Date: Wed, 29 Apr 2009 10:02:06 +0200
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Andrew Gallatin <gallatin@myri.com>, linux-kernel@vger.kernel.org,
       rick.jones2@hp.com, brice@myri.com, Paul Mackerras <paulus@samba.org>,
       Benjamin Herrenschmidt <benh@kernel.crashing.org>,
       Ingo Molnar <mingo@elte.hu>
Subject: Re: IRQ / SoftIRQ CPU time accounting broken by
 457533a7d3402d1d91fbc125c8bd1bd16dcd3cd4
Message-ID: <20090429100206.6371b093@skybase>
In-Reply-To: <20090428163004.46733752.akpm@linux-foundation.org>
References: <49F078FA.6010507@myri.com>
	<20090428163004.46733752.akpm@linux-foundation.org>
Organization: IBM Corporation
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1885
Lines: 52

On Tue, 28 Apr 2009 16:30:04 -0700
Andrew Morton <akpm@linux-foundation.org> wrote:

> (cc's added)
> 
> On Thu, 23 Apr 2009 10:19:38 -0400
> Andrew Gallatin <gallatin@myri.com> wrote:
> 
> > When running netperf for some 10GbE tests, I noticed
> > that  IRQ and SOFTIRQ CPU time is no longer reported for an
> > otherwise idle CPU on recent kernels, at least for x86_64.
> > 
> > I spent a few hours git-bisecting until I finally got here:
> > 
> > % git-bisect bad
> > Bisecting: 0 revisions left to test after this
> > [457533a7d3402d1d91fbc125c8bd1bd16dcd3cd4] fix scaled & unscaled cputime 
> > accounting
> > 
> > I have neither CONFIG_NO_HZ, CONFIG_VIRT_CPU_ACCOUNTING, or XEN configured.
> > 
> 
> Thanks for doing the bisection.
> 
> 457533a7d3402d1d91fbc125c8bd1bd16dcd3cd4 was merged late last year, so
> this regression has been around for a while.
> 
> We might have fixed it in more recent kernels - stranger things have
> happened ;)

Hmm, are you sure that this commit caused the problem? I can't see how.
In regard to the accounting of system ticks on x86_64 the only change
is the bug fix that a system tick of a guest cpu has incorrectly been
added to p->stimescaled instead of p->utimescaled.
Without guest cpus in the picture nothing changes. For x86_64
CONFIG_VIRT_CPU_ACCOUNTING=n so account_process_tick is used. Before
the patch accout_process_tick accounted a system tick by calling
account_system_time and account_system_time_scaled. The second function
just added the cputime to p->stimescaled. That is now done in
accout_system_time.

-- 
blue skies,
   Martin.

"Reality continues to ruin my life." - Calvin.

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