[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20151219085137.GV28521@esperanza>
Date: Sat, 19 Dec 2015 11:51:38 +0300
From: Vladimir Davydov <vdavydov@...tuozzo.com>
To: Andrew Morton <akpm@...ux-foundation.org>
CC: Johannes Weiner <hannes@...xchg.org>, <stable@...r.kernel.org>,
Michal Hocko <mhocko@...nel.org>, <linux-mm@...ck.org>,
<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] mm: memcontrol: fix possible memcg leak due to
interrupted reclaim
On Fri, Dec 18, 2015 at 02:40:04PM -0800, Andrew Morton wrote:
> On Fri, 18 Dec 2015 19:24:05 +0300 Vladimir Davydov <vdavydov@...tuozzo.com> wrote:
>
> >
> > OK, got it, thanks. Here goes the incremental patch (it should also fix
> > the warning regarding unused cmpxchg returned value):
> > ---
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index fc25dc211eaf..908c075e04eb 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -864,7 +864,7 @@ struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *root,
> > * might block it. So we clear iter->position right
> > * away.
> > */
> > - cmpxchg(&iter->position, pos, NULL);
> > + (void)cmpxchg(&iter->position, pos, NULL);
>
> No, this doesn't actually squish the __must_check warning.
The warning was caused not by a __must_check annotation - using it for
cmpxchg would be just wrong - it was caused by type conversion done in
expansion of cmpxchg macro:
arch/m68k/include/asm/cmpxchg.h:121:3: warning: value computed is not used [-Wunused-value]
((__typeof__(*(ptr)))__cmpxchg((ptr), (unsigned long)(o), \
^
(see http://www.spinics.net/lists/linux-mm/msg99133.html)
Type conversion to (void) helps suppressing this warning, and it seems
this is what is done commonly (e.g. see kernel/rcu/tree_plugin.h)
Thanks,
Vladimir
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists