Re: [patch 12/23] SLUB: Trigger defragmentation from memory reclaim

[Christoph Lameter]

On Wed, 7 Nov 2007, Johannes Weiner wrote:

> > @@ -210,6 +218,8 @@ unsigned long shrink_slab(unsigned long
> >                 shrinker->nr += total_scan;
> >         }
> >         up_read(&shrinker_rwsem);
> > +       if (gfp_mask & __GFP_FS)
> > +               kmem_cache_defrag(zone ? zone_to_nid(zone) : -1);
> >         return ret;
> >  }
> 
> What about the objects that kmem_cache_defrag() releases?  Shouldn't
> they be counted too?
> 
>      ret += kmem_cache_defrag(...)
> 
> Or am I overseeing something here?

kmem_cache_defrag returns the number of pages that were released by defrag 
actions.

shrink_slab returns the number of objects released by the shrinkers. 
kmem_cache_defrag has no way of knowing how many objects where released by 
the kick methods. The kick method may have chosen to reallocate the 
object.


-
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/