GNOME Bugzilla – Bug 500835
gst/gstobject unit test fails on alpha
Last modified: 2009-02-03 15:00:06 UTC
Hi, since we enabled to run the unit test suite on the Debian build servers there was a unit test failure on alpha: Running suite(s): GstObject 85%: Checks: 7, Failures: 1, Errors: 0 gst/gstobject.c:75:F:general:test_fail_abstract_new:0: Early exit with return value 0 FAIL: gst/gstobject This needs of course further investigation. Any thoughts on how this could happen? Maybe a bug in check? Apart from that there was only the known bug on ppc/s390 that is probably compiler related and probably not worth noting as there is another bug about this I was told: Running suite(s): data protocol 80%: Checks: 5, Failures: 1, Errors: 0 libs/gdp.c:150:F:general:tf:0: GST_BUFFER_IN_CAPS flag should have been copied ! FAIL: libs/gdp
The ppc/s390 failure is handled in bug #348114, can be ignored here :)
> Running suite(s): GstObject > 85%: Checks: 7, Failures: 1, Errors: 0 > gst/gstobject.c:75:F:general:test_fail_abstract_new:0: Early exit with return > value 0 > FAIL: gst/gstobject > > This needs of course further investigation. Any thoughts on how this could > happen? Maybe a bug in check? I've seen this before with empty or almost-empty test functions, but can't remember what the problem was. Do you have access to a test machine? Can you investigate? (In particular: does it also happen with CK_FORK=no?)
Hey Sebastian, are there any news on this one? I tend to close this bug, if you can't provide the necessary information, as this bug is NEEDINFO for quite some time now.
The alpha builds listed on http://experimental.debian.net/build.php?pkg=gstreamer0.10 don't show this problem any more afaict (the tee stress test timeout is probably due to the machine being busy/slow).
Yes, it's working now :)