From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: Christoph Lameter <clameter@sgi.com>
Cc: Matt Mackall <mpm@selenic.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Thomas Graf <tgraf@suug.ch>, David Miller <davem@davemloft.net>,
Andrew Morton <akpm@linux-foundation.org>,
Daniel Phillips <phillips@google.com>,
Pekka Enberg <penberg@cs.helsinki.fi>
Subject: Re: [PATCH 0/5] make slab gfp fair
Date: Wed, 16 May 2007 08:59:31 +0200 [thread overview]
Message-ID: <1179298771.7173.16.camel@twins> (raw)
In-Reply-To: <Pine.LNX.4.64.0705151457060.3155@schroedinger.engr.sgi.com>
On Tue, 2007-05-15 at 15:02 -0700, Christoph Lameter wrote:
> On Tue, 15 May 2007, Peter Zijlstra wrote:
>
> > How about something like this; it seems to sustain a little stress.
>
> Argh again mods to kmem_cache.
Hmm, I had not understood you minded that very much; I did stay away
from all the fast paths this time.
The thing is, I wanted to fold all the emergency allocs into a single
slab, not a per cpu thing. And once you loose the per cpu thing, you
need some extra serialization. Currently the top level lock is
slab_lock(page), but that only works because we have interrupts disabled
and work per cpu.
Why is it bad to extend kmem_cache a bit?
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: Christoph Lameter <clameter@sgi.com>
Cc: Matt Mackall <mpm@selenic.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Thomas Graf <tgraf@suug.ch>, David Miller <davem@davemloft.net>,
Andrew Morton <akpm@linux-foundation.org>,
Daniel Phillips <phillips@google.com>,
Pekka Enberg <penberg@cs.helsinki.fi>
Subject: Re: [PATCH 0/5] make slab gfp fair
Date: Wed, 16 May 2007 08:59:31 +0200 [thread overview]
Message-ID: <1179298771.7173.16.camel@twins> (raw)
In-Reply-To: <Pine.LNX.4.64.0705151457060.3155@schroedinger.engr.sgi.com>
On Tue, 2007-05-15 at 15:02 -0700, Christoph Lameter wrote:
> On Tue, 15 May 2007, Peter Zijlstra wrote:
>
> > How about something like this; it seems to sustain a little stress.
>
> Argh again mods to kmem_cache.
Hmm, I had not understood you minded that very much; I did stay away
from all the fast paths this time.
The thing is, I wanted to fold all the emergency allocs into a single
slab, not a per cpu thing. And once you loose the per cpu thing, you
need some extra serialization. Currently the top level lock is
slab_lock(page), but that only works because we have interrupts disabled
and work per cpu.
Why is it bad to extend kmem_cache a bit?
--
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:[~2007-05-16 6:59 UTC|newest]
Thread overview: 138+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-05-14 13:19 [PATCH 0/5] make slab gfp fair Peter Zijlstra
2007-05-14 13:19 ` Peter Zijlstra
2007-05-14 13:19 ` [PATCH 1/5] mm: page allocation rank Peter Zijlstra
2007-05-14 13:19 ` Peter Zijlstra
2007-05-14 13:19 ` [PATCH 2/5] mm: slab allocation fairness Peter Zijlstra
2007-05-14 13:19 ` Peter Zijlstra
2007-05-14 15:51 ` Christoph Lameter
2007-05-14 15:51 ` Christoph Lameter
2007-05-14 13:19 ` [PATCH 3/5] mm: slub " Peter Zijlstra
2007-05-14 13:19 ` Peter Zijlstra
2007-05-14 15:49 ` Christoph Lameter
2007-05-14 15:49 ` Christoph Lameter
2007-05-14 16:14 ` Peter Zijlstra
2007-05-14 16:14 ` Peter Zijlstra
2007-05-14 16:35 ` Christoph Lameter
2007-05-14 16:35 ` Christoph Lameter
2007-05-14 13:19 ` [PATCH 4/5] mm: slob " Peter Zijlstra
2007-05-14 13:19 ` Peter Zijlstra
2007-05-14 13:19 ` [PATCH 5/5] mm: allow mempool to fall back to memalloc reserves Peter Zijlstra
2007-05-14 13:19 ` Peter Zijlstra
2007-05-14 15:53 ` [PATCH 0/5] make slab gfp fair Christoph Lameter
2007-05-14 15:53 ` Christoph Lameter
2007-05-14 16:10 ` Peter Zijlstra
2007-05-14 16:10 ` Peter Zijlstra
2007-05-14 16:37 ` Christoph Lameter
2007-05-14 16:37 ` Christoph Lameter
2007-05-14 16:12 ` Matt Mackall
2007-05-14 16:12 ` Matt Mackall
2007-05-14 16:29 ` Christoph Lameter
2007-05-14 16:29 ` Christoph Lameter
2007-05-14 17:40 ` Peter Zijlstra
2007-05-14 17:40 ` Peter Zijlstra
2007-05-14 17:57 ` Christoph Lameter
2007-05-14 17:57 ` Christoph Lameter
2007-05-14 19:28 ` Peter Zijlstra
2007-05-14 19:28 ` Peter Zijlstra
2007-05-14 19:56 ` Christoph Lameter
2007-05-14 19:56 ` Christoph Lameter
2007-05-14 20:03 ` Peter Zijlstra
2007-05-14 20:03 ` Peter Zijlstra
2007-05-14 20:06 ` Christoph Lameter
2007-05-14 20:06 ` Christoph Lameter
2007-05-14 20:12 ` Peter Zijlstra
2007-05-14 20:12 ` Peter Zijlstra
2007-05-14 20:25 ` Christoph Lameter
2007-05-14 20:25 ` Christoph Lameter
2007-05-15 17:27 ` Peter Zijlstra
2007-05-15 17:27 ` Peter Zijlstra
2007-05-15 22:02 ` Christoph Lameter
2007-05-15 22:02 ` Christoph Lameter
2007-05-16 6:59 ` Peter Zijlstra [this message]
2007-05-16 6:59 ` Peter Zijlstra
2007-05-16 18:43 ` Christoph Lameter
2007-05-16 18:43 ` Christoph Lameter
2007-05-16 19:25 ` Peter Zijlstra
2007-05-16 19:25 ` Peter Zijlstra
2007-05-16 19:53 ` Christoph Lameter
2007-05-16 19:53 ` Christoph Lameter
2007-05-16 20:18 ` Peter Zijlstra
2007-05-16 20:18 ` Peter Zijlstra
2007-05-16 20:27 ` Christoph Lameter
2007-05-16 20:27 ` Christoph Lameter
2007-05-16 20:40 ` Peter Zijlstra
2007-05-16 20:40 ` Peter Zijlstra
2007-05-16 20:44 ` Christoph Lameter
2007-05-16 20:44 ` Christoph Lameter
2007-05-16 20:54 ` Peter Zijlstra
2007-05-16 20:54 ` Peter Zijlstra
2007-05-16 20:59 ` Christoph Lameter
2007-05-16 20:59 ` Christoph Lameter
2007-05-16 21:04 ` Peter Zijlstra
2007-05-16 21:04 ` Peter Zijlstra
2007-05-16 21:13 ` Christoph Lameter
2007-05-16 21:13 ` Christoph Lameter
2007-05-16 21:20 ` Peter Zijlstra
2007-05-16 21:20 ` Peter Zijlstra
2007-05-16 21:42 ` Christoph Lameter
2007-05-16 21:42 ` Christoph Lameter
2007-05-17 7:28 ` Peter Zijlstra
2007-05-17 7:28 ` Peter Zijlstra
2007-05-17 17:30 ` Christoph Lameter
2007-05-17 17:30 ` Christoph Lameter
2007-05-17 17:53 ` Peter Zijlstra
2007-05-17 17:53 ` Peter Zijlstra
2007-05-17 18:01 ` Christoph Lameter
2007-05-17 18:01 ` Christoph Lameter
2007-05-14 19:44 ` Andrew Morton
2007-05-14 19:44 ` Andrew Morton
2007-05-14 20:01 ` Matt Mackall
2007-05-14 20:01 ` Matt Mackall
2007-05-14 20:05 ` Peter Zijlstra
2007-05-14 20:05 ` Peter Zijlstra
2007-05-17 3:02 ` Christoph Lameter
2007-05-17 3:02 ` Christoph Lameter
2007-05-17 7:08 ` Peter Zijlstra
2007-05-17 7:08 ` Peter Zijlstra
2007-05-17 17:29 ` Christoph Lameter
2007-05-17 17:29 ` Christoph Lameter
2007-05-17 17:52 ` Peter Zijlstra
2007-05-17 17:52 ` Peter Zijlstra
2007-05-17 17:59 ` Christoph Lameter
2007-05-17 17:59 ` Christoph Lameter
2007-05-17 17:53 ` Matt Mackall
2007-05-17 17:53 ` Matt Mackall
2007-05-17 18:02 ` Christoph Lameter
2007-05-17 18:02 ` Christoph Lameter
2007-05-17 19:18 ` Peter Zijlstra
2007-05-17 19:18 ` Peter Zijlstra
2007-05-17 19:24 ` Christoph Lameter
2007-05-17 19:24 ` Christoph Lameter
2007-05-17 21:26 ` Peter Zijlstra
2007-05-17 21:26 ` Peter Zijlstra
2007-05-17 21:44 ` Paul Jackson
2007-05-17 21:44 ` Paul Jackson
2007-05-17 22:27 ` Christoph Lameter
2007-05-17 22:27 ` Christoph Lameter
2007-05-18 9:54 ` Peter Zijlstra
2007-05-18 9:54 ` Peter Zijlstra
2007-05-18 17:11 ` Paul Jackson
2007-05-18 17:11 ` Paul Jackson
2007-05-18 17:11 ` Christoph Lameter
2007-05-18 17:11 ` Christoph Lameter
2007-05-20 8:39 ` Peter Zijlstra
2007-05-20 8:39 ` Peter Zijlstra
2007-05-21 16:45 ` Christoph Lameter
2007-05-21 16:45 ` Christoph Lameter
2007-05-21 19:33 ` Peter Zijlstra
2007-05-21 19:33 ` Peter Zijlstra
2007-05-21 19:43 ` Christoph Lameter
2007-05-21 19:43 ` Christoph Lameter
2007-05-21 20:08 ` Peter Zijlstra
2007-05-21 20:08 ` Peter Zijlstra
2007-05-21 20:32 ` Christoph Lameter
2007-05-21 20:32 ` Christoph Lameter
2007-05-21 20:54 ` Peter Zijlstra
2007-05-21 20:54 ` Peter Zijlstra
2007-05-21 21:04 ` Christoph Lameter
2007-05-21 21:04 ` 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=1179298771.7173.16.camel@twins \
--to=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=clameter@sgi.com \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mpm@selenic.com \
--cc=penberg@cs.helsinki.fi \
--cc=phillips@google.com \
--cc=tgraf@suug.ch \
/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.