On Fri, Apr 29, 2011 at 6:05 AM, Michal Hocko <mhocko@xxxxxxx> wrote:
> On Thu 28-04-11 15:37:05, Ying Han wrote:
>> We recently added the change in global background reclaim which
>> counts the return value of soft_limit reclaim. Now this patch adds
>> the similar logic on global direct reclaim.
>>
>> We should skip scanning global LRU on shrink_zone if soft_limit reclaim
>> does enough work. This is the first step where we start with counting
>> the nr_scanned and nr_reclaimed from soft_limit reclaim into global
>> scan_control.
>
> Makes sense.
>
>>
>> Signed-off-by: Ying Han <yinghan@xxxxxxxxxx>
>> ---
>> mm/vmscan.c | 16 ++++++++++++++--
>> 1 files changed, 14 insertions(+), 2 deletions(-)
>>
>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>> index b3a569f..84003cc 100644
>> --- a/mm/vmscan.c
>> +++ b/mm/vmscan.c
>> @@ -1959,11 +1959,14 @@ restart:
>> * If a zone is deemed to be full of pinned pages then just give it a light
>> * scan then give up on it.
>> */
>> -static void shrink_zones(int priority, struct zonelist *zonelist,
>> +static unsigned long shrink_zones(int priority, struct zonelist *zonelist,
>> struct scan_control *sc)
>> {
>> struct zoneref *z;
>> struct zone *zone;
>> + unsigned long nr_soft_reclaimed;
>> + unsigned long nr_soft_scanned;
>> + unsigned long total_scanned = 0;
>>
>> for_each_zone_zonelist_nodemask(zone, z, zonelist,
>> gfp_zone(sc->gfp_mask), sc->nodemask) {
>> @@ -1980,8 +1983,17 @@ static void shrink_zones(int priority, struct zonelist *zonelist,
>> continue; /* Let kswapd poll it */
>> }
>>
>> + nr_soft_scanned = 0;
>> + nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone,
>> + sc->order, sc->gfp_mask,
>> + &nr_soft_scanned);
>> + sc->nr_reclaimed += nr_soft_reclaimed;
>> + total_scanned += nr_soft_scanned;
>> +
>> shrink_zone(priority, zone, sc);
>
> This can cause more aggressive reclaiming, right? Shouldn't we check
> whether shrink_zone is still needed?
We decided to leave the shrink_zone for now before making further
changes for soft_limit reclaim. The same
patch I did last time for global background reclaim. It is safer to do
this step-by-step :)
--Ying
>
> --
> Michal Hocko
> SUSE Labs
> SUSE LINUX s.r.o.
> Lihovarska 1060/12
> 190 00 Praha 9
> Czech Republic
>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href