Re: [PATCH] memcg: Fix race condition in memcg_check_events() withthis_cpu usage

[Andrew Morton]

On Mon, 26 Sep 2011 09:43:22 +0900
KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> wrote:

> On Fri, 23 Sep 2011 20:54:42 -0400
> Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> 
> > From: Steven Rostedt <srostedt@xxxxxxxxxx>
> > 
> > The code in memcg_check_events() calls this_cpu_read() on
> > different variables without disabling preemption, and can cause
> > the calculations to be done from two different CPU variables.
> > 
> > Disable preemption throughout the check to keep apples and oranges
> > from becoming a mixed drink.
> > 
> > [ Added this_cpu to __this_cpu conversion by Johannes ]
> > 
> > Cc: Greg Thelen <gthelen@xxxxxxxxxx>
> > Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
> > Cc: Balbir Singh <balbir@xxxxxxxxxxxxxxxxxx>
> > Cc: Daisuke Nishimura <nishimura@xxxxxxxxxxxxxxxxx>
> > Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> > Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> > Cc: Christoph Lameter <cl@xxxxxxxxx>
> > Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
> > Signed-off-by: Steven Rostedt <rostedt@xxxxxxxxxxx>
> > Link: http://lkml.kernel.org/r/20110919212641.015320989@xxxxxxxxxxx
> 
> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
> 
> Andrew, could you pick this up ?

The patch needed rework due to other changes we have pending in there.


From: Steven Rostedt <srostedt@xxxxxxxxxx>
Subject: memcg: Fix race condition in memcg_check_events() with this_cpu usage

Various code in memcontrol.c () calls this_cpu_read() on the calculations
to be done from two different percpu variables, or does an open-coded
read-modify-write on a single percpu variable.

Disable preemption throughout these operations so that the writes go to
the correct palces.

[ Added this_cpu to __this_cpu conversion by Johannes ]

Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Signed-off-by: Steven Rostedt <rostedt@xxxxxxxxxxx>
Cc: Greg Thelen <gthelen@xxxxxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: Balbir Singh <balbir@xxxxxxxxxxxxxxxxxx>
Cc: Daisuke Nishimura <nishimura@xxxxxxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Christoph Lameter <cl@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---

 mm/memcontrol.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff -puN mm/memcontrol.c~memcg-fix-race-condition-in-memcg_check_events-with-this_cpu-usage mm/memcontrol.c
--- a/mm/memcontrol.c~memcg-fix-race-condition-in-memcg_check_events-with-this_cpu-usage
+++ a/mm/memcontrol.c
@@ -687,8 +687,8 @@ static bool __memcg_event_check(struct m
 {
 	unsigned long val, next;
 
-	val = this_cpu_read(memcg->stat->events[MEM_CGROUP_EVENTS_COUNT]);
-	next = this_cpu_read(memcg->stat->targets[target]);
+	val = __this_cpu_read(memcg->stat->events[MEM_CGROUP_EVENTS_COUNT]);
+	next = __this_cpu_read(memcg->stat->targets[target]);
 	/* from time_after() in jiffies.h */
 	return ((long)next - (long)val < 0);
 }
@@ -697,7 +697,7 @@ static void __mem_cgroup_target_update(s
 {
 	unsigned long val, next;
 
-	val = this_cpu_read(memcg->stat->events[MEM_CGROUP_EVENTS_COUNT]);
+	val = __this_cpu_read(memcg->stat->events[MEM_CGROUP_EVENTS_COUNT]);
 
 	switch (target) {
 	case MEM_CGROUP_TARGET_THRESH:
@@ -713,7 +713,7 @@ static void __mem_cgroup_target_update(s
 		return;
 	}
 
-	this_cpu_write(memcg->stat->targets[target], next);
+	__this_cpu_write(memcg->stat->targets[target], next);
 }
 
 /*
@@ -722,6 +722,7 @@ static void __mem_cgroup_target_update(s
  */
 static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
 {
+	preempt_disable();
 	/* threshold event is triggered in finer grain than soft limit */
 	if (unlikely(__memcg_event_check(memcg, MEM_CGROUP_TARGET_THRESH))) {
 		mem_cgroup_threshold(memcg);
@@ -741,6 +742,7 @@ static void memcg_check_events(struct me
 		}
 #endif
 	}
+	preempt_enable();
 }
 
 static struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
_

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/