless/chan.c b/net/wireless/chan.c index 2adf7b2eccbc..84d686e2dbd0 100644 --- a/net/wireless/chan.c +++ b/net/wireless/chan.c @@ -616,12 +616,14 @@ bool cfg80211_chandef_usable(struct wiphy *wiphy, width = 5; break; case NL80211_CHAN_WIDTH_10: + prohibited_flags |= IEEE80211_CHAN_NO_10MHZ; width = 10; break; case NL80211_CHAN_WIDTH_20: if (!ht_cap->ht_supported) return false; case NL80211_CHAN_WIDTH_20_NOHT: + prohibited_flags |= IEEE80211_CHAN_NO_20MHZ; width = 20; break; case NL80211_CHAN_WIDTH_40: diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index fce423a4e96a..ca75f60041d2 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -567,6 +567,13 @@ static int nl80211_msg_put_channel(struct sk_buff *msg, struct ieee80211_channel *chan, bool large) { + /* Some channels must be completely excluded from the + * list to protect old user-space tools from breaking + */ + if (!large && chan->flags & + (IEEE80211_CHAN_NO_10MHZ | IEEE80211_CHAN_NO_20MHZ)) + return 0; + if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_FREQ, chan->center_freq)) goto nla_put_failure; @@ -620,6 +627,12 @@ static int nl80211_msg_put_channel(struct sk_buff *msg, if ((chan->flags & IEEE80211_CHAN_GO_CONCURRENT) && nla_put_flag(msg, NL80211_FREQUENCY_ATTR_GO_CONCURRENT)) goto nla_put_failure; + if ((chan->flags & IEEE80211_CHAN_NO_20MHZ) && + nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_20MHZ)) + goto nla_put_failure; + if ((chan->flags & IEEE80211_CHAN_NO_10MHZ) && + nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_10MHZ)) + goto nla_put_failure; } if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_MAX_TX_POWER, -- cgit v1.2.3 From a89778d8baf19cd7e728d81121a294a06cedaad1 Mon Sep 17 00:00:00 2001 From: Erik Hugne Date: Thu, 24 Apr 2014 16:26:46 +0200 Subject: tipc: add support for link state subscriptions When links are established over a bearer plane, we create a node local publication containing information about the peer node and bearer plane. This allows TIPC applications to use the standard TIPC topology server subscription mechanism to get notifications when a link goes up or down. Signed-off-by: Erik Hugne Reviewed-by: Ying Xue Reviewed-by: Jon Maloy Signed-off-by: David S. Miller --- include/uapi/linux/tipc.h | 1 + net/tipc/node.c | 6 +++++- 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/tipc.h b/include/uapi/linux/tipc.h index 852373d27dbb..53cd7902d34e 100644 --- a/include/uapi/linux/tipc.h +++ b/include/uapi/linux/tipc.h @@ -87,6 +87,7 @@ static inline unsigned int tipc_node(__u32 addr) #define TIPC_CFG_SRV 0 /* configuration service name type */ #define TIPC_TOP_SRV 1 /* topology service name type */ +#define TIPC_LINK_STATE 2 /* link state name type */ #define TIPC_RESERVED_TYPES 64 /* lowest user-publishable name type */ /* diff --git a/net/tipc/node.c b/net/tipc/node.c index be90115cda1a..3b86a74cb31f 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -144,9 +144,11 @@ void tipc_node_stop(void) void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { struct tipc_link **active = &n_ptr->active_links[0]; + u32 addr = n_ptr->addr; n_ptr->working_links++; - + tipc_nametbl_publish(TIPC_LINK_STATE, addr, addr, TIPC_NODE_SCOPE, + l_ptr->bearer_id, addr); pr_info("Established link <%s> on network plane %c\n", l_ptr->name, l_ptr->net_plane); @@ -203,8 +205,10 @@ static void node_select_active_links(struct tipc_node *n_ptr) void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { struct tipc_link **active; + u32 addr = n_ptr->addr; n_ptr->working_links--; + tipc_nametbl_withdraw(TIPC_LINK_STATE, addr, l_ptr->bearer_id, addr); if (!tipc_link_is_active(l_ptr)) { pr_info("Lost standby link <%s> on network plane %c\n", -- cgit v1.2.3 From 78acb1f9b898e85fa2c1e28e700b54b66b288e8d Mon Sep 17 00:00:00 2001 From: Erik Hugne Date: Thu, 24 Apr 2014 16:26:47 +0200 Subject: tipc: add ioctl to fetch link names We add a new ioctl for AF_TIPC that can be used to fetch the logical name for a link to a remote node on a given bearer. This should be used in combination with link state subscriptions. The logical name size limit definitions are moved to tipc.h, as they are now also needed by the new ioctl. Signed-off-by: Erik Hugne Reviewed-by: Ying Xue Reviewed-by: Jon Maloy Signed-off-by: David S. Miller --- include/uapi/linux/tipc.h | 22 ++++++++++++++++++++++ include/uapi/linux/tipc_config.h | 10 +--------- net/tipc/node.c | 27 +++++++++++++++++++++++++++ net/tipc/node.h | 1 + net/tipc/socket.c | 29 ++++++++++++++++++++++++++--- 5 files changed, 77 insertions(+), 12 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/tipc.h b/include/uapi/linux/tipc.h index 53cd7902d34e..6f71b9b41595 100644 --- a/include/uapi/linux/tipc.h +++ b/include/uapi/linux/tipc.h @@ -38,6 +38,7 @@ #define _LINUX_TIPC_H_ #include +#include /* * TIPC addressing primitives @@ -207,4 +208,25 @@ struct sockaddr_tipc { #define TIPC_NODE_RECVQ_DEPTH 131 /* Default: none (read only) */ #define TIPC_SOCK_RECVQ_DEPTH 132 /* Default: none (read only) */ +/* + * Maximum sizes of TIPC bearer-related names (including terminating NULL) + * The string formatting for each name element is: + * media: media + * interface: media:interface name + * link: Z.C.N:interface-Z.C.N:interface + * + */ + +#define TIPC_MAX_MEDIA_NAME 16 +#define TIPC_MAX_IF_NAME 16 +#define TIPC_MAX_BEARER_NAME 32 +#define TIPC_MAX_LINK_NAME 60 + +#define SIOCGETLINKNAME SIOCPROTOPRIVATE + +struct tipc_sioc_ln_req { + __u32 peer; + __u32 bearer_id; + char linkname[TIPC_MAX_LINK_NAME]; +}; #endif diff --git a/include/uapi/linux/tipc_config.h b/include/uapi/linux/tipc_config.h index 6b0bff09b3a7..41a76acbb305 100644 --- a/include/uapi/linux/tipc_config.h +++ b/include/uapi/linux/tipc_config.h @@ -39,6 +39,7 @@ #include #include +#include #include #ifndef __KERNEL__ @@ -154,15 +155,6 @@ #define TIPC_TLV_NAME_TBL_QUERY 25 /* struct tipc_name_table_query */ #define TIPC_TLV_PORT_REF 26 /* 32-bit port reference */ -/* - * Maximum sizes of TIPC bearer-related names (including terminating NUL) - */ - -#define TIPC_MAX_MEDIA_NAME 16 /* format = media */ -#define TIPC_MAX_IF_NAME 16 /* format = interface */ -#define TIPC_MAX_BEARER_NAME 32 /* format = media:interface */ -#define TIPC_MAX_LINK_NAME 60 /* format = Z.C.N:interface-Z.C.N:interface */ - /* * Link priority limits (min, default, max, media default) */ diff --git a/net/tipc/node.c b/net/tipc/node.c index 3b86a74cb31f..1f938f3dba4b 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -438,3 +438,30 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) rcu_read_unlock(); return buf; } + +/** + * tipc_node_get_linkname - get the name of a link + * + * @bearer_id: id of the bearer + * @node: peer node address + * @linkname: link name output buffer + * + * Returns 0 on success + */ +int tipc_node_get_linkname(u32 bearer_id, u32 addr, char *linkname, size_t len) +{ + struct tipc_link *link; + struct tipc_node *node = tipc_node_find(addr); + + if ((bearer_id > MAX_BEARERS) || !node) + return -EINVAL; + tipc_node_lock(node); + link = node->links[bearer_id]; + if (link) { + strncpy(linkname, link->name, len); + tipc_node_unlock(node); + return 0; + } + tipc_node_unlock(node); + return -EINVAL; +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 7cbb8cec1a93..411b19114064 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -118,6 +118,7 @@ int tipc_node_active_links(struct tipc_node *n_ptr); int tipc_node_is_up(struct tipc_node *n_ptr); struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space); struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space); +int tipc_node_get_linkname(u32 bearer_id, u32 node, char *linkname, size_t len); static inline void tipc_node_lock(struct tipc_node *n_ptr) { diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 3c0256962f7d..3f9912f87d0d 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -36,6 +36,7 @@ #include "core.h" #include "port.h" +#include "node.h" #include @@ -1905,6 +1906,28 @@ static int tipc_getsockopt(struct socket *sock, int lvl, int opt, return put_user(sizeof(value), ol); } +int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) +{ + struct tipc_sioc_ln_req lnr; + void __user *argp = (void __user *)arg; + + switch (cmd) { + case SIOCGETLINKNAME: + if (copy_from_user(&lnr, argp, sizeof(lnr))) + return -EFAULT; + if (!tipc_node_get_linkname(lnr.bearer_id, lnr.peer, + lnr.linkname, TIPC_MAX_LINK_NAME)) { + if (copy_to_user(argp, &lnr, sizeof(lnr))) + return -EFAULT; + return 0; + } + return -EADDRNOTAVAIL; + break; + default: + return -ENOIOCTLCMD; + } +} + /* Protocol switches for the various types of TIPC sockets */ static const struct proto_ops msg_ops = { @@ -1917,7 +1940,7 @@ static const struct proto_ops msg_ops = { .accept = sock_no_accept, .getname = tipc_getname, .poll = tipc_poll, - .ioctl = sock_no_ioctl, + .ioctl = tipc_ioctl, .listen = sock_no_listen, .shutdown = tipc_shutdown, .setsockopt = tipc_setsockopt, @@ -1938,7 +1961,7 @@ static const struct proto_ops packet_ops = { .accept = tipc_accept, .getname = tipc_getname, .poll = tipc_poll, - .ioctl = sock_no_ioctl, + .ioctl = tipc_ioctl, .listen = tipc_listen, .shutdown = tipc_shutdown, .setsockopt = tipc_setsockopt, @@ -1959,7 +1982,7 @@ static const struct proto_ops stream_ops = { .accept = tipc_accept, .getname = tipc_getname, .poll = tipc_poll, - .ioctl = sock_no_ioctl, + .ioctl = tipc_ioctl, .listen = tipc_listen, .shutdown = tipc_shutdown, .setsockopt = tipc_setsockopt, -- cgit v1.2.3 From eb11022dca4eb22def72d5d4e3140caa357b34e1 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Fri, 25 Apr 2014 10:35:07 +0100 Subject: FDDI: Reformat for 8-character tabs Some of our FDDI support code has been apparently written with an assumption that tabs are 4-character wide. In preparation to the next change this update reformats so that it stays within 79 columns and otherwise renders correctly with 8-character tabs. No functional change. Signed-off-by: Maciej W. Rozycki Signed-off-by: David S. Miller --- include/uapi/linux/if_fddi.h | 90 ++++++++++++++++++++++---------------------- 1 file changed, 46 insertions(+), 44 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_fddi.h b/include/uapi/linux/if_fddi.h index 0d36909c3aef..1086cd9f6754 100644 --- a/include/uapi/linux/if_fddi.h +++ b/include/uapi/linux/if_fddi.h @@ -30,74 +30,76 @@ * Define max and min legal sizes. The frame sizes do not include * 4 byte FCS/CRC (frame check sequence). */ -#define FDDI_K_ALEN 6 /* Octets in one FDDI address */ -#define FDDI_K_8022_HLEN 16 /* Total octets in 802.2 header */ -#define FDDI_K_SNAP_HLEN 21 /* Total octets in 802.2 SNAP header */ -#define FDDI_K_8022_ZLEN 16 /* Min octets in 802.2 frame sans FCS */ -#define FDDI_K_SNAP_ZLEN 21 /* Min octets in 802.2 SNAP frame sans FCS */ +#define FDDI_K_ALEN 6 /* Octets in one FDDI address */ +#define FDDI_K_8022_HLEN 16 /* Total octets in 802.2 header */ +#define FDDI_K_SNAP_HLEN 21 /* Total octets in 802.2 SNAP header */ +#define FDDI_K_8022_ZLEN 16 /* Min octets in 802.2 frame sans + FCS */ +#define FDDI_K_SNAP_ZLEN 21 /* Min octets in 802.2 SNAP frame sans + FCS */ #define FDDI_K_8022_DLEN 4475 /* Max octets in 802.2 payload */ #define FDDI_K_SNAP_DLEN 4470 /* Max octets in 802.2 SNAP payload */ -#define FDDI_K_LLC_ZLEN 13 /* Min octets in LLC frame sans FCS */ +#define FDDI_K_LLC_ZLEN 13 /* Min octets in LLC frame sans FCS */ #define FDDI_K_LLC_LEN 4491 /* Max octets in LLC frame sans FCS */ +#define FDDI_K_OUI_LEN 3 /* Octets in OUI in 802.2 SNAP + header */ /* Define FDDI Frame Control (FC) Byte values */ -#define FDDI_FC_K_VOID 0x00 -#define FDDI_FC_K_NON_RESTRICTED_TOKEN 0x80 -#define FDDI_FC_K_RESTRICTED_TOKEN 0xC0 -#define FDDI_FC_K_SMT_MIN 0x41 -#define FDDI_FC_K_SMT_MAX 0x4F -#define FDDI_FC_K_MAC_MIN 0xC1 -#define FDDI_FC_K_MAC_MAX 0xCF -#define FDDI_FC_K_ASYNC_LLC_MIN 0x50 -#define FDDI_FC_K_ASYNC_LLC_DEF 0x54 -#define FDDI_FC_K_ASYNC_LLC_MAX 0x5F -#define FDDI_FC_K_SYNC_LLC_MIN 0xD0 -#define FDDI_FC_K_SYNC_LLC_MAX 0xD7 -#define FDDI_FC_K_IMPLEMENTOR_MIN 0x60 -#define FDDI_FC_K_IMPLEMENTOR_MAX 0x6F -#define FDDI_FC_K_RESERVED_MIN 0x70 -#define FDDI_FC_K_RESERVED_MAX 0x7F +#define FDDI_FC_K_VOID 0x00 +#define FDDI_FC_K_NON_RESTRICTED_TOKEN 0x80 +#define FDDI_FC_K_RESTRICTED_TOKEN 0xC0 +#define FDDI_FC_K_SMT_MIN 0x41 +#define FDDI_FC_K_SMT_MAX 0x4F +#define FDDI_FC_K_MAC_MIN 0xC1 +#define FDDI_FC_K_MAC_MAX 0xCF +#define FDDI_FC_K_ASYNC_LLC_MIN 0x50 +#define FDDI_FC_K_ASYNC_LLC_DEF 0x54 +#define FDDI_FC_K_ASYNC_LLC_MAX 0x5F +#define FDDI_FC_K_SYNC_LLC_MIN 0xD0 +#define FDDI_FC_K_SYNC_LLC_MAX 0xD7 +#define FDDI_FC_K_IMPLEMENTOR_MIN 0x60 +#define FDDI_FC_K_IMPLEMENTOR_MAX 0x6F +#define FDDI_FC_K_RESERVED_MIN 0x70 +#define FDDI_FC_K_RESERVED_MAX 0x7F /* Define LLC and SNAP constants */ -#define FDDI_EXTENDED_SAP 0xAA +#define FDDI_EXTENDED_SAP 0xAA #define FDDI_UI_CMD 0x03 /* Define 802.2 Type 1 header */ struct fddi_8022_1_hdr { - __u8 dsap; /* destination service access point */ - __u8 ssap; /* source service access point */ - __u8 ctrl; /* control byte #1 */ + __u8 dsap; /* destination service access point */ + __u8 ssap; /* source service access point */ + __u8 ctrl; /* control byte #1 */ } __attribute__((packed)); /* Define 802.2 Type 2 header */ struct fddi_8022_2_hdr { - __u8 dsap; /* destination service access point */ - __u8 ssap; /* source service access point */ - __u8 ctrl_1; /* control byte #1 */ - __u8 ctrl_2; /* control byte #2 */ + __u8 dsap; /* destination service access point */ + __u8 ssap; /* source service access point */ + __u8 ctrl_1; /* control byte #1 */ + __u8 ctrl_2; /* control byte #2 */ } __attribute__((packed)); /* Define 802.2 SNAP header */ -#define FDDI_K_OUI_LEN 3 struct fddi_snap_hdr { - __u8 dsap; /* always 0xAA */ - __u8 ssap; /* always 0xAA */ - __u8 ctrl; /* always 0x03 */ + __u8 dsap; /* always 0xAA */ + __u8 ssap; /* always 0xAA */ + __u8 ctrl; /* always 0x03 */ __u8 oui[FDDI_K_OUI_LEN]; /* organizational universal id */ - __be16 ethertype; /* packet type ID field */ + __be16 ethertype; /* packet type ID field */ } __attribute__((packed)); /* Define FDDI LLC frame header */ struct fddihdr { - __u8 fc; /* frame control */ - __u8 daddr[FDDI_K_ALEN]; /* destination address */ - __u8 saddr[FDDI_K_ALEN]; /* source address */ - union - { - struct fddi_8022_1_hdr llc_8022_1; - struct fddi_8022_2_hdr llc_8022_2; - struct fddi_snap_hdr llc_snap; - } hdr; + __u8 fc; /* frame control */ + __u8 daddr[FDDI_K_ALEN]; /* destination address */ + __u8 saddr[FDDI_K_ALEN]; /* source address */ + union { + struct fddi_8022_1_hdr llc_8022_1; + struct fddi_8022_2_hdr llc_8022_2; + struct fddi_snap_hdr llc_snap; + } hdr; } __attribute__((packed)); -- cgit v1.2.3 From e16821bcfb364b0c41142db275dc74b39fa42c30 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Mon, 28 Apr 2014 11:22:08 +0300 Subject: cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 8 ++++++++ include/uapi/linux/nl80211.h | 4 ++++ net/wireless/nl80211.c | 40 ++++++++++++++++++++++++++++------------ net/wireless/rdev-ops.h | 13 +++++++++++++ net/wireless/trace.h | 18 ++++++++++++++++++ 5 files changed, 71 insertions(+), 12 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index c98cf08538b9..7eae46ccec01 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2290,6 +2290,10 @@ struct cfg80211_qos_map { * @channel_switch: initiate channel-switch procedure (with CSA) * * @set_qos_map: Set QoS mapping information to the driver + * + * @set_ap_chanwidth: Set the AP (including P2P GO) mode channel width for the + * given interface This is used e.g. for dynamic HT 20/40 MHz channel width + * changes during the lifetime of the BSS. */ struct cfg80211_ops { int (*suspend)(struct wiphy *wiphy, struct cfg80211_wowlan *wow); @@ -2533,9 +2537,13 @@ struct cfg80211_ops { int (*channel_switch)(struct wiphy *wiphy, struct net_device *dev, struct cfg80211_csa_settings *params); + int (*set_qos_map)(struct wiphy *wiphy, struct net_device *dev, struct cfg80211_qos_map *qos_map); + + int (*set_ap_chanwidth)(struct wiphy *wiphy, struct net_device *dev, + struct cfg80211_chan_def *chandef); }; /* diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 0592032ff160..406010d4def0 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -3929,6 +3929,9 @@ enum nl80211_ap_sme_features { * interface. An active monitor interface behaves like a normal monitor * interface, but gets added to the driver. It ensures that incoming * unicast packets directed at the configured interface address get ACKed. + * @NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE: This driver supports dynamic + * channel bandwidth change (e.g., HT 20 <-> 40 MHz channel) during the + * lifetime of a BSS. */ enum nl80211_feature_flags { NL80211_FEATURE_SK_TX_STATUS = 1 << 0, @@ -3949,6 +3952,7 @@ enum nl80211_feature_flags { NL80211_FEATURE_FULL_AP_CLIENT_STATE = 1 << 15, NL80211_FEATURE_USERSPACE_MPM = 1 << 16, NL80211_FEATURE_ACTIVE_MONITOR = 1 << 17, + NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE = 1 << 18, }; /** diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index ca75f60041d2..0f1b18f209d6 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1928,18 +1928,20 @@ static int nl80211_parse_chandef(struct cfg80211_registered_device *rdev, } static int __nl80211_set_channel(struct cfg80211_registered_device *rdev, - struct wireless_dev *wdev, + struct net_device *dev, struct genl_info *info) { struct cfg80211_chan_def chandef; int result; enum nl80211_iftype iftype = NL80211_IFTYPE_MONITOR; + struct wireless_dev *wdev = NULL; - if (wdev) - iftype = wdev->iftype; - + if (dev) + wdev = dev->ieee80211_ptr; if (!nl80211_can_set_dev_channel(wdev)) return -EOPNOTSUPP; + if (wdev) + iftype = wdev->iftype; result = nl80211_parse_chandef(rdev, info, &chandef); if (result) @@ -1948,14 +1950,27 @@ static int __nl80211_set_channel(struct cfg80211_registered_device *rdev, switch (iftype) { case NL80211_IFTYPE_AP: case NL80211_IFTYPE_P2P_GO: - if (wdev->beacon_interval) { - result = -EBUSY; - break; - } if (!cfg80211_reg_can_beacon(&rdev->wiphy, &chandef, iftype)) { result = -EINVAL; break; } + if (wdev->beacon_interval) { + if (!dev || !rdev->ops->set_ap_chanwidth || + !(rdev->wiphy.features & + NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE)) { + result = -EBUSY; + break; + } + + /* Only allow dynamic channel width changes */ + if (chandef.chan != wdev->preset_chandef.chan) { + result = -EBUSY; + break; + } + result = rdev_set_ap_chanwidth(rdev, dev, &chandef); + if (result) + break; + } wdev->preset_chandef = chandef; result = 0; break; @@ -1977,7 +1992,7 @@ static int nl80211_set_channel(struct sk_buff *skb, struct genl_info *info) struct cfg80211_registered_device *rdev = info->user_ptr[0]; struct net_device *netdev = info->user_ptr[1]; - return __nl80211_set_channel(rdev, netdev->ieee80211_ptr, info); + return __nl80211_set_channel(rdev, netdev, info); } static int nl80211_set_wds_peer(struct sk_buff *skb, struct genl_info *info) @@ -2099,9 +2114,10 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) } if (info->attrs[NL80211_ATTR_WIPHY_FREQ]) { - result = __nl80211_set_channel(rdev, - nl80211_can_set_dev_channel(wdev) ? wdev : NULL, - info); + result = __nl80211_set_channel( + rdev, + nl80211_can_set_dev_channel(wdev) ? netdev : NULL, + info); if (result) return result; } diff --git a/net/wireless/rdev-ops.h b/net/wireless/rdev-ops.h index 74d97d33c938..00cdf73ba6c4 100644 --- a/net/wireless/rdev-ops.h +++ b/net/wireless/rdev-ops.h @@ -950,4 +950,17 @@ static inline int rdev_set_qos_map(struct cfg80211_registered_device *rdev, return ret; } +static inline int +rdev_set_ap_chanwidth(struct cfg80211_registered_device *rdev, + struct net_device *dev, struct cfg80211_chan_def *chandef) +{ + int ret; + + trace_rdev_set_ap_chanwidth(&rdev->wiphy, dev, chandef); + ret = rdev->ops->set_ap_chanwidth(&rdev->wiphy, dev, chandef); + trace_rdev_return_int(&rdev->wiphy, ret); + + return ret; +} + #endif /* __CFG80211_RDEV_OPS */ diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 47b499f8f54d..f3c13ff4d04c 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -1919,6 +1919,24 @@ TRACE_EVENT(rdev_set_qos_map, WIPHY_PR_ARG, NETDEV_PR_ARG, __entry->num_des) ); +TRACE_EVENT(rdev_set_ap_chanwidth, + TP_PROTO(struct wiphy *wiphy, struct net_device *netdev, + struct cfg80211_chan_def *chandef), + TP_ARGS(wiphy, netdev, chandef), + TP_STRUCT__entry( + WIPHY_ENTRY + NETDEV_ENTRY + CHAN_DEF_ENTRY + ), + TP_fast_assign( + WIPHY_ASSIGN; + NETDEV_ASSIGN; + CHAN_DEF_ASSIGN(chandef); + ), + TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", " CHAN_DEF_PR_FMT, + WIPHY_PR_ARG, NETDEV_PR_ARG, CHAN_DEF_PR_ARG) +); + /************************************************************* * cfg80211 exported functions traces * *************************************************************/ -- cgit v1.2.3 From 683399eddb9fff742b1a14c5a5d03e12bfc0afff Mon Sep 17 00:00:00 2001 From: Mathieu Poirier Date: Sun, 20 Apr 2014 18:57:36 -0600 Subject: netfilter: nfnetlink_acct: Adding quota support to accounting framework nfacct objects already support accounting at the byte and packet level. As such it is a natural extension to add the possiblity to define a ceiling limit for both metrics. All the support for quotas itself is added to nfnetlink acctounting framework to stay coherent with current accounting object management. Quota limit checks are implemented in xt_nfacct filter where statistic collection is already done. Pablo Neira Ayuso has also contributed to this feature. Signed-off-by: Mathieu Poirier Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nfnetlink_acct.h | 8 ++- include/uapi/linux/netfilter/nfnetlink.h | 2 + include/uapi/linux/netfilter/nfnetlink_acct.h | 9 +++ net/netfilter/nfnetlink_acct.c | 85 +++++++++++++++++++++++++++ net/netfilter/xt_nfacct.c | 5 +- 5 files changed, 107 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/netfilter/nfnetlink_acct.h b/include/linux/netfilter/nfnetlink_acct.h index b2e85e59f760..6ec975748742 100644 --- a/include/linux/netfilter/nfnetlink_acct.h +++ b/include/linux/netfilter/nfnetlink_acct.h @@ -3,11 +3,17 @@ #include +enum { + NFACCT_NO_QUOTA = -1, + NFACCT_UNDERQUOTA, + NFACCT_OVERQUOTA, +}; struct nf_acct; struct nf_acct *nfnl_acct_find_get(const char *filter_name); void nfnl_acct_put(struct nf_acct *acct); void nfnl_acct_update(const struct sk_buff *skb, struct nf_acct *nfacct); - +extern int nfnl_acct_overquota(const struct sk_buff *skb, + struct nf_acct *nfacct); #endif /* _NFNL_ACCT_H */ diff --git a/include/uapi/linux/netfilter/nfnetlink.h b/include/uapi/linux/netfilter/nfnetlink.h index 596ddd45253c..354a7e5e50f2 100644 --- a/include/uapi/linux/netfilter/nfnetlink.h +++ b/include/uapi/linux/netfilter/nfnetlink.h @@ -20,6 +20,8 @@ enum nfnetlink_groups { #define NFNLGRP_CONNTRACK_EXP_DESTROY NFNLGRP_CONNTRACK_EXP_DESTROY NFNLGRP_NFTABLES, #define NFNLGRP_NFTABLES NFNLGRP_NFTABLES + NFNLGRP_ACCT_QUOTA, +#define NFNLGRP_ACCT_QUOTA NFNLGRP_ACCT_QUOTA __NFNLGRP_MAX, }; #define NFNLGRP_MAX (__NFNLGRP_MAX - 1) diff --git a/include/uapi/linux/netfilter/nfnetlink_acct.h b/include/uapi/linux/netfilter/nfnetlink_acct.h index c7b6269e760b..51404ec19022 100644 --- a/include/uapi/linux/netfilter/nfnetlink_acct.h +++ b/include/uapi/linux/netfilter/nfnetlink_acct.h @@ -10,15 +10,24 @@ enum nfnl_acct_msg_types { NFNL_MSG_ACCT_GET, NFNL_MSG_ACCT_GET_CTRZERO, NFNL_MSG_ACCT_DEL, + NFNL_MSG_ACCT_OVERQUOTA, NFNL_MSG_ACCT_MAX }; +enum nfnl_acct_flags { + NFACCT_F_QUOTA_PKTS = (1 << 0), + NFACCT_F_QUOTA_BYTES = (1 << 1), + NFACCT_F_OVERQUOTA = (1 << 2), /* can't be set from userspace */ +}; + enum nfnl_acct_type { NFACCT_UNSPEC, NFACCT_NAME, NFACCT_PKTS, NFACCT_BYTES, NFACCT_USE, + NFACCT_FLAGS, + NFACCT_QUOTA, __NFACCT_MAX }; #define NFACCT_MAX (__NFACCT_MAX - 1) diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c index c7b6d466a662..70e86bbb3637 100644 --- a/net/netfilter/nfnetlink_acct.c +++ b/net/netfilter/nfnetlink_acct.c @@ -32,18 +32,24 @@ static LIST_HEAD(nfnl_acct_list); struct nf_acct { atomic64_t pkts; atomic64_t bytes; + unsigned long flags; struct list_head head; atomic_t refcnt; char name[NFACCT_NAME_MAX]; struct rcu_head rcu_head; + char data[0]; }; +#define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS | NFACCT_F_QUOTA_BYTES) + static int nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const tb[]) { struct nf_acct *nfacct, *matching = NULL; char *acct_name; + unsigned int size = 0; + u32 flags = 0; if (!tb[NFACCT_NAME]) return -EINVAL; @@ -68,15 +74,39 @@ nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, /* reset counters if you request a replacement. */ atomic64_set(&matching->pkts, 0); atomic64_set(&matching->bytes, 0); + smp_mb__before_clear_bit(); + /* reset overquota flag if quota is enabled. */ + if ((matching->flags & NFACCT_F_QUOTA)) + clear_bit(NFACCT_F_OVERQUOTA, &matching->flags); return 0; } return -EBUSY; } nfacct = kzalloc(sizeof(struct nf_acct), GFP_KERNEL); + if (tb[NFACCT_FLAGS]) { + flags = ntohl(nla_get_be32(tb[NFACCT_FLAGS])); + if (flags & ~NFACCT_F_QUOTA) + return -EOPNOTSUPP; + if ((flags & NFACCT_F_QUOTA) == NFACCT_F_QUOTA) + return -EINVAL; + if (flags & NFACCT_F_OVERQUOTA) + return -EINVAL; + + size += sizeof(u64); + } + + nfacct = kzalloc(sizeof(struct nf_acct) + size, GFP_KERNEL); if (nfacct == NULL) return -ENOMEM; + if (flags & NFACCT_F_QUOTA) { + u64 *quota = (u64 *)nfacct->data; + + *quota = be64_to_cpu(nla_get_be64(tb[NFACCT_QUOTA])); + nfacct->flags = flags; + } + strncpy(nfacct->name, nla_data(tb[NFACCT_NAME]), NFACCT_NAME_MAX); if (tb[NFACCT_BYTES]) { @@ -117,6 +147,9 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, if (type == NFNL_MSG_ACCT_GET_CTRZERO) { pkts = atomic64_xchg(&acct->pkts, 0); bytes = atomic64_xchg(&acct->bytes, 0); + smp_mb__before_clear_bit(); + if (acct->flags & NFACCT_F_QUOTA) + clear_bit(NFACCT_F_OVERQUOTA, &acct->flags); } else { pkts = atomic64_read(&acct->pkts); bytes = atomic64_read(&acct->bytes); @@ -125,7 +158,13 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes)) || nla_put_be32(skb, NFACCT_USE, htonl(atomic_read(&acct->refcnt)))) goto nla_put_failure; + if (acct->flags & NFACCT_F_QUOTA) { + u64 *quota = (u64 *)acct->data; + if (nla_put_be32(skb, NFACCT_FLAGS, htonl(acct->flags)) || + nla_put_be64(skb, NFACCT_QUOTA, cpu_to_be64(*quota))) + goto nla_put_failure; + } nlmsg_end(skb, nlh); return skb->len; @@ -270,6 +309,8 @@ static const struct nla_policy nfnl_acct_policy[NFACCT_MAX+1] = { [NFACCT_NAME] = { .type = NLA_NUL_STRING, .len = NFACCT_NAME_MAX-1 }, [NFACCT_BYTES] = { .type = NLA_U64 }, [NFACCT_PKTS] = { .type = NLA_U64 }, + [NFACCT_FLAGS] = { .type = NLA_U32 }, + [NFACCT_QUOTA] = { .type = NLA_U64 }, }; static const struct nfnl_callback nfnl_acct_cb[NFNL_MSG_ACCT_MAX] = { @@ -336,6 +377,50 @@ void nfnl_acct_update(const struct sk_buff *skb, struct nf_acct *nfacct) } EXPORT_SYMBOL_GPL(nfnl_acct_update); +static void nfnl_overquota_report(struct nf_acct *nfacct) +{ + int ret; + struct sk_buff *skb; + + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC); + if (skb == NULL) + return; + + ret = nfnl_acct_fill_info(skb, 0, 0, NFNL_MSG_ACCT_OVERQUOTA, 0, + nfacct); + if (ret <= 0) { + kfree_skb(skb); + return; + } + netlink_broadcast(init_net.nfnl, skb, 0, NFNLGRP_ACCT_QUOTA, + GFP_ATOMIC); +} + +int nfnl_acct_overquota(const struct sk_buff *skb, struct nf_acct *nfacct) +{ + u64 now; + u64 *quota; + int ret = NFACCT_UNDERQUOTA; + + /* no place here if we don't have a quota */ + if (!(nfacct->flags & NFACCT_F_QUOTA)) + return NFACCT_NO_QUOTA; + + quota = (u64 *)nfacct->data; + now = (nfacct->flags & NFACCT_F_QUOTA_PKTS) ? + atomic64_read(&nfacct->pkts) : atomic64_read(&nfacct->bytes); + + ret = now > *quota; + + if (now >= *quota && + !test_and_set_bit(NFACCT_F_OVERQUOTA, &nfacct->flags)) { + nfnl_overquota_report(nfacct); + } + + return ret; +} +EXPORT_SYMBOL_GPL(nfnl_acct_overquota); + static int __init nfnl_acct_init(void) { int ret; diff --git a/net/netfilter/xt_nfacct.c b/net/netfilter/xt_nfacct.c index b3be0ef21f19..8c646ed9c921 100644 --- a/net/netfilter/xt_nfacct.c +++ b/net/netfilter/xt_nfacct.c @@ -21,11 +21,14 @@ MODULE_ALIAS("ip6t_nfacct"); static bool nfacct_mt(const struct sk_buff *skb, struct xt_action_param *par) { + int overquota; const struct xt_nfacct_match_info *info = par->targinfo; nfnl_acct_update(skb, info->nfacct); - return true; + overquota = nfnl_acct_overquota(skb, info->nfacct); + + return overquota == NFACCT_UNDERQUOTA ? false : true; } static int -- cgit v1.2.3 From 8757ad65d30f009fe0beeb2d70d3cd834cb998f2 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 11 Apr 2014 10:37:39 -0400 Subject: NVMe: Update copyright headers Make the copyright dates accurate and remove the final paragraph that includes the address of the FSF. Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 4 ---- drivers/block/nvme-scsi.c | 6 +----- include/linux/nvme.h | 6 +----- include/uapi/linux/nvme.h | 6 +----- 4 files changed, 3 insertions(+), 19 deletions(-) (limited to 'include/uapi') diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 7c64fa756cce..eacd64e36be3 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -10,10 +10,6 @@ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. */ #include diff --git a/drivers/block/nvme-scsi.c b/drivers/block/nvme-scsi.c index 2c3f5be06da1..da3b252dea6f 100644 --- a/drivers/block/nvme-scsi.c +++ b/drivers/block/nvme-scsi.c @@ -1,6 +1,6 @@ /* * NVM Express device driver - * Copyright (c) 2011, Intel Corporation. + * Copyright (c) 2011-2014, Intel Corporation. * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU General Public License, @@ -10,10 +10,6 @@ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. */ /* diff --git a/include/linux/nvme.h b/include/linux/nvme.h index a50173ca1d72..cfd084cab22b 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -1,6 +1,6 @@ /* * Definitions for the NVM Express interface - * Copyright (c) 2011-2013, Intel Corporation. + * Copyright (c) 2011-2014, Intel Corporation. * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU General Public License, @@ -10,10 +10,6 @@ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. */ #ifndef _LINUX_NVME_H diff --git a/include/uapi/linux/nvme.h b/include/uapi/linux/nvme.h index 096fe1c6f83d..ad9014e49693 100644 --- a/include/uapi/linux/nvme.h +++ b/include/uapi/linux/nvme.h @@ -1,6 +1,6 @@ /* * Definitions for the NVM Express interface - * Copyright (c) 2011-2013, Intel Corporation. + * Copyright (c) 2011-2014, Intel Corporation. * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU General Public License, @@ -10,10 +10,6 @@ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. */ #ifndef _UAPI_LINUX_NVME_H -- cgit v1.2.3 From a7d2ce2832d84e0182585f63bf96ca7323b3aee7 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 29 Apr 2014 11:41:28 -0600 Subject: NVMe: Configure support for block flush This configures an nvme request_queue as flush capable if the device has a volatile write cache present. Signed-off-by: Keith Busch Signed-off-by: Matthew Wilcox --- drivers/block/nvme-core.c | 3 +++ include/linux/nvme.h | 1 + include/uapi/linux/nvme.h | 1 + 3 files changed, 5 insertions(+) (limited to 'include/uapi') diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index 025dd4cad4a6..e7c4fdb6a651 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -1897,6 +1897,8 @@ static struct nvme_ns *nvme_alloc_ns(struct nvme_dev *dev, unsigned nsid, blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift); if (dev->max_hw_sectors) blk_queue_max_hw_sectors(ns->queue, dev->max_hw_sectors); + if (dev->vwc & NVME_CTRL_VWC_PRESENT) + blk_queue_flush(ns->queue, REQ_FLUSH | REQ_FUA); disk->major = nvme_major; disk->first_minor = 0; @@ -2201,6 +2203,7 @@ static int nvme_dev_add(struct nvme_dev *dev) nn = le32_to_cpup(&ctrl->nn); dev->oncs = le16_to_cpup(&ctrl->oncs); dev->abort_limit = ctrl->acl + 1; + dev->vwc = ctrl->vwc; memcpy(dev->serial, ctrl->sn, sizeof(ctrl->sn)); memcpy(dev->model, ctrl->mn, sizeof(ctrl->mn)); memcpy(dev->firmware_rev, ctrl->fr, sizeof(ctrl->fr)); diff --git a/include/linux/nvme.h b/include/linux/nvme.h index cfd084cab22b..6266373d3147 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -99,6 +99,7 @@ struct nvme_dev { u32 stripe_size; u16 oncs; u16 abort_limit; + u8 vwc; u8 initialized; }; diff --git a/include/uapi/linux/nvme.h b/include/uapi/linux/nvme.h index ad9014e49693..f090336d5bad 100644 --- a/include/uapi/linux/nvme.h +++ b/include/uapi/linux/nvme.h @@ -73,6 +73,7 @@ enum { NVME_CTRL_ONCS_COMPARE = 1 << 0, NVME_CTRL_ONCS_WRITE_UNCORRECTABLE = 1 << 1, NVME_CTRL_ONCS_DSM = 1 << 2, + NVME_CTRL_VWC_PRESENT = 1 << 0, }; struct nvme_lbaf { -- cgit v1.2.3 From 5409e46f1bcf960c651f3fff35f2f25e539655cf Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 7 May 2014 13:49:44 +0200 Subject: nfsd: clean up fh_auth usage Use fh_fsid when reffering to the fsid part of the filehandle. The variable length auth field envisioned in nfsfh wasn't ever implemented. Also clean up some lose ends around this and document the file handle format better. Btw, why do we even export nfsfh.h to userspace? The file handle very much is kernel private, and nothing in nfs-utils include the header either. Signed-off-by: Christoph Hellwig Signed-off-by: J. Bruce Fields --- fs/nfsd/nfsfh.c | 20 +++++++++----------- include/uapi/linux/nfsd/nfsfh.h | 32 +++++++------------------------- 2 files changed, 16 insertions(+), 36 deletions(-) (limited to 'include/uapi') diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 3c37b160dcad..a337106d6875 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -169,8 +169,8 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp) data_left -= len; if (data_left < 0) return error; - exp = rqst_exp_find(rqstp, fh->fh_fsid_type, fh->fh_auth); - fid = (struct fid *)(fh->fh_auth + len); + exp = rqst_exp_find(rqstp, fh->fh_fsid_type, fh->fh_fsid); + fid = (struct fid *)(fh->fh_fsid + len); } else { __u32 tfh[2]; dev_t xdev; @@ -385,7 +385,7 @@ static void _fh_update(struct svc_fh *fhp, struct svc_export *exp, { if (dentry != exp->ex_path.dentry) { struct fid *fid = (struct fid *) - (fhp->fh_handle.fh_auth + fhp->fh_handle.fh_size/4 - 1); + (fhp->fh_handle.fh_fsid + fhp->fh_handle.fh_size/4 - 1); int maxsize = (fhp->fh_maxsize - fhp->fh_handle.fh_size)/4; int subtreecheck = !(exp->ex_flags & NFSEXP_NOSUBTREECHECK); @@ -513,7 +513,6 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, */ struct inode * inode = dentry->d_inode; - __u32 *datap; dev_t ex_dev = exp_sb(exp)->s_dev; dprintk("nfsd: fh_compose(exp %02x:%02x/%ld %pd2, ino=%ld)\n", @@ -557,17 +556,16 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, if (inode) _fh_update_old(dentry, exp, &fhp->fh_handle); } else { - int len; + fhp->fh_handle.fh_size = + key_len(fhp->fh_handle.fh_fsid_type) + 4; fhp->fh_handle.fh_auth_type = 0; - datap = fhp->fh_handle.fh_auth+0; - mk_fsid(fhp->fh_handle.fh_fsid_type, datap, ex_dev, + + mk_fsid(fhp->fh_handle.fh_fsid_type, + fhp->fh_handle.fh_fsid, + ex_dev, exp->ex_path.dentry->d_inode->i_ino, exp->ex_fsid, exp->ex_uuid); - len = key_len(fhp->fh_handle.fh_fsid_type); - datap += len/4; - fhp->fh_handle.fh_size = 4 + len; - if (inode) _fh_update(fhp, exp, dentry); if (fhp->fh_handle.fh_fileid_type == FILEID_INVALID) { diff --git a/include/uapi/linux/nfsd/nfsfh.h b/include/uapi/linux/nfsd/nfsfh.h index 616e3b396476..20391235d088 100644 --- a/include/uapi/linux/nfsd/nfsfh.h +++ b/include/uapi/linux/nfsd/nfsfh.h @@ -1,13 +1,7 @@ /* - * include/linux/nfsd/nfsfh.h - * * This file describes the layout of the file handles as passed * over the wire. * - * Earlier versions of knfsd used to sign file handles using keyed MD5 - * or SHA. I've removed this code, because it doesn't give you more - * security than blocking external access to port 2049 on your firewall. - * * Copyright (C) 1995, 1996, 1997 Olaf Kirch */ @@ -37,7 +31,7 @@ struct nfs_fhbase_old { }; /* - * This is the new flexible, extensible style NFSv2/v3 file handle. + * This is the new flexible, extensible style NFSv2/v3/v4 file handle. * by Neil Brown - March 2000 * * The file handle starts with a sequence of four-byte words. @@ -47,14 +41,7 @@ struct nfs_fhbase_old { * * All four-byte values are in host-byte-order. * - * The auth_type field specifies how the filehandle can be authenticated - * This might allow a file to be confirmed to be in a writable part of a - * filetree without checking the path from it up to the root. - * Current values: - * 0 - No authentication. fb_auth is 0 bytes long - * Possible future values: - * 1 - 4 bytes taken from MD5 hash of the remainer of the file handle - * prefixed by a secret and with the important export flags. + * The auth_type field is deprecated and must be set to 0. * * The fsid_type identifies how the filesystem (or export point) is * encoded. @@ -71,14 +58,9 @@ struct nfs_fhbase_old { * 7 - 8 byte inode number and 16 byte uuid * * The fileid_type identified how the file within the filesystem is encoded. - * This is (will be) passed to, and set by, the underlying filesystem if it supports - * filehandle operations. The filesystem must not use the value '0' or '0xff' and may - * only use the values 1 and 2 as defined below: - * Current values: - * 0 - The root, or export point, of the filesystem. fb_fileid is 0 bytes. - * 1 - 32bit inode number, 32 bit generation number. - * 2 - 32bit inode number, 32 bit generation number, 32 bit parent directory inode number. - * + * The values for this field are filesystem specific, exccept that + * filesystems must not use the values '0' or '0xff'. 'See enum fid_type' + * in include/linux/exportfs.h for currently registered values. */ struct nfs_fhbase_new { __u8 fb_version; /* == 1, even => nfs_fhbase_old */ @@ -114,9 +96,9 @@ struct knfsd_fh { #define fh_fsid_type fh_base.fh_new.fb_fsid_type #define fh_auth_type fh_base.fh_new.fb_auth_type #define fh_fileid_type fh_base.fh_new.fb_fileid_type -#define fh_auth fh_base.fh_new.fb_auth #define fh_fsid fh_base.fh_new.fb_auth - +/* Do not use, provided for userspace compatiblity. */ +#define fh_auth fh_base.fh_new.fb_auth #endif /* _UAPI_LINUX_NFSD_FH_H */ -- cgit v1.2.3 From f2727f7eb9132803d06309839a95de3dad82d237 Mon Sep 17 00:00:00 2001 From: Dimitri John Ledkov Date: Wed, 7 May 2014 20:55:30 +0100 Subject: NVMe: Update namespace and controller identify structures to the 1.1a spec Controller: add CNTLID, AVSCC, APSTA, NVSCC, ACWU, SGLS fields. Namespace: add NMIC, RESCAP, EUI64 fields. EUI64 is specifically interesting, since it can be used to construct an UEFI NVMe device path for a boot entry. As per NVM Express 1.1a spec: http://www.nvmexpress.org/wp-content/uploads/NVM-Express-1_1a.pdf Signed-off-by: Dimitri John Ledkov Signed-off-by: Matthew Wilcox --- include/uapi/linux/nvme.h | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/nvme.h b/include/uapi/linux/nvme.h index f090336d5bad..e74da782d69c 100644 --- a/include/uapi/linux/nvme.h +++ b/include/uapi/linux/nvme.h @@ -45,7 +45,8 @@ struct nvme_id_ctrl { __u8 ieee[3]; __u8 mic; __u8 mdts; - __u8 rsvd78[178]; + __u16 cntlid; + __u8 rsvd80[176]; __le16 oacs; __u8 acl; __u8 aerl; @@ -53,7 +54,9 @@ struct nvme_id_ctrl { __u8 lpa; __u8 elpe; __u8 npss; - __u8 rsvd264[248]; + __u8 avscc; + __u8 apsta; + __u8 rsvd266[246]; __u8 sqes; __u8 cqes; __u8 rsvd514[2]; @@ -64,7 +67,12 @@ struct nvme_id_ctrl { __u8 vwc; __le16 awun; __le16 awupf; - __u8 rsvd530[1518]; + __u8 nvscc; + __u8 rsvd531; + __le16 acwu; + __u8 rsvd534[2]; + __le32 sgls; + __u8 rsvd540[1508]; struct nvme_id_power_state psd[32]; __u8 vs[1024]; }; @@ -92,7 +100,10 @@ struct nvme_id_ns { __u8 mc; __u8 dpc; __u8 dps; - __u8 rsvd30[98]; + __u8 nmic; + __u8 rescap; + __u8 rsvd32[88]; + __u8 eui64[8]; struct nvme_lbaf lbaf[16]; __u8 rsvd192[192]; __u8 vs[3712]; -- cgit v1.2.3 From 8c48b50a1a888ac5511fe856d63f72fb688c6bb4 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 5 May 2014 11:48:40 +0200 Subject: cfg80211: allow restricting supported dfs regions At the moment, the ath9k/ath10k DFS module only supports detecting ETSI radar patterns. Add a bitmap in the interface combinations, indicating which DFS regions are supported by the detector. If unset, support for all regions is assumed. Signed-off-by: Felix Fietkau Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 2 ++ include/uapi/linux/nl80211.h | 3 +++ net/wireless/nl80211.c | 6 ++++-- net/wireless/util.c | 14 ++++++++++++++ 4 files changed, 23 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 5c7169b0ac57..e3a48b0a2b3b 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2638,6 +2638,7 @@ struct ieee80211_iface_limit { * between infrastructure and AP types must match. This is required * only in special cases. * @radar_detect_widths: bitmap of channel widths supported for radar detection + * @radar_detect_regions: bitmap of regions supported for radar detection * * With this structure the driver can describe which interface * combinations it supports concurrently. @@ -2695,6 +2696,7 @@ struct ieee80211_iface_combination { u8 n_limits; bool beacon_int_infra_match; u8 radar_detect_widths; + u8 radar_detect_regions; }; struct ieee80211_txrx_stypes { diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 406010d4def0..b65095a85dee 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -3688,6 +3688,8 @@ enum nl80211_iface_limit_attrs { * different channels may be used within this group. * @NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS: u32 attribute containing the bitmap * of supported channel widths for radar detection. + * @NL80211_IFACE_COMB_RADAR_DETECT_REGIONS: u32 attribute containing the bitmap + * of supported regulatory regions for radar detection. * @NUM_NL80211_IFACE_COMB: number of attributes * @MAX_NL80211_IFACE_COMB: highest attribute number * @@ -3721,6 +3723,7 @@ enum nl80211_if_combination_attrs { NL80211_IFACE_COMB_STA_AP_BI_MATCH, NL80211_IFACE_COMB_NUM_CHANNELS, NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS, + NL80211_IFACE_COMB_RADAR_DETECT_REGIONS, /* keep last */ NUM_NL80211_IFACE_COMB, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 0f1b18f209d6..c0833830cfe7 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -970,8 +970,10 @@ static int nl80211_put_iface_combinations(struct wiphy *wiphy, c->max_interfaces)) goto nla_put_failure; if (large && - nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS, - c->radar_detect_widths)) + (nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS, + c->radar_detect_widths) || + nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_REGIONS, + c->radar_detect_regions))) goto nla_put_failure; nla_nest_end(msg, nl_combi); diff --git a/net/wireless/util.c b/net/wireless/util.c index a756429b3a0a..8c61d5c6fad3 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -1274,10 +1274,20 @@ int cfg80211_iter_combinations(struct wiphy *wiphy, void *data), void *data) { + const struct ieee80211_regdomain *regdom; + enum nl80211_dfs_regions region = 0; int i, j, iftype; int num_interfaces = 0; u32 used_iftypes = 0; + if (radar_detect) { + rcu_read_lock(); + regdom = rcu_dereference(cfg80211_regdomain); + if (regdom) + region = regdom->dfs_region; + rcu_read_unlock(); + } + for (iftype = 0; iftype < NUM_NL80211_IFTYPES; iftype++) { num_interfaces += iftype_num[iftype]; if (iftype_num[iftype] > 0 && @@ -1318,6 +1328,10 @@ int cfg80211_iter_combinations(struct wiphy *wiphy, if (radar_detect != (c->radar_detect_widths & radar_detect)) goto cont; + if (radar_detect && c->radar_detect_regions && + !(c->radar_detect_regions & BIT(region))) + goto cont; + /* Finally check that all iftypes that we're currently * using are actually part of this combination. If they * aren't then we can't use this combination and have -- cgit v1.2.3 From 34d22ce22b0b249804816990a3b62b08b1a62546 Mon Sep 17 00:00:00 2001 From: Andrei Otcheretianski Date: Fri, 9 May 2014 14:11:44 +0300 Subject: cfg80211: Add API to update CSA counters in mgmt frames Add NL80211_ATTR_CSA_C_OFFSETS_TX which holds an array of offsets to the CSA counters which should be updated when sending a management frames with NL80211_CMD_FRAME. This API should be used by the drivers that wish to keep the CSA counter updated in probe responses, but do not implement probe response offloading and so, do not use ieee80211_proberesp_get function. Signed-off-by: Andrei Otcheretianski Signed-off-by: Luciano Coelho Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 4 ++++ include/uapi/linux/nl80211.h | 8 ++++++++ net/wireless/nl80211.c | 24 ++++++++++++++++++++++-- 3 files changed, 34 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index e3a48b0a2b3b..f46e1e15746d 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1986,6 +1986,8 @@ struct cfg80211_update_ft_ies_params { * @len: buffer length * @no_cck: don't use cck rates for this frame * @dont_wait_for_ack: tells the low level not to wait for an ack + * @n_csa_offsets: length of csa_offsets array + * @csa_offsets: array of all the csa offsets in the frame */ struct cfg80211_mgmt_tx_params { struct ieee80211_channel *chan; @@ -1995,6 +1997,8 @@ struct cfg80211_mgmt_tx_params { size_t len; bool no_cck; bool dont_wait_for_ack; + int n_csa_offsets; + const u16 *csa_offsets; }; /** diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index b65095a85dee..ec90fc9d2358 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -503,6 +503,9 @@ * TX status event pertaining to the TX request. * %NL80211_ATTR_TX_NO_CCK_RATE is used to decide whether to send the * management frames at CCK rate or not in 2GHz band. + * %NL80211_ATTR_CSA_C_OFFSETS_TX is an array of offsets to CSA + * counters which will be updated to the current value. This attribute + * is used during CSA period. * @NL80211_CMD_FRAME_WAIT_CANCEL: When an off-channel TX was requested, this * command may be used with the corresponding cookie to cancel the wait * time if it is known that it is no longer necessary. @@ -1576,6 +1579,9 @@ enum nl80211_commands { * advertise values that cannot always be met. In such cases, an attempt * to add a new station entry with @NL80211_CMD_NEW_STATION may fail. * + * @NL80211_ATTR_CSA_C_OFFSETS_TX: An array of csa counter offsets (u16) which + * should be updated when the frame is transmitted. + * * @NL80211_ATTR_TDLS_PEER_CAPABILITY: flags for TDLS peer capabilities, u32. * As specified in the &enum nl80211_tdls_peer_capability. * @@ -1920,6 +1926,8 @@ enum nl80211_attrs { NL80211_ATTR_IFACE_SOCKET_OWNER, + NL80211_ATTR_CSA_C_OFFSETS_TX, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 74e7299e4add..4c0ca40ef90e 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -386,6 +386,7 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_WIPHY_FREQ_HINT] = { .type = NLA_U32 }, [NL80211_ATTR_TDLS_PEER_CAPABILITY] = { .type = NLA_U32 }, [NL80211_ATTR_IFACE_SOCKET_OWNER] = { .type = NLA_FLAG }, + [NL80211_ATTR_CSA_C_OFFSETS_TX] = { .type = NLA_BINARY }, }; /* policy for the key attributes */ @@ -7786,6 +7787,27 @@ static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info) if (!chandef.chan && params.offchan) return -EINVAL; + params.buf = nla_data(info->attrs[NL80211_ATTR_FRAME]); + params.len = nla_len(info->attrs[NL80211_ATTR_FRAME]); + + if (info->attrs[NL80211_ATTR_CSA_C_OFFSETS_TX]) { + int len = nla_len(info->attrs[NL80211_ATTR_CSA_C_OFFSETS_TX]); + int i; + + if (len % sizeof(u16)) + return -EINVAL; + + params.n_csa_offsets = len / sizeof(u16); + params.csa_offsets = + nla_data(info->attrs[NL80211_ATTR_CSA_C_OFFSETS_TX]); + + /* check that all the offsets fit the frame */ + for (i = 0; i < params.n_csa_offsets; i++) { + if (params.csa_offsets[i] >= params.len) + return -EINVAL; + } + } + if (!params.dont_wait_for_ack) { msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!msg) @@ -7799,8 +7821,6 @@ static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info) } } - params.buf = nla_data(info->attrs[NL80211_ATTR_FRAME]); - params.len = nla_len(info->attrs[NL80211_ATTR_FRAME]); params.chan = chandef.chan; err = cfg80211_mlme_mgmt_tx(rdev, wdev, ¶ms, &cookie); if (err) -- cgit v1.2.3 From 9a774c78e2114c7e8605e3a168ccd552cbe3d922 Mon Sep 17 00:00:00 2001 From: Andrei Otcheretianski Date: Fri, 9 May 2014 14:11:46 +0300 Subject: cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski Signed-off-by: Luciano Coelho Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 18 ++++++++++-- include/uapi/linux/nl80211.h | 11 +++++--- net/mac80211/cfg.c | 10 +++++-- net/wireless/core.c | 2 ++ net/wireless/nl80211.c | 65 ++++++++++++++++++++++++++++++++++---------- net/wireless/trace.h | 22 +++++++++------ 6 files changed, 96 insertions(+), 32 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index f46e1e15746d..447cb58f0d77 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -694,8 +694,10 @@ struct cfg80211_ap_settings { * * @chandef: defines the channel to use after the switch * @beacon_csa: beacon data while performing the switch - * @counter_offset_beacon: offset for the counter within the beacon (tail) - * @counter_offset_presp: offset for the counter within the probe response + * @counter_offsets_beacon: offsets of the counters within the beacon (tail) + * @counter_offsets_presp: offsets of the counters within the probe response + * @n_counter_offsets_beacon: number of csa counters the beacon (tail) + * @n_counter_offsets_presp: number of csa counters in the probe response * @beacon_after: beacon data to be used on the new channel * @radar_required: whether radar detection is required on the new channel * @block_tx: whether transmissions should be blocked while changing @@ -704,7 +706,10 @@ struct cfg80211_ap_settings { struct cfg80211_csa_settings { struct cfg80211_chan_def chandef; struct cfg80211_beacon_data beacon_csa; - u16 counter_offset_beacon, counter_offset_presp; + const u16 *counter_offsets_beacon; + const u16 *counter_offsets_presp; + unsigned int n_counter_offsets_beacon; + unsigned int n_counter_offsets_presp; struct cfg80211_beacon_data beacon_after; bool radar_required; bool block_tx; @@ -3048,6 +3053,13 @@ struct wiphy { u16 max_ap_assoc_sta; + /* + * Number of supported csa_counters in beacons and probe responses. + * This value should be set if the driver wishes to limit the number of + * csa counters. Default (0) means infinite. + */ + u8 max_num_csa_counters; + char priv[0] __aligned(NETDEV_ALIGN); }; diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index ec90fc9d2358..0cfa827123ff 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1528,10 +1528,10 @@ enum nl80211_commands { * operation). * @NL80211_ATTR_CSA_IES: Nested set of attributes containing the IE information * for the time while performing a channel switch. - * @NL80211_ATTR_CSA_C_OFF_BEACON: Offset of the channel switch counter - * field in the beacons tail (%NL80211_ATTR_BEACON_TAIL). - * @NL80211_ATTR_CSA_C_OFF_PRESP: Offset of the channel switch counter - * field in the probe response (%NL80211_ATTR_PROBE_RESP). + * @NL80211_ATTR_CSA_C_OFF_BEACON: An array of offsets (u16) to the channel + * switch counters in the beacons tail (%NL80211_ATTR_BEACON_TAIL). + * @NL80211_ATTR_CSA_C_OFF_PRESP: An array of offsets (u16) to the channel + * switch counters in the probe response (%NL80211_ATTR_PROBE_RESP). * * @NL80211_ATTR_RXMGMT_FLAGS: flags for nl80211_send_mgmt(), u32. * As specified in the &enum nl80211_rxmgmt_flags. @@ -1581,6 +1581,8 @@ enum nl80211_commands { * * @NL80211_ATTR_CSA_C_OFFSETS_TX: An array of csa counter offsets (u16) which * should be updated when the frame is transmitted. + * @NL80211_ATTR_MAX_CSA_COUNTERS: U8 attribute used to advertise the maximum + * supported number of csa counters. * * @NL80211_ATTR_TDLS_PEER_CAPABILITY: flags for TDLS peer capabilities, u32. * As specified in the &enum nl80211_tdls_peer_capability. @@ -1927,6 +1929,7 @@ enum nl80211_attrs { NL80211_ATTR_IFACE_SOCKET_OWNER, NL80211_ATTR_CSA_C_OFFSETS_TX, + NL80211_ATTR_MAX_CSA_COUNTERS, /* add attributes here, update the policy in nl80211.c */ diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 8e4754ebc2d8..7a6f8aba5c46 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -3192,8 +3192,14 @@ static int ieee80211_set_csa_beacon(struct ieee80211_sub_if_data *sdata, break; sdata->csa_counter_offset_beacon = - params->counter_offset_beacon; - sdata->csa_counter_offset_presp = params->counter_offset_presp; + params->counter_offsets_beacon[0]; + + if (params->n_counter_offsets_presp) + sdata->csa_counter_offset_presp = + params->counter_offsets_presp[0]; + else + sdata->csa_counter_offset_presp = 0; + err = ieee80211_assign_beacon(sdata, ¶ms->beacon_csa); if (err < 0) { kfree(sdata->u.ap.next_beacon); diff --git a/net/wireless/core.c b/net/wireless/core.c index 39788711ce6e..d03d8bdb29ca 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -402,6 +402,8 @@ struct wiphy *wiphy_new(const struct cfg80211_ops *ops, int sizeof_priv) rdev->wiphy.rts_threshold = (u32) -1; rdev->wiphy.coverage_class = 0; + rdev->wiphy.max_num_csa_counters = 1; + return &rdev->wiphy; } EXPORT_SYMBOL(wiphy_new); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 4c0ca40ef90e..ca19b1520389 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -371,8 +371,8 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_CH_SWITCH_COUNT] = { .type = NLA_U32 }, [NL80211_ATTR_CH_SWITCH_BLOCK_TX] = { .type = NLA_FLAG }, [NL80211_ATTR_CSA_IES] = { .type = NLA_NESTED }, - [NL80211_ATTR_CSA_C_OFF_BEACON] = { .type = NLA_U16 }, - [NL80211_ATTR_CSA_C_OFF_PRESP] = { .type = NLA_U16 }, + [NL80211_ATTR_CSA_C_OFF_BEACON] = { .type = NLA_BINARY }, + [NL80211_ATTR_CSA_C_OFF_PRESP] = { .type = NLA_BINARY }, [NL80211_ATTR_STA_SUPPORTED_CHANNELS] = { .type = NLA_BINARY }, [NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES] = { .type = NLA_BINARY }, [NL80211_ATTR_HANDLE_DFS] = { .type = NLA_FLAG }, @@ -1670,6 +1670,13 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev, } nla_nest_end(msg, nested); } + state->split_start++; + break; + case 12: + if (rdev->wiphy.flags & WIPHY_FLAG_HAS_CHANNEL_SWITCH && + nla_put_u8(msg, NL80211_ATTR_MAX_CSA_COUNTERS, + rdev->wiphy.max_num_csa_counters)) + goto nla_put_failure; /* done */ state->split_start = 0; @@ -5864,6 +5871,7 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) u8 radar_detect_width = 0; int err; bool need_new_beacon = false; + int len, i; if (!rdev->ops->channel_switch || !(rdev->wiphy.flags & WIPHY_FLAG_HAS_CHANNEL_SWITCH)) @@ -5922,26 +5930,55 @@ static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info) if (!csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON]) return -EINVAL; - params.counter_offset_beacon = - nla_get_u16(csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON]); - if (params.counter_offset_beacon >= params.beacon_csa.tail_len) + len = nla_len(csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON]); + if (!len || (len % sizeof(u16))) return -EINVAL; - /* sanity check - counters should be the same */ - if (params.beacon_csa.tail[params.counter_offset_beacon] != - params.count) + params.n_counter_offsets_beacon = len / sizeof(u16); + if (rdev->wiphy.max_num_csa_counters && + (params.n_counter_offsets_beacon > + rdev->wiphy.max_num_csa_counters)) return -EINVAL; + params.counter_offsets_beacon = + nla_data(csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON]); + + /* sanity checks - counters should fit and be the same */ + for (i = 0; i < params.n_counter_offsets_beacon; i++) { + u16 offset = params.counter_offsets_beacon[i]; + + if (offset >= params.beacon_csa.tail_len) + return -EINVAL; + + if (params.beacon_csa.tail[offset] != params.count) + return -EINVAL; + } + if (csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]) { - params.counter_offset_presp = - nla_get_u16(csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]); - if (params.counter_offset_presp >= - params.beacon_csa.probe_resp_len) + len = nla_len(csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]); + if (!len || (len % sizeof(u16))) return -EINVAL; - if (params.beacon_csa.probe_resp[params.counter_offset_presp] != - params.count) + params.n_counter_offsets_presp = len / sizeof(u16); + if (rdev->wiphy.max_num_csa_counters && + (params.n_counter_offsets_beacon > + rdev->wiphy.max_num_csa_counters)) return -EINVAL; + + params.counter_offsets_presp = + nla_data(csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]); + + /* sanity checks - counters should fit and be the same */ + for (i = 0; i < params.n_counter_offsets_presp; i++) { + u16 offset = params.counter_offsets_presp[i]; + + if (offset >= params.beacon_csa.probe_resp_len) + return -EINVAL; + + if (params.beacon_csa.probe_resp[offset] != + params.count) + return -EINVAL; + } } skip_beacons: diff --git a/net/wireless/trace.h b/net/wireless/trace.h index cdfbb00e1b37..560ed77084e9 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -1876,29 +1876,33 @@ TRACE_EVENT(rdev_channel_switch, WIPHY_ENTRY NETDEV_ENTRY CHAN_DEF_ENTRY - __field(u16, counter_offset_beacon) - __field(u16, counter_offset_presp) __field(bool, radar_required) __field(bool, block_tx) __field(u8, count) + __dynamic_array(u16, bcn_ofs, params->n_counter_offsets_beacon) + __dynamic_array(u16, pres_ofs, params->n_counter_offsets_presp) ), TP_fast_assign( WIPHY_ASSIGN; NETDEV_ASSIGN; CHAN_DEF_ASSIGN(¶ms->chandef); - __entry->counter_offset_beacon = params->counter_offset_beacon; - __entry->counter_offset_presp = params->counter_offset_presp; __entry->radar_required = params->radar_required; __entry->block_tx = params->block_tx; __entry->count = params->count; + memcpy(__get_dynamic_array(bcn_ofs), + params->counter_offsets_beacon, + params->n_counter_offsets_beacon * sizeof(u16)); + + /* probe response offsets are optional */ + if (params->n_counter_offsets_presp) + memcpy(__get_dynamic_array(pres_ofs), + params->counter_offsets_presp, + params->n_counter_offsets_presp * sizeof(u16)); ), TP_printk(WIPHY_PR_FMT ", " NETDEV_PR_FMT ", " CHAN_DEF_PR_FMT - ", block_tx: %d, count: %u, radar_required: %d" - ", counter offsets (beacon/presp): %u/%u", + ", block_tx: %d, count: %u, radar_required: %d", WIPHY_PR_ARG, NETDEV_PR_ARG, CHAN_DEF_PR_ARG, - __entry->block_tx, __entry->count, __entry->radar_required, - __entry->counter_offset_beacon, - __entry->counter_offset_presp) + __entry->block_tx, __entry->count, __entry->radar_required) ); TRACE_EVENT(rdev_set_qos_map, -- cgit v1.2.3 From 5cc9ed4b9a7ac579362ccebac67f7a4cdb36de06 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 16 May 2014 14:22:37 +0100 Subject: drm/i915: Introduce mapping of user pages into video memory (userptr) ioctl By exporting the ability to map user address and inserting PTEs representing their backing pages into the GTT, we can exploit UMA in order to utilize normal application data as a texture source or even as a render target (depending upon the capabilities of the chipset). This has a number of uses, with zero-copy downloads to the GPU and efficient readback making the intermixed streaming of CPU and GPU operations fairly efficient. This ability has many widespread implications from faster rendering of client-side software rasterisers (chromium), mitigation of stalls due to read back (firefox) and to faster pipelining of texture data (such as pixel buffer objects in GL or data blobs in CL). v2: Compile with CONFIG_MMU_NOTIFIER v3: We can sleep while performing invalidate-range, which we can utilise to drop our page references prior to the kernel manipulating the vma (for either discard or cloning) and so protect normal users. v4: Only run the invalidate notifier if the range intercepts the bo. v5: Prevent userspace from attempting to GTT mmap non-page aligned buffers v6: Recheck after reacquire mutex for lost mmu. v7: Fix implicit padding of ioctl struct by rounding to next 64bit boundary. v8: Fix rebasing error after forwarding porting the back port. v9: Limit the userptr to page aligned entries. We now expect userspace to handle all the offset-in-page adjustments itself. v10: Prevent vma from being copied across fork to avoid issues with cow. v11: Drop vma behaviour changes -- locking is nigh on impossible. Use a worker to load user pages to avoid lock inversions. v12: Use get_task_mm()/mmput() for correct refcounting of mm. v13: Use a worker to release the mmu_notifier to avoid lock inversion v14: Decouple mmu_notifier from struct_mutex using a custom mmu_notifer with its own locking and tree of objects for each mm/mmu_notifier. v15: Prevent overlapping userptr objects, and invalidate all objects within the mmu_notifier range v16: Fix a typo for iterating over multiple objects in the range and rearrange error path to destroy the mmu_notifier locklessly. Also close a race between invalidate_range and the get_pages_worker. v17: Close a race between get_pages_worker/invalidate_range and fresh allocations of the same userptr range - and notice that struct_mutex was presumed to be held when during creation it wasn't. v18: Sigh. Fix the refactor of st_set_pages() to allocate enough memory for the struct sg_table and to clear it before reporting an error. v19: Always error out on read-only userptr requests as we don't have the hardware infrastructure to support them at the moment. v20: Refuse to implement read-only support until we have the required infrastructure - but reserve the bit in flags for future use. v21: use_mm() is not required for get_user_pages(). It is only meant to be used to fix up the kernel thread's current->mm for use with copy_user(). v22: Use sg_alloc_table_from_pages for that chunky feeling v23: Export a function for sanity checking dma-buf rather than encode userptr details elsewhere, and clean up comments based on suggestions by Bradley. Signed-off-by: Chris Wilson Cc: Tvrtko Ursulin Cc: "Gong, Zhipeng" Cc: Akash Goel Cc: "Volkin, Bradley D" Reviewed-by: Tvrtko Ursulin Reviewed-by: Brad Volkin [danvet: Frob ioctl allocation to pick the next one - will cause a bit of fuss with create2 apparently, but such are the rules.] [danvet2: oops, forgot to git add after manual patch application] [danvet3: Appease sparse.] Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/Kconfig | 1 + drivers/gpu/drm/i915/Makefile | 1 + drivers/gpu/drm/i915/i915_dma.c | 1 + drivers/gpu/drm/i915/i915_drv.h | 25 +- drivers/gpu/drm/i915/i915_gem.c | 4 + drivers/gpu/drm/i915/i915_gem_dmabuf.c | 8 + drivers/gpu/drm/i915/i915_gem_userptr.c | 711 ++++++++++++++++++++++++++++++++ drivers/gpu/drm/i915/i915_gpu_error.c | 2 + include/uapi/drm/i915_drm.h | 16 + 9 files changed, 768 insertions(+), 1 deletion(-) create mode 100644 drivers/gpu/drm/i915/i915_gem_userptr.c (limited to 'include/uapi') diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig index e4e3c01b8cbc..437e1824d0bf 100644 --- a/drivers/gpu/drm/i915/Kconfig +++ b/drivers/gpu/drm/i915/Kconfig @@ -5,6 +5,7 @@ config DRM_I915 depends on (AGP || AGP=n) select INTEL_GTT select AGP_INTEL if AGP + select INTERVAL_TREE # we need shmfs for the swappable backing store, and in particular # the shmem_readpage() which depends upon tmpfs select SHMEM diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 24469168d550..7b2f3bee3518 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -27,6 +27,7 @@ i915-y += i915_cmd_parser.o \ i915_gem.o \ i915_gem_stolen.o \ i915_gem_tiling.o \ + i915_gem_userptr.o \ i915_gpu_error.o \ i915_irq.o \ i915_trace_points.o \ diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 46f1decab76f..dea455bd889a 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -1983,6 +1983,7 @@ const struct drm_ioctl_desc i915_ioctls[] = { DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(I915_GET_RESET_STATS, i915_get_reset_stats_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(I915_GEM_USERPTR, i915_gem_userptr_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW), }; int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 9e9e8166a1f3..a270e0773e2d 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -41,6 +41,7 @@ #include #include #include +#include #include #include #include @@ -178,6 +179,7 @@ enum hpd_pin { if ((intel_connector)->base.encoder == (__encoder)) struct drm_i915_private; +struct i915_mmu_object; enum intel_dpll_id { DPLL_ID_PRIVATE = -1, /* non-shared dpll in use */ @@ -403,6 +405,7 @@ struct drm_i915_error_state { u32 tiling:2; u32 dirty:1; u32 purgeable:1; + u32 userptr:1; s32 ring:4; u32 cache_level:3; } **active_bo, **pinned_bo; @@ -1447,6 +1450,9 @@ struct drm_i915_private { struct i915_gtt gtt; /* VM representing the global address space */ struct i915_gem_mm mm; +#if defined(CONFIG_MMU_NOTIFIER) + DECLARE_HASHTABLE(mmu_notifiers, 7); +#endif /* Kernel Modesetting */ @@ -1580,6 +1586,8 @@ struct drm_i915_gem_object_ops { */ int (*get_pages)(struct drm_i915_gem_object *); void (*put_pages)(struct drm_i915_gem_object *); + int (*dmabuf_export)(struct drm_i915_gem_object *); + void (*release)(struct drm_i915_gem_object *); }; struct drm_i915_gem_object { @@ -1693,8 +1701,20 @@ struct drm_i915_gem_object { /** for phy allocated objects */ struct drm_i915_gem_phys_object *phys_obj; -}; + union { + struct i915_gem_userptr { + uintptr_t ptr; + unsigned read_only :1; + unsigned workers :4; +#define I915_GEM_USERPTR_MAX_WORKERS 15 + + struct mm_struct *mm; + struct i915_mmu_object *mn; + struct work_struct *work; + } userptr; + }; +}; #define to_intel_bo(x) container_of(x, struct drm_i915_gem_object, base) /** @@ -2119,6 +2139,9 @@ int i915_gem_set_tiling(struct drm_device *dev, void *data, struct drm_file *file_priv); int i915_gem_get_tiling(struct drm_device *dev, void *data, struct drm_file *file_priv); +int i915_gem_init_userptr(struct drm_device *dev); +int i915_gem_userptr_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); int i915_gem_wait_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index f6c03513de52..d5aad0bc71f7 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -4263,6 +4263,9 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj) if (obj->base.import_attach) drm_prime_gem_destroy(&obj->base, NULL); + if (obj->ops->release) + obj->ops->release(obj); + drm_gem_object_release(&obj->base); i915_gem_info_remove_obj(dev_priv, obj->base.size); @@ -4542,6 +4545,7 @@ int i915_gem_init(struct drm_device *dev) DRM_DEBUG_DRIVER("allow wake ack timed out\n"); } + i915_gem_init_userptr(dev); i915_gem_init_global_gtt(dev); ret = i915_gem_context_init(dev); diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c index 321102a8374b..580aa42443ed 100644 --- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c @@ -229,6 +229,14 @@ static const struct dma_buf_ops i915_dmabuf_ops = { struct dma_buf *i915_gem_prime_export(struct drm_device *dev, struct drm_gem_object *gem_obj, int flags) { + struct drm_i915_gem_object *obj = to_intel_bo(gem_obj); + + if (obj->ops->dmabuf_export) { + int ret = obj->ops->dmabuf_export(obj); + if (ret) + return ERR_PTR(ret); + } + return dma_buf_export(gem_obj, &i915_dmabuf_ops, gem_obj->size, flags); } diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c new file mode 100644 index 000000000000..21ea92886a56 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -0,0 +1,711 @@ +/* + * Copyright © 2012-2014 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "drmP.h" +#include "i915_drm.h" +#include "i915_drv.h" +#include "i915_trace.h" +#include "intel_drv.h" +#include +#include +#include +#include + +#if defined(CONFIG_MMU_NOTIFIER) +#include + +struct i915_mmu_notifier { + spinlock_t lock; + struct hlist_node node; + struct mmu_notifier mn; + struct rb_root objects; + struct drm_device *dev; + struct mm_struct *mm; + struct work_struct work; + unsigned long count; + unsigned long serial; +}; + +struct i915_mmu_object { + struct i915_mmu_notifier *mmu; + struct interval_tree_node it; + struct drm_i915_gem_object *obj; +}; + +static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn, + struct mm_struct *mm, + unsigned long start, + unsigned long end) +{ + struct i915_mmu_notifier *mn = container_of(_mn, struct i915_mmu_notifier, mn); + struct interval_tree_node *it = NULL; + unsigned long serial = 0; + + end--; /* interval ranges are inclusive, but invalidate range is exclusive */ + while (start < end) { + struct drm_i915_gem_object *obj; + + obj = NULL; + spin_lock(&mn->lock); + if (serial == mn->serial) + it = interval_tree_iter_next(it, start, end); + else + it = interval_tree_iter_first(&mn->objects, start, end); + if (it != NULL) { + obj = container_of(it, struct i915_mmu_object, it)->obj; + drm_gem_object_reference(&obj->base); + serial = mn->serial; + } + spin_unlock(&mn->lock); + if (obj == NULL) + return; + + mutex_lock(&mn->dev->struct_mutex); + /* Cancel any active worker and force us to re-evaluate gup */ + obj->userptr.work = NULL; + + if (obj->pages != NULL) { + struct drm_i915_private *dev_priv = to_i915(mn->dev); + struct i915_vma *vma, *tmp; + bool was_interruptible; + + was_interruptible = dev_priv->mm.interruptible; + dev_priv->mm.interruptible = false; + + list_for_each_entry_safe(vma, tmp, &obj->vma_list, vma_link) { + int ret = i915_vma_unbind(vma); + WARN_ON(ret && ret != -EIO); + } + WARN_ON(i915_gem_object_put_pages(obj)); + + dev_priv->mm.interruptible = was_interruptible; + } + + start = obj->userptr.ptr + obj->base.size; + + drm_gem_object_unreference(&obj->base); + mutex_unlock(&mn->dev->struct_mutex); + } +} + +static const struct mmu_notifier_ops i915_gem_userptr_notifier = { + .invalidate_range_start = i915_gem_userptr_mn_invalidate_range_start, +}; + +static struct i915_mmu_notifier * +__i915_mmu_notifier_lookup(struct drm_device *dev, struct mm_struct *mm) +{ + struct drm_i915_private *dev_priv = to_i915(dev); + struct i915_mmu_notifier *mmu; + + /* Protected by dev->struct_mutex */ + hash_for_each_possible(dev_priv->mmu_notifiers, mmu, node, (unsigned long)mm) + if (mmu->mm == mm) + return mmu; + + return NULL; +} + +static struct i915_mmu_notifier * +i915_mmu_notifier_get(struct drm_device *dev, struct mm_struct *mm) +{ + struct drm_i915_private *dev_priv = to_i915(dev); + struct i915_mmu_notifier *mmu; + int ret; + + lockdep_assert_held(&dev->struct_mutex); + + mmu = __i915_mmu_notifier_lookup(dev, mm); + if (mmu) + return mmu; + + mmu = kmalloc(sizeof(*mmu), GFP_KERNEL); + if (mmu == NULL) + return ERR_PTR(-ENOMEM); + + spin_lock_init(&mmu->lock); + mmu->dev = dev; + mmu->mn.ops = &i915_gem_userptr_notifier; + mmu->mm = mm; + mmu->objects = RB_ROOT; + mmu->count = 0; + mmu->serial = 0; + + /* Protected by mmap_sem (write-lock) */ + ret = __mmu_notifier_register(&mmu->mn, mm); + if (ret) { + kfree(mmu); + return ERR_PTR(ret); + } + + /* Protected by dev->struct_mutex */ + hash_add(dev_priv->mmu_notifiers, &mmu->node, (unsigned long)mm); + return mmu; +} + +static void +__i915_mmu_notifier_destroy_worker(struct work_struct *work) +{ + struct i915_mmu_notifier *mmu = container_of(work, typeof(*mmu), work); + mmu_notifier_unregister(&mmu->mn, mmu->mm); + kfree(mmu); +} + +static void +__i915_mmu_notifier_destroy(struct i915_mmu_notifier *mmu) +{ + lockdep_assert_held(&mmu->dev->struct_mutex); + + /* Protected by dev->struct_mutex */ + hash_del(&mmu->node); + + /* Our lock ordering is: mmap_sem, mmu_notifier_scru, struct_mutex. + * We enter the function holding struct_mutex, therefore we need + * to drop our mutex prior to calling mmu_notifier_unregister in + * order to prevent lock inversion (and system-wide deadlock) + * between the mmap_sem and struct-mutex. Hence we defer the + * unregistration to a workqueue where we hold no locks. + */ + INIT_WORK(&mmu->work, __i915_mmu_notifier_destroy_worker); + schedule_work(&mmu->work); +} + +static void __i915_mmu_notifier_update_serial(struct i915_mmu_notifier *mmu) +{ + if (++mmu->serial == 0) + mmu->serial = 1; +} + +static void +i915_mmu_notifier_del(struct i915_mmu_notifier *mmu, + struct i915_mmu_object *mn) +{ + lockdep_assert_held(&mmu->dev->struct_mutex); + + spin_lock(&mmu->lock); + interval_tree_remove(&mn->it, &mmu->objects); + __i915_mmu_notifier_update_serial(mmu); + spin_unlock(&mmu->lock); + + /* Protected against _add() by dev->struct_mutex */ + if (--mmu->count == 0) + __i915_mmu_notifier_destroy(mmu); +} + +static int +i915_mmu_notifier_add(struct i915_mmu_notifier *mmu, + struct i915_mmu_object *mn) +{ + struct interval_tree_node *it; + int ret; + + ret = i915_mutex_lock_interruptible(mmu->dev); + if (ret) + return ret; + + /* Make sure we drop the final active reference (and thereby + * remove the objects from the interval tree) before we do + * the check for overlapping objects. + */ + i915_gem_retire_requests(mmu->dev); + + /* Disallow overlapping userptr objects */ + spin_lock(&mmu->lock); + it = interval_tree_iter_first(&mmu->objects, + mn->it.start, mn->it.last); + if (it) { + struct drm_i915_gem_object *obj; + + /* We only need to check the first object in the range as it + * either has cancelled gup work queued and we need to + * return back to the user to give time for the gup-workers + * to flush their object references upon which the object will + * be removed from the interval-tree, or the the range is + * still in use by another client and the overlap is invalid. + */ + + obj = container_of(it, struct i915_mmu_object, it)->obj; + ret = obj->userptr.workers ? -EAGAIN : -EINVAL; + } else { + interval_tree_insert(&mn->it, &mmu->objects); + __i915_mmu_notifier_update_serial(mmu); + ret = 0; + } + spin_unlock(&mmu->lock); + mutex_unlock(&mmu->dev->struct_mutex); + + return ret; +} + +static void +i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj) +{ + struct i915_mmu_object *mn; + + mn = obj->userptr.mn; + if (mn == NULL) + return; + + i915_mmu_notifier_del(mn->mmu, mn); + obj->userptr.mn = NULL; +} + +static int +i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj, + unsigned flags) +{ + struct i915_mmu_notifier *mmu; + struct i915_mmu_object *mn; + int ret; + + if (flags & I915_USERPTR_UNSYNCHRONIZED) + return capable(CAP_SYS_ADMIN) ? 0 : -EPERM; + + down_write(&obj->userptr.mm->mmap_sem); + ret = i915_mutex_lock_interruptible(obj->base.dev); + if (ret == 0) { + mmu = i915_mmu_notifier_get(obj->base.dev, obj->userptr.mm); + if (!IS_ERR(mmu)) + mmu->count++; /* preemptive add to act as a refcount */ + else + ret = PTR_ERR(mmu); + mutex_unlock(&obj->base.dev->struct_mutex); + } + up_write(&obj->userptr.mm->mmap_sem); + if (ret) + return ret; + + mn = kzalloc(sizeof(*mn), GFP_KERNEL); + if (mn == NULL) { + ret = -ENOMEM; + goto destroy_mmu; + } + + mn->mmu = mmu; + mn->it.start = obj->userptr.ptr; + mn->it.last = mn->it.start + obj->base.size - 1; + mn->obj = obj; + + ret = i915_mmu_notifier_add(mmu, mn); + if (ret) + goto free_mn; + + obj->userptr.mn = mn; + return 0; + +free_mn: + kfree(mn); +destroy_mmu: + mutex_lock(&obj->base.dev->struct_mutex); + if (--mmu->count == 0) + __i915_mmu_notifier_destroy(mmu); + mutex_unlock(&obj->base.dev->struct_mutex); + return ret; +} + +#else + +static void +i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj) +{ +} + +static int +i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj, + unsigned flags) +{ + if ((flags & I915_USERPTR_UNSYNCHRONIZED) == 0) + return -ENODEV; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + return 0; +} +#endif + +struct get_pages_work { + struct work_struct work; + struct drm_i915_gem_object *obj; + struct task_struct *task; +}; + + +#if IS_ENABLED(CONFIG_SWIOTLB) +#define swiotlb_active() swiotlb_nr_tbl() +#else +#define swiotlb_active() 0 +#endif + +static int +st_set_pages(struct sg_table **st, struct page **pvec, int num_pages) +{ + struct scatterlist *sg; + int ret, n; + + *st = kmalloc(sizeof(**st), GFP_KERNEL); + if (*st == NULL) + return -ENOMEM; + + if (swiotlb_active()) { + ret = sg_alloc_table(*st, num_pages, GFP_KERNEL); + if (ret) + goto err; + + for_each_sg((*st)->sgl, sg, num_pages, n) + sg_set_page(sg, pvec[n], PAGE_SIZE, 0); + } else { + ret = sg_alloc_table_from_pages(*st, pvec, num_pages, + 0, num_pages << PAGE_SHIFT, + GFP_KERNEL); + if (ret) + goto err; + } + + return 0; + +err: + kfree(*st); + *st = NULL; + return ret; +} + +static void +__i915_gem_userptr_get_pages_worker(struct work_struct *_work) +{ + struct get_pages_work *work = container_of(_work, typeof(*work), work); + struct drm_i915_gem_object *obj = work->obj; + struct drm_device *dev = obj->base.dev; + const int num_pages = obj->base.size >> PAGE_SHIFT; + struct page **pvec; + int pinned, ret; + + ret = -ENOMEM; + pinned = 0; + + pvec = kmalloc(num_pages*sizeof(struct page *), + GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY); + if (pvec == NULL) + pvec = drm_malloc_ab(num_pages, sizeof(struct page *)); + if (pvec != NULL) { + struct mm_struct *mm = obj->userptr.mm; + + down_read(&mm->mmap_sem); + while (pinned < num_pages) { + ret = get_user_pages(work->task, mm, + obj->userptr.ptr + pinned * PAGE_SIZE, + num_pages - pinned, + !obj->userptr.read_only, 0, + pvec + pinned, NULL); + if (ret < 0) + break; + + pinned += ret; + } + up_read(&mm->mmap_sem); + } + + mutex_lock(&dev->struct_mutex); + if (obj->userptr.work != &work->work) { + ret = 0; + } else if (pinned == num_pages) { + ret = st_set_pages(&obj->pages, pvec, num_pages); + if (ret == 0) { + list_add_tail(&obj->global_list, &to_i915(dev)->mm.unbound_list); + pinned = 0; + } + } + + obj->userptr.work = ERR_PTR(ret); + obj->userptr.workers--; + drm_gem_object_unreference(&obj->base); + mutex_unlock(&dev->struct_mutex); + + release_pages(pvec, pinned, 0); + drm_free_large(pvec); + + put_task_struct(work->task); + kfree(work); +} + +static int +i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj) +{ + const int num_pages = obj->base.size >> PAGE_SHIFT; + struct page **pvec; + int pinned, ret; + + /* If userspace should engineer that these pages are replaced in + * the vma between us binding this page into the GTT and completion + * of rendering... Their loss. If they change the mapping of their + * pages they need to create a new bo to point to the new vma. + * + * However, that still leaves open the possibility of the vma + * being copied upon fork. Which falls under the same userspace + * synchronisation issue as a regular bo, except that this time + * the process may not be expecting that a particular piece of + * memory is tied to the GPU. + * + * Fortunately, we can hook into the mmu_notifier in order to + * discard the page references prior to anything nasty happening + * to the vma (discard or cloning) which should prevent the more + * egregious cases from causing harm. + */ + + pvec = NULL; + pinned = 0; + if (obj->userptr.mm == current->mm) { + pvec = kmalloc(num_pages*sizeof(struct page *), + GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY); + if (pvec == NULL) { + pvec = drm_malloc_ab(num_pages, sizeof(struct page *)); + if (pvec == NULL) + return -ENOMEM; + } + + pinned = __get_user_pages_fast(obj->userptr.ptr, num_pages, + !obj->userptr.read_only, pvec); + } + if (pinned < num_pages) { + if (pinned < 0) { + ret = pinned; + pinned = 0; + } else { + /* Spawn a worker so that we can acquire the + * user pages without holding our mutex. Access + * to the user pages requires mmap_sem, and we have + * a strict lock ordering of mmap_sem, struct_mutex - + * we already hold struct_mutex here and so cannot + * call gup without encountering a lock inversion. + * + * Userspace will keep on repeating the operation + * (thanks to EAGAIN) until either we hit the fast + * path or the worker completes. If the worker is + * cancelled or superseded, the task is still run + * but the results ignored. (This leads to + * complications that we may have a stray object + * refcount that we need to be wary of when + * checking for existing objects during creation.) + * If the worker encounters an error, it reports + * that error back to this function through + * obj->userptr.work = ERR_PTR. + */ + ret = -EAGAIN; + if (obj->userptr.work == NULL && + obj->userptr.workers < I915_GEM_USERPTR_MAX_WORKERS) { + struct get_pages_work *work; + + work = kmalloc(sizeof(*work), GFP_KERNEL); + if (work != NULL) { + obj->userptr.work = &work->work; + obj->userptr.workers++; + + work->obj = obj; + drm_gem_object_reference(&obj->base); + + work->task = current; + get_task_struct(work->task); + + INIT_WORK(&work->work, __i915_gem_userptr_get_pages_worker); + schedule_work(&work->work); + } else + ret = -ENOMEM; + } else { + if (IS_ERR(obj->userptr.work)) { + ret = PTR_ERR(obj->userptr.work); + obj->userptr.work = NULL; + } + } + } + } else { + ret = st_set_pages(&obj->pages, pvec, num_pages); + if (ret == 0) { + obj->userptr.work = NULL; + pinned = 0; + } + } + + release_pages(pvec, pinned, 0); + drm_free_large(pvec); + return ret; +} + +static void +i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj) +{ + struct scatterlist *sg; + int i; + + BUG_ON(obj->userptr.work != NULL); + + if (obj->madv != I915_MADV_WILLNEED) + obj->dirty = 0; + + for_each_sg(obj->pages->sgl, sg, obj->pages->nents, i) { + struct page *page = sg_page(sg); + + if (obj->dirty) + set_page_dirty(page); + + mark_page_accessed(page); + page_cache_release(page); + } + obj->dirty = 0; + + sg_free_table(obj->pages); + kfree(obj->pages); +} + +static void +i915_gem_userptr_release(struct drm_i915_gem_object *obj) +{ + i915_gem_userptr_release__mmu_notifier(obj); + + if (obj->userptr.mm) { + mmput(obj->userptr.mm); + obj->userptr.mm = NULL; + } +} + +static int +i915_gem_userptr_dmabuf_export(struct drm_i915_gem_object *obj) +{ + if (obj->userptr.mn) + return 0; + + return i915_gem_userptr_init__mmu_notifier(obj, 0); +} + +static const struct drm_i915_gem_object_ops i915_gem_userptr_ops = { + .dmabuf_export = i915_gem_userptr_dmabuf_export, + .get_pages = i915_gem_userptr_get_pages, + .put_pages = i915_gem_userptr_put_pages, + .release = i915_gem_userptr_release, +}; + +/** + * Creates a new mm object that wraps some normal memory from the process + * context - user memory. + * + * We impose several restrictions upon the memory being mapped + * into the GPU. + * 1. It must be page aligned (both start/end addresses, i.e ptr and size). + * 2. It cannot overlap any other userptr object in the same address space. + * 3. It must be normal system memory, not a pointer into another map of IO + * space (e.g. it must not be a GTT mmapping of another object). + * 4. We only allow a bo as large as we could in theory map into the GTT, + * that is we limit the size to the total size of the GTT. + * 5. The bo is marked as being snoopable. The backing pages are left + * accessible directly by the CPU, but reads and writes by the GPU may + * incur the cost of a snoop (unless you have an LLC architecture). + * + * Synchronisation between multiple users and the GPU is left to userspace + * through the normal set-domain-ioctl. The kernel will enforce that the + * GPU relinquishes the VMA before it is returned back to the system + * i.e. upon free(), munmap() or process termination. However, the userspace + * malloc() library may not immediately relinquish the VMA after free() and + * instead reuse it whilst the GPU is still reading and writing to the VMA. + * Caveat emptor. + * + * Also note, that the object created here is not currently a "first class" + * object, in that several ioctls are banned. These are the CPU access + * ioctls: mmap(), pwrite and pread. In practice, you are expected to use + * direct access via your pointer rather than use those ioctls. + * + * If you think this is a good interface to use to pass GPU memory between + * drivers, please use dma-buf instead. In fact, wherever possible use + * dma-buf instead. + */ +int +i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_gem_userptr *args = data; + struct drm_i915_gem_object *obj; + int ret; + u32 handle; + + if (args->flags & ~(I915_USERPTR_READ_ONLY | + I915_USERPTR_UNSYNCHRONIZED)) + return -EINVAL; + + if (offset_in_page(args->user_ptr | args->user_size)) + return -EINVAL; + + if (args->user_size > dev_priv->gtt.base.total) + return -E2BIG; + + if (!access_ok(args->flags & I915_USERPTR_READ_ONLY ? VERIFY_READ : VERIFY_WRITE, + (char __user *)(unsigned long)args->user_ptr, args->user_size)) + return -EFAULT; + + if (args->flags & I915_USERPTR_READ_ONLY) { + /* On almost all of the current hw, we cannot tell the GPU that a + * page is readonly, so this is just a placeholder in the uAPI. + */ + return -ENODEV; + } + + /* Allocate the new object */ + obj = i915_gem_object_alloc(dev); + if (obj == NULL) + return -ENOMEM; + + drm_gem_private_object_init(dev, &obj->base, args->user_size); + i915_gem_object_init(obj, &i915_gem_userptr_ops); + obj->cache_level = I915_CACHE_LLC; + obj->base.write_domain = I915_GEM_DOMAIN_CPU; + obj->base.read_domains = I915_GEM_DOMAIN_CPU; + + obj->userptr.ptr = args->user_ptr; + obj->userptr.read_only = !!(args->flags & I915_USERPTR_READ_ONLY); + + /* And keep a pointer to the current->mm for resolving the user pages + * at binding. This means that we need to hook into the mmu_notifier + * in order to detect if the mmu is destroyed. + */ + ret = -ENOMEM; + if ((obj->userptr.mm = get_task_mm(current))) + ret = i915_gem_userptr_init__mmu_notifier(obj, args->flags); + if (ret == 0) + ret = drm_gem_handle_create(file, &obj->base, &handle); + + /* drop reference from allocate - handle holds it now */ + drm_gem_object_unreference_unlocked(&obj->base); + if (ret) + return ret; + + args->handle = handle; + return 0; +} + +int +i915_gem_init_userptr(struct drm_device *dev) +{ +#if defined(CONFIG_MMU_NOTIFIER) + struct drm_i915_private *dev_priv = to_i915(dev); + hash_init(dev_priv->mmu_notifiers); +#endif + return 0; +} diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 2d819858c19b..dcbec692f60f 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -205,6 +205,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m, err_puts(m, tiling_flag(err->tiling)); err_puts(m, dirty_flag(err->dirty)); err_puts(m, purgeable_flag(err->purgeable)); + err_puts(m, err->userptr ? " userptr" : ""); err_puts(m, err->ring != -1 ? " " : ""); err_puts(m, ring_str(err->ring)); err_puts(m, i915_cache_level_str(err->cache_level)); @@ -641,6 +642,7 @@ static void capture_bo(struct drm_i915_error_buffer *err, err->tiling = obj->tiling_mode; err->dirty = obj->dirty; err->purgeable = obj->madv != I915_MADV_WILLNEED; + err->userptr = obj->userptr.mm != NULL; err->ring = obj->ring ? obj->ring->id : -1; err->cache_level = obj->cache_level; } diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 8a3e4ef00c3d..ff57f07c3249 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -223,6 +223,7 @@ typedef struct _drm_i915_sarea { #define DRM_I915_GEM_GET_CACHING 0x30 #define DRM_I915_REG_READ 0x31 #define DRM_I915_GET_RESET_STATS 0x32 +#define DRM_I915_GEM_USERPTR 0x33 #define DRM_IOCTL_I915_INIT DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t) #define DRM_IOCTL_I915_FLUSH DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH) @@ -273,6 +274,7 @@ typedef struct _drm_i915_sarea { #define DRM_IOCTL_I915_GEM_CONTEXT_DESTROY DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_CONTEXT_DESTROY, struct drm_i915_gem_context_destroy) #define DRM_IOCTL_I915_REG_READ DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_REG_READ, struct drm_i915_reg_read) #define DRM_IOCTL_I915_GET_RESET_STATS DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GET_RESET_STATS, struct drm_i915_reset_stats) +#define DRM_IOCTL_I915_GEM_USERPTR DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_USERPTR, struct drm_i915_gem_userptr) /* Allow drivers to submit batchbuffers directly to hardware, relying * on the security mechanisms provided by hardware. @@ -1050,4 +1052,18 @@ struct drm_i915_reset_stats { __u32 pad; }; +struct drm_i915_gem_userptr { + __u64 user_ptr; + __u64 user_size; + __u32 flags; +#define I915_USERPTR_READ_ONLY 0x1 +#define I915_USERPTR_UNSYNCHRONIZED 0x80000000 + /** + * Returned handle for the object. + * + * Object handles are nonzero. + */ + __u32 handle; +}; + #endif /* _UAPI_I915_DRM_H_ */ -- cgit v1.2.3 From 7455fa2422898eee3464032351d20695930d9542 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 15 May 2014 01:41:23 +0100 Subject: ethtool: Name the 'no change' value for setting RSS hash key but not indir table We usually allocate special values of u32 fields starting from the top down, so also change the value to 0xffffffff. As these operations haven't been included in a stable release yet, it's not too late to change. Signed-off-by: Ben Hutchings --- include/uapi/linux/ethtool.h | 12 +++++++----- net/core/ethtool.c | 12 +++++++----- 2 files changed, 14 insertions(+), 10 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index d47d31d6fa0e..cba18e3f8fab 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -850,7 +850,8 @@ struct ethtool_rxfh_indir { * struct ethtool_rxfh - command to get/set RX flow hash indir or/and hash key. * @cmd: Specific command number - %ETHTOOL_GRSSH or %ETHTOOL_SRSSH * @rss_context: RSS context identifier. - * @indir_size: On entry, the array size of the user buffer, which may be zero. + * @indir_size: On entry, the array size of the user buffer, which may be zero, + * or (for %ETHTOOL_SRSSH), %ETH_RXFH_INDIR_NO_CHANGE. * On return from %ETHTOOL_GRSSH, the array size of the hardware * indirection table. * @key_size: On entry, the array size of the user buffer in bytes, @@ -861,10 +862,10 @@ struct ethtool_rxfh_indir { * of size @indir_size followed by hash key of size @key_size. * * For %ETHTOOL_GRSSH, a @indir_size and key_size of zero means that only the - * size should be returned. For %ETHTOOL_SRSSH, a @indir_size of 0xDEADBEEF - * means that indir table setting is not requested and a @indir_size of zero - * means the indir table should be reset to default values. This last feature - * is not supported by the original implementations. + * size should be returned. For %ETHTOOL_SRSSH, an @indir_size of + * %ETH_RXFH_INDIR_NO_CHANGE means that indir table setting is not requested + * and a @indir_size of zero means the indir table should be reset to default + * values. */ struct ethtool_rxfh { __u32 cmd; @@ -874,6 +875,7 @@ struct ethtool_rxfh { __u32 rsvd[2]; __u32 rss_config[0]; }; +#define ETH_RXFH_INDIR_NO_CHANGE 0xffffffff /** * struct ethtool_rx_ntuple_flow_spec - specification for RX flow filter diff --git a/net/core/ethtool.c b/net/core/ethtool.c index c834cb29f682..7156fe5ca876 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -803,12 +803,13 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, /* If either indir or hash key is valid, proceed further. */ - if ((user_indir_size && ((user_indir_size != 0xDEADBEEF) && - user_indir_size != dev_indir_size)) || + if ((user_indir_size && + user_indir_size != ETH_RXFH_INDIR_NO_CHANGE && + user_indir_size != dev_indir_size) || (user_key_size && (user_key_size != dev_key_size))) return -EINVAL; - if (user_indir_size != 0xDEADBEEF) + if (user_indir_size != ETH_RXFH_INDIR_NO_CHANGE) indir_bytes = dev_indir_size * sizeof(indir[0]); rss_config = kzalloc(indir_bytes + user_key_size, GFP_USER); @@ -821,9 +822,10 @@ static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, goto out; /* user_indir_size == 0 means reset the indir table to default. - * user_indir_size == 0xDEADBEEF means indir setting is not requested. + * user_indir_size == ETH_RXFH_INDIR_NO_CHANGE means leave it unchanged. */ - if (user_indir_size && user_indir_size != 0xDEADBEEF) { + if (user_indir_size && + user_indir_size != ETH_RXFH_INDIR_NO_CHANGE) { indir = (u32 *)rss_config; ret = ethtool_copy_validate_indir(indir, useraddr + rss_cfg_offset, -- cgit v1.2.3 From 38c891a49dec43dbb1575cc40d10dbd49c4961ab Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 15 May 2014 01:07:16 +0100 Subject: ethtool: Improve explanation of the two arrays following struct ethtool_rxfh The use of two variable-length arrays is unusual so deserves a bit more explanation. Signed-off-by: Ben Hutchings --- include/uapi/linux/ethtool.h | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index cba18e3f8fab..e3c7a719c76b 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -850,16 +850,17 @@ struct ethtool_rxfh_indir { * struct ethtool_rxfh - command to get/set RX flow hash indir or/and hash key. * @cmd: Specific command number - %ETHTOOL_GRSSH or %ETHTOOL_SRSSH * @rss_context: RSS context identifier. - * @indir_size: On entry, the array size of the user buffer, which may be zero, - * or (for %ETHTOOL_SRSSH), %ETH_RXFH_INDIR_NO_CHANGE. - * On return from %ETHTOOL_GRSSH, the array size of the hardware - * indirection table. - * @key_size: On entry, the array size of the user buffer in bytes, - * which may be zero. - * On return from %ETHTOOL_GRSSH, the size of the RSS hash key. + * @indir_size: On entry, the array size of the user buffer for the + * indirection table, which may be zero, or (for %ETHTOOL_SRSSH), + * %ETH_RXFH_INDIR_NO_CHANGE. On return from %ETHTOOL_GRSSH, + * the array size of the hardware indirection table. + * @key_size: On entry, the array size of the user buffer for the hash key, + * which may be zero. On return from %ETHTOOL_GRSSH, the size of the + * hardware hash key. * @rsvd: Reserved for future extensions. * @rss_config: RX ring/queue index for each hash value i.e., indirection table - * of size @indir_size followed by hash key of size @key_size. + * of @indir_size __u32 elements, followed by hash key of @key_size + * bytes. * * For %ETHTOOL_GRSSH, a @indir_size and key_size of zero means that only the * size should be returned. For %ETHTOOL_SRSSH, an @indir_size of -- cgit v1.2.3 From 42193e3efb632c84d686acacd7b2327f2b1f8c63 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Thu, 15 May 2014 20:31:56 +0200 Subject: can: unify identifiers to ensure unique include processing Armin pointed me to the fact that the identifier which is used to ensure the unique include processing in lunux/include/uapi/linux/can.h is CAN_H. This clashed with his own source as includes from libraries and APIs should use an underscore '_' at the identifier start. This patch fixes the protection identifiers in all CAN relavant includes. Reported-by: Armin Burchardt Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde --- include/linux/can/core.h | 6 +++--- include/linux/can/dev.h | 6 +++--- include/linux/can/led.h | 6 +++--- include/linux/can/platform/cc770.h | 6 +++--- include/linux/can/platform/mcp251x.h | 6 +++--- include/linux/can/platform/sja1000.h | 6 +++--- include/linux/can/platform/ti_hecc.h | 6 +++--- include/linux/can/skb.h | 6 +++--- include/uapi/linux/can.h | 6 +++--- include/uapi/linux/can/bcm.h | 6 +++--- include/uapi/linux/can/error.h | 6 +++--- include/uapi/linux/can/gw.h | 6 +++--- include/uapi/linux/can/netlink.h | 6 +++--- include/uapi/linux/can/raw.h | 6 +++--- 14 files changed, 42 insertions(+), 42 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/can/core.h b/include/linux/can/core.h index 78c6c52073ad..a0875001b13c 100644 --- a/include/linux/can/core.h +++ b/include/linux/can/core.h @@ -10,8 +10,8 @@ * */ -#ifndef CAN_CORE_H -#define CAN_CORE_H +#ifndef _CAN_CORE_H +#define _CAN_CORE_H #include #include @@ -58,4 +58,4 @@ extern void can_rx_unregister(struct net_device *dev, canid_t can_id, extern int can_send(struct sk_buff *skb, int loop); extern int can_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg); -#endif /* CAN_CORE_H */ +#endif /* !_CAN_CORE_H */ diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index 3ce5e526525f..6992afc6ba7f 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -10,8 +10,8 @@ * */ -#ifndef CAN_DEV_H -#define CAN_DEV_H +#ifndef _CAN_DEV_H +#define _CAN_DEV_H #include #include @@ -132,4 +132,4 @@ struct sk_buff *alloc_canfd_skb(struct net_device *dev, struct sk_buff *alloc_can_err_skb(struct net_device *dev, struct can_frame **cf); -#endif /* CAN_DEV_H */ +#endif /* !_CAN_DEV_H */ diff --git a/include/linux/can/led.h b/include/linux/can/led.h index 9c1167baf273..e0475c5cbb92 100644 --- a/include/linux/can/led.h +++ b/include/linux/can/led.h @@ -6,8 +6,8 @@ * published by the Free Software Foundation. */ -#ifndef CAN_LED_H -#define CAN_LED_H +#ifndef _CAN_LED_H +#define _CAN_LED_H #include #include @@ -48,4 +48,4 @@ static inline void can_led_notifier_exit(void) #endif -#endif +#endif /* !_CAN_LED_H */ diff --git a/include/linux/can/platform/cc770.h b/include/linux/can/platform/cc770.h index 7702641f87ee..78b2d44f04cf 100644 --- a/include/linux/can/platform/cc770.h +++ b/include/linux/can/platform/cc770.h @@ -1,5 +1,5 @@ -#ifndef _CAN_PLATFORM_CC770_H_ -#define _CAN_PLATFORM_CC770_H_ +#ifndef _CAN_PLATFORM_CC770_H +#define _CAN_PLATFORM_CC770_H /* CPU Interface Register (0x02) */ #define CPUIF_CEN 0x01 /* Clock Out Enable */ @@ -30,4 +30,4 @@ struct cc770_platform_data { u8 bcr; /* Bus Configuration Register */ }; -#endif /* !_CAN_PLATFORM_CC770_H_ */ +#endif /* !_CAN_PLATFORM_CC770_H */ diff --git a/include/linux/can/platform/mcp251x.h b/include/linux/can/platform/mcp251x.h index dc029dba7a03..d44fcae274ff 100644 --- a/include/linux/can/platform/mcp251x.h +++ b/include/linux/can/platform/mcp251x.h @@ -1,5 +1,5 @@ -#ifndef __CAN_PLATFORM_MCP251X_H__ -#define __CAN_PLATFORM_MCP251X_H__ +#ifndef _CAN_PLATFORM_MCP251X_H +#define _CAN_PLATFORM_MCP251X_H /* * @@ -18,4 +18,4 @@ struct mcp251x_platform_data { unsigned long oscillator_frequency; }; -#endif /* __CAN_PLATFORM_MCP251X_H__ */ +#endif /* !_CAN_PLATFORM_MCP251X_H */ diff --git a/include/linux/can/platform/sja1000.h b/include/linux/can/platform/sja1000.h index 96f8fcc78d78..93570b61ec6c 100644 --- a/include/linux/can/platform/sja1000.h +++ b/include/linux/can/platform/sja1000.h @@ -1,5 +1,5 @@ -#ifndef _CAN_PLATFORM_SJA1000_H_ -#define _CAN_PLATFORM_SJA1000_H_ +#ifndef _CAN_PLATFORM_SJA1000_H +#define _CAN_PLATFORM_SJA1000_H /* clock divider register */ #define CDR_CLKOUT_MASK 0x07 @@ -32,4 +32,4 @@ struct sja1000_platform_data { u8 cdr; /* clock divider register */ }; -#endif /* !_CAN_PLATFORM_SJA1000_H_ */ +#endif /* !_CAN_PLATFORM_SJA1000_H */ diff --git a/include/linux/can/platform/ti_hecc.h b/include/linux/can/platform/ti_hecc.h index af17cb3f7a84..a52f47ca6c8a 100644 --- a/include/linux/can/platform/ti_hecc.h +++ b/include/linux/can/platform/ti_hecc.h @@ -1,5 +1,5 @@ -#ifndef __CAN_PLATFORM_TI_HECC_H__ -#define __CAN_PLATFORM_TI_HECC_H__ +#ifndef _CAN_PLATFORM_TI_HECC_H +#define _CAN_PLATFORM_TI_HECC_H /* * TI HECC (High End CAN Controller) driver platform header @@ -41,4 +41,4 @@ struct ti_hecc_platform_data { u32 version; void (*transceiver_switch) (int); }; -#endif +#endif /* !_CAN_PLATFORM_TI_HECC_H */ diff --git a/include/linux/can/skb.h b/include/linux/can/skb.h index f9bbbb472663..cc00d15c6107 100644 --- a/include/linux/can/skb.h +++ b/include/linux/can/skb.h @@ -7,8 +7,8 @@ * */ -#ifndef CAN_SKB_H -#define CAN_SKB_H +#ifndef _CAN_SKB_H +#define _CAN_SKB_H #include #include @@ -80,4 +80,4 @@ static inline struct sk_buff *can_create_echo_skb(struct sk_buff *skb) return skb; } -#endif /* CAN_SKB_H */ +#endif /* !_CAN_SKB_H */ diff --git a/include/uapi/linux/can.h b/include/uapi/linux/can.h index 5d9d1d140718..41892f720057 100644 --- a/include/uapi/linux/can.h +++ b/include/uapi/linux/can.h @@ -42,8 +42,8 @@ * DAMAGE. */ -#ifndef CAN_H -#define CAN_H +#ifndef _UAPI_CAN_H +#define _UAPI_CAN_H #include #include @@ -191,4 +191,4 @@ struct can_filter { #define CAN_INV_FILTER 0x20000000U /* to be set in can_filter.can_id */ -#endif /* CAN_H */ +#endif /* !_UAPI_CAN_H */ diff --git a/include/uapi/linux/can/bcm.h b/include/uapi/linux/can/bcm.h index 382251a1d214..89ddb9dc9bdf 100644 --- a/include/uapi/linux/can/bcm.h +++ b/include/uapi/linux/can/bcm.h @@ -41,8 +41,8 @@ * DAMAGE. */ -#ifndef CAN_BCM_H -#define CAN_BCM_H +#ifndef _UAPI_CAN_BCM_H +#define _UAPI_CAN_BCM_H #include #include @@ -95,4 +95,4 @@ enum { #define TX_RESET_MULTI_IDX 0x0200 #define RX_RTR_FRAME 0x0400 -#endif /* CAN_BCM_H */ +#endif /* !_UAPI_CAN_BCM_H */ diff --git a/include/uapi/linux/can/error.h b/include/uapi/linux/can/error.h index b63204545320..c247446ab25a 100644 --- a/include/uapi/linux/can/error.h +++ b/include/uapi/linux/can/error.h @@ -41,8 +41,8 @@ * DAMAGE. */ -#ifndef CAN_ERROR_H -#define CAN_ERROR_H +#ifndef _UAPI_CAN_ERROR_H +#define _UAPI_CAN_ERROR_H #define CAN_ERR_DLC 8 /* dlc for error message frames */ @@ -120,4 +120,4 @@ /* controller specific additional information / data[5..7] */ -#endif /* CAN_ERROR_H */ +#endif /* _UAPI_CAN_ERROR_H */ diff --git a/include/uapi/linux/can/gw.h b/include/uapi/linux/can/gw.h index 844c8964bdfe..3e6184cf2f6d 100644 --- a/include/uapi/linux/can/gw.h +++ b/include/uapi/linux/can/gw.h @@ -41,8 +41,8 @@ * DAMAGE. */ -#ifndef CAN_GW_H -#define CAN_GW_H +#ifndef _UAPI_CAN_GW_H +#define _UAPI_CAN_GW_H #include #include @@ -200,4 +200,4 @@ enum { * Beware of sending unpacked or aligned structs! */ -#endif +#endif /* !_UAPI_CAN_GW_H */ diff --git a/include/uapi/linux/can/netlink.h b/include/uapi/linux/can/netlink.h index 7e2e1863db16..813d11f54977 100644 --- a/include/uapi/linux/can/netlink.h +++ b/include/uapi/linux/can/netlink.h @@ -15,8 +15,8 @@ * GNU General Public License for more details. */ -#ifndef CAN_NETLINK_H -#define CAN_NETLINK_H +#ifndef _UAPI_CAN_NETLINK_H +#define _UAPI_CAN_NETLINK_H #include @@ -130,4 +130,4 @@ enum { #define IFLA_CAN_MAX (__IFLA_CAN_MAX - 1) -#endif /* CAN_NETLINK_H */ +#endif /* !_UAPI_CAN_NETLINK_H */ diff --git a/include/uapi/linux/can/raw.h b/include/uapi/linux/can/raw.h index c7d8c334e0ce..78ec76fd89a6 100644 --- a/include/uapi/linux/can/raw.h +++ b/include/uapi/linux/can/raw.h @@ -42,8 +42,8 @@ * DAMAGE. */ -#ifndef CAN_RAW_H -#define CAN_RAW_H +#ifndef _UAPI_CAN_RAW_H +#define _UAPI_CAN_RAW_H #include @@ -59,4 +59,4 @@ enum { CAN_RAW_FD_FRAMES, /* allow CAN FD frames (default:off) */ }; -#endif +#endif /* !_UAPI_CAN_RAW_H */ -- cgit v1.2.3 From 958bee14d0718ca7a5002c0f48a099d1d345812a Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 3 Apr 2014 11:48:44 +0200 Subject: netfilter: nf_tables: use new transaction infrastructure to handle sets This patch reworks the nf_tables API so set updates are included in the same batch that contains rule updates. This speeds up rule-set updates since we skip a dialog of four messages between kernel and user-space (two on each direction), from: 1) create the set and send netlink message to the kernel 2) process the response from the kernel that contains the allocated name. 3) add the set elements and send netlink message to the kernel. 4) process the response from the kernel (to check for errors). To: 1) add the set to the batch. 2) add the set elements to the batch. 3) add the rule that points to the set. 4) send batch to the kernel. This also introduces an internal set ID (NFTA_SET_ID) that is unique in the batch so set elements and rules can refer to new sets. Backward compatibility has been only retained in userspace, this means that new nft versions can talk to the kernel both in the new and the old fashion. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 12 +++ include/uapi/linux/netfilter/nf_tables.h | 6 ++ net/netfilter/nf_tables_api.c | 123 +++++++++++++++++++++++++++---- net/netfilter/nft_lookup.c | 10 ++- 4 files changed, 133 insertions(+), 18 deletions(-) (limited to 'include/uapi') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index d8dfb2695e0f..0f472d668cbe 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -268,6 +268,8 @@ static inline void *nft_set_priv(const struct nft_set *set) struct nft_set *nf_tables_set_lookup(const struct nft_table *table, const struct nlattr *nla); +struct nft_set *nf_tables_set_lookup_byid(const struct net *net, + const struct nlattr *nla); /** * struct nft_set_binding - nf_tables set binding @@ -408,6 +410,16 @@ struct nft_trans_rule { #define nft_trans_rule(trans) \ (((struct nft_trans_rule *)trans->data)->rule) +struct nft_trans_set { + struct nft_set *set; + u32 set_id; +}; + +#define nft_trans_set(trans) \ + (((struct nft_trans_set *)trans->data)->set) +#define nft_trans_set_id(trans) \ + (((struct nft_trans_set *)trans->data)->set_id) + static inline struct nft_expr *nft_expr_first(const struct nft_rule *rule) { return (struct nft_expr *)&rule->data[0]; diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 7d6433f66bf8..2a88f645a5d8 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -246,6 +246,7 @@ enum nft_set_desc_attributes { * @NFTA_SET_DATA_LEN: mapping data length (NLA_U32) * @NFTA_SET_POLICY: selection policy (NLA_U32) * @NFTA_SET_DESC: set description (NLA_NESTED) + * @NFTA_SET_ID: uniquely identifies a set in a transaction (NLA_U32) */ enum nft_set_attributes { NFTA_SET_UNSPEC, @@ -258,6 +259,7 @@ enum nft_set_attributes { NFTA_SET_DATA_LEN, NFTA_SET_POLICY, NFTA_SET_DESC, + NFTA_SET_ID, __NFTA_SET_MAX }; #define NFTA_SET_MAX (__NFTA_SET_MAX - 1) @@ -293,12 +295,14 @@ enum nft_set_elem_attributes { * @NFTA_SET_ELEM_LIST_TABLE: table of the set to be changed (NLA_STRING) * @NFTA_SET_ELEM_LIST_SET: name of the set to be changed (NLA_STRING) * @NFTA_SET_ELEM_LIST_ELEMENTS: list of set elements (NLA_NESTED: nft_set_elem_attributes) + * @NFTA_SET_ELEM_LIST_SET_ID: uniquely identifies a set in a transaction (NLA_U32) */ enum nft_set_elem_list_attributes { NFTA_SET_ELEM_LIST_UNSPEC, NFTA_SET_ELEM_LIST_TABLE, NFTA_SET_ELEM_LIST_SET, NFTA_SET_ELEM_LIST_ELEMENTS, + NFTA_SET_ELEM_LIST_SET_ID, __NFTA_SET_ELEM_LIST_MAX }; #define NFTA_SET_ELEM_LIST_MAX (__NFTA_SET_ELEM_LIST_MAX - 1) @@ -484,12 +488,14 @@ enum nft_cmp_attributes { * @NFTA_LOOKUP_SET: name of the set where to look for (NLA_STRING) * @NFTA_LOOKUP_SREG: source register of the data to look for (NLA_U32: nft_registers) * @NFTA_LOOKUP_DREG: destination register (NLA_U32: nft_registers) + * @NFTA_LOOKUP_SET_ID: uniquely identifies a set in a transaction (NLA_U32) */ enum nft_lookup_attributes { NFTA_LOOKUP_UNSPEC, NFTA_LOOKUP_SET, NFTA_LOOKUP_SREG, NFTA_LOOKUP_DREG, + NFTA_LOOKUP_SET_ID, __NFTA_LOOKUP_MAX }; #define NFTA_LOOKUP_MAX (__NFTA_LOOKUP_MAX - 1) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 34aa3d507542..02d3cc65690c 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1934,6 +1934,7 @@ static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = { [NFTA_SET_DATA_LEN] = { .type = NLA_U32 }, [NFTA_SET_POLICY] = { .type = NLA_U32 }, [NFTA_SET_DESC] = { .type = NLA_NESTED }, + [NFTA_SET_ID] = { .type = NLA_U32 }, }; static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = { @@ -1984,6 +1985,20 @@ struct nft_set *nf_tables_set_lookup(const struct nft_table *table, return ERR_PTR(-ENOENT); } +struct nft_set *nf_tables_set_lookup_byid(const struct net *net, + const struct nlattr *nla) +{ + struct nft_trans *trans; + u32 id = ntohl(nla_get_be32(nla)); + + list_for_each_entry(trans, &net->nft.commit_list, list) { + if (trans->msg_type == NFT_MSG_NEWSET && + id == nft_trans_set_id(trans)) + return nft_trans_set(trans); + } + return ERR_PTR(-ENOENT); +} + static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, const char *name) { @@ -2259,6 +2274,8 @@ static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) return ret; } +#define NFT_SET_INACTIVE (1 << 15) /* Internal set flag */ + static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2288,6 +2305,8 @@ static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (skb2 == NULL) @@ -2321,6 +2340,26 @@ static int nf_tables_set_desc_parse(const struct nft_ctx *ctx, return 0; } +static int nft_trans_set_add(struct nft_ctx *ctx, int msg_type, + struct nft_set *set) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_set)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] != NULL) { + nft_trans_set_id(trans) = + ntohl(nla_get_be32(ctx->nla[NFTA_SET_ID])); + set->flags |= NFT_SET_INACTIVE; + } + nft_trans_set(trans) = set; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + + return 0; +} + static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2341,7 +2380,8 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (nla[NFTA_SET_TABLE] == NULL || nla[NFTA_SET_NAME] == NULL || - nla[NFTA_SET_KEY_LEN] == NULL) + nla[NFTA_SET_KEY_LEN] == NULL || + nla[NFTA_SET_ID] == NULL) return -EINVAL; memset(&desc, 0, sizeof(desc)); @@ -2458,8 +2498,11 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (err < 0) goto err2; + err = nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set); + if (err < 0) + goto err2; + list_add_tail(&set->list, &table->sets); - nf_tables_set_notify(&ctx, set, NFT_MSG_NEWSET); return 0; err2: @@ -2469,16 +2512,20 @@ err1: return err; } -static void nf_tables_set_destroy(const struct nft_ctx *ctx, struct nft_set *set) +static void nft_set_destroy(struct nft_set *set) { - list_del(&set->list); - nf_tables_set_notify(ctx, set, NFT_MSG_DELSET); - set->ops->destroy(set); module_put(set->ops->owner); kfree(set); } +static void nf_tables_set_destroy(const struct nft_ctx *ctx, struct nft_set *set) +{ + list_del(&set->list); + nf_tables_set_notify(ctx, set, NFT_MSG_DELSET); + nft_set_destroy(set); +} + static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2500,10 +2547,16 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; if (!list_empty(&set->bindings)) return -EBUSY; - nf_tables_set_destroy(&ctx, set); + err = nft_trans_set_add(&ctx, NFT_MSG_DELSET, set); + if (err < 0) + return err; + + list_del(&set->list); return 0; } @@ -2563,7 +2616,8 @@ void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, { list_del(&binding->list); - if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS) + if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS && + !(set->flags & NFT_SET_INACTIVE)) nf_tables_set_destroy(ctx, set); } @@ -2581,6 +2635,7 @@ static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + [NFTA_SET_ELEM_LIST_TABLE] = { .type = NLA_STRING }, [NFTA_SET_ELEM_LIST_SET] = { .type = NLA_STRING }, [NFTA_SET_ELEM_LIST_ELEMENTS] = { .type = NLA_NESTED }, + [NFTA_SET_ELEM_LIST_SET_ID] = { .type = NLA_U32 }, }; static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, @@ -2680,6 +2735,8 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; event = NFT_MSG_NEWSETELEM; event |= NFNL_SUBSYS_NFTABLES << 8; @@ -2743,6 +2800,8 @@ static int nf_tables_getsetelem(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -2928,6 +2987,7 @@ static int nf_tables_newsetelem(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + struct net *net = sock_net(skb->sk); const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; @@ -2938,8 +2998,15 @@ static int nf_tables_newsetelem(struct sock *nlsk, struct sk_buff *skb, return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); - if (IS_ERR(set)) - return PTR_ERR(set); + if (IS_ERR(set)) { + if (nla[NFTA_SET_ELEM_LIST_SET_ID]) { + set = nf_tables_set_lookup_byid(net, + nla[NFTA_SET_ELEM_LIST_SET_ID]); + } + if (IS_ERR(set)) + return PTR_ERR(set); + } + if (!list_empty(&set->bindings) && set->flags & NFT_SET_CONSTANT) return -EBUSY; @@ -3069,7 +3136,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_rule_policy, }, [NFT_MSG_NEWSET] = { - .call = nf_tables_newset, + .call_batch = nf_tables_newset, .attr_count = NFTA_SET_MAX, .policy = nft_set_policy, }, @@ -3079,12 +3146,12 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_set_policy, }, [NFT_MSG_DELSET] = { - .call = nf_tables_delset, + .call_batch = nf_tables_delset, .attr_count = NFTA_SET_MAX, .policy = nft_set_policy, }, [NFT_MSG_NEWSETELEM] = { - .call = nf_tables_newsetelem, + .call_batch = nf_tables_newsetelem, .attr_count = NFTA_SET_ELEM_LIST_MAX, .policy = nft_set_elem_list_policy, }, @@ -3094,7 +3161,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_set_elem_list_policy, }, [NFT_MSG_DELSETELEM] = { - .call = nf_tables_delsetelem, + .call_batch = nf_tables_delsetelem, .attr_count = NFTA_SET_ELEM_LIST_MAX, .policy = nft_set_elem_list_policy, }, @@ -3136,6 +3203,16 @@ static int nf_tables_commit(struct sk_buff *skb) nft_trans_rule(trans), NFT_MSG_DELRULE, 0, trans->ctx.afi->family); break; + case NFT_MSG_NEWSET: + nft_trans_set(trans)->flags &= ~NFT_SET_INACTIVE; + nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), + NFT_MSG_NEWSET); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSET: + nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), + NFT_MSG_DELSET); + break; } } @@ -3148,9 +3225,12 @@ static int nf_tables_commit(struct sk_buff *skb) case NFT_MSG_DELRULE: nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); - nft_trans_destroy(trans); + break; + case NFT_MSG_DELSET: + nft_set_destroy(nft_trans_set(trans)); break; } + nft_trans_destroy(trans); } return 0; @@ -3170,6 +3250,14 @@ static int nf_tables_abort(struct sk_buff *skb) nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); nft_trans_destroy(trans); break; + case NFT_MSG_NEWSET: + list_del(&nft_trans_set(trans)->list); + break; + case NFT_MSG_DELSET: + list_add_tail(&nft_trans_set(trans)->list, + &trans->ctx.table->sets); + nft_trans_destroy(trans); + break; } } @@ -3181,9 +3269,12 @@ static int nf_tables_abort(struct sk_buff *skb) case NFT_MSG_NEWRULE: nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); - nft_trans_destroy(trans); + break; + case NFT_MSG_NEWSET: + nft_set_destroy(nft_trans_set(trans)); break; } + nft_trans_destroy(trans); } return 0; diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 7fd2bea8aa23..6404a726d17b 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -56,8 +56,14 @@ static int nft_lookup_init(const struct nft_ctx *ctx, return -EINVAL; set = nf_tables_set_lookup(ctx->table, tb[NFTA_LOOKUP_SET]); - if (IS_ERR(set)) - return PTR_ERR(set); + if (IS_ERR(set)) { + if (tb[NFTA_LOOKUP_SET_ID]) { + set = nf_tables_set_lookup_byid(ctx->net, + tb[NFTA_LOOKUP_SET_ID]); + } + if (IS_ERR(set)) + return PTR_ERR(set); + } priv->sreg = ntohl(nla_get_be32(tb[NFTA_LOOKUP_SREG])); err = nft_validate_input_register(priv->sreg); -- cgit v1.2.3 From 57be1f3f3ec1ccab6432615ca161c4c9ece2a2aa Mon Sep 17 00:00:00 2001 From: Hiren Tandel Date: Mon, 5 May 2014 19:43:31 +0900 Subject: NFC: Add RAW socket type support for SOCKPROTO_RAW This allows for a more generic NFC sniffing by using SOCKPROTO_RAW SOCK_RAW to read RAW NFC frames. This is for sniffing anything but LLCP (HCI, NCI, etc...). Signed-off-by: Hiren Tandel Signed-off-by: Rahul Tank Signed-off-by: Samuel Ortiz --- include/net/nfc/nfc.h | 3 ++ include/uapi/linux/nfc.h | 16 ++++++--- net/nfc/llcp_commands.c | 2 +- net/nfc/llcp_core.c | 11 +++--- net/nfc/nfc.h | 6 ++++ net/nfc/rawsock.c | 94 +++++++++++++++++++++++++++++++++++++++++++++--- 6 files changed, 117 insertions(+), 15 deletions(-) (limited to 'include/uapi') diff --git a/include/net/nfc/nfc.h b/include/net/nfc/nfc.h index 2e8b40c16274..6c583e244de2 100644 --- a/include/net/nfc/nfc.h +++ b/include/net/nfc/nfc.h @@ -264,4 +264,7 @@ int nfc_add_se(struct nfc_dev *dev, u32 se_idx, u16 type); int nfc_remove_se(struct nfc_dev *dev, u32 se_idx); struct nfc_se *nfc_find_se(struct nfc_dev *dev, u32 se_idx); +void nfc_send_to_raw_sock(struct nfc_dev *dev, struct sk_buff *skb, + u8 payload_type, u8 direction); + #endif /* __NET_NFC_H */ diff --git a/include/uapi/linux/nfc.h b/include/uapi/linux/nfc.h index 9789dc95b6a8..9b19b4461928 100644 --- a/include/uapi/linux/nfc.h +++ b/include/uapi/linux/nfc.h @@ -273,11 +273,19 @@ struct sockaddr_nfc_llcp { * First byte is the adapter index * Second byte contains flags * - 0x01 - Direction (0=RX, 1=TX) - * - 0x02-0x80 - Reserved + * - 0x02-0x04 - Payload type (000=LLCP, 001=NCI, 010=HCI, 011=Digital, + * 100=Proprietary) + * - 0x05-0x80 - Reserved **/ -#define NFC_LLCP_RAW_HEADER_SIZE 2 -#define NFC_LLCP_DIRECTION_RX 0x00 -#define NFC_LLCP_DIRECTION_TX 0x01 +#define NFC_RAW_HEADER_SIZE 2 +#define NFC_DIRECTION_RX 0x00 +#define NFC_DIRECTION_TX 0x01 + +#define RAW_PAYLOAD_LLCP 0 +#define RAW_PAYLOAD_NCI 1 +#define RAW_PAYLOAD_HCI 2 +#define RAW_PAYLOAD_DIGITAL 3 +#define RAW_PAYLOAD_PROPRIETARY 4 /* socket option names */ #define NFC_LLCP_RW 0 diff --git a/net/nfc/llcp_commands.c b/net/nfc/llcp_commands.c index bec6ed15f503..a3ad69a4c648 100644 --- a/net/nfc/llcp_commands.c +++ b/net/nfc/llcp_commands.c @@ -387,7 +387,7 @@ int nfc_llcp_send_symm(struct nfc_dev *dev) __net_timestamp(skb); - nfc_llcp_send_to_raw_sock(local, skb, NFC_LLCP_DIRECTION_TX); + nfc_llcp_send_to_raw_sock(local, skb, NFC_DIRECTION_TX); return nfc_data_exchange(dev, local->target_idx, skb, nfc_llcp_recv, local); diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c index b4671958fcf9..f6278da68763 100644 --- a/net/nfc/llcp_core.c +++ b/net/nfc/llcp_core.c @@ -680,16 +680,17 @@ void nfc_llcp_send_to_raw_sock(struct nfc_llcp_local *local, continue; if (skb_copy == NULL) { - skb_copy = __pskb_copy(skb, NFC_LLCP_RAW_HEADER_SIZE, + skb_copy = __pskb_copy(skb, NFC_RAW_HEADER_SIZE, GFP_ATOMIC); if (skb_copy == NULL) continue; - data = skb_push(skb_copy, NFC_LLCP_RAW_HEADER_SIZE); + data = skb_push(skb_copy, NFC_RAW_HEADER_SIZE); data[0] = local->dev ? local->dev->idx : 0xFF; - data[1] = direction; + data[1] = direction & 0x01; + data[1] |= (RAW_PAYLOAD_LLCP << 1); } nskb = skb_clone(skb_copy, GFP_ATOMIC); @@ -747,7 +748,7 @@ static void nfc_llcp_tx_work(struct work_struct *work) __net_timestamp(skb); nfc_llcp_send_to_raw_sock(local, skb, - NFC_LLCP_DIRECTION_TX); + NFC_DIRECTION_TX); ret = nfc_data_exchange(local->dev, local->target_idx, skb, nfc_llcp_recv, local); @@ -1476,7 +1477,7 @@ static void nfc_llcp_rx_work(struct work_struct *work) __net_timestamp(skb); - nfc_llcp_send_to_raw_sock(local, skb, NFC_LLCP_DIRECTION_RX); + nfc_llcp_send_to_raw_sock(local, skb, NFC_DIRECTION_RX); nfc_llcp_rx_skb(local, skb); diff --git a/net/nfc/nfc.h b/net/nfc/nfc.h index 9d6e74f7e6b3..88d60064890e 100644 --- a/net/nfc/nfc.h +++ b/net/nfc/nfc.h @@ -40,6 +40,12 @@ struct nfc_rawsock { struct work_struct tx_work; bool tx_work_scheduled; }; + +struct nfc_sock_list { + struct hlist_head head; + rwlock_t lock; +}; + #define nfc_rawsock(sk) ((struct nfc_rawsock *) sk) #define to_rawsock_sk(_tx_work) \ ((struct sock *) container_of(_tx_work, struct nfc_rawsock, tx_work)) diff --git a/net/nfc/rawsock.c b/net/nfc/rawsock.c index c27a6e86cae4..8627c75063e2 100644 --- a/net/nfc/rawsock.c +++ b/net/nfc/rawsock.c @@ -27,6 +27,24 @@ #include "nfc.h" +static struct nfc_sock_list raw_sk_list = { + .lock = __RW_LOCK_UNLOCKED(raw_sk_list.lock) +}; + +void nfc_sock_link(struct nfc_sock_list *l, struct sock *sk) +{ + write_lock(&l->lock); + sk_add_node(sk, &l->head); + write_unlock(&l->lock); +} + +void nfc_sock_unlink(struct nfc_sock_list *l, struct sock *sk) +{ + write_lock(&l->lock); + sk_del_node_init(sk); + write_unlock(&l->lock); +} + static void rawsock_write_queue_purge(struct sock *sk) { pr_debug("sk=%p\n", sk); @@ -57,6 +75,9 @@ static int rawsock_release(struct socket *sock) if (!sk) return 0; + if (sock->type == SOCK_RAW) + nfc_sock_unlink(&raw_sk_list, sk); + sock_orphan(sk); sock_put(sk); @@ -275,6 +296,26 @@ static const struct proto_ops rawsock_ops = { .mmap = sock_no_mmap, }; +static const struct proto_ops rawsock_raw_ops = { + .family = PF_NFC, + .owner = THIS_MODULE, + .release = rawsock_release, + .bind = sock_no_bind, + .connect = sock_no_connect, + .socketpair = sock_no_socketpair, + .accept = sock_no_accept, + .getname = sock_no_getname, + .poll = datagram_poll, + .ioctl = sock_no_ioctl, + .listen = sock_no_listen, + .shutdown = sock_no_shutdown, + .setsockopt = sock_no_setsockopt, + .getsockopt = sock_no_getsockopt, + .sendmsg = sock_no_sendmsg, + .recvmsg = rawsock_recvmsg, + .mmap = sock_no_mmap, +}; + static void rawsock_destruct(struct sock *sk) { pr_debug("sk=%p\n", sk); @@ -300,10 +341,13 @@ static int rawsock_create(struct net *net, struct socket *sock, pr_debug("sock=%p\n", sock); - if (sock->type != SOCK_SEQPACKET) + if ((sock->type != SOCK_SEQPACKET) && (sock->type != SOCK_RAW)) return -ESOCKTNOSUPPORT; - sock->ops = &rawsock_ops; + if (sock->type == SOCK_RAW) + sock->ops = &rawsock_raw_ops; + else + sock->ops = &rawsock_ops; sk = sk_alloc(net, PF_NFC, GFP_ATOMIC, nfc_proto->proto); if (!sk) @@ -313,13 +357,53 @@ static int rawsock_create(struct net *net, struct socket *sock, sk->sk_protocol = nfc_proto->id; sk->sk_destruct = rawsock_destruct; sock->state = SS_UNCONNECTED; - - INIT_WORK(&nfc_rawsock(sk)->tx_work, rawsock_tx_work); - nfc_rawsock(sk)->tx_work_scheduled = false; + if (sock->type == SOCK_RAW) + nfc_sock_link(&raw_sk_list, sk); + else { + INIT_WORK(&nfc_rawsock(sk)->tx_work, rawsock_tx_work); + nfc_rawsock(sk)->tx_work_scheduled = false; + } return 0; } +void nfc_send_to_raw_sock(struct nfc_dev *dev, struct sk_buff *skb, + u8 payload_type, u8 direction) +{ + struct sk_buff *skb_copy = NULL, *nskb; + struct sock *sk; + u8 *data; + + read_lock(&raw_sk_list.lock); + + sk_for_each(sk, &raw_sk_list.head) { + if (!skb_copy) { + skb_copy = __pskb_copy(skb, NFC_RAW_HEADER_SIZE, + GFP_ATOMIC); + if (!skb_copy) + continue; + + data = skb_push(skb_copy, NFC_RAW_HEADER_SIZE); + + data[0] = dev ? dev->idx : 0xFF; + data[1] = direction & 0x01; + data[1] |= (payload_type << 1); + } + + nskb = skb_clone(skb_copy, GFP_ATOMIC); + if (!nskb) + continue; + + if (sock_queue_rcv_skb(sk, nskb)) + kfree_skb(nskb); + } + + read_unlock(&raw_sk_list.lock); + + kfree_skb(skb_copy); +} +EXPORT_SYMBOL(nfc_send_to_raw_sock); + static struct proto rawsock_proto = { .name = "NFC_RAW", .owner = THIS_MODULE, -- cgit v1.2.3 From 867d849fc844623a88ec7b380442952b5ffe5e68 Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Mon, 19 May 2014 21:53:19 +0200 Subject: cfg80211: export expected throughput through get_station() Users may need information about the expected throughput towards a given peer. This value is supposed to consider the size overhead generated by the 802.11 header. This value is exported in kbps through the get_station() API by including it into the station_info object. Moreover, it is sent to user space when replying to the nl80211 GET_STATION command. This information will be useful to the batman-adv module which will use it for its new metric computation. Signed-off-by: Antonio Quartulli Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 62 ++++++++++++++++++++++++-------------------- include/uapi/linux/nl80211.h | 3 +++ net/wireless/nl80211.c | 4 +++ 3 files changed, 41 insertions(+), 28 deletions(-) (limited to 'include/uapi') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index fe4fa287f788..857d6476a128 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -873,36 +873,38 @@ int cfg80211_check_station_change(struct wiphy *wiphy, * @STATION_INFO_NONPEER_PM: @nonpeer_pm filled * @STATION_INFO_CHAIN_SIGNAL: @chain_signal filled * @STATION_INFO_CHAIN_SIGNAL_AVG: @chain_signal_avg filled + * @STATION_INFO_EXPECTED_THROUGHPUT: @expected_throughput filled */ enum station_info_flags { - STATION_INFO_INACTIVE_TIME = 1<<0, - STATION_INFO_RX_BYTES = 1<<1, - STATION_INFO_TX_BYTES = 1<<2, - STATION_INFO_LLID = 1<<3, - STATION_INFO_PLID = 1<<4, - STATION_INFO_PLINK_STATE = 1<<5, - STATION_INFO_SIGNAL = 1<<6, - STATION_INFO_TX_BITRATE = 1<<7, - STATION_INFO_RX_PACKETS = 1<<8, - STATION_INFO_TX_PACKETS = 1<<9, - STATION_INFO_TX_RETRIES = 1<<10, - STATION_INFO_TX_FAILED = 1<<11, - STATION_INFO_RX_DROP_MISC = 1<<12, - STATION_INFO_SIGNAL_AVG = 1<<13, - STATION_INFO_RX_BITRATE = 1<<14, - STATION_INFO_BSS_PARAM = 1<<15, - STATION_INFO_CONNECTED_TIME = 1<<16, - STATION_INFO_ASSOC_REQ_IES = 1<<17, - STATION_INFO_STA_FLAGS = 1<<18, - STATION_INFO_BEACON_LOSS_COUNT = 1<<19, - STATION_INFO_T_OFFSET = 1<<20, - STATION_INFO_LOCAL_PM = 1<<21, - STATION_INFO_PEER_PM = 1<<22, - STATION_INFO_NONPEER_PM = 1<<23, - STATION_INFO_RX_BYTES64 = 1<<24, - STATION_INFO_TX_BYTES64 = 1<<25, - STATION_INFO_CHAIN_SIGNAL = 1<<26, - STATION_INFO_CHAIN_SIGNAL_AVG = 1<<27, + STATION_INFO_INACTIVE_TIME = BIT(0), + STATION_INFO_RX_BYTES = BIT(1), + STATION_INFO_TX_BYTES = BIT(2), + STATION_INFO_LLID = BIT(3), + STATION_INFO_PLID = BIT(4), + STATION_INFO_PLINK_STATE = BIT(5), + STATION_INFO_SIGNAL = BIT(6), + STATION_INFO_TX_BITRATE = BIT(7), + STATION_INFO_RX_PACKETS = BIT(8), + STATION_INFO_TX_PACKETS = BIT(9), + STATION_INFO_TX_RETRIES = BIT(10), + STATION_INFO_TX_FAILED = BIT(11), + STATION_INFO_RX_DROP_MISC = BIT(12), + STATION_INFO_SIGNAL_AVG = BIT(13), + STATION_INFO_RX_BITRATE = BIT(14), + STATION_INFO_BSS_PARAM = BIT(15), + STATION_INFO_CONNECTED_TIME = BIT(16), + STATION_INFO_ASSOC_REQ_IES = BIT(17), + STATION_INFO_STA_FLAGS = BIT(18), + STATION_INFO_BEACON_LOSS_COUNT = BIT(19), + STATION_INFO_T_OFFSET = BIT(20), + STATION_INFO_LOCAL_PM = BIT(21), + STATION_INFO_PEER_PM = BIT(22), + STATION_INFO_NONPEER_PM = BIT(23), + STATION_INFO_RX_BYTES64 = BIT(24), + STATION_INFO_TX_BYTES64 = BIT(25), + STATION_INFO_CHAIN_SIGNAL = BIT(26), + STATION_INFO_CHAIN_SIGNAL_AVG = BIT(27), + STATION_INFO_EXPECTED_THROUGHPUT = BIT(28), }; /** @@ -1024,6 +1026,8 @@ struct sta_bss_parameters { * @local_pm: local mesh STA power save mode * @peer_pm: peer mesh STA power save mode * @nonpeer_pm: non-peer mesh STA power save mode + * @expected_throughput: expected throughput in kbps (including 802.11 headers) + * towards this station. */ struct station_info { u32 filled; @@ -1062,6 +1066,8 @@ struct station_info { enum nl80211_mesh_power_mode peer_pm; enum nl80211_mesh_power_mode nonpeer_pm; + u32 expected_throughput; + /* * Note: Add a new enum station_info_flags value for each new field and * use it to check which fields are initialized. diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 0cfa827123ff..fb0efa1f9066 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -2199,6 +2199,8 @@ enum nl80211_sta_bss_param { * Contains a nested array of signal strength attributes (u8, dBm) * @NL80211_STA_INFO_CHAIN_SIGNAL_AVG: per-chain signal strength average * Same format as NL80211_STA_INFO_CHAIN_SIGNAL. + * @NL80211_STA_EXPECTED_THROUGHPUT: expected throughput considering also the + * 802.11 header (u32, kbps) * @__NL80211_STA_INFO_AFTER_LAST: internal * @NL80211_STA_INFO_MAX: highest possible station info attribute */ @@ -2230,6 +2232,7 @@ enum nl80211_sta_info { NL80211_STA_INFO_TX_BYTES64, NL80211_STA_INFO_CHAIN_SIGNAL, NL80211_STA_INFO_CHAIN_SIGNAL_AVG, + NL80211_STA_INFO_EXPECTED_THROUGHPUT, /* keep last */ __NL80211_STA_INFO_AFTER_LAST, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 49adf58646e6..62bdb1adaa4d 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -3650,6 +3650,10 @@ static int nl80211_send_station(struct sk_buff *msg, u32 portid, u32 seq, nla_put_u32(msg, NL80211_STA_INFO_TX_FAILED, sinfo->tx_failed)) goto nla_put_failure; + if ((sinfo->filled & STATION_INFO_EXPECTED_THROUGHPUT) && + nla_put_u32(msg, NL80211_STA_INFO_EXPECTED_THROUGHPUT, + sinfo->expected_throughput)) + goto nla_put_failure; if ((sinfo->filled & STATION_INFO_BEACON_LOSS_COUNT) && nla_put_u32(msg, NL80211_STA_INFO_BEACON_LOSS, sinfo->beacon_loss_count)) -- cgit v1.2.3 From 96ba9dd65788a0bd2a7d1e57ec78b7642f0ccc25 Mon Sep 17 00:00:00 2001 From: Vincenzo Aliberti Date: Thu, 15 May 2014 23:19:32 +0200 Subject: mtd: lpddr: add driver for LPDDR2-NVM PCM memories Signed-off-by: Vincenzo Aliberti Signed-off-by: Brian Norris --- drivers/mtd/lpddr/Kconfig | 12 +- drivers/mtd/lpddr/Makefile | 1 + drivers/mtd/lpddr/lpddr2_nvm.c | 507 +++++++++++++++++++++++++++++++++++++++++ include/uapi/mtd/mtd-abi.h | 1 + 4 files changed, 519 insertions(+), 2 deletions(-) create mode 100644 drivers/mtd/lpddr/lpddr2_nvm.c (limited to 'include/uapi') diff --git a/drivers/mtd/lpddr/Kconfig b/drivers/mtd/lpddr/Kconfig index 265f969817e3..468f06dea45d 100644 --- a/drivers/mtd/lpddr/Kconfig +++ b/drivers/mtd/lpddr/Kconfig @@ -1,5 +1,5 @@ -menu "LPDDR flash memory drivers" - depends on MTD!=n +menu "LPDDR & LPDDR2 PCM memory drivers" + depends on MTD config MTD_LPDDR tristate "Support for LPDDR flash chips" @@ -17,4 +17,12 @@ config MTD_QINFO_PROBE Window QINFO interface, permits software to be used for entire families of devices. This serves similar purpose of CFI on legacy Flash products + +config MTD_LPDDR2_NVM + depends on MTD + tristate "Support for LPDDR2-NVM flash chips" + help + This option enables support of PCM memories with a LPDDR2-NVM + (Low power double data rate 2) interface. + endmenu diff --git a/drivers/mtd/lpddr/Makefile b/drivers/mtd/lpddr/Makefile index da48e46b5812..881d440d483e 100644 --- a/drivers/mtd/lpddr/Makefile +++ b/drivers/mtd/lpddr/Makefile @@ -4,3 +4,4 @@ obj-$(CONFIG_MTD_QINFO_PROBE) += qinfo_probe.o obj-$(CONFIG_MTD_LPDDR) += lpddr_cmds.o +obj-$(CONFIG_MTD_LPDDR2_NVM) += lpddr2_nvm.o diff --git a/drivers/mtd/lpddr/lpddr2_nvm.c b/drivers/mtd/lpddr/lpddr2_nvm.c new file mode 100644 index 000000000000..063cec40d0ae --- /dev/null +++ b/drivers/mtd/lpddr/lpddr2_nvm.c @@ -0,0 +1,507 @@ +/* + * LPDDR2-NVM MTD driver. This module provides read, write, erase, lock/unlock + * support for LPDDR2-NVM PCM memories + * + * Copyright © 2012 Micron Technology, Inc. + * + * Vincenzo Aliberti + * Domenico Manna + * Many thanks to Andrea Vigilante for initial enabling + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": %s: " fmt, __func__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* Parameters */ +#define ERASE_BLOCKSIZE (0x00020000/2) /* in Word */ +#define WRITE_BUFFSIZE (0x00000400/2) /* in Word */ +#define OW_BASE_ADDRESS 0x00000000 /* OW offset */ +#define BUS_WIDTH 0x00000020 /* x32 devices */ + +/* PFOW symbols address offset */ +#define PFOW_QUERY_STRING_P (0x0000/2) /* in Word */ +#define PFOW_QUERY_STRING_F (0x0002/2) /* in Word */ +#define PFOW_QUERY_STRING_O (0x0004/2) /* in Word */ +#define PFOW_QUERY_STRING_W (0x0006/2) /* in Word */ + +/* OW registers address */ +#define CMD_CODE_OFS (0x0080/2) /* in Word */ +#define CMD_DATA_OFS (0x0084/2) /* in Word */ +#define CMD_ADD_L_OFS (0x0088/2) /* in Word */ +#define CMD_ADD_H_OFS (0x008A/2) /* in Word */ +#define MPR_L_OFS (0x0090/2) /* in Word */ +#define MPR_H_OFS (0x0092/2) /* in Word */ +#define CMD_EXEC_OFS (0x00C0/2) /* in Word */ +#define STATUS_REG_OFS (0x00CC/2) /* in Word */ +#define PRG_BUFFER_OFS (0x0010/2) /* in Word */ + +/* Datamask */ +#define MR_CFGMASK 0x8000 +#define SR_OK_DATAMASK 0x0080 + +/* LPDDR2-NVM Commands */ +#define LPDDR2_NVM_LOCK 0x0061 +#define LPDDR2_NVM_UNLOCK 0x0062 +#define LPDDR2_NVM_SW_PROGRAM 0x0041 +#define LPDDR2_NVM_SW_OVERWRITE 0x0042 +#define LPDDR2_NVM_BUF_PROGRAM 0x00E9 +#define LPDDR2_NVM_BUF_OVERWRITE 0x00EA +#define LPDDR2_NVM_ERASE 0x0020 + +/* LPDDR2-NVM Registers offset */ +#define LPDDR2_MODE_REG_DATA 0x0040 +#define LPDDR2_MODE_REG_CFG 0x0050 + +/* + * Internal Type Definitions + * pcm_int_data contains memory controller details: + * @reg_data : LPDDR2_MODE_REG_DATA register address after remapping + * @reg_cfg : LPDDR2_MODE_REG_CFG register address after remapping + * &bus_width: memory bus-width (eg: x16 2 Bytes, x32 4 Bytes) + */ +struct pcm_int_data { + void __iomem *ctl_regs; + int bus_width; +}; + +static DEFINE_MUTEX(lpdd2_nvm_mutex); + +/* + * Build a map_word starting from an u_long + */ +static inline map_word build_map_word(u_long myword) +{ + map_word val = { {0} }; + val.x[0] = myword; + return val; +} + +/* + * Build Mode Register Configuration DataMask based on device bus-width + */ +static inline u_int build_mr_cfgmask(u_int bus_width) +{ + u_int val = MR_CFGMASK; + + if (bus_width == 0x0004) /* x32 device */ + val = val << 16; + + return val; +} + +/* + * Build Status Register OK DataMask based on device bus-width + */ +static inline u_int build_sr_ok_datamask(u_int bus_width) +{ + u_int val = SR_OK_DATAMASK; + + if (bus_width == 0x0004) /* x32 device */ + val = (val << 16)+val; + + return val; +} + +/* + * Evaluates Overlay Window Control Registers address + */ +static inline u_long ow_reg_add(struct map_info *map, u_long offset) +{ + u_long val = 0; + struct pcm_int_data *pcm_data = map->fldrv_priv; + + val = map->pfow_base + offset*pcm_data->bus_width; + + return val; +} + +/* + * Enable lpddr2-nvm Overlay Window + * Overlay Window is a memory mapped area containing all LPDDR2-NVM registers + * used by device commands as well as uservisible resources like Device Status + * Register, Device ID, etc + */ +static inline void ow_enable(struct map_info *map) +{ + struct pcm_int_data *pcm_data = map->fldrv_priv; + + writel_relaxed(build_mr_cfgmask(pcm_data->bus_width) | 0x18, + pcm_data->ctl_regs + LPDDR2_MODE_REG_CFG); + writel_relaxed(0x01, pcm_data->ctl_regs + LPDDR2_MODE_REG_DATA); +} + +/* + * Disable lpddr2-nvm Overlay Window + * Overlay Window is a memory mapped area containing all LPDDR2-NVM registers + * used by device commands as well as uservisible resources like Device Status + * Register, Device ID, etc + */ +static inline void ow_disable(struct map_info *map) +{ + struct pcm_int_data *pcm_data = map->fldrv_priv; + + writel_relaxed(build_mr_cfgmask(pcm_data->bus_width) | 0x18, + pcm_data->ctl_regs + LPDDR2_MODE_REG_CFG); + writel_relaxed(0x02, pcm_data->ctl_regs + LPDDR2_MODE_REG_DATA); +} + +/* + * Execute lpddr2-nvm operations + */ +static int lpddr2_nvm_do_op(struct map_info *map, u_long cmd_code, + u_long cmd_data, u_long cmd_add, u_long cmd_mpr, u_char *buf) +{ + map_word add_l = { {0} }, add_h = { {0} }, mpr_l = { {0} }, + mpr_h = { {0} }, data_l = { {0} }, cmd = { {0} }, + exec_cmd = { {0} }, sr; + map_word data_h = { {0} }; /* only for 2x x16 devices stacked */ + u_long i, status_reg, prg_buff_ofs; + struct pcm_int_data *pcm_data = map->fldrv_priv; + u_int sr_ok_datamask = build_sr_ok_datamask(pcm_data->bus_width); + + /* Builds low and high words for OW Control Registers */ + add_l.x[0] = cmd_add & 0x0000FFFF; + add_h.x[0] = (cmd_add >> 16) & 0x0000FFFF; + mpr_l.x[0] = cmd_mpr & 0x0000FFFF; + mpr_h.x[0] = (cmd_mpr >> 16) & 0x0000FFFF; + cmd.x[0] = cmd_code & 0x0000FFFF; + exec_cmd.x[0] = 0x0001; + data_l.x[0] = cmd_data & 0x0000FFFF; + data_h.x[0] = (cmd_data >> 16) & 0x0000FFFF; /* only for 2x x16 */ + + /* Set Overlay Window Control Registers */ + map_write(map, cmd, ow_reg_add(map, CMD_CODE_OFS)); + map_write(map, data_l, ow_reg_add(map, CMD_DATA_OFS)); + map_write(map, add_l, ow_reg_add(map, CMD_ADD_L_OFS)); + map_write(map, add_h, ow_reg_add(map, CMD_ADD_H_OFS)); + map_write(map, mpr_l, ow_reg_add(map, MPR_L_OFS)); + map_write(map, mpr_h, ow_reg_add(map, MPR_H_OFS)); + if (pcm_data->bus_width == 0x0004) { /* 2x16 devices stacked */ + map_write(map, cmd, ow_reg_add(map, CMD_CODE_OFS) + 2); + map_write(map, data_h, ow_reg_add(map, CMD_DATA_OFS) + 2); + map_write(map, add_l, ow_reg_add(map, CMD_ADD_L_OFS) + 2); + map_write(map, add_h, ow_reg_add(map, CMD_ADD_H_OFS) + 2); + map_write(map, mpr_l, ow_reg_add(map, MPR_L_OFS) + 2); + map_write(map, mpr_h, ow_reg_add(map, MPR_H_OFS) + 2); + } + + /* Fill Program Buffer */ + if ((cmd_code == LPDDR2_NVM_BUF_PROGRAM) || + (cmd_code == LPDDR2_NVM_BUF_OVERWRITE)) { + prg_buff_ofs = (map_read(map, + ow_reg_add(map, PRG_BUFFER_OFS))).x[0]; + for (i = 0; i < cmd_mpr; i++) { + map_write(map, build_map_word(buf[i]), map->pfow_base + + prg_buff_ofs + i); + } + } + + /* Command Execute */ + map_write(map, exec_cmd, ow_reg_add(map, CMD_EXEC_OFS)); + if (pcm_data->bus_width == 0x0004) /* 2x16 devices stacked */ + map_write(map, exec_cmd, ow_reg_add(map, CMD_EXEC_OFS) + 2); + + /* Status Register Check */ + do { + sr = map_read(map, ow_reg_add(map, STATUS_REG_OFS)); + status_reg = sr.x[0]; + if (pcm_data->bus_width == 0x0004) {/* 2x16 devices stacked */ + sr = map_read(map, ow_reg_add(map, + STATUS_REG_OFS) + 2); + status_reg += sr.x[0] << 16; + } + } while ((status_reg & sr_ok_datamask) != sr_ok_datamask); + + return (((status_reg & sr_ok_datamask) == sr_ok_datamask) ? 0 : -EIO); +} + +/* + * Execute lpddr2-nvm operations @ block level + */ +static int lpddr2_nvm_do_block_op(struct mtd_info *mtd, loff_t start_add, + uint64_t len, u_char block_op) +{ + struct map_info *map = mtd->priv; + u_long add, end_add; + int ret = 0; + + mutex_lock(&lpdd2_nvm_mutex); + + ow_enable(map); + + add = start_add; + end_add = add + len; + + do { + ret = lpddr2_nvm_do_op(map, block_op, 0x00, add, add, NULL); + if (ret) + goto out; + add += mtd->erasesize; + } while (add < end_add); + +out: + ow_disable(map); + mutex_unlock(&lpdd2_nvm_mutex); + return ret; +} + +/* + * verify presence of PFOW string + */ +static int lpddr2_nvm_pfow_present(struct map_info *map) +{ + map_word pfow_val[4]; + unsigned int found = 1; + + mutex_lock(&lpdd2_nvm_mutex); + + ow_enable(map); + + /* Load string from array */ + pfow_val[0] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_P)); + pfow_val[1] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_F)); + pfow_val[2] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_O)); + pfow_val[3] = map_read(map, ow_reg_add(map, PFOW_QUERY_STRING_W)); + + /* Verify the string loaded vs expected */ + if (!map_word_equal(map, build_map_word('P'), pfow_val[0])) + found = 0; + if (!map_word_equal(map, build_map_word('F'), pfow_val[1])) + found = 0; + if (!map_word_equal(map, build_map_word('O'), pfow_val[2])) + found = 0; + if (!map_word_equal(map, build_map_word('W'), pfow_val[3])) + found = 0; + + ow_disable(map); + + mutex_unlock(&lpdd2_nvm_mutex); + + return found; +} + +/* + * lpddr2_nvm driver read method + */ +static int lpddr2_nvm_read(struct mtd_info *mtd, loff_t start_add, + size_t len, size_t *retlen, u_char *buf) +{ + struct map_info *map = mtd->priv; + + mutex_lock(&lpdd2_nvm_mutex); + + *retlen = len; + + map_copy_from(map, buf, start_add, *retlen); + + mutex_unlock(&lpdd2_nvm_mutex); + return 0; +} + +/* + * lpddr2_nvm driver write method + */ +static int lpddr2_nvm_write(struct mtd_info *mtd, loff_t start_add, + size_t len, size_t *retlen, const u_char *buf) +{ + struct map_info *map = mtd->priv; + struct pcm_int_data *pcm_data = map->fldrv_priv; + u_long add, current_len, tot_len, target_len, my_data; + u_char *write_buf = (u_char *)buf; + int ret = 0; + + mutex_lock(&lpdd2_nvm_mutex); + + ow_enable(map); + + /* Set start value for the variables */ + add = start_add; + target_len = len; + tot_len = 0; + + while (tot_len < target_len) { + if (!(IS_ALIGNED(add, mtd->writesize))) { /* do sw program */ + my_data = write_buf[tot_len]; + my_data += (write_buf[tot_len+1]) << 8; + if (pcm_data->bus_width == 0x0004) {/* 2x16 devices */ + my_data += (write_buf[tot_len+2]) << 16; + my_data += (write_buf[tot_len+3]) << 24; + } + ret = lpddr2_nvm_do_op(map, LPDDR2_NVM_SW_OVERWRITE, + my_data, add, 0x00, NULL); + if (ret) + goto out; + + add += pcm_data->bus_width; + tot_len += pcm_data->bus_width; + } else { /* do buffer program */ + current_len = min(target_len - tot_len, + (u_long) mtd->writesize); + ret = lpddr2_nvm_do_op(map, LPDDR2_NVM_BUF_OVERWRITE, + 0x00, add, current_len, write_buf + tot_len); + if (ret) + goto out; + + add += current_len; + tot_len += current_len; + } + } + +out: + *retlen = tot_len; + ow_disable(map); + mutex_unlock(&lpdd2_nvm_mutex); + return ret; +} + +/* + * lpddr2_nvm driver erase method + */ +static int lpddr2_nvm_erase(struct mtd_info *mtd, struct erase_info *instr) +{ + int ret = lpddr2_nvm_do_block_op(mtd, instr->addr, instr->len, + LPDDR2_NVM_ERASE); + if (!ret) { + instr->state = MTD_ERASE_DONE; + mtd_erase_callback(instr); + } + + return ret; +} + +/* + * lpddr2_nvm driver unlock method + */ +static int lpddr2_nvm_unlock(struct mtd_info *mtd, loff_t start_add, + uint64_t len) +{ + return lpddr2_nvm_do_block_op(mtd, start_add, len, LPDDR2_NVM_UNLOCK); +} + +/* + * lpddr2_nvm driver lock method + */ +static int lpddr2_nvm_lock(struct mtd_info *mtd, loff_t start_add, + uint64_t len) +{ + return lpddr2_nvm_do_block_op(mtd, start_add, len, LPDDR2_NVM_LOCK); +} + +/* + * lpddr2_nvm driver probe method + */ +static int lpddr2_nvm_probe(struct platform_device *pdev) +{ + struct map_info *map; + struct mtd_info *mtd; + struct resource *add_range; + struct resource *control_regs; + struct pcm_int_data *pcm_data; + + /* Allocate memory control_regs data structures */ + pcm_data = devm_kzalloc(&pdev->dev, sizeof(*pcm_data), GFP_KERNEL); + if (!pcm_data) + return -ENOMEM; + + pcm_data->bus_width = BUS_WIDTH; + + /* Allocate memory for map_info & mtd_info data structures */ + map = devm_kzalloc(&pdev->dev, sizeof(*map), GFP_KERNEL); + if (!map) + return -ENOMEM; + + mtd = devm_kzalloc(&pdev->dev, sizeof(*mtd), GFP_KERNEL); + if (!mtd) + return -ENOMEM; + + /* lpddr2_nvm address range */ + add_range = platform_get_resource(pdev, IORESOURCE_MEM, 0); + + /* Populate map_info data structure */ + *map = (struct map_info) { + .virt = devm_ioremap_resource(&pdev->dev, add_range), + .name = pdev->dev.init_name, + .phys = add_range->start, + .size = resource_size(add_range), + .bankwidth = pcm_data->bus_width / 2, + .pfow_base = OW_BASE_ADDRESS, + .fldrv_priv = pcm_data, + }; + if (IS_ERR(map->virt)) + return PTR_ERR(map->virt); + + simple_map_init(map); /* fill with default methods */ + + control_regs = platform_get_resource(pdev, IORESOURCE_MEM, 1); + pcm_data->ctl_regs = devm_ioremap_resource(&pdev->dev, control_regs); + if (IS_ERR(pcm_data->ctl_regs)) + return PTR_ERR(pcm_data->ctl_regs); + + /* Populate mtd_info data structure */ + *mtd = (struct mtd_info) { + .name = pdev->dev.init_name, + .type = MTD_RAM, + .priv = map, + .size = resource_size(add_range), + .erasesize = ERASE_BLOCKSIZE * pcm_data->bus_width, + .writesize = 1, + .writebufsize = WRITE_BUFFSIZE * pcm_data->bus_width, + .flags = (MTD_CAP_NVRAM | MTD_POWERUP_LOCK), + ._read = lpddr2_nvm_read, + ._write = lpddr2_nvm_write, + ._erase = lpddr2_nvm_erase, + ._unlock = lpddr2_nvm_unlock, + ._lock = lpddr2_nvm_lock, + }; + + /* Verify the presence of the device looking for PFOW string */ + if (!lpddr2_nvm_pfow_present(map)) { + pr_err("device not recognized\n"); + return -EINVAL; + } + /* Parse partitions and register the MTD device */ + return mtd_device_parse_register(mtd, NULL, NULL, NULL, 0); +} + +/* + * lpddr2_nvm driver remove method + */ +static int lpddr2_nvm_remove(struct platform_device *pdev) +{ + return mtd_device_unregister(dev_get_drvdata(&pdev->dev)); +} + +/* Initialize platform_driver data structure for lpddr2_nvm */ +static struct platform_driver lpddr2_nvm_drv = { + .driver = { + .name = "lpddr2_nvm", + }, + .probe = lpddr2_nvm_probe, + .remove = lpddr2_nvm_remove, +}; + +module_platform_driver(lpddr2_nvm_drv); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Vincenzo Aliberti "); +MODULE_DESCRIPTION("MTD driver for LPDDR2-NVM PCM memories"); diff --git a/include/uapi/mtd/mtd-abi.h b/include/uapi/mtd/mtd-abi.h index e272ea060e38..763bb6950402 100644 --- a/include/uapi/mtd/mtd-abi.h +++ b/include/uapi/mtd/mtd-abi.h @@ -109,6 +109,7 @@ struct mtd_write_req { #define MTD_CAP_RAM (MTD_WRITEABLE | MTD_BIT_WRITEABLE | MTD_NO_ERASE) #define MTD_CAP_NORFLASH (MTD_WRITEABLE | MTD_BIT_WRITEABLE) #define MTD_CAP_NANDFLASH (MTD_WRITEABLE) +#define MTD_CAP_NVRAM (MTD_WRITEABLE | MTD_BIT_WRITEABLE | MTD_NO_ERASE) /* Obsolete ECC byte placement modes (used with obsolete MEMGETOOBSEL) */ #define MTD_NANDECC_OFF 0 // Switch off ECC (Not recommended) -- cgit v1.2.3 From be52c9e96a6657d117bb0ec6e11438fb246af5c7 Mon Sep 17 00:00:00 2001 From: Jarno Rajahalme Date: Mon, 5 May 2014 09:59:40 -0700 Subject: openvswitch: Avoid assigning a NULL pointer to flow actions. Flow SET can accept an empty set of actions, with the intended semantics of leaving existing actions unmodified. This seems to have been brokin after OVS 1.7, as we have assigned the flow's actions pointer to NULL in this case, but we never check for the NULL pointer later on. This patch restores the intended behavior and documents it in the include/linux/openvswitch.h. Signed-off-by: Jarno Rajahalme Signed-off-by: Pravin B Shelar --- include/uapi/linux/openvswitch.h | 4 +++- net/openvswitch/datapath.c | 14 ++++++++------ 2 files changed, 11 insertions(+), 7 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index 970553cbbc8e..0b979ee4bfc0 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -395,7 +395,9 @@ struct ovs_key_nd { * @OVS_FLOW_ATTR_ACTIONS: Nested %OVS_ACTION_ATTR_* attributes specifying * the actions to take for packets that match the key. Always present in * notifications. Required for %OVS_FLOW_CMD_NEW requests, optional for - * %OVS_FLOW_CMD_SET requests. + * %OVS_FLOW_CMD_SET requests. An %OVS_FLOW_CMD_SET without + * %OVS_FLOW_ATTR_ACTIONS will not modify the actions. To clear the actions, + * an %OVS_FLOW_ATTR_ACTIONS without any nested attributes must be given. * @OVS_FLOW_ATTR_STATS: &struct ovs_flow_stats giving statistics for this * flow. Present in notifications if the stats would be nonzero. Ignored in * requests. diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 8867d7e2d65b..90a1e5e66287 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -810,6 +810,7 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) goto err_kfree; } } else if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW) { + /* OVS_FLOW_CMD_NEW must have actions. */ error = -EINVAL; goto error; } @@ -849,8 +850,6 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) reply = ovs_flow_cmd_build_info(flow, dp, info, OVS_FLOW_CMD_NEW); } else { /* We found a matching flow. */ - struct sw_flow_actions *old_acts; - /* Bail out if we're not allowed to modify an existing flow. * We accept NLM_F_CREATE in place of the intended NLM_F_EXCL * because Generic Netlink treats the latter as a dump @@ -866,11 +865,14 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) if (!ovs_flow_cmp_unmasked_key(flow, &match)) goto err_unlock_ovs; - /* Update actions. */ - old_acts = ovsl_dereference(flow->sf_acts); - rcu_assign_pointer(flow->sf_acts, acts); - ovs_nla_free_flow_actions(old_acts); + /* Update actions, if present. */ + if (acts) { + struct sw_flow_actions *old_acts; + old_acts = ovsl_dereference(flow->sf_acts); + rcu_assign_pointer(flow->sf_acts, acts); + ovs_nla_free_flow_actions(old_acts); + } reply = ovs_flow_cmd_build_info(flow, dp, info, OVS_FLOW_CMD_NEW); /* Clear stats. */ -- cgit v1.2.3 From 8b2e62cc34feaaf1cac9440a93fb18ac0b1e81bc Mon Sep 17 00:00:00 2001 From: Huacai Chen Date: Wed, 21 May 2014 10:49:19 +0800 Subject: MIPS: Fix a typo error in AUDIT_ARCH definition Missing a "|" in AUDIT_ARCH_MIPSEL64N32 macro definition. Signed-off-by: Huacai Chen Reviewed-by: Markos Chandras Cc: John Crispin Cc: Steven J. Hill Cc: Aurelien Jarno Cc: linux-mips@linux-mips.org Cc: Fuxin Zhang Cc: Zhangjin Wu Patchwork: https://patchwork.linux-mips.org/patch/6978/ Signed-off-by: Ralf Baechle --- include/uapi/linux/audit.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h index 1b1efddb91cd..4c31a366be16 100644 --- a/include/uapi/linux/audit.h +++ b/include/uapi/linux/audit.h @@ -357,7 +357,7 @@ enum { #define AUDIT_ARCH_MIPS64N32 (EM_MIPS|__AUDIT_ARCH_64BIT|\ __AUDIT_ARCH_CONVENTION_MIPS64_N32) #define AUDIT_ARCH_MIPSEL64 (EM_MIPS|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) -#define AUDIT_ARCH_MIPSEL64N32 (EM_MIPS|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE\ +#define AUDIT_ARCH_MIPSEL64N32 (EM_MIPS|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE|\ __AUDIT_ARCH_CONVENTION_MIPS64_N32) #define AUDIT_ARCH_OPENRISC (EM_OPENRISC) #define AUDIT_ARCH_PARISC (EM_PARISC) -- cgit v1.2.3 From ed616689a3d95eb6c9bdbb1ef74b0f50cbdf276a Mon Sep 17 00:00:00 2001 From: Sucheta Chakraborty Date: Thu, 22 May 2014 09:59:05 -0400 Subject: net-next:v4: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool. o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed to have a bandwidth of at least this value. max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth of up to this value. o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced which takes 4 arguments: netdev, VF number, min_tx_rate, max_tx_rate o ndo_set_vf_rate replaces ndo_set_vf_tx_rate handler. o Drivers that currently implement ndo_set_vf_tx_rate should now call ndo_set_vf_rate instead and reject attempt to set a minimum bandwidth greater than 0 for IFLA_VF_TX_RATE when IFLA_VF_RATE is not yet implemented by driver. o If user enters only one of either min_tx_rate or max_tx_rate, then, userland should read back the other value from driver and set both for IFLA_VF_RATE. Drivers that have not yet implemented IFLA_VF_RATE should always return min_tx_rate as 0 when read from ip tool. o If both IFLA_VF_TX_RATE and IFLA_VF_RATE options are specified, then IFLA_VF_RATE should override. o Idea is to have consistent display of rate values to user. o Usage example: - ./ip link set p4p1 vf 0 rate 900 ./ip link show p4p1 32: p4p1: mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 900 (Mbps), max_tx_rate 900Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 300 min_tx_rate 200 ./ip link show p4p1 32: p4p1: mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f0 brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5a, tx rate 300 (Mbps), max_tx_rate 300Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a ./ip link set p4p1 vf 0 max_tx_rate 600 rate 300 ./ip link show p4p1 32: p4p1: mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000 link/ether 00:0e:1e:08:b0:f brd ff:ff:ff:ff:ff:ff vf 0 MAC 3e:a0:ca:bd:ae:5, tx rate 600 (Mbps), max_tx_rate 600Mbps, min_tx_rate 200Mbps vf 1 MAC f6:c6:7c:3f:3d:6c vf 2 MAC 56:32:43:98:d7:71 vf 3 MAC d6:be:c3:b5:85:ff vf 4 MAC ee:a9:9a:1e:19:14 vf 5 MAC 4a:d0:4c:07:52:18 vf 6 MAC 3a:76:44:93:62:f9 vf 7 MAC 82:e9:e7:e3:15:1a Signed-off-by: Sucheta Chakraborty Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c | 3 +- drivers/net/ethernet/emulex/benet/be_main.c | 21 +++++---- drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 26 +++++++---- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h | 3 +- drivers/net/ethernet/intel/igb/igb_main.c | 20 ++++++--- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 +- drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c | 13 ++++-- drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h | 3 +- drivers/net/ethernet/mellanox/mlx4/cmd.c | 11 ++--- drivers/net/ethernet/qlogic/qlcnic/qlcnic.h | 1 + drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c | 2 +- drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h | 2 +- .../ethernet/qlogic/qlcnic/qlcnic_sriov_common.c | 1 + .../net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c | 52 +++++++++++++++------- drivers/net/ethernet/sfc/siena_sriov.c | 3 +- include/linux/if_link.h | 3 +- include/linux/netdevice.h | 8 ++-- include/uapi/linux/if_link.h | 9 +++- net/core/rtnetlink.c | 38 +++++++++++++--- 20 files changed, 156 insertions(+), 67 deletions(-) (limited to 'include/uapi') diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c index 81cc2d9831c2..8d0479d5be8e 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c @@ -2576,7 +2576,8 @@ int bnx2x_get_vf_config(struct net_device *dev, int vfidx, ivi->vf = vfidx; ivi->qos = 0; - ivi->tx_rate = 10000; /* always 10G. TBA take from link struct */ + ivi->max_tx_rate = 10000; /* always 10G. TBA take from link struct */ + ivi->min_tx_rate = 0; ivi->spoofchk = 1; /*always enabled */ if (vf->state == VF_ENABLED) { /* mac and vlan are in vlan_mac objects */ diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index dcc5e5c69743..4693d004a223 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1302,7 +1302,8 @@ static int be_get_vf_config(struct net_device *netdev, int vf, return -EINVAL; vi->vf = vf; - vi->tx_rate = vf_cfg->tx_rate; + vi->max_tx_rate = vf_cfg->tx_rate; + vi->min_tx_rate = 0; vi->vlan = vf_cfg->vlan_tag & VLAN_VID_MASK; vi->qos = vf_cfg->vlan_tag >> VLAN_PRIO_SHIFT; memcpy(&vi->mac, vf_cfg->mac_addr, ETH_ALEN); @@ -1342,7 +1343,8 @@ static int be_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos) return status; } -static int be_set_vf_tx_rate(struct net_device *netdev, int vf, int rate) +static int be_set_vf_tx_rate(struct net_device *netdev, int vf, + int min_tx_rate, int max_tx_rate) { struct be_adapter *adapter = netdev_priv(netdev); int status = 0; @@ -1353,18 +1355,21 @@ static int be_set_vf_tx_rate(struct net_device *netdev, int vf, int rate) if (vf >= adapter->num_vfs) return -EINVAL; - if (rate < 100 || rate > 10000) { + if (min_tx_rate) + return -EINVAL; + + if (max_tx_rate < 100 || max_tx_rate > 10000) { dev_err(&adapter->pdev->dev, - "tx rate must be between 100 and 10000 Mbps\n"); + "max tx rate must be between 100 and 10000 Mbps\n"); return -EINVAL; } - status = be_cmd_config_qos(adapter, rate / 10, vf + 1); + status = be_cmd_config_qos(adapter, max_tx_rate / 10, vf + 1); if (status) dev_err(&adapter->pdev->dev, - "tx rate %d on VF %d failed\n", rate, vf); + "max tx rate %d on VF %d failed\n", max_tx_rate, vf); else - adapter->vf_cfg[vf].tx_rate = rate; + adapter->vf_cfg[vf].tx_rate = max_tx_rate; return status; } static int be_set_vf_link_state(struct net_device *netdev, int vf, @@ -4257,7 +4262,7 @@ static const struct net_device_ops be_netdev_ops = { .ndo_vlan_rx_kill_vid = be_vlan_rem_vid, .ndo_set_vf_mac = be_set_vf_mac, .ndo_set_vf_vlan = be_set_vf_vlan, - .ndo_set_vf_tx_rate = be_set_vf_tx_rate, + .ndo_set_vf_rate = be_set_vf_tx_rate, .ndo_get_vf_config = be_get_vf_config, .ndo_set_vf_link_state = be_set_vf_link_state, #ifdef CONFIG_NET_POLL_CONTROLLER diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index e0e5c6a867b1..96f7fabd8758 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -6724,7 +6724,7 @@ static const struct net_device_ops i40e_netdev_ops = { .ndo_set_features = i40e_set_features, .ndo_set_vf_mac = i40e_ndo_set_vf_mac, .ndo_set_vf_vlan = i40e_ndo_set_vf_port_vlan, - .ndo_set_vf_tx_rate = i40e_ndo_set_vf_bw, + .ndo_set_vf_rate = i40e_ndo_set_vf_bw, .ndo_get_vf_config = i40e_ndo_get_vf_config, .ndo_set_vf_link_state = i40e_ndo_set_vf_link_state, #ifdef CONFIG_I40E_VXLAN diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 4d219566a04d..8564b0939dc4 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -2205,7 +2205,8 @@ error_pvid: * * configure vf tx rate **/ -int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate) +int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate, + int max_tx_rate) { struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_pf *pf = np->vsi->back; @@ -2221,6 +2222,12 @@ int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate) goto error; } + if (min_tx_rate) { + dev_err(&pf->pdev->dev, "Invalid min tx rate (%d) (greater than 0) specified for vf %d.\n", + min_tx_rate, vf_id); + return -EINVAL; + } + vf = &(pf->vf[vf_id]); vsi = pf->vsi[vf->lan_vsi_index]; if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states)) { @@ -2243,23 +2250,23 @@ int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate) break; } - if (tx_rate > speed) { - dev_err(&pf->pdev->dev, "Invalid tx rate %d specified for vf %d.", - tx_rate, vf->vf_id); + if (max_tx_rate > speed) { + dev_err(&pf->pdev->dev, "Invalid max tx rate %d specified for vf %d.", + max_tx_rate, vf->vf_id); ret = -EINVAL; goto error; } /* Tx rate credits are in values of 50Mbps, 0 is disabled*/ - ret = i40e_aq_config_vsi_bw_limit(&pf->hw, vsi->seid, tx_rate / 50, 0, - NULL); + ret = i40e_aq_config_vsi_bw_limit(&pf->hw, vsi->seid, max_tx_rate / 50, + 0, NULL); if (ret) { - dev_err(&pf->pdev->dev, "Unable to set tx rate, error code %d.\n", + dev_err(&pf->pdev->dev, "Unable to set max tx rate, error code %d.\n", ret); ret = -EIO; goto error; } - vf->tx_rate = tx_rate; + vf->tx_rate = max_tx_rate; error: return ret; } @@ -2301,7 +2308,8 @@ int i40e_ndo_get_vf_config(struct net_device *netdev, memcpy(&ivi->mac, vf->default_lan_addr.addr, ETH_ALEN); - ivi->tx_rate = vf->tx_rate; + ivi->max_tx_rate = vf->tx_rate; + ivi->min_tx_rate = 0; ivi->vlan = le16_to_cpu(vsi->info.pvid) & I40E_VLAN_MASK; ivi->qos = (le16_to_cpu(vsi->info.pvid) & I40E_PRIORITY_MASK) >> I40E_VLAN_PRIORITY_SHIFT; diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h index ba3d1f8414be..5a559be4ba2c 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h @@ -116,7 +116,8 @@ void i40e_vc_notify_vf_reset(struct i40e_vf *vf); int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac); int i40e_ndo_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos); -int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int tx_rate); +int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate, + int max_tx_rate); int i40e_ndo_get_vf_config(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi); int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link); diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index bfcda8a455f4..1075b3f8c415 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -169,7 +169,7 @@ static void igb_restore_vf_multicasts(struct igb_adapter *adapter); static int igb_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac); static int igb_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos); -static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate); +static int igb_ndo_set_vf_bw(struct net_device *, int, int, int); static int igb_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting); static int igb_ndo_get_vf_config(struct net_device *netdev, int vf, @@ -2084,7 +2084,7 @@ static const struct net_device_ops igb_netdev_ops = { .ndo_vlan_rx_kill_vid = igb_vlan_rx_kill_vid, .ndo_set_vf_mac = igb_ndo_set_vf_mac, .ndo_set_vf_vlan = igb_ndo_set_vf_vlan, - .ndo_set_vf_tx_rate = igb_ndo_set_vf_bw, + .ndo_set_vf_rate = igb_ndo_set_vf_bw, .ndo_set_vf_spoofchk = igb_ndo_set_vf_spoofchk, .ndo_get_vf_config = igb_ndo_get_vf_config, #ifdef CONFIG_NET_POLL_CONTROLLER @@ -7879,7 +7879,8 @@ static void igb_check_vf_rate_limit(struct igb_adapter *adapter) } } -static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate) +static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, + int min_tx_rate, int max_tx_rate) { struct igb_adapter *adapter = netdev_priv(netdev); struct e1000_hw *hw = &adapter->hw; @@ -7888,15 +7889,19 @@ static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate) if (hw->mac.type != e1000_82576) return -EOPNOTSUPP; + if (min_tx_rate) + return -EINVAL; + actual_link_speed = igb_link_mbps(adapter->link_speed); if ((vf >= adapter->vfs_allocated_count) || (!(rd32(E1000_STATUS) & E1000_STATUS_LU)) || - (tx_rate < 0) || (tx_rate > actual_link_speed)) + (max_tx_rate < 0) || + (max_tx_rate > actual_link_speed)) return -EINVAL; adapter->vf_rate_link_speed = actual_link_speed; - adapter->vf_data[vf].tx_rate = (u16)tx_rate; - igb_set_vf_rate_limit(hw, vf, tx_rate, actual_link_speed); + adapter->vf_data[vf].tx_rate = (u16)max_tx_rate; + igb_set_vf_rate_limit(hw, vf, max_tx_rate, actual_link_speed); return 0; } @@ -7936,7 +7941,8 @@ static int igb_ndo_get_vf_config(struct net_device *netdev, return -EINVAL; ivi->vf = vf; memcpy(&ivi->mac, adapter->vf_data[vf].vf_mac_addresses, ETH_ALEN); - ivi->tx_rate = adapter->vf_data[vf].tx_rate; + ivi->max_tx_rate = adapter->vf_data[vf].tx_rate; + ivi->min_tx_rate = 0; ivi->vlan = adapter->vf_data[vf].pf_vlan; ivi->qos = adapter->vf_data[vf].pf_qos; ivi->spoofchk = adapter->vf_data[vf].spoofchk_enabled; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 8089ea9f2fba..a5332389620a 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -7926,7 +7926,7 @@ static const struct net_device_ops ixgbe_netdev_ops = { .ndo_do_ioctl = ixgbe_ioctl, .ndo_set_vf_mac = ixgbe_ndo_set_vf_mac, .ndo_set_vf_vlan = ixgbe_ndo_set_vf_vlan, - .ndo_set_vf_tx_rate = ixgbe_ndo_set_vf_bw, + .ndo_set_vf_rate = ixgbe_ndo_set_vf_bw, .ndo_set_vf_spoofchk = ixgbe_ndo_set_vf_spoofchk, .ndo_get_vf_config = ixgbe_ndo_get_vf_config, .ndo_get_stats64 = ixgbe_get_stats64, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c index a01417c06620..3248e208c9dc 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c @@ -1222,7 +1222,8 @@ void ixgbe_check_vf_rate_limit(struct ixgbe_adapter *adapter) } } -int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate) +int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int min_tx_rate, + int max_tx_rate) { struct ixgbe_adapter *adapter = netdev_priv(netdev); int link_speed; @@ -1240,13 +1241,16 @@ int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate) if (link_speed != 10000) return -EINVAL; + if (min_tx_rate) + return -EINVAL; + /* rate limit cannot be less than 10Mbs or greater than link speed */ - if (tx_rate && ((tx_rate <= 10) || (tx_rate > link_speed))) + if (max_tx_rate && ((max_tx_rate <= 10) || (max_tx_rate > link_speed))) return -EINVAL; /* store values */ adapter->vf_rate_link_speed = link_speed; - adapter->vfinfo[vf].tx_rate = tx_rate; + adapter->vfinfo[vf].tx_rate = max_tx_rate; /* update hardware configuration */ ixgbe_set_vf_rate_limit(adapter, vf); @@ -1288,7 +1292,8 @@ int ixgbe_ndo_get_vf_config(struct net_device *netdev, return -EINVAL; ivi->vf = vf; memcpy(&ivi->mac, adapter->vfinfo[vf].vf_mac_addresses, ETH_ALEN); - ivi->tx_rate = adapter->vfinfo[vf].tx_rate; + ivi->max_tx_rate = adapter->vfinfo[vf].tx_rate; + ivi->min_tx_rate = 0; ivi->vlan = adapter->vfinfo[vf].pf_vlan; ivi->qos = adapter->vfinfo[vf].pf_qos; ivi->spoofchk = adapter->vfinfo[vf].spoofchk_enabled; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h index cea640147604..32c26d586c01 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h @@ -44,7 +44,8 @@ void ixgbe_ping_all_vfs(struct ixgbe_adapter *adapter); int ixgbe_ndo_set_vf_mac(struct net_device *netdev, int queue, u8 *mac); int ixgbe_ndo_set_vf_vlan(struct net_device *netdev, int queue, u16 vlan, u8 qos); -int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate); +int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int min_tx_rate, + int max_tx_rate); int ixgbe_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting); int ixgbe_ndo_get_vf_config(struct net_device *netdev, int vf, struct ifla_vf_info *ivi); diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 59c7fd406805..ca8e7cb5a8e4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -2486,11 +2486,12 @@ int mlx4_get_vf_config(struct mlx4_dev *dev, int port, int vf, struct ifla_vf_in ivf->mac[4] = ((s_info->mac >> (1*8)) & 0xff); ivf->mac[5] = ((s_info->mac) & 0xff); - ivf->vlan = s_info->default_vlan; - ivf->qos = s_info->default_qos; - ivf->tx_rate = s_info->tx_rate; - ivf->spoofchk = s_info->spoofchk; - ivf->linkstate = s_info->link_state; + ivf->vlan = s_info->default_vlan; + ivf->qos = s_info->default_qos; + ivf->max_tx_rate = s_info->tx_rate; + ivf->min_tx_rate = 0; + ivf->spoofchk = s_info->spoofchk; + ivf->linkstate = s_info->link_state; return 0; } diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index 6e7527e2b595..41abe6070466 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -1319,6 +1319,7 @@ struct qlcnic_eswitch { #define QL_STATUS_INVALID_PARAM -1 #define MAX_BW 100 /* % of link speed */ +#define MIN_BW 1 /* % of link speed */ #define MAX_VLAN_ID 4095 #define MIN_VLAN_ID 2 #define DEFAULT_MAC_LEARN 1 diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index f0a285359e66..f06ba90b4282 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -525,7 +525,7 @@ static const struct net_device_ops qlcnic_netdev_ops = { #endif #ifdef CONFIG_QLCNIC_SRIOV .ndo_set_vf_mac = qlcnic_sriov_set_vf_mac, - .ndo_set_vf_tx_rate = qlcnic_sriov_set_vf_tx_rate, + .ndo_set_vf_rate = qlcnic_sriov_set_vf_tx_rate, .ndo_get_vf_config = qlcnic_sriov_get_vf_config, .ndo_set_vf_vlan = qlcnic_sriov_set_vf_vlan, .ndo_set_vf_spoofchk = qlcnic_sriov_set_vf_spoofchk, diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h index 335b50f7bd3e..4677b2edccca 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov.h @@ -233,7 +233,7 @@ bool qlcnic_sriov_soft_flr_check(struct qlcnic_adapter *, void qlcnic_sriov_pf_reset(struct qlcnic_adapter *); int qlcnic_sriov_pf_reinit(struct qlcnic_adapter *); int qlcnic_sriov_set_vf_mac(struct net_device *, int, u8 *); -int qlcnic_sriov_set_vf_tx_rate(struct net_device *, int, int); +int qlcnic_sriov_set_vf_tx_rate(struct net_device *, int, int, int); int qlcnic_sriov_get_vf_config(struct net_device *, int , struct ifla_vf_info *); int qlcnic_sriov_set_vf_vlan(struct net_device *, int, u16, u8); diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c index 498fa6350c8d..2bdd9deffb38 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c @@ -201,6 +201,7 @@ int qlcnic_sriov_init(struct qlcnic_adapter *adapter, int num_vfs) sriov->vf_info[i].vp = vp; vp->vlan_mode = QLC_GUEST_VLAN_MODE; vp->max_tx_bw = MAX_BW; + vp->min_tx_bw = MIN_BW; vp->spoofchk = false; random_ether_addr(vp->mac); dev_info(&adapter->pdev->dev, diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c index 6d2f72f114f2..a29538b86edf 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c @@ -1848,7 +1848,8 @@ int qlcnic_sriov_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) return 0; } -int qlcnic_sriov_set_vf_tx_rate(struct net_device *netdev, int vf, int tx_rate) +int qlcnic_sriov_set_vf_tx_rate(struct net_device *netdev, int vf, + int min_tx_rate, int max_tx_rate) { struct qlcnic_adapter *adapter = netdev_priv(netdev); struct qlcnic_sriov *sriov = adapter->ahw->sriov; @@ -1863,35 +1864,52 @@ int qlcnic_sriov_set_vf_tx_rate(struct net_device *netdev, int vf, int tx_rate) if (vf >= sriov->num_vfs) return -EINVAL; - if (tx_rate >= 10000 || tx_rate < 100) { + vf_info = &sriov->vf_info[vf]; + vp = vf_info->vp; + vpid = vp->handle; + + if (!min_tx_rate) + min_tx_rate = QLC_VF_MIN_TX_RATE; + + if (max_tx_rate && + (max_tx_rate >= 10000 || max_tx_rate < min_tx_rate)) { netdev_err(netdev, - "Invalid Tx rate, allowed range is [%d - %d]", - QLC_VF_MIN_TX_RATE, QLC_VF_MAX_TX_RATE); + "Invalid max Tx rate, allowed range is [%d - %d]", + min_tx_rate, QLC_VF_MAX_TX_RATE); return -EINVAL; } - if (tx_rate == 0) - tx_rate = 10000; + if (!max_tx_rate) + max_tx_rate = 10000; - vf_info = &sriov->vf_info[vf]; - vp = vf_info->vp; - vpid = vp->handle; + if (min_tx_rate && + (min_tx_rate > max_tx_rate || min_tx_rate < QLC_VF_MIN_TX_RATE)) { + netdev_err(netdev, + "Invalid min Tx rate, allowed range is [%d - %d]", + QLC_VF_MIN_TX_RATE, max_tx_rate); + return -EINVAL; + } if (test_bit(QLC_BC_VF_STATE, &vf_info->state)) { if (qlcnic_sriov_get_vf_vport_info(adapter, &nic_info, vpid)) return -EIO; - nic_info.max_tx_bw = tx_rate / 100; + nic_info.max_tx_bw = max_tx_rate / 100; + nic_info.min_tx_bw = min_tx_rate / 100; nic_info.bit_offsets = BIT_0; if (qlcnic_sriov_pf_set_vport_info(adapter, &nic_info, vpid)) return -EIO; } - vp->max_tx_bw = tx_rate / 100; + vp->max_tx_bw = max_tx_rate / 100; + netdev_info(netdev, + "Setting Max Tx rate %d (Mbps), %d %% of PF bandwidth, for VF %d\n", + max_tx_rate, vp->max_tx_bw, vf); + vp->min_tx_bw = min_tx_rate / 100; netdev_info(netdev, - "Setting Tx rate %d (Mbps), %d %% of PF bandwidth, for VF %d\n", - tx_rate, vp->max_tx_bw, vf); + "Setting Min Tx rate %d (Mbps), %d %% of PF bandwidth, for VF %d\n", + min_tx_rate, vp->min_tx_bw, vf); return 0; } @@ -1990,9 +2008,13 @@ int qlcnic_sriov_get_vf_config(struct net_device *netdev, ivi->qos = vp->qos; ivi->spoofchk = vp->spoofchk; if (vp->max_tx_bw == MAX_BW) - ivi->tx_rate = 0; + ivi->max_tx_rate = 0; + else + ivi->max_tx_rate = vp->max_tx_bw * 100; + if (vp->min_tx_bw == MIN_BW) + ivi->min_tx_rate = 0; else - ivi->tx_rate = vp->max_tx_bw * 100; + ivi->min_tx_rate = vp->min_tx_bw * 100; ivi->vf = vf; return 0; diff --git a/drivers/net/ethernet/sfc/siena_sriov.c b/drivers/net/ethernet/sfc/siena_sriov.c index 9a9205e77896..43d2e64546ed 100644 --- a/drivers/net/ethernet/sfc/siena_sriov.c +++ b/drivers/net/ethernet/sfc/siena_sriov.c @@ -1633,7 +1633,8 @@ int efx_sriov_get_vf_config(struct net_device *net_dev, int vf_i, ivi->vf = vf_i; ether_addr_copy(ivi->mac, vf->addr.mac_addr); - ivi->tx_rate = 0; + ivi->max_tx_rate = 0; + ivi->min_tx_rate = 0; tci = ntohs(vf->addr.tci); ivi->vlan = tci & VLAN_VID_MASK; ivi->qos = (tci >> VLAN_PRIO_SHIFT) & 0x7; diff --git a/include/linux/if_link.h b/include/linux/if_link.h index a86784dec3d3..119130e9298b 100644 --- a/include/linux/if_link.h +++ b/include/linux/if_link.h @@ -10,8 +10,9 @@ struct ifla_vf_info { __u8 mac[32]; __u32 vlan; __u32 qos; - __u32 tx_rate; __u32 spoofchk; __u32 linkstate; + __u32 min_tx_rate; + __u32 max_tx_rate; }; #endif /* _LINUX_IF_LINK_H */ diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f4ad247fd324..9ec3a945caf2 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -850,7 +850,8 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev, * SR-IOV management functions. * int (*ndo_set_vf_mac)(struct net_device *dev, int vf, u8* mac); * int (*ndo_set_vf_vlan)(struct net_device *dev, int vf, u16 vlan, u8 qos); - * int (*ndo_set_vf_tx_rate)(struct net_device *dev, int vf, int rate); + * int (*ndo_set_vf_rate)(struct net_device *dev, int vf, int min_tx_rate, + * int max_tx_rate); * int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting); * int (*ndo_get_vf_config)(struct net_device *dev, * int vf, struct ifla_vf_info *ivf); @@ -1044,8 +1045,9 @@ struct net_device_ops { int queue, u8 *mac); int (*ndo_set_vf_vlan)(struct net_device *dev, int queue, u16 vlan, u8 qos); - int (*ndo_set_vf_tx_rate)(struct net_device *dev, - int vf, int rate); + int (*ndo_set_vf_rate)(struct net_device *dev, + int vf, int min_tx_rate, + int max_tx_rate); int (*ndo_set_vf_spoofchk)(struct net_device *dev, int vf, bool setting); int (*ndo_get_vf_config)(struct net_device *dev, diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 9a7f7ace6649..622e7910b8cc 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -399,9 +399,10 @@ enum { IFLA_VF_UNSPEC, IFLA_VF_MAC, /* Hardware queue specific attributes */ IFLA_VF_VLAN, - IFLA_VF_TX_RATE, /* TX Bandwidth Allocation */ + IFLA_VF_TX_RATE, /* Max TX Bandwidth Allocation */ IFLA_VF_SPOOFCHK, /* Spoof Checking on/off switch */ IFLA_VF_LINK_STATE, /* link state enable/disable/auto switch */ + IFLA_VF_RATE, /* Min and Max TX Bandwidth Allocation */ __IFLA_VF_MAX, }; @@ -423,6 +424,12 @@ struct ifla_vf_tx_rate { __u32 rate; /* Max TX bandwidth in Mbps, 0 disables throttling */ }; +struct ifla_vf_rate { + __u32 vf; + __u32 min_tx_rate; /* Min Bandwidth in Mbps */ + __u32 max_tx_rate; /* Max Bandwidth in Mbps */ +}; + struct ifla_vf_spoofchk { __u32 vf; __u32 setting; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 9837bebf93ce..d6417464dc66 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -767,8 +767,8 @@ static inline int rtnl_vfinfo_size(const struct net_device *dev, size += num_vfs * (nla_total_size(sizeof(struct ifla_vf_mac)) + nla_total_size(sizeof(struct ifla_vf_vlan)) + - nla_total_size(sizeof(struct ifla_vf_tx_rate)) + - nla_total_size(sizeof(struct ifla_vf_spoofchk))); + nla_total_size(sizeof(struct ifla_vf_spoofchk)) + + nla_total_size(sizeof(struct ifla_vf_rate))); return size; } else return 0; @@ -1034,6 +1034,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, struct ifla_vf_info ivi; struct ifla_vf_mac vf_mac; struct ifla_vf_vlan vf_vlan; + struct ifla_vf_rate vf_rate; struct ifla_vf_tx_rate vf_tx_rate; struct ifla_vf_spoofchk vf_spoofchk; struct ifla_vf_link_state vf_linkstate; @@ -1054,6 +1055,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, break; vf_mac.vf = vf_vlan.vf = + vf_rate.vf = vf_tx_rate.vf = vf_spoofchk.vf = vf_linkstate.vf = ivi.vf; @@ -1061,7 +1063,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, memcpy(vf_mac.mac, ivi.mac, sizeof(ivi.mac)); vf_vlan.vlan = ivi.vlan; vf_vlan.qos = ivi.qos; - vf_tx_rate.rate = ivi.tx_rate; + vf_tx_rate.rate = ivi.max_tx_rate; + vf_rate.min_tx_rate = ivi.min_tx_rate; + vf_rate.max_tx_rate = ivi.max_tx_rate; vf_spoofchk.setting = ivi.spoofchk; vf_linkstate.link_state = ivi.linkstate; vf = nla_nest_start(skb, IFLA_VF_INFO); @@ -1071,6 +1075,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, } if (nla_put(skb, IFLA_VF_MAC, sizeof(vf_mac), &vf_mac) || nla_put(skb, IFLA_VF_VLAN, sizeof(vf_vlan), &vf_vlan) || + nla_put(skb, IFLA_VF_RATE, sizeof(vf_rate), + &vf_rate) || nla_put(skb, IFLA_VF_TX_RATE, sizeof(vf_tx_rate), &vf_tx_rate) || nla_put(skb, IFLA_VF_SPOOFCHK, sizeof(vf_spoofchk), @@ -1177,6 +1183,8 @@ static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = { .len = sizeof(struct ifla_vf_tx_rate) }, [IFLA_VF_SPOOFCHK] = { .type = NLA_BINARY, .len = sizeof(struct ifla_vf_spoofchk) }, + [IFLA_VF_RATE] = { .type = NLA_BINARY, + .len = sizeof(struct ifla_vf_rate) }, }; static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = { @@ -1336,11 +1344,29 @@ static int do_setvfinfo(struct net_device *dev, struct nlattr *attr) } case IFLA_VF_TX_RATE: { struct ifla_vf_tx_rate *ivt; + struct ifla_vf_info ivf; ivt = nla_data(vf); err = -EOPNOTSUPP; - if (ops->ndo_set_vf_tx_rate) - err = ops->ndo_set_vf_tx_rate(dev, ivt->vf, - ivt->rate); + if (ops->ndo_get_vf_config) + err = ops->ndo_get_vf_config(dev, ivt->vf, + &ivf); + if (err) + break; + err = -EOPNOTSUPP; + if (ops->ndo_set_vf_rate) + err = ops->ndo_set_vf_rate(dev, ivt->vf, + ivf.min_tx_rate, + ivt->rate); + break; + } + case IFLA_VF_RATE: { + struct ifla_vf_rate *ivt; + ivt = nla_data(vf); + err = -EOPNOTSUPP; + if (ops->ndo_set_vf_rate) + err = ops->ndo_set_vf_rate(dev, ivt->vf, + ivt->min_tx_rate, + ivt->max_tx_rate); break; } case IFLA_VF_SPOOFCHK: { -- cgit v1.2.3 From 1c19448c9ba6545b80ded18488a64a7f3d8e6998 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Fri, 23 May 2014 08:47:32 -0700 Subject: net: Make enabling of zero UDP6 csums more restrictive RFC 6935 permits zero checksums to be used in IPv6 however this is recommended only for certain tunnel protocols, it does not make checksums completely optional like they are in IPv4. This patch restricts the use of IPv6 zero checksums that was previously intoduced. no_check6_tx and no_check6_rx have been added to control the use of checksums in UDP6 RX and TX path. The normal sk_no_check_{rx,tx} settings are not used (this avoids ambiguity when dealing with a dual stack socket). A helper function has been added (udp_set_no_check6) which can be called by tunnel impelmentations to all zero checksums (send on the socket, and accept them as valid). Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/linux/udp.h | 24 +++++++++++++++++++++++- include/uapi/linux/udp.h | 2 ++ net/ipv4/udp.c | 20 +++++++++++++++++++- net/ipv6/udp.c | 8 ++++---- 4 files changed, 48 insertions(+), 6 deletions(-) (limited to 'include/uapi') diff --git a/include/linux/udp.h b/include/linux/udp.h index 42278bbf7a88..247cfdcc4b08 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -47,7 +47,9 @@ struct udp_sock { #define udp_portaddr_node inet.sk.__sk_common.skc_portaddr_node int pending; /* Any pending frames ? */ unsigned int corkflag; /* Cork is required */ - __u16 encap_type; /* Is this an Encapsulation socket? */ + __u8 encap_type; /* Is this an Encapsulation socket? */ + unsigned char no_check6_tx:1,/* Send zero UDP6 checksums on TX? */ + no_check6_rx:1;/* Allow zero UDP6 checksums on RX? */ /* * Following member retains the information to create a UDP header * when the socket is uncorked. @@ -76,6 +78,26 @@ static inline struct udp_sock *udp_sk(const struct sock *sk) return (struct udp_sock *)sk; } +static inline void udp_set_no_check6_tx(struct sock *sk, bool val) +{ + udp_sk(sk)->no_check6_tx = val; +} + +static inline void udp_set_no_check6_rx(struct sock *sk, bool val) +{ + udp_sk(sk)->no_check6_rx = val; +} + +static inline bool udp_get_no_check6_tx(struct sock *sk) +{ + return udp_sk(sk)->no_check6_tx; +} + +static inline bool udp_get_no_check6_rx(struct sock *sk) +{ + return udp_sk(sk)->no_check6_rx; +} + #define udp_portaddr_for_each_entry(__sk, node, list) \ hlist_nulls_for_each_entry(__sk, node, list, __sk_common.skc_portaddr_node) diff --git a/include/uapi/linux/udp.h b/include/uapi/linux/udp.h index e2bcfd75a30d..16574ea18f0c 100644 --- a/include/uapi/linux/udp.h +++ b/include/uapi/linux/udp.h @@ -29,6 +29,8 @@ struct udphdr { /* UDP socket options */ #define UDP_CORK 1 /* Never send partially complete segments */ #define UDP_ENCAP 100 /* Set the socket to accept encapsulated packets */ +#define UDP_NO_CHECK6_TX 101 /* Disable sending checksum for UDP6X */ +#define UDP_NO_CHECK6_RX 102 /* Disable accpeting checksum for UDP6 */ /* UDP encapsulation types */ #define UDP_ENCAP_ESPINUDP_NON_IKE 1 /* draft-ietf-ipsec-nat-t-ike-00/01 */ diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 12c6175b29cd..e07d52b8617a 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1968,7 +1968,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, int (*push_pending_frames)(struct sock *)) { struct udp_sock *up = udp_sk(sk); - int val; + int val, valbool; int err = 0; int is_udplite = IS_UDPLITE(sk); @@ -1978,6 +1978,8 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, if (get_user(val, (int __user *)optval)) return -EFAULT; + valbool = val ? 1 : 0; + switch (optname) { case UDP_CORK: if (val != 0) { @@ -2007,6 +2009,14 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, } break; + case UDP_NO_CHECK6_TX: + up->no_check6_tx = valbool; + break; + + case UDP_NO_CHECK6_RX: + up->no_check6_rx = valbool; + break; + /* * UDP-Lite's partial checksum coverage (RFC 3828). */ @@ -2089,6 +2099,14 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname, val = up->encap_type; break; + case UDP_NO_CHECK6_TX: + val = up->no_check6_tx; + break; + + case UDP_NO_CHECK6_RX: + val = up->no_check6_rx; + break; + /* The following two cannot be changed on UDP sockets, the return is * always 0 (which corresponds to the full checksum coverage of UDP). */ case UDPLITE_SEND_CSCOV: diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index b8db453133aa..60325236446a 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -794,10 +794,10 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb, dif = inet6_iif(skb); sk = udp_v6_mcast_next(net, sk, uh->dest, daddr, uh->source, saddr, dif); while (sk) { - /* If zero checksum and sk_no_check is not on for + /* If zero checksum and no_check is not on for * the socket then skip it. */ - if (uh->check || sk->sk_no_check_rx) + if (uh->check || udp_sk(sk)->no_check6_rx) stack[count++] = sk; sk = udp_v6_mcast_next(net, sk_nulls_next(sk), uh->dest, daddr, @@ -887,7 +887,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (sk != NULL) { int ret; - if (!uh->check && !sk->sk_no_check_rx) { + if (!uh->check && !udp_sk(sk)->no_check6_rx) { sock_put(sk); udp6_csum_zero_error(skb); goto csum_error; @@ -1037,7 +1037,7 @@ static int udp_v6_push_pending_frames(struct sock *sk) if (is_udplite) csum = udplite_csum_outgoing(sk, skb); - else if (sk->sk_no_check_tx) { /* UDP csum disabled */ + else if (up->no_check6_tx) { /* UDP csum disabled */ skb->ip_summed = CHECKSUM_NONE; goto send; } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ -- cgit v1.2.3 From 6b649feafe10b293f4bd5a74aca95faf625ae525 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Fri, 23 May 2014 08:47:40 -0700 Subject: l2tp: Add support for zero IPv6 checksums Added new L2TP configuration options to allow TX and RX of zero checksums in IPv6. Default is not to use them. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- include/uapi/linux/l2tp.h | 2 ++ net/l2tp/l2tp_core.c | 9 ++++++++- net/l2tp/l2tp_core.h | 4 +++- net/l2tp/l2tp_netlink.c | 7 +++++++ 4 files changed, 20 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/l2tp.h b/include/uapi/linux/l2tp.h index 8adb68160327..21caa2631c20 100644 --- a/include/uapi/linux/l2tp.h +++ b/include/uapi/linux/l2tp.h @@ -124,6 +124,8 @@ enum { L2TP_ATTR_STATS, /* nested */ L2TP_ATTR_IP6_SADDR, /* struct in6_addr */ L2TP_ATTR_IP6_DADDR, /* struct in6_addr */ + L2TP_ATTR_UDP_ZERO_CSUM6_TX, /* u8 */ + L2TP_ATTR_UDP_ZERO_CSUM6_RX, /* u8 */ __L2TP_ATTR_MAX, }; diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index a1186105f537..379558014b60 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1102,7 +1102,9 @@ static void l2tp_xmit_ipv6_csum(struct sock *sk, struct sk_buff *skb, struct ipv6_pinfo *np = inet6_sk(sk); struct udphdr *uh = udp_hdr(skb); - if (!skb_dst(skb) || !skb_dst(skb)->dev || + if (udp_get_no_check6_tx(sk)) + skb->ip_summed = CHECKSUM_NONE; + else if (!skb_dst(skb) || !skb_dst(skb)->dev || !(skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) { __wsum csum = skb_checksum(skb, 0, udp_len, 0); skb->ip_summed = CHECKSUM_UNNECESSARY; @@ -1435,6 +1437,11 @@ static int l2tp_tunnel_sock_create(struct net *net, sizeof(udp6_addr), 0); if (err < 0) goto out; + + if (cfg->udp6_zero_tx_checksums) + udp_set_no_check6_tx(sock->sk, true); + if (cfg->udp6_zero_rx_checksums) + udp_set_no_check6_rx(sock->sk, true); } else #endif { diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h index 3f93ccd6ba97..68aa9ffd4ae4 100644 --- a/net/l2tp/l2tp_core.h +++ b/net/l2tp/l2tp_core.h @@ -162,7 +162,9 @@ struct l2tp_tunnel_cfg { #endif u16 local_udp_port; u16 peer_udp_port; - unsigned int use_udp_checksums:1; + unsigned int use_udp_checksums:1, + udp6_zero_tx_checksums:1, + udp6_zero_rx_checksums:1; }; struct l2tp_tunnel { diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index f3d331bdd706..0ac907adb2f4 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -161,6 +161,13 @@ static int l2tp_nl_cmd_tunnel_create(struct sk_buff *skb, struct genl_info *info cfg.peer_udp_port = nla_get_u16(info->attrs[L2TP_ATTR_UDP_DPORT]); if (info->attrs[L2TP_ATTR_UDP_CSUM]) cfg.use_udp_checksums = nla_get_flag(info->attrs[L2TP_ATTR_UDP_CSUM]); + +#if IS_ENABLED(CONFIG_IPV6) + if (info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_TX]) + cfg.udp6_zero_tx_checksums = nla_get_flag(info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_TX]); + if (info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_RX]) + cfg.udp6_zero_rx_checksums = nla_get_flag(info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_RX]); +#endif } if (info->attrs[L2TP_ATTR_DEBUG]) -- cgit v1.2.3 From 182407a6ed5333fc37dd980a8de91a8f826a94f6 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Fri, 2 May 2014 11:09:54 +1000 Subject: drm: add DP MST encoder type This adds an encoder type for DP MST encoders. Reviewed-by: Todd Previte Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_crtc.c | 1 + include/uapi/drm/drm_mode.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 37a3e0791ddf..edee61bccd14 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -227,6 +227,7 @@ static const struct drm_prop_enum_list drm_encoder_enum_list[] = { DRM_MODE_ENCODER_TVDAC, "TV" }, { DRM_MODE_ENCODER_VIRTUAL, "Virtual" }, { DRM_MODE_ENCODER_DSI, "DSI" }, + { DRM_MODE_ENCODER_DPMST, "DP MST" }, }; static const struct drm_prop_enum_list drm_subpixel_enum_list[] = diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h index f104c2603ebe..719add464f95 100644 --- a/include/uapi/drm/drm_mode.h +++ b/include/uapi/drm/drm_mode.h @@ -181,6 +181,7 @@ struct drm_mode_get_plane_res { #define DRM_MODE_ENCODER_TVDAC 4 #define DRM_MODE_ENCODER_VIRTUAL 5 #define DRM_MODE_ENCODER_DSI 6 +#define DRM_MODE_ENCODER_DPMST 7 struct drm_mode_get_encoder { __u32 encoder_id; -- cgit v1.2.3 From 90dfdc7ceb577c0d7b7635def3c62039a091e50d Mon Sep 17 00:00:00 2001 From: David Daney Date: Wed, 28 May 2014 23:52:12 +0200 Subject: MIPS: Add functions for hypervisor call Introduce kvm_hypercall[0-3]. Define three new hypercalls for MIPS: GET_CLOCK_FREQ, EXIT_VM, and CONSOLE_OUTPUT. [andreas.herrmann: * Properly define hypercalls and HC numbers for MIPS in kvm_para.h header files] Signed-off-by: David Daney Signed-off-by: Andreas Herrmann Cc: linux-mips@linux-mips.org Cc: James Hogan Cc: kvm@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/7005/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/kvm_para.h | 109 ++++++++++++++++++++++++++++++++++ arch/mips/include/uapi/asm/kvm_para.h | 6 +- include/uapi/linux/kvm_para.h | 3 + 3 files changed, 117 insertions(+), 1 deletion(-) create mode 100644 arch/mips/include/asm/kvm_para.h (limited to 'include/uapi') diff --git a/arch/mips/include/asm/kvm_para.h b/arch/mips/include/asm/kvm_para.h new file mode 100644 index 000000000000..5a9aa918abe6 --- /dev/null +++ b/arch/mips/include/asm/kvm_para.h @@ -0,0 +1,109 @@ +#ifndef _ASM_MIPS_KVM_PARA_H +#define _ASM_MIPS_KVM_PARA_H + +#include + +#define KVM_HYPERCALL ".word 0x42000028" + +/* + * Hypercalls for KVM. + * + * Hypercall number is passed in v0. + * Return value will be placed in v0. + * Up to 3 arguments are passed in a0, a1, and a2. + */ +static inline unsigned long kvm_hypercall0(unsigned long num) +{ + register unsigned long n asm("v0"); + register unsigned long r asm("v0"); + + n = num; + __asm__ __volatile__( + KVM_HYPERCALL + : "=r" (r) : "r" (n) : "memory" + ); + + return r; +} + +static inline unsigned long kvm_hypercall1(unsigned long num, + unsigned long arg0) +{ + register unsigned long n asm("v0"); + register unsigned long r asm("v0"); + register unsigned long a0 asm("a0"); + + n = num; + a0 = arg0; + __asm__ __volatile__( + KVM_HYPERCALL + : "=r" (r) : "r" (n), "r" (a0) : "memory" + ); + + return r; +} + +static inline unsigned long kvm_hypercall2(unsigned long num, + unsigned long arg0, unsigned long arg1) +{ + register unsigned long n asm("v0"); + register unsigned long r asm("v0"); + register unsigned long a0 asm("a0"); + register unsigned long a1 asm("a1"); + + n = num; + a0 = arg0; + a1 = arg1; + __asm__ __volatile__( + KVM_HYPERCALL + : "=r" (r) : "r" (n), "r" (a0), "r" (a1) : "memory" + ); + + return r; +} + +static inline unsigned long kvm_hypercall3(unsigned long num, + unsigned long arg0, unsigned long arg1, unsigned long arg2) +{ + register unsigned long n asm("v0"); + register unsigned long r asm("v0"); + register unsigned long a0 asm("a0"); + register unsigned long a1 asm("a1"); + register unsigned long a2 asm("a2"); + + n = num; + a0 = arg0; + a1 = arg1; + a2 = arg2; + __asm__ __volatile__( + KVM_HYPERCALL + : "=r" (r) : "r" (n), "r" (a0), "r" (a1), "r" (a2) : "memory" + ); + + return r; +} + +static inline bool kvm_check_and_clear_guest_paused(void) +{ + return false; +} + +static inline unsigned int kvm_arch_para_features(void) +{ + return 0; +} + +#ifdef CONFIG_MIPS_PARAVIRT +static inline bool kvm_para_available(void) +{ + return true; +} +#else +static inline bool kvm_para_available(void) +{ + return false; +} +#endif + + +#endif /* _ASM_MIPS_KVM_PARA_H */ diff --git a/arch/mips/include/uapi/asm/kvm_para.h b/arch/mips/include/uapi/asm/kvm_para.h index 14fab8f0b957..7e16d7c42e65 100644 --- a/arch/mips/include/uapi/asm/kvm_para.h +++ b/arch/mips/include/uapi/asm/kvm_para.h @@ -1 +1,5 @@ -#include +#ifndef _UAPI_ASM_MIPS_KVM_PARA_H +#define _UAPI_ASM_MIPS_KVM_PARA_H + + +#endif /* _UAPI_ASM_MIPS_KVM_PARA_H */ diff --git a/include/uapi/linux/kvm_para.h b/include/uapi/linux/kvm_para.h index 2841f86eae0b..bf6cd7d5cac2 100644 --- a/include/uapi/linux/kvm_para.h +++ b/include/uapi/linux/kvm_para.h @@ -20,6 +20,9 @@ #define KVM_HC_FEATURES 3 #define KVM_HC_PPC_MAP_MAGIC_PAGE 4 #define KVM_HC_KICK_CPU 5 +#define KVM_HC_MIPS_GET_CLOCK_FREQ 6 +#define KVM_HC_MIPS_EXIT_VM 7 +#define KVM_HC_MIPS_CONSOLE_OUTPUT 8 /* * hypercalls use architecture specific -- cgit v1.2.3 From 41c389d72cf0756957450c25c1dbc7d026324df8 Mon Sep 17 00:00:00 2001 From: Roopa Prabhu Date: Tue, 27 May 2014 22:39:37 -0700 Subject: bridge: Add bridge ifindex to bridge fdb notify msgs (This patch was previously posted as RFC at http://patchwork.ozlabs.org/patch/352677/) This patch adds NDA_MASTER attribute to neighbour attributes enum for bridge/master ifindex. And adds NDA_MASTER to bridge fdb notify msgs. Today bridge fdb notifications dont contain bridge information. Userspace can derive it from the port information in the fdb notification. However this is tricky in some scenarious. Example, bridge port delete notification comes before bridge fdb delete notifications. And we have seen problems in userspace when using libnl where, the bridge fdb delete notification handling code does not understand which bridge this fdb entry is part of because the bridge and port association has already been deleted. And these notifications (port membership and fdb) are generated on separate rtnl groups. Fixing the order of notifications could possibly solve the problem for some cases (I can submit a separate patch for that). This patch chooses to add NDA_MASTER to bridge fdb notify msgs because it not only solves the problem described above, but also helps userspace avoid another lookup into link msgs to derive the master index. Signed-off-by: Roopa Prabhu Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/uapi/linux/neighbour.h | 1 + net/bridge/br_fdb.c | 3 +++ 2 files changed, 4 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/neighbour.h b/include/uapi/linux/neighbour.h index d3ef583104e0..4a1d7e96dfe3 100644 --- a/include/uapi/linux/neighbour.h +++ b/include/uapi/linux/neighbour.h @@ -24,6 +24,7 @@ enum { NDA_PORT, NDA_VNI, NDA_IFINDEX, + NDA_MASTER, __NDA_MAX }; diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 648d0e849595..2c45c069ea1a 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -616,6 +616,8 @@ static int fdb_fill_info(struct sk_buff *skb, const struct net_bridge *br, if (nla_put(skb, NDA_LLADDR, ETH_ALEN, &fdb->addr)) goto nla_put_failure; + if (nla_put_u32(skb, NDA_MASTER, br->dev->ifindex)) + goto nla_put_failure; ci.ndm_used = jiffies_to_clock_t(now - fdb->used); ci.ndm_confirmed = 0; ci.ndm_updated = jiffies_to_clock_t(now - fdb->updated); @@ -637,6 +639,7 @@ static inline size_t fdb_nlmsg_size(void) { return NLMSG_ALIGN(sizeof(struct ndmsg)) + nla_total_size(ETH_ALEN) /* NDA_LLADDR */ + + nla_total_size(sizeof(u32)) /* NDA_MASTER */ + nla_total_size(sizeof(u16)) /* NDA_VLAN */ + nla_total_size(sizeof(struct nda_cacheinfo)); } -- cgit v1.2.3 From 23372af15e638bf3ce0764554db3b5e58bf7ceb8 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 9 May 2014 22:45:08 -0400 Subject: NVMe: Update data structures for NVMe 1.2 Include changes from the current set of ratified Technical Proposals for NVMe 1.2. Signed-off-by: Matthew Wilcox --- include/uapi/linux/nvme.h | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/nvme.h b/include/uapi/linux/nvme.h index e74da782d69c..a6fb2a360577 100644 --- a/include/uapi/linux/nvme.h +++ b/include/uapi/linux/nvme.h @@ -27,7 +27,12 @@ struct nvme_id_power_state { __u8 read_lat; __u8 write_tput; __u8 write_lat; - __u8 rsvd16[16]; + __le16 idle_power; + __u8 idle_scale; + __u8 rsvd19; + __le16 active_power; + __u8 active_work_scale; + __u8 rsvd23[9]; }; enum { @@ -46,7 +51,8 @@ struct nvme_id_ctrl { __u8 mic; __u8 mdts; __u16 cntlid; - __u8 rsvd80[176]; + __u32 ver; + __u8 rsvd84[172]; __le16 oacs; __u8 acl; __u8 aerl; @@ -56,7 +62,9 @@ struct nvme_id_ctrl { __u8 npss; __u8 avscc; __u8 apsta; - __u8 rsvd266[246]; + __le16 wctemp; + __le16 cctemp; + __u8 rsvd270[242]; __u8 sqes; __u8 cqes; __u8 rsvd514[2]; @@ -102,7 +110,12 @@ struct nvme_id_ns { __u8 dps; __u8 nmic; __u8 rescap; - __u8 rsvd32[88]; + __u8 fpi; + __u8 rsvd33; + __le16 nawun; + __le16 nawupf; + __le16 nacwu; + __u8 rsvd40[80]; __u8 eui64[8]; struct nvme_lbaf lbaf[16]; __u8 rsvd192[192]; @@ -134,7 +147,10 @@ struct nvme_smart_log { __u8 unsafe_shutdowns[16]; __u8 media_errors[16]; __u8 num_err_log_entries[16]; - __u8 rsvd192[320]; + __le32 warning_temp_time; + __le32 critical_comp_time; + __le16 temp_sensor[8]; + __u8 rsvd216[296]; }; enum { -- cgit v1.2.3 From 5ea22f24d77b511d68c4ecaf4e6fd5d6ab462b8f Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Fri, 30 May 2014 11:34:01 -0400 Subject: drm: add extended property types If we continue to use bitmask for type, we will quickly run out of room to add new types. Split this up so existing part of bitmask range continues to function as before, but reserve a chunk of the remaining space for an integer type-id. Wrap this all up in some type-check helpers to keep the backwards-compat uglyness contained. Signed-off-by: Rob Clark Reviewed-by: Daniel Vetter --- drivers/gpu/drm/drm_crtc.c | 26 +++++++++++++++++--------- include/drm/drm_crtc.h | 17 +++++++++++++++++ include/uapi/drm/drm_mode.h | 13 +++++++++++++ 3 files changed, 47 insertions(+), 9 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index c4b44be7d464..73f543af4924 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -3094,6 +3094,9 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags, } list_add_tail(&property->head, &dev->mode_config.property_list); + + WARN_ON(!drm_property_type_valid(property)); + return property; fail: kfree(property->values); @@ -3251,14 +3254,16 @@ int drm_property_add_enum(struct drm_property *property, int index, { struct drm_property_enum *prop_enum; - if (!(property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK))) + if (!(drm_property_type_is(property, DRM_MODE_PROP_ENUM) || + drm_property_type_is(property, DRM_MODE_PROP_BITMASK))) return -EINVAL; /* * Bitmask enum properties have the additional constraint of values * from 0 to 63 */ - if ((property->flags & DRM_MODE_PROP_BITMASK) && (value > 63)) + if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK) && + (value > 63)) return -EINVAL; if (!list_empty(&property->enum_blob_list)) { @@ -3439,10 +3444,11 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev, goto done; } - if (property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)) { + if (drm_property_type_is(property, DRM_MODE_PROP_ENUM) || + drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) { list_for_each_entry(prop_enum, &property->enum_blob_list, head) enum_count++; - } else if (property->flags & DRM_MODE_PROP_BLOB) { + } else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) { list_for_each_entry(prop_blob, &property->enum_blob_list, head) blob_count++; } @@ -3464,7 +3470,8 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev, } out_resp->count_values = value_count; - if (property->flags & (DRM_MODE_PROP_ENUM | DRM_MODE_PROP_BITMASK)) { + if (drm_property_type_is(property, DRM_MODE_PROP_ENUM) || + drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) { if ((out_resp->count_enum_blobs >= enum_count) && enum_count) { copied = 0; enum_ptr = (struct drm_mode_property_enum __user *)(unsigned long)out_resp->enum_blob_ptr; @@ -3486,7 +3493,7 @@ int drm_mode_getproperty_ioctl(struct drm_device *dev, out_resp->count_enum_blobs = enum_count; } - if (property->flags & DRM_MODE_PROP_BLOB) { + if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) { if ((out_resp->count_enum_blobs >= blob_count) && blob_count) { copied = 0; blob_id_ptr = (uint32_t __user *)(unsigned long)out_resp->enum_blob_ptr; @@ -3640,17 +3647,18 @@ static bool drm_property_change_is_valid(struct drm_property *property, { if (property->flags & DRM_MODE_PROP_IMMUTABLE) return false; - if (property->flags & DRM_MODE_PROP_RANGE) { + + if (drm_property_type_is(property, DRM_MODE_PROP_RANGE)) { if (value < property->values[0] || value > property->values[1]) return false; return true; - } else if (property->flags & DRM_MODE_PROP_BITMASK) { + } else if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) { int i; uint64_t valid_mask = 0; for (i = 0; i < property->num_values; i++) valid_mask |= (1ULL << property->values[i]); return !(value & ~valid_mask); - } else if (property->flags & DRM_MODE_PROP_BLOB) { + } else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) { /* Only the driver knows */ return true; } else { diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 90bd1a25e2c6..92f6ec2de86a 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -930,6 +930,23 @@ extern void drm_mode_config_cleanup(struct drm_device *dev); extern int drm_mode_connector_update_edid_property(struct drm_connector *connector, struct edid *edid); + +static inline bool drm_property_type_is(struct drm_property *property, + uint32_t type) +{ + /* instanceof for props.. handles extended type vs original types: */ + if (property->flags & DRM_MODE_PROP_EXTENDED_TYPE) + return (property->flags & DRM_MODE_PROP_EXTENDED_TYPE) == type; + return property->flags & type; +} + +static inline bool drm_property_type_valid(struct drm_property *property) +{ + if (property->flags & DRM_MODE_PROP_EXTENDED_TYPE) + return !(property->flags & DRM_MODE_PROP_LEGACY_TYPE); + return !!(property->flags & DRM_MODE_PROP_LEGACY_TYPE); +} + extern int drm_object_property_set_value(struct drm_mode_object *obj, struct drm_property *property, uint64_t val); diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h index 719add464f95..caeaa6f7ebde 100644 --- a/include/uapi/drm/drm_mode.h +++ b/include/uapi/drm/drm_mode.h @@ -252,6 +252,19 @@ struct drm_mode_get_connector { #define DRM_MODE_PROP_BLOB (1<<4) #define DRM_MODE_PROP_BITMASK (1<<5) /* bitmask of enumerated types */ +/* non-extended types: legacy bitmask, one bit per type: */ +#define DRM_MODE_PROP_LEGACY_TYPE ( \ + DRM_MODE_PROP_RANGE | \ + DRM_MODE_PROP_ENUM | \ + DRM_MODE_PROP_BLOB | \ + DRM_MODE_PROP_BITMASK) + +/* extended-types: rather than continue to consume a bit per type, + * grab a chunk of the bits to use as integer type id. + */ +#define DRM_MODE_PROP_EXTENDED_TYPE 0x0000ffc0 +#define DRM_MODE_PROP_TYPE(n) ((n) << 6) + struct drm_mode_property_enum { __u64 value; char name[DRM_PROP_NAME_LEN]; -- cgit v1.2.3 From 98f75de40e9d83c3a90d294b8fd25fa2874212a9 Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Fri, 30 May 2014 11:37:03 -0400 Subject: drm: add object property type An object property is an id (idr) for a drm mode object. This will allow a property to be used set/get a framebuffer, CRTC, etc. Signed-off-by: Rob Clark Reviewed-by: Daniel Vetter --- drivers/gpu/drm/drm_crtc.c | 61 +++++++++++++++++++++++++++++++++++++++------ include/drm/drm_crtc.h | 5 ++++ include/uapi/drm/drm_mode.h | 1 + 3 files changed, 60 insertions(+), 7 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 73f543af4924..f990d9f180f0 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -370,6 +370,21 @@ void drm_mode_object_put(struct drm_device *dev, mutex_unlock(&dev->mode_config.idr_mutex); } +static struct drm_mode_object *_object_find(struct drm_device *dev, + uint32_t id, uint32_t type) +{ + struct drm_mode_object *obj = NULL; + + mutex_lock(&dev->mode_config.idr_mutex); + obj = idr_find(&dev->mode_config.crtc_idr, id); + if (!obj || (type != DRM_MODE_OBJECT_ANY && obj->type != type) || + (obj->id != id)) + obj = NULL; + mutex_unlock(&dev->mode_config.idr_mutex); + + return obj; +} + /** * drm_mode_object_find - look up a drm object with static lifetime * @dev: drm device @@ -377,7 +392,9 @@ void drm_mode_object_put(struct drm_device *dev, * @type: type of the mode object * * Note that framebuffers cannot be looked up with this functions - since those - * are reference counted, they need special treatment. + * are reference counted, they need special treatment. Even with + * DRM_MODE_OBJECT_ANY (although that will simply return NULL + * rather than WARN_ON()). */ struct drm_mode_object *drm_mode_object_find(struct drm_device *dev, uint32_t id, uint32_t type) @@ -387,13 +404,10 @@ struct drm_mode_object *drm_mode_object_find(struct drm_device *dev, /* Framebuffers are reference counted and need their own lookup * function.*/ WARN_ON(type == DRM_MODE_OBJECT_FB); - - mutex_lock(&dev->mode_config.idr_mutex); - obj = idr_find(&dev->mode_config.crtc_idr, id); - if (!obj || (obj->type != type) || (obj->id != id)) + obj = _object_find(dev, id, type); + /* don't leak out unref'd fb's */ + if (obj && (obj->type == DRM_MODE_OBJECT_FB)) obj = NULL; - mutex_unlock(&dev->mode_config.idr_mutex); - return obj; } EXPORT_SYMBOL(drm_mode_object_find); @@ -3074,6 +3088,8 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags, if (!property) return NULL; + property->dev = dev; + if (num_values) { property->values = kzalloc(sizeof(uint64_t)*num_values, GFP_KERNEL); if (!property->values) @@ -3234,6 +3250,23 @@ struct drm_property *drm_property_create_range(struct drm_device *dev, int flags } EXPORT_SYMBOL(drm_property_create_range); +struct drm_property *drm_property_create_object(struct drm_device *dev, + int flags, const char *name, uint32_t type) +{ + struct drm_property *property; + + flags |= DRM_MODE_PROP_OBJECT; + + property = drm_property_create(dev, flags, name, 1); + if (!property) + return NULL; + + property->values[0] = type; + + return property; +} +EXPORT_SYMBOL(drm_property_create_object); + /** * drm_property_add_enum - add a possible value to an enumeration property * @property: enumeration property to change @@ -3661,6 +3694,20 @@ static bool drm_property_change_is_valid(struct drm_property *property, } else if (drm_property_type_is(property, DRM_MODE_PROP_BLOB)) { /* Only the driver knows */ return true; + } else if (drm_property_type_is(property, DRM_MODE_PROP_OBJECT)) { + struct drm_mode_object *obj; + /* a zero value for an object property translates to null: */ + if (value == 0) + return true; + /* + * NOTE: use _object_find() directly to bypass restriction on + * looking up refcnt'd objects (ie. fb's). For a refcnt'd + * object this could race against object finalization, so it + * simply tells us that the object *was* valid. Which is good + * enough. + */ + obj = _object_find(property->dev, value, property->values[0]); + return obj != NULL; } else { int i; for (i = 0; i < property->num_values; i++) diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 92f6ec2de86a..cb2e599f769f 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -50,6 +50,7 @@ struct drm_clip_rect; #define DRM_MODE_OBJECT_BLOB 0xbbbbbbbb #define DRM_MODE_OBJECT_PLANE 0xeeeeeeee #define DRM_MODE_OBJECT_BRIDGE 0xbdbdbdbd +#define DRM_MODE_OBJECT_ANY 0 struct drm_mode_object { uint32_t id; @@ -190,6 +191,7 @@ struct drm_property { char name[DRM_PROP_NAME_LEN]; uint32_t num_values; uint64_t *values; + struct drm_device *dev; struct list_head enum_blob_list; }; @@ -980,6 +982,8 @@ struct drm_property *drm_property_create_bitmask(struct drm_device *dev, struct drm_property *drm_property_create_range(struct drm_device *dev, int flags, const char *name, uint64_t min, uint64_t max); +struct drm_property *drm_property_create_object(struct drm_device *dev, + int flags, const char *name, uint32_t type); extern void drm_property_destroy(struct drm_device *dev, struct drm_property *property); extern int drm_property_add_enum(struct drm_property *property, int index, uint64_t value, const char *name); @@ -995,6 +999,7 @@ extern int drm_mode_crtc_set_gamma_size(struct drm_crtc *crtc, int gamma_size); extern struct drm_mode_object *drm_mode_object_find(struct drm_device *dev, uint32_t id, uint32_t type); + /* IOCTLs */ extern int drm_mode_getresources(struct drm_device *dev, void *data, struct drm_file *file_priv); diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h index caeaa6f7ebde..57f46348befe 100644 --- a/include/uapi/drm/drm_mode.h +++ b/include/uapi/drm/drm_mode.h @@ -264,6 +264,7 @@ struct drm_mode_get_connector { */ #define DRM_MODE_PROP_EXTENDED_TYPE 0x0000ffc0 #define DRM_MODE_PROP_TYPE(n) ((n) << 6) +#define DRM_MODE_PROP_OBJECT DRM_MODE_PROP_TYPE(1) struct drm_mode_property_enum { __u64 value; -- cgit v1.2.3 From ebc44cf386734374983922c6fc1ae8ac47f88bef Mon Sep 17 00:00:00 2001 From: Rob Clark Date: Wed, 12 Sep 2012 22:22:31 -0500 Subject: drm: add signed-range property type Like range, but values are signed. Signed-off-by: Rob Clark Reviewed-by: David Herrmann --- drivers/gpu/drm/drm_crtc.c | 45 +++++++++++++++++++++++++++++++++------------ include/drm/drm_crtc.h | 12 ++++++++++++ include/uapi/drm/drm_mode.h | 1 + 3 files changed, 46 insertions(+), 12 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index f990d9f180f0..6127073407e0 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -3213,6 +3213,22 @@ struct drm_property *drm_property_create_bitmask(struct drm_device *dev, } EXPORT_SYMBOL(drm_property_create_bitmask); +static struct drm_property *property_create_range(struct drm_device *dev, + int flags, const char *name, + uint64_t min, uint64_t max) +{ + struct drm_property *property; + + property = drm_property_create(dev, flags, name, 2); + if (!property) + return NULL; + + property->values[0] = min; + property->values[1] = max; + + return property; +} + /** * drm_property_create - create a new ranged property type * @dev: drm device @@ -3235,21 +3251,20 @@ struct drm_property *drm_property_create_range(struct drm_device *dev, int flags const char *name, uint64_t min, uint64_t max) { - struct drm_property *property; - - flags |= DRM_MODE_PROP_RANGE; - - property = drm_property_create(dev, flags, name, 2); - if (!property) - return NULL; - - property->values[0] = min; - property->values[1] = max; - - return property; + return property_create_range(dev, DRM_MODE_PROP_RANGE | flags, + name, min, max); } EXPORT_SYMBOL(drm_property_create_range); +struct drm_property *drm_property_create_signed_range(struct drm_device *dev, + int flags, const char *name, + int64_t min, int64_t max) +{ + return property_create_range(dev, DRM_MODE_PROP_SIGNED_RANGE | flags, + name, I642U64(min), I642U64(max)); +} +EXPORT_SYMBOL(drm_property_create_signed_range); + struct drm_property *drm_property_create_object(struct drm_device *dev, int flags, const char *name, uint32_t type) { @@ -3685,6 +3700,12 @@ static bool drm_property_change_is_valid(struct drm_property *property, if (value < property->values[0] || value > property->values[1]) return false; return true; + } else if (drm_property_type_is(property, DRM_MODE_PROP_SIGNED_RANGE)) { + int64_t svalue = U642I64(value); + if (svalue < U642I64(property->values[0]) || + svalue > U642I64(property->values[1])) + return false; + return true; } else if (drm_property_type_is(property, DRM_MODE_PROP_BITMASK)) { int i; uint64_t valid_mask = 0; diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index cb2e599f769f..849cfdccff09 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -65,6 +65,15 @@ struct drm_object_properties { uint64_t values[DRM_OBJECT_MAX_PROPERTY]; }; +static inline int64_t U642I64(uint64_t val) +{ + return (int64_t)*((int64_t *)&val); +} +static inline uint64_t I642U64(int64_t val) +{ + return (uint64_t)*((uint64_t *)&val); +} + enum drm_connector_force { DRM_FORCE_UNSPECIFIED, DRM_FORCE_OFF, @@ -982,6 +991,9 @@ struct drm_property *drm_property_create_bitmask(struct drm_device *dev, struct drm_property *drm_property_create_range(struct drm_device *dev, int flags, const char *name, uint64_t min, uint64_t max); +struct drm_property *drm_property_create_signed_range(struct drm_device *dev, + int flags, const char *name, + int64_t min, int64_t max); struct drm_property *drm_property_create_object(struct drm_device *dev, int flags, const char *name, uint32_t type); extern void drm_property_destroy(struct drm_device *dev, struct drm_property *property); diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h index 57f46348befe..def54f9e07ca 100644 --- a/include/uapi/drm/drm_mode.h +++ b/include/uapi/drm/drm_mode.h @@ -265,6 +265,7 @@ struct drm_mode_get_connector { #define DRM_MODE_PROP_EXTENDED_TYPE 0x0000ffc0 #define DRM_MODE_PROP_TYPE(n) ((n) << 6) #define DRM_MODE_PROP_OBJECT DRM_MODE_PROP_TYPE(1) +#define DRM_MODE_PROP_SIGNED_RANGE DRM_MODE_PROP_TYPE(2) struct drm_mode_property_enum { __u64 value; -- cgit v1.2.3 From 359a0ea9875ef4f32c8425bbe1ae348e1fd2ed2a Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Wed, 4 Jun 2014 17:20:29 -0700 Subject: vxlan: Add support for UDP checksums (v4 sending, v6 zero csums) Added VXLAN link configuration for sending UDP checksums, and allowing TX and RX of UDP6 checksums. Also, call common iptunnel_handle_offloads and added GSO support for checksums. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- drivers/net/vxlan.c | 120 +++++++++++++++++++++--------------------- include/net/vxlan.h | 12 ++++- include/uapi/linux/if_link.h | 3 ++ net/openvswitch/vport-vxlan.c | 2 +- 4 files changed, 74 insertions(+), 63 deletions(-) (limited to 'include/uapi') diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index e68c8eb4ea8e..4e2caaf8b5da 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -135,7 +135,7 @@ struct vxlan_dev { __u16 port_max; __u8 tos; /* TOS override */ __u8 ttl; - u32 flags; /* VXLAN_F_* below */ + u32 flags; /* VXLAN_F_* in vxlan.h */ struct work_struct sock_work; struct work_struct igmp_join; @@ -150,13 +150,6 @@ struct vxlan_dev { struct hlist_head fdb_head[FDB_HASH_SIZE]; }; -#define VXLAN_F_LEARN 0x01 -#define VXLAN_F_PROXY 0x02 -#define VXLAN_F_RSC 0x04 -#define VXLAN_F_L2MISS 0x08 -#define VXLAN_F_L3MISS 0x10 -#define VXLAN_F_IPV6 0x20 /* internal flag */ - /* salt for hash table */ static u32 vxlan_salt __read_mostly; static struct workqueue_struct *vxlan_wq; @@ -1601,18 +1594,11 @@ __be16 vxlan_src_port(__u16 port_min, __u16 port_max, struct sk_buff *skb) } EXPORT_SYMBOL_GPL(vxlan_src_port); -static int handle_offloads(struct sk_buff *skb) +static inline struct sk_buff *vxlan_handle_offloads(struct sk_buff *skb, + bool udp_csum) { - if (skb_is_gso(skb)) { - int err = skb_unclone(skb, GFP_ATOMIC); - if (unlikely(err)) - return err; - - skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL; - } else if (skb->ip_summed != CHECKSUM_PARTIAL) - skb->ip_summed = CHECKSUM_NONE; - - return 0; + int type = udp_csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL; + return iptunnel_handle_offloads(skb, udp_csum, type); } #if IS_ENABLED(CONFIG_IPV6) @@ -1629,10 +1615,9 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs, int min_headroom; int err; - if (!skb->encapsulation) { - skb_reset_inner_headers(skb); - skb->encapsulation = 1; - } + skb = vxlan_handle_offloads(skb, !udp_get_no_check6_tx(vs->sock->sk)); + if (IS_ERR(skb)) + return -EINVAL; skb_scrub_packet(skb, xnet); @@ -1666,27 +1651,14 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs, uh->source = src_port; uh->len = htons(skb->len); - uh->check = 0; memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED | IPSKB_REROUTED); skb_dst_set(skb, dst); - if (!skb_is_gso(skb) && !(dst->dev->features & NETIF_F_IPV6_CSUM)) { - __wsum csum = skb_checksum(skb, 0, skb->len, 0); - skb->ip_summed = CHECKSUM_UNNECESSARY; - uh->check = csum_ipv6_magic(saddr, daddr, skb->len, - IPPROTO_UDP, csum); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - } else { - skb->ip_summed = CHECKSUM_PARTIAL; - skb->csum_start = skb_transport_header(skb) - skb->head; - skb->csum_offset = offsetof(struct udphdr, check); - uh->check = ~csum_ipv6_magic(saddr, daddr, - skb->len, IPPROTO_UDP, 0); - } + udp6_set_csum(udp_get_no_check6_tx(vs->sock->sk), skb, + saddr, daddr, skb->len); __skb_push(skb, sizeof(*ip6h)); skb_reset_network_header(skb); @@ -1702,10 +1674,6 @@ static int vxlan6_xmit_skb(struct vxlan_sock *vs, ip6h->daddr = *daddr; ip6h->saddr = *saddr; - err = handle_offloads(skb); - if (err) - return err; - ip6tunnel_xmit(skb, dev); return 0; } @@ -1721,10 +1689,9 @@ int vxlan_xmit_skb(struct vxlan_sock *vs, int min_headroom; int err; - if (!skb->encapsulation) { - skb_reset_inner_headers(skb); - skb->encapsulation = 1; - } + skb = vxlan_handle_offloads(skb, !vs->sock->sk->sk_no_check_tx); + if (IS_ERR(skb)) + return -EINVAL; min_headroom = LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len + VXLAN_HLEN + sizeof(struct iphdr) @@ -1756,11 +1723,9 @@ int vxlan_xmit_skb(struct vxlan_sock *vs, uh->source = src_port; uh->len = htons(skb->len); - uh->check = 0; - err = handle_offloads(skb); - if (err) - return err; + udp_set_csum(vs->sock->sk->sk_no_check_tx, skb, + src, dst, skb->len); return iptunnel_xmit(vs->sock->sk, rt, skb, src, dst, IPPROTO_UDP, tos, ttl, df, xnet); @@ -2405,7 +2370,7 @@ static void vxlan_del_work(struct work_struct *work) * could be used for both IPv4 and IPv6 communications, but * users may set bindv6only=1. */ -static struct socket *create_v6_sock(struct net *net, __be16 port) +static struct socket *create_v6_sock(struct net *net, __be16 port, u32 flags) { struct sock *sk; struct socket *sock; @@ -2442,18 +2407,25 @@ static struct socket *create_v6_sock(struct net *net, __be16 port) /* Disable multicast loopback */ inet_sk(sk)->mc_loop = 0; + + if (flags & VXLAN_F_UDP_ZERO_CSUM6_TX) + udp_set_no_check6_tx(sk, true); + + if (flags & VXLAN_F_UDP_ZERO_CSUM6_RX) + udp_set_no_check6_rx(sk, true); + return sock; } #else -static struct socket *create_v6_sock(struct net *net, __be16 port) +static struct socket *create_v6_sock(struct net *net, __be16 port, u32 flags) { return ERR_PTR(-EPFNOSUPPORT); } #endif -static struct socket *create_v4_sock(struct net *net, __be16 port) +static struct socket *create_v4_sock(struct net *net, __be16 port, u32 flags) { struct sock *sk; struct socket *sock; @@ -2486,18 +2458,24 @@ static struct socket *create_v4_sock(struct net *net, __be16 port) /* Disable multicast loopback */ inet_sk(sk)->mc_loop = 0; + + if (!(flags & VXLAN_F_UDP_CSUM)) + sock->sk->sk_no_check_tx = 1; + return sock; } /* Create new listen socket if needed */ static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port, - vxlan_rcv_t *rcv, void *data, bool ipv6) + vxlan_rcv_t *rcv, void *data, + u32 flags) { struct vxlan_net *vn = net_generic(net, vxlan_net_id); struct vxlan_sock *vs; struct socket *sock; struct sock *sk; unsigned int h; + bool ipv6 = !!(flags & VXLAN_F_IPV6); vs = kzalloc(sizeof(*vs), GFP_KERNEL); if (!vs) @@ -2509,9 +2487,9 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port, INIT_WORK(&vs->del_work, vxlan_del_work); if (ipv6) - sock = create_v6_sock(net, port); + sock = create_v6_sock(net, port, flags); else - sock = create_v4_sock(net, port); + sock = create_v4_sock(net, port, flags); if (IS_ERR(sock)) { kfree(vs); return ERR_CAST(sock); @@ -2549,12 +2527,12 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, __be16 port, struct vxlan_sock *vxlan_sock_add(struct net *net, __be16 port, vxlan_rcv_t *rcv, void *data, - bool no_share, bool ipv6) + bool no_share, u32 flags) { struct vxlan_net *vn = net_generic(net, vxlan_net_id); struct vxlan_sock *vs; - vs = vxlan_socket_create(net, port, rcv, data, ipv6); + vs = vxlan_socket_create(net, port, rcv, data, flags); if (!IS_ERR(vs)) return vs; @@ -2587,7 +2565,7 @@ static void vxlan_sock_work(struct work_struct *work) __be16 port = vxlan->dst_port; struct vxlan_sock *nvs; - nvs = vxlan_sock_add(net, port, vxlan_rcv, NULL, false, vxlan->flags & VXLAN_F_IPV6); + nvs = vxlan_sock_add(net, port, vxlan_rcv, NULL, false, vxlan->flags); spin_lock(&vn->sock_lock); if (!IS_ERR(nvs)) vxlan_vs_add_dev(nvs, vxlan); @@ -2711,6 +2689,17 @@ static int vxlan_newlink(struct net *net, struct net_device *dev, if (data[IFLA_VXLAN_PORT]) vxlan->dst_port = nla_get_be16(data[IFLA_VXLAN_PORT]); + if (data[IFLA_VXLAN_UDP_CSUM] && nla_get_u8(data[IFLA_VXLAN_UDP_CSUM])) + vxlan->flags |= VXLAN_F_UDP_CSUM; + + if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX] && + nla_get_u8(data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX])) + vxlan->flags |= VXLAN_F_UDP_ZERO_CSUM6_TX; + + if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX] && + nla_get_u8(data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX])) + vxlan->flags |= VXLAN_F_UDP_ZERO_CSUM6_RX; + if (vxlan_find_vni(net, vni, vxlan->dst_port)) { pr_info("duplicate VNI %u\n", vni); return -EEXIST; @@ -2774,7 +2763,10 @@ static size_t vxlan_get_size(const struct net_device *dev) nla_total_size(sizeof(__u32)) + /* IFLA_VXLAN_AGEING */ nla_total_size(sizeof(__u32)) + /* IFLA_VXLAN_LIMIT */ nla_total_size(sizeof(struct ifla_vxlan_port_range)) + - nla_total_size(sizeof(__be16))+ /* IFLA_VXLAN_PORT */ + nla_total_size(sizeof(__be16)) + /* IFLA_VXLAN_PORT */ + nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_UDP_CSUM */ + nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_UDP_ZERO_CSUM6_TX */ + nla_total_size(sizeof(__u8)) + /* IFLA_VXLAN_UDP_ZERO_CSUM6_RX */ 0; } @@ -2834,7 +2826,13 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev) !!(vxlan->flags & VXLAN_F_L3MISS)) || nla_put_u32(skb, IFLA_VXLAN_AGEING, vxlan->age_interval) || nla_put_u32(skb, IFLA_VXLAN_LIMIT, vxlan->addrmax) || - nla_put_be16(skb, IFLA_VXLAN_PORT, vxlan->dst_port)) + nla_put_be16(skb, IFLA_VXLAN_PORT, vxlan->dst_port) || + nla_put_u8(skb, IFLA_VXLAN_UDP_CSUM, + !!(vxlan->flags & VXLAN_F_UDP_CSUM)) || + nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_TX, + !!(vxlan->flags & VXLAN_F_UDP_ZERO_CSUM6_TX)) || + nla_put_u8(skb, IFLA_VXLAN_UDP_ZERO_CSUM6_RX, + !!(vxlan->flags & VXLAN_F_UDP_ZERO_CSUM6_RX))) goto nla_put_failure; if (nla_put(skb, IFLA_VXLAN_PORT_RANGE, sizeof(ports), &ports)) diff --git a/include/net/vxlan.h b/include/net/vxlan.h index 7bb4084b1bd0..12196ce661d9 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -24,9 +24,19 @@ struct vxlan_sock { struct udp_offload udp_offloads; }; +#define VXLAN_F_LEARN 0x01 +#define VXLAN_F_PROXY 0x02 +#define VXLAN_F_RSC 0x04 +#define VXLAN_F_L2MISS 0x08 +#define VXLAN_F_L3MISS 0x10 +#define VXLAN_F_IPV6 0x20 +#define VXLAN_F_UDP_CSUM 0x40 +#define VXLAN_F_UDP_ZERO_CSUM6_TX 0x80 +#define VXLAN_F_UDP_ZERO_CSUM6_RX 0x100 + struct vxlan_sock *vxlan_sock_add(struct net *net, __be16 port, vxlan_rcv_t *rcv, void *data, - bool no_share, bool ipv6); + bool no_share, u32 flags); void vxlan_sock_release(struct vxlan_sock *vs); diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 622e7910b8cc..b38534895db5 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -319,6 +319,9 @@ enum { IFLA_VXLAN_PORT, /* destination port */ IFLA_VXLAN_GROUP6, IFLA_VXLAN_LOCAL6, + IFLA_VXLAN_UDP_CSUM, + IFLA_VXLAN_UDP_ZERO_CSUM6_TX, + IFLA_VXLAN_UDP_ZERO_CSUM6_RX, __IFLA_VXLAN_MAX }; #define IFLA_VXLAN_MAX (__IFLA_VXLAN_MAX - 1) diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c index a93efa3f64c3..0edbd95c60e7 100644 --- a/net/openvswitch/vport-vxlan.c +++ b/net/openvswitch/vport-vxlan.c @@ -122,7 +122,7 @@ static struct vport *vxlan_tnl_create(const struct vport_parms *parms) vxlan_port = vxlan_vport(vport); strncpy(vxlan_port->name, parms->name, IFNAMSIZ); - vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true, false); + vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true, 0); if (IS_ERR(vs)) { ovs_vport_free(vport); return (void *)vs; -- cgit v1.2.3 From bac52139f0b7ab31330e98fd87fc5a2664951050 Mon Sep 17 00:00:00 2001 From: Anshuman Khandual Date: Thu, 22 May 2014 12:50:07 +0530 Subject: perf: Add new conditional branch filter 'PERF_SAMPLE_BRANCH_COND' This patch introduces new branch filter PERF_SAMPLE_BRANCH_COND which will extend the existing perf ABI. This will filter branches which are conditional. Various architectures can provide this functionality either with HW filtering support (if present) or with SW filtering of captured branch instructions. Signed-off-by: Anshuman Khandual Reviewed-by: Stephane Eranian Reviewed-by: Andi Kleen Signed-off-by: Peter Zijlstra Cc: mpe@ellerman.id.au Cc: benh@kernel.crashing.org Cc: Arnaldo Carvalho de Melo Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1400743210-32289-1-git-send-email-khandual@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- include/uapi/linux/perf_event.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index e3fc8f09d110..d9cd853818ad 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -163,8 +163,9 @@ enum perf_branch_sample_type { PERF_SAMPLE_BRANCH_ABORT_TX = 1U << 7, /* transaction aborts */ PERF_SAMPLE_BRANCH_IN_TX = 1U << 8, /* in transaction */ PERF_SAMPLE_BRANCH_NO_TX = 1U << 9, /* not in transaction */ + PERF_SAMPLE_BRANCH_COND = 1U << 10, /* conditional branches */ - PERF_SAMPLE_BRANCH_MAX = 1U << 10, /* non-ABI */ + PERF_SAMPLE_BRANCH_MAX = 1U << 11, /* non-ABI */ }; #define PERF_SAMPLE_BRANCH_PLM_ALL \ -- cgit v1.2.3 From 82b897782d10fcc4930c9d4a15b175348fdd2871 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 28 May 2014 11:45:04 +0300 Subject: perf: Differentiate exec() and non-exec() comm events perf tools like 'perf report' can aggregate samples by comm strings, which generally works. However, there are other potential use-cases. For example, to pair up 'calls' with 'returns' accurately (from branch events like Intel BTS) it is necessary to identify whether the process has exec'd. Although a comm event is generated when an 'exec' happens it is also generated whenever the comm string is changed on a whim (e.g. by prctl PR_SET_NAME). This patch adds a flag to the comm event to differentiate one case from the other. In order to determine whether the kernel supports the new flag, a selection bit named 'exec' is added to struct perf_event_attr. The bit does nothing but will cause perf_event_open() to fail if the bit is set on kernels that do not have it defined. Signed-off-by: Adrian Hunter Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/537D9EBE.7030806@intel.com Cc: Paul Mackerras Cc: Dave Jones Cc: Arnaldo Carvalho de Melo Cc: David Ahern Cc: Jiri Olsa Cc: Alexander Viro Cc: Linus Torvalds Cc: linux-fsdevel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar --- fs/exec.c | 6 +++--- include/linux/perf_event.h | 4 ++-- include/linux/sched.h | 6 +++++- include/uapi/linux/perf_event.h | 9 +++++++-- kernel/events/core.c | 4 ++-- 5 files changed, 19 insertions(+), 10 deletions(-) (limited to 'include/uapi') diff --git a/fs/exec.c b/fs/exec.c index a038a41a3677..a3d33fe592d6 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1046,13 +1046,13 @@ EXPORT_SYMBOL_GPL(get_task_comm); * so that a new one can be started */ -void set_task_comm(struct task_struct *tsk, const char *buf) +void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec) { task_lock(tsk); trace_task_rename(tsk, buf); strlcpy(tsk->comm, buf, sizeof(tsk->comm)); task_unlock(tsk); - perf_event_comm(tsk); + perf_event_comm(tsk, exec); } int flush_old_exec(struct linux_binprm * bprm) @@ -1111,7 +1111,7 @@ void setup_new_exec(struct linux_binprm * bprm) set_dumpable(current->mm, suid_dumpable); perf_event_exec(); - set_task_comm(current, kbasename(bprm->filename)); + __set_task_comm(current, kbasename(bprm->filename), true); /* Set the new mm task size. We have to do that late because it may * depend on TIF_32BIT which is only updated in flush_thread() on diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index b4c1d4685bf0..707617a8c0f6 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -707,7 +707,7 @@ extern int perf_register_guest_info_callbacks(struct perf_guest_info_callbacks * extern int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks *callbacks); extern void perf_event_exec(void); -extern void perf_event_comm(struct task_struct *tsk); +extern void perf_event_comm(struct task_struct *tsk, bool exec); extern void perf_event_fork(struct task_struct *tsk); /* Callchains */ @@ -815,7 +815,7 @@ static inline int perf_unregister_guest_info_callbacks static inline void perf_event_mmap(struct vm_area_struct *vma) { } static inline void perf_event_exec(void) { } -static inline void perf_event_comm(struct task_struct *tsk) { } +static inline void perf_event_comm(struct task_struct *tsk, bool exec) { } static inline void perf_event_fork(struct task_struct *tsk) { } static inline void perf_event_init(void) { } static inline int perf_swevent_get_recursion_context(void) { return -1; } diff --git a/include/linux/sched.h b/include/linux/sched.h index 221b2bde3723..ad86e1d7dbc2 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2379,7 +2379,11 @@ extern long do_fork(unsigned long, unsigned long, unsigned long, int __user *, i struct task_struct *fork_idle(int); extern pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags); -extern void set_task_comm(struct task_struct *tsk, const char *from); +extern void __set_task_comm(struct task_struct *tsk, const char *from, bool exec); +static inline void set_task_comm(struct task_struct *tsk, const char *from) +{ + __set_task_comm(tsk, from, false); +} extern char *get_task_comm(char *to, struct task_struct *tsk); #ifdef CONFIG_SMP diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index d9cd853818ad..5312fae47218 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -302,8 +302,8 @@ struct perf_event_attr { exclude_callchain_kernel : 1, /* exclude kernel callchains */ exclude_callchain_user : 1, /* exclude user callchains */ mmap2 : 1, /* include mmap with inode data */ - - __reserved_1 : 40; + comm_exec : 1, /* flag comm events that are due to an exec */ + __reserved_1 : 39; union { __u32 wakeup_events; /* wakeup every n events */ @@ -502,7 +502,12 @@ struct perf_event_mmap_page { #define PERF_RECORD_MISC_GUEST_KERNEL (4 << 0) #define PERF_RECORD_MISC_GUEST_USER (5 << 0) +/* + * PERF_RECORD_MISC_MMAP_DATA and PERF_RECORD_MISC_COMM_EXEC are used on + * different events so can reuse the same bit position. + */ #define PERF_RECORD_MISC_MMAP_DATA (1 << 13) +#define PERF_RECORD_MISC_COMM_EXEC (1 << 13) /* * Indicates that the content of PERF_SAMPLE_IP points to * the actual instruction that triggered the event. See also diff --git a/kernel/events/core.c b/kernel/events/core.c index 8fac2056d51e..7da5e561e89a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5090,7 +5090,7 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event) NULL); } -void perf_event_comm(struct task_struct *task) +void perf_event_comm(struct task_struct *task, bool exec) { struct perf_comm_event comm_event; @@ -5104,7 +5104,7 @@ void perf_event_comm(struct task_struct *task) .event_id = { .header = { .type = PERF_RECORD_COMM, - .misc = 0, + .misc = exec ? PERF_RECORD_MISC_COMM_EXEC : 0, /* .size */ }, /* .pid */ -- cgit v1.2.3 From f972eb63b1003fae68d7b7e9b674d4ba5db681c2 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 19 May 2014 15:13:47 -0400 Subject: perf: Pass protection and flags bits through mmap2 interface The mmap2 interface was missing the protection and flags bits needed to accurately determine if a mmap memory area was shared or private and if it was readable or not. Signed-off-by: Peter Zijlstra [tweaked patch to compile and wrote changelog] Signed-off-by: Don Zickus Link: http://lkml.kernel.org/r/1400526833-141779-2-git-send-email-dzickus@redhat.com Signed-off-by: Jiri Olsa --- include/uapi/linux/perf_event.h | 1 + kernel/events/core.c | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 34 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index 5312fae47218..9269de254874 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -705,6 +705,7 @@ enum perf_event_type { * u32 min; * u64 ino; * u64 ino_generation; + * u32 prot, flags; * char filename[]; * struct sample_id sample_id; * }; diff --git a/kernel/events/core.c b/kernel/events/core.c index 7da5e561e89a..eea1955c7868 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -40,6 +40,7 @@ #include #include #include +#include #include "internal.h" @@ -5127,6 +5128,7 @@ struct perf_mmap_event { int maj, min; u64 ino; u64 ino_generation; + u32 prot, flags; struct { struct perf_event_header header; @@ -5168,6 +5170,8 @@ static void perf_event_mmap_output(struct perf_event *event, mmap_event->event_id.header.size += sizeof(mmap_event->min); mmap_event->event_id.header.size += sizeof(mmap_event->ino); mmap_event->event_id.header.size += sizeof(mmap_event->ino_generation); + mmap_event->event_id.header.size += sizeof(mmap_event->prot); + mmap_event->event_id.header.size += sizeof(mmap_event->flags); } perf_event_header__init_id(&mmap_event->event_id.header, &sample, event); @@ -5186,6 +5190,8 @@ static void perf_event_mmap_output(struct perf_event *event, perf_output_put(&handle, mmap_event->min); perf_output_put(&handle, mmap_event->ino); perf_output_put(&handle, mmap_event->ino_generation); + perf_output_put(&handle, mmap_event->prot); + perf_output_put(&handle, mmap_event->flags); } __output_copy(&handle, mmap_event->file_name, @@ -5204,6 +5210,7 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) struct file *file = vma->vm_file; int maj = 0, min = 0; u64 ino = 0, gen = 0; + u32 prot = 0, flags = 0; unsigned int size; char tmp[16]; char *buf = NULL; @@ -5234,6 +5241,28 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) gen = inode->i_generation; maj = MAJOR(dev); min = MINOR(dev); + + if (vma->vm_flags & VM_READ) + prot |= PROT_READ; + if (vma->vm_flags & VM_WRITE) + prot |= PROT_WRITE; + if (vma->vm_flags & VM_EXEC) + prot |= PROT_EXEC; + + if (vma->vm_flags & VM_MAYSHARE) + flags = MAP_SHARED; + else + flags = MAP_PRIVATE; + + if (vma->vm_flags & VM_DENYWRITE) + flags |= MAP_DENYWRITE; + if (vma->vm_flags & VM_MAYEXEC) + flags |= MAP_EXECUTABLE; + if (vma->vm_flags & VM_LOCKED) + flags |= MAP_LOCKED; + if (vma->vm_flags & VM_HUGETLB) + flags |= MAP_HUGETLB; + goto got_name; } else { name = (char *)arch_vma_name(vma); @@ -5274,6 +5303,8 @@ got_name: mmap_event->min = min; mmap_event->ino = ino; mmap_event->ino_generation = gen; + mmap_event->prot = prot; + mmap_event->flags = flags; if (!(vma->vm_flags & VM_EXEC)) mmap_event->event_id.header.misc |= PERF_RECORD_MISC_MMAP_DATA; @@ -5314,6 +5345,8 @@ void perf_event_mmap(struct vm_area_struct *vma) /* .min (attr_mmap2 only) */ /* .ino (attr_mmap2 only) */ /* .ino_generation (attr_mmap2 only) */ + /* .prot (attr_mmap2 only) */ + /* .flags (attr_mmap2 only) */ }; perf_event_mmap_event(&mmap_event); -- cgit v1.2.3 From 7d824b6f9cf28917d8a05891ef423fb0e4e34c69 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Wed, 7 May 2014 17:37:51 +0200 Subject: btrfs: balance filter: add limit of processed chunks This started as debugging helper, to watch the effects of converting between raid levels on multiple devices, but could be useful standalone. In my case the usage filter was not finegrained enough and led to converting too many chunks at once. Another example use is in connection with drange+devid or vrange filters that allow to work with a specific chunk or even with a chunk on a given device. The limit filter applies last, the value of 0 means no limiting. CC: Ilya Dryomov CC: Hugo Mills Signed-off-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/ctree.h | 7 ++++++- fs/btrfs/volumes.c | 18 ++++++++++++++++++ fs/btrfs/volumes.h | 1 + include/uapi/linux/btrfs.h | 3 ++- 4 files changed, 27 insertions(+), 2 deletions(-) (limited to 'include/uapi') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index ba6b88528dc7..e6f899dc5e47 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -840,7 +840,10 @@ struct btrfs_disk_balance_args { /* BTRFS_BALANCE_ARGS_* */ __le64 flags; - __le64 unused[8]; + /* BTRFS_BALANCE_ARGS_LIMIT value */ + __le64 limit; + + __le64 unused[7]; } __attribute__ ((__packed__)); /* @@ -2897,6 +2900,7 @@ btrfs_disk_balance_args_to_cpu(struct btrfs_balance_args *cpu, cpu->vend = le64_to_cpu(disk->vend); cpu->target = le64_to_cpu(disk->target); cpu->flags = le64_to_cpu(disk->flags); + cpu->limit = le64_to_cpu(disk->limit); } static inline void @@ -2914,6 +2918,7 @@ btrfs_cpu_balance_args_to_disk(struct btrfs_disk_balance_args *disk, disk->vend = cpu_to_le64(cpu->vend); disk->target = cpu_to_le64(cpu->target); disk->flags = cpu_to_le64(cpu->flags); + disk->limit = cpu_to_le64(cpu->limit); } /* struct btrfs_super_block */ diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 49d7fab73360..3b761a456acd 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -2922,6 +2922,16 @@ static int should_balance_chunk(struct btrfs_root *root, return 0; } + /* + * limited by count, must be the last filter + */ + if ((bargs->flags & BTRFS_BALANCE_ARGS_LIMIT)) { + if (bargs->limit == 0) + return 0; + else + bargs->limit--; + } + return 1; } @@ -2944,6 +2954,9 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info) int ret; int enospc_errors = 0; bool counting = true; + u64 limit_data = bctl->data.limit; + u64 limit_meta = bctl->meta.limit; + u64 limit_sys = bctl->sys.limit; /* step one make some room on all the devices */ devices = &fs_info->fs_devices->devices; @@ -2982,6 +2995,11 @@ static int __btrfs_balance(struct btrfs_fs_info *fs_info) memset(&bctl->stat, 0, sizeof(bctl->stat)); spin_unlock(&fs_info->balance_lock); again: + if (!counting) { + bctl->data.limit = limit_data; + bctl->meta.limit = limit_meta; + bctl->sys.limit = limit_sys; + } key.objectid = BTRFS_FIRST_CHUNK_TREE_OBJECTID; key.offset = (u64)-1; key.type = BTRFS_CHUNK_ITEM_KEY; diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 80754f9dd3df..1a15bbeb65e2 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -255,6 +255,7 @@ struct map_lookup { #define BTRFS_BALANCE_ARGS_DEVID (1ULL << 2) #define BTRFS_BALANCE_ARGS_DRANGE (1ULL << 3) #define BTRFS_BALANCE_ARGS_VRANGE (1ULL << 4) +#define BTRFS_BALANCE_ARGS_LIMIT (1ULL << 5) /* * Profile changing flags. When SOFT is set we won't relocate chunk if diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index b4d69092fbdb..901a3c563f60 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -211,7 +211,8 @@ struct btrfs_balance_args { __u64 flags; - __u64 unused[8]; + __u64 limit; /* limit number of processed chunks */ + __u64 unused[7]; } __attribute__ ((__packed__)); /* report balance progress to userspace */ -- cgit v1.2.3 From 80a773fbfc2d6b5b2478377e8ac271d495f55e73 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Wed, 7 May 2014 18:17:06 +0200 Subject: btrfs: retrieve more info from FS_INFO ioctl Provide the basic information about filesystem through the ioctl: * b-tree node size (same as leaf size) * sector size * expected alignment of CLONE_RANGE and EXTENT_SAME ioctl arguments Backward compatibility: if the values are 0, kernel does not provide this information, the applications should ignore them. Signed-off-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 4 ++++ include/uapi/linux/btrfs.h | 6 +++++- 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 2f6d7b13b5bd..c6c8e3560e73 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2574,6 +2574,10 @@ static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg) } mutex_unlock(&fs_devices->device_list_mutex); + fi_args->nodesize = root->fs_info->super_copy->nodesize; + fi_args->sectorsize = root->fs_info->super_copy->sectorsize; + fi_args->clone_alignment = root->fs_info->super_copy->sectorsize; + if (copy_to_user(arg, fi_args, sizeof(*fi_args))) ret = -EFAULT; diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index 901a3c563f60..7554fd381a56 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -181,7 +181,11 @@ struct btrfs_ioctl_fs_info_args { __u64 max_id; /* out */ __u64 num_devices; /* out */ __u8 fsid[BTRFS_FSID_SIZE]; /* out */ - __u64 reserved[124]; /* pad to 1k */ + __u32 nodesize; /* out */ + __u32 sectorsize; /* out */ + __u32 clone_alignment; /* out */ + __u32 reserved32; + __u64 reserved[122]; /* pad to 1k */ }; struct btrfs_ioctl_feature_flags { -- cgit v1.2.3 From 65fcf668ee7f2de2fbd580e1297336045f1ef6f4 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 2 Jun 2014 16:13:21 -0400 Subject: drm/radeon: add query for number of active CUs Query to find out how many compute units on a GPU. Useful for OpenCL usermode drivers. Signed-off-by: Alex Deucher --- drivers/gpu/drm/radeon/cik.c | 12 +++++++++++- drivers/gpu/drm/radeon/evergreen.c | 12 ++++++++++++ drivers/gpu/drm/radeon/ni.c | 12 ++++++++++++ drivers/gpu/drm/radeon/r600.c | 3 +++ drivers/gpu/drm/radeon/radeon.h | 6 ++++++ drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- drivers/gpu/drm/radeon/radeon_kms.c | 16 ++++++++++++++++ drivers/gpu/drm/radeon/rv770.c | 3 +++ drivers/gpu/drm/radeon/si.c | 11 ++++++++++- include/uapi/drm/radeon_drm.h | 2 +- 10 files changed, 76 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index e4b2f2b51bb8..dcd4518a9b08 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -80,6 +80,7 @@ extern int sumo_rlc_init(struct radeon_device *rdev); extern void si_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc); extern void si_rlc_reset(struct radeon_device *rdev); extern void si_init_uvd_internal_cg(struct radeon_device *rdev); +static u32 cik_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh); extern int cik_sdma_resume(struct radeon_device *rdev); extern void cik_sdma_enable(struct radeon_device *rdev, bool enable); extern void cik_sdma_fini(struct radeon_device *rdev); @@ -3257,7 +3258,7 @@ static void cik_gpu_init(struct radeon_device *rdev) u32 mc_shared_chmap, mc_arb_ramcfg; u32 hdp_host_path_cntl; u32 tmp; - int i, j; + int i, j, k; switch (rdev->family) { case CHIP_BONAIRE: @@ -3446,6 +3447,15 @@ static void cik_gpu_init(struct radeon_device *rdev) rdev->config.cik.max_sh_per_se, rdev->config.cik.max_backends_per_se); + for (i = 0; i < rdev->config.cik.max_shader_engines; i++) { + for (j = 0; j < rdev->config.cik.max_sh_per_se; j++) { + for (k = 0; k < rdev->config.cik.max_cu_per_sh; k++) { + rdev->config.cik.active_cus += + hweight32(cik_get_cu_active_bitmap(rdev, i, j)); + } + } + } + /* set HW defaults for 3D engine */ WREG32(CP_MEQ_THRESHOLDS, MEQ1_START(0x30) | MEQ2_START(0x60)); diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 653eff814504..e2f605224e8c 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -3337,6 +3337,18 @@ static void evergreen_gpu_init(struct radeon_device *rdev) disabled_rb_mask &= ~(1 << i); } + for (i = 0; i < rdev->config.evergreen.num_ses; i++) { + u32 simd_disable_bitmap; + + WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i)); + WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i)); + simd_disable_bitmap = (RREG32(CC_GC_SHADER_PIPE_CONFIG) & 0xffff0000) >> 16; + simd_disable_bitmap |= 0xffffffff << rdev->config.evergreen.max_simds; + tmp <<= 16; + tmp |= simd_disable_bitmap; + } + rdev->config.evergreen.active_simds = hweight32(~tmp); + WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES); WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES); diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index c0fd8f647cf0..5a33ca681867 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -1057,6 +1057,18 @@ static void cayman_gpu_init(struct radeon_device *rdev) disabled_rb_mask &= ~(1 << i); } + for (i = 0; i < rdev->config.cayman.max_shader_engines; i++) { + u32 simd_disable_bitmap; + + WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i)); + WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_INDEX(i)); + simd_disable_bitmap = (RREG32(CC_GC_SHADER_PIPE_CONFIG) & 0xffff0000) >> 16; + simd_disable_bitmap |= 0xffffffff << rdev->config.cayman.max_simds_per_se; + tmp <<= 16; + tmp |= simd_disable_bitmap; + } + rdev->config.cayman.active_simds = hweight32(~tmp); + WREG32(GRBM_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES); WREG32(RLC_GFX_INDEX, INSTANCE_BROADCAST_WRITES | SE_BROADCAST_WRITES); diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index c2ff17cebd91..c66952d4b00c 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -1958,6 +1958,9 @@ static void r600_gpu_init(struct radeon_device *rdev) if (tmp < rdev->config.r600.max_simds) { rdev->config.r600.max_simds = tmp; } + tmp = rdev->config.r600.max_simds - + r600_count_pipe_bits((cc_gc_shader_pipe_config >> 16) & R6XX_MAX_SIMDS_MASK); + rdev->config.r600.active_simds = tmp; disabled_rb_mask = (RREG32(CC_RB_BACKEND_DISABLE) >> 16) & R6XX_MAX_BACKENDS_MASK; tmp = (tiling_config & PIPE_TILING__MASK) >> PIPE_TILING__SHIFT; diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index dd77111096bd..4b0bbf88d5c0 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1932,6 +1932,7 @@ struct r600_asic { unsigned tiling_group_size; unsigned tile_config; unsigned backend_map; + unsigned active_simds; }; struct rv770_asic { @@ -1957,6 +1958,7 @@ struct rv770_asic { unsigned tiling_group_size; unsigned tile_config; unsigned backend_map; + unsigned active_simds; }; struct evergreen_asic { @@ -1983,6 +1985,7 @@ struct evergreen_asic { unsigned tiling_group_size; unsigned tile_config; unsigned backend_map; + unsigned active_simds; }; struct cayman_asic { @@ -2021,6 +2024,7 @@ struct cayman_asic { unsigned multi_gpu_tile_size; unsigned tile_config; + unsigned active_simds; }; struct si_asic { @@ -2051,6 +2055,7 @@ struct si_asic { unsigned tile_config; uint32_t tile_mode_array[32]; + uint32_t active_cus; }; struct cik_asic { @@ -2082,6 +2087,7 @@ struct cik_asic { unsigned tile_config; uint32_t tile_mode_array[32]; uint32_t macrotile_mode_array[16]; + uint32_t active_cus; }; union radeon_asic_config { diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index b7a2ec2d1598..6e3017413386 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -81,9 +81,10 @@ * 2.37.0 - allow GS ring setup on r6xx/r7xx * 2.38.0 - RADEON_GEM_OP (GET_INITIAL_DOMAIN, SET_INITIAL_DOMAIN), * CIK: 1D and linear tiling modes contain valid PIPE_CONFIG + * 2.39.0 - Add INFO query for number of active CUs */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 38 +#define KMS_DRIVER_MINOR 39 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index eaaedba04675..5cd70f9e7311 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -513,6 +513,22 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file value_size = sizeof(uint64_t); value64 = atomic64_read(&rdev->gtt_usage); break; + case RADEON_INFO_ACTIVE_CU_COUNT: + if (rdev->family >= CHIP_BONAIRE) + *value = rdev->config.cik.active_cus; + else if (rdev->family >= CHIP_TAHITI) + *value = rdev->config.si.active_cus; + else if (rdev->family >= CHIP_CAYMAN) + *value = rdev->config.cayman.active_simds; + else if (rdev->family >= CHIP_CEDAR) + *value = rdev->config.evergreen.active_simds; + else if (rdev->family >= CHIP_RV770) + *value = rdev->config.rv770.active_simds; + else if (rdev->family >= CHIP_R600) + *value = rdev->config.r600.active_simds; + else + *value = 1; + break; default: DRM_DEBUG_KMS("Invalid request %d\n", info->request); return -EINVAL; diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index 97b776666b75..da8703d8d455 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -1327,6 +1327,9 @@ static void rv770_gpu_init(struct radeon_device *rdev) if (tmp < rdev->config.rv770.max_simds) { rdev->config.rv770.max_simds = tmp; } + tmp = rdev->config.rv770.max_simds - + r600_count_pipe_bits((cc_gc_shader_pipe_config >> 16) & R7XX_MAX_SIMDS_MASK); + rdev->config.rv770.active_simds = tmp; switch (rdev->config.rv770.max_tile_pipes) { case 1: diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index ec13e8df4c30..730cee2c34cf 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -71,6 +71,7 @@ MODULE_FIRMWARE("radeon/HAINAN_mc2.bin"); MODULE_FIRMWARE("radeon/HAINAN_rlc.bin"); MODULE_FIRMWARE("radeon/HAINAN_smc.bin"); +static u32 si_get_cu_active_bitmap(struct radeon_device *rdev, u32 se, u32 sh); static void si_pcie_gen3_enable(struct radeon_device *rdev); static void si_program_aspm(struct radeon_device *rdev); extern void sumo_rlc_fini(struct radeon_device *rdev); @@ -2900,7 +2901,7 @@ static void si_gpu_init(struct radeon_device *rdev) u32 sx_debug_1; u32 hdp_host_path_cntl; u32 tmp; - int i, j; + int i, j, k; switch (rdev->family) { case CHIP_TAHITI: @@ -3098,6 +3099,14 @@ static void si_gpu_init(struct radeon_device *rdev) rdev->config.si.max_sh_per_se, rdev->config.si.max_cu_per_sh); + for (i = 0; i < rdev->config.si.max_shader_engines; i++) { + for (j = 0; j < rdev->config.si.max_sh_per_se; j++) { + for (k = 0; k < rdev->config.si.max_cu_per_sh; k++) { + rdev->config.si.active_cus += + hweight32(si_get_cu_active_bitmap(rdev, i, j)); + } + } + } /* set HW defaults for 3D engine */ WREG32(CP_QUEUE_THRESHOLDS, (ROQ_IB1_START(0x16) | diff --git a/include/uapi/drm/radeon_drm.h b/include/uapi/drm/radeon_drm.h index aefa2f6afa3b..1cc0b610f162 100644 --- a/include/uapi/drm/radeon_drm.h +++ b/include/uapi/drm/radeon_drm.h @@ -1007,7 +1007,7 @@ struct drm_radeon_cs { #define RADEON_INFO_NUM_BYTES_MOVED 0x1d #define RADEON_INFO_VRAM_USAGE 0x1e #define RADEON_INFO_GTT_USAGE 0x1f - +#define RADEON_INFO_ACTIVE_CU_COUNT 0x20 struct drm_radeon_info { uint32_t request; -- cgit v1.2.3 From 30dc5e63d6a5ad24894b5512d10b228d73645a44 Mon Sep 17 00:00:00 2001 From: Tatyana Nikolova Date: Wed, 26 Mar 2014 17:07:35 -0500 Subject: RDMA/core: Add support for iWARP Port Mapper user space service This patch adds iWARP Port Mapper (IWPM) Version 2 support. The iWARP Port Mapper implementation is based on the port mapper specification section in the Sockets Direct Protocol paper - http://www.rdmaconsortium.org/home/draft-pinkerton-iwarp-sdp-v1.0.pdf Existing iWARP RDMA providers use the same IP address as the native TCP/IP stack when creating RDMA connections. They need a mechanism to claim the TCP ports used for RDMA connections to prevent TCP port collisions when other host applications use TCP ports. The iWARP Port Mapper provides a standard mechanism to accomplish this. Without this service it is possible for RDMA application to bind/listen on the same port which is already being used by native TCP host application. If that happens the incoming TCP connection data can be passed to the RDMA stack with error. The iWARP Port Mapper solution doesn't contain any changes to the existing network stack in the kernel space. All the changes are contained with the infiniband tree and also in user space. The iWARP Port Mapper service is implemented as a user space daemon process. Source for the IWPM service is located at http://git.openfabrics.org/git?p=~tnikolova/libiwpm-1.0.0/.git;a=summary The iWARP driver (port mapper client) sends to the IWPM service the local IP address and TCP port it has received from the RDMA application, when starting a connection. The IWPM service performs a socket bind from user space to get an available TCP port, called a mapped port, and communicates it back to the client. In that sense, the IWPM service is used to map the TCP port, which the RDMA application uses to any port available from the host TCP port space. The mapped ports are used in iWARP RDMA connections to avoid collisions with native TCP stack which is aware that these ports are taken. When an RDMA connection using a mapped port is terminated, the client notifies the IWPM service, which then releases the TCP port. The message exchange between the IWPM service and the iWARP drivers (between user space and kernel space) is implemented using netlink sockets. 1) Netlink interface functions are added: ibnl_unicast() and ibnl_mulitcast() for sending netlink messages to user space 2) The signature of the existing ibnl_put_msg() is changed to be more generic 3) Two netlink clients are added: RDMA_NL_NES, RDMA_NL_C4IW corresponding to the two iWarp drivers - nes and cxgb4 which use the IWPM service 4) Enums are added to enumerate the attributes in the netlink messages, which are exchanged between the user space IWPM service and the iWARP drivers Signed-off-by: Tatyana Nikolova Signed-off-by: Steve Wise Reviewed-by: PJ Waskiewicz [ Fold in range checking fixes and nlh_next removal as suggested by Dan Carpenter and Steve Wise. Fix sparse endianness in hash. - Roland ] Signed-off-by: Roland Dreier --- drivers/infiniband/core/Makefile | 2 +- drivers/infiniband/core/cma.c | 3 +- drivers/infiniband/core/iwpm_msg.c | 685 ++++++++++++++++++++++++++++++++++++ drivers/infiniband/core/iwpm_util.c | 607 ++++++++++++++++++++++++++++++++ drivers/infiniband/core/iwpm_util.h | 238 +++++++++++++ drivers/infiniband/core/netlink.c | 18 +- include/rdma/iw_portmap.h | 199 +++++++++++ include/rdma/rdma_netlink.h | 23 +- include/uapi/rdma/rdma_netlink.h | 96 ++++- 9 files changed, 1865 insertions(+), 6 deletions(-) create mode 100644 drivers/infiniband/core/iwpm_msg.c create mode 100644 drivers/infiniband/core/iwpm_util.c create mode 100644 drivers/infiniband/core/iwpm_util.h create mode 100644 include/rdma/iw_portmap.h (limited to 'include/uapi') diff --git a/drivers/infiniband/core/Makefile b/drivers/infiniband/core/Makefile index 3ab3865544bb..ffd0af6734af 100644 --- a/drivers/infiniband/core/Makefile +++ b/drivers/infiniband/core/Makefile @@ -18,7 +18,7 @@ ib_sa-y := sa_query.o multicast.o ib_cm-y := cm.o -iw_cm-y := iwcm.o +iw_cm-y := iwcm.o iwpm_util.o iwpm_msg.o rdma_cm-y := cma.o diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 42c3058e6e9c..d570030d899c 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -3607,7 +3607,8 @@ static int cma_get_id_stats(struct sk_buff *skb, struct netlink_callback *cb) id_stats = ibnl_put_msg(skb, &nlh, cb->nlh->nlmsg_seq, sizeof *id_stats, RDMA_NL_RDMA_CM, - RDMA_NL_RDMA_CM_ID_STATS); + RDMA_NL_RDMA_CM_ID_STATS, + NLM_F_MULTI); if (!id_stats) goto out; diff --git a/drivers/infiniband/core/iwpm_msg.c b/drivers/infiniband/core/iwpm_msg.c new file mode 100644 index 000000000000..b85ddbc979e0 --- /dev/null +++ b/drivers/infiniband/core/iwpm_msg.c @@ -0,0 +1,685 @@ +/* + * Copyright (c) 2014 Intel Corporation. All rights reserved. + * Copyright (c) 2014 Chelsio, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "iwpm_util.h" + +static const char iwpm_ulib_name[] = "iWarpPortMapperUser"; +static int iwpm_ulib_version = 3; +static int iwpm_user_pid = IWPM_PID_UNDEFINED; +static atomic_t echo_nlmsg_seq; + +int iwpm_valid_pid(void) +{ + return iwpm_user_pid > 0; +} +EXPORT_SYMBOL(iwpm_valid_pid); + +/* + * iwpm_register_pid - Send a netlink query to user space + * for the iwarp port mapper pid + * + * nlmsg attributes: + * [IWPM_NLA_REG_PID_SEQ] + * [IWPM_NLA_REG_IF_NAME] + * [IWPM_NLA_REG_IBDEV_NAME] + * [IWPM_NLA_REG_ULIB_NAME] + */ +int iwpm_register_pid(struct iwpm_dev_data *pm_msg, u8 nl_client) +{ + struct sk_buff *skb = NULL; + struct iwpm_nlmsg_request *nlmsg_request = NULL; + struct nlmsghdr *nlh; + u32 msg_seq; + const char *err_str = ""; + int ret = -EINVAL; + + if (!iwpm_valid_client(nl_client)) { + err_str = "Invalid port mapper client"; + goto pid_query_error; + } + if (iwpm_registered_client(nl_client)) + return 0; + skb = iwpm_create_nlmsg(RDMA_NL_IWPM_REG_PID, &nlh, nl_client); + if (!skb) { + err_str = "Unable to create a nlmsg"; + goto pid_query_error; + } + nlh->nlmsg_seq = iwpm_get_nlmsg_seq(); + nlmsg_request = iwpm_get_nlmsg_request(nlh->nlmsg_seq, nl_client, GFP_KERNEL); + if (!nlmsg_request) { + err_str = "Unable to allocate netlink request"; + goto pid_query_error; + } + msg_seq = atomic_read(&echo_nlmsg_seq); + + /* fill in the pid request message */ + err_str = "Unable to put attribute of the nlmsg"; + ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, IWPM_NLA_REG_PID_SEQ); + if (ret) + goto pid_query_error; + ret = ibnl_put_attr(skb, nlh, IWPM_IFNAME_SIZE, + pm_msg->if_name, IWPM_NLA_REG_IF_NAME); + if (ret) + goto pid_query_error; + ret = ibnl_put_attr(skb, nlh, IWPM_DEVNAME_SIZE, + pm_msg->dev_name, IWPM_NLA_REG_IBDEV_NAME); + if (ret) + goto pid_query_error; + ret = ibnl_put_attr(skb, nlh, IWPM_ULIBNAME_SIZE, + (char *)iwpm_ulib_name, IWPM_NLA_REG_ULIB_NAME); + if (ret) + goto pid_query_error; + + pr_debug("%s: Multicasting a nlmsg (dev = %s ifname = %s iwpm = %s)\n", + __func__, pm_msg->dev_name, pm_msg->if_name, iwpm_ulib_name); + + ret = ibnl_multicast(skb, nlh, RDMA_NL_GROUP_IWPM, GFP_KERNEL); + if (ret) { + skb = NULL; /* skb is freed in the netlink send-op handling */ + iwpm_set_registered(nl_client, 1); + iwpm_user_pid = IWPM_PID_UNAVAILABLE; + err_str = "Unable to send a nlmsg"; + goto pid_query_error; + } + nlmsg_request->req_buffer = pm_msg; + ret = iwpm_wait_complete_req(nlmsg_request); + return ret; +pid_query_error: + pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client); + if (skb) + dev_kfree_skb(skb); + if (nlmsg_request) + iwpm_free_nlmsg_request(&nlmsg_request->kref); + return ret; +} +EXPORT_SYMBOL(iwpm_register_pid); + +/* + * iwpm_add_mapping - Send a netlink add mapping message + * to the port mapper + * nlmsg attributes: + * [IWPM_NLA_MANAGE_MAPPING_SEQ] + * [IWPM_NLA_MANAGE_ADDR] + */ +int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client) +{ + struct sk_buff *skb = NULL; + struct iwpm_nlmsg_request *nlmsg_request = NULL; + struct nlmsghdr *nlh; + u32 msg_seq; + const char *err_str = ""; + int ret = -EINVAL; + + if (!iwpm_valid_client(nl_client)) { + err_str = "Invalid port mapper client"; + goto add_mapping_error; + } + if (!iwpm_registered_client(nl_client)) { + err_str = "Unregistered port mapper client"; + goto add_mapping_error; + } + if (!iwpm_valid_pid()) + return 0; + skb = iwpm_create_nlmsg(RDMA_NL_IWPM_ADD_MAPPING, &nlh, nl_client); + if (!skb) { + err_str = "Unable to create a nlmsg"; + goto add_mapping_error; + } + nlh->nlmsg_seq = iwpm_get_nlmsg_seq(); + nlmsg_request = iwpm_get_nlmsg_request(nlh->nlmsg_seq, nl_client, GFP_KERNEL); + if (!nlmsg_request) { + err_str = "Unable to allocate netlink request"; + goto add_mapping_error; + } + msg_seq = atomic_read(&echo_nlmsg_seq); + /* fill in the add mapping message */ + err_str = "Unable to put attribute of the nlmsg"; + ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, + IWPM_NLA_MANAGE_MAPPING_SEQ); + if (ret) + goto add_mapping_error; + ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage), + &pm_msg->loc_addr, IWPM_NLA_MANAGE_ADDR); + if (ret) + goto add_mapping_error; + nlmsg_request->req_buffer = pm_msg; + + ret = ibnl_unicast(skb, nlh, iwpm_user_pid); + if (ret) { + skb = NULL; /* skb is freed in the netlink send-op handling */ + iwpm_user_pid = IWPM_PID_UNDEFINED; + err_str = "Unable to send a nlmsg"; + goto add_mapping_error; + } + ret = iwpm_wait_complete_req(nlmsg_request); + return ret; +add_mapping_error: + pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client); + if (skb) + dev_kfree_skb(skb); + if (nlmsg_request) + iwpm_free_nlmsg_request(&nlmsg_request->kref); + return ret; +} +EXPORT_SYMBOL(iwpm_add_mapping); + +/* + * iwpm_add_and_query_mapping - Send a netlink add and query + * mapping message to the port mapper + * nlmsg attributes: + * [IWPM_NLA_QUERY_MAPPING_SEQ] + * [IWPM_NLA_QUERY_LOCAL_ADDR] + * [IWPM_NLA_QUERY_REMOTE_ADDR] + */ +int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client) +{ + struct sk_buff *skb = NULL; + struct iwpm_nlmsg_request *nlmsg_request = NULL; + struct nlmsghdr *nlh; + u32 msg_seq; + const char *err_str = ""; + int ret = -EINVAL; + + if (!iwpm_valid_client(nl_client)) { + err_str = "Invalid port mapper client"; + goto query_mapping_error; + } + if (!iwpm_registered_client(nl_client)) { + err_str = "Unregistered port mapper client"; + goto query_mapping_error; + } + if (!iwpm_valid_pid()) + return 0; + ret = -ENOMEM; + skb = iwpm_create_nlmsg(RDMA_NL_IWPM_QUERY_MAPPING, &nlh, nl_client); + if (!skb) { + err_str = "Unable to create a nlmsg"; + goto query_mapping_error; + } + nlh->nlmsg_seq = iwpm_get_nlmsg_seq(); + nlmsg_request = iwpm_get_nlmsg_request(nlh->nlmsg_seq, + nl_client, GFP_KERNEL); + if (!nlmsg_request) { + err_str = "Unable to allocate netlink request"; + goto query_mapping_error; + } + msg_seq = atomic_read(&echo_nlmsg_seq); + + /* fill in the query message */ + err_str = "Unable to put attribute of the nlmsg"; + ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, + IWPM_NLA_QUERY_MAPPING_SEQ); + if (ret) + goto query_mapping_error; + ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage), + &pm_msg->loc_addr, IWPM_NLA_QUERY_LOCAL_ADDR); + if (ret) + goto query_mapping_error; + ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage), + &pm_msg->rem_addr, IWPM_NLA_QUERY_REMOTE_ADDR); + if (ret) + goto query_mapping_error; + nlmsg_request->req_buffer = pm_msg; + + ret = ibnl_unicast(skb, nlh, iwpm_user_pid); + if (ret) { + skb = NULL; /* skb is freed in the netlink send-op handling */ + err_str = "Unable to send a nlmsg"; + goto query_mapping_error; + } + ret = iwpm_wait_complete_req(nlmsg_request); + return ret; +query_mapping_error: + pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client); + if (skb) + dev_kfree_skb(skb); + if (nlmsg_request) + iwpm_free_nlmsg_request(&nlmsg_request->kref); + return ret; +} +EXPORT_SYMBOL(iwpm_add_and_query_mapping); + +/* + * iwpm_remove_mapping - Send a netlink remove mapping message + * to the port mapper + * nlmsg attributes: + * [IWPM_NLA_MANAGE_MAPPING_SEQ] + * [IWPM_NLA_MANAGE_ADDR] + */ +int iwpm_remove_mapping(struct sockaddr_storage *local_addr, u8 nl_client) +{ + struct sk_buff *skb = NULL; + struct nlmsghdr *nlh; + u32 msg_seq; + const char *err_str = ""; + int ret = -EINVAL; + + if (!iwpm_valid_client(nl_client)) { + err_str = "Invalid port mapper client"; + goto remove_mapping_error; + } + if (!iwpm_registered_client(nl_client)) { + err_str = "Unregistered port mapper client"; + goto remove_mapping_error; + } + if (!iwpm_valid_pid()) + return 0; + skb = iwpm_create_nlmsg(RDMA_NL_IWPM_REMOVE_MAPPING, &nlh, nl_client); + if (!skb) { + ret = -ENOMEM; + err_str = "Unable to create a nlmsg"; + goto remove_mapping_error; + } + msg_seq = atomic_read(&echo_nlmsg_seq); + nlh->nlmsg_seq = iwpm_get_nlmsg_seq(); + err_str = "Unable to put attribute of the nlmsg"; + ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, + IWPM_NLA_MANAGE_MAPPING_SEQ); + if (ret) + goto remove_mapping_error; + ret = ibnl_put_attr(skb, nlh, sizeof(struct sockaddr_storage), + local_addr, IWPM_NLA_MANAGE_ADDR); + if (ret) + goto remove_mapping_error; + + ret = ibnl_unicast(skb, nlh, iwpm_user_pid); + if (ret) { + skb = NULL; /* skb is freed in the netlink send-op handling */ + iwpm_user_pid = IWPM_PID_UNDEFINED; + err_str = "Unable to send a nlmsg"; + goto remove_mapping_error; + } + iwpm_print_sockaddr(local_addr, + "remove_mapping: Local sockaddr:"); + return 0; +remove_mapping_error: + pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client); + if (skb) + dev_kfree_skb_any(skb); + return ret; +} +EXPORT_SYMBOL(iwpm_remove_mapping); + +/* netlink attribute policy for the received response to register pid request */ +static const struct nla_policy resp_reg_policy[IWPM_NLA_RREG_PID_MAX] = { + [IWPM_NLA_RREG_PID_SEQ] = { .type = NLA_U32 }, + [IWPM_NLA_RREG_IBDEV_NAME] = { .type = NLA_STRING, + .len = IWPM_DEVNAME_SIZE - 1 }, + [IWPM_NLA_RREG_ULIB_NAME] = { .type = NLA_STRING, + .len = IWPM_ULIBNAME_SIZE - 1 }, + [IWPM_NLA_RREG_ULIB_VER] = { .type = NLA_U16 }, + [IWPM_NLA_RREG_PID_ERR] = { .type = NLA_U16 } +}; + +/* + * iwpm_register_pid_cb - Process a port mapper response to + * iwpm_register_pid() + */ +int iwpm_register_pid_cb(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct iwpm_nlmsg_request *nlmsg_request = NULL; + struct nlattr *nltb[IWPM_NLA_RREG_PID_MAX]; + struct iwpm_dev_data *pm_msg; + char *dev_name, *iwpm_name; + u32 msg_seq; + u8 nl_client; + u16 iwpm_version; + const char *msg_type = "Register Pid response"; + + if (iwpm_parse_nlmsg(cb, IWPM_NLA_RREG_PID_MAX, + resp_reg_policy, nltb, msg_type)) + return -EINVAL; + + msg_seq = nla_get_u32(nltb[IWPM_NLA_RREG_PID_SEQ]); + nlmsg_request = iwpm_find_nlmsg_request(msg_seq); + if (!nlmsg_request) { + pr_info("%s: Could not find a matching request (seq = %u)\n", + __func__, msg_seq); + return -EINVAL; + } + pm_msg = nlmsg_request->req_buffer; + nl_client = nlmsg_request->nl_client; + dev_name = (char *)nla_data(nltb[IWPM_NLA_RREG_IBDEV_NAME]); + iwpm_name = (char *)nla_data(nltb[IWPM_NLA_RREG_ULIB_NAME]); + iwpm_version = nla_get_u16(nltb[IWPM_NLA_RREG_ULIB_VER]); + + /* check device name, ulib name and version */ + if (strcmp(pm_msg->dev_name, dev_name) || + strcmp(iwpm_ulib_name, iwpm_name) || + iwpm_version != iwpm_ulib_version) { + + pr_info("%s: Incorrect info (dev = %s name = %s version = %d)\n", + __func__, dev_name, iwpm_name, iwpm_version); + nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR; + goto register_pid_response_exit; + } + iwpm_user_pid = cb->nlh->nlmsg_pid; + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + pr_debug("%s: iWarp Port Mapper (pid = %d) is available!\n", + __func__, iwpm_user_pid); + if (iwpm_valid_client(nl_client)) + iwpm_set_registered(nl_client, 1); +register_pid_response_exit: + nlmsg_request->request_done = 1; + /* always for found nlmsg_request */ + kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request); + barrier(); + wake_up(&nlmsg_request->waitq); + return 0; +} +EXPORT_SYMBOL(iwpm_register_pid_cb); + +/* netlink attribute policy for the received response to add mapping request */ +static const struct nla_policy resp_add_policy[IWPM_NLA_RMANAGE_MAPPING_MAX] = { + [IWPM_NLA_MANAGE_MAPPING_SEQ] = { .type = NLA_U32 }, + [IWPM_NLA_MANAGE_ADDR] = { .len = sizeof(struct sockaddr_storage) }, + [IWPM_NLA_MANAGE_MAPPED_LOC_ADDR] = { .len = sizeof(struct sockaddr_storage) }, + [IWPM_NLA_RMANAGE_MAPPING_ERR] = { .type = NLA_U16 } +}; + +/* + * iwpm_add_mapping_cb - Process a port mapper response to + * iwpm_add_mapping() + */ +int iwpm_add_mapping_cb(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct iwpm_sa_data *pm_msg; + struct iwpm_nlmsg_request *nlmsg_request = NULL; + struct nlattr *nltb[IWPM_NLA_RMANAGE_MAPPING_MAX]; + struct sockaddr_storage *local_sockaddr; + struct sockaddr_storage *mapped_sockaddr; + const char *msg_type; + u32 msg_seq; + + msg_type = "Add Mapping response"; + if (iwpm_parse_nlmsg(cb, IWPM_NLA_RMANAGE_MAPPING_MAX, + resp_add_policy, nltb, msg_type)) + return -EINVAL; + + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + + msg_seq = nla_get_u32(nltb[IWPM_NLA_MANAGE_MAPPING_SEQ]); + nlmsg_request = iwpm_find_nlmsg_request(msg_seq); + if (!nlmsg_request) { + pr_info("%s: Could not find a matching request (seq = %u)\n", + __func__, msg_seq); + return -EINVAL; + } + pm_msg = nlmsg_request->req_buffer; + local_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_MANAGE_ADDR]); + mapped_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_MANAGE_MAPPED_LOC_ADDR]); + + if (iwpm_compare_sockaddr(local_sockaddr, &pm_msg->loc_addr)) { + nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR; + goto add_mapping_response_exit; + } + if (mapped_sockaddr->ss_family != local_sockaddr->ss_family) { + pr_info("%s: Sockaddr family doesn't match the requested one\n", + __func__); + nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR; + goto add_mapping_response_exit; + } + memcpy(&pm_msg->mapped_loc_addr, mapped_sockaddr, + sizeof(*mapped_sockaddr)); + iwpm_print_sockaddr(&pm_msg->loc_addr, + "add_mapping: Local sockaddr:"); + iwpm_print_sockaddr(&pm_msg->mapped_loc_addr, + "add_mapping: Mapped local sockaddr:"); + +add_mapping_response_exit: + nlmsg_request->request_done = 1; + /* always for found request */ + kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request); + barrier(); + wake_up(&nlmsg_request->waitq); + return 0; +} +EXPORT_SYMBOL(iwpm_add_mapping_cb); + +/* netlink attribute policy for the response to add and query mapping request */ +static const struct nla_policy resp_query_policy[IWPM_NLA_RQUERY_MAPPING_MAX] = { + [IWPM_NLA_QUERY_MAPPING_SEQ] = { .type = NLA_U32 }, + [IWPM_NLA_QUERY_LOCAL_ADDR] = { .len = sizeof(struct sockaddr_storage) }, + [IWPM_NLA_QUERY_REMOTE_ADDR] = { .len = sizeof(struct sockaddr_storage) }, + [IWPM_NLA_RQUERY_MAPPED_LOC_ADDR] = { .len = sizeof(struct sockaddr_storage) }, + [IWPM_NLA_RQUERY_MAPPED_REM_ADDR] = { .len = sizeof(struct sockaddr_storage) }, + [IWPM_NLA_RQUERY_MAPPING_ERR] = { .type = NLA_U16 } +}; + +/* + * iwpm_add_and_query_mapping_cb - Process a port mapper response to + * iwpm_add_and_query_mapping() + */ +int iwpm_add_and_query_mapping_cb(struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct iwpm_sa_data *pm_msg; + struct iwpm_nlmsg_request *nlmsg_request = NULL; + struct nlattr *nltb[IWPM_NLA_RQUERY_MAPPING_MAX]; + struct sockaddr_storage *local_sockaddr, *remote_sockaddr; + struct sockaddr_storage *mapped_loc_sockaddr, *mapped_rem_sockaddr; + const char *msg_type; + u32 msg_seq; + u16 err_code; + + msg_type = "Query Mapping response"; + if (iwpm_parse_nlmsg(cb, IWPM_NLA_RQUERY_MAPPING_MAX, + resp_query_policy, nltb, msg_type)) + return -EINVAL; + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + + msg_seq = nla_get_u32(nltb[IWPM_NLA_QUERY_MAPPING_SEQ]); + nlmsg_request = iwpm_find_nlmsg_request(msg_seq); + if (!nlmsg_request) { + pr_info("%s: Could not find a matching request (seq = %u)\n", + __func__, msg_seq); + return -EINVAL; + } + pm_msg = nlmsg_request->req_buffer; + local_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_QUERY_LOCAL_ADDR]); + remote_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_QUERY_REMOTE_ADDR]); + mapped_loc_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR]); + mapped_rem_sockaddr = (struct sockaddr_storage *) + nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_REM_ADDR]); + + err_code = nla_get_u16(nltb[IWPM_NLA_RQUERY_MAPPING_ERR]); + if (err_code == IWPM_REMOTE_QUERY_REJECT) { + pr_info("%s: Received a Reject (pid = %u, echo seq = %u)\n", + __func__, cb->nlh->nlmsg_pid, msg_seq); + nlmsg_request->err_code = IWPM_REMOTE_QUERY_REJECT; + } + if (iwpm_compare_sockaddr(local_sockaddr, &pm_msg->loc_addr) || + iwpm_compare_sockaddr(remote_sockaddr, &pm_msg->rem_addr)) { + pr_info("%s: Incorrect local sockaddr\n", __func__); + nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR; + goto query_mapping_response_exit; + } + if (mapped_loc_sockaddr->ss_family != local_sockaddr->ss_family || + mapped_rem_sockaddr->ss_family != remote_sockaddr->ss_family) { + pr_info("%s: Sockaddr family doesn't match the requested one\n", + __func__); + nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR; + goto query_mapping_response_exit; + } + memcpy(&pm_msg->mapped_loc_addr, mapped_loc_sockaddr, + sizeof(*mapped_loc_sockaddr)); + memcpy(&pm_msg->mapped_rem_addr, mapped_rem_sockaddr, + sizeof(*mapped_rem_sockaddr)); + + iwpm_print_sockaddr(&pm_msg->loc_addr, + "query_mapping: Local sockaddr:"); + iwpm_print_sockaddr(&pm_msg->mapped_loc_addr, + "query_mapping: Mapped local sockaddr:"); + iwpm_print_sockaddr(&pm_msg->rem_addr, + "query_mapping: Remote sockaddr:"); + iwpm_print_sockaddr(&pm_msg->mapped_rem_addr, + "query_mapping: Mapped remote sockaddr:"); +query_mapping_response_exit: + nlmsg_request->request_done = 1; + /* always for found request */ + kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request); + barrier(); + wake_up(&nlmsg_request->waitq); + return 0; +} +EXPORT_SYMBOL(iwpm_add_and_query_mapping_cb); + +/* netlink attribute policy for the received request for mapping info */ +static const struct nla_policy resp_mapinfo_policy[IWPM_NLA_MAPINFO_REQ_MAX] = { + [IWPM_NLA_MAPINFO_ULIB_NAME] = { .type = NLA_STRING, + .len = IWPM_ULIBNAME_SIZE - 1 }, + [IWPM_NLA_MAPINFO_ULIB_VER] = { .type = NLA_U16 } +}; + +/* + * iwpm_mapping_info_cb - Process a port mapper request for mapping info + */ +int iwpm_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct nlattr *nltb[IWPM_NLA_MAPINFO_REQ_MAX]; + const char *msg_type = "Mapping Info response"; + int iwpm_pid; + u8 nl_client; + char *iwpm_name; + u16 iwpm_version; + int ret = -EINVAL; + + if (iwpm_parse_nlmsg(cb, IWPM_NLA_MAPINFO_REQ_MAX, + resp_mapinfo_policy, nltb, msg_type)) { + pr_info("%s: Unable to parse nlmsg\n", __func__); + return ret; + } + iwpm_name = (char *)nla_data(nltb[IWPM_NLA_MAPINFO_ULIB_NAME]); + iwpm_version = nla_get_u16(nltb[IWPM_NLA_MAPINFO_ULIB_VER]); + if (strcmp(iwpm_ulib_name, iwpm_name) || + iwpm_version != iwpm_ulib_version) { + pr_info("%s: Invalid port mapper name = %s version = %d\n", + __func__, iwpm_name, iwpm_version); + return ret; + } + nl_client = RDMA_NL_GET_CLIENT(cb->nlh->nlmsg_type); + if (!iwpm_valid_client(nl_client)) { + pr_info("%s: Invalid port mapper client = %d\n", + __func__, nl_client); + return ret; + } + iwpm_set_registered(nl_client, 0); + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + if (!iwpm_mapinfo_available()) + return 0; + iwpm_pid = cb->nlh->nlmsg_pid; + pr_debug("%s: iWarp Port Mapper (pid = %d) is available!\n", + __func__, iwpm_pid); + ret = iwpm_send_mapinfo(nl_client, iwpm_pid); + return ret; +} +EXPORT_SYMBOL(iwpm_mapping_info_cb); + +/* netlink attribute policy for the received mapping info ack */ +static const struct nla_policy ack_mapinfo_policy[IWPM_NLA_MAPINFO_NUM_MAX] = { + [IWPM_NLA_MAPINFO_SEQ] = { .type = NLA_U32 }, + [IWPM_NLA_MAPINFO_SEND_NUM] = { .type = NLA_U32 }, + [IWPM_NLA_MAPINFO_ACK_NUM] = { .type = NLA_U32 } +}; + +/* + * iwpm_ack_mapping_info_cb - Process a port mapper ack for + * the provided mapping info records + */ +int iwpm_ack_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct nlattr *nltb[IWPM_NLA_MAPINFO_NUM_MAX]; + u32 mapinfo_send, mapinfo_ack; + const char *msg_type = "Mapping Info Ack"; + + if (iwpm_parse_nlmsg(cb, IWPM_NLA_MAPINFO_NUM_MAX, + ack_mapinfo_policy, nltb, msg_type)) + return -EINVAL; + mapinfo_send = nla_get_u32(nltb[IWPM_NLA_MAPINFO_SEND_NUM]); + mapinfo_ack = nla_get_u32(nltb[IWPM_NLA_MAPINFO_ACK_NUM]); + if (mapinfo_ack != mapinfo_send) + pr_info("%s: Invalid mapinfo number (sent = %u ack-ed = %u)\n", + __func__, mapinfo_send, mapinfo_ack); + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + return 0; +} +EXPORT_SYMBOL(iwpm_ack_mapping_info_cb); + +/* netlink attribute policy for the received port mapper error message */ +static const struct nla_policy map_error_policy[IWPM_NLA_ERR_MAX] = { + [IWPM_NLA_ERR_SEQ] = { .type = NLA_U32 }, + [IWPM_NLA_ERR_CODE] = { .type = NLA_U16 }, +}; + +/* + * iwpm_mapping_error_cb - Process a port mapper error message + */ +int iwpm_mapping_error_cb(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct iwpm_nlmsg_request *nlmsg_request = NULL; + int nl_client = RDMA_NL_GET_CLIENT(cb->nlh->nlmsg_type); + struct nlattr *nltb[IWPM_NLA_ERR_MAX]; + u32 msg_seq; + u16 err_code; + const char *msg_type = "Mapping Error Msg"; + + if (iwpm_parse_nlmsg(cb, IWPM_NLA_ERR_MAX, + map_error_policy, nltb, msg_type)) + return -EINVAL; + + msg_seq = nla_get_u32(nltb[IWPM_NLA_ERR_SEQ]); + err_code = nla_get_u16(nltb[IWPM_NLA_ERR_CODE]); + pr_info("%s: Received msg seq = %u err code = %u client = %d\n", + __func__, msg_seq, err_code, nl_client); + /* look for nlmsg_request */ + nlmsg_request = iwpm_find_nlmsg_request(msg_seq); + if (!nlmsg_request) { + /* not all errors have associated requests */ + pr_debug("Could not find matching req (seq = %u)\n", msg_seq); + return 0; + } + atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq); + nlmsg_request->err_code = err_code; + nlmsg_request->request_done = 1; + /* always for found request */ + kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request); + barrier(); + wake_up(&nlmsg_request->waitq); + return 0; +} +EXPORT_SYMBOL(iwpm_mapping_error_cb); diff --git a/drivers/infiniband/core/iwpm_util.c b/drivers/infiniband/core/iwpm_util.c new file mode 100644 index 000000000000..69e9f84c1605 --- /dev/null +++ b/drivers/infiniband/core/iwpm_util.c @@ -0,0 +1,607 @@ +/* + * Copyright (c) 2014 Chelsio, Inc. All rights reserved. + * Copyright (c) 2014 Intel Corporation. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "iwpm_util.h" + +#define IWPM_HASH_BUCKET_SIZE 512 +#define IWPM_HASH_BUCKET_MASK (IWPM_HASH_BUCKET_SIZE - 1) + +static LIST_HEAD(iwpm_nlmsg_req_list); +static DEFINE_SPINLOCK(iwpm_nlmsg_req_lock); + +static struct hlist_head *iwpm_hash_bucket; +static DEFINE_SPINLOCK(iwpm_mapinfo_lock); + +static DEFINE_MUTEX(iwpm_admin_lock); +static struct iwpm_admin_data iwpm_admin; + +int iwpm_init(u8 nl_client) +{ + if (iwpm_valid_client(nl_client)) + return -EINVAL; + mutex_lock(&iwpm_admin_lock); + if (atomic_read(&iwpm_admin.refcount) == 0) { + iwpm_hash_bucket = kzalloc(IWPM_HASH_BUCKET_SIZE * + sizeof(struct hlist_head), GFP_KERNEL); + if (!iwpm_hash_bucket) { + mutex_unlock(&iwpm_admin_lock); + pr_err("%s Unable to create mapinfo hash table\n", __func__); + return -ENOMEM; + } + } + atomic_inc(&iwpm_admin.refcount); + mutex_unlock(&iwpm_admin_lock); + iwpm_set_valid(nl_client, 1); + return 0; +} +EXPORT_SYMBOL(iwpm_init); + +static void free_hash_bucket(void); + +int iwpm_exit(u8 nl_client) +{ + + if (!iwpm_valid_client(nl_client)) + return -EINVAL; + mutex_lock(&iwpm_admin_lock); + if (atomic_read(&iwpm_admin.refcount) == 0) { + mutex_unlock(&iwpm_admin_lock); + pr_err("%s Incorrect usage - negative refcount\n", __func__); + return -EINVAL; + } + if (atomic_dec_and_test(&iwpm_admin.refcount)) { + free_hash_bucket(); + pr_debug("%s: Mapinfo hash table is destroyed\n", __func__); + } + mutex_unlock(&iwpm_admin_lock); + iwpm_set_valid(nl_client, 0); + return 0; +} +EXPORT_SYMBOL(iwpm_exit); + +static struct hlist_head *get_hash_bucket_head(struct sockaddr_storage *, + struct sockaddr_storage *); + +int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, + struct sockaddr_storage *mapped_sockaddr, + u8 nl_client) +{ + struct hlist_head *hash_bucket_head; + struct iwpm_mapping_info *map_info; + unsigned long flags; + + if (!iwpm_valid_client(nl_client)) + return -EINVAL; + map_info = kzalloc(sizeof(struct iwpm_mapping_info), GFP_KERNEL); + if (!map_info) { + pr_err("%s: Unable to allocate a mapping info\n", __func__); + return -ENOMEM; + } + memcpy(&map_info->local_sockaddr, local_sockaddr, + sizeof(struct sockaddr_storage)); + memcpy(&map_info->mapped_sockaddr, mapped_sockaddr, + sizeof(struct sockaddr_storage)); + map_info->nl_client = nl_client; + + spin_lock_irqsave(&iwpm_mapinfo_lock, flags); + if (iwpm_hash_bucket) { + hash_bucket_head = get_hash_bucket_head( + &map_info->local_sockaddr, + &map_info->mapped_sockaddr); + hlist_add_head(&map_info->hlist_node, hash_bucket_head); + } + spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); + return 0; +} +EXPORT_SYMBOL(iwpm_create_mapinfo); + +int iwpm_remove_mapinfo(struct sockaddr_storage *local_sockaddr, + struct sockaddr_storage *mapped_local_addr) +{ + struct hlist_node *tmp_hlist_node; + struct hlist_head *hash_bucket_head; + struct iwpm_mapping_info *map_info = NULL; + unsigned long flags; + int ret = -EINVAL; + + spin_lock_irqsave(&iwpm_mapinfo_lock, flags); + if (iwpm_hash_bucket) { + hash_bucket_head = get_hash_bucket_head( + local_sockaddr, + mapped_local_addr); + hlist_for_each_entry_safe(map_info, tmp_hlist_node, + hash_bucket_head, hlist_node) { + + if (!iwpm_compare_sockaddr(&map_info->mapped_sockaddr, + mapped_local_addr)) { + + hlist_del_init(&map_info->hlist_node); + kfree(map_info); + ret = 0; + break; + } + } + } + spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); + return ret; +} +EXPORT_SYMBOL(iwpm_remove_mapinfo); + +static void free_hash_bucket(void) +{ + struct hlist_node *tmp_hlist_node; + struct iwpm_mapping_info *map_info; + unsigned long flags; + int i; + + /* remove all the mapinfo data from the list */ + spin_lock_irqsave(&iwpm_mapinfo_lock, flags); + for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) { + hlist_for_each_entry_safe(map_info, tmp_hlist_node, + &iwpm_hash_bucket[i], hlist_node) { + + hlist_del_init(&map_info->hlist_node); + kfree(map_info); + } + } + /* free the hash list */ + kfree(iwpm_hash_bucket); + iwpm_hash_bucket = NULL; + spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); +} + +struct iwpm_nlmsg_request *iwpm_get_nlmsg_request(__u32 nlmsg_seq, + u8 nl_client, gfp_t gfp) +{ + struct iwpm_nlmsg_request *nlmsg_request = NULL; + unsigned long flags; + + nlmsg_request = kzalloc(sizeof(struct iwpm_nlmsg_request), gfp); + if (!nlmsg_request) { + pr_err("%s Unable to allocate a nlmsg_request\n", __func__); + return NULL; + } + spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags); + list_add_tail(&nlmsg_request->inprocess_list, &iwpm_nlmsg_req_list); + spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags); + + kref_init(&nlmsg_request->kref); + kref_get(&nlmsg_request->kref); + nlmsg_request->nlmsg_seq = nlmsg_seq; + nlmsg_request->nl_client = nl_client; + nlmsg_request->request_done = 0; + nlmsg_request->err_code = 0; + return nlmsg_request; +} + +void iwpm_free_nlmsg_request(struct kref *kref) +{ + struct iwpm_nlmsg_request *nlmsg_request; + unsigned long flags; + + nlmsg_request = container_of(kref, struct iwpm_nlmsg_request, kref); + + spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags); + list_del_init(&nlmsg_request->inprocess_list); + spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags); + + if (!nlmsg_request->request_done) + pr_debug("%s Freeing incomplete nlmsg request (seq = %u).\n", + __func__, nlmsg_request->nlmsg_seq); + kfree(nlmsg_request); +} + +struct iwpm_nlmsg_request *iwpm_find_nlmsg_request(__u32 echo_seq) +{ + struct iwpm_nlmsg_request *nlmsg_request; + struct iwpm_nlmsg_request *found_request = NULL; + unsigned long flags; + + spin_lock_irqsave(&iwpm_nlmsg_req_lock, flags); + list_for_each_entry(nlmsg_request, &iwpm_nlmsg_req_list, + inprocess_list) { + if (nlmsg_request->nlmsg_seq == echo_seq) { + found_request = nlmsg_request; + kref_get(&nlmsg_request->kref); + break; + } + } + spin_unlock_irqrestore(&iwpm_nlmsg_req_lock, flags); + return found_request; +} + +int iwpm_wait_complete_req(struct iwpm_nlmsg_request *nlmsg_request) +{ + int ret; + init_waitqueue_head(&nlmsg_request->waitq); + + ret = wait_event_timeout(nlmsg_request->waitq, + (nlmsg_request->request_done != 0), IWPM_NL_TIMEOUT); + if (!ret) { + ret = -EINVAL; + pr_info("%s: Timeout %d sec for netlink request (seq = %u)\n", + __func__, (IWPM_NL_TIMEOUT/HZ), nlmsg_request->nlmsg_seq); + } else { + ret = nlmsg_request->err_code; + } + kref_put(&nlmsg_request->kref, iwpm_free_nlmsg_request); + return ret; +} + +int iwpm_get_nlmsg_seq(void) +{ + return atomic_inc_return(&iwpm_admin.nlmsg_seq); +} + +int iwpm_valid_client(u8 nl_client) +{ + if (nl_client >= RDMA_NL_NUM_CLIENTS) + return 0; + return iwpm_admin.client_list[nl_client]; +} + +void iwpm_set_valid(u8 nl_client, int valid) +{ + if (nl_client >= RDMA_NL_NUM_CLIENTS) + return; + iwpm_admin.client_list[nl_client] = valid; +} + +/* valid client */ +int iwpm_registered_client(u8 nl_client) +{ + return iwpm_admin.reg_list[nl_client]; +} + +/* valid client */ +void iwpm_set_registered(u8 nl_client, int reg) +{ + iwpm_admin.reg_list[nl_client] = reg; +} + +int iwpm_compare_sockaddr(struct sockaddr_storage *a_sockaddr, + struct sockaddr_storage *b_sockaddr) +{ + if (a_sockaddr->ss_family != b_sockaddr->ss_family) + return 1; + if (a_sockaddr->ss_family == AF_INET) { + struct sockaddr_in *a4_sockaddr = + (struct sockaddr_in *)a_sockaddr; + struct sockaddr_in *b4_sockaddr = + (struct sockaddr_in *)b_sockaddr; + if (!memcmp(&a4_sockaddr->sin_addr, + &b4_sockaddr->sin_addr, sizeof(struct in_addr)) + && a4_sockaddr->sin_port == b4_sockaddr->sin_port) + return 0; + + } else if (a_sockaddr->ss_family == AF_INET6) { + struct sockaddr_in6 *a6_sockaddr = + (struct sockaddr_in6 *)a_sockaddr; + struct sockaddr_in6 *b6_sockaddr = + (struct sockaddr_in6 *)b_sockaddr; + if (!memcmp(&a6_sockaddr->sin6_addr, + &b6_sockaddr->sin6_addr, sizeof(struct in6_addr)) + && a6_sockaddr->sin6_port == b6_sockaddr->sin6_port) + return 0; + + } else { + pr_err("%s: Invalid sockaddr family\n", __func__); + } + return 1; +} + +struct sk_buff *iwpm_create_nlmsg(u32 nl_op, struct nlmsghdr **nlh, + int nl_client) +{ + struct sk_buff *skb = NULL; + + skb = dev_alloc_skb(NLMSG_GOODSIZE); + if (!skb) { + pr_err("%s Unable to allocate skb\n", __func__); + goto create_nlmsg_exit; + } + if (!(ibnl_put_msg(skb, nlh, 0, 0, nl_client, nl_op, + NLM_F_REQUEST))) { + pr_warn("%s: Unable to put the nlmsg header\n", __func__); + dev_kfree_skb(skb); + skb = NULL; + } +create_nlmsg_exit: + return skb; +} + +int iwpm_parse_nlmsg(struct netlink_callback *cb, int policy_max, + const struct nla_policy *nlmsg_policy, + struct nlattr *nltb[], const char *msg_type) +{ + int nlh_len = 0; + int ret; + const char *err_str = ""; + + ret = nlmsg_validate(cb->nlh, nlh_len, policy_max-1, nlmsg_policy); + if (ret) { + err_str = "Invalid attribute"; + goto parse_nlmsg_error; + } + ret = nlmsg_parse(cb->nlh, nlh_len, nltb, policy_max-1, nlmsg_policy); + if (ret) { + err_str = "Unable to parse the nlmsg"; + goto parse_nlmsg_error; + } + ret = iwpm_validate_nlmsg_attr(nltb, policy_max); + if (ret) { + err_str = "Invalid NULL attribute"; + goto parse_nlmsg_error; + } + return 0; +parse_nlmsg_error: + pr_warn("%s: %s (msg type %s ret = %d)\n", + __func__, err_str, msg_type, ret); + return ret; +} + +void iwpm_print_sockaddr(struct sockaddr_storage *sockaddr, char *msg) +{ + struct sockaddr_in6 *sockaddr_v6; + struct sockaddr_in *sockaddr_v4; + + switch (sockaddr->ss_family) { + case AF_INET: + sockaddr_v4 = (struct sockaddr_in *)sockaddr; + pr_debug("%s IPV4 %pI4: %u(0x%04X)\n", + msg, &sockaddr_v4->sin_addr, + ntohs(sockaddr_v4->sin_port), + ntohs(sockaddr_v4->sin_port)); + break; + case AF_INET6: + sockaddr_v6 = (struct sockaddr_in6 *)sockaddr; + pr_debug("%s IPV6 %pI6: %u(0x%04X)\n", + msg, &sockaddr_v6->sin6_addr, + ntohs(sockaddr_v6->sin6_port), + ntohs(sockaddr_v6->sin6_port)); + break; + default: + break; + } +} + +static u32 iwpm_ipv6_jhash(struct sockaddr_in6 *ipv6_sockaddr) +{ + u32 ipv6_hash = jhash(&ipv6_sockaddr->sin6_addr, sizeof(struct in6_addr), 0); + u32 hash = jhash_2words(ipv6_hash, (__force u32) ipv6_sockaddr->sin6_port, 0); + return hash; +} + +static u32 iwpm_ipv4_jhash(struct sockaddr_in *ipv4_sockaddr) +{ + u32 ipv4_hash = jhash(&ipv4_sockaddr->sin_addr, sizeof(struct in_addr), 0); + u32 hash = jhash_2words(ipv4_hash, (__force u32) ipv4_sockaddr->sin_port, 0); + return hash; +} + +static struct hlist_head *get_hash_bucket_head(struct sockaddr_storage + *local_sockaddr, + struct sockaddr_storage + *mapped_sockaddr) +{ + u32 local_hash, mapped_hash, hash; + + if (local_sockaddr->ss_family == AF_INET) { + local_hash = iwpm_ipv4_jhash((struct sockaddr_in *) local_sockaddr); + mapped_hash = iwpm_ipv4_jhash((struct sockaddr_in *) mapped_sockaddr); + + } else if (local_sockaddr->ss_family == AF_INET6) { + local_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) local_sockaddr); + mapped_hash = iwpm_ipv6_jhash((struct sockaddr_in6 *) mapped_sockaddr); + } else { + pr_err("%s: Invalid sockaddr family\n", __func__); + return NULL; + } + + if (local_hash == mapped_hash) /* if port mapper isn't available */ + hash = local_hash; + else + hash = jhash_2words(local_hash, mapped_hash, 0); + + return &iwpm_hash_bucket[hash & IWPM_HASH_BUCKET_MASK]; +} + +static int send_mapinfo_num(u32 mapping_num, u8 nl_client, int iwpm_pid) +{ + struct sk_buff *skb = NULL; + struct nlmsghdr *nlh; + u32 msg_seq; + const char *err_str = ""; + int ret = -EINVAL; + + skb = iwpm_create_nlmsg(RDMA_NL_IWPM_MAPINFO_NUM, &nlh, nl_client); + if (!skb) { + err_str = "Unable to create a nlmsg"; + goto mapinfo_num_error; + } + nlh->nlmsg_seq = iwpm_get_nlmsg_seq(); + msg_seq = 0; + err_str = "Unable to put attribute of mapinfo number nlmsg"; + ret = ibnl_put_attr(skb, nlh, sizeof(u32), &msg_seq, IWPM_NLA_MAPINFO_SEQ); + if (ret) + goto mapinfo_num_error; + ret = ibnl_put_attr(skb, nlh, sizeof(u32), + &mapping_num, IWPM_NLA_MAPINFO_SEND_NUM); + if (ret) + goto mapinfo_num_error; + ret = ibnl_unicast(skb, nlh, iwpm_pid); + if (ret) { + skb = NULL; + err_str = "Unable to send a nlmsg"; + goto mapinfo_num_error; + } + pr_debug("%s: Sent mapping number = %d\n", __func__, mapping_num); + return 0; +mapinfo_num_error: + pr_info("%s: %s\n", __func__, err_str); + if (skb) + dev_kfree_skb(skb); + return ret; +} + +static int send_nlmsg_done(struct sk_buff *skb, u8 nl_client, int iwpm_pid) +{ + struct nlmsghdr *nlh = NULL; + int ret = 0; + + if (!skb) + return ret; + if (!(ibnl_put_msg(skb, &nlh, 0, 0, nl_client, + RDMA_NL_IWPM_MAPINFO, NLM_F_MULTI))) { + pr_warn("%s Unable to put NLMSG_DONE\n", __func__); + return -ENOMEM; + } + nlh->nlmsg_type = NLMSG_DONE; + ret = ibnl_unicast(skb, (struct nlmsghdr *)skb->data, iwpm_pid); + if (ret) + pr_warn("%s Unable to send a nlmsg\n", __func__); + return ret; +} + +int iwpm_send_mapinfo(u8 nl_client, int iwpm_pid) +{ + struct iwpm_mapping_info *map_info; + struct sk_buff *skb = NULL; + struct nlmsghdr *nlh; + int skb_num = 0, mapping_num = 0; + int i = 0, nlmsg_bytes = 0; + unsigned long flags; + const char *err_str = ""; + int ret; + + skb = dev_alloc_skb(NLMSG_GOODSIZE); + if (!skb) { + ret = -ENOMEM; + err_str = "Unable to allocate skb"; + goto send_mapping_info_exit; + } + skb_num++; + spin_lock_irqsave(&iwpm_mapinfo_lock, flags); + for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) { + hlist_for_each_entry(map_info, &iwpm_hash_bucket[i], + hlist_node) { + if (map_info->nl_client != nl_client) + continue; + nlh = NULL; + if (!(ibnl_put_msg(skb, &nlh, 0, 0, nl_client, + RDMA_NL_IWPM_MAPINFO, NLM_F_MULTI))) { + ret = -ENOMEM; + err_str = "Unable to put the nlmsg header"; + goto send_mapping_info_unlock; + } + err_str = "Unable to put attribute of the nlmsg"; + ret = ibnl_put_attr(skb, nlh, + sizeof(struct sockaddr_storage), + &map_info->local_sockaddr, + IWPM_NLA_MAPINFO_LOCAL_ADDR); + if (ret) + goto send_mapping_info_unlock; + + ret = ibnl_put_attr(skb, nlh, + sizeof(struct sockaddr_storage), + &map_info->mapped_sockaddr, + IWPM_NLA_MAPINFO_MAPPED_ADDR); + if (ret) + goto send_mapping_info_unlock; + + iwpm_print_sockaddr(&map_info->local_sockaddr, + "send_mapping_info: Local sockaddr:"); + iwpm_print_sockaddr(&map_info->mapped_sockaddr, + "send_mapping_info: Mapped local sockaddr:"); + mapping_num++; + nlmsg_bytes += nlh->nlmsg_len; + + /* check if all mappings can fit in one skb */ + if (NLMSG_GOODSIZE - nlmsg_bytes < nlh->nlmsg_len * 2) { + /* and leave room for NLMSG_DONE */ + nlmsg_bytes = 0; + skb_num++; + spin_unlock_irqrestore(&iwpm_mapinfo_lock, + flags); + /* send the skb */ + ret = send_nlmsg_done(skb, nl_client, iwpm_pid); + skb = NULL; + if (ret) { + err_str = "Unable to send map info"; + goto send_mapping_info_exit; + } + if (skb_num == IWPM_MAPINFO_SKB_COUNT) { + ret = -ENOMEM; + err_str = "Insufficient skbs for map info"; + goto send_mapping_info_exit; + } + skb = dev_alloc_skb(NLMSG_GOODSIZE); + if (!skb) { + ret = -ENOMEM; + err_str = "Unable to allocate skb"; + goto send_mapping_info_exit; + } + spin_lock_irqsave(&iwpm_mapinfo_lock, flags); + } + } + } +send_mapping_info_unlock: + spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); +send_mapping_info_exit: + if (ret) { + pr_warn("%s: %s (ret = %d)\n", __func__, err_str, ret); + if (skb) + dev_kfree_skb(skb); + return ret; + } + send_nlmsg_done(skb, nl_client, iwpm_pid); + return send_mapinfo_num(mapping_num, nl_client, iwpm_pid); +} + +int iwpm_mapinfo_available(void) +{ + unsigned long flags; + int full_bucket = 0, i = 0; + + spin_lock_irqsave(&iwpm_mapinfo_lock, flags); + if (iwpm_hash_bucket) { + for (i = 0; i < IWPM_HASH_BUCKET_SIZE; i++) { + if (!hlist_empty(&iwpm_hash_bucket[i])) { + full_bucket = 1; + break; + } + } + } + spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); + return full_bucket; +} diff --git a/drivers/infiniband/core/iwpm_util.h b/drivers/infiniband/core/iwpm_util.h new file mode 100644 index 000000000000..9777c869a140 --- /dev/null +++ b/drivers/infiniband/core/iwpm_util.h @@ -0,0 +1,238 @@ +/* + * Copyright (c) 2014 Intel Corporation. All rights reserved. + * Copyright (c) 2014 Chelsio, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _IWPM_UTIL_H +#define _IWPM_UTIL_H + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +#define IWPM_NL_RETRANS 3 +#define IWPM_NL_TIMEOUT (10*HZ) +#define IWPM_MAPINFO_SKB_COUNT 20 + +#define IWPM_PID_UNDEFINED -1 +#define IWPM_PID_UNAVAILABLE -2 + +struct iwpm_nlmsg_request { + struct list_head inprocess_list; + __u32 nlmsg_seq; + void *req_buffer; + u8 nl_client; + u8 request_done; + u16 err_code; + wait_queue_head_t waitq; + struct kref kref; +}; + +struct iwpm_mapping_info { + struct hlist_node hlist_node; + struct sockaddr_storage local_sockaddr; + struct sockaddr_storage mapped_sockaddr; + u8 nl_client; +}; + +struct iwpm_admin_data { + atomic_t refcount; + atomic_t nlmsg_seq; + int client_list[RDMA_NL_NUM_CLIENTS]; + int reg_list[RDMA_NL_NUM_CLIENTS]; +}; + +/** + * iwpm_get_nlmsg_request - Allocate and initialize netlink message request + * @nlmsg_seq: Sequence number of the netlink message + * @nl_client: The index of the netlink client + * @gfp: Indicates how the memory for the request should be allocated + * + * Returns the newly allocated netlink request object if successful, + * otherwise returns NULL + */ +struct iwpm_nlmsg_request *iwpm_get_nlmsg_request(__u32 nlmsg_seq, + u8 nl_client, gfp_t gfp); + +/** + * iwpm_free_nlmsg_request - Deallocate netlink message request + * @kref: Holds reference of netlink message request + */ +void iwpm_free_nlmsg_request(struct kref *kref); + +/** + * iwpm_find_nlmsg_request - Find netlink message request in the request list + * @echo_seq: Sequence number of the netlink request to find + * + * Returns the found netlink message request, + * if not found, returns NULL + */ +struct iwpm_nlmsg_request *iwpm_find_nlmsg_request(__u32 echo_seq); + +/** + * iwpm_wait_complete_req - Block while servicing the netlink request + * @nlmsg_request: Netlink message request to service + * + * Wakes up, after the request is completed or expired + * Returns 0 if the request is complete without error + */ +int iwpm_wait_complete_req(struct iwpm_nlmsg_request *nlmsg_request); + +/** + * iwpm_get_nlmsg_seq - Get the sequence number for a netlink + * message to send to the port mapper + * + * Returns the sequence number for the netlink message. + */ +int iwpm_get_nlmsg_seq(void); + +/** + * iwpm_valid_client - Check if the port mapper client is valid + * @nl_client: The index of the netlink client + * + * Valid clients need to call iwpm_init() before using + * the port mapper + */ +int iwpm_valid_client(u8 nl_client); + +/** + * iwpm_set_valid - Set the port mapper client to valid or not + * @nl_client: The index of the netlink client + * @valid: 1 if valid or 0 if invalid + */ +void iwpm_set_valid(u8 nl_client, int valid); + +/** + * iwpm_registered_client - Check if the port mapper client is registered + * @nl_client: The index of the netlink client + * + * Call iwpm_register_pid() to register a client + */ +int iwpm_registered_client(u8 nl_client); + +/** + * iwpm_set_registered - Set the port mapper client to registered or not + * @nl_client: The index of the netlink client + * @reg: 1 if registered or 0 if not + */ +void iwpm_set_registered(u8 nl_client, int reg); + +/** + * iwpm_send_mapinfo - Send local and mapped IPv4/IPv6 address info of + * a client to the user space port mapper + * @nl_client: The index of the netlink client + * @iwpm_pid: The pid of the user space port mapper + * + * If successful, returns the number of sent mapping info records + */ +int iwpm_send_mapinfo(u8 nl_client, int iwpm_pid); + +/** + * iwpm_mapinfo_available - Check if any mapping info records is available + * in the hash table + * + * Returns 1 if mapping information is available, otherwise returns 0 + */ +int iwpm_mapinfo_available(void); + +/** + * iwpm_compare_sockaddr - Compare two sockaddr storage structs + * + * Returns 0 if they are holding the same ip/tcp address info, + * otherwise returns 1 + */ +int iwpm_compare_sockaddr(struct sockaddr_storage *a_sockaddr, + struct sockaddr_storage *b_sockaddr); + +/** + * iwpm_validate_nlmsg_attr - Check for NULL netlink attributes + * @nltb: Holds address of each netlink message attributes + * @nla_count: Number of netlink message attributes + * + * Returns error if any of the nla_count attributes is NULL + */ +static inline int iwpm_validate_nlmsg_attr(struct nlattr *nltb[], + int nla_count) +{ + int i; + for (i = 1; i < nla_count; i++) { + if (!nltb[i]) + return -EINVAL; + } + return 0; +} + +/** + * iwpm_create_nlmsg - Allocate skb and form a netlink message + * @nl_op: Netlink message opcode + * @nlh: Holds address of the netlink message header in skb + * @nl_client: The index of the netlink client + * + * Returns the newly allcated skb, or NULL if the tailroom of the skb + * is insufficient to store the message header and payload + */ +struct sk_buff *iwpm_create_nlmsg(u32 nl_op, struct nlmsghdr **nlh, + int nl_client); + +/** + * iwpm_parse_nlmsg - Validate and parse the received netlink message + * @cb: Netlink callback structure + * @policy_max: Maximum attribute type to be expected + * @nlmsg_policy: Validation policy + * @nltb: Array to store policy_max parsed elements + * @msg_type: Type of netlink message + * + * Returns 0 on success or a negative error code + */ +int iwpm_parse_nlmsg(struct netlink_callback *cb, int policy_max, + const struct nla_policy *nlmsg_policy, + struct nlattr *nltb[], const char *msg_type); + +/** + * iwpm_print_sockaddr - Print IPv4/IPv6 address and TCP port + * @sockaddr: Socket address to print + * @msg: Message to print + */ +void iwpm_print_sockaddr(struct sockaddr_storage *sockaddr, char *msg); +#endif diff --git a/drivers/infiniband/core/netlink.c b/drivers/infiniband/core/netlink.c index a1e9cba84944..23dd5a5c7597 100644 --- a/drivers/infiniband/core/netlink.c +++ b/drivers/infiniband/core/netlink.c @@ -103,13 +103,13 @@ int ibnl_remove_client(int index) EXPORT_SYMBOL(ibnl_remove_client); void *ibnl_put_msg(struct sk_buff *skb, struct nlmsghdr **nlh, int seq, - int len, int client, int op) + int len, int client, int op, int flags) { unsigned char *prev_tail; prev_tail = skb_tail_pointer(skb); *nlh = nlmsg_put(skb, 0, seq, RDMA_NL_GET_TYPE(client, op), - len, NLM_F_MULTI); + len, flags); if (!*nlh) goto out_nlmsg_trim; (*nlh)->nlmsg_len = skb_tail_pointer(skb) - prev_tail; @@ -172,6 +172,20 @@ static void ibnl_rcv(struct sk_buff *skb) mutex_unlock(&ibnl_mutex); } +int ibnl_unicast(struct sk_buff *skb, struct nlmsghdr *nlh, + __u32 pid) +{ + return nlmsg_unicast(nls, skb, pid); +} +EXPORT_SYMBOL(ibnl_unicast); + +int ibnl_multicast(struct sk_buff *skb, struct nlmsghdr *nlh, + unsigned int group, gfp_t flags) +{ + return nlmsg_multicast(nls, skb, 0, group, flags); +} +EXPORT_SYMBOL(ibnl_multicast); + int __init ibnl_init(void) { struct netlink_kernel_cfg cfg = { diff --git a/include/rdma/iw_portmap.h b/include/rdma/iw_portmap.h new file mode 100644 index 000000000000..928b2775e992 --- /dev/null +++ b/include/rdma/iw_portmap.h @@ -0,0 +1,199 @@ +/* + * Copyright (c) 2014 Intel Corporation. All rights reserved. + * Copyright (c) 2014 Chelsio, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef _IW_PORTMAP_H +#define _IW_PORTMAP_H + +#define IWPM_ULIBNAME_SIZE 32 +#define IWPM_DEVNAME_SIZE 32 +#define IWPM_IFNAME_SIZE 16 +#define IWPM_IPADDR_SIZE 16 + +enum { + IWPM_INVALID_NLMSG_ERR = 10, + IWPM_CREATE_MAPPING_ERR, + IWPM_DUPLICATE_MAPPING_ERR, + IWPM_UNKNOWN_MAPPING_ERR, + IWPM_CLIENT_DEV_INFO_ERR, + IWPM_USER_LIB_INFO_ERR, + IWPM_REMOTE_QUERY_REJECT +}; + +struct iwpm_dev_data { + char dev_name[IWPM_DEVNAME_SIZE]; + char if_name[IWPM_IFNAME_SIZE]; +}; + +struct iwpm_sa_data { + struct sockaddr_storage loc_addr; + struct sockaddr_storage mapped_loc_addr; + struct sockaddr_storage rem_addr; + struct sockaddr_storage mapped_rem_addr; +}; + +/** + * iwpm_init - Allocate resources for the iwarp port mapper + * + * Should be called when network interface goes up. + */ +int iwpm_init(u8); + +/** + * iwpm_exit - Deallocate resources for the iwarp port mapper + * + * Should be called when network interface goes down. + */ +int iwpm_exit(u8); + +/** + * iwpm_valid_pid - Check if the userspace iwarp port mapper pid is valid + * + * Returns true if the pid is greater than zero, otherwise returns false + */ +int iwpm_valid_pid(void); + +/** + * iwpm_register_pid - Send a netlink query to userspace + * to get the iwarp port mapper pid + * @pm_msg: Contains driver info to send to the userspace port mapper + * @nl_client: The index of the netlink client + */ +int iwpm_register_pid(struct iwpm_dev_data *pm_msg, u8 nl_client); + +/** + * iwpm_add_mapping - Send a netlink add mapping request to + * the userspace port mapper + * @pm_msg: Contains the local ip/tcp address info to send + * @nl_client: The index of the netlink client + * + * If the request is successful, the pm_msg stores + * the port mapper response (mapped address info) + */ +int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client); + +/** + * iwpm_add_and_query_mapping - Send a netlink add and query mapping request + * to the userspace port mapper + * @pm_msg: Contains the local and remote ip/tcp address info to send + * @nl_client: The index of the netlink client + * + * If the request is successful, the pm_msg stores the + * port mapper response (mapped local and remote address info) + */ +int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client); + +/** + * iwpm_remove_mapping - Send a netlink remove mapping request + * to the userspace port mapper + * + * @local_addr: Local ip/tcp address to remove + * @nl_client: The index of the netlink client + */ +int iwpm_remove_mapping(struct sockaddr_storage *local_addr, u8 nl_client); + +/** + * iwpm_register_pid_cb - Process the port mapper response to + * iwpm_register_pid query + * @skb: + * @cb: Contains the received message (payload and netlink header) + * + * If successful, the function receives the userspace port mapper pid + * which is used in future communication with the port mapper + */ +int iwpm_register_pid_cb(struct sk_buff *, struct netlink_callback *); + +/** + * iwpm_add_mapping_cb - Process the port mapper response to + * iwpm_add_mapping request + * @skb: + * @cb: Contains the received message (payload and netlink header) + */ +int iwpm_add_mapping_cb(struct sk_buff *, struct netlink_callback *); + +/** + * iwpm_add_and_query_mapping_cb - Process the port mapper response to + * iwpm_add_and_query_mapping request + * @skb: + * @cb: Contains the received message (payload and netlink header) + */ +int iwpm_add_and_query_mapping_cb(struct sk_buff *, struct netlink_callback *); + +/** + * iwpm_mapping_error_cb - Process port mapper notification for error + * + * @skb: + * @cb: Contains the received message (payload and netlink header) + */ +int iwpm_mapping_error_cb(struct sk_buff *, struct netlink_callback *); + +/** + * iwpm_mapping_info_cb - Process a notification that the userspace + * port mapper daemon is started + * @skb: + * @cb: Contains the received message (payload and netlink header) + * + * Using the received port mapper pid, send all the local mapping + * info records to the userspace port mapper + */ +int iwpm_mapping_info_cb(struct sk_buff *, struct netlink_callback *); + +/** + * iwpm_ack_mapping_info_cb - Process the port mapper ack for + * the provided local mapping info records + * @skb: + * @cb: Contains the received message (payload and netlink header) + */ +int iwpm_ack_mapping_info_cb(struct sk_buff *, struct netlink_callback *); + +/** + * iwpm_create_mapinfo - Store local and mapped IPv4/IPv6 address + * info in a hash table + * @local_addr: Local ip/tcp address + * @mapped_addr: Mapped local ip/tcp address + * @nl_client: The index of the netlink client + */ +int iwpm_create_mapinfo(struct sockaddr_storage *local_addr, + struct sockaddr_storage *mapped_addr, u8 nl_client); + +/** + * iwpm_remove_mapinfo - Remove local and mapped IPv4/IPv6 address + * info from the hash table + * @local_addr: Local ip/tcp address + * @mapped_addr: Mapped local ip/tcp address + * + * Returns err code if mapping info is not found in the hash table, + * otherwise returns 0 + */ +int iwpm_remove_mapinfo(struct sockaddr_storage *local_addr, + struct sockaddr_storage *mapped_addr); + +#endif /* _IW_PORTMAP_H */ diff --git a/include/rdma/rdma_netlink.h b/include/rdma/rdma_netlink.h index e38de79eeb48..0790882e0c9b 100644 --- a/include/rdma/rdma_netlink.h +++ b/include/rdma/rdma_netlink.h @@ -43,7 +43,7 @@ int ibnl_remove_client(int index); * Returns the allocated buffer on success and NULL on failure. */ void *ibnl_put_msg(struct sk_buff *skb, struct nlmsghdr **nlh, int seq, - int len, int client, int op); + int len, int client, int op, int flags); /** * Put a new attribute in a supplied skb. * @skb: The netlink skb. @@ -56,4 +56,25 @@ void *ibnl_put_msg(struct sk_buff *skb, struct nlmsghdr **nlh, int seq, int ibnl_put_attr(struct sk_buff *skb, struct nlmsghdr *nlh, int len, void *data, int type); +/** + * Send the supplied skb to a specific userspace PID. + * @skb: The netlink skb + * @nlh: Header of the netlink message to send + * @pid: Userspace netlink process ID + * Returns 0 on success or a negative error code. + */ +int ibnl_unicast(struct sk_buff *skb, struct nlmsghdr *nlh, + __u32 pid); + +/** + * Send the supplied skb to a netlink group. + * @skb: The netlink skb + * @nlh: Header of the netlink message to send + * @group: Netlink group ID + * @flags: allocation flags + * Returns 0 on success or a negative error code. + */ +int ibnl_multicast(struct sk_buff *skb, struct nlmsghdr *nlh, + unsigned int group, gfp_t flags); + #endif /* _RDMA_NETLINK_H */ diff --git a/include/uapi/rdma/rdma_netlink.h b/include/uapi/rdma/rdma_netlink.h index 8297285b6288..de69170a30ce 100644 --- a/include/uapi/rdma/rdma_netlink.h +++ b/include/uapi/rdma/rdma_netlink.h @@ -4,7 +4,16 @@ #include enum { - RDMA_NL_RDMA_CM = 1 + RDMA_NL_RDMA_CM = 1, + RDMA_NL_NES, + RDMA_NL_C4IW, + RDMA_NL_NUM_CLIENTS +}; + +enum { + RDMA_NL_GROUP_CM = 1, + RDMA_NL_GROUP_IWPM, + RDMA_NL_NUM_GROUPS }; #define RDMA_NL_GET_CLIENT(type) ((type & (((1 << 6) - 1) << 10)) >> 10) @@ -22,6 +31,18 @@ enum { RDMA_NL_RDMA_CM_NUM_ATTR, }; +/* iwarp port mapper op-codes */ +enum { + RDMA_NL_IWPM_REG_PID = 0, + RDMA_NL_IWPM_ADD_MAPPING, + RDMA_NL_IWPM_QUERY_MAPPING, + RDMA_NL_IWPM_REMOVE_MAPPING, + RDMA_NL_IWPM_HANDLE_ERR, + RDMA_NL_IWPM_MAPINFO, + RDMA_NL_IWPM_MAPINFO_NUM, + RDMA_NL_IWPM_NUM_OPS +}; + struct rdma_cm_id_stats { __u32 qp_num; __u32 bound_dev_if; @@ -33,5 +54,78 @@ struct rdma_cm_id_stats { __u8 qp_type; }; +enum { + IWPM_NLA_REG_PID_UNSPEC = 0, + IWPM_NLA_REG_PID_SEQ, + IWPM_NLA_REG_IF_NAME, + IWPM_NLA_REG_IBDEV_NAME, + IWPM_NLA_REG_ULIB_NAME, + IWPM_NLA_REG_PID_MAX +}; + +enum { + IWPM_NLA_RREG_PID_UNSPEC = 0, + IWPM_NLA_RREG_PID_SEQ, + IWPM_NLA_RREG_IBDEV_NAME, + IWPM_NLA_RREG_ULIB_NAME, + IWPM_NLA_RREG_ULIB_VER, + IWPM_NLA_RREG_PID_ERR, + IWPM_NLA_RREG_PID_MAX + +}; + +enum { + IWPM_NLA_MANAGE_MAPPING_UNSPEC = 0, + IWPM_NLA_MANAGE_MAPPING_SEQ, + IWPM_NLA_MANAGE_ADDR, + IWPM_NLA_MANAGE_MAPPED_LOC_ADDR, + IWPM_NLA_RMANAGE_MAPPING_ERR, + IWPM_NLA_RMANAGE_MAPPING_MAX +}; + +#define IWPM_NLA_MANAGE_MAPPING_MAX 3 +#define IWPM_NLA_QUERY_MAPPING_MAX 4 +#define IWPM_NLA_MAPINFO_SEND_MAX 3 + +enum { + IWPM_NLA_QUERY_MAPPING_UNSPEC = 0, + IWPM_NLA_QUERY_MAPPING_SEQ, + IWPM_NLA_QUERY_LOCAL_ADDR, + IWPM_NLA_QUERY_REMOTE_ADDR, + IWPM_NLA_RQUERY_MAPPED_LOC_ADDR, + IWPM_NLA_RQUERY_MAPPED_REM_ADDR, + IWPM_NLA_RQUERY_MAPPING_ERR, + IWPM_NLA_RQUERY_MAPPING_MAX +}; + +enum { + IWPM_NLA_MAPINFO_REQ_UNSPEC = 0, + IWPM_NLA_MAPINFO_ULIB_NAME, + IWPM_NLA_MAPINFO_ULIB_VER, + IWPM_NLA_MAPINFO_REQ_MAX +}; + +enum { + IWPM_NLA_MAPINFO_UNSPEC = 0, + IWPM_NLA_MAPINFO_LOCAL_ADDR, + IWPM_NLA_MAPINFO_MAPPED_ADDR, + IWPM_NLA_MAPINFO_MAX +}; + +enum { + IWPM_NLA_MAPINFO_NUM_UNSPEC = 0, + IWPM_NLA_MAPINFO_SEQ, + IWPM_NLA_MAPINFO_SEND_NUM, + IWPM_NLA_MAPINFO_ACK_NUM, + IWPM_NLA_MAPINFO_NUM_MAX +}; + +enum { + IWPM_NLA_ERR_UNSPEC = 0, + IWPM_NLA_ERR_SEQ, + IWPM_NLA_ERR_CODE, + IWPM_NLA_ERR_MAX +}; + #endif /* _UAPI_RDMA_NETLINK_H */ -- cgit v1.2.3 From efd0f11d85e16d375dddeb77e8e78f43c67c5b13 Mon Sep 17 00:00:00 2001 From: Dmitry Popov Date: Wed, 11 Jun 2014 15:09:14 +0400 Subject: ip_vti: fix sparse warnings for VTI_ISVTI This patch fixes the following sparse warnings: net/ipv4/ip_tunnel.c:245:53: warning: restricted __be16 degrades to integer net/ipv4/ip_vti.c:321:19: warning: incorrect type in assignment (different base types) net/ipv4/ip_vti.c:321:19: expected restricted __be16 [addressable] [assigned] [usertype] i_flags net/ipv4/ip_vti.c:321:19: got int net/ipv4/ip_vti.c:447:24: warning: incorrect type in assignment (different base types) net/ipv4/ip_vti.c:447:24: expected restricted __be16 [usertype] i_flags net/ipv4/ip_vti.c:447:24: got int Since VTI_ISVTI is always used with ip_tunnel_parm->i_flags (which is __be16), we can __force cast VTI_ISVTI to __be16 in header file. Signed-off-by: Dmitry Popov Signed-off-by: David S. Miller --- include/uapi/linux/if_tunnel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index aee73d0611fb..3bce9e9d9f7c 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -100,7 +100,7 @@ enum { #define IFLA_GRE_MAX (__IFLA_GRE_MAX - 1) /* VTI-mode i_flags */ -#define VTI_ISVTI 0x0001 +#define VTI_ISVTI ((__force __be16)0x0001) enum { IFLA_VTI_UNSPEC, -- cgit v1.2.3 From 2bd0ae464a6cf7363bbf72c8545e0aa43caa57f0 Mon Sep 17 00:00:00 2001 From: "Wang, Xiaoming" Date: Thu, 12 Jun 2014 18:47:07 -0400 Subject: ALSA: compress: Cancel the optimization of compiler and fix the size of struct for all platform. Cancel the optimization of compiler for struct snd_compr_avail which size will be 0x1c in 32bit kernel while 0x20 in 64bit kernel under the optimizer. That will make compaction between 32bit and 64bit. So add packed to fix the size of struct snd_compr_avail to 0x1c for all platform. Signed-off-by: Zhang Dongxing Signed-off-by: xiaoming wang Acked-by: Vinod Koul Cc: Signed-off-by: Takashi Iwai --- include/uapi/sound/compress_offload.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi') diff --git a/include/uapi/sound/compress_offload.h b/include/uapi/sound/compress_offload.h index 5759810e1c1b..21eed488783f 100644 --- a/include/uapi/sound/compress_offload.h +++ b/include/uapi/sound/compress_offload.h @@ -80,7 +80,7 @@ struct snd_compr_tstamp { struct snd_compr_avail { __u64 avail; struct snd_compr_tstamp tstamp; -}; +} __attribute__((packed)); enum snd_compr_direction { SND_COMPRESS_PLAYBACK = 0, -- cgit v1.2.3 From 3d69bb6e4699251102dc145b7800dd012ddec375 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 13 Jun 2014 10:53:49 -0400 Subject: NVMe: Define Log Page constants Taken from the 1.1a version of the spec Signed-off-by: Matthew Wilcox --- include/uapi/linux/nvme.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/nvme.h b/include/uapi/linux/nvme.h index a6fb2a360577..29a7d8619d8d 100644 --- a/include/uapi/linux/nvme.h +++ b/include/uapi/linux/nvme.h @@ -306,6 +306,10 @@ enum { NVME_FEAT_WRITE_ATOMIC = 0x0a, NVME_FEAT_ASYNC_EVENT = 0x0b, NVME_FEAT_SW_PROGRESS = 0x0c, + NVME_LOG_ERROR = 0x01, + NVME_LOG_SMART = 0x02, + NVME_LOG_FW_SLOT = 0x03, + NVME_LOG_RESERVATION = 0x80, NVME_FWACT_REPL = (0 << 3), NVME_FWACT_REPL_ACTV = (1 << 3), NVME_FWACT_ACTV = (2 << 3), -- cgit v1.2.3 From cc68a8a5a4330a4bb72922d0c7a7044ae13ee692 Mon Sep 17 00:00:00 2001 From: Gerhard Heift Date: Thu, 30 Jan 2014 16:24:03 +0100 Subject: btrfs: new ioctl TREE_SEARCH_V2 This new ioctl call allows the user to supply a buffer of varying size in which a tree search can store its results. This is much more flexible if you want to receive items which are larger than the current fixed buffer of 3992 bytes or if you want to fetch more items at once. Items larger than this buffer are for example some of the type EXTENT_CSUM. Signed-off-by: Gerhard Heift Signed-off-by: Chris Mason Acked-by: David Sterba --- fs/btrfs/ioctl.c | 41 +++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/btrfs.h | 10 ++++++++++ 2 files changed, 51 insertions(+) (limited to 'include/uapi') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 393a543a519e..6ea15469c63f 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2172,6 +2172,45 @@ static noinline int btrfs_ioctl_tree_search(struct file *file, return ret; } +static noinline int btrfs_ioctl_tree_search_v2(struct file *file, + void __user *argp) +{ + struct btrfs_ioctl_search_args_v2 __user *uarg; + struct btrfs_ioctl_search_args_v2 args; + struct inode *inode; + int ret; + size_t buf_size; + const size_t buf_limit = 16 * 1024 * 1024; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + /* copy search header and buffer size */ + uarg = (struct btrfs_ioctl_search_args_v2 __user *)argp; + if (copy_from_user(&args, uarg, sizeof(args))) + return -EFAULT; + + buf_size = args.buf_size; + + if (buf_size < sizeof(struct btrfs_ioctl_search_header)) + return -EOVERFLOW; + + /* limit result size to 16MB */ + if (buf_size > buf_limit) + buf_size = buf_limit; + + inode = file_inode(file); + ret = search_ioctl(inode, &args.key, &buf_size, + (char *)(&uarg->buf[0])); + if (ret == 0 && copy_to_user(&uarg->key, &args.key, sizeof(args.key))) + ret = -EFAULT; + else if (ret == -EOVERFLOW && + copy_to_user(&uarg->buf_size, &buf_size, sizeof(buf_size))) + ret = -EFAULT; + + return ret; +} + /* * Search INODE_REFs to identify path name of 'dirid' directory * in a 'tree_id' tree. and sets path name to 'name'. @@ -5252,6 +5291,8 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_trans_end(file); case BTRFS_IOC_TREE_SEARCH: return btrfs_ioctl_tree_search(file, argp); + case BTRFS_IOC_TREE_SEARCH_V2: + return btrfs_ioctl_tree_search_v2(file, argp); case BTRFS_IOC_INO_LOOKUP: return btrfs_ioctl_ino_lookup(file, argp); case BTRFS_IOC_INO_PATHS: diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index 7554fd381a56..6f9c38ce45c7 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -306,6 +306,14 @@ struct btrfs_ioctl_search_args { char buf[BTRFS_SEARCH_ARGS_BUFSIZE]; }; +struct btrfs_ioctl_search_args_v2 { + struct btrfs_ioctl_search_key key; /* in/out - search parameters */ + __u64 buf_size; /* in - size of buffer + * out - on EOVERFLOW: needed size + * to store item */ + __u64 buf[0]; /* out - found items */ +}; + struct btrfs_ioctl_clone_range_args { __s64 src_fd; __u64 src_offset, src_length; @@ -558,6 +566,8 @@ static inline char *btrfs_err_str(enum btrfs_err_code err_code) struct btrfs_ioctl_defrag_range_args) #define BTRFS_IOC_TREE_SEARCH _IOWR(BTRFS_IOCTL_MAGIC, 17, \ struct btrfs_ioctl_search_args) +#define BTRFS_IOC_TREE_SEARCH_V2 _IOWR(BTRFS_IOCTL_MAGIC, 17, \ + struct btrfs_ioctl_search_args_v2) #define BTRFS_IOC_INO_LOOKUP _IOWR(BTRFS_IOCTL_MAGIC, 18, \ struct btrfs_ioctl_ino_lookup_args) #define BTRFS_IOC_DEFAULT_SUBVOL _IOW(BTRFS_IOCTL_MAGIC, 19, __u64) -- cgit v1.2.3 From f2af74123f8c5a735248547f4286a3adc28633c1 Mon Sep 17 00:00:00 2001 From: Michal Nazarewicz Date: Tue, 20 May 2014 09:38:03 -1000 Subject: tools: ffs-test: convert to new descriptor format fixing compilation error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit [ac8dde11: “usb: gadget: f_fs: Add flags to descriptors block”] which introduced a new descriptor format for FunctionFS removed the usb_functionfs_descs_head structure, which is still used by ffs-test. tool. Convert ffs-test by converting it to use the new header format. For testing kernels prior to 3.14 (when the new format was introduced) and parsing of the legacy headers in the new kernels, provide a compilation flag to make the tool use the old format. Finally, include information as to when the legacy FunctionFS headers format has been deprecated (which is also when the new one has been introduced). Reported-by: Lad, Prabhakar Signed-off-by: Michal Nazarewicz Signed-off-by: Felipe Balbi --- include/uapi/linux/usb/functionfs.h | 2 +- tools/usb/Makefile | 6 +++++- tools/usb/ffs-test.c | 20 ++++++++++++++++++-- 3 files changed, 24 insertions(+), 4 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/usb/functionfs.h b/include/uapi/linux/usb/functionfs.h index 2a4b4a72a4f9..ecb3a31f7ca6 100644 --- a/include/uapi/linux/usb/functionfs.h +++ b/include/uapi/linux/usb/functionfs.h @@ -53,7 +53,7 @@ struct usb_endpoint_descriptor_no_audio { * structure. Any flags that are not recognised cause the whole block to be * rejected with -ENOSYS. * - * Legacy descriptors format: + * Legacy descriptors format (deprecated as of 3.14): * * | off | name | type | description | * |-----+-----------+--------------+--------------------------------------| diff --git a/tools/usb/Makefile b/tools/usb/Makefile index acf2165c04e6..d576b3bac3cf 100644 --- a/tools/usb/Makefile +++ b/tools/usb/Makefile @@ -6,7 +6,11 @@ WARNINGS = -Wall -Wextra CFLAGS = $(WARNINGS) -g -I../include LDFLAGS = $(PTHREAD_LIBS) -all: testusb ffs-test +all: testusb ffs-test ffs-test-legacy + +ffs-test-legacy: ffs-test.c + $(CC) $(CFLAGS) -o $@ $^ $(LDFLAGS) -DUSE_LEGACY_DESC_HEAD + %: %.c $(CC) $(CFLAGS) -o $@ $^ $(LDFLAGS) diff --git a/tools/usb/ffs-test.c b/tools/usb/ffs-test.c index fe1e66b6ef40..74b353d9eb50 100644 --- a/tools/usb/ffs-test.c +++ b/tools/usb/ffs-test.c @@ -1,5 +1,5 @@ /* - * ffs-test.c.c -- user mode filesystem api for usb composite function + * ffs-test.c -- user mode filesystem api for usb composite function * * Copyright (C) 2010 Samsung Electronics * Author: Michal Nazarewicz @@ -21,6 +21,8 @@ /* $(CROSS_COMPILE)cc -Wall -Wextra -g -o ffs-test ffs-test.c -lpthread */ +/* Uncomment to make the tool use legacy FFS descriptor headers. */ +/* #define USE_LEGACY_DESC_HEAD */ #define _BSD_SOURCE /* for endian.h */ @@ -106,7 +108,15 @@ static void _msg(unsigned level, const char *fmt, ...) /******************** Descriptors and Strings *******************************/ static const struct { - struct usb_functionfs_descs_head header; + struct { + __le32 magic; + __le32 length; +#ifndef USE_LEGACY_DESC_HEAD + __le32 flags; +#endif + __le32 fs_count; + __le32 hs_count; + } __attribute__((packed)) header; struct { struct usb_interface_descriptor intf; struct usb_endpoint_descriptor_no_audio sink; @@ -114,7 +124,13 @@ static const struct { } __attribute__((packed)) fs_descs, hs_descs; } __attribute__((packed)) descriptors = { .header = { +#ifdef USE_LEGACY_DESC_HEAD .magic = cpu_to_le32(FUNCTIONFS_DESCRIPTORS_MAGIC), +#else + .magic = cpu_to_le32(FUNCTIONFS_DESCRIPTORS_MAGIC_V2), + .flags = cpu_to_le32(FUNCTIONFS_HAS_FS_DESC | + FUNCTIONFS_HAS_HS_DESC), +#endif .length = cpu_to_le32(sizeof descriptors), .fs_count = 3, .hs_count = 3, -- cgit v1.2.3 From 2da38e0c9465b89518b29328daeb7da0ca1690b7 Mon Sep 17 00:00:00 2001 From: Vinod Koul Date: Thu, 19 Jun 2014 14:41:00 +0530 Subject: ALSA: compress: fix the struct alignment to 4 bytes In 64bit systems the compiler can default align to 8bytes causing mis-match with 32bit usermode. Avoid this is future by ensuring all the structures shared with usermode are packed and aligned to 4 bytes irrespective of arch used [coding style fixes by tiwai] Signed-off-by: Vinod Koul Signed-off-by: Takashi Iwai --- include/uapi/sound/compress_offload.h | 14 +++++++------- include/uapi/sound/compress_params.h | 14 +++++++------- 2 files changed, 14 insertions(+), 14 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/sound/compress_offload.h b/include/uapi/sound/compress_offload.h index 21eed488783f..1964026b5e09 100644 --- a/include/uapi/sound/compress_offload.h +++ b/include/uapi/sound/compress_offload.h @@ -39,7 +39,7 @@ struct snd_compressed_buffer { __u32 fragment_size; __u32 fragments; -}; +} __attribute__((packed, aligned(4))); /** * struct snd_compr_params: compressed stream params @@ -51,7 +51,7 @@ struct snd_compr_params { struct snd_compressed_buffer buffer; struct snd_codec codec; __u8 no_wake_mode; -}; +} __attribute__((packed, aligned(4))); /** * struct snd_compr_tstamp: timestamp descriptor @@ -70,7 +70,7 @@ struct snd_compr_tstamp { __u32 pcm_frames; __u32 pcm_io_frames; __u32 sampling_rate; -}; +} __attribute__((packed, aligned(4))); /** * struct snd_compr_avail: avail descriptor @@ -80,7 +80,7 @@ struct snd_compr_tstamp { struct snd_compr_avail { __u64 avail; struct snd_compr_tstamp tstamp; -} __attribute__((packed)); +} __attribute__((packed, aligned(4))); enum snd_compr_direction { SND_COMPRESS_PLAYBACK = 0, @@ -107,7 +107,7 @@ struct snd_compr_caps { __u32 max_fragments; __u32 codecs[MAX_NUM_CODECS]; __u32 reserved[11]; -}; +} __attribute__((packed, aligned(4))); /** * struct snd_compr_codec_caps: query capability of codec @@ -119,7 +119,7 @@ struct snd_compr_codec_caps { __u32 codec; __u32 num_descriptors; struct snd_codec_desc descriptor[MAX_NUM_CODEC_DESCRIPTORS]; -}; +} __attribute__((packed, aligned(4))); /** * @SNDRV_COMPRESS_ENCODER_PADDING: no of samples appended by the encoder at the @@ -140,7 +140,7 @@ enum { struct snd_compr_metadata { __u32 key; __u32 value[8]; -}; +} __attribute__((packed, aligned(4))); /** * compress path ioctl definitions diff --git a/include/uapi/sound/compress_params.h b/include/uapi/sound/compress_params.h index 165e7059de75..d9bd9ca0d5b0 100644 --- a/include/uapi/sound/compress_params.h +++ b/include/uapi/sound/compress_params.h @@ -268,7 +268,7 @@ struct snd_enc_vorbis { __u32 max_bit_rate; __u32 min_bit_rate; __u32 downmix; -}; +} __attribute__((packed, aligned(4))); /** @@ -284,7 +284,7 @@ struct snd_enc_real { __u32 quant_bits; __u32 start_region; __u32 num_regions; -}; +} __attribute__((packed, aligned(4))); /** * struct snd_enc_flac @@ -308,12 +308,12 @@ struct snd_enc_real { struct snd_enc_flac { __u32 num; __u32 gain; -}; +} __attribute__((packed, aligned(4))); struct snd_enc_generic { __u32 bw; /* encoder bandwidth */ __s32 reserved[15]; -}; +} __attribute__((packed, aligned(4))); union snd_codec_options { struct snd_enc_wma wma; @@ -321,7 +321,7 @@ union snd_codec_options { struct snd_enc_real real; struct snd_enc_flac flac; struct snd_enc_generic generic; -}; +} __attribute__((packed, aligned(4))); /** struct snd_codec_desc - description of codec capabilities * @max_ch: Maximum number of audio channels @@ -358,7 +358,7 @@ struct snd_codec_desc { __u32 formats; __u32 min_buffer; __u32 reserved[15]; -}; +} __attribute__((packed, aligned(4))); /** struct snd_codec * @id: Identifies the supported audio encoder/decoder. @@ -399,6 +399,6 @@ struct snd_codec { __u32 align; union snd_codec_options options; __u32 reserved[3]; -}; +} __attribute__((packed, aligned(4))); #endif -- cgit v1.2.3 From 9ad7860450ea65c6bbcbd52a9a25b54b07e35941 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Fri, 27 Jun 2014 10:41:00 -0500 Subject: Revert "tools: ffs-test: convert to new descriptor format fixing compilation error" This reverts commit f2af74123f8c5a735248547f4286a3adc28633c1. There is a better fix for this build error coming in a following patch. Signed-of-by: Felipe Balbi --- include/uapi/linux/usb/functionfs.h | 2 +- tools/usb/Makefile | 6 +----- tools/usb/ffs-test.c | 20 ++------------------ 3 files changed, 4 insertions(+), 24 deletions(-) (limited to 'include/uapi') diff --git a/include/uapi/linux/usb/functionfs.h b/include/uapi/linux/usb/functionfs.h index ecb3a31f7ca6..2a4b4a72a4f9 100644 --- a/include/uapi/linux/usb/functionfs.h +++ b/include/uapi/linux/usb/functionfs.h @@ -53,7 +53,7 @@ struct usb_endpoint_descriptor_no_audio { * structure. Any flags that are not recognised cause the whole block to be * rejected with -ENOSYS. * - * Legacy descriptors format (deprecated as of 3.14): + * Legacy descriptors format: * * | off | name | type | description | * |-----+-----------+--------------+--------------------------------------| diff --git a/tools/usb/Makefile b/tools/usb/Makefile index d576b3bac3cf..acf2165c04e6 100644 --- a/tools/usb/Makefile +++ b/tools/usb/Makefile @@ -6,11 +6,7 @@ WARNINGS = -Wall -Wextra CFLAGS = $(WARNINGS) -g -I../include LDFLAGS = $(PTHREAD_LIBS) -all: testusb ffs-test ffs-test-legacy - -ffs-test-legacy: ffs-test.c - $(CC) $(CFLAGS) -o $@ $^ $(LDFLAGS) -DUSE_LEGACY_DESC_HEAD - +all: testusb ffs-test %: %.c $(CC) $(CFLAGS) -o $@ $^ $(LDFLAGS) diff --git a/tools/usb/ffs-test.c b/tools/usb/ffs-test.c index 74b353d9eb50..fe1e66b6ef40 100644 --- a/tools/usb/ffs-test.c +++ b/tools/usb/ffs-test.c @@ -1,5 +1,5 @@ /* - * ffs-test.c -- user mode filesystem api for usb composite function + * ffs-test.c.c -- user mode filesystem api for usb composite function * * Copyright (C) 2010 Samsung Electronics * Author: Michal Nazarewicz @@ -21,8 +21,6 @@ /* $(CROSS_COMPILE)cc -Wall -Wextra -g -o ffs-test ffs-test.c -lpthread */ -/* Uncomment to make the tool use legacy FFS descriptor headers. */ -/* #define USE_LEGACY_DESC_HEAD */ #define _BSD_SOURCE /* for endian.h */ @@ -108,15 +106,7 @@ static void _msg(unsigned level, const char *fmt, ...) /******************** Descriptors and Strings *******************************/ static const struct { - struct { - __le32 magic; - __le32 length; -#ifndef USE_LEGACY_DESC_HEAD - __le32 flags; -#endif - __le32 fs_count; - __le32 hs_count; - } __attribute__((packed)) header; + struct usb_functionfs_descs_head header; struct { struct usb_interface_descriptor intf; struct usb_endpoint_descriptor_no_audio sink; @@ -124,13 +114,7 @@ static const struct { } __attribute__((packed)) fs_descs, hs_descs; } __attribute__((packed)) descriptors = { .header = { -#ifdef USE_LEGACY_DESC_HEAD .magic = cpu_to_le32(FUNCTIONFS_DESCRIPTORS_MAGIC), -#else - .magic = cpu_to_le32(FUNCTIONFS_DESCRIPTORS_MAGIC_V2), - .flags = cpu_to_le32(FUNCTIONFS_HAS_FS_DESC | - FUNCTIONFS_HAS_HS_DESC), -#endif .length = cpu_to_le32(sizeof descriptors), .fs_count = 3, .hs_count = 3, -- cgit v1.2.3 From 09122141785348bf9539762a5f5dbbae3761c783 Mon Sep 17 00:00:00 2001 From: Michal Nazarewicz Date: Fri, 13 Jun 2014 15:38:04 +0200 Subject: usb: gadget: f_fs: resurect usb_functionfs_descs_head structure MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Even though usb_functionfs_descs_head structure is now deprecated, it has been used by some user space tools. Its removel in commit [ac8dde1: “Add flags to descriptors block”] was an oversight leading to build breakage for such tools. Bring it back so that old user space tools can still be build without problems on newer kernel versions. Cc: # 3.14 Reported-by: Lad, Prabhakar Reported-by: Krzysztof Opasiak Signed-off-by: Michal Nazarewicz Signed-off-by: Felipe Balbi --- include/uapi/linux/usb/functionfs.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/uapi') diff --git a/include/uapi/linux/usb/functionfs.h b/include/uapi/linux/usb/functionfs.h index 2a4b4a72a4f9..24b68c59dcf8 100644 --- a/include/uapi/linux/usb/functionfs.h +++ b/include/uapi/linux/usb/functionfs.h @@ -33,6 +33,13 @@ struct usb_endpoint_descriptor_no_audio { __u8 bInterval; } __attribute__((packed)); +/* Legacy format, deprecated as of 3.14. */ +struct usb_functionfs_descs_head { + __le32 magic; + __le32 length; + __le32 fs_count; + __le32 hs_count; +} __attribute__((packed, deprecated)); /* * Descriptors format: -- cgit v1.2.3 From b2373f255cacdc1ea4da25e75a5a78949ffd9d66 Mon Sep 17 00:00:00 2001 From: Anand Jain Date: Tue, 3 Jun 2014 11:36:03 +0800 Subject: btrfs: create sprout should rename fsid on the sysfs as well Creating sprout will change the fsid of the mounted root. do the same on the sysfs as well. reproducer: mount /dev/sdb /btrfs (seed disk) btrfs dev add /dev/sdc /btrfs mount -o rw,remount /btrfs btrfs dev del /dev/sdb /btrfs mount /dev/sdb /btrfs Error: kobject_add_internal failed for fe350492-dc28-4051-a601-e017b17e6145 with -EEXIST, don't try to register things with the same name in the same directory. Signed-off-by: Anand Jain Reviewed-by: David Sterba Signed-off-by: Chris Mason --- fs/btrfs/volumes.c | 9 +++++++++ include/uapi/linux/btrfs.h | 1 + 2 files changed, 10 insertions(+) (limited to 'include/uapi') diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 19b67e969b52..6ca0d9cc46f9 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -2154,6 +2154,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path) mutex_unlock(&root->fs_info->fs_devices->device_list_mutex); if (seeding_dev) { + char fsid_buf[BTRFS_UUID_UNPARSED_SIZE]; ret = init_first_rw_device(trans, root, device); if (ret) { btrfs_abort_transaction(trans, root, ret); @@ -2164,6 +2165,14 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path) btrfs_abort_transaction(trans, root, ret); goto error_trans; } + + /* Sprouting would change fsid of the mounted root, + * so rename the fsid on the sysfs + */ + snprintf(fsid_buf, BTRFS_UUID_UNPARSED_SIZE, "%pU", + root->fs_info->fsid); + if (kobject_rename(&root->fs_info->super_kobj, fsid_buf)) + goto error_trans; } else { ret = btrfs_add_device(trans, root, device); if (ret) { diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index 6f9c38ce45c7..2f47824e7a36 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -38,6 +38,7 @@ struct btrfs_ioctl_vol_args { #define BTRFS_SUBVOL_QGROUP_INHERIT (1ULL << 2) #define BTRFS_FSID_SIZE 16 #define BTRFS_UUID_SIZE 16 +#define BTRFS_UUID_UNPARSED_SIZE 37 #define BTRFS_QGROUP_INHERIT_SET_LIMITS (1ULL << 0) -- cgit v1.2.3