linux/net/ipv4/ip_input.c, branch v3.12.10

net: Fix memory leak if TPROXY used with TCP early demux

2014-02-06T19:22:21Z

[ Upstream commit a452ce345d63ddf92cd101e4196569f8718ad319 ] I see a memory leak when using a transparent HTTP proxy using TPROXY together with TCP early demux and Kernel v3.8.13.15 (Ubuntu stable): unreferenced object 0xffff88008cba4a40 (size 1696): comm "softirq", pid 0, jiffies 4294944115 (age 8907.520s) hex dump (first 32 bytes): 0a e0 20 6a 40 04 1b 37 92 be 32 e2 e8 b4 00 00 .. j@..7..2..... 02 00 07 01 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmem_cache_alloc+0xad/0xb9 [] sk_prot_alloc+0x29/0xc5 [] sk_clone_lock+0x14/0x283 [] inet_csk_clone_lock+0xf/0x7b [] netlink_broadcast+0x14/0x16 [] tcp_create_openreq_child+0x1b/0x4c3 [] tcp_v4_syn_recv_sock+0x38/0x25d [] tcp_check_req+0x25c/0x3d0 [] tcp_v4_do_rcv+0x287/0x40e [] ip_route_input_noref+0x843/0xa55 [] tcp_v4_rcv+0x4c9/0x725 [] ip_local_deliver_finish+0xe9/0x154 [] __netif_receive_skb+0x4b2/0x514 [] process_backlog+0xee/0x1c5 [] net_rx_action+0xa7/0x200 [] add_interrupt_randomness+0x39/0x157 But there are many more, resulting in the machine going OOM after some days. From looking at the TPROXY code, and with help from Florian, I see that the memory leak is introduced in tcp_v4_early_demux(): void tcp_v4_early_demux(struct sk_buff *skb) { /* ... */ iph = ip_hdr(skb); th = tcp_hdr(skb); if (th->doff < sizeof(struct tcphdr) / 4) return; sk = __inet_lookup_established(dev_net(skb->dev), &tcp_hashinfo, iph->saddr, th->source, iph->daddr, ntohs(th->dest), skb->skb_iif); if (sk) { skb->sk = sk; where the socket is assigned unconditionally to skb->sk, also bumping the refcnt on it. This is problematic, because in our case the skb has already a socket assigned in the TPROXY target. This then results in the leak I see. The very same issue seems to be with IPv6, but haven't tested. Reviewed-by: Florian Westphal Signed-off-by: Holger Eitzenberger Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

net: add SNMP counters tracking incoming ECN bits

2013-08-09T05:24:59Z

With GRO/LRO processing, there is a problem because Ip[6]InReceives SNMP counters do not count the number of frames, but number of aggregated segments. Its probably too late to change this now. This patch adds four new counters, tracking number of frames, regardless of LRO/GRO, and on a per ECN status basis, for IPv4 and IPv6. Ip[6]NoECTPkts : Number of packets received with NOECT Ip[6]ECT1Pkts : Number of packets received with ECT(1) Ip[6]ECT0Pkts : Number of packets received with ECT(0) Ip[6]CEPkts : Number of packets received with Congestion Experienced lph37:~# nstat | egrep "Pkts|InReceive" IpInReceives 1634137 0.0 Ip6InReceives 3714107 0.0 Ip6InNoECTPkts 19205 0.0 Ip6InECT0Pkts 52651828 0.0 IpExtInNoECTPkts 33630 0.0 IpExtInECT0Pkts 15581379 0.0 IpExtInCEPkts 6 0.0 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

ipv4: set transport header earlier

2013-07-16T19:59:28Z

commit 45f00f99d6e ("ipv4: tcp: clean up tcp_v4_early_demux()") added a performance regression for non GRO traffic, basically disabling IP early demux. IPv6 stack resets transport header in ip6_rcv() before calling IP early demux in ip6_rcv_finish(), while IPv4 does this only in ip_local_deliver_finish(), _after_ IP early demux. GRO traffic happened to enable IP early demux because transport header is also set in inet_gro_receive() Instead of reverting the faulty commit, we can make IPv4/IPv6 behave the same : transport_header should be set in ip_rcv() instead of ip_local_deliver_finish() ip_local_deliver_finish() can also use skb_network_header_len() which is faster than ip_hdrlen() Signed-off-by: Eric Dumazet Cc: Neal Cardwell Cc: Tom Herbert Signed-off-by: David S. Miller

net: Add MIB counters for checksum errors

2013-04-29T19:14:03Z

Add MIB counters for checksum errors in IP layer, and TCP/UDP/ICMP layers, to help diagnose problems. $ nstat -a | grep Csum IcmpInCsumErrors 72 0.0 TcpInCsumErrors 382 0.0 UdpInCsumErrors 463221 0.0 Icmp6InCsumErrors 75 0.0 Udp6InCsumErrors 173442 0.0 IpExtInCsumErrors 10884 0.0 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

ipv[4|6]: correct dropwatch false positive in local_deliver_finish

2013-03-01T20:56:29Z

I had a report recently of a user trying to use dropwatch to localise some frame loss, and they were getting false positives. Turned out they were using a user space SCTP stack that used raw sockets to grab frames. When we don't have a registered protocol for a given packet, we record it as a drop, even if a raw socket receieves the frame. We should only record the drop in the event a raw socket doesnt exist to receive the frames Tested by the reported successfully Signed-off-by: Neil Horman Reported-by: William Reich Tested-by: William Reich CC: "David S. Miller" CC: William Reich CC: eric.dumazet@gmail.com Acked-by: Eric Dumazet Signed-off-by: David S. Miller

ipv4: Disallow non-namespace aware protocols to register.

2013-02-05T19:42:23Z

All in-tree ipv4 protocol implementations are now namespace aware. Therefore all the run-time checks are superfluous. Reject registry of any non-namespace aware ipv4 protocol. Eventually we'll remove prot->netns_ok and this registry time check as well. Signed-off-by: David S. Miller

net: TCP early demux cleanup

2012-07-30T21:53:21Z

early_demux() handlers should be called in RCU context, and as we use skb_dst_set_noref(skb, dst), caller must not exit from RCU context before dst use (skb_dst(skb)) or release (skb_drop(dst)) Therefore, rcu_read_lock()/rcu_read_unlock() pairs around ->early_demux() are confusing and not needed : Protocol handlers are already in an RCU read lock section. (__netif_receive_skb() does the rcu_read_lock() ) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

ipv6: Early TCP socket demux

2012-07-26T22:50:39Z

This is the IPv6 missing bits for infrastructure added in commit 41063e9dd1195 (ipv4: Early TCP socket demux.) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

ipv4: Fix input route performance regression.

2012-07-26T22:50:39Z

With the routing cache removal we lost the "noref" code paths on input, and this can kill some routing workloads. Reinstate the noref path when we hit a cached route in the FIB nexthops. With help from Eric Dumazet. Reported-by: Alexander Duyck Signed-off-by: David S. Miller Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

tcp: early_demux fixes

2012-07-24T20:54:15Z

1) Remove a non needed pskb_may_pull() in tcp_v4_early_demux() and fix a potential bug if skb->head was reallocated (iph & th pointers were not reloaded) TCP stack will pull/check headers anyway. 2) must reload iph in ip_rcv_finish() after early_demux() call since skb->head might have changed. 3) skb->dev->ifindex can be now replaced by skb->skb_iif Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller