test_memcg_sock() currently requires that memory.stat's "sock " counter
is exactly zero immediately after the TCP server exits. On a busy system
this assumption is too strict:
- Socket memory may be freed with a small delay (e.g. RCU callbacks).
- memcg statistics are updated asynchronously via the rstat flushing
worker, so the "sock " value in memory.stat can stay non-zero for a
short period of time even after all socket memory has been uncharged.
As a result, test_memcg_sock() can intermittently fail even though socket
memory accounting is working correctly.
Make the test more robust by polling memory.stat for the "sock "
counter and allowing it some time to drop to zero instead of checking
it only once. The timeout is set to 3 seconds to cover the periodic
rstat flush interval (FLUSH_TIME = 2*HZ by default) plus some
scheduling slack. If the counter does not become zero within the
timeout, the test still fails as before.
On my test system, running test_memcontrol 50 times produced:
- Before this patch: 6/50 runs passed.
- After this patch: 50/50 runs passed.
Signed-off-by: Guopeng Zhang <zhangguopeng@kylinos.cn>
Suggested-by: Lance Yang <lance.yang@linux.dev>
---
.../selftests/cgroup/test_memcontrol.c | 20 ++++++++++++++++++-
1 file changed, 19 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
index 4e1647568c5b..dda12e5c6457 100644
--- a/tools/testing/selftests/cgroup/test_memcontrol.c
+++ b/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -21,6 +21,8 @@
#include "kselftest.h"
#include "cgroup_util.h"
+#define MEMCG_SOCKSTAT_WAIT_RETRIES 30 /* 3s total */
+
static bool has_localevents;
static bool has_recursiveprot;
@@ -1384,6 +1386,7 @@ static int test_memcg_sock(const char *root)
int bind_retries = 5, ret = KSFT_FAIL, pid, err;
unsigned short port;
char *memcg;
+ long sock_post = -1;
memcg = cg_name(root, "memcg_test");
if (!memcg)
@@ -1432,7 +1435,22 @@ static int test_memcg_sock(const char *root)
if (cg_read_long(memcg, "memory.current") < 0)
goto cleanup;
- if (cg_read_key_long(memcg, "memory.stat", "sock "))
+ /*
+ * memory.stat is updated asynchronously via the memcg rstat
+ * flushing worker, which runs periodically (every 2 seconds,
+ * see FLUSH_TIME). On a busy system, the "sock " counter may
+ * stay non-zero for a short period of time after the TCP
+ * connection is closed and all socket memory has been
+ * uncharged.
+ *
+ * Poll memory.stat for up to 3 seconds (~FLUSH_TIME plus some
+ * scheduling slack) and require that the "sock " counter
+ * eventually drops to zero.
+ */
+ sock_post = cg_read_key_long_poll(memcg, "memory.stat", "sock ", 0,
+ MEMCG_SOCKSTAT_WAIT_RETRIES,
+ DEFAULT_WAIT_INTERVAL_US);
+ if (sock_post)
goto cleanup;
ret = KSFT_PASS;
--
2.25.1On 2025/11/24 20:38, Guopeng Zhang wrote: > test_memcg_sock() currently requires that memory.stat's "sock " counter > is exactly zero immediately after the TCP server exits. On a busy system > this assumption is too strict: > > - Socket memory may be freed with a small delay (e.g. RCU callbacks). > - memcg statistics are updated asynchronously via the rstat flushing > worker, so the "sock " value in memory.stat can stay non-zero for a > short period of time even after all socket memory has been uncharged. > > As a result, test_memcg_sock() can intermittently fail even though socket > memory accounting is working correctly. > > Make the test more robust by polling memory.stat for the "sock " > counter and allowing it some time to drop to zero instead of checking > it only once. The timeout is set to 3 seconds to cover the periodic > rstat flush interval (FLUSH_TIME = 2*HZ by default) plus some > scheduling slack. If the counter does not become zero within the > timeout, the test still fails as before. > > On my test system, running test_memcontrol 50 times produced: > > - Before this patch: 6/50 runs passed. > - After this patch: 50/50 runs passed. > > Signed-off-by: Guopeng Zhang <zhangguopeng@kylinos.cn> > Suggested-by: Lance Yang <lance.yang@linux.dev> > --- > .../selftests/cgroup/test_memcontrol.c | 20 ++++++++++++++++++- > 1 file changed, 19 insertions(+), 1 deletion(-) > > diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c > index 4e1647568c5b..dda12e5c6457 100644 > --- a/tools/testing/selftests/cgroup/test_memcontrol.c > +++ b/tools/testing/selftests/cgroup/test_memcontrol.c > @@ -21,6 +21,8 @@ > #include "kselftest.h" This patch fails to apply to mm-new ... Hmm, it expects #include "kselftest.h" here, but the tree uses #include "../kselftest.h". Which is odd, as that line hasn't been touched in years ...
On 11/27/25 18:55, Lance Yang wrote:
>
>
> On 2025/11/24 20:38, Guopeng Zhang wrote:
>> test_memcg_sock() currently requires that memory.stat's "sock " counter
>> is exactly zero immediately after the TCP server exits. On a busy system
>> this assumption is too strict:
>>
>> - Socket memory may be freed with a small delay (e.g. RCU callbacks).
>> - memcg statistics are updated asynchronously via the rstat flushing
>> worker, so the "sock " value in memory.stat can stay non-zero for a
>> short period of time even after all socket memory has been uncharged.
>>
>> As a result, test_memcg_sock() can intermittently fail even though socket
>> memory accounting is working correctly.
>>
>> Make the test more robust by polling memory.stat for the "sock "
>> counter and allowing it some time to drop to zero instead of checking
>> it only once. The timeout is set to 3 seconds to cover the periodic
>> rstat flush interval (FLUSH_TIME = 2*HZ by default) plus some
>> scheduling slack. If the counter does not become zero within the
>> timeout, the test still fails as before.
>>
>> On my test system, running test_memcontrol 50 times produced:
>>
>> - Before this patch: 6/50 runs passed.
>> - After this patch: 50/50 runs passed.
>>
>> Signed-off-by: Guopeng Zhang <zhangguopeng@kylinos.cn>
>> Suggested-by: Lance Yang <lance.yang@linux.dev>
>> ---
>> .../selftests/cgroup/test_memcontrol.c | 20 ++++++++++++++++++-
>> 1 file changed, 19 insertions(+), 1 deletion(-)
>>
>> diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
>> index 4e1647568c5b..dda12e5c6457 100644
>> --- a/tools/testing/selftests/cgroup/test_memcontrol.c
>> +++ b/tools/testing/selftests/cgroup/test_memcontrol.c
>> @@ -21,6 +21,8 @@
>> #include "kselftest.h"
>
> This patch fails to apply to mm-new ...
>
> Hmm, it expects #include "kselftest.h" here, but the tree uses
> #include "../kselftest.h".
>
> Which is odd, as that line hasn't been touched in years ...
Hi,lance
Thanks for your review.
When I prepared this patch I was working on linux-next, where
tools/testing/selftests/cgroup/test_memcontrol.c already uses:
#include "kselftest.h"
I just checked, and this change comes from the following commit:
1aaedc385b9b278dcf91f4e9d0c3e1a078804ff1
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20251127&id=1aaedc385b9b278dcf91f4e9d0c3e1a078804ff1
So the patch applies cleanly on top of the latest linux-next, but not on
mm-new which still has `#include "../kselftest.h"`.
Thanks,
Guopeng
On 2025/11/27 19:18, Guopeng Zhang wrote: > > > On 11/27/25 18:55, Lance Yang wrote: >> >> >> On 2025/11/24 20:38, Guopeng Zhang wrote: >>> test_memcg_sock() currently requires that memory.stat's "sock " counter >>> is exactly zero immediately after the TCP server exits. On a busy system >>> this assumption is too strict: >>> >>> - Socket memory may be freed with a small delay (e.g. RCU callbacks). >>> - memcg statistics are updated asynchronously via the rstat flushing >>> worker, so the "sock " value in memory.stat can stay non-zero for a >>> short period of time even after all socket memory has been uncharged. >>> >>> As a result, test_memcg_sock() can intermittently fail even though socket >>> memory accounting is working correctly. >>> >>> Make the test more robust by polling memory.stat for the "sock " >>> counter and allowing it some time to drop to zero instead of checking >>> it only once. The timeout is set to 3 seconds to cover the periodic >>> rstat flush interval (FLUSH_TIME = 2*HZ by default) plus some >>> scheduling slack. If the counter does not become zero within the >>> timeout, the test still fails as before. >>> >>> On my test system, running test_memcontrol 50 times produced: >>> >>> - Before this patch: 6/50 runs passed. >>> - After this patch: 50/50 runs passed. >>> >>> Signed-off-by: Guopeng Zhang <zhangguopeng@kylinos.cn> >>> Suggested-by: Lance Yang <lance.yang@linux.dev> >>> --- >>> .../selftests/cgroup/test_memcontrol.c | 20 ++++++++++++++++++- >>> 1 file changed, 19 insertions(+), 1 deletion(-) >>> >>> diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c >>> index 4e1647568c5b..dda12e5c6457 100644 >>> --- a/tools/testing/selftests/cgroup/test_memcontrol.c >>> +++ b/tools/testing/selftests/cgroup/test_memcontrol.c >>> @@ -21,6 +21,8 @@ >>> #include "kselftest.h" >> >> This patch fails to apply to mm-new ... >> >> Hmm, it expects #include "kselftest.h" here, but the tree uses >> #include "../kselftest.h". >> >> Which is odd, as that line hasn't been touched in years ... > Hi,lance > > Thanks for your review. > > When I prepared this patch I was working on linux-next, where > tools/testing/selftests/cgroup/test_memcontrol.c already uses: > > #include "kselftest.h" > > I just checked, and this change comes from the following commit: > > 1aaedc385b9b278dcf91f4e9d0c3e1a078804ff1 > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20251127&id=1aaedc385b9b278dcf91f4e9d0c3e1a078804ff1 > > So the patch applies cleanly on top of the latest linux-next, but not on > mm-new which still has `#include "../kselftest.h"`. Ahh, I see, thanks!
© 2016 - 2025 Red Hat, Inc.