From: Mel Gorman Date: Tue, 16 Jun 2009 22:33:23 +0000 (-0700) Subject: vmscan: count the number of times zone_reclaim() scans and fails X-Git-Tag: v2.6.30.1~10 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=cc2b6aacfc0e7a9288ed7ea518ff122454ccf7fb;p=thirdparty%2Fkernel%2Fstable.git vmscan: count the number of times zone_reclaim() scans and fails commit 24cf72518c79cdcda486ed26074ff8151291cf65 upstream. On NUMA machines, the administrator can configure zone_reclaim_mode that is a more targetted form of direct reclaim. On machines with large NUMA distances for example, a zone_reclaim_mode defaults to 1 meaning that clean unmapped pages will be reclaimed if the zone watermarks are not being met. There is a heuristic that determines if the scan is worthwhile but it is possible that the heuristic will fail and the CPU gets tied up scanning uselessly. Detecting the situation requires some guesswork and experimentation so this patch adds a counter "zreclaim_failed" to /proc/vmstat. If during high CPU utilisation this counter is increasing rapidly, then the resolution to the problem may be to set /proc/sys/vm/zone_reclaim_mode to 0. [akpm@linux-foundation.org: name things consistently] Signed-off-by: Mel Gorman Reviewed-by: Rik van Riel Cc: Christoph Lameter Reviewed-by: KOSAKI Motohiro Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index 524cd1b28ecb0..cfdd3cae88892 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -36,6 +36,9 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT, FOR_ALL_ZONES(PGSTEAL), FOR_ALL_ZONES(PGSCAN_KSWAPD), FOR_ALL_ZONES(PGSCAN_DIRECT), +#ifdef CONFIG_NUMA + PGSCAN_ZONE_RECLAIM_FAILED, +#endif PGINODESTEAL, SLABS_SCANNED, KSWAPD_STEAL, KSWAPD_INODESTEAL, PAGEOUTRUN, ALLOCSTALL, PGROTATED, #ifdef CONFIG_HUGETLB_PAGE diff --git a/mm/vmscan.c b/mm/vmscan.c index d51e9937ff20a..2500b01e7a781 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2452,6 +2452,9 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) ret = __zone_reclaim(zone, gfp_mask, order); zone_clear_flag(zone, ZONE_RECLAIM_LOCKED); + if (!ret) + count_vm_event(PGSCAN_ZONE_RECLAIM_FAILED); + return ret; } #endif diff --git a/mm/vmstat.c b/mm/vmstat.c index 74d66dba0cbe4..ddb360a09efd0 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -675,6 +675,9 @@ static const char * const vmstat_text[] = { TEXTS_FOR_ZONES("pgscan_kswapd") TEXTS_FOR_ZONES("pgscan_direct") +#ifdef CONFIG_NUMA + "zone_reclaim_failed", +#endif "pginodesteal", "slabs_scanned", "kswapd_steal",