On Fri 03-02-17 20:32:22, Mel Gorman wrote:
> Hillf Danton pointed out that since commit 1d82de618dd ("mm, vmscan:
> make kswapd reclaim in terms of nodes") that PGDAT_WRITEBACK is no longer
> cleared. It was not noticed as triggering it requires pages under writeback
> to cycle twice through the LRU and before kswapd gets stalled. Historically,
> such issues tended to occur on small machines writing heavily to slow
> storage such as a USB stick. Once kswapd stalls, direct reclaim stalls may
> be higher but due to the fact that memory pressure is requires, it would not
> be very noticable. Michal Hocko suggested removing the flag entirely but
> the conservative fix is to restore the intended PGDAT_WRITEBACK behaviour
> and clear the flag when a suitable zone is balanced.
>
> Signed-off-by: Mel Gorman <mgor...@suse.de>
I agree this is a more conservative approach but I think removing
PGDAT_WRITEBACK should simplify things a bit.
Acked-by: Michal Hocko <mho...@suse.com>
> ---
> mm/vmscan.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 532a2a750952..3379fa5ce6d8 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -3103,6 +3103,7 @@ static bool zone_balanced(struct zone *zone, int order,
> int classzone_idx)
> */
> clear_bit(PGDAT_CONGESTED, &zone->zone_pgdat->flags);
> clear_bit(PGDAT_DIRTY, &zone->zone_pgdat->flags);
> + clear_bit(PGDAT_WRITEBACK, &zone->zone_pgdat->flags);
>
> return true;
> }
>
> --
> Mel Gorman
> SUSE Labs
--
Michal Hocko
SUSE Labs
