| Message ID | 20180815003620.15678-2-guro@fb.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [RFC,1/2] mm: rework memcg kernel stack accounting | expand |
On Tue, Aug 14, 2018 at 5:36 PM Roman Gushchin <guro@fb.com> wrote: > > Memcg charge is batched using per-cpu stocks, so an offline memcg > can be pinned by a cached charge up to a moment, when a process > belonging to some other cgroup will charge some memory on the same > cpu. In other words, cached charges can prevent a memory cgroup > from being reclaimed for some time, without any clear need. > > Let's optimize it by explicit draining of all stocks on css offlining. > As draining is performed asynchronously, and is skipped if any > parallel draining is happening, it's cheap. > > Signed-off-by: Roman Gushchin <guro@fb.com> Seems reasonable. Reviewed-by: Shakeel Butt <shakeelb@google.com> > Cc: Johannes Weiner <hannes@cmpxchg.org> > Cc: Michal Hocko <mhocko@kernel.org> > Cc: Konstantin Khlebnikov <koct9i@gmail.com> > Cc: Tejun Heo <tj@kernel.org> > --- > mm/memcontrol.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index 4e3c1315b1de..cfb64b5b9957 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -4575,6 +4575,8 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css) > memcg_offline_kmem(memcg); > wb_memcg_offline(memcg); > > + drain_all_stock(memcg); > + > mem_cgroup_id_put(memcg); > } > > -- > 2.14.4 >
On Tue 14-08-18 17:36:20, Roman Gushchin wrote: > Memcg charge is batched using per-cpu stocks, so an offline memcg > can be pinned by a cached charge up to a moment, when a process > belonging to some other cgroup will charge some memory on the same > cpu. In other words, cached charges can prevent a memory cgroup > from being reclaimed for some time, without any clear need. > > Let's optimize it by explicit draining of all stocks on css offlining. > As draining is performed asynchronously, and is skipped if any > parallel draining is happening, it's cheap. Yes this makes sense. > Signed-off-by: Roman Gushchin <guro@fb.com> > Cc: Johannes Weiner <hannes@cmpxchg.org> > Cc: Michal Hocko <mhocko@kernel.org> > Cc: Konstantin Khlebnikov <koct9i@gmail.com> > Cc: Tejun Heo <tj@kernel.org> Acked-by: Michal Hocko <mhocko@suse.com> > --- > mm/memcontrol.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index 4e3c1315b1de..cfb64b5b9957 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -4575,6 +4575,8 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css) > memcg_offline_kmem(memcg); > wb_memcg_offline(memcg); > > + drain_all_stock(memcg); > + > mem_cgroup_id_put(memcg); > } > > -- > 2.14.4
diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 4e3c1315b1de..cfb64b5b9957 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4575,6 +4575,8 @@ static void mem_cgroup_css_offline(struct cgroup_subsys_state *css) memcg_offline_kmem(memcg); wb_memcg_offline(memcg); + drain_all_stock(memcg); + mem_cgroup_id_put(memcg); }
Memcg charge is batched using per-cpu stocks, so an offline memcg can be pinned by a cached charge up to a moment, when a process belonging to some other cgroup will charge some memory on the same cpu. In other words, cached charges can prevent a memory cgroup from being reclaimed for some time, without any clear need. Let's optimize it by explicit draining of all stocks on css offlining. As draining is performed asynchronously, and is skipped if any parallel draining is happening, it's cheap. Signed-off-by: Roman Gushchin <guro@fb.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Konstantin Khlebnikov <koct9i@gmail.com> Cc: Tejun Heo <tj@kernel.org> --- mm/memcontrol.c | 2 ++ 1 file changed, 2 insertions(+)