https://bugs.winehq.org/show_bug.cgi?id=47876
Bug ID: 47876 Summary: mmdevapi:capture issues an invalid and fake todo Product: Wine Version: 4.17 Hardware: x86 OS: Windows Status: NEW Severity: normal Priority: P2 Component: mmdevapi Assignee: wine-bugs@winehq.org Reporter: fgouget@codeweavers.com
mmdevapi:capture issues the following todo on Windows:
capture.c:395: Test marked todo: only PulseAudio gets here
https://test.winehq.org/data/tests/mmdevapi:capture.html (look for +1 lines)
First tests should only be marked todo on Wine so far so there should be no todo on Windows. Second this line is generated by a trace():
trace("Test marked todo: only PulseAudio gets here\n");
So this is not a todo test at all but a simple trace. The problem is that it looks like a real 'todo_wine ok()' test and is counted as such by test.winehq.org and the TestBot, which both then complain that the todo test tally does not match the result at the end:
15bc:capture: 255 tests executed (0 marked as todo, 6 failures), 0 skipped. The test has unaccounted for todo messages
Finally, the message says it is only issued when running with PulseAudio. PulseAudio is not available on Windows so obviously the tests that lead there are wrong.
This issue happens randomly on a variety of TestBot Windows 7+ VMs, both in 32 and 64 bit tests: * cw-rx460 both when running Windows 8.1 and 10 1809 * cw1-hd6800 Windows 10 1507 & 1709 & 1809 * cw2-gtx560 when running Windows 10 1809 * w7pro64 * w1064v1507 and w1064v1809
This message was added by the following commit:
commit f24b31fcf2047e8b4d0d8b910c54c0815c54d1cf Author: Jörg Höhle hoehle@users.sourceforge.net AuthorDate: Thu Jan 12 22:36:31 2012 +0100 Commit: Alexandre Julliard julliard@winehq.org CommitDate: Thu Jan 26 21:27:54 2012 +0100
mmdevapi/tests: Perform capture protocol tests in shared mode.
https://bugs.winehq.org/show_bug.cgi?id=47876
Vijay Kamuju infyquest@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |infyquest@gmail.com
--- Comment #1 from Vijay Kamuju infyquest@gmail.com --- patch sent https://source.winehq.org/patches/data/171485
https://bugs.winehq.org/show_bug.cgi?id=47876
Andrew Eikum aeikum@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |aeikum@codeweavers.com
--- Comment #2 from Andrew Eikum aeikum@codeweavers.com --- Created attachment 65464 --> https://bugs.winehq.org/attachment.cgi?id=65464 mmdevapi/tests: Don't depend on having no data in the buffer after Reset
The test predates the winepulse driver, and also looks a little race-y. It depends on Start() and GetBuffer() happening so quickly that there is less than one period in the buffer. Which is probably always the case, but I'm not sure it's useful to depend on that.
I suggest a patch like this. It succeeds on both winepulse and winealsa for me.
https://bugs.winehq.org/show_bug.cgi?id=47876
--- Comment #3 from Andrew Eikum aeikum@codeweavers.com --- Actually re-reading the tests and the test failures, this will still fail later on the 2nd GetBuffer call. Maybe we should just remove this whole test sequence (everything from that last Reset onwards). I'm not sure what it is trying to prove.
https://bugs.winehq.org/show_bug.cgi?id=47876
--- Comment #4 from Andrew Eikum aeikum@codeweavers.com --- I believe this should be fixed by ba340e741ab888a567e3e78716a93da3bbc9c96f. I also cleaned up a bunch of other failures in the following commits.
https://bugs.winehq.org/show_bug.cgi?id=47876
--- Comment #5 from Andrew Eikum aeikum@codeweavers.com --- François, ok to close this?
https://bugs.winehq.org/show_bug.cgi?id=47876
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED Keywords| |source, testcase
--- Comment #6 from François Gouget fgouget@codeweavers.com --- Yes. There are still failures on some platforms but the pseudo-todos are gone and I'll trust you on the fix details. Thanks.
https://bugs.winehq.org/show_bug.cgi?id=47876
Gijs Vermeulen gijsvrm@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |ba340e741ab888a567e3e78716a | |93da3bbc9c96f
https://bugs.winehq.org/show_bug.cgi?id=47876
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #7 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 4.20.