diff options
| author | joth@chromium.org <joth@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-06-09 09:36:54 +0000 |
|---|---|---|
| committer | joth@chromium.org <joth@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2010-06-09 09:36:54 +0000 |
| commit | 5946850109f6eaece3c402b4073836c33bbac23b (patch) | |
| tree | ffb5e0f665d04bef0ad91d5e0f0fc2e04e5fde31 | |
| parent | 38a8cd88f853d0a30e6537db10b12a07998728a8 (diff) | |
| download | chromium_src-5946850109f6eaece3c402b4073836c33bbac23b.zip chromium_src-5946850109f6eaece3c402b4073836c33bbac23b.tar.gz chromium_src-5946850109f6eaece3c402b4073836c33bbac23b.tar.bz2 | |
Revert 48102 - this test is now failing again, apparently since r49188 landed. As it is not clear this test was enabled due to the functionality it test being required (indeed, the comment still states it is not), disabling it again to get the tree open.
Original description from reverted change r48102:
TTF: Re-enable the FileDescriptorWatcherOutlivesMessageLoop test as it now appears
to be passing.
BUG=none
TEST=MessageLoopTest.FileDescriptorWatcherOutlivesMessageLoop
Patch by Doug Muir <dmuir@google.com>.
Review URL: http://codereview.chromium.org/2077014
TBR=jhawkins@chromium.org
Review URL: http://codereview.chromium.org/2789002
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@49241 0039d316-1c4b-4281-b951-d872f2087c98
| -rw-r--r-- | base/message_loop_unittest.cc | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/base/message_loop_unittest.cc b/base/message_loop_unittest.cc index 44d545f..de039f2 100644 --- a/base/message_loop_unittest.cc +++ b/base/message_loop_unittest.cc @@ -1492,11 +1492,12 @@ class QuitDelegate : public } // namespace -TEST(MessageLoopTest, FileDescriptorWatcherOutlivesMessageLoop) { +TEST(MessageLoopTest, DISABLED_FileDescriptorWatcherOutlivesMessageLoop) { // Simulate a MessageLoop that dies before an FileDescriptorWatcher. // This could happen when people use the Singleton pattern or atexit. // This is disabled for now because it fails (valgrind shows // invalid reads), and it's not clear any code relies on this... + // TODO(dkegel): enable if it turns out we rely on this // Create a file descriptor. Doesn't need to be readable or writable, // as we don't need to actually get any notifications. |