public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org,
	Jan Glauber <jang@linux.vnet.ibm.com>,
	heiko.carstens@de.ibm.com, Paul Mackerras <paulus@samba.org>
Subject: Re: [accounting regression since rc1]  scheduler updates
Date: Tue, 21 Aug 2007 11:11:16 +0200	[thread overview]
Message-ID: <1187687476.7623.8.camel@localhost> (raw)
In-Reply-To: <20070821084243.GB1144@elte.hu>

On Tue, 2007-08-21 at 10:42 +0200, Ingo Molnar wrote:
> * Christian Borntraeger <borntraeger@de.ibm.com> wrote:
> 
> > Am Montag, 20. August 2007 schrieb Ingo Molnar:
> > > could you send that precise sched_clock() patch? It should be an order 
> > > of magnitude simpler than the high-precision stime/utime tracking you 
> > > already do, and it's needed for quality scheduling anyway.
> > 
> > I have a question about that. I just played with sched_clock, and even 
> > when I intentionally slow down sched_clock by a factor of 2, my cpu 
> > bound process gets 100 % in top. If this is intentional, I dont 
> > understand how a virtualized sched_clock would fix the accounting 
> > change?
> 
> hm, does on s390 scheduler_tick() get driven in virtual time or in real 
> time? The very latest scheduler code will enforce a minimum rate of 
> sched_clock() across two scheduler_tick() calls (in rc3 and later 
> kernels). If sched_clock() "slows down" but scheduler_tick() still has a 
> real-time frequency then that impacts the quality of scheduling. So 
> scheduler_tick() and sched_clock() must really have the same behavior 
> (either both are virtual or both are real), so that scheduling becomes 
> invariant to steal-time.

scheduler_tick() is based on the HZ timer which uses the TOD clock =
real time. sched_clock() currently uses the TOD clock as well so in
regard to the new scheduler we currently do not have a problem. We have
a problem with cpu time accounting, the change to the /proc code breaks
the precise accounting on s390. To solve the cpu time accounting we need
to change sched_clock() to the cpu timer = virtual time. To change the
scheduler_tick() as well requires another patch and I fear it would
complicate things in the s390 backend.

And if you say that the scheduling becomes invariant to steal-time, how
is the cpu time accounting via sum_exec supposed to work if it does not
take steal-time into account ?

-- 
blue skies,
  Martin.

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



  reply	other threads:[~2007-08-21  9:07 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-08-12 16:32 [git pull request] scheduler updates Ingo Molnar
2007-08-14  8:37 ` [accounting regression since rc1] " Christian Borntraeger
2007-08-16  8:17   ` [PATCH][RFC] Re: accounting regression since rc1 Christian Borntraeger
2007-08-20 15:45   ` [accounting regression since rc1] scheduler updates Ingo Molnar
2007-08-20 17:03     ` Martin Schwidefsky
2007-08-20 18:08       ` Ingo Molnar
2007-08-20 18:33         ` Martin Schwidefsky
2007-08-20 19:00           ` Balbir Singh
2007-08-20 19:05           ` Ingo Molnar
2007-08-21  7:20             ` Christian Borntraeger
2007-08-20 19:12           ` Ingo Molnar
2007-08-21  7:00           ` Christian Borntraeger
2007-08-21  9:18             ` Martin Schwidefsky
2007-08-20 23:07         ` Paul Mackerras
2007-08-21  2:18         ` Andi Kleen
2007-08-21  7:09           ` Ingo Molnar
2007-08-21 10:07             ` Andi Kleen
2007-08-21 10:20               ` Ingo Molnar
2007-08-21 11:15                 ` Andi Kleen
2007-08-21 11:20                   ` Ingo Molnar
2007-08-21  8:17     ` Christian Borntraeger
2007-08-21  8:42       ` Ingo Molnar
2007-08-21  9:11         ` Martin Schwidefsky [this message]
2007-08-21  9:34           ` Ingo Molnar
2007-08-21  9:48             ` Paul Mackerras
2007-08-21 10:38             ` Martin Schwidefsky
2007-08-21 11:36               ` Ingo Molnar
2007-08-21 11:58                 ` Martin Schwidefsky
2007-08-21 10:39             ` Christian Borntraeger
2007-08-21 10:43             ` Christian Borntraeger
2007-08-21 11:15               ` Ingo Molnar
2007-08-21 11:24                 ` Christian Borntraeger
2007-08-21 11:30                   ` Ingo Molnar
2007-08-21 11:58                     ` Christian Borntraeger
2007-08-21 12:21                       ` Ingo Molnar
2007-08-21 12:57                         ` Martin Schwidefsky
2007-08-21 11:25       ` Ingo Molnar
2007-08-22  7:50         ` Christian Borntraeger
2007-08-22  7:59           ` Ingo Molnar
     [not found] ` <200708141032.47235.borntraeger@de.ibm.com>
     [not found]   ` <alpine.LFD.0.999.0708140835240.30176@woody.linux-foundation.org>
2007-08-14 18:19     ` Christian Borntraeger

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1187687476.7623.8.camel@localhost \
    --to=schwidefsky@de.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=borntraeger@de.ibm.com \
    --cc=heiko.carstens@de.ibm.com \
    --cc=jang@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=paulus@samba.org \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox