diff options
| author | Eric Dumazet <edumazet@google.com> | 2023-09-11 17:05:30 +0000 |
|---|---|---|
| committer | Paolo Abeni <pabeni@redhat.com> | 2023-09-12 19:10:01 +0200 |
| commit | 4505dc2a522826975167823f64f0896bac1323fb (patch) | |
| tree | ca24379daf4c8c19ce173fdfd3bc510366cb9ad5 | |
| parent | 11445469dec803730ac7f78a5ad2a5ed131fa941 (diff) | |
net: call prot->release_cb() when processing backlog
__sk_flush_backlog() / sk_flush_backlog() are used
when TCP recvmsg()/sendmsg() process large chunks,
to not let packets in the backlog too long.
It makes sense to call tcp_release_cb() to also
process actions held in sk->sk_tsq_flags for smoother
scheduling.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
| -rw-r--r-- | net/core/sock.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/net/core/sock.c b/net/core/sock.c index 21610e3845a5..bb89b88bc1e8 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3001,6 +3001,9 @@ void __sk_flush_backlog(struct sock *sk) { spin_lock_bh(&sk->sk_lock.slock); __release_sock(sk); + + if (sk->sk_prot->release_cb) + sk->sk_prot->release_cb(sk); spin_unlock_bh(&sk->sk_lock.slock); } EXPORT_SYMBOL_GPL(__sk_flush_backlog); |