[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1184879223.5516.45.camel@localhost.localdomain>
Date: Thu, 19 Jul 2007 21:07:03 +0000
From: Jan Glauber <jang@...ux.vnet.ibm.com>
To: Ingo Molnar <mingo@...e.hu>
Cc: Jeremy Fitzhardinge <jeremy@...p.org>,
LKML <linux-kernel@...r.kernel.org>, vatsa@...ux.vnet.ibm.com,
mschwid2@...ux.vnet.ibm.com, efault@....de,
dmitry.adamushko@...il.com, paulus@...ba.org, anton@...ba.org
Subject: Re: [PATCH] virtual sched_clock() for s390
On Thu, 2007-07-19 at 21:38 +0200, Ingo Molnar wrote:
> * Jan Glauber <jang@...ux.vnet.ibm.com> wrote:
>
> > > still, CFS needs time measurement across idle periods as well, for
> > > another purpose: to be able to do precise task statistics for /proc.
> > > (for top, ps, etc.) So it's still true that sched_clock() should
> > > include idle periods too.
> >
> > I'm not sure, s390 already has an implemetation for precise accounting
> > in the architecture code, does CFS also improve accounting data?
>
> what kind of precise accounting does s390 have in the architecture code?
> CFS changes task (and load) accounting to be sched_clock() driven in
> essence.
s390 has per-process accounting that is aware of virtual cpu time, implemented in
arch/s390/kernel/time.c: account_ticks() and arch/s390/kernel/vtime.c:
account_*_vtime(). Timestamps are taken in entry.S for system calls, interrupts
and other system entries and are accounted later, we don't call update_process_times().
Jan
> Ingo
> -
> 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/
-
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/
Powered by blists - more mailing lists