https://bugs.winehq.org/show_bug.cgi?id=55110
Bug ID: 55110 Summary: quartz:vmr7 & quartz:vmr9 sometimes exit prematurely in Wine Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: quartz Assignee: wine-bugs@winehq.org Reporter: fgouget@codeweavers.com Distribution: ---
quartz:vmr7 & quartz:vmr9 sometimes exit prematurely in Wine:
vmr7.c:2393: Test marked todo: Got hr 0. vmr7.c:2424: Test marked todo: Got hr 0x8004020a. quartz:vmr7:0890 done (0) in 4s 2250B
Notice the missing ' xxx tests executed' line. And also in the Wine traces:
01e4:fixme:d3d:wined3d_guess_card No card selector available for card vendor 0000 (using GL_RENDERER "llvmpipe (LLVM 11.0.1, 256 bits)"). X Error of failed request: BadWindow (invalid Window parameter) Major opcode of failed request: 7 (X_ReparentWindow) Resource id in failed request: 0x140019d Serial number of failed request: 2252 Current serial number in output stream: 2253
See https://test.winehq.org/data/patterns.html#quartz:vmr7 https://test.winehq.org/data/patterns.html#quartz:vmr9
So it is most likely the X error that causes the test to be prematurely killed. I did not notice an X error for quartz:vmr9 but for now I will consider that it either has the same cause or is a consequence of the vmr7 failure.
The first failure happened on 2023-06-13 and, across all test configurations, it has been happening about once a day since.
Further, a bisect shows that the vmr7 failure started with the commit below:
commit 9f8d9eef216d4ede3b53d3caab222ad8f896f10b Author: Rémi Bernon rbernon@codeweavers.com Date: Mon Jan 25 15:22:08 2021 +0100
winex11: Set the window cursor immediately in SetCursor.
Instead of synchronizing the cursor on mouse changes. The SetCursor call should now only be made when cursor or window handle have changed.
https://bugs.winehq.org/show_bug.cgi?id=55110
François Gouget fgouget@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |9f8d9eef216d4ede3b53d3caab2 | |22ad8f896f10b Keywords| |regression, source, | |testcase
https://bugs.winehq.org/show_bug.cgi?id=55110
Rémi Bernon rbernon@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |rbernon@codeweavers.com
--- Comment #1 from Rémi Bernon rbernon@codeweavers.com --- Created attachment 74674 --> https://bugs.winehq.org/attachment.cgi?id=74674 Possible fix
I'm not able to reproduce that exact error, but I have been able to trigger X errors when running this test. I believe there's some race conditions between X displays and client / whole windows requests. The attached patch tries to mitigate this, and as it's also about window parents I think it might help?
https://bugs.winehq.org/show_bug.cgi?id=55110
--- Comment #2 from Rémi Bernon rbernon@codeweavers.com --- Fwiw the relation with the mentioned commit is likely an indirect effect of the reduced number of X requests from the cursor sync change.
https://bugs.winehq.org/show_bug.cgi?id=55110
Rémi Bernon rbernon@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |a6bc5f34b87393e04ff46659f51 | |8f2e7094cc7e4 Status|NEW |RESOLVED Resolution|--- |FIXED
--- Comment #3 from Rémi Bernon rbernon@codeweavers.com --- I believe this should be fixed after a6bc5f34b87393e04ff46659f518f2e7094cc7e4 though it's only been two days without failures, so feel free to reopen if not.
https://bugs.winehq.org/show_bug.cgi?id=55110
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #4 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 8.12.
-
WineHQ Bugzilla