diff options
author | Dean Deng <deandeng@google.com> | 2020-04-06 09:50:13 -0700 |
---|---|---|
committer | gVisor bot <gvisor-bot@google.com> | 2020-04-06 09:51:26 -0700 |
commit | 4baa7e70795edbb350d55a9365807341515d3af4 (patch) | |
tree | c1f482c75e21e225277e3cd538922dd299c2ef61 | |
parent | 00d9776a4bb1cc1d7125af7d3e54a939a4f3847a (diff) |
Bump up acceptable sample count for flaky itimer test.
Running the test 1000x almost always produces 1+ test failures where
the sample count is slightly more than 60.
PiperOrigin-RevId: 305051754
-rw-r--r-- | test/syscalls/linux/itimer.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/test/syscalls/linux/itimer.cc b/test/syscalls/linux/itimer.cc index 8b48f0804..dd981a278 100644 --- a/test/syscalls/linux/itimer.cc +++ b/test/syscalls/linux/itimer.cc @@ -246,7 +246,7 @@ int TestSIGPROFFairness(absl::Duration sleep) { // The number of samples on the main thread should be very low as it did // nothing. - TEST_CHECK(result.main_thread_samples < 60); + TEST_CHECK(result.main_thread_samples < 80); // Both workers should get roughly equal number of samples. TEST_CHECK(result.worker_samples.size() == 2); |