From: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Glauber Costa <glommer@parallels.com>
Cc: Michal Hocko <mhocko@suse.cz>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Andrew Morton <akpm@linux-foundation.org>,
Mel Gorman <mgorman@suse.de>,
Suleiman Souhlal <suleiman@google.com>, Tejun Heo <tj@kernel.org>,
cgroups@vger.kernel.org, Johannes Weiner <hannes@cmpxchg.org>,
Greg Thelen <gthelen@google.com>,
devel@openvz.org, Frederic Weisbecker <fweisbec@gmail.com>,
Christoph Lameter <cl@linux.com>,
Pekka Enberg <penberg@cs.helsinki.fi>
Subject: Re: [PATCH v4 10/14] memcg: use static branches when code not in use
Date: Tue, 16 Oct 2012 17:48:29 +0900 [thread overview]
Message-ID: <507D1F5D.8080709@jp.fujitsu.com> (raw)
In-Reply-To: <5077CB29.2000505@parallels.com>
(2012/10/12 16:47), Glauber Costa wrote:
> On 10/11/2012 05:40 PM, Michal Hocko wrote:
>> On Mon 08-10-12 14:06:16, Glauber Costa wrote:
>>> We can use static branches to patch the code in or out when not used.
>>>
>>> Because the _ACTIVE bit on kmem_accounted is only set after the
>>> increment is done, we guarantee that the root memcg will always be
>>> selected for kmem charges until all call sites are patched (see
>>> memcg_kmem_enabled). This guarantees that no mischarges are applied.
>>>
>>> static branch decrement happens when the last reference count from the
>>> kmem accounting in memcg dies. This will only happen when the charges
>>> drop down to 0.
>>>
>>> When that happen, we need to disable the static branch only on those
>>> memcgs that enabled it. To achieve this, we would be forced to
>>> complicate the code by keeping track of which memcgs were the ones
>>> that actually enabled limits, and which ones got it from its parents.
>>>
>>> It is a lot simpler just to do static_key_slow_inc() on every child
>>> that is accounted.
>>>
>>> [ v4: adapted this patch to the changes in kmem_accounted ]
>>>
>>> Signed-off-by: Glauber Costa <glommer@parallels.com>
>>> CC: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>>> CC: Christoph Lameter <cl@linux.com>
>>> CC: Pekka Enberg <penberg@cs.helsinki.fi>
>>> CC: Michal Hocko <mhocko@suse.cz>
>>> CC: Johannes Weiner <hannes@cmpxchg.org>
>>> CC: Suleiman Souhlal <suleiman@google.com>
>>
>> Looks reasonable to me
>> Acked-by: Michal Hocko <mhocko@suse.cz>
>>
>> Just a little nit.
>>
>> [...]
>>
>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>>> index 634c7b5..724a08b 100644
>>> --- a/mm/memcontrol.c
>>> +++ b/mm/memcontrol.c
>>> @@ -344,11 +344,15 @@ struct mem_cgroup {
>>> /* internal only representation about the status of kmem accounting. */
>>> enum {
>>> KMEM_ACCOUNTED_ACTIVE = 0, /* accounted by this cgroup itself */
>>> + KMEM_ACCOUNTED_ACTIVATED, /* static key enabled. */
>>> KMEM_ACCOUNTED_DEAD, /* dead memcg, pending kmem charges */
>>> };
>>>
>>> -/* first bit */
>>> -#define KMEM_ACCOUNTED_MASK 0x1
>>> +/*
>>> + * first two bits. We account when limit is on, but only after
>>> + * call sites are patched
>>> + */
>>> +#define KMEM_ACCOUNTED_MASK 0x3
>>
>> The names are long but why not use KMEM_ACCOUNTED_ACTIVE*
>> #define KMEM_ACCOUNTED_MASK 1<<KMEM_ACCOUNTED_ACTIVE | 1<<KMEM_ACCOUNTED_ACTIVATED
>>
> Because the names are long! =)
>
please use "long" macros ;) it's not bad.
Anyway,
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
--
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>
WARNING: multiple messages have this Message-ID (diff)
From: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Glauber Costa <glommer@parallels.com>
Cc: Michal Hocko <mhocko@suse.cz>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Andrew Morton <akpm@linux-foundation.org>,
Mel Gorman <mgorman@suse.de>,
Suleiman Souhlal <suleiman@google.com>, Tejun Heo <tj@kernel.org>,
cgroups@vger.kernel.org, Johannes Weiner <hannes@cmpxchg.org>,
Greg Thelen <gthelen@google.com>,
devel@openvz.org, Frederic Weisbecker <fweisbec@gmail.com>,
Christoph Lameter <cl@linux.com>,
Pekka Enberg <penberg@cs.helsinki.fi>
Subject: Re: [PATCH v4 10/14] memcg: use static branches when code not in use
Date: Tue, 16 Oct 2012 17:48:29 +0900 [thread overview]
Message-ID: <507D1F5D.8080709@jp.fujitsu.com> (raw)
In-Reply-To: <5077CB29.2000505@parallels.com>
(2012/10/12 16:47), Glauber Costa wrote:
> On 10/11/2012 05:40 PM, Michal Hocko wrote:
>> On Mon 08-10-12 14:06:16, Glauber Costa wrote:
>>> We can use static branches to patch the code in or out when not used.
>>>
>>> Because the _ACTIVE bit on kmem_accounted is only set after the
>>> increment is done, we guarantee that the root memcg will always be
>>> selected for kmem charges until all call sites are patched (see
>>> memcg_kmem_enabled). This guarantees that no mischarges are applied.
>>>
>>> static branch decrement happens when the last reference count from the
>>> kmem accounting in memcg dies. This will only happen when the charges
>>> drop down to 0.
>>>
>>> When that happen, we need to disable the static branch only on those
>>> memcgs that enabled it. To achieve this, we would be forced to
>>> complicate the code by keeping track of which memcgs were the ones
>>> that actually enabled limits, and which ones got it from its parents.
>>>
>>> It is a lot simpler just to do static_key_slow_inc() on every child
>>> that is accounted.
>>>
>>> [ v4: adapted this patch to the changes in kmem_accounted ]
>>>
>>> Signed-off-by: Glauber Costa <glommer@parallels.com>
>>> CC: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>>> CC: Christoph Lameter <cl@linux.com>
>>> CC: Pekka Enberg <penberg@cs.helsinki.fi>
>>> CC: Michal Hocko <mhocko@suse.cz>
>>> CC: Johannes Weiner <hannes@cmpxchg.org>
>>> CC: Suleiman Souhlal <suleiman@google.com>
>>
>> Looks reasonable to me
>> Acked-by: Michal Hocko <mhocko@suse.cz>
>>
>> Just a little nit.
>>
>> [...]
>>
>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>>> index 634c7b5..724a08b 100644
>>> --- a/mm/memcontrol.c
>>> +++ b/mm/memcontrol.c
>>> @@ -344,11 +344,15 @@ struct mem_cgroup {
>>> /* internal only representation about the status of kmem accounting. */
>>> enum {
>>> KMEM_ACCOUNTED_ACTIVE = 0, /* accounted by this cgroup itself */
>>> + KMEM_ACCOUNTED_ACTIVATED, /* static key enabled. */
>>> KMEM_ACCOUNTED_DEAD, /* dead memcg, pending kmem charges */
>>> };
>>>
>>> -/* first bit */
>>> -#define KMEM_ACCOUNTED_MASK 0x1
>>> +/*
>>> + * first two bits. We account when limit is on, but only after
>>> + * call sites are patched
>>> + */
>>> +#define KMEM_ACCOUNTED_MASK 0x3
>>
>> The names are long but why not use KMEM_ACCOUNTED_ACTIVE*
>> #define KMEM_ACCOUNTED_MASK 1<<KMEM_ACCOUNTED_ACTIVE | 1<<KMEM_ACCOUNTED_ACTIVATED
>>
> Because the names are long! =)
>
please use "long" macros ;) it's not bad.
Anyway,
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
next prev parent reply other threads:[~2012-10-16 8:48 UTC|newest]
Thread overview: 136+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-08 10:06 [PATCH v4 00/14] kmem controller for memcg Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-1-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-08 10:06 ` [PATCH v4 01/14] memcg: Make it possible to use the stock for more than one page Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` [PATCH v4 02/14] memcg: Reclaim when more than one page needed Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-3-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-16 3:22 ` Kamezawa Hiroyuki
2012-10-16 3:22 ` Kamezawa Hiroyuki
2012-10-16 3:22 ` Kamezawa Hiroyuki
2012-10-08 10:06 ` [PATCH v4 03/14] memcg: change defines to an enum Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` [PATCH v4 04/14] kmem accounting basic infrastructure Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-5-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-11 10:11 ` Michal Hocko
2012-10-11 10:11 ` Michal Hocko
2012-10-11 10:11 ` Michal Hocko
2012-10-11 12:53 ` Michal Hocko
2012-10-11 12:53 ` Michal Hocko
2012-10-11 13:38 ` Michal Hocko
2012-10-11 13:38 ` Michal Hocko
[not found] ` <20121011101119.GB29295-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-12 7:36 ` Glauber Costa
2012-10-12 7:36 ` Glauber Costa
2012-10-12 7:36 ` Glauber Costa
[not found] ` <5077C886.2030609-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-12 8:27 ` Michal Hocko
2012-10-12 8:27 ` Michal Hocko
2012-10-12 8:27 ` Michal Hocko
2012-10-08 10:06 ` [PATCH v4 05/14] Add a __GFP_KMEMCG flag Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-6-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-09 15:04 ` Michal Hocko
2012-10-09 15:04 ` Michal Hocko
2012-10-09 15:04 ` Michal Hocko
2012-10-08 10:06 ` [PATCH v4 06/14] memcg: kmem controller infrastructure Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-11 12:42 ` Michal Hocko
2012-10-11 12:42 ` Michal Hocko
2012-10-11 12:56 ` Michal Hocko
2012-10-11 12:56 ` Michal Hocko
[not found] ` <20121011124212.GC29295-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-12 7:45 ` Glauber Costa
2012-10-12 7:45 ` Glauber Costa
2012-10-12 7:45 ` Glauber Costa
2012-10-12 8:39 ` Michal Hocko
2012-10-12 8:39 ` Michal Hocko
[not found] ` <20121012083944.GD10110-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-12 8:44 ` Glauber Costa
2012-10-12 8:44 ` Glauber Costa
2012-10-12 8:44 ` Glauber Costa
[not found] ` <5077D889.2040100-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-12 8:57 ` Michal Hocko
2012-10-12 8:57 ` Michal Hocko
2012-10-12 8:57 ` Michal Hocko
[not found] ` <20121012085740.GG10110-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-12 9:13 ` Glauber Costa
2012-10-12 9:13 ` Glauber Costa
2012-10-12 9:13 ` Glauber Costa
[not found] ` <5077DF20.7020200-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-12 9:47 ` Michal Hocko
2012-10-12 9:47 ` Michal Hocko
2012-10-12 9:47 ` Michal Hocko
2012-10-16 8:00 ` Kamezawa Hiroyuki
2012-10-16 8:00 ` Kamezawa Hiroyuki
2012-10-16 8:00 ` Kamezawa Hiroyuki
2012-10-08 10:06 ` [PATCH v4 07/14] mm: Allocate kernel pages to the right memcg Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` [PATCH v4 08/14] res_counter: return amount of charges after res_counter_uncharge Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-09 15:08 ` Michal Hocko
2012-10-09 15:08 ` Michal Hocko
2012-10-09 15:14 ` Glauber Costa
2012-10-09 15:14 ` Glauber Costa
2012-10-09 15:35 ` Michal Hocko
2012-10-09 15:35 ` Michal Hocko
[not found] ` <20121009153506.GD7655-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-10 9:03 ` Glauber Costa
2012-10-10 9:03 ` Glauber Costa
2012-10-10 9:03 ` Glauber Costa
[not found] ` <507539EB.90006-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-10 11:24 ` Michal Hocko
2012-10-10 11:24 ` Michal Hocko
2012-10-10 11:24 ` Michal Hocko
[not found] ` <1349690780-15988-9-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-10 11:25 ` Michal Hocko
2012-10-10 11:25 ` Michal Hocko
2012-10-10 11:25 ` Michal Hocko
2012-10-16 8:20 ` Kamezawa Hiroyuki
2012-10-16 8:20 ` Kamezawa Hiroyuki
2012-10-16 8:20 ` Kamezawa Hiroyuki
2012-10-08 10:06 ` [PATCH v4 09/14] memcg: kmem accounting lifecycle management Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-10-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-11 13:11 ` Michal Hocko
2012-10-11 13:11 ` Michal Hocko
2012-10-11 13:11 ` Michal Hocko
[not found] ` <20121011131143.GF29295-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-12 7:47 ` Glauber Costa
2012-10-12 7:47 ` Glauber Costa
2012-10-12 7:47 ` Glauber Costa
[not found] ` <5077CB05.907-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-12 8:41 ` Michal Hocko
2012-10-12 8:41 ` Michal Hocko
2012-10-12 8:41 ` Michal Hocko
2012-10-16 8:41 ` Kamezawa Hiroyuki
2012-10-16 8:41 ` Kamezawa Hiroyuki
2012-10-08 10:06 ` [PATCH v4 10/14] memcg: use static branches when code not in use Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-11-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-11 13:40 ` Michal Hocko
2012-10-11 13:40 ` Michal Hocko
2012-10-11 13:40 ` Michal Hocko
2012-10-12 7:47 ` Glauber Costa
2012-10-12 7:47 ` Glauber Costa
2012-10-16 8:48 ` Kamezawa Hiroyuki [this message]
2012-10-16 8:48 ` Kamezawa Hiroyuki
2012-10-08 10:06 ` [PATCH v4 11/14] memcg: allow a memcg with kmem charges to be destructed Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` [PATCH v4 12/14] execute the whole memcg freeing in free_worker Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
[not found] ` <1349690780-15988-13-git-send-email-glommer-bzQdu9zFT3WakBO8gow8eQ@public.gmane.org>
2012-10-11 14:21 ` Michal Hocko
2012-10-11 14:21 ` Michal Hocko
2012-10-11 14:21 ` Michal Hocko
2012-10-08 10:06 ` [PATCH v4 13/14] protect architectures where THREAD_SIZE >= PAGE_SIZE against fork bombs Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` [PATCH v4 14/14] Add documentation about the kmem controller Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-08 10:06 ` Glauber Costa
2012-10-11 14:35 ` Michal Hocko
2012-10-11 14:35 ` Michal Hocko
[not found] ` <20121011143559.GJ29295-2MMpYkNvuYDjFM9bn6wA6Q@public.gmane.org>
2012-10-12 7:53 ` Glauber Costa
2012-10-12 7:53 ` Glauber Costa
2012-10-12 7:53 ` Glauber Costa
2012-10-12 8:44 ` Michal Hocko
2012-10-12 8:44 ` Michal Hocko
2012-10-17 7:29 ` Kamezawa Hiroyuki
2012-10-17 7:29 ` Kamezawa Hiroyuki
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=507D1F5D.8080709@jp.fujitsu.com \
--to=kamezawa.hiroyu@jp.fujitsu.com \
--cc=akpm@linux-foundation.org \
--cc=cgroups@vger.kernel.org \
--cc=cl@linux.com \
--cc=devel@openvz.org \
--cc=fweisbec@gmail.com \
--cc=glommer@parallels.com \
--cc=gthelen@google.com \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@suse.cz \
--cc=penberg@cs.helsinki.fi \
--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 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.