From: Rik van Riel <riel@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Preeti Murthy <preeti.lkml@gmail.com>,
umgwanakikbuti@gmail.com, LKML <linux-kernel@vger.kernel.org>,
Morten Rasmussen <morten.rasmussen@arm.com>,
Ingo Molnar <mingo@kernel.org>,
george.mccollister@gmail.com, ktkhai@parallels.com,
Preeti U Murthy <preeti@linux.vnet.ibm.com>
Subject: Re: [PATCH RFC/TEST] sched: make sync affine wakeups work
Date: Tue, 06 May 2014 16:20:59 -0400 [thread overview]
Message-ID: <5369442B.9010505@redhat.com> (raw)
In-Reply-To: <20140506132516.GJ11096@twins.programming.kicks-ass.net>
On 05/06/2014 09:25 AM, Peter Zijlstra wrote:
> On Sun, May 04, 2014 at 08:41:09AM -0400, Rik van Riel wrote:
>> Even on 8-node DL980 systems, the NUMA distance in the
>> SLIT table is less than RECLAIM_DISTANCE, and we will
>> do wake_affine across the entire system.
>
> Yeah, so the problem is that (AFAIK) ACPI doesn't actually specify a
> metric for the SLIT distance. This (in as far as BIOS people would care
> to stick to specs anyhow) has lead to the 'fun' situation where BIOS
> engineers tweak SLIT table values to make OSes behave as they thing it
> should.
>
> So if the BIOS engineer finds that this system should have <
> RECLAIM_DISTANCE it will simply make the table such that the max SLIT
> value is below that.
>
> And yes, I've seen this :-(
It appears to be the case on the vast majority of the NUMA
systems that are actually in use.
To me, this suggests that we should probably deal with it.
--
All rights reversed
next prev parent reply other threads:[~2014-05-06 20:21 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-02 4:42 [PATCH RFC/TEST] sched: make sync affine wakeups work Rik van Riel
2014-05-02 5:32 ` Mike Galbraith
2014-05-02 5:41 ` Mike Galbraith
2014-05-02 5:58 ` Mike Galbraith
2014-05-02 6:08 ` Rik van Riel
2014-05-02 6:36 ` Mike Galbraith
2014-05-02 6:51 ` Mike Galbraith
2014-05-02 6:13 ` Mike Galbraith
2014-05-02 6:30 ` Rik van Riel
2014-05-02 7:37 ` Mike Galbraith
2014-05-02 10:56 ` Rik van Riel
2014-05-02 11:27 ` Mike Galbraith
2014-05-02 12:51 ` Mike Galbraith
[not found] ` <5363B793.9010208@redhat.com>
2014-05-06 11:54 ` Peter Zijlstra
2014-05-06 20:19 ` Rik van Riel
2014-05-06 20:39 ` Peter Zijlstra
2014-05-06 23:46 ` Rik van Riel
2014-05-09 2:20 ` Rik van Riel
2014-05-09 5:27 ` [PATCH] sched: wake up task on prev_cpu if not in SD_WAKE_AFFINE domain with cpu Rik van Riel
2014-05-09 6:04 ` [PATCH] sched: clean up select_task_rq_fair conditionals and indentation Rik van Riel
2014-05-09 7:34 ` [PATCH] sched: wake up task on prev_cpu if not in SD_WAKE_AFFINE domain with cpu Mike Galbraith
2014-05-09 14:22 ` Rik van Riel
2014-05-09 15:24 ` Mike Galbraith
2014-05-09 15:24 ` Rik van Riel
2014-05-09 17:55 ` Mike Galbraith
2014-05-09 18:16 ` Rik van Riel
2014-05-10 3:54 ` Mike Galbraith
2014-05-13 14:08 ` Rik van Riel
2014-05-14 4:08 ` Mike Galbraith
2014-05-14 15:40 ` [PATCH] sched: call select_idle_sibling when not affine_sd Rik van Riel
2014-05-14 15:45 ` Peter Zijlstra
2014-05-19 13:08 ` [tip:sched/core] " tip-bot for Rik van Riel
2014-05-22 12:27 ` [tip:sched/core] sched: Call select_idle_sibling() " tip-bot for Rik van Riel
2014-05-04 11:44 ` [PATCH RFC/TEST] sched: make sync affine wakeups work Preeti Murthy
2014-05-04 12:04 ` Mike Galbraith
2014-05-05 4:38 ` Preeti U Murthy
2014-05-04 12:41 ` Rik van Riel
2014-05-05 4:50 ` Preeti U Murthy
2014-05-05 6:43 ` Preeti U Murthy
2014-05-05 11:28 ` Rik van Riel
2014-05-06 13:26 ` Peter Zijlstra
2014-05-06 13:25 ` Peter Zijlstra
2014-05-06 20:20 ` Rik van Riel [this message]
2014-05-06 20:41 ` Peter Zijlstra
2014-05-07 12:17 ` Ingo Molnar
2014-05-06 11:56 ` Peter Zijlstra
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=5369442B.9010505@redhat.com \
--to=riel@redhat.com \
--cc=george.mccollister@gmail.com \
--cc=ktkhai@parallels.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=peterz@infradead.org \
--cc=preeti.lkml@gmail.com \
--cc=preeti@linux.vnet.ibm.com \
--cc=umgwanakikbuti@gmail.com \
/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;
as well as URLs for NNTP newsgroup(s).