On Fri 30-10-15 12:10:15, Hillf Danton wrote:
[...]
> > + for_each_zone_zonelist_nodemask(zone, z, ac->zonelist, ac->high_zoneidx, ac->nodemask) {
> > + unsigned long free = zone_page_state(zone, NR_FREE_PAGES);
> > + unsigned long reclaimable;
> > + unsigned long target;
> > +
> > + reclaimable = zone_reclaimable_pages(zone) +
> > + zone_page_state(zone, NR_ISOLATED_FILE) +
> > + zone_page_state(zone, NR_ISOLATED_ANON);
> > + target = reclaimable;
> > + target -= stall_backoff * (1 + target/MAX_STALL_BACKOFF);
>
> target = reclaimable - stall_backoff * (1 + target/MAX_STALL_BACKOFF);
> = reclaimable - stall_backoff - stall_backoff * (target/MAX_STALL_BACKOFF);
>
> then the first stall_backoff looks unreasonable.
First stall_backoff is off by 1 but that shouldn't make any difference.
> I guess you mean
> target = reclaimable - target * (stall_backoff/MAX_STALL_BACKOFF);
> = reclaimable - stall_back * (target/MAX_STALL_BACKOFF);
No the reason I used the bias is to converge for MAX_STALL_BACKOFF. If
you have target which is not divisible by MAX_STALL_BACKOFF then the
rounding would get target > 0 and so we wouldn't converge. With the +1
you underflow which is MAX_STALL_BACKOFF in maximum which should be
fixed up by the free memory. Maybe a check for free < MAX_STALL_BACKOFF
would be good but I didn't get that far with this.
[...]
> /*
> > @@ -2734,10 +2730,6 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
> > goto retry;
> > }
> >
> > - /* Any of the zones still reclaimable? Don't OOM. */
> > - if (zones_reclaimable)
> > - return 1;
> > -
>
> Looks cleanup of zones_reclaimable left.
Removed. Thanks!
--
Michal Hocko
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>