diff options
| author | Zach Koopmans <zkoopmans@google.com> | 2019-07-29 16:46:28 -0700 |
|---|---|---|
| committer | gVisor bot <gvisor-bot@google.com> | 2019-07-29 16:47:58 -0700 |
| commit | f0507e1db1574ff165000fa5e34b651413f37aae (patch) | |
| tree | 5ab5d1e9ccef86134feab06ecf64747faca7eb78 /test/syscalls | |
| parent | 8e8b6096116eb490b438a53061195f737d4eca8b (diff) | |
Fix flaky stat.cc test.
This test flaked on my current CL. Linux makes no guarantee
that two inodes will consecutive (overflows happen).
https://github.com/avagin/linux-task-diag/blob/master/fs/inode.c#L880
PiperOrigin-RevId: 260608240
Diffstat (limited to 'test/syscalls')
| -rw-r--r-- | test/syscalls/linux/stat.cc | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/test/syscalls/linux/stat.cc b/test/syscalls/linux/stat.cc index 510f7bee5..88ab90b5b 100644 --- a/test/syscalls/linux/stat.cc +++ b/test/syscalls/linux/stat.cc @@ -539,9 +539,8 @@ TEST(SimpleStatTest, AnonDeviceAllocatesUniqueInodesAcrossSaveRestore) { ASSERT_THAT(fstat(fd1.get(), &st1), SyscallSucceeds()); ASSERT_THAT(fstat(fd2.get(), &st2), SyscallSucceeds()); - // The two fds should have different inode numbers. Specifically, since fd2 - // was created later, it should have a higher inode number. - EXPECT_GT(st2.st_ino, st1.st_ino); + // The two fds should have different inode numbers. + EXPECT_NE(st2.st_ino, st1.st_ino); // Verify again after another S/R cycle. The inode numbers should remain the // same. |