From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: Christoph Lameter <clameter@sgi.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>, Paul Jackson <pj@sgi.com>,
akpm@osdl.org, linux-mm@kvack.org, rientjes@google.com,
ak@suse.de
Subject: Re: [PATCH] mm: exempt pcp alloc from watermarks
Date: Mon, 18 Sep 2006 22:43:26 +0200 [thread overview]
Message-ID: <1158612206.3278.5.camel@taijtu> (raw)
In-Reply-To: <Pine.LNX.4.64.0609181317520.28726@schroedinger.engr.sgi.com>
On Mon, 2006-09-18 at 13:20 -0700, Christoph Lameter wrote:
> On Mon, 18 Sep 2006, Peter Zijlstra wrote:
>
> > On Sun, 2006-09-17 at 23:52 +1000, Nick Piggin wrote:
> >
> > > What we could do then, is allocate pages in batches (we already do),
> > > but only check watermarks if we have to go to the buddly allocator
> > > (we don't currently do this, but really should anyway, considering
> > > that the watermark checks are based on pages in the buddy allocator
> > > rather than pages in buddy + pcp).
>
> buffered_rmqueue has never checked watermarks. Seems that this is a
> fragment of a larger discussion and someone added those checks?
get_page_from_freelist() seems to check the watermarks before calling
buffered_rmqueue(), so if the watermarks fail, we never get to
buffered_rmqueue().
This patch adds a path to the per cpu pagelists before checking the
watermarks, however it will not refill the pcps when empty.
So now we can deplete the pcps even though we fail the watermark; which
is correct since the free_pages count is excluding the pcp pages.
--
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:[~2006-09-18 20:43 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-13 23:50 [PATCH] GFP_THISNODE for the slab allocator Christoph Lameter
2006-09-15 5:00 ` Andrew Morton
2006-09-15 6:49 ` Paul Jackson
2006-09-15 7:23 ` Andrew Morton
2006-09-15 7:44 ` Paul Jackson
2006-09-15 8:06 ` Andrew Morton
2006-09-15 15:53 ` David Rientjes
2006-09-15 23:03 ` David Rientjes
2006-09-16 0:04 ` Paul Jackson
2006-09-16 1:36 ` Andrew Morton
2006-09-16 2:23 ` Christoph Lameter
2006-09-16 4:34 ` Andrew Morton
2006-09-16 3:28 ` [PATCH] Add node to zone for the NUMA case Christoph Lameter
2006-09-16 3:40 ` Paul Jackson
2006-09-16 3:45 ` [PATCH] GFP_THISNODE for the slab allocator Paul Jackson
2006-09-16 2:47 ` Christoph Lameter
2006-09-17 3:45 ` David Rientjes
2006-09-17 11:17 ` Paul Jackson
2006-09-17 12:41 ` Christoph Lameter
2006-09-17 13:03 ` Paul Jackson
2006-09-17 20:36 ` David Rientjes
2006-09-17 21:20 ` Paul Jackson
2006-09-17 22:27 ` Paul Jackson
2006-09-17 23:49 ` David Rientjes
2006-09-18 2:20 ` Paul Jackson
2006-09-18 16:34 ` Paul Jackson
2006-09-18 17:49 ` David Rientjes
2006-09-18 20:46 ` Paul Jackson
2006-09-19 20:52 ` David Rientjes
2006-09-19 21:26 ` Christoph Lameter
2006-09-19 21:50 ` David Rientjes
2006-09-21 22:11 ` David Rientjes
2006-09-22 10:10 ` Nick Piggin
2006-09-22 16:26 ` Paul Jackson
2006-09-22 16:36 ` Christoph Lameter
2006-09-15 8:28 ` Andrew Morton
2006-09-16 3:38 ` Paul Jackson
2006-09-16 4:42 ` Andi Kleen
2006-09-16 11:38 ` Paul Jackson
2006-09-16 4:48 ` Andrew Morton
2006-09-16 11:30 ` Paul Jackson
2006-09-16 15:18 ` Andrew Morton
2006-09-17 9:28 ` Paul Jackson
2006-09-17 9:51 ` Nick Piggin
2006-09-17 11:15 ` Paul Jackson
2006-09-17 12:44 ` Nick Piggin
2006-09-17 13:19 ` Paul Jackson
2006-09-17 13:52 ` Nick Piggin
2006-09-17 21:19 ` Paul Jackson
2006-09-18 12:44 ` [PATCH] mm: exempt pcp alloc from watermarks Peter Zijlstra
2006-09-18 20:20 ` Christoph Lameter
2006-09-18 20:43 ` Peter Zijlstra [this message]
2006-09-19 14:35 ` Nick Piggin
2006-09-19 14:44 ` Christoph Lameter
2006-09-19 15:02 ` Nick Piggin
2006-09-19 14:51 ` Peter Zijlstra
2006-09-19 15:10 ` Nick Piggin
2006-09-19 15:05 ` Peter Zijlstra
2006-09-19 15:39 ` Christoph Lameter
2006-09-17 16:29 ` [PATCH] GFP_THISNODE for the slab allocator Andrew Morton
2006-09-18 2:11 ` Paul Jackson
2006-09-18 5:09 ` Andrew Morton
2006-09-18 7:49 ` Paul Jackson
2006-09-16 11:48 ` Paul Jackson
2006-09-16 15:38 ` Andrew Morton
2006-09-16 21:51 ` Paul Jackson
2006-09-16 23:10 ` Andrew Morton
2006-09-17 4:37 ` Christoph Lameter
2006-09-17 4:55 ` Andrew Morton
2006-09-17 12:09 ` Paul Jackson
2006-09-17 12:36 ` Christoph Lameter
2006-09-17 13:06 ` Paul Jackson
2006-09-19 19:17 ` David Rientjes
2006-09-19 19:19 ` David Rientjes
2006-09-19 19:31 ` Christoph Lameter
2006-09-19 21:12 ` David Rientjes
2006-09-19 21:28 ` Christoph Lameter
2006-09-19 21:53 ` Paul Jackson
2006-09-15 17:08 ` Christoph Lameter
2006-09-15 17:37 ` [PATCH] Add NUMA_BUILD definition in kernel.h to avoid #ifdef CONFIG_NUMA Christoph Lameter
2006-09-15 17:38 ` [PATCH] Disable GFP_THISNODE in the non-NUMA case Christoph Lameter
2006-09-15 17:42 ` [PATCH] GFP_THISNODE for the slab allocator V2 Christoph Lameter
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=1158612206.3278.5.camel@taijtu \
--to=a.p.zijlstra@chello.nl \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=clameter@sgi.com \
--cc=linux-mm@kvack.org \
--cc=nickpiggin@yahoo.com.au \
--cc=pj@sgi.com \
--cc=rientjes@google.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