[085/163] mm: memcontrol: avoid workload stalls when lowering memory.high

Commit Message

Andrew Morton Aug. 7, 2020, 6:21 a.m. UTC

From: Roman Gushchin <[e-mail]>
Subject: mm: memcontrol: avoid workload stalls when lowering memory.high

Memory.high limit is implemented in a way such that the kernel penalizes
all threads which are allocating a memory over the limit.  Forcing all
threads into the synchronous reclaim and adding some artificial delays
allows to slow down the memory consumption and potentially give some time
for userspace oom handlers/resource control agents to react.

It works nicely if the memory usage is hitting the limit from below,
however it works sub-optimal if a user adjusts memory.high to a value way
below the current memory usage.  It basically forces all workload threads
(doing any memory allocations) into the synchronous reclaim and sleep. 
This makes the workload completely unresponsive for a long period of time
and can also lead to a system-wide contention on lru locks.  It can happen
even if the workload is not actually tight on memory and has, for example,
a ton of cold pagecache.

In the current implementation writing to memory.high causes an atomic
update of page counter's high value followed by an attempt to reclaim
enough memory to fit into the new limit.  To fix the problem described
above, all we need is to change the order of execution: try to push the
memory usage under the limit first, and only then set the new high limit.

Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Roman Gushchin <[e-mail]>
Reported-by: Domas Mituzas <[e-mail]>
Acked-by: Michal Hocko <[e-mail]>
Reviewed-by: Shakeel Butt <[e-mail]>
Cc: Johannes Weiner <[e-mail]>
Cc: Tejun Heo <[e-mail]>
Cc: Chris Down <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
---

 mm/memcontrol.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Patch

--- a/mm/memcontrol.c~mm-memcontrol-avoid-workload-stalls-when-lowering-memoryhigh
+++ a/mm/memcontrol.c
@@ -6213,8 +6213,6 @@  static ssize_t memory_high_write(struct
 	if (err)
 		return err;
 
-	page_counter_set_high(&memcg->memory, high);
-
 	for (;;) {
 		unsigned long nr_pages = page_counter_read(&memcg->memory);
 		unsigned long reclaimed;
@@ -6238,6 +6236,8 @@  static ssize_t memory_high_write(struct
 			break;
 	}
 
+	page_counter_set_high(&memcg->memory, high);
+
 	return nbytes;
 }