diff options
| author |   dlg <dlg@openbsd.org> | 2019-04-14 06:57:00 +0000 |
|---|---|---|
| committer | dlg <dlg@openbsd.org> | 2019-04-14 06:57:00 +0000 |
| commit | 9b6e3526c472fd403c07739b9915bd87151f8ed3 (patch) | |
| tree | e8dbaf43c9cf1aa06256725867d4216747790a6e /sys/net/if.c | |
| parent | md5 auth has useful information on the wire that helps when fixing issues (diff) | |
| download | wireguard-openbsd-9b6e3526c472fd403c07739b9915bd87151f8ed3.tar.xz wireguard-openbsd-9b6e3526c472fd403c07739b9915bd87151f8ed3.zip | |
don't take splnet when running the network stack
the stack uses the NET_LOCK for most protection now, so it doesnt
need to block actual hardware interrupts. blocking hw interrupts
can cause huge latency spikes, which in turn works against the rx
ring moderation.
im putting this in early in the release cycle so it can get the
most testing possible.
ok mpi@ (a while back)
Diffstat (limited to 'sys/net/if.c')
| -rw-r--r-- | sys/net/if.c | 5 |
1 files changed, 1 insertions, 4 deletions
diff --git a/sys/net/if.c b/sys/net/if.c index 3526e56f535..3289400486e 100644 --- a/sys/net/if.c +++ b/sys/net/if.c @@ -1,4 +1,4 @@ -/* $OpenBSD: if.c,v 1.574 2019/04/10 09:51:35 dlg Exp $ */ +/* $OpenBSD: if.c,v 1.575 2019/04/14 06:57:00 dlg Exp $ */ /* $NetBSD: if.c,v 1.35 1996/05/07 05:26:04 thorpej Exp $ */ /* @@ -902,7 +902,6 @@ if_input_process(struct ifnet *ifp, struct mbuf_list *ml) struct mbuf *m; struct ifih *ifih; struct srp_ref sr; - int s; if (ml_empty(ml)) return; @@ -923,7 +922,6 @@ if_input_process(struct ifnet *ifp, struct mbuf_list *ml) * lists. */ NET_RLOCK(); - s = splnet(); while ((m = ml_dequeue(ml)) != NULL) { /* * Pass this mbuf to all input handlers of its @@ -938,7 +936,6 @@ if_input_process(struct ifnet *ifp, struct mbuf_list *ml) if (ifih == NULL) m_freem(m); } - splx(s); NET_RUNLOCK(); } |