After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 440682 - gst/gstsystemclock unit test occasionally fails
gst/gstsystemclock unit test occasionally fails
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gstreamer (core)
git master
Other Linux
: High critical
: 0.10.13
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2007-05-23 12:23 UTC by Tim-Philipp Müller
Modified: 2007-05-24 15:03 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
gzipped debug log of ':waiting:test_periodic_shot:0: Waiting did not return OK' failure (57.68 KB, application/x-gzip)
2007-05-23 12:47 UTC, Tim-Philipp Müller
Details
gzipped debug log of 'test_async_order:0: Expected notification for id2 to come first' failure (130.72 KB, application/x-gzip)
2007-05-23 12:48 UTC, Tim-Philipp Müller
Details

Description Tim-Philipp Müller 2007-05-23 12:23:37 UTC
Running suite(s): GstSystemClock
85%: Checks: 7, Failures: 1, Errors: 0
gst/gstsystemclock.c:239:F:waiting:test_async_order:0: Missing notification for id1
FAIL: gst/gstsystemclock

Running suite(s): GstSystemClock
71%: Checks: 7, Failures: 2, Errors: 0
gst/gstsystemclock.c:179:F:waiting:test_periodic_shot:0: Waiting did not return OK
gst/gstsystemclock.c:237:F:waiting:test_async_order:0: Expected notification for id2 to come first
Comment 1 Tim-Philipp Müller 2007-05-23 12:47:00 UTC
Created attachment 88665 [details]
gzipped debug log of ':waiting:test_periodic_shot:0: Waiting did not return OK' failure
Comment 2 Tim-Philipp Müller 2007-05-23 12:48:04 UTC
Created attachment 88666 [details]
gzipped debug log of 'test_async_order:0: Expected notification for id2 to come first' failure
Comment 3 Wim Taymans 2007-05-23 13:29:51 UTC
Seems the deadline is missed (100ms) in the first case. increasing the TIME_UNIT might help to reduce this problem. We could also ignore GST_CLOCK_EARLY..
Comment 4 Wim Taymans 2007-05-23 13:45:00 UTC
the second test could be written better, patch follows..
Comment 5 Wim Taymans 2007-05-23 13:57:00 UTC
this patch should improve the test.

        * tests/check/gst/gstsystemclock.c: (store_callback),
        (GST_START_TEST):
        Improve test_async_order to also work when both timers are already
        expired when we get scheduled to check it.
Comment 6 Tim-Philipp Müller 2007-05-23 15:51:47 UTC
FWIW, it works fine for me now. One of the tests still fails occasionally under high cpu load, but that's to be expected for timing-sensitive stuff, I guess. The order test always passes for me now.
Comment 7 René Stadler 2007-05-23 17:15:18 UTC
Confirming that, too.  The tests ran here for almost three hours on a calm CPU without a single failure now.