[PATCH 1/2] memcg: dirty pages accounting and limiting infrastructure

Minchan Kim minchan.kim at gmail.com
Thu Feb 25 20:50:04 PST 2010


Hi, Kame.

On Fri, Feb 26, 2010 at 9:23 AM, KAMEZAWA Hiroyuki
<kamezawa.hiroyu at jp.fujitsu.com> wrote:
> On Fri, 26 Feb 2010 00:36:15 +0900
> Minchan Kim <minchan.kim at gmail.com> wrote:
>
>> Hi
>>
>> On Tue, Feb 23, 2010 at 8:58 PM, Andrea Righi <arighi at develer.com> wrote:
>> > On Mon, Feb 22, 2010 at 01:07:32PM -0500, Vivek Goyal wrote:
>> >> > > > +unsigned long mem_cgroup_dirty_bytes(void)
>> >> > > > +{
>> >> > > > +       struct mem_cgroup *memcg;
>> >> > > > +       unsigned long dirty_bytes;
>> >> > > > +
>> >> > > > +       if (mem_cgroup_disabled())
>> >> > > > +               return vm_dirty_bytes;
>> >> > > > +
>> >> > > > +       rcu_read_lock();
>> >> > > > +       memcg = mem_cgroup_from_task(current);
>> >> > > > +       if (memcg == NULL)
>> >> > > > +               dirty_bytes = vm_dirty_bytes;
>> >> > > > +       else
>> >> > > > +               dirty_bytes = get_dirty_bytes(memcg);
>> >> > > > +       rcu_read_unlock();
>> >> > >
>> >> > > The rcu_read_lock() isn't protecting anything here.
>> >> >
>> >> > Right!
>> >>
>> >> Are we not protecting "memcg" pointer using rcu here?
>> >
>> > Vivek, you are right:
>> >
>> >  mem_cgroup_from_task() -> task_subsys_state() -> rcu_dereference()
>> >
>> > So, this *must* be RCU protected.
>>
>> So, Doesn't mem_cgroup_from_task in mem_cgroup_can_attach need RCU, too?
>>
> Hm ? I don't read the whole thread but can_attach() is called under
> cgroup_mutex(). So, it doesn't need to use RCU.

Vivek mentioned memcg is protected by RCU if I understand his intention right.
So I commented that without enough knowledge of memcg.
After your comment, I dive into the code.

Just out of curiosity.

Really, memcg is protected by RCU?
I think most of RCU around memcg is for protecting task_struct and
cgroup_subsys_state.
The memcg is protected by cgroup_mutex as you mentioned.
Am I missing something?

> Thanks,
> -Kame
>
>
>



-- 
Kind regards,
Minchan Kim


More information about the Containers mailing list