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

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 <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 5 insertions, 0 deletions

diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c
index a3095fdefbed..90c0e8386116 100644
--- a/net/ipv4/datagram.c
+++ b/net/ipv4/datagram.c
@@ -76,6 +76,7 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
         inet->inet_daddr = fl4->daddr;
         inet->inet_dport = usin->sin_port;
         sk->sk_state = TCP_ESTABLISHED;
+        inet_set_txhash(sk);
         inet->inet_id = jiffies;
 
         sk_dst_set(sk, &rt->dst);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 52d0f6a1ec2c..1edc739b9da5 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -208,6 +208,8 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
         inet->inet_dport = usin->sin_port;
         inet->inet_daddr = daddr;
 
+        inet_set_txhash(sk);
+
         inet_csk(sk)->icsk_ext_hdr_len = 0;
         if (inet_opt)
                 inet_csk(sk)->icsk_ext_hdr_len = inet_opt->opt.optlen;
@@ -1334,6 +1336,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
         newinet->mc_ttl       = ip_hdr(skb)->ttl;
         newinet->rcv_tos      = ip_hdr(skb)->tos;
         inet_csk(newsk)->icsk_ext_hdr_len = 0;
+        inet_set_txhash(newsk);
         if (inet_opt)
                 inet_csk(newsk)->icsk_ext_hdr_len = inet_opt->opt.optlen;
         newinet->inet_id = newtp->write_seq ^ jiffies;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index f8f2a944a1ce..bcee13c4627c 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -916,6 +916,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
         skb_orphan(skb);
         skb->sk = sk;
         skb->destructor = tcp_wfree;
+        skb_set_hash_from_sk(skb, sk);
         atomic_add(skb->truesize, &sk->sk_wmem_alloc);
 
         /* Build TCP header and checksum it. */