linux/net, branch v3.12.14

SUNRPC: Ensure that gss_auth isn't freed before its upcall messages

2014-03-05T16:13:45Z

commit 9eb2ddb48ce3a7bd745c14a933112994647fa3cd upstream. Fix a race in which the RPC client is shutting down while the gss daemon is processing a downcall. If the RPC client manages to shut down before the gss daemon is done, then the struct gss_auth used in gss_release_msg() may have already been freed. Link: http://lkml.kernel.org/r/1392494917.71728.YahooMailNeo@web140002.mail.bf1.yahoo.com Reported-by: John Reported-by: Borislav Petkov Signed-off-by: Trond Myklebust Signed-off-by: Jiri Slaby

SUNRPC: Fix races in xs_nospace()

2014-03-05T16:13:44Z

commit 06ea0bfe6e6043cb56a78935a19f6f8ebc636226 upstream. When a send failure occurs due to the socket being out of buffer space, we call xs_nospace() in order to have the RPC task wait until the socket has drained enough to make it worth while trying again. The current patch fixes a race in which the socket is drained before we get round to setting up the machinery in xs_nospace(), and which is reported to cause hangs. Link: http://lkml.kernel.org/r/20140210170315.33dfc621@notabene.brown Fixes: a9a6b52ee1ba (SUNRPC: Don't start the retransmission timer...) Reported-by: Neil Brown Signed-off-by: Trond Myklebust Signed-off-by: Jiri Slaby

net: use __GFP_NORETRY for high order allocations

2014-02-26T09:22:53Z

[ Upstream commit ed98df3361f059db42786c830ea96e2d18b8d4db ] sock_alloc_send_pskb() & sk_page_frag_refill() have a loop trying high order allocations to prepare skb with low number of fragments as this increases performance. Problem is that under memory pressure/fragmentation, this can trigger OOM while the intent was only to try the high order allocations, then fallback to order-0 allocations. We had various reports from unexpected regressions. According to David, setting __GFP_NORETRY should be fine, as the asynchronous compaction is still enabled, and this will prevent OOM from kicking as in : CFSClientEventm invoked oom-killer: gfp_mask=0x42d0, order=3, oom_adj=0, oom_score_adj=0, oom_score_badness=2 (enabled),memcg_scoring=disabled CFSClientEventm Call Trace: [] dump_header+0xe1/0x23e [] oom_kill_process+0x6a/0x323 [] out_of_memory+0x4b3/0x50d [] __alloc_pages_may_oom+0xa2/0xc7 [] __alloc_pages_nodemask+0x1002/0x17f0 [] alloc_pages_current+0x103/0x2b0 [] sk_page_frag_refill+0x8f/0x160 [] tcp_sendmsg+0x560/0xee0 [] inet_sendmsg+0x67/0x100 [] __sock_sendmsg_nosec+0x6c/0x90 [] sock_sendmsg+0xc5/0xf0 [] __sys_sendmsg+0x136/0x430 [] sys_sendmsg+0x88/0x110 [] system_call_fastpath+0x16/0x1b Out of Memory: Kill process 2856 (bash) score 9999 or sacrifice child Signed-off-by: Eric Dumazet Acked-by: David Rientjes Acked-by: "Eric W. Biederman" Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

net: ip, ipv6: handle gso skbs in forwarding path

2014-02-26T09:22:53Z

commit fe6cc55f3a9a053482a76f5a6b2257cee51b4663 upstream. Marcelo Ricardo Leitner reported problems when the forwarding link path has a lower mtu than the incoming one if the inbound interface supports GRO. Given: Host R1 R2 Host sends tcp stream which is routed via R1 and R2. R1 performs GRO. In this case, the kernel will fail to send ICMP fragmentation needed messages (or pkt too big for ipv6), as GSO packets currently bypass dstmtu checks in forward path. Instead, Linux tries to send out packets exceeding the mtu. When locking route MTU on Host (i.e., no ipv4 DF bit set), R1 does not fragment the packets when forwarding, and again tries to send out packets exceeding R1-R2 link mtu. This alters the forwarding dstmtu checks to take the individual gso segment lengths into account. For ipv6, we send out pkt too big error for gso if the individual segments are too big. For ipv4, we either send icmp fragmentation needed, or, if the DF bit is not set, perform software segmentation and let the output path create fragments when the packet is leaving the machine. It is not 100% correct as the error message will contain the headers of the GRO skb instead of the original/segmented one, but it seems to work fine in my (limited) tests. Eric Dumazet suggested to simply shrink mss via ->gso_size to avoid sofware segmentation. However it turns out that skb_segment() assumes skb nr_frags is related to mss size so we would BUG there. I don't want to mess with it considering Herbert and Eric disagree on what the correct behavior should be. Hannes Frederic Sowa notes that when we would shrink gso_size skb_segment would then also need to deal with the case where SKB_MAX_FRAGS would be exceeded. This uses sofware segmentation in the forward path when we hit ipv4 non-DF packets and the outgoing link mtu is too small. Its not perfect, but given the lack of bug reports wrt. GRO fwd being broken this is a rare case anyway. Also its not like this could not be improved later once the dust settles. Acked-by: Herbert Xu Reported-by: Marcelo Ricardo Leitner Signed-off-by: Florian Westphal Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

net: core: introduce netif_skb_dev_features

2014-02-26T09:22:53Z

commit d206940319c41df4299db75ed56142177bb2e5f6 upstream. Will be used by upcoming ipv4 forward path change that needs to determine feature mask using skb->dst->dev instead of skb->dev. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

net: add and use skb_gso_transport_seglen()

2014-02-26T09:22:53Z

commit de960aa9ab4decc3304959f69533eef64d05d8e8 upstream. This moves part of Eric Dumazets skb_gso_seglen helper from tbf sched to skbuff core so it may be reused by upcoming ip forwarding path patch. Signed-off-by: Florian Westphal Acked-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

net: sctp: fix sctp_connectx abi for ia32 emulation/compat mode

2014-02-26T09:22:53Z

[ Upstream commit ffd5939381c609056b33b7585fb05a77b4c695f3 ] SCTP's sctp_connectx() abi breaks for 64bit kernels compiled with 32bit emulation (e.g. ia32 emulation or x86_x32). Due to internal usage of 'struct sctp_getaddrs_old' which includes a struct sockaddr pointer, sizeof(param) check will always fail in kernel as the structure in 64bit kernel space is 4bytes larger than for user binaries compiled in 32bit mode. Thus, applications making use of sctp_connectx() won't be able to run under such circumstances. Introduce a compat interface in the kernel to deal with such situations by using a 'struct compat_sctp_getaddrs_old' structure where user data is copied into it, and then sucessively transformed into a 'struct sctp_getaddrs_old' structure with the help of compat_ptr(). That fixes sctp_connectx() abi without any changes needed in user space, and lets the SCTP test suite pass when compiled in 32bit and run on 64bit kernels. Fixes: f9c67811ebc0 ("sctp: Fix regression introduced by new sctp_connectx api") Signed-off-by: Daniel Borkmann Acked-by: Neil Horman Acked-by: Vlad Yasevich Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

ipv4: fix counter in_slow_tot

2014-02-26T09:22:53Z

[ Upstream commit a6254864c08109c66a194612585afc0439005286 ] since commit 89aef8921bf("ipv4: Delete routing cache."), the counter in_slow_tot can't work correctly. The counter in_slow_tot increase by one when fib_lookup() return successfully in ip_route_input_slow(), but actually the dst struct maybe not be created and cached, so we can increase in_slow_tot after the dst struct is created. Signed-off-by: Duan Jiong Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

tcp: tsq: fix nonagle handling

2014-02-26T09:22:51Z

[ Upstream commit bf06200e732de613a1277984bf34d1a21c2de03d ] Commit 46d3ceabd8d9 ("tcp: TCP Small Queues") introduced a possible regression for applications using TCP_NODELAY. If TCP session is throttled because of tsq, we should consult tp->nonagle when TX completion is done and allow us to send additional segment, especially if this segment is not a full MSS. Otherwise this segment is sent after an RTO. [edumazet] : Cooked the changelog, added another fix about testing sk_wmem_alloc twice because TX completion can happen right before setting TSQ_THROTTLED bit. This problem is particularly visible with recent auto corking, but might also be triggered with low tcp_limit_output_bytes values or NIC drivers delaying TX completion by hundred of usec, and very low rtt. Thomas Glanzmann for example reported an iscsi regression, caused by tcp auto corking making this bug quite visible. Fixes: 46d3ceabd8d9 ("tcp: TCP Small Queues") Signed-off-by: John Ogness Signed-off-by: Eric Dumazet Reported-by: Thomas Glanzmann Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby

netpoll: fix netconsole IPv6 setup

2014-02-26T09:22:51Z

[ Upstream commit 00fe11b3c67dc670fe6391d22f1fe64e7c99a8ec ] Currently, to make netconsole start over IPv6, the source address needs to be specified. Without a source address, netpoll_parse_options assumes we're setting up over IPv4 and the destination IPv6 address is rejected. Check if the IP version has been forced by a source address before checking for a version mismatch when parsing the destination address. Signed-off-by: Sabrina Dubroca Acked-by: Cong Wang Signed-off-by: David S. Miller Signed-off-by: Jiri Slaby