diff options
author | Eric Dumazet <eric.dumazet@gmail.com> | 2010-04-08 19:03:29 -0400 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2010-04-13 04:41:33 -0400 |
commit | b6c6712a42ca3f9fa7f4a3d7c40e3a9dd1fd9e03 (patch) | |
tree | 42032b4978874e8ffcf6c851d13324b8c8c7c113 /net/ipv4/af_inet.c | |
parent | 7a161ea92471087a1579239d7a58dd06eaa5601c (diff) |
net: sk_dst_cache RCUification
With latest CONFIG_PROVE_RCU stuff, I felt more comfortable to make this
work.
sk->sk_dst_cache is currently protected by a rwlock (sk_dst_lock)
This rwlock is readlocked for a very small amount of time, and dst
entries are already freed after RCU grace period. This calls for RCU
again :)
This patch converts sk_dst_lock to a spinlock, and use RCU for readers.
__sk_dst_get() is supposed to be called with rcu_read_lock() or if
socket locked by user, so use appropriate rcu_dereference_check()
condition (rcu_read_lock_held() || sock_owned_by_user(sk))
This patch avoids two atomic ops per tx packet on UDP connected sockets,
for example, and permits sk_dst_lock to be much less dirtied.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/af_inet.c')
-rw-r--r-- | net/ipv4/af_inet.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a0beb32beaa3..193dcd6ed64f 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c | |||
@@ -154,7 +154,7 @@ void inet_sock_destruct(struct sock *sk) | |||
154 | WARN_ON(sk->sk_forward_alloc); | 154 | WARN_ON(sk->sk_forward_alloc); |
155 | 155 | ||
156 | kfree(inet->opt); | 156 | kfree(inet->opt); |
157 | dst_release(sk->sk_dst_cache); | 157 | dst_release(rcu_dereference_check(sk->sk_dst_cache, 1)); |
158 | sk_refcnt_debug_dec(sk); | 158 | sk_refcnt_debug_dec(sk); |
159 | } | 159 | } |
160 | EXPORT_SYMBOL(inet_sock_destruct); | 160 | EXPORT_SYMBOL(inet_sock_destruct); |