From: Mel Gorman <[email protected]> When deciding if kswapd is sleeping prematurely, the classzone is taken into account but this is different to what balance_pgdat() and the allocator are doing. Specifically, the DMA zone will be checked based on the classzone used when waking kswapd which could be for a GFP_KERNEL or GFP_HIGHMEM request. The lowmem reserve limit kicks in, the watermark is not met and kswapd thinks it's sleeping prematurely keeping kswapd awake in error.
Signed-off-by: Mel Gorman <[email protected]> Reported-by: Pádraig Brady <[email protected]> Tested-by: Pádraig Brady <[email protected]> Tested-by: Andrew Lutomirski <[email protected]> Acked-by: Rik van Riel <[email protected]> Reviewed-by: Minchan Kim <[email protected]> Cc: KOSAKI Motohiro <[email protected]> Cc: Johannes Weiner <[email protected]> Cc: <[email protected]> Signed-off-by: Andrew Morton <[email protected]> --- mm/vmscan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff -puN mm/vmscan.c~mm-vmscan-evaluate-the-watermarks-against-the-correct-classzone mm/vmscan.c --- a/mm/vmscan.c~mm-vmscan-evaluate-the-watermarks-against-the-correct-classzone +++ a/mm/vmscan.c @@ -2344,7 +2344,7 @@ static bool sleeping_prematurely(pg_data } if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), - classzone_idx, 0)) + i, 0)) all_zones_ok = false; else balanced += zone->present_pages; _ _______________________________________________ stable mailing list [email protected] http://linux.kernel.org/mailman/listinfo/stable
