From ade004edddcdab33faee7c7b5833d2bf917e71c0 Mon Sep 17 00:00:00 2001 From: deraadt Date: Sat, 31 Oct 2009 06:40:14 +0000 Subject: Add missing KNOTE() calls after selwakeup(), until we decide if the KNOTE() calls can go directly into selwakeup() safely long discussion with nicm, murmers of consent from tedu and miod, noone else seems to care of kqueue is busted as long as it makes their sockets move data fast... pretty sad. --- sys/dev/systrace.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'sys/dev/systrace.c') diff --git a/sys/dev/systrace.c b/sys/dev/systrace.c index 2ad38420036..f3ede4462d1 100644 --- a/sys/dev/systrace.c +++ b/sys/dev/systrace.c @@ -1,4 +1,4 @@ -/* $OpenBSD: systrace.c,v 1.50 2009/07/09 22:29:55 thib Exp $ */ +/* $OpenBSD: systrace.c,v 1.51 2009/10/31 06:40:16 deraadt Exp $ */ /* * Copyright 2002 Niels Provos * All rights reserved. @@ -609,6 +609,7 @@ systrace_wakeup(struct fsystrace *fst) { wakeup((caddr_t)fst); selwakeup(&fst->si); + KNOTE(&fst->si.si_note, 0); } struct proc * -- cgit v1.2.3-59-g8ed1b