From: Ingo Molnar <mingo@elte.hu>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
bugzilla-daemon@bugzilla.kernel.org,
bugme-daemon@bugzilla.kernel.org, ant.starikov@gmail.com,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [Bugme-new] [Bug 15618] New: 2.6.18->2.6.32->2.6.33 huge regression in performance
Date: Tue, 23 Mar 2010 19:27:35 +0100 [thread overview]
Message-ID: <20100323182735.GA10897@elte.hu> (raw)
In-Reply-To: <20100323111351.756c8752.akpm@linux-foundation.org>
* Andrew Morton <akpm@linux-foundation.org> wrote:
> On Tue, 23 Mar 2010 18:34:09 +0100
> Ingo Molnar <mingo@elte.hu> wrote:
>
> >
> > It shows a very brutal amount of page fault invoked mmap_sem spinning
> > overhead.
> >
>
> Yes. Note that we fall off a cliff at nine threads on a 16-way. As soon as
> a core gets two threads scheduled onto it?
it's AMD Opterons so no SMT.
My (wild) guess would be that 8 cpus can still do cacheline ping-pong
reasonably efficiently, but it starts breaking down very seriously with 9 or
more cores bouncing the same single cache-line.
Breakdowns in scalability are usually very non-linear, for hardware and
software reasons. '8 threads' sounds like a hw limit to me. From the scheduler
POV there's no big difference between 8 or 9 CPUs used [this is non-HT] - with
8 or 7 cores still idle.
Ingo
WARNING: multiple messages have this Message-ID (diff)
From: Ingo Molnar <mingo@elte.hu>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
bugzilla-daemon@bugzilla.kernel.org,
bugme-daemon@bugzilla.kernel.org, ant.starikov@gmail.com,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [Bugme-new] [Bug 15618] New: 2.6.18->2.6.32->2.6.33 huge regression in performance
Date: Tue, 23 Mar 2010 19:27:35 +0100 [thread overview]
Message-ID: <20100323182735.GA10897@elte.hu> (raw)
In-Reply-To: <20100323111351.756c8752.akpm@linux-foundation.org>
* Andrew Morton <akpm@linux-foundation.org> wrote:
> On Tue, 23 Mar 2010 18:34:09 +0100
> Ingo Molnar <mingo@elte.hu> wrote:
>
> >
> > It shows a very brutal amount of page fault invoked mmap_sem spinning
> > overhead.
> >
>
> Yes. Note that we fall off a cliff at nine threads on a 16-way. As soon as
> a core gets two threads scheduled onto it?
it's AMD Opterons so no SMT.
My (wild) guess would be that 8 cpus can still do cacheline ping-pong
reasonably efficiently, but it starts breaking down very seriously with 9 or
more cores bouncing the same single cache-line.
Breakdowns in scalability are usually very non-linear, for hardware and
software reasons. '8 threads' sounds like a hw limit to me. From the scheduler
POV there's no big difference between 8 or 9 CPUs used [this is non-HT] - with
8 or 7 cores still idle.
Ingo
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2010-03-23 18:27 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <bug-15618-10286@https.bugzilla.kernel.org/>
2010-03-23 14:22 ` [Bugme-new] [Bug 15618] New: 2.6.18->2.6.32->2.6.33 huge regression in performance Andrew Morton
2010-03-23 14:22 ` Andrew Morton
2010-03-23 17:34 ` Ingo Molnar
2010-03-23 17:34 ` Ingo Molnar
2010-03-23 17:45 ` Linus Torvalds
2010-03-23 17:45 ` Linus Torvalds
2010-03-23 17:57 ` Anton Starikov
2010-03-23 17:57 ` Anton Starikov
2010-03-23 18:00 ` Ingo Molnar
2010-03-23 18:00 ` Ingo Molnar
2010-03-23 18:03 ` Anton Starikov
2010-03-23 18:03 ` Anton Starikov
2010-03-23 18:21 ` Andrew Morton
2010-03-23 18:21 ` Andrew Morton
2010-03-23 18:25 ` Anton Starikov
2010-03-23 18:25 ` Anton Starikov
2010-03-23 19:22 ` Robin Holt
2010-03-23 19:22 ` Robin Holt
2010-03-23 19:30 ` Anton Starikov
2010-03-23 19:30 ` Anton Starikov
2010-03-23 19:49 ` Robin Holt
2010-03-23 19:49 ` Robin Holt
2010-03-23 19:57 ` Robin Holt
2010-03-23 19:57 ` Robin Holt
2010-03-23 19:50 ` Anton Starikov
2010-03-23 19:50 ` Anton Starikov
2010-03-23 19:52 ` Linus Torvalds
2010-03-23 19:52 ` Linus Torvalds
2010-03-24 16:40 ` Roland Dreier
2010-03-24 16:40 ` Roland Dreier
2010-03-26 3:24 ` Anton Starikov
2010-03-26 3:24 ` Anton Starikov
2010-03-23 19:14 ` Anton Starikov
2010-03-23 19:14 ` Anton Starikov
2010-03-23 19:17 ` Peter Zijlstra
2010-03-23 19:17 ` Peter Zijlstra
2010-03-23 19:42 ` Anton Starikov
2010-03-23 19:54 ` Linus Torvalds
2010-03-23 19:54 ` Linus Torvalds
2010-03-23 20:43 ` Anton Starikov
2010-03-23 20:43 ` Anton Starikov
2010-03-23 23:04 ` Linus Torvalds
2010-03-23 23:04 ` Linus Torvalds
2010-03-23 23:19 ` Anton Starikov
2010-03-23 23:19 ` Anton Starikov
2010-03-23 23:36 ` Ingo Molnar
2010-03-23 23:36 ` Ingo Molnar
2010-03-23 23:55 ` Linus Torvalds
2010-03-23 23:55 ` Linus Torvalds
2010-03-24 0:03 ` Anton Starikov
2010-03-24 0:03 ` Anton Starikov
2010-03-24 2:15 ` Andi Kleen
2010-03-24 2:15 ` Andi Kleen
2010-03-24 3:00 ` Linus Torvalds
2010-03-24 3:00 ` Linus Torvalds
2010-04-19 18:19 ` Greg KH
2010-04-19 18:19 ` Greg KH
2010-03-23 18:13 ` Andrew Morton
2010-03-23 18:13 ` Andrew Morton
2010-03-23 18:19 ` Anton Starikov
2010-03-23 18:19 ` Anton Starikov
2010-03-23 18:27 ` Ingo Molnar [this message]
2010-03-23 18:27 ` Ingo Molnar
2010-03-23 21:19 ` Anton Starikov
2010-03-23 21:19 ` Anton Starikov
2010-04-02 18:57 ` Lee Schermerhorn
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=20100323182735.GA10897@elte.hu \
--to=mingo@elte.hu \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=ant.starikov@gmail.com \
--cc=bugme-daemon@bugzilla.kernel.org \
--cc=bugzilla-daemon@bugzilla.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.