lkml.org 
[lkml]   [2011]   [Jun]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [BUGFIX][PATCH v3] memcg: fix behavior of per cpu charge cache draining.
On Fri 10-06-11 17:39:58, KAMEZAWA Hiroyuki wrote:
> On Fri, 10 Jun 2011 10:12:19 +0200
> Michal Hocko <mhocko@suse.cz> wrote:
>
> > On Thu 09-06-11 09:30:45, KAMEZAWA Hiroyuki wrote:
[...]
> > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > > index bd9052a..3baddcb 100644
> > > --- a/mm/memcontrol.c
> > > +++ b/mm/memcontrol.c
> > [...]
> > > static struct mem_cgroup_per_zone *
> > > mem_cgroup_zoneinfo(struct mem_cgroup *mem, int nid, int zid)
> > > @@ -1670,8 +1670,6 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
> > > victim = mem_cgroup_select_victim(root_mem);
> > > if (victim == root_mem) {
> > > loop++;
> > > - if (loop >= 1)
> > > - drain_all_stock_async();
> > > if (loop >= 2) {
> > > /*
> > > * If we have not been able to reclaim
> > > @@ -1723,6 +1721,7 @@ static int mem_cgroup_hierarchical_reclaim(struct mem_cgroup *root_mem,
> > > return total;
> > > } else if (mem_cgroup_margin(root_mem))
> > > return total;
> > > + drain_all_stock_async(root_mem);
> > > }
> > > return total;
> > > }
> >
> > I still think that we pointlessly reclaim even though we could have a
> > lot of pages pre-charged in the cache (the more CPUs we have the more
> > significant this might be).
>
> The more CPUs, the more scan cost for each per-cpu memory, which makes
> cache-miss.
>
> I know placement of drain_all_stock_async() is not big problem on my host,
> which has 2socket/8core cpus. But, assuming 1000+ cpu host,

Hmm, it really depends what you want to optimize for. Reclaim path is
already slow path and cache misses, while not good, are not the most
significant issue, I guess.
What I would see as a much bigger problem is that there might be a lot
of memory pre-charged at those per-cpu caches. Falling into a reclaim
costs us much more IMO and we can evict something that could be useful
for no good reason.

> "when you hit limit, you'll see 1000*128bytes cache miss and need to
> call test_and_set for 1000+ cpus in bad case." doesn't seem much win.
>
> If we implement "call-drain-only-nearby-cpus", I think we can call it before
> calling try_to_free_mem_cgroup_pages(). I'll add it to my TO-DO-LIST.

It would just consider cpus at the same node?

> How do you think ?

I am afraid we would need two versions then. One for complete draining
(rmdir and company) while the other for reclaim purposes. Which sounds
like a more code complexity.

--
Michal Hocko
SUSE Labs
SUSE LINUX s.r.o.
Lihovarska 1060/12
190 00 Praha 9
Czech Republic


\
 
 \ /
  Last update: 2011-06-10 11:11    [W:0.062 / U:0.328 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site