diff mbox series

[06/19] mm: memcontrol: prevent starvation when writing memory.high

Message ID 20210124050119.34lm1Gw1Q%akpm@linux-foundation.org (mailing list archive)
State New, archived
Headers show
Series [01/19] x86/setup: don't remove E820_TYPE_RAM for pfn 0 | expand

Commit Message

Andrew Morton Jan. 24, 2021, 5:01 a.m. UTC 
From: Johannes Weiner <hannes@cmpxchg.org>
Subject: mm: memcontrol: prevent starvation when writing memory.high

When a value is written to a cgroup's memory.high control file, the
write() context first tries to reclaim the cgroup to size before putting
the limit in place for the workload.  Concurrent charges from the workload
can keep such a write() looping in reclaim indefinitely.

In the past, a write to memory.high would first put the limit in place for
the workload, then do targeted reclaim until the new limit has been met -
similar to how we do it for memory.max.  This wasn't prone to the
described starvation issue.  However, this sequence could cause excessive
latencies in the workload, when allocating threads could be put into long
penalty sleeps on the sudden memory.high overage created by the write(),
before that had a chance to work it off.

Now that memory_high_write() performs reclaim before enforcing the new
limit, reflect that the cgroup may well fail to converge due to concurrent
workload activity.  Bail out of the loop after a few tries.

Link: https://lkml.kernel.org/r/20210112163011.127833-1-hannes@cmpxchg.org
Fixes: 536d3bf261a2 ("mm: memcontrol: avoid workload stalls when lowering memory.high")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Reported-by: Tejun Heo <tj@kernel.org>
Acked-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>	[5.8+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/memcontrol.c |    7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

Comments

Shakeel Butt Jan. 24, 2021, 6:01 p.m. UTC | #1 
On Sat, Jan 23, 2021 at 9:01 PM Andrew Morton <akpm@linux-foundation.org> wrote:
>
> From: Johannes Weiner <hannes@cmpxchg.org>
> Subject: mm: memcontrol: prevent starvation when writing memory.high
>
> When a value is written to a cgroup's memory.high control file, the
> write() context first tries to reclaim the cgroup to size before putting
> the limit in place for the workload.  Concurrent charges from the workload
> can keep such a write() looping in reclaim indefinitely.
>
> In the past, a write to memory.high would first put the limit in place for
> the workload, then do targeted reclaim until the new limit has been met -
> similar to how we do it for memory.max.  This wasn't prone to the
> described starvation issue.  However, this sequence could cause excessive
> latencies in the workload, when allocating threads could be put into long
> penalty sleeps on the sudden memory.high overage created by the write(),
> before that had a chance to work it off.
>
> Now that memory_high_write() performs reclaim before enforcing the new
> limit, reflect that the cgroup may well fail to converge due to concurrent
> workload activity.  Bail out of the loop after a few tries.
>
> Link: https://lkml.kernel.org/r/20210112163011.127833-1-hannes@cmpxchg.org
> Fixes: 536d3bf261a2 ("mm: memcontrol: avoid workload stalls when lowering memory.high")
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
> Reviewed-by: Shakeel Butt <shakeelb@google.com>
> Reported-by: Tejun Heo <tj@kernel.org>
> Acked-by: Roman Gushchin <guro@fb.com>
> Reviewed-by: Michal Koutný <mkoutny@suse.com>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: <stable@vger.kernel.org>    [5.8+]
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

Johannes requested to replace this patch with
https://lore.kernel.org/linux-mm/20210122184341.292461-1-hannes@cmpxchg.org/
Linus Torvalds Jan. 24, 2021, 6:35 p.m. UTC | #2 
On Sun, Jan 24, 2021 at 10:02 AM Shakeel Butt <shakeelb@google.com> wrote:
>
> Johannes requested to replace this patch with
> https://lore.kernel.org/linux-mm/20210122184341.292461-1-hannes@cmpxchg.org/

I've dropped it (not replaced it - will wait for Andrew to
comment/send) from my queue.

          Linus
diff mbox series

Patch

--- a/mm/memcontrol.c~mm-memcontrol-prevent-starvation-when-writing-memoryhigh
+++ a/mm/memcontrol.c
@@ -6273,7 +6273,6 @@  static ssize_t memory_high_write(struct
 
 	for (;;) {
 		unsigned long nr_pages = page_counter_read(&memcg->memory);
-		unsigned long reclaimed;
 
 		if (nr_pages <= high)
 			break;
@@ -6287,10 +6286,10 @@  static ssize_t memory_high_write(struct
 			continue;
 		}
 
-		reclaimed = try_to_free_mem_cgroup_pages(memcg, nr_pages - high,
-							 GFP_KERNEL, true);
+		try_to_free_mem_cgroup_pages(memcg, nr_pages - high,
+					     GFP_KERNEL, true);
 
-		if (!reclaimed && !nr_retries--)
+		if (!nr_retries--)
 			break;
 	}