linux/include/net/ipv6.h, branch v3.17

inet: frag: don't account number of fragment queues

2014-07-28T05:34:36Z

The 'nqueues' counter is protected by the lru list lock, once thats removed this needs to be converted to atomic counter. Given this isn't used for anything except for reporting it to userspace via /proc, just remove it. We still report the memory currently used by fragment reassembly queues. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller

inet: frag: constify match, hashfn and constructor arguments

2014-07-28T05:34:35Z

Signed-off-by: Florian Westphal Signed-off-by: David S. Miller

net: clean up some sparse endianness warnings in ipv6.h

2014-07-17T06:31:38Z

sparse is throwing warnings when building sunrpc modules due to some endianness shenanigans in ipv6.h. Specifically: CHECK net/sunrpc/addr.c include/net/ipv6.h:573:17: warning: restricted __be64 degrades to integer include/net/ipv6.h:577:34: warning: restricted __be32 degrades to integer include/net/ipv6.h:573:17: warning: restricted __be64 degrades to integer include/net/ipv6.h:577:34: warning: restricted __be32 degrades to integer Sprinkle some endianness fixups to silence them. These should all get fixed up at compile time, so I don't think this will add any extra work to be done at runtime. Signed-off-by: Jeff Layton Signed-off-by: David S. Miller

net: provide stubs for ip6_set_txhash and ip6_make_flowlabel

2014-07-09T03:43:35Z

Commit cb1ce2ef387b ("ipv6: Implement automatic flow label generation on transmit") introduced ip6_make_flowlabel, while commit b73c3d0e4f0e ("net: Save TX flow hash in sock and set in skbuf on xmit") introduced ip6_set_txhash. ip6_set_tx_hash() uses sk_v6_daddr which references __sk_common.skc_v6_daddr from struct sock_common, which is gated with IS_ENABLED(CONFIG_IPV6). ip6_make_flowlabel() uses the ipv6 member from struct net which is also gated with IS_ENABLED(CONFIG_IPV6). When CONFIG_IPV6 is disabled, we will hit a build failure that looks like this when the compiler attempts inlining these functions: CC [M] drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.o In file included from include/net/inet_sock.h:27:0, from include/net/ip.h:30, from drivers/net/ethernet/broadcom/cnic.c:37: include/net/ipv6.h: In function 'ip6_set_txhash': include/net/sock.h:327:33: error: 'struct sock_common' has no member named 'skc_v6_daddr' #define sk_v6_daddr __sk_common.skc_v6_daddr ^ include/net/ipv6.h:696:49: note: in expansion of macro 'sk_v6_daddr' keys.dst = (__force __be32)ipv6_addr_hash(&sk->sk_v6_daddr); ^ In file included from include/net/inetpeer.h:15:0, from include/net/route.h:28, from include/net/ip.h:31, from drivers/net/ethernet/broadcom/cnic.c:37: include/net/ipv6.h: In function 'ip6_make_flowlabel': include/net/ipv6.h:706:37: error: 'struct net' has no member named 'ipv6' if (!flowlabel && (autolabel || net->ipv6.sysctl.auto_flowlabels)) { ^ Fixes: cb1ce2ef387b ("ipv6: Implement automatic flow label generation on transmit") Fixes: b73c3d0e4f0e ("net: Save TX flow hash in sock and set in skbuf on xmit") Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller

ipv6: Implement automatic flow label generation on transmit

2014-07-08T04:14:21Z

Automatically generate flow labels for IPv6 packets on transmit. The flow label is computed based on skb_get_hash. The flow label will only automatically be set when it is zero otherwise (i.e. flow label manager hasn't set one). This supports the transmit side functionality of RFC 6438. Added an IPv6 sysctl auto_flowlabels to enable/disable this behavior system wide, and added IPV6_AUTOFLOWLABEL socket option to enable this functionality per socket. By default, auto flowlabels are disabled to avoid possible conflicts with flow label manager, however if this feature proves useful we may want to enable it by default. It should also be noted that FreeBSD has already implemented automatic flow labels (including the sysctl and socket option). In FreeBSD, automatic flow labels default to enabled. Performance impact: Running super_netperf with 200 flows for TCP_RR and UDP_RR for IPv6. Note that in UDP case, __skb_get_hash will be called for every packet with explains slight regression. In the TCP case the hash is saved in the socket so there is no regression. Automatic flow labels disabled: TCP_RR: 86.53% CPU utilization 127/195/322 90/95/99% latencies 1.40498e+06 tps UDP_RR: 90.70% CPU utilization 118/168/243 90/95/99% latencies 1.50309e+06 tps Automatic flow labels enabled: TCP_RR: 85.90% CPU utilization 128/199/337 90/95/99% latencies 1.40051e+06 UDP_RR 92.61% CPU utilization 115/164/236 90/95/99% latencies 1.4687e+06 Signed-off-by: Tom Herbert Signed-off-by: David S. Miller

net: Save TX flow hash in sock and set in skbuf on xmit

2014-07-08T04:14:21Z

For a connected socket we can precompute the flow hash for setting in skb->hash on output. This is a performance advantage over calculating the skb->hash for every packet on the connection. The computation is done using the common hash algorithm to be consistent with computations done for packets of the connection in other states where thers is no socket (e.g. time-wait, syn-recv, syn-cookies). This patch adds sk_txhash to the sock structure. inet_set_txhash and ip6_set_txhash functions are added which are called from points in TCP and UDP where socket moves to established state. skb_set_hash_from_sk is a function which sets skb->hash from the sock txhash value. This is called in UDP and TCP transmit path when transmitting within the context of a socket. Tested: ran super_netperf with 200 TCP_RR streams over a vxlan interface (in this case skb_get_hash called on every TX packet to create a UDP source port). Before fix: 95.02% CPU utilization 154/256/505 90/95/99% latencies 1.13042e+06 tps Time in functions: 0.28% skb_flow_dissect 0.21% __skb_get_hash After fix: 94.95% CPU utilization 156/254/485 90/95/99% latencies 1.15447e+06 Neither __skb_get_hash nor skb_flow_dissect appear in perf Signed-off-by: Tom Herbert Signed-off-by: David S. Miller

inetpeer: get rid of ip_id_count

2014-06-02T18:00:41Z

Ideally, we would need to generate IP ID using a per destination IP generator. linux kernels used inet_peer cache for this purpose, but this had a huge cost on servers disabling MTU discovery. 1) each inet_peer struct consumes 192 bytes 2) inetpeer cache uses a binary tree of inet_peer structs, with a nominal size of ~66000 elements under load. 3) lookups in this tree are hitting a lot of cache lines, as tree depth is about 20. 4) If server deals with many tcp flows, we have a high probability of not finding the inet_peer, allocating a fresh one, inserting it in the tree with same initial ip_id_count, (cf secure_ip_id()) 5) We garbage collect inet_peer aggressively. IP ID generation do not have to be 'perfect' Goal is trying to avoid duplicates in a short period of time, so that reassembly units have a chance to complete reassembly of fragments belonging to one message before receiving other fragments with a recycled ID. We simply use an array of generators, and a Jenkin hash using the dst IP as a key. ipv6_select_ident() is put back into net/ipv6/ip6_output.c where it belongs (it is only used from this file) secure_ip_id() and secure_ipv6_id() no longer are needed. Rename ip_select_ident_more() to ip_select_ident_segs() to avoid unnecessary decrement/increment of the number of segments. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

net: add a sysctl to reflect the fwmark on replies

2014-05-13T22:35:08Z

Kernel-originated IP packets that have no user socket associated with them (e.g., ICMP errors and echo replies, TCP RSTs, etc.) are emitted with a mark of zero. Add a sysctl to make them have the same mark as the packet they are replying to. This allows an administrator that wishes to do so to use mark-based routing, firewalling, etc. for these replies by marking the original packets inbound. Tested using user-mode linux: - ICMP/ICMPv6 echo replies and errors. - TCP RST packets (IPv4 and IPv6). Signed-off-by: Lorenzo Colitti Signed-off-by: David S. Miller

net: ipv6: Introduce ip6_sk_dst_hoplimit.

2014-04-30T17:31:26Z

This replaces 6 identical code snippets with a call to a new static inline function. Signed-off-by: Lorenzo Colitti Signed-off-by: David S. Miller

ipv4: add a sock pointer to dst->output() path.

2014-04-15T17:47:15Z

In the dst->output() path for ipv4, the code assumes the skb it has to transmit is attached to an inet socket, specifically via ip_mc_output() : The sk_mc_loop() test triggers a WARN_ON() when the provider of the packet is an AF_PACKET socket. The dst->output() method gets an additional 'struct sock *sk' parameter. This needs a cascade of changes so that this parameter can be propagated from vxlan to final consumer. Fixes: 8f646c922d55 ("vxlan: keep original skb ownership") Reported-by: lucien xin Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller