From: Rik van Riel <riel@redhat.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: mingo@kernel.org, oleg@redhat.com, pjt@google.com,
akpm@linux-foundation.org, torvalds@linux-foundation.org,
tglx@linutronix.de, Lee.Schermerhorn@hp.com,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 18/19] sched, numa: Per task memory placement for big processes
Date: Tue, 31 Jul 2012 17:56:11 -0400 [thread overview]
Message-ID: <5018547B.4020000@redhat.com> (raw)
In-Reply-To: <20120731192809.428855038@chello.nl>
On 07/31/2012 03:12 PM, Peter Zijlstra wrote:
> Probability says that the task faulting on a page after we protect it,
> is most likely to be the task that uses that page most.
>
> To decrease the likelyhood of acting on a false relation, we only
> migrate a page when two consecutive samples are from the same task.
>
> I'm still not entirely convinced this scheme is sound, esp. for things
> like virtualization and n:m threading solutions in general the
> compute<->task relation is fundamentally untrue.
Again, we may need some additional code on top in the future,
eg. something like Andrea's policy that tries grouping related
tasks/threads together, but this looks like a very good way
to start.
We can introduce complexity if it is needed. Simplicity is good.
Acked-by: Rik van Riel
--
All rights reversed
next prev parent reply other threads:[~2012-07-31 21:56 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-31 19:12 [PATCH 00/19] sched-numa rewrite Peter Zijlstra
2012-07-31 19:12 ` [PATCH 01/19] task_work: Remove dependency on sched.h Peter Zijlstra
2012-07-31 20:52 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 02/19] mm/mpol: Remove NUMA_INTERLEAVE_HIT Peter Zijlstra
2012-07-31 20:52 ` Rik van Riel
2012-08-09 21:41 ` Andrea Arcangeli
2012-08-10 0:50 ` Andi Kleen
2012-07-31 19:12 ` [PATCH 03/19] mm/mpol: Make MPOL_LOCAL a real policy Peter Zijlstra
2012-07-31 20:52 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 04/19] mm, thp: Preserve pgprot across huge page split Peter Zijlstra
2012-07-31 20:53 ` Rik van Riel
2012-08-09 21:42 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 05/19] mm, mpol: Create special PROT_NONE infrastructure Peter Zijlstra
2012-07-31 20:55 ` Rik van Riel
2012-08-09 21:43 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 06/19] mm/mpol: Add MPOL_MF_LAZY Peter Zijlstra
2012-07-31 21:04 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 07/19] mm/mpol: Add MPOL_MF_NOOP Peter Zijlstra
2012-07-31 21:06 ` Rik van Riel
2012-08-09 21:44 ` Andrea Arcangeli
2012-10-01 9:36 ` Michael Kerrisk
2012-10-01 9:45 ` Ingo Molnar
2012-07-31 19:12 ` [PATCH 08/19] mm/mpol: Check for misplaced page Peter Zijlstra
2012-07-31 21:13 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 09/19] mm, migrate: Introduce migrate_misplaced_page() Peter Zijlstra
2012-07-31 21:16 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 10/19] mm, mpol: Use special PROT_NONE to migrate pages Peter Zijlstra
2012-07-31 21:24 ` Rik van Riel
2012-08-09 21:44 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 11/19] sched, mm: Introduce tsk_home_node() Peter Zijlstra
2012-07-31 21:30 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 12/19] mm/mpol: Make mempolicy home-node aware Peter Zijlstra
2012-07-31 21:33 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 13/19] sched: Introduce sched_feat_numa() Peter Zijlstra
2012-07-31 21:34 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 14/19] sched: Make find_busiest_queue() a method Peter Zijlstra
2012-07-31 21:34 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 15/19] sched: Implement home-node awareness Peter Zijlstra
2012-07-31 21:52 ` Rik van Riel
2012-08-09 21:51 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 16/19] sched, numa: NUMA home-node selection code Peter Zijlstra
2012-07-31 21:52 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 17/19] sched, numa: Detect big processes Peter Zijlstra
2012-07-31 21:53 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 18/19] sched, numa: Per task memory placement for " Peter Zijlstra
2012-07-31 21:56 ` Rik van Riel [this message]
2012-08-08 21:35 ` Peter Zijlstra
2012-08-09 21:57 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 19/19] mm, numa: retry failed page migrations Peter Zijlstra
2012-08-02 20:40 ` Christoph Lameter
2012-08-08 17:17 ` [PATCH 00/19] sched-numa rewrite Andrea Arcangeli
2012-08-08 18:43 ` Rik van Riel
2012-08-17 18:08 ` Andrea Arcangeli
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=5018547B.4020000@redhat.com \
--to=riel@redhat.com \
--cc=Lee.Schermerhorn@hp.com \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=pjt@google.com \
--cc=tglx@linutronix.de \
--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.