[LTP] [PATCH] mem/lib: Skip oom(KSM) if child_alloc() makes single TESTMEM+MB allocation

[LTP] [PATCH] mem/lib: Skip oom(KSM) if child_alloc() makes single TESTMEM+MB allocation

[Li Wang]

We occasionally catch errors like:
oom03       0  TINFO  :  start OOM testing for KSM pages.
oom03       0  TINFO  :  expected victim is 3490.
oom03       6  TFAIL  :  mem.c:163: victim unexpectedly ended with retcode: 0, expected: 12
oom03       0  TINFO  :  set overcommit_memory to 0

It cames from the caller testoom(0, 1, ENOMEM, 1). The issue was occurred
since child_alloc() go into single(lite == 1) thread mode but successfully
finish the memory(TESTMEM+MB) allocation and return 0. That's not what we
expected because oom03 has already set 'limit_in_byte = TESTMEM' in cgroup.
A probably reason is that KSM scan merged samepages and disturbed the OOM
test.

By this patch, Let's skip the oom(KSM) test in cgroup.

Signed-off-by: Li Wang <liwang@redhat.com>
---
 testcases/kernel/mem/lib/mem.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/testcases/kernel/mem/lib/mem.c b/testcases/kernel/mem/lib/mem.c
index 3d853a3..10eb0f7 100644
--- a/testcases/kernel/mem/lib/mem.c
+++ b/testcases/kernel/mem/lib/mem.c
@@ -224,8 +224,10 @@ void testoom(int mempolicy, int lite, int retcode, int allow_sigkill)
 	tst_resm(TINFO, "start OOM testing for mlocked pages.");
 	oom(MLOCK, lite, retcode, allow_sigkill);
 
-	if (access(PATH_KSM, F_OK) == -1) {
-		tst_resm(TINFO, "KSM configuration is not enabled, "
+	/* Skip oom(KSM) if lite == 1, since limit_in_bytes may vary from
+	 * run to run, which isn't reliable for oom03 cgroup test.*/
+	if (access(PATH_KSM, F_OK) == -1 || lite == 1) {
+		tst_resm(TINFO, "KSM is not configed or lite == 1, "
 			 "skip OOM test for KSM pags");
 	} else {
 		tst_resm(TINFO, "start OOM testing for KSM pages.");
-- 
1.8.3.1

[LTP] [PATCH] mem/lib: Skip oom(KSM) if child_alloc() makes single TESTMEM+MB allocation

[Jan Stancek]

----- Original Message -----
> From: "Li Wang" <liwang@redhat.com>
> To: jstancek@redhat.com
> Cc: ltp@lists.linux.it
> Sent: Tuesday, 21 June, 2016 12:54:43 PM
> Subject: [PATCH] mem/lib: Skip oom(KSM) if child_alloc() makes single TESTMEM+MB allocation
> 

I reworded the description a little and pushed.

Thanks,
Jan