From: Eric Dumazet <eric.dumazet@gmail.com>
To: JoonSoo Kim <js1304@gmail.com>
Cc: Pekka Enberg <penberg@kernel.org>,
Christoph Lameter <cl@linux-foundation.org>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
David Rientjes <rientjes@google.com>
Subject: Re: [PATCH 1/3 v2] slub: prefetch next freelist pointer in __slab_alloc()
Date: Wed, 04 Jul 2012 18:15:17 +0200 [thread overview]
Message-ID: <1341418517.2583.2252.camel@edumazet-glaptop> (raw)
In-Reply-To: <CAAmzW4P8itKqMLLUqAAtT7GakKecCixd0PV8y0LgFOL+=g_tZQ@mail.gmail.com>
On Thu, 2012-07-05 at 00:48 +0900, JoonSoo Kim wrote:
> 2012/7/5 Eric Dumazet <eric.dumazet@gmail.com>:
> > Its the slow path. I am not convinced its useful on real workloads (not
> > a benchmark)
> >
> > I mean, if a workload hits badly slow path, some more important work
> > should be done to avoid this at a higher level.
> >
>
> In hackbench test, fast path allocation is about to 93%.
> Is it insufficient?
7% is insufficient I am afraid.
One prefetch() in the fast path serves 93% of the allocations,
so added icache pressure is ok.
One prefetch() in slow path serves 7% of the allocations, do you see the
difference ?
Adding a prefetch() is usually a win when a benchmark uses the path one
million times per second.
But adding prefetches also increases kernel size and it hurts globally.
(Latency of the kernel depends on its size, when cpu caches are cold)
--
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:[~2012-07-04 16:15 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <yes>
2012-06-08 17:23 ` [PATCH 1/4] slub: change declare of get_slab() to inline at all times Joonsoo Kim
2012-06-08 17:23 ` [PATCH 2/4] slub: use __cmpxchg_double_slab() at interrupt disabled place Joonsoo Kim
2012-06-08 17:23 ` [PATCH 3/4] slub: refactoring unfreeze_partials() Joonsoo Kim
2012-06-20 7:19 ` Pekka Enberg
2012-06-08 17:23 ` [PATCH 4/4] slub: deactivate freelist of kmem_cache_cpu all at once in deactivate_slab() Joonsoo Kim
2012-06-08 19:04 ` Christoph Lameter
2012-06-10 10:27 ` JoonSoo Kim
2012-06-22 18:34 ` JoonSoo Kim
2012-06-08 19:02 ` [PATCH 1/4] slub: change declare of get_slab() to inline at all times Christoph Lameter
2012-06-09 15:57 ` JoonSoo Kim
2012-06-11 15:04 ` Christoph Lameter
2012-06-22 18:22 ` [PATCH 1/3] slub: prefetch next freelist pointer in __slab_alloc() Joonsoo Kim
2012-06-22 18:22 ` [PATCH 2/3] slub: reduce failure of this_cpu_cmpxchg in put_cpu_partial() after unfreezing Joonsoo Kim
2012-07-04 13:05 ` Pekka Enberg
2012-07-05 14:20 ` Christoph Lameter
2012-08-16 7:06 ` Pekka Enberg
2012-06-22 18:22 ` [PATCH 3/3] slub: release a lock if freeing object with a lock is failed in __slab_free() Joonsoo Kim
2012-07-04 13:10 ` Pekka Enberg
2012-07-04 14:48 ` JoonSoo Kim
2012-07-05 14:26 ` Christoph Lameter
2012-07-06 14:19 ` JoonSoo Kim
2012-07-06 14:34 ` Christoph Lameter
2012-07-06 14:59 ` JoonSoo Kim
2012-07-06 15:10 ` Christoph Lameter
2012-07-08 16:19 ` JoonSoo Kim
2012-06-22 18:45 ` [PATCH 1/3 v2] slub: prefetch next freelist pointer in __slab_alloc() Joonsoo Kim
2012-07-04 12:58 ` JoonSoo Kim
2012-07-04 13:00 ` Pekka Enberg
2012-07-04 14:30 ` JoonSoo Kim
2012-07-04 15:08 ` Pekka Enberg
2012-07-04 15:26 ` Eric Dumazet
2012-07-04 15:48 ` JoonSoo Kim
2012-07-04 16:15 ` Eric Dumazet [this message]
2012-07-04 16:24 ` JoonSoo Kim
2012-07-04 15:45 ` JoonSoo Kim
2012-07-04 15:59 ` Pekka Enberg
2012-07-04 16:04 ` JoonSoo Kim
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=1341418517.2583.2252.camel@edumazet-glaptop \
--to=eric.dumazet@gmail.com \
--cc=cl@linux-foundation.org \
--cc=js1304@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=penberg@kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox