summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorRainer Weikusat <rweikusat@mobileactivedefense.com>2015-11-26 19:23:15 +0000
committerDavid S. Miller <davem@davemloft.net>2015-12-01 14:57:43 -0500
commit77b75f4d8cf105b599beef38724f8171e557919d (patch)
treec16747ed0980aa1f8542745a3111bc5ed09f18b1
parent75c4a57ca9ee747f1cbb24591d92a41b2eddff52 (diff)
downloadlinux-77b75f4d8cf105b599beef38724f8171e557919d.tar.gz
linux-77b75f4d8cf105b599beef38724f8171e557919d.tar.bz2
linux-77b75f4d8cf105b599beef38724f8171e557919d.zip
unix: use wq_has_sleeper in unix_dgram_recvmsg
The current unix_dgram_recvmsg does a wake up for every received datagram. This seems wasteful as only SOCK_DGRAM client sockets in an n:1 association with a server socket will ever wait because of the associated condition. The patch below changes the function such that the wake up only happens if wq_has_sleeper indicates that someone actually wants to be notified. Testing with SOCK_SEQPACKET and SOCK_DGRAM socket seems to confirm that this is an improvment. Signed-Off-By: Rainer Weikusat <rweikusat@mobileactivedefense.com> Signed-off-by: David S. Miller <davem@davemloft.net>
-rw-r--r--net/unix/af_unix.c6
1 files changed, 4 insertions, 2 deletions
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index efb706e1d1c0..ac011b97097d 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1914,8 +1914,10 @@ static int unix_dgram_recvmsg(struct socket *sock, struct msghdr *msg,
goto out_unlock;
}
- wake_up_interruptible_sync_poll(&u->peer_wait,
- POLLOUT | POLLWRNORM | POLLWRBAND);
+ if (wq_has_sleeper(&u->peer_wait))
+ wake_up_interruptible_sync_poll(&u->peer_wait,
+ POLLOUT | POLLWRNORM |
+ POLLWRBAND);
if (msg->msg_name)
unix_copy_addr(msg, skb->sk);