From: KOSAKI Motohiro Date: Tue, 10 Aug 2010 00:19:54 +0000 (-0700) Subject: vmscan: shrink_slab() requires the number of lru_pages, not the page order X-Git-Tag: v2.6.36-rc1~397 X-Git-Url: http://git.openpandora.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=4dc4b3d971b23e12d483ba9f3b93b648c54b298a;p=pandora-kernel.git vmscan: shrink_slab() requires the number of lru_pages, not the page order Presently shrink_slab() has the following scanning equation. lru_scanned max_pass basic_scan_objects = 4 x ------------- x ----------------------------- lru_pages shrinker->seeks (default:2) scan_objects = min(basic_scan_objects, max_pass * 2) If we pass very small value as lru_pages instead real number of lru pages, shrink_slab() drop much objects rather than necessary. And now, __zone_reclaim() pass 'order' as lru_pages by mistake. That produces a bad result. For example, if we receive very low memory pressure (scan = 32, order = 0), shrink_slab() via zone_reclaim() always drop _all_ icache/dcache objects. (see above equation, very small lru_pages make very big scan_objects result). This patch fixes it. [akpm@linux-foundation.org: fix layout, typos] Signed-off-by: KOSAKI Motohiro Reviewed-by: Minchan Kim Acked-by: Christoph Lameter Acked-by: Rik van Riel Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Reading git-diff-tree failed