user/sven/linux.git/net/core, branch v3.10.44

net-gro: reset skb->truesize in napi_reuse_skb()

2014-05-31T04:52:16Z

[ Upstream commit e33d0ba8047b049c9262fdb1fcafb93cb52ceceb ] Recycling skb always had been very tough... This time it appears GRO layer can accumulate skb->truesize adjustments made by drivers when they attach a fragment to skb. skb_gro_receive() can only subtract from skb->truesize the used part of a fragment. I spotted this problem seeing TcpExtPruneCalled and TcpExtTCPRcvCollapsed that were unexpected with a recent kernel, where TCP receive window should be sized properly to accept traffic coming from a driver not overshooting skb->truesize. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

rtnetlink: Only supply IFLA_VF_PORTS information when RTEXT_FILTER_VF is set

2014-05-31T04:52:15Z

[ Upstream commit c53864fd60227de025cb79e05493b13f69843971 ] Since 115c9b81928360d769a76c632bae62d15206a94a (rtnetlink: Fix problem with buffer allocation), RTM_NEWLINK messages only contain the IFLA_VFINFO_LIST attribute if they were solicited by a GETLINK message containing an IFLA_EXT_MASK attribute with the RTEXT_FILTER_VF flag. That was done because some user programs broke when they received more data than expected - because IFLA_VFINFO_LIST contains information for each VF it can become large if there are many VFs. However, the IFLA_VF_PORTS attribute, supplied for devices which implement ndo_get_vf_port (currently the 'enic' driver only), has the same problem. It supplies per-VF information and can therefore become large, but it is not currently conditional on the IFLA_EXT_MASK value. Worse, it interacts badly with the existing EXT_MASK handling. When IFLA_EXT_MASK is not supplied, the buffer for netlink replies is fixed at NLMSG_GOODSIZE. If the information for IFLA_VF_PORTS exceeds this, then rtnl_fill_ifinfo() returns -EMSGSIZE on the first message in a packet. netlink_dump() will misinterpret this as having finished the listing and omit data for this interface and all subsequent ones. That can cause getifaddrs(3) to enter an infinite loop. This patch addresses the problem by only supplying IFLA_VF_PORTS when IFLA_EXT_MASK is supplied with the RTEXT_FILTER_VF flag set. Signed-off-by: David Gibson Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

rtnetlink: Warn when interface's information won't fit in our packet

2014-05-31T04:52:15Z

[ Upstream commit 973462bbde79bb827824c73b59027a0aed5c9ca6 ] Without IFLA_EXT_MASK specified, the information reported for a single interface in response to RTM_GETLINK is expected to fit within a netlink packet of NLMSG_GOODSIZE. If it doesn't, however, things will go badly wrong, When listing all interfaces, netlink_dump() will incorrectly treat -EMSGSIZE on the first message in a packet as the end of the listing and omit information for that interface and all subsequent ones. This can cause getifaddrs(3) to enter an infinite loop. This patch won't fix the problem, but it will WARN_ON() making it easier to track down what's going wrong. Signed-off-by: David Gibson Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

net: Fix ns_capable check in sock_diag_put_filterinfo

2014-05-31T04:52:15Z

[ Upstream commit 78541c1dc60b65ecfce5a6a096fc260219d6784e ] The caller needs capabilities on the namespace being queried, not on their own namespace. This is a security bug, although it likely has only a minor impact. Cc: stable@vger.kernel.org Signed-off-by: Andy Lutomirski Acked-by: Nicolas Dichtel Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

vlan: Fix lockdep warning when vlan dev handle notification

2014-05-31T04:52:15Z

[ Upstream commit dc8eaaa006350d24030502a4521542e74b5cb39f ] When I open the LOCKDEP config and run these steps: modprobe 8021q vconfig add eth2 20 vconfig add eth2.20 30 ifconfig eth2 xx.xx.xx.xx then the Call Trace happened: [32524.386288] ============================================= [32524.386293] [ INFO: possible recursive locking detected ] [32524.386298] 3.14.0-rc2-0.7-default+ #35 Tainted: G O [32524.386302] --------------------------------------------- [32524.386306] ifconfig/3103 is trying to acquire lock: [32524.386310] (&vlan_netdev_addr_lock_key/1){+.....}, at: [] dev_mc_sync+0x64/0xb0 [32524.386326] [32524.386326] but task is already holding lock: [32524.386330] (&vlan_netdev_addr_lock_key/1){+.....}, at: [] dev_set_rx_mode+0x23/0x40 [32524.386341] [32524.386341] other info that might help us debug this: [32524.386345] Possible unsafe locking scenario: [32524.386345] [32524.386350] CPU0 [32524.386352] ---- [32524.386354] lock(&vlan_netdev_addr_lock_key/1); [32524.386359] lock(&vlan_netdev_addr_lock_key/1); [32524.386364] [32524.386364] *** DEADLOCK *** [32524.386364] [32524.386368] May be due to missing lock nesting notation [32524.386368] [32524.386373] 2 locks held by ifconfig/3103: [32524.386376] #0: (rtnl_mutex){+.+.+.}, at: [] rtnl_lock+0x12/0x20 [32524.386387] #1: (&vlan_netdev_addr_lock_key/1){+.....}, at: [] dev_set_rx_mode+0x23/0x40 [32524.386398] [32524.386398] stack backtrace: [32524.386403] CPU: 1 PID: 3103 Comm: ifconfig Tainted: G O 3.14.0-rc2-0.7-default+ #35 [32524.386409] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007 [32524.386414] ffffffff81ffae40 ffff8800d9625ae8 ffffffff814f68a2 ffff8800d9625bc8 [32524.386421] ffffffff810a35fb ffff8800d8a8d9d0 00000000d9625b28 ffff8800d8a8e5d0 [32524.386428] 000003cc00000000 0000000000000002 ffff8800d8a8e5f8 0000000000000000 [32524.386435] Call Trace: [32524.386441] [] dump_stack+0x6a/0x78 [32524.386448] [] __lock_acquire+0x7ab/0x1940 [32524.386454] [] ? __lock_acquire+0x3ea/0x1940 [32524.386459] [] lock_acquire+0xe4/0x110 [32524.386464] [] ? dev_mc_sync+0x64/0xb0 [32524.386471] [] _raw_spin_lock_nested+0x2a/0x40 [32524.386476] [] ? dev_mc_sync+0x64/0xb0 [32524.386481] [] dev_mc_sync+0x64/0xb0 [32524.386489] [] vlan_dev_set_rx_mode+0x2b/0x50 [8021q] [32524.386495] [] __dev_set_rx_mode+0x5f/0xb0 [32524.386500] [] dev_set_rx_mode+0x2b/0x40 [32524.386506] [] __dev_open+0xef/0x150 [32524.386511] [] __dev_change_flags+0xa7/0x190 [32524.386516] [] dev_change_flags+0x32/0x80 [32524.386524] [] devinet_ioctl+0x7d6/0x830 [32524.386532] [] ? dev_ioctl+0x34b/0x660 [32524.386540] [] inet_ioctl+0x80/0xa0 [32524.386550] [] sock_do_ioctl+0x2d/0x60 [32524.386558] [] sock_ioctl+0x82/0x2a0 [32524.386568] [] do_vfs_ioctl+0x93/0x590 [32524.386578] [] ? rcu_read_lock_held+0x45/0x50 [32524.386586] [] ? __fget_light+0x105/0x110 [32524.386594] [] SyS_ioctl+0x91/0xb0 [32524.386604] [] system_call_fastpath+0x16/0x1b ======================================================================== The reason is that all of the addr_lock_key for vlan dev have the same class, so if we change the status for vlan dev, the vlan dev and its real dev will hold the same class of addr_lock_key together, so the warning happened. we should distinguish the lock depth for vlan dev and its real dev. v1->v2: Convert the vlan_netdev_addr_lock_key to an array of eight elements, which could support to add 8 vlan id on a same vlan dev, I think it is enough for current scene, because a netdev's name is limited to IFNAMSIZ which could not hold 8 vlan id, and the vlan dev would not meet the same class key with its real dev. The new function vlan_dev_get_lockdep_subkey() will return the subkey and make the vlan dev could get a suitable class key. v2->v3: According David's suggestion, I use the subclass to distinguish the lock key for vlan dev and its real dev, but it make no sense, because the difference for subclass in the lock_class_key doesn't mean that the difference class for lock_key, so I use lock_depth to distinguish the different depth for every vlan dev, the same depth of the vlan dev could have the same lock_class_key, I import the MAX_LOCK_DEPTH from the include/linux/sched.h, I think it is enough here, the lockdep should never exceed that value. v3->v4: Add a huge array of locking keys will waste static kernel memory and is not a appropriate method, we could use _nested() variants to fix the problem, calculate the depth for every vlan dev, and use the depth as the subclass for addr_lock_key. Signed-off-by: Ding Tianhong Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

filter: prevent nla extensions to peek beyond the end of the message

2014-05-31T04:52:15Z

[ Upstream commit 05ab8f2647e4221cbdb3856dd7d32bd5407316b3 ] The BPF_S_ANC_NLATTR and BPF_S_ANC_NLATTR_NEST extensions fail to check for a minimal message length before testing the supplied offset to be within the bounds of the message. This allows the subtraction of the nla header to underflow and therefore -- as the data type is unsigned -- allowing far to big offset and length values for the search of the netlink attribute. The remainder calculation for the BPF_S_ANC_NLATTR_NEST extension is also wrong. It has the minuend and subtrahend mixed up, therefore calculates a huge length value, allowing to overrun the end of the message while looking for the netlink attribute. The following three BPF snippets will trigger the bugs when attached to a UNIX datagram socket and parsing a message with length 1, 2 or 3. ,-[ PoC for missing size check in BPF_S_ANC_NLATTR ]-- | ld #0x87654321 | ldx #42 | ld #nla | ret a `--- ,-[ PoC for the same bug in BPF_S_ANC_NLATTR_NEST ]-- | ld #0x87654321 | ldx #42 | ld #nlan | ret a `--- ,-[ PoC for wrong remainder calculation in BPF_S_ANC_NLATTR_NEST ]-- | ; (needs a fake netlink header at offset 0) | ld #0 | ldx #42 | ld #nlan | ret a `--- Fix the first issue by ensuring the message length fulfills the minimal size constrains of a nla header. Fix the second bug by getting the math for the remainder calculation right. Fixes: 4738c1db15 ("[SKFILTER]: Add SKF_ADF_NLATTR instruction") Fixes: d214c7537b ("filter: add SKF_AD_NLATTR_NEST to look for nested..") Cc: Patrick McHardy Cc: Pablo Neira Ayuso Signed-off-by: Mathias Krause Acked-by: Daniel Borkmann Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

net: core: don't account for udp header size when computing seglen

2014-05-31T04:52:14Z

[ Upstream commit 6d39d589bb76ee8a1c6cde6822006ae0053decff ] In case of tcp, gso_size contains the tcpmss. For UFO (udp fragmentation offloading) skbs, gso_size is the fragment payload size, i.e. we must not account for udp header size. Otherwise, when using virtio drivers, a to-be-forwarded UFO GSO packet will be needlessly fragmented in the forward path, because we think its individual segments are too large for the outgoing link. Fixes: fe6cc55f3a9a053 ("net: ip, ipv6: handle gso skbs in forwarding path") Cc: Eric Dumazet Reported-by: Tobias Brunner Signed-off-by: Florian Westphal Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

netpoll: fix the skb check in pkt_is_ns

2014-04-14T13:42:17Z

[ Not applicable upstream commit, the code here has been removed upstream. ] Neighbor Solicitation is ipv6 protocol, so we should check skb->protocol with ETH_P_IPV6 Signed-off-by: Li RongQing Cc: WANG Cong Signed-off-by: Greg Kroah-Hartman

rtnetlink: fix fdb notification flags

2014-04-14T13:42:17Z

[ Upstream commit 1c104a6bebf3c16b6248408b84f91d09ac8a26b6 ] Commit 3ff661c38c84 ("net: rtnetlink notify events for FDB NTF_SELF adds and deletes") reuses the function nlmsg_populate_fdb_fill() to notify fdb events. But this function was used only for dump and thus was always setting the flag NLM_F_MULTI, which is wrong in case of a single notification. Libraries like libnl will wait forever for NLMSG_DONE. CC: Thomas Graf Signed-off-by: Nicolas Dichtel Acked-by: Thomas Graf Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

tcp: tcp_release_cb() should release socket ownership

2014-04-14T13:42:16Z

[ Upstream commit c3f9b01849ef3bc69024990092b9f42e20df7797 ] Lars Persson reported following deadlock : -000 |M:0x0:0x802B6AF8(asm) <-- arch_spin_lock -001 |tcp_v4_rcv(skb = 0x8BD527A0) <-- sk = 0x8BE6B2A0 -002 |ip_local_deliver_finish(skb = 0x8BD527A0) -003 |__netif_receive_skb_core(skb = 0x8BD527A0, ?) -004 |netif_receive_skb(skb = 0x8BD527A0) -005 |elk_poll(napi = 0x8C770500, budget = 64) -006 |net_rx_action(?) -007 |__do_softirq() -008 |do_softirq() -009 |local_bh_enable() -010 |tcp_rcv_established(sk = 0x8BE6B2A0, skb = 0x87D3A9E0, th = 0x814EBE14, ?) -011 |tcp_v4_do_rcv(sk = 0x8BE6B2A0, skb = 0x87D3A9E0) -012 |tcp_delack_timer_handler(sk = 0x8BE6B2A0) -013 |tcp_release_cb(sk = 0x8BE6B2A0) -014 |release_sock(sk = 0x8BE6B2A0) -015 |tcp_sendmsg(?, sk = 0x8BE6B2A0, ?, ?) -016 |sock_sendmsg(sock = 0x8518C4C0, msg = 0x87D8DAA8, size = 4096) -017 |kernel_sendmsg(?, ?, ?, ?, size = 4096) -018 |smb_send_kvec() -019 |smb_send_rqst(server = 0x87C4D400, rqst = 0x87D8DBA0) -020 |cifs_call_async() -021 |cifs_async_writev(wdata = 0x87FD6580) -022 |cifs_writepages(mapping = 0x852096E4, wbc = 0x87D8DC88) -023 |__writeback_single_inode(inode = 0x852095D0, wbc = 0x87D8DC88) -024 |writeback_sb_inodes(sb = 0x87D6D800, wb = 0x87E4A9C0, work = 0x87D8DD88) -025 |__writeback_inodes_wb(wb = 0x87E4A9C0, work = 0x87D8DD88) -026 |wb_writeback(wb = 0x87E4A9C0, work = 0x87D8DD88) -027 |wb_do_writeback(wb = 0x87E4A9C0, force_wait = 0) -028 |bdi_writeback_workfn(work = 0x87E4A9CC) -029 |process_one_work(worker = 0x8B045880, work = 0x87E4A9CC) -030 |worker_thread(__worker = 0x8B045880) -031 |kthread(_create = 0x87CADD90) -032 |ret_from_kernel_thread(asm) Bug occurs because __tcp_checksum_complete_user() enables BH, assuming it is running from softirq context. Lars trace involved a NIC without RX checksum support but other points are problematic as well, like the prequeue stuff. Problem is triggered by a timer, that found socket being owned by user. tcp_release_cb() should call tcp_write_timer_handler() or tcp_delack_timer_handler() in the appropriate context : BH disabled and socket lock held, but 'owned' field cleared, as if they were running from timer handlers. Fixes: 6f458dfb4092 ("tcp: improve latencies of timer triggered events") Reported-by: Lars Persson Tested-by: Lars Persson Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman