On Wed, 17 Aug 2011 22:51:26 +0200
Johannes Weiner <[email protected]> wrote:

> > Signed-off-by: Johannes Weiner <[email protected]>
> > Cc: KAMEZAWA Hiroyuki <[email protected]
> > Cc: Michal Hocko <[email protected]>
> > Cc: <[email protected]>             [3.0.x]
> 
> The bug was introduced only after the 3.0 release, I think a stable
> backport is not required.

OK.

But 3.0's drain_all_stock_async() has a similar bug.


:       curcpu = raw_smp_processor_id();
:       for_each_online_cpu(cpu) {
:               struct memcg_stock_pcp *stock = &per_cpu(memcg_stock, cpu);
:               struct mem_cgroup *mem;

-> reschedule

:               if (cpu == curcpu)
:                       continue;

What are the consequences?

_______________________________________________
stable mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/stable

Reply via email to