The patch titled
Fix all_unreclaimable in memcg.
has been added to the -mm tree. Its filename is
vmscan-check-all_unreclaimable-in-direct-reclaim-path-fix.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find
out what to do about this
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: Fix all_unreclaimable in memcg.
From: Minchan Kim <minchan.kim@xxxxxxxxx>
vmscan-check-all_unreclaimable-in-direct-reclaim-path.patch has a problem.
It changed old behavior of memcg.
When memory pressure in memcg is high, do_try_to_free_pages returns 0. It
causes mem_cgroup_out_of_memory so that any process in mem group would be
killed.
But vmscan-check-all_unreclaimable-in-direct-reclaim-path.patch changed
the old behavior. It returns 1 unconditionally regardless of considering
global reclaim or memcg relcaim. It causes hang without triggering OOM in
case of memcg direct reclaim.
This patch fixes it.
Signed-off-by: Minchan Kim <minchan.kim@xxxxxxxxx>
Reported-by: <caiqian@xxxxxxxxxx>
Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Acked-by: Rik van Riel <riel@xxxxxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: Balbir Singh <balbir@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/vmscan.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff -puN mm/vmscan.c~vmscan-check-all_unreclaimable-in-direct-reclaim-path-fix mm/vmscan.c
--- a/mm/vmscan.c~vmscan-check-all_unreclaimable-in-direct-reclaim-path-fix
+++ a/mm/vmscan.c
@@ -1846,9 +1846,6 @@ static bool all_unreclaimable(struct zon
struct zone *zone;
bool all_unreclaimable = true;
- if (!scanning_global_lru(sc))
- return false;
-
for_each_zone_zonelist_nodemask(zone, z, zonelist,
gfp_zone(sc->gfp_mask), sc->nodemask) {
if (!populated_zone(zone))
@@ -1962,7 +1959,7 @@ out:
return sc->nr_reclaimed;
/* top priority shrink_zones still had more to do? don't OOM, then */
- if (!all_unreclaimable(zonelist, sc))
+ if (scanning_global_lru(sc) && !all_unreclaimable(zonelist, sc))
return 1;
return 0;
_
Patches currently in -mm which might be from minchan.kim@xxxxxxxxx are
oom-always-return-a-badness-score-of-non-zero-for-eligible-tasks.patch
vmscan-check-all_unreclaimable-in-direct-reclaim-path.patch
vmscan-check-all_unreclaimable-in-direct-reclaim-path-v2.patch
vmscan-check-all_unreclaimable-in-direct-reclaim-path-fix.patch
mm-vmap-area-cache.patch
mm-compaction-fix-compactpagefailed-counting.patch
vmscan-prevent-background-aging-of-anon-page-in-no-swap-system.patch
tracing-vmscan-add-trace-events-for-lru-list-shrinking.patch
writeback-account-for-time-spent-congestion_waited.patch
vmscan-synchronous-lumpy-reclaim-should-not-call-congestion_wait.patch
vmscan-narrow-the-scenarios-lumpy-reclaim-uses-synchrounous-reclaim.patch
vmscan-remove-dead-code-in-shrink_inactive_list.patch
vmscan-isolated_lru_pages-stop-neighbour-search-if-neighbour-cannot-be-isolated.patch
writeback-do-not-sleep-on-the-congestion-queue-if-there-are-no-congested-bdis.patch
writeback-do-not-sleep-on-the-congestion-queue-if-there-are-no-congested-bdis-or-if-significant-congestion-is-not-being-encountered-in-the-current-zone.patch
vmscantmpfs-treat-used-once-pages-on-tmpfs-as-used-once.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
