+ cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function.patch added to -mm tree

* + cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function.patch added to -mm tree
@ 2022-04-22 20:20 Andrew Morton
  0 siblings, 0 replies; 2+ messages in thread
From: Andrew Morton @ 2022-04-22 20:20 UTC (permalink / raw)
  To: mm-commits, tj, shakeelb, roman.gushchin, mhocko, hannes, void, akpm


The patch titled
     Subject: cgroup: fix racy check in alloc_pagecache_max_30M() helper function
has been added to the -mm tree.  Its filename is
     cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function.patch

This patch should soon appear at
    https://ozlabs.org/~akpm/mmots/broken-out/cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function.patch
and later at
    https://ozlabs.org/~akpm/mmotm/broken-out/cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function.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/process/submit-checklist.rst when testing your code ***

The -mm tree is included into linux-next and is updated
there every 3-4 working days

------------------------------------------------------
From: David Vernet <void@manifault.com>
Subject: cgroup: fix racy check in alloc_pagecache_max_30M() helper function

alloc_pagecache_max_30M() in the cgroup memcg tests performs a 50MB
pagecache allocation, which it expects to be capped at 30MB due to the
calling process having a memory.high setting of 30MB.  After the
allocation, the function contains a check that verifies that MB(29) <
memory.current <= MB(30).  This check can actually fail
non-deterministically.

The testcases that use this function are test_memcg_high() and
test_memcg_max(), which set memory.min and memory.max to 30MB respectively
for the cgroup under test.  The allocation can slightly exceed this number
in both cases, and for memory.max, the process performing the allocation
will not have the OOM killer invoked as it's performing a pagecache
allocation.  This patchset therefore updates the above check to instead
use the verify_close() helper function.

Link: https://lkml.kernel.org/r/20220422155728.3055914-6-void@manifault.com
Signed-off-by: David Vernet <void@manifault.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 tools/testing/selftests/cgroup/test_memcontrol.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

--- a/tools/testing/selftests/cgroup/test_memcontrol.c~cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function
+++ a/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -564,9 +564,14 @@ static int alloc_pagecache_max_30M(const
 {
 	size_t size = MB(50);
 	int ret = -1;
-	long current;
+	long current, high, max;
 	int fd;
 
+	high = cg_read_long(cgroup, "memory.high");
+	max = cg_read_long(cgroup, "memory.max");
+	if (high != MB(30) && max != MB(30))
+		goto cleanup;
+
 	fd = get_temp_fd();
 	if (fd < 0)
 		return -1;
@@ -575,7 +580,7 @@ static int alloc_pagecache_max_30M(const
 		goto cleanup;
 
 	current = cg_read_long(cgroup, "memory.current");
-	if (current <= MB(29) || current > MB(30))
+	if (!values_close(current, MB(30), 5))
 		goto cleanup;
 
 	ret = 0;
_

Patches currently in -mm which might be from void@manifault.com are

cgroups-refactor-children-cgroups-in-memcg-tests.patch
cgroup-account-for-memory_recursiveprot-in-test_memcg_low.patch
cgroup-account-for-memory_localevents-in-test_memcg_oom_group_leaf_events.patch
cgroup-removing-racy-check-in-test_memcg_sock.patch
cgroup-fix-racy-check-in-alloc_pagecache_max_30m-helper-function.patch


^ permalink raw reply	[flat|nested] 2+ messages in thread