linux-kernel - Re: Very high CPU values in top on idle system (3.0-rc1)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 31 May 2011 11:55:45 +0200
From:	Peter Zijlstra <peterz@...radead.org>
To:	Markus Trippelsdorf <markus@...ppelsdorf.de>
Cc:	linux-kernel@...r.kernel.org, Mike Galbraith <efault@....de>,
	Ingo Molnar <mingo@...e.hu>
Subject: Re: Very high CPU values in top on idle system (3.0-rc1)

On Tue, 2011-05-31 at 00:12 +0200, Peter Zijlstra wrote:
> 
> I poked around with some skip_rq_update bits but couldn't make it go
> away, will try more tomorrow.

Gah, that was annoying, and explains why I wasn't seeing it on my main
dev box but could see it on my desktop (wsm tsc is much better synced
than the core2 tsc).

Could you confirm that this indeed cures the problem? I'll probably
commit a version without the toggle, but that was easy for verifying it
indeed made a difference..

---
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -2604,6 +2601,8 @@ static void ttwu_queue(struct task_struc
 
 #if defined(CONFIG_SMP)
 	if (sched_feat(TTWU_QUEUE) && cpu != smp_processor_id()) {
+		if (sched_feat(TTWU_SYNC_CLOCK))
+			sched_clock_cpu(cpu); /* sync clocks x-cpu */
 		ttwu_queue_remote(p, cpu);
 		return;
 	}
Index: linux-2.6/kernel/sched_features.h
===================================================================
--- linux-2.6.orig/kernel/sched_features.h
+++ linux-2.6/kernel/sched_features.h
@@ -70,3 +70,4 @@ SCHED_FEAT(NONIRQ_POWER, 1)
  * using the scheduler IPI. Reduces rq->lock contention/bounces.
  */
 SCHED_FEAT(TTWU_QUEUE, 1)
+SCHED_FEAT(TTWU_SYNC_CLOCK, 1)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/