From: Roman Gushchin <guro@xxxxxx> Subject: mm,page_alloc,cma: conditionally prefer cma pageblocks for movable allocations Currently a cma area is barely used by the page allocator because it's used only as a fallback from movable, however kswapd tries hard to make sure that the fallback path isn't used. This results in a system evicting memory and pushing data into swap, while lots of CMA memory is still available. This happens despite the fact that alloc_contig_range is perfectly capable of moving any movable allocations out of the way of an allocation. To effectively use the cma area let's alter the rules: if the zone has more free cma pages than the half of total free pages in the zone, use cma pageblocks first and fallback to movable blocks in the case of failure. [guro@xxxxxx: ifdef the cma-specific code] Link: http://lkml.kernel.org/r/20200311225832.GA178154@xxxxxxxxxxxxxxxxxxxxxxxxxxx Link: http://lkml.kernel.org/r/20200306150102.3e77354b@xxxxxxxxxxxxxxxxxxxx Signed-off-by: Roman Gushchin <guro@xxxxxx> Signed-off-by: Rik van Riel <riel@xxxxxxxxxxx> Co-developed-by: Rik van Riel <riel@xxxxxxxxxxx> Acked-by: Vlastimil Babka <vbabka@xxxxxxx> Acked-by: Minchan Kim <minchan@xxxxxxxxxx> Cc: Qian Cai <cai@xxxxxx> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> Cc: Anshuman Khandual <anshuman.khandual@xxxxxxx> Cc: Joonsoo Kim <js1304@xxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/page_alloc.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) --- a/mm/page_alloc.c~mmpage_alloccma-conditionally-prefer-cma-pageblocks-for-movable-allocations +++ a/mm/page_alloc.c @@ -2752,6 +2752,20 @@ __rmqueue(struct zone *zone, unsigned in { struct page *page; +#ifdef CONFIG_CMA + /* + * Balance movable allocations between regular and CMA areas by + * allocating from CMA when over half of the zone's free memory + * is in the CMA area. + */ + if (migratetype == MIGRATE_MOVABLE && + zone_page_state(zone, NR_FREE_CMA_PAGES) > + zone_page_state(zone, NR_FREE_PAGES) / 2) { + page = __rmqueue_cma_fallback(zone, order); + if (page) + return page; + } +#endif retry: page = __rmqueue_smallest(zone, order, migratetype); if (unlikely(!page)) { _
