linux-kernel - Re: [tip:sched/core] sched: cpuacct: Use bigger percpu counter batch values for stats counters

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]

Message-Id: <20090820141638.bcd298c0.kamezawa.hiroyu@jp.fujitsu.com>
Date:	Thu, 20 Aug 2009 14:16:38 +0900
From:	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
To:	Anton Blanchard <anton@...ba.org>
Cc:	Bharata B Rao <bharata@...ux.vnet.ibm.com>,
	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
	Ingo Molnar <mingo@...e.hu>,
	Balbir Singh <balbir@...ux.vnet.ibm.com>, mingo@...hat.com,
	hpa@...or.com, linux-kernel@...r.kernel.org,
	a.p.zijlstra@...llo.nl, schwidefsky@...ibm.com,
	balajirrao@...il.com, dhaval@...ux.vnet.ibm.com,
	tglx@...utronix.de, akpm@...ux-foundation.org
Subject: Re: [tip:sched/core] sched: cpuacct: Use bigger percpu counter
 batch values for stats counters

On Thu, 20 Aug 2009 15:10:38 +1000
Anton Blanchard <anton@...ba.org> wrote:

> 
> Hi,
> 
> Looks like this issue is still present. I tested on a 32 core box and 
> the patch improved maximum context switch rate from from 76k/sec to 9.5M/sec.
> Thats over 100x faster, or 50x per line of code. That's got to be some sort of
> record :)
> 
> Any chance we can get a fix in for 2.6.31? Don't make me find an even bigger
> box so I can break the 200x mark :)
> 
I like the patch itself, I think it does a correct fix.

Hmm, but, IIRC, there was a discssion "what happens at reading
this counter once per 1ms ?" or some.

What overhead at read on big system ? Can you show it ?

Thanks,
-Kame


> Anton
> 
> > --
> > 
> > When CONFIG_VIRT_CPU_ACCOUNTING is enabled we can call cpuacct_update_stats
> > with values much larger than percpu_counter_batch. This means the
> > call to percpu_counter_add will always add to the global count which is
> > protected by a spinlock.
> > 
> > Since reading of the CPU accounting cgroup counters is not performance
> > critical, we can use a maximum size batch of INT_MAX and use
> > percpu_counter_sum on the read side which will add all the percpu
> > counters.
> > 
> > With this patch an 8 core POWER6 with CONFIG_VIRT_CPU_ACCOUNTING and
> > CONFIG_CGROUP_CPUACCT shows an improvement in aggregate context switch rate of
> > 397k/sec to 3.9M/sec, a 10x improvement.
> > 
> > Signed-off-by: Anton Blanchard <anton@...ba.org>
> > ---
> > 
> > Index: linux.trees.git/kernel/sched.c
> > ===================================================================
> > --- linux.trees.git.orig/kernel/sched.c	2009-07-16 10:11:02.000000000 +1000
> > +++ linux.trees.git/kernel/sched.c	2009-07-16 10:16:41.000000000 +1000
> > @@ -10551,7 +10551,7 @@
> >  	int i;
> >  
> >  	for (i = 0; i < CPUACCT_STAT_NSTATS; i++) {
> > -		s64 val = percpu_counter_read(&ca->cpustat[i]);
> > +		s64 val = percpu_counter_sum(&ca->cpustat[i]);
> >  		val = cputime64_to_clock_t(val);
> >  		cb->fill(cb, cpuacct_stat_desc[i], val);
> >  	}
> > @@ -10621,7 +10621,7 @@
> >  	ca = task_ca(tsk);
> >  
> >  	do {
> > -		percpu_counter_add(&ca->cpustat[idx], val);
> > +		__percpu_counter_add(&ca->cpustat[idx], val, INT_MAX);
> >  		ca = ca->parent;
> >  	} while (ca);
> >  	rcu_read_unlock();
> 

--
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/