linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: David Rientjes <rientjes@google.com>
To: Shaohua Li <shaohua.li@intel.com>
Cc: "Shi, Alex" <alex.shi@intel.com>,
	Christoph Lameter <cl@linux.com>,
	"penberg@kernel.org" <penberg@kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Andi Kleen <ak@linux.intel.com>
Subject: Re: [PATCH 1/3] slub: set a criteria for slub node partial adding
Date: Tue, 13 Dec 2011 17:29:22 -0800 (PST)	[thread overview]
Message-ID: <alpine.DEB.2.00.1112131726140.8593@chino.kir.corp.google.com> (raw)
In-Reply-To: <1323657793.22361.383.camel@sli10-conroe>

On Mon, 12 Dec 2011, Shaohua Li wrote:

> With the per-cpu partial list, I didn't see any workload which is still
> suffering from the list lock, so I suppose both the trashing approach
> and pick 25% used slab approach don't help.

This doesn't necessarily have anything to do with contention on list_lock, 
it has to do with the fact that ~99% of allocations come from the slowpath 
since the cpu slab only has one free object when it is activated, that's 
what the statistics indicated for kmalloc-256 and kmalloc-2k.  That's what 
I called "slab thrashing": the continual deactivation of the cpu slab and 
picking from the partial list that would only have one or two free objects 
causing the vast majority of allocations to require the slowpath.

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2011-12-14  1:29 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-02  8:23 Alex Shi
2011-12-02  8:23 ` [PATCH 2/3] slub: remove unnecessary statistics, deactivate_to_head/tail Alex Shi
2011-12-02  8:23   ` [PATCH 3/3] slub: fill per cpu partial only when free objects larger than one quarter Alex Shi
2011-12-02 14:44   ` [PATCH 2/3] slub: remove unnecessary statistics, deactivate_to_head/tail Christoph Lameter
2011-12-06 21:08   ` David Rientjes
2011-12-02 11:36 ` [PATCH 1/3] slub: set a criteria for slub node partial adding Eric Dumazet
2011-12-02 20:02   ` Christoph Lameter
2011-12-05  2:21     ` Shaohua Li
2011-12-05 10:01     ` Alex,Shi
2011-12-05  3:28   ` Alex,Shi
2011-12-02 14:43 ` Christoph Lameter
2011-12-05  9:22   ` Alex,Shi
2011-12-06 21:06     ` David Rientjes
2011-12-07  5:11       ` Shaohua Li
2011-12-07  7:28         ` David Rientjes
2011-12-12  2:43           ` Shaohua Li
2011-12-12  4:14             ` Alex,Shi
2011-12-12  4:35               ` Shaohua Li
2011-12-12  4:25                 ` Alex,Shi
2011-12-12  4:48                   ` Shaohua Li
2011-12-12  6:17                     ` Alex,Shi
2011-12-12  6:09             ` Eric Dumazet
2011-12-14  1:29             ` David Rientjes [this message]
2011-12-14  2:43               ` Shaohua Li
2011-12-14  2:38                 ` David Rientjes
2011-12-09  8:30   ` Alex,Shi
2011-12-09 10:10     ` David Rientjes
2011-12-09 13:40       ` Shi, Alex
2011-12-14  1:38         ` David Rientjes
2011-12-14  2:36           ` David Rientjes
2011-12-14  6:06             ` Alex,Shi
2011-12-14  6:44               ` Eric Dumazet
2011-12-14  6:47                 ` Pekka Enberg
2011-12-14 14:53                   ` Christoph Lameter
2011-12-14  6:56                 ` Alex,Shi
2011-12-14 14:59                   ` Christoph Lameter
2011-12-14 17:33                     ` Eric Dumazet
2011-12-14 18:26                       ` Christoph Lameter
2011-12-13 13:01       ` Shi, Alex

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=alpine.DEB.2.00.1112131726140.8593@chino.kir.corp.google.com \
    --to=rientjes@google.com \
    --cc=ak@linux.intel.com \
    --cc=alex.shi@intel.com \
    --cc=cl@linux.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=penberg@kernel.org \
    --cc=shaohua.li@intel.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