Re: commit 0bf1457f0cfca7b " mm: vmscan: do not swap anon pages just because free+file is low" causes heavy performance regression on paging

[Rik van Riel <riel@redhat.com>]

On 04/22/2014 07:57 AM, Christian Borntraeger wrote:
> On 22/04/14 12:55, Christian Borntraeger wrote:
>> While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very severe performance hickup on guest paging (all anonymous memory).
>>
>> All memory bound guests are in "D" state now and the system is barely unusable.
>>
>> Reverting commit 0bf1457f0cfca7bc026a82323ad34bcf58ad035d
>> "mm: vmscan: do not swap anon pages just because free+file is low" makes the problem go away.
>>
>> According to /proc/vmstat the system is now in direct reclaim almost all the time for every page fault (more than 10x more direct reclaims than kswap reclaims)
>> With the patch being reverted everything is fine again.
>>
>> Any ideas?
> 
> Here is an idea to tackle my problem and the original problem:
> 
> reverting  0bf1457f0cfca7bc026a82323ad34bcf58ad035d + checking against low, also seems to make my system usable.
> 
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1923,7 +1923,7 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
>          */
>         if (global_reclaim(sc)) {
>                 free = zone_page_state(zone, NR_FREE_PAGES);
> -               if (unlikely(file + free <= high_wmark_pages(zone))) {
> +               if (unlikely(file + free <= low_wmark_pages(zone))) {
>                         scan_balance = SCAN_ANON;
>                         goto out;
>                 }
> 

Looks reasonable to me.  Johannes?

-- 
All rights reversed