From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: David Rientjes <rientjes@google.com>,
Matt Mackall <mpm@selenic.com>,
Pekka Enberg <penberg@cs.helsinki.fi>,
linux-mm@kvack.org, Christoph Lameter <cl@linux-foundation.org>,
LKML <linux-kernel@vger.kernel.org>,
Nick Piggin <npiggin@suse.de>
Subject: Re: lockdep complaints in slab allocator
Date: Tue, 24 Nov 2009 13:48:14 -0800 [thread overview]
Message-ID: <20091124214814.GK6831@linux.vnet.ibm.com> (raw)
In-Reply-To: <1259097150.4531.1822.camel@laptop>
On Tue, Nov 24, 2009 at 10:12:30PM +0100, Peter Zijlstra wrote:
> On Tue, 2009-11-24 at 13:03 -0800, David Rientjes wrote:
> > On Tue, 24 Nov 2009, Peter Zijlstra wrote:
> >
> > > Merge SLQB and rm mm/sl[ua]b.c include/linux/sl[ua]b.h for .33-rc1
> > >
> >
> > slqb still has a 5-10% performance regression compared to slab for
> > benchmarks such as netperf TCP_RR on machines with high cpu counts,
> > forcing that type of regression isn't acceptable.
>
> Having _4_ slab allocators is equally unacceptable.
I completely agree. We need at least ten. ;-)
Thanx, Paul
WARNING: multiple messages have this Message-ID (diff)
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: David Rientjes <rientjes@google.com>,
Matt Mackall <mpm@selenic.com>,
Pekka Enberg <penberg@cs.helsinki.fi>,
linux-mm@kvack.org, Christoph Lameter <cl@linux-foundation.org>,
LKML <linux-kernel@vger.kernel.org>,
Nick Piggin <npiggin@suse.de>
Subject: Re: lockdep complaints in slab allocator
Date: Tue, 24 Nov 2009 13:48:14 -0800 [thread overview]
Message-ID: <20091124214814.GK6831@linux.vnet.ibm.com> (raw)
In-Reply-To: <1259097150.4531.1822.camel@laptop>
On Tue, Nov 24, 2009 at 10:12:30PM +0100, Peter Zijlstra wrote:
> On Tue, 2009-11-24 at 13:03 -0800, David Rientjes wrote:
> > On Tue, 24 Nov 2009, Peter Zijlstra wrote:
> >
> > > Merge SLQB and rm mm/sl[ua]b.c include/linux/sl[ua]b.h for .33-rc1
> > >
> >
> > slqb still has a 5-10% performance regression compared to slab for
> > benchmarks such as netperf TCP_RR on machines with high cpu counts,
> > forcing that type of regression isn't acceptable.
>
> Having _4_ slab allocators is equally unacceptable.
I completely agree. We need at least ten. ;-)
Thanx, Paul
--
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:[~2009-11-24 21:48 UTC|newest]
Thread overview: 121+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-11-18 18:12 lockdep complaints in slab allocator Paul E. McKenney
2009-11-20 6:49 ` Pekka Enberg
2009-11-20 6:49 ` Pekka Enberg
2009-11-20 9:25 ` Peter Zijlstra
2009-11-20 9:25 ` Peter Zijlstra
2009-11-20 10:38 ` Pekka Enberg
2009-11-20 10:38 ` Pekka Enberg
2009-11-20 10:52 ` Peter Zijlstra
2009-11-20 10:52 ` Peter Zijlstra
2009-11-20 11:05 ` Pekka Enberg
2009-11-20 11:05 ` Pekka Enberg
2009-11-20 14:48 ` Paul E. McKenney
2009-11-20 14:48 ` Paul E. McKenney
2009-11-20 15:17 ` Peter Zijlstra
2009-11-20 15:17 ` Peter Zijlstra
2009-11-20 16:25 ` Paul E. McKenney
2009-11-20 16:25 ` Paul E. McKenney
2009-11-20 15:09 ` Peter Zijlstra
2009-11-20 15:09 ` Peter Zijlstra
2009-11-23 19:00 ` Pekka Enberg
2009-11-23 19:00 ` Pekka Enberg
2009-11-23 19:10 ` Matt Mackall
2009-11-23 19:10 ` Matt Mackall
2009-11-23 19:13 ` Pekka Enberg
2009-11-23 19:13 ` Pekka Enberg
2009-11-24 16:33 ` Peter Zijlstra
2009-11-24 16:33 ` Peter Zijlstra
2009-11-24 17:00 ` Paul E. McKenney
2009-11-24 17:00 ` Paul E. McKenney
2009-11-24 17:12 ` Matt Mackall
2009-11-24 17:12 ` Matt Mackall
2009-11-24 17:58 ` Paul E. McKenney
2009-11-24 17:58 ` Paul E. McKenney
2009-11-24 18:14 ` Peter Zijlstra
2009-11-24 18:14 ` Peter Zijlstra
2009-11-24 18:25 ` Paul E. McKenney
2009-11-24 18:25 ` Paul E. McKenney
2009-11-24 18:31 ` Peter Zijlstra
2009-11-24 18:31 ` Peter Zijlstra
2009-11-24 18:53 ` Christoph Lameter
2009-11-24 18:53 ` Christoph Lameter
2009-11-24 18:54 ` Paul E. McKenney
2009-11-24 18:54 ` Paul E. McKenney
2009-11-24 19:23 ` Matt Mackall
2009-11-24 19:23 ` Matt Mackall
2009-11-24 19:50 ` Paul E. McKenney
2009-11-24 19:50 ` Paul E. McKenney
2009-11-24 20:46 ` Peter Zijlstra
2009-11-24 20:46 ` Peter Zijlstra
2009-11-24 20:53 ` Matt Mackall
2009-11-24 20:53 ` Matt Mackall
2009-11-24 21:01 ` Peter Zijlstra
2009-11-24 21:01 ` Peter Zijlstra
2009-11-24 21:03 ` David Rientjes
2009-11-24 21:03 ` David Rientjes
2009-11-24 21:12 ` Peter Zijlstra
2009-11-24 21:12 ` Peter Zijlstra
2009-11-24 21:19 ` Pekka Enberg
2009-11-24 21:19 ` Pekka Enberg
2009-11-24 21:22 ` David Rientjes
2009-11-24 21:22 ` David Rientjes
2009-11-24 21:35 ` Peter Zijlstra
2009-11-24 21:35 ` Peter Zijlstra
2009-11-24 21:46 ` David Rientjes
2009-11-24 21:46 ` David Rientjes
2009-11-24 22:23 ` Paul E. McKenney
2009-11-24 22:23 ` Paul E. McKenney
2009-11-25 7:12 ` Pekka Enberg
2009-11-25 7:12 ` Pekka Enberg
2009-11-25 7:25 ` Pekka Enberg
2009-11-25 7:25 ` Pekka Enberg
2009-11-27 17:22 ` Christoph Lameter
2009-11-27 17:22 ` Christoph Lameter
2009-11-24 21:48 ` Paul E. McKenney [this message]
2009-11-24 21:48 ` Paul E. McKenney
2009-11-24 21:16 ` Pekka Enberg
2009-11-24 21:16 ` Pekka Enberg
2009-11-24 21:07 ` Pekka Enberg
2009-11-24 21:07 ` Pekka Enberg
2009-11-24 22:55 ` Matt Mackall
2009-11-24 22:55 ` Matt Mackall
2009-11-25 21:59 ` David Rientjes
2009-11-25 21:59 ` David Rientjes
2009-11-25 23:06 ` Matt Mackall
2009-11-25 23:06 ` Matt Mackall
2009-11-27 17:28 ` Christoph Lameter
2009-11-27 17:28 ` Christoph Lameter
2009-11-30 23:14 ` David Rientjes
2009-11-30 23:14 ` David Rientjes
2009-12-01 0:21 ` Matt Mackall
2009-12-01 0:21 ` Matt Mackall
2009-12-01 22:41 ` David Rientjes
2009-12-01 22:41 ` David Rientjes
2009-12-01 16:47 ` Christoph Lameter
2009-12-01 16:47 ` Christoph Lameter
2009-11-27 17:26 ` Christoph Lameter
2009-11-27 17:26 ` Christoph Lameter
2009-11-23 19:30 ` Christoph Lameter
2009-11-23 19:30 ` Christoph Lameter
2009-11-23 19:43 ` Paul E. McKenney
2009-11-23 19:43 ` Paul E. McKenney
2009-11-23 19:50 ` Pekka Enberg
2009-11-23 19:50 ` Pekka Enberg
2009-11-23 20:01 ` Pekka Enberg
2009-11-23 20:01 ` Pekka Enberg
2009-11-23 20:57 ` Paul E. McKenney
2009-11-23 20:57 ` Paul E. McKenney
2009-11-23 21:01 ` Matt Mackall
2009-11-23 21:01 ` Matt Mackall
2009-11-24 16:23 ` Paul E. McKenney
2009-11-24 16:23 ` Paul E. McKenney
2009-11-24 20:59 ` Pekka Enberg
2009-11-24 20:59 ` Pekka Enberg
2009-11-24 21:26 ` Peter Zijlstra
2009-11-24 21:26 ` Peter Zijlstra
2009-11-25 10:42 ` Pekka Enberg
2009-11-25 10:42 ` Pekka Enberg
2009-11-24 21:47 ` Paul E. McKenney
2009-11-24 21:47 ` Paul E. McKenney
2009-11-30 16:18 ` Paul E. McKenney
2009-11-30 16:18 ` Paul E. McKenney
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=20091124214814.GK6831@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=cl@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mpm@selenic.com \
--cc=npiggin@suse.de \
--cc=penberg@cs.helsinki.fi \
--cc=peterz@infradead.org \
--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 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.