lkml.org 
[lkml]   [2010]   [Jul]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH] vmscan: stop meaningless loop iteration when no reclaimable slab
Date
> On Fri, Jul 9, 2010 at 7:13 PM, KOSAKI Motohiro
> <kosaki.motohiro@jp.fujitsu.com> wrote:
> > If number of reclaimable slabs are zero, shrink_icache_memory() and
> > shrink_dcache_memory() return 0. but strangely shrink_slab() ignore
> > it and continue meaningless loop iteration.
> >
> > This patch fixes it.
> >
> > Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> > ---
> >  mm/vmscan.c |    5 +++++
> >  1 files changed, 5 insertions(+), 0 deletions(-)
> >
> > diff --git a/mm/vmscan.c b/mm/vmscan.c
> > index 0f9f624..8f61adb 100644
> > --- a/mm/vmscan.c
> > +++ b/mm/vmscan.c
> > @@ -243,6 +243,11 @@ unsigned long shrink_slab(unsigned long scanned, gfp_t gfp_mask,
> >                        int nr_before;
> >
> >                        nr_before = (*shrinker->shrink)(0, gfp_mask);
> > +                       /* no slab objects, no more reclaim. */
> > +                       if (nr_before == 0) {
> > +                               total_scan = 0;
>
> Why do you reset totoal_scan to 0?

If shab objects are zero, we don't need more reclaim.

> I don't know exact meaning of shrinker->nr.

similar meaning of reclaim_stat->nr_saved_scan.
If total_scan can't divide SHRINK_BATCH(128), saving remainder and using at next shrink_slab().

> AFAIU, it can affect next shrinker's total_scan.
> Isn't it harmful?

No. This loop is

total_scan = shrinker->nr; /* Reset and init total_scan */
shrinker->nr = 0;

while (total_scan >= SHRINK_BATCH) {
nr_before = (*shrinker->shrink)(0, gfp_mask);
/* no slab objects, no more reclaim. */
if (nr_before == 0) {
total_scan = 0;
break;
}
shrink_ret = (*shrinker->shrink)(this_scan, gfp_mask);
if (shrink_ret == -1)
break;
if (shrink_ret < nr_before)
ret += nr_before - shrink_ret;
total_scan -= this_scan;
}

shrinker->nr += total_scan; /* save remainder #of-scan */



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2010-07-09 13:07    [W:0.089 / U:0.572 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site