From: Huang Ying <ying.huang(a)intel.com> ANBZ: #80 commit bfe9d006c971a5daefe7a8b27819ccd497090fd8 upstream When zone_watermark_ok() is called in migrate_balanced_pgdat() to check migration target node, the parameter classzone_idx (for requested zone) is specified as 0 (ZONE_DMA). But when allocating memory for autonuma in alloc_misplaced_dst_page(), the requested zone from GFP flags is ZONE_MOVABLE. That is, the requested zone is different. The size of lowmem_reserve for the different requested zone is different. And this may cause some issues. For example, in the zoneinfo of a test machine as below, Node 0, zone DMA32 pages free 61592 min 29 low 454 high 879 spanned 1044480 present 442306 managed 425921 protection: (0, 0, 62457, 62457, 62457) The free page number of ZONE_DMA32 is greater than "high watermark + lowmem_reserve[ZONE_DMA]", but less than "high watermark + lowmem_reserve[ZONE_MOVABLE]". And because __alloc_pages_node() in alloc_misplaced_dst_page() requests ZONE_MOVABLE, the zone_watermark_ok() on ZONE_DMA32 in migrate_balanced_pgdat() may always return true. So, autonuma may not stop even when memory pressure in node 0 is heavy. To fix the issue, ZONE_MOVABLE is used as parameter to call zone_watermark_ok() in migrate_balanced_pgdat(). This makes it same as requested zone in alloc_misplaced_dst_page(). So that migrate_balanced_pgdat() returns false when memory pressure is heavy. Link: http://lkml.kernel.org/r/20191101075727.26683-2-ying.huang@intel.com Signed-off-by: "Huang, Ying" <ying.huang(a)intel.com> Acked-by: Mel Gorman <mgorman(a)suse.de> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Rik van Riel <riel(a)redhat.com> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Ingo Molnar <mingo(a)kernel.org> Cc: Dave Hansen <dave.hansen(a)linux.intel.com> Cc: Dan Williams <dan.j.williams(a)intel.com> Cc: Fengguang Wu <fengguang.wu(a)intel.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Baolin Wang <baolin.wang(a)linux.alibaba.com> --- Note: this patch fixes the problem that the DRAM node's kswapd is not waked up in time, and improves about 12% with mysql performance testing. --- mm/migrate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/migrate.c b/mm/migrate.c index 6d25ea0..e2dbf24 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1969,7 +1969,7 @@ static bool migrate_balanced_pgdat(struct pglist_data *pgdat, int order) /* Avoid waking kswapd by allocating pages to migrate. */ if (!zone_watermark_ok(zone, order, high_wmark_pages(zone), - 0, 0)) + ZONE_MOVABLE, 0)) continue; return true; }