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.
Suggested-by: Lance Yang <lance.yang(a)linux.dev>
Signed-off-by: Guopeng Zhang <zhangguopeng(a)kylinos.cn>
---
v2:
- Mention the periodic rstat flush interval (FLUSH_TIME = 2*HZ) in
the comment and clarify the rationale for the 3s timeout.
- Replace the hard-coded retry count and wait interval with macros
to avoid magic numbers and make the 3s timeout calculation explicit.
---
.../selftests/cgroup/test_memcontrol.c | 30 ++++++++++++++++++-
1 file changed, 29 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
index 4e1647568c5b..7bea656658a2 100644
--- a/tools/testing/selftests/cgroup/test_memcontrol.c
+++ b/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -24,6 +24,9 @@
static bool has_localevents;
static bool has_recursiveprot;
+#define MEMCG_SOCKSTAT_WAIT_RETRIES 30 /* 3s total */
+#define MEMCG_SOCKSTAT_WAIT_INTERVAL_US (100 * 1000) /* 100 ms */
+
int get_temp_fd(void)
{
return open(".", O_TMPFILE | O_RDWR | O_EXCL);
@@ -1384,6 +1387,8 @@ 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;
+ int i;
memcg = cg_name(root, "memcg_test");
if (!memcg)
@@ -1432,7 +1437,30 @@ 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.
+ */
+ for (i = 0; i < MEMCG_SOCKSTAT_WAIT_RETRIES; i++) {
+ sock_post = cg_read_key_long(memcg, "memory.stat", "sock ");
+ if (sock_post < 0)
+ goto cleanup;
+
+ if (!sock_post)
+ break;
+
+ usleep(MEMCG_SOCKSTAT_WAIT_INTERVAL_US);
+ }
+
+ if (sock_post)
goto cleanup;
ret = KSFT_PASS;
--
2.25.1
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. 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(a)kylinos.cn>
---
.../selftests/cgroup/test_memcontrol.c | 24 ++++++++++++++++++-
1 file changed, 23 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
index 4e1647568c5b..86d9981cddd8 100644
--- a/tools/testing/selftests/cgroup/test_memcontrol.c
+++ b/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -1384,6 +1384,8 @@ 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;
+ int i, retries = 30;
memcg = cg_name(root, "memcg_test");
if (!memcg)
@@ -1432,7 +1434,27 @@ 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, so 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 and require that the
+ * "sock " counter eventually drops to zero.
+ */
+ for (i = 0; i < retries; i++) {
+ sock_post = cg_read_key_long(memcg, "memory.stat", "sock ");
+ if (sock_post < 0)
+ goto cleanup;
+
+ if (!sock_post)
+ break;
+
+ usleep(100 * 1000); /* 100ms */
+ }
+
+ if (sock_post)
goto cleanup;
ret = KSFT_PASS;
--
2.25.1
Here are various unrelated fixes:
- Patch 1: Fix window space computation for fallback connections which
can affect ACK generation. A fix for v5.11.
- Patch 2: Avoid unneeded subflow-level drops due to unsynced received
window. A fix for v5.11.
- Patch 3: Avoid premature close for fallback connections with PREEMPT
kernels. A fix for v5.12.
- Patch 4: Reset instead of fallback in case of data in the MPTCP
out-of-order queue. A fix for v5.7.
- Patches 5-7: Avoid also sending "plain" TCP reset when closing with an
MP_FASTCLOSE. A fix for v6.1.
- Patches 8-9: Longer timeout for background connections in MPTCP Join
selftests. An additional fix for recent patches for v5.13/v6.1.
- Patches 10-11: Fix typo in a check introduce in a recent refactoring.
A fix for v6.15.
Signed-off-by: Matthieu Baerts (NGI0) <matttbe(a)kernel.org>
---
Gang Yan (2):
mptcp: fix address removal logic in mptcp_pm_nl_rm_addr
selftests: mptcp: add a check for 'add_addr_accepted'
Matthieu Baerts (NGI0) (3):
selftests: mptcp: join: fastclose: remove flaky marks
selftests: mptcp: join: endpoints: longer timeout
selftests: mptcp: join: userspace: longer timeout
Paolo Abeni (6):
mptcp: fix ack generation for fallback msk
mptcp: avoid unneeded subflow-level drops
mptcp: fix premature close in case of fallback
mptcp: do not fallback when OoO is present
mptcp: decouple mptcp fastclose from tcp close
mptcp: fix duplicate reset on fastclose
net/mptcp/options.c | 54 +++++++++++++++++++++-
net/mptcp/pm_kernel.c | 2 +-
net/mptcp/protocol.c | 59 +++++++++++++++++--------
net/mptcp/protocol.h | 3 +-
tools/testing/selftests/net/mptcp/mptcp_join.sh | 27 ++++++-----
5 files changed, 113 insertions(+), 32 deletions(-)
---
base-commit: 8e0a754b0836d996802713bbebc87bc1cc17925c
change-id: 20251117-net-mptcp-misc-fixes-6-18-rc6-835d94cdc095
Best regards,
--
Matthieu Baerts (NGI0) <matttbe(a)kernel.org>