Some tweaks to the cpu affinity code.

 - Split up choosing of cpu between fork and "normal" cases. Fork is
   very different and should be treated as such.
 - Instead of implicitly choosing a cpu in setrunqueue, do it outside
   where it actually makes sense.
 - Just because a cpu is marked as idle doesn't mean it will be soon.
   There could be a thundering herd effect if we call wakeup from an
   interrupt handler, so subtract cpus with queued processes when
   deciding which cpu is actually idle.
 - some simplifications allowed by the above.

kettenis@ ok (except one bugfix that was not in the intial diff)

1 files changed, 2 insertions, 1 deletions

diff --git a/sys/kern/kern_fork.c b/sys/kern/kern_fork.c
index 10dd9a5f8eb..b2947bd4421 100644
--- a/sys/kern/kern_fork.c
+++ b/sys/kern/kern_fork.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: kern_fork.c,v 1.102 2009/03/23 13:25:11 art Exp $	*/
+/*	$OpenBSD: kern_fork.c,v 1.103 2009/04/14 09:13:25 art Exp $	*/
 /*	$NetBSD: kern_fork.c,v 1.29 1996/02/09 18:59:34 christos Exp $	*/
 
 /*
@@ -432,6 +432,7 @@ fork1(struct proc *p1, int exitsig, int flags, void *stack, size_t stacksize,
  	getmicrotime(&p2->p_stats->p_start);
 	p2->p_acflag = AFORK;
 	p2->p_stat = SRUN;
+	p2->p_cpu = sched_choosecpu_fork(p1, flags);
 	setrunqueue(p2);
 	SCHED_UNLOCK(s);