Re: SLUB regression in current Linus

[Eric Dumazet]

Le mercredi 25 mai 2011 Ã 10:04 -0500, Christoph Lameter a Ãcrit :
> On Wed, 25 May 2011, Linus Torvalds wrote:
> 
> > On Wed, May 25, 2011 at 7:47 AM, Christoph Lameter <cl@xxxxxxxxx> wrote:
> > >
> > > So simply return the object. In the debug case we do not need all the other
> > > processing that unlock_out: does.
> >
> > So that patch looks like it should explain things.
> >
> > But exactly why don't we need the ->tid update and the ALLOC_SLOWPATH
> > stats? Both of them would seem to be equally valid for the debug case.
> 
> We always fall back to the slow path in the debug case. So the tid update
> is useless.

tid change is updated anyway at the end of deactivate_slab()


--
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/