From: Pekka Enberg <penberg@cs.helsinki.fi>
To: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>,
Andi Kleen <andi@firstfloor.org>,
tj@kernel.org, Metathronius Galabant <m.galabant@googlemail.com>,
Matt Mackall <mpm@selenic.com>,
Eric Dumazet <eric.dumazet@gmail.com>,
Adrian Drzewiecki <z@drze.net>,
linux-kernel@vger.kernel.org
Subject: Re: [slub p2 0/4] SLUB: [RFC] Per cpu partial lists V2
Date: Thu, 07 Jul 2011 22:05:03 +0300 [thread overview]
Message-ID: <1310065503.21902.61.camel@jaguar> (raw)
In-Reply-To: <20110620153244.214038140@linux.com>
On Mon, 2011-06-20 at 10:32 -0500, Christoph Lameter wrote:
> The following patchset applied on top of the lockless patchset V7. It
> introduces per cpu partial lists which allow a performance increase of
> around ~15 during contention for the nodelock (can be tested using
> hackbench).
>
> These lists help to avoid per nodelocking overhead. Allocator latency
> could be further reduced by making these operations work without
> disabling interrupts (like the fastpath and the free slowpath) as well as
> implementing better ways of handling ther cpu array with partial pages.
>
> I am still not satisfied with the cleanliness of the code after these
> changes. Some review with suggestions as to how to restructure the
> code given these changes in operations would be appreciated.
>
> It is interesting to note that BSD has gone to a scheme with partial
> pages only per cpu (source: Adrian). Transfer of cpu ownerships is
> done using IPIs. Probably too much overhead for our taste. The use
> of a few per cpu partial pages looks to be beneficial though.
>
> Note that there is no performance gain when there is no contention.
>
> Performance:
>
> Before After
> ./hackbench 100 process 200000
> Time: 2299.072 1742.454
> ./hackbench 100 process 20000
> Time: 224.654 182.393
> ./hackbench 100 process 20000
> Time: 227.126 182.780
> ./hackbench 100 process 20000
> Time: 219.608 182.899
> ./hackbench 10 process 20000
> Time: 21.769 18.756
> ./hackbench 10 process 20000
> Time: 21.657 18.938
> ./hackbench 10 process 20000
> Time: 23.193 19.537
> ./hackbench 1 process 20000
> Time: 2.337 2.263
> ./hackbench 1 process 20000
> Time: 2.223 2.271
> ./hackbench 1 process 20000
> Time: 2.269 2.301
Impressive numbers! David, comments on the series?
Pekka
prev parent reply other threads:[~2011-07-07 19:05 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-20 15:32 Christoph Lameter
2011-06-20 15:32 ` [slub p2 1/4] slub: Prepare inuse field in new_slab() Christoph Lameter
2011-06-20 15:32 ` [slub p2 2/4] slub: pass kmem_cache_cpu pointer to get_partial() Christoph Lameter
2011-06-20 15:32 ` [slub p2 3/4] slub: return object pointer from get_partial() / new_slab() Christoph Lameter
2011-06-20 15:32 ` [slub p2 4/4] slub: [RFC] per cpu cache for partial pages Christoph Lameter
2011-06-20 19:42 ` [slub p2 0/4] SLUB: [RFC] Per cpu partial lists V2 Andi Kleen
2011-06-20 20:01 ` Christoph Lameter
2011-07-07 19:05 ` Pekka Enberg [this message]
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=1310065503.21902.61.camel@jaguar \
--to=penberg@cs.helsinki.fi \
--cc=andi@firstfloor.org \
--cc=cl@linux.com \
--cc=eric.dumazet@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=m.galabant@googlemail.com \
--cc=mpm@selenic.com \
--cc=rientjes@google.com \
--cc=tj@kernel.org \
--cc=z@drze.net \
/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
Powered by JetHome