From: Glauber Costa <glommer@parallels.com>
To: Christoph Lameter <cl@linux.com>
Cc: <linux-kernel@vger.kernel.org>, <cgroups@vger.kernel.org>,
<kamezawa.hiroyu@jp.fujitsu.com>, <devel@openvz.org>,
Tejun Heo <tj@kernel.org>, <linux-mm@kvack.org>,
Suleiman Souhlal <suleiman@google.com>,
Frederic Weisbecker <fweisbec@gmail.com>,
Mel Gorman <mgorman@suse.de>,
David Rientjes <rientjes@google.com>,
Pekka Enberg <penberg@cs.helsinki.fi>
Subject: Re: [PATCH v3 09/16] sl[au]b: always get the cache from its page in kfree
Date: Wed, 19 Sep 2012 11:42:54 +0400 [thread overview]
Message-ID: <5059777E.8060906@parallels.com> (raw)
In-Reply-To: <00000139d9fe8595-8905906d-18ed-4d41-afdb-f4c632c2d50a-000000@email.amazonses.com>
On 09/18/2012 07:28 PM, Christoph Lameter wrote:
> On Tue, 18 Sep 2012, Glauber Costa wrote:
>
>> index f2d760c..18de3f6 100644
>> --- a/mm/slab.c
>> +++ b/mm/slab.c
>> @@ -3938,9 +3938,12 @@ EXPORT_SYMBOL(__kmalloc);
>> * Free an object which was previously allocated from this
>> * cache.
>> */
>> -void kmem_cache_free(struct kmem_cache *cachep, void *objp)
>> +void kmem_cache_free(struct kmem_cache *s, void *objp)
>> {
>> unsigned long flags;
>> + struct kmem_cache *cachep = virt_to_cache(objp);
>> +
>> + VM_BUG_ON(!slab_equal_or_parent(cachep, s));
>>
>
> This is an extremely hot path of the kernel and you are adding significant
> processing. Check how the benchmarks are influenced by this change.
> virt_to_cache can be a bit expensive.
>
Would it be enough for you to have a separate code path for
!CONFIG_MEMCG_KMEM?
I don't really see another way to do it, aside from deriving the cache
from the object in our case. I am open to suggestions if you do.
>> diff --git a/mm/slub.c b/mm/slub.c
>> index 4778548..a045dfc 100644
>> --- a/mm/slub.c
>> +++ b/mm/slub.c
>> @@ -2604,7 +2604,9 @@ void kmem_cache_free(struct kmem_cache *s, void *x)
>>
>> page = virt_to_head_page(x);
>>
>> - slab_free(s, page, x, _RET_IP_);
>> + VM_BUG_ON(!slab_equal_or_parent(page->slab, s));
>> +
>> + slab_free(page->slab, page, x, _RET_IP_);
>>
>
> Less of a problem here but you are eroding one advantage that slab has had
> in the past over slub in terms of freeing objects.
>
likewise.
next prev parent reply other threads:[~2012-09-19 7:49 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-18 14:11 [PATCH v3 00/16] slab accounting for memcg Glauber Costa
2012-09-18 14:11 ` [PATCH v3 01/16] slab/slub: struct memcg_params Glauber Costa
2012-09-18 14:11 ` [PATCH v3 02/16] slub: use free_page instead of put_page for freeing kmalloc allocation Glauber Costa
2012-09-18 14:11 ` [PATCH v3 03/16] slab: Ignore the cflgs bit in cache creation Glauber Costa
2012-09-18 15:20 ` Christoph Lameter
2012-09-21 17:33 ` Tejun Heo
2012-09-18 14:11 ` [PATCH v3 04/16] provide a common place for initcall processing in kmem_cache Glauber Costa
2012-09-18 15:22 ` Christoph Lameter
2012-09-18 14:11 ` [PATCH v3 05/16] consider a memcg parameter in kmem_create_cache Glauber Costa
2012-09-21 18:14 ` Tejun Heo
2012-09-24 8:12 ` Glauber Costa
2012-09-24 12:41 ` Christoph
2012-09-24 12:41 ` Glauber Costa
2012-09-24 13:42 ` Christoph Lameter
2012-09-24 13:44 ` Glauber Costa
2012-09-24 13:56 ` Christoph Lameter
2012-09-24 13:57 ` Glauber Costa
2012-09-24 15:38 ` Pekka Enberg
2012-09-24 15:36 ` Glauber Costa
2012-09-24 17:38 ` Christoph Lameter
2012-09-24 17:39 ` Christoph Lameter
2012-10-02 14:46 ` Michal Hocko
2012-09-18 14:12 ` [PATCH v3 06/16] memcg: infrastructure to match an allocation to the right cache Glauber Costa
2012-09-21 18:32 ` Tejun Heo
2012-09-24 8:46 ` Glauber Costa
2012-09-24 17:56 ` Tejun Heo
2012-09-25 13:57 ` Glauber Costa
2012-09-25 16:28 ` Christoph Lameter
2012-09-21 20:52 ` Tejun Heo
2012-09-24 8:17 ` Glauber Costa
2012-09-24 17:58 ` Tejun Heo
2012-09-18 14:12 ` [PATCH v3 07/16] memcg: skip memcg kmem allocations in specified code regions Glauber Costa
2012-09-21 19:59 ` Tejun Heo
2012-09-24 9:09 ` Glauber Costa
2012-09-24 17:47 ` Tejun Heo
2012-09-18 14:12 ` [PATCH v3 08/16] slab: allow enable_cpu_cache to use preset values for its tunables Glauber Costa
2012-09-18 15:25 ` Christoph Lameter
2012-09-19 7:44 ` Glauber Costa
2012-09-21 9:29 ` Pekka Enberg
2012-09-18 14:12 ` [PATCH v3 09/16] sl[au]b: always get the cache from its page in kfree Glauber Costa
2012-09-18 15:28 ` Christoph Lameter
2012-09-19 7:42 ` Glauber Costa [this message]
2012-09-19 14:14 ` Christoph Lameter
2012-09-21 9:33 ` Pekka Enberg
2012-09-21 9:30 ` Glauber Costa
2012-09-21 9:41 ` Pekka Enberg
2012-09-21 20:07 ` Tejun Heo
2012-09-21 20:14 ` Pekka Enberg
2012-09-21 20:16 ` Tejun Heo
2012-09-18 14:12 ` [PATCH v3 10/16] sl[au]b: Allocate objects from memcg cache Glauber Costa
2012-09-18 14:12 ` [PATCH v3 11/16] memcg: destroy memcg caches Glauber Costa
2012-09-21 20:22 ` Tejun Heo
2012-09-18 14:12 ` [PATCH v3 12/16] memcg/sl[au]b Track all the memcg children of a kmem_cache Glauber Costa
2012-09-21 20:31 ` Tejun Heo
2012-09-18 14:12 ` [PATCH v3 13/16] slab: slab-specific propagation changes Glauber Costa
2012-09-18 17:00 ` Christoph Lameter
2012-09-19 7:41 ` Glauber Costa
2012-09-18 14:12 ` [PATCH v3 14/16] slub: slub-specific " Glauber Costa
2012-09-18 14:12 ` [PATCH v3 15/16] memcg/sl[au]b: shrink dead caches Glauber Costa
2012-09-18 17:02 ` Christoph Lameter
2012-09-19 7:40 ` Glauber Costa
2012-09-21 4:48 ` JoonSoo Kim
2012-09-21 8:40 ` Glauber Costa
2012-09-21 9:28 ` JoonSoo Kim
2012-09-21 9:31 ` Glauber Costa
2012-09-21 20:40 ` Tejun Heo
2012-09-24 8:25 ` Glauber Costa
2012-09-24 17:43 ` Tejun Heo
2012-09-18 14:12 ` [PATCH v3 16/16] Add documentation about the kmem controller Glauber Costa
2012-09-21 9:40 ` [PATCH v3 00/16] slab accounting for memcg Pekka Enberg
2012-09-21 9:43 ` Glauber Costa
2012-09-21 20:46 ` Tejun Heo
2012-09-21 20:47 ` Tejun Heo
2012-09-24 8:15 ` Glauber Costa
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=5059777E.8060906@parallels.com \
--to=glommer@parallels.com \
--cc=cgroups@vger.kernel.org \
--cc=cl@linux.com \
--cc=devel@openvz.org \
--cc=fweisbec@gmail.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=penberg@cs.helsinki.fi \
--cc=rientjes@google.com \
--cc=suleiman@google.com \
--cc=tj@kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).