http://bugs.winehq.org/show_bug.cgi?id=28198
Summary: ddraw produces an error in glibc: double free Product: Wine Version: 1.3.27 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-ddraw AssignedTo: wine-bugs@winehq.org ReportedBy: toralf.foerster@gmx.de
git bisect blames
commit a04f0e9c789295c22aab6e0d3b3239d0c794b304 Author: Henri Verbeet hverbeet@codeweavers.com Date: Sun Aug 21 21:35:00 2011 +0200
wined3d: Move FBO based color blits from IWineD3DSurfaceImpl_BltOverride() to wined3d_surface_blt().
for this bug :
$> ../../../tools/runtest -q -P wine -M ddraw.dll -T ../../.. -p ddraw_test.exe.so d3d.c && touch d3d.ok
wine: configuration in '/home/tfoerste/.wine' has been updated. VERT DCL IN[0] DCL IN[1] DCL OUT[0], POSITION DCL OUT[1], COLOR 0: MOV OUT[0], IN[0] 1: MOV OUT[1], IN[1] 2: END *** glibc detected *** ddraw_test.exe.so: corrupted double-linked list: 0x7c96b4a0 *** make: *** [d3d.ok] Terminated Terminated
(terminated by "pkill make")
http://bugs.winehq.org/show_bug.cgi?id=28198
Toralf Förster toralf.foerster@gmx.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |a04f0e9c789295c22aab6e0d3b3 | |239d0c794b304
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #1 from Henri Verbeet hverbeet@gmail.com 2011-08-27 00:29:06 CDT --- There's a fairly large chance this isn't a bug in Wine. Which driver is this? Your output suggests something Mesa based, but it shouldn't actually dump vertex / fragment shaders unless you explicitly ask for it.
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #2 from Toralf Förster toralf.foerster@gmx.de 2011-08-27 02:56:31 CDT --- I'm running an almost stable Gentoo w/ integrated intel GM45 graphic, current intel driver 2.16.0 (+ commit d330f37), mesa 7.11, xorg 1.10.4
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #3 from Toralf Förster toralf.foerster@gmx.de 2011-08-28 04:25:22 CDT --- Well, switching from gallium to classic for the i965 mesa module solved this issue (and few others as seen here : http://test.winehq.org/data/6a6aab27633405aeb5e464943110960f0099dffe/index_L... ).
Just curious that user32:static and user32:msg have more failures in the classic path.
Nevertheless how can I produce more and better output to tell the Intel driver devs and/or the Mesa devs what happened ?
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #4 from Henri Verbeet hverbeet@gmail.com 2011-08-28 06:39:01 CDT --- Valgrind might help.
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #5 from Toralf Förster toralf.foerster@gmx.de 2011-08-28 11:00:22 CDT --- Created an attachment (id=36135) --> (http://bugs.winehq.org/attachment.cgi?id=36135) valgrind output
(In reply to comment #4)
Valgrind might help.
I fear not in this case, or ?
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #6 from Henri Verbeet hverbeet@gmail.com 2011-08-28 11:16:52 CDT --- That doesn't look like it ran Valgrind on the actual test. You probably need at least the "--trace-children=yes" option.
http://bugs.winehq.org/show_bug.cgi?id=28198
Toralf Förster toralf.foerster@gmx.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #36135|0 |1 is obsolete| |
--- Comment #7 from Toralf Förster toralf.foerster@gmx.de 2011-08-28 15:08:19 CDT --- Created an attachment (id=36143) --> (http://bugs.winehq.org/attachment.cgi?id=36143) valgrind output for mesa gallium arch
I'm wondering whether this should be addressed directly to the Mesa people b/c under valgrind the file /sys/kernel/debug/dri/0/i915_error_state is empty - therefore the Intel driver devs would complain about that - or whether this is useful for your wine devs too ?
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #8 from Toralf Förster toralf.foerster@gmx.de 2011-08-28 15:09:04 CDT --- Created an attachment (id=36144) --> (http://bugs.winehq.org/attachment.cgi?id=36144) valgrind output for mesa classic arch
for completeness this is the case w/o a glibc error but the test case fails anyway
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #9 from Henri Verbeet hverbeet@gmail.com 2011-08-28 15:57:16 CDT --- (In reply to comment #7)
I'm wondering whether this should be addressed directly to the Mesa people b/c under valgrind the file /sys/kernel/debug/dri/0/i915_error_state is empty - therefore the Intel driver devs would complain about that - or whether this is useful for your wine devs too ?
The invalid reads and writes are potentially bad, so that's probably something you want to ask the Intel people to look at. I'm not familiar enough with the i965g driver to say something useful about that without looking at the source first though.
http://bugs.winehq.org/show_bug.cgi?id=28198
--- Comment #10 from Toralf Förster toralf.foerster@gmx.de 2011-08-29 03:58:02 CDT --- ok, I filed these 2 bugs
mesa : https://bugs.freedesktop.org/show_bug.cgi?id=40440 intel: https://bugs.freedesktop.org/show_bug.cgi?id=40439
http://bugs.winehq.org/show_bug.cgi?id=28198
Toralf Förster toralf.foerster@gmx.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |FIXED
--- Comment #11 from Toralf Förster toralf.foerster@gmx.de 2011-08-29 13:55:58 CDT --- intel provided a fix
http://bugs.winehq.org/show_bug.cgi?id=28198
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|FIXED |INVALID
--- Comment #12 from Austin English austinenglish@gmail.com 2011-08-29 14:35:20 CDT --- (In reply to comment #11)
intel provided a fix
Nothing was fixed in Wine.
http://bugs.winehq.org/show_bug.cgi?id=28198
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED CC| |nerv@dawncrow.de
--- Comment #13 from André H. nerv@dawncrow.de 2011-09-08 15:57:36 CDT --- closing invalid
http://bugs.winehq.org/show_bug.cgi?id=28198
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|directx-ddraw |directx-d3d
-
wine-bugs＠winehq.org