Messages in this thread | | | Date | Thu, 25 Oct 2012 11:08:12 -0700 | From | Tejun Heo <> | Subject | Re: [PATCH v5 08/18] memcg: infrastructure to match an allocation to the right cache |
| |
On Thu, Oct 25, 2012 at 11:06:52AM -0700, Tejun Heo wrote: > Hello, Glauber. > > On Thu, Oct 25, 2012 at 03:05:22PM +0400, Glauber Costa wrote: > > > Is there any rmb() pair? > > > As far as I know, without rmb(), wmb() doesn't guarantee anything. > > > > > > > There should be. But it seems I missed it. Speaking of which, I should > > You probably can use read_barrier_depends().
Ooh, and when you use memory barrier pairs, please make sure to comment on both sides explaining how they're paired and why. There's nothing more frustrating than barriers without clear annotation.
Thanks.
-- tejun
| |