Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

PR kern/38707 scheduler related deadlock during build.sh

Browse Source 
Fail sched_catchlwp() if mutex_tryenter() on the remote CPU's state fails.
Seems to work around the issue described in this PR.

XXX Stealing jobs from remote CPUs could probably be moved into the idle
loop, making the locking quite a bit simpler.
This commit is contained in:
ad 2008-05-25 23:46:55 +00:00
parent 99f94fd98d
commit c7615c48c8
1 changed files with 3 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
sys/kern/kern_runq.c
View File

@ -1,4 +1,4 @@
/* $NetBSD: kern_runq.c,v 1.8 2008/05/21 15:41:03 ad Exp $ */
/* $NetBSD: kern_runq.c,v 1.9 2008/05/25 23:46:55 ad Exp $ */
/*
* Copyright (c) 2007, 2008 Mindaugas Rasiukevicius <rmind at NetBSD org>
@ -27,7 +27,7 @@
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: kern_runq.c,v 1.8 2008/05/21 15:41:03 ad Exp $");
__KERNEL_RCSID(0, "$NetBSD: kern_runq.c,v 1.9 2008/05/25 23:46:55 ad Exp $");
#include <sys/param.h>
#include <sys/kernel.h>
@ -417,33 +417,11 @@ sched_catchlwp(void)
runqueue_t *ci_rq;
struct lwp *l;
if (curci == ci)
if (curci == ci || !mutex_tryenter(ci->ci_schedstate.spc_mutex))
return NULL;
/* Lockless check */
spc = &ci->ci_schedstate;
ci_rq = spc->spc_sched_info;
if (ci_rq->r_mcount < min_catch)
return NULL;
/*
* Double-lock the runqueues.
*/
if (curci < ci) {
spc_lock(ci);
} else if (!mutex_tryenter(ci->ci_schedstate.spc_mutex)) {
const runqueue_t *cur_rq = curci->ci_schedstate.spc_sched_info;
spc_unlock(curci);
spc_lock(ci);
spc_lock(curci);
if (cur_rq->r_count) {
spc_unlock(ci);
return NULL;
}
}
if (ci_rq->r_mcount < min_catch) {
spc_unlock(ci);
return NULL;
@ -599,8 +577,6 @@ sched_nextlwp(void)
/* Reset the counter, and call the balancer */
ci_rq->r_avgcount = 0;
sched_balance(ci);
/* The re-locking will be done inside */
return sched_catchlwp();
}
#else