Re: [PATCH 05/27] mm, vmscan: Have kswapd only scan based on the highest requested zone

[Vlastimil Babka]

On 06/09/2016 08:04 PM, Mel Gorman wrote:
> kswapd checks all eligible zones to see if they need balancing even if it was
> woken for a lower zone. This made sense when we reclaimed on a per-zone basis
> because we wanted to shrink zones fairly so avoid age-inversion problems.

Now we reclaim a single lru, but still will skip over pages from the 
higher zones than reclaim_idx, so this is not much different from 
per-zone basis wrt age-inversion?

> Ideally this is completely unnecessary when reclaiming on a per-node basis.
> In theory, there may still be anomalies when all requests are for lower
> zones and very old pages are preserved in higher zones but this should be
> the exceptional case.
>
> Signed-off-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>

I don't see the argument, but agree it should be exceptional in any 
case, and if there's such a case, it's better to focus on pages from the 
zone(s) where a pending (potentially atomic) allocation is restricted 
to. Or rather, this is the only way we can focus reclaim on such pages 
now that there's a single lru list.

Acked-by: Vlastimil Babka <vbabka@xxxxxxx>

> ---
>  mm/vmscan.c | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index ab1b28e7e20a..0a619241c576 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -3171,11 +3171,8 @@ static int balance_pgdat(pg_data_t *pgdat, int order, int classzone_idx)
>
>  		sc.nr_reclaimed = 0;
>
> -		/*
> -		 * Scan in the highmem->dma direction for the highest
> -		 * zone which needs scanning
> -		 */
> -		for (i = pgdat->nr_zones - 1; i >= 0; i--) {
> +		/* Scan from the highest requested zone to dma */
> +		for (i = classzone_idx; i >= 0; i--) {
>  			struct zone *zone = pgdat->node_zones + i;
>
>  			if (!populated_zone(zone))