The upstream change reverts the other change as listed below:
commit d1908362ae0b97374eb8328fbb471576332f9fb1
Author: Minchan Kim <minchan.kim@xxxxxxxxx>
Date: Wed Sep 22 13:05:01 2010 -0700
vmscan: check all_unreclaimable in direct reclaim path
The zone->all_unreclaimable flag is set by kswapd by checking zone->pages_scanned in
zone_reclaimable(). It is possible to have zone->all_unreclaimable == false while
the zone is actually unreclaimable, and it will cause machine to stuck.
1. while kswapd in reclaim priority loop, someone frees a page on the zone. It
will end up resetting the pages_scanned.
2. kswapd is frozen for whatever reason. This happens in hibernation where we are
not interested in google.
Especially we need to keep Minchan's patch after the softlimit reclaim support.
On a system which over-commit the softlimit, it is easily to make the system hang
w/o it.
Signed-off-by: Ying Han <yinghan@xxxxxxxxxx>
---
mm/vmscan.c | 9 ++++++---
1 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5d036f5..65febc1 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2318,6 +2318,7 @@ static bool all_unreclaimable(struct zonelist *zonelist,
{
struct zoneref *z;
struct zone *zone;
+ bool all_unreclaimable = true;
for_each_zone_zonelist_nodemask(zone, z, zonelist,
gfp_zone(sc->gfp_mask), sc->nodemask) {
@@ -2325,11 +2326,13 @@ static bool all_unreclaimable(struct zonelist *zonelist,
continue;
if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL))
continue;
- if (!zone->all_unreclaimable)
- return false;
+ if (zone_reclaimable(zone)) {
+ all_unreclaimable = false;
+ break;
+ }
}
- return true;
+ return all_unreclaimable;
}
/*
--
1.7.7.3
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>