Subject: [merged] mm-memcg-protect-mem_cgroup_read_events-for-cpu-hotplug.patch removed from -mm tree
To: rientjes@xxxxxxxxxx,hannes@xxxxxxxxxxx,kamezawa.hiroyu@xxxxxxxxxxxxxx,kosaki.motohiro@xxxxxxxxxxxxxx,mhocko@xxxxxxx,mm-commits@xxxxxxxxxxxxxxx
From: akpm@xxxxxxxxxxxxxxxxxxxx
Date: Thu, 17 Oct 2013 11:12:53 -0700
The patch titled
Subject: mm, memcg: protect mem_cgroup_read_events for cpu hotplug
has been removed from the -mm tree. Its filename was
mm-memcg-protect-mem_cgroup_read_events-for-cpu-hotplug.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: David Rientjes <rientjes@xxxxxxxxxx>
Subject: mm, memcg: protect mem_cgroup_read_events for cpu hotplug
for_each_online_cpu() needs the protection of {get,put}_online_cpus() so
cpu_online_mask doesn't change during the iteration.
cpu_hotplug.lock is held while a cpu is going down, it's a coarse lock
that is used kernel-wide to synchronize cpu hotplug activity. Memcg has a
cpu hotplug notifier, called while there may not be any cpu hotplug
refcounts, which drains per-cpu event counts to memcg->nocpu_base.events
to maintain a cumulative event count as cpus disappear. Without
get_online_cpus() in mem_cgroup_read_events(), it's possible to account
for the event count on a dying cpu twice, and this value may be
significantly large.
In fact, all memcg->pcp_counter_lock use should be nested by
{get,put}_online_cpus().
This fixes that issue and ensures the reported statistics are not vastly
over-reported during cpu hotplug.
Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/memcontrol.c | 2 ++
1 file changed, 2 insertions(+)
diff -puN mm/memcontrol.c~mm-memcg-protect-mem_cgroup_read_events-for-cpu-hotplug mm/memcontrol.c
--- a/mm/memcontrol.c~mm-memcg-protect-mem_cgroup_read_events-for-cpu-hotplug
+++ a/mm/memcontrol.c
@@ -866,6 +866,7 @@ static unsigned long mem_cgroup_read_eve
unsigned long val = 0;
int cpu;
+ get_online_cpus();
for_each_online_cpu(cpu)
val += per_cpu(memcg->stat->events[idx], cpu);
#ifdef CONFIG_HOTPLUG_CPU
@@ -873,6 +874,7 @@ static unsigned long mem_cgroup_read_eve
val += memcg->nocpu_base.events[idx];
spin_unlock(&memcg->pcp_counter_lock);
#endif
+ put_online_cpus();
return val;
}
_
Patches currently in -mm which might be from rientjes@xxxxxxxxxx are
origin.patch
linux-next.patch
kthread-make-kthread_create-killable.patch
x86-srat-use-numa_no_node.patch
mm-vmalloc-use-numa_no_node.patch
mm-add-a-helper-function-to-check-may-oom-condition.patch
mm-arch-use-numa_no_node.patch
mm-mempolicy-make-mpol_to_str-robust-and-always-succeed.patch
mm-vmalloc-dont-set-area-caller-twice.patch
mm-vmalloc-fix-show-vmap_area-information-race-with-vmap_area-tear-down.patch
mm-vmalloc-revert-mm-vmallocc-check-vm_uninitialized-flag-in-s_show-instead-of-show_numa_info.patch
revert-mm-vmallocc-emit-the-failure-message-before-return.patch
mm-mempolicy-use-numa_no_node.patch
mm-rearrange-madvise-code-to-allow-for-reuse.patch
mm-add-a-field-to-store-names-for-private-anonymous-memory.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
