http://bugs.winehq.org/show_bug.cgi?id=29245
Bug #: 29245 Summary: WinUAE graphics trashed/corrupted, regression between 1.3.15 and 1.3.28? Product: Wine Version: 1.3.34 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d AssignedTo: wine-bugs@winehq.org ReportedBy: markk@clara.co.uk Classification: Unclassified
Hi,
[I set the component of this bug to directx-d3d, but I'm not certain that's correct.]
WinUAE is an open-source Amiga emulator. I tested WinUAE version 2.3.3 in Wine. The test system was running Lubuntu 11.10 x86-64 with Nvidia graphics driver version 285.
There has been a regression, probably some time between 1.3.15 and 1.3.28. (I don't recall seeing this issue with Wine 1.3.15 in Xubuntu 9.10 32-bit.)
First test, using Wine 1.2.3. This works correctly: - Download and run the WinUAE installer. - Run winuae.exe - Click OK at the dialog box about system ROMs. - The WinUAE Properties window appears. Click Start. - The emulation window opens. It shows shades of grey for several seconds, then "No bootable media found..." text. There's a red mouse pointer which you can move.
Second test, running Wine 1.3.28: - First four steps as above. - The emulation window opens. There is just graphical trash/garbage in it. The window contents are static, and don't have any relation to the graphics which should be there.
Testing with Wine 1.3.34 shows the same problem as 1.3.28.
There's little difference in console output testing with 1.3.28 vs 1.2.3. There was this line in the 1.3.28 output: fixme:d3d:swapchain_init Add OpenGL context recreation support to context_validate_onscreen_formats However I've seen similar lines with other programs before when there was no corruption.
http://bugs.winehq.org/show_bug.cgi?id=29245
Mark K markk@clara.co.uk changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download, regression, | |source URL| |http://www.winuae.net/files | |/InstallWinUAE2330.exe CC| |markk@clara.co.uk
http://bugs.winehq.org/show_bug.cgi?id=29245
Dmitry Timoshkov dmitry@baikal.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|markk@clara.co.uk | Component|directx-d3d |-unknown Summary|WinUAE graphics |WinUAE graphics |trashed/corrupted, |trashed/corrupted |regression between 1.3.15 | |and 1.3.28? |
--- Comment #1 from Dmitry Timoshkov dmitry@baikal.ru 2011-12-05 09:13:24 CST --- If this is a regression please post the results of your regression testing: http://wiki.winehq.org/RegressionTesting
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #2 from Mark K markk@clara.co.uk 2011-12-05 11:16:01 CST --- Created attachment 37811 --> http://bugs.winehq.org/attachment.cgi?id=37811 Console output running WinUAE 2.3.3 in Wine 1.3.27 and 1.3.28
http://bugs.winehq.org/show_bug.cgi?id=29245
Mark K markk@clara.co.uk changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |markk@clara.co.uk
--- Comment #3 from Mark K markk@clara.co.uk 2011-12-05 11:16:49 CST --- I hope to be able to do a proper regression test shortly.
In the mean time, I tested several Wine release versions. The problem first appeared with Wine 1.3.28. I have attached a text file with console output when running WinUAE 2.3.3 in Wine 1.3.27 (works) and 1.3.28 (first bad release).
The basic difference is that the 1.3.27 console output includes this line repeated 6 times: fixme:d3d_surface:surface_cpu_blt Filter WINED3DTEXF_LINEAR not supported in software blit.
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #4 from Mark K markk@clara.co.uk 2011-12-05 13:44:14 CST --- I ran a regression test and got this result:
94ae743ea668e49d40ae4e2dc5fe1f5d9be018cb is the first bad commit commit 94ae743ea668e49d40ae4e2dc5fe1f5d9be018cb Author: Henri Verbeet hverbeet@codeweavers.com Date: Tue Aug 30 20:12:31 2011 +0200
ddraw: Make the OpenGL renderer the default one.
:040000 040000 fbce2dc172258a4a7eabc65b6f86bd58522b4667 dfb3b24196ac3ad09100758a48b2d4ef7e45e26b M dlls
http://bugs.winehq.org/show_bug.cgi?id=29245
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|-unknown |directx-d3d Regression SHA1| |94ae743ea668e49d40ae4e2dc5f | |e1f5d9be018cb
http://bugs.winehq.org/show_bug.cgi?id=29245
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |hverbeet@gmail.com
--- Comment #5 from Austin English austinenglish@gmail.com 2011-12-05 17:08:40 CST --- winetricks ddr=gdi should workaround it, in the meantime.
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #6 from Henri Verbeet hverbeet@gmail.com 2011-12-07 04:23:18 CST --- This seems to work for me with 1.3.34. Are you using a clean wineprefix? You're using the binary nvidia driver, right? Which graphics card do you have?
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #7 from Mark K markk@clara.co.uk 2011-12-07 07:51:00 CST --- I did some more testing. Wine 1.3.34 with clean wineprefix. Nvidia Geforce Go 7300 graphics chip. I tested with Lubuntu 11.10 64-bit, with several Nvidia driver versions.
Nouveau: Seems to work okay Nvidia 96.43.20 (nvidia-96): Seems to work okay Nvidia 173.14.30 (nvidia-173): Crash on starting emulation with Null filter (which is the default). No problem when filter is set to none before starting emulation. Nvidia 280.13 (nvidia-current): Problem present, same as with Nvidia 285 driver. Nvidia 285.05.09 (nvidia-current-updates): Problem present as described above
So it looks like this could be an Nvidia driver bug. (Unless later drivers support some feature not present in earlier versions, and there's a bug in the part of Wine which uses that feature???)
Can anyone else reproduce the problem I'm seeing with Nvidia driver 280 or later?
The crash with driver version 173 did not cause the X server to crash/hang; the program just prints some console output and exits. WinUAE opens its Properties window, the crash happens when you click Start with the Null filter (which is the default). Relevant part of the console output:
*** glibc detected *** C:\Program Files\WinUAE\winuae.exe: corrupted double-linked list: 0xf598f140 *** ======= Backtrace: ========= /lib32/libc.so.6(+0x6eb72)[0xf7487b72] /lib32/libc.so.6(+0x6f9d8)[0xf74889d8] /lib32/libc.so.6(cfree+0x6d)[0xf748b8cd] /usr/lib32/nvidia-173-updates/libGLcore.so.1(+0x5964ef)[0xf6a104ef] [0x200000] /usr/lib32/nvidia-173-updates/libGLcore.so.1(+0x5ed9c0)[0xf6a679c0] [0x138808b] ======= Memory map: ======== ... [rest omitted] ...
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #8 from Henri Verbeet hverbeet@gmail.com 2011-12-07 08:23:17 CST --- Maybe, but it would have to be specific to the hardware you're using. I tried this with 285.05.09 and a GeForce 9600M GT.
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #9 from Henri Verbeet hverbeet@gmail.com 2012-01-27 11:57:19 CST --- Since my last comment I've tried this on a GF7800 as well, and it works fine there for me as well. It would be helpful if someone else could verify this happens / doesn't happen on GF7 hardware, but I'm inclined to think this is INVALID.
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #10 from Mark K markk@clara.co.uk 2012-01-28 02:55:31 CST --- I'll re-test with Wine 1.4rc1 and various Nvidia driver versions in a few days time.
In the changelog for 1.4rc1, bug 20089 "Theme Park World: Loading screen displayed wrong on GeForce 4 MX" and bug 29605 "Theme Park World: Lobby screen is black on GeForce 4 MX" are listed as fixed. That (hopefully) could mean the variety of graphics issues I've seen with Wine and various Nvidia driver versions are fixed.
The reporter of those bugs was using a Geforce 4MX with Nvidia driver v96. That chip is significantly older than my Geforce Go 7300, but the Go 7300 does work with the v96 driver (as well as the other versions provided by Ubuntu; 173, 280 and 285). The Geforce 7 generation is the last chip supported by the v96 driver. If you have access to a Geforce 7-series card it might be helpful for testing purposes.
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #11 from Henri Verbeet hverbeet@gmail.com 2012-02-16 07:49:21 CST --- (In reply to comment #10)
I'll re-test with Wine 1.4rc1 and various Nvidia driver versions in a few days time.
Any luck?
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #12 from Mark K markk@clara.co.uk 2012-02-16 10:59:10 CST --- The problem is still present with Wine 1.4rc1 and Nvidia driver 285. If an Ubuntu package for driver version 295 becomes available soon I'll try that too.
I did a sort-of regression test, by compiling various older Wine versions and doing "winetricks ddr=opengl" before testing (since the default was gdi in older Wine versions). I got the same issue back to Wine 1.1 (from memory). Didn't try any earlier because I got a compilation error.
Still not sure whether it's a driver bug or Wine issue. Maybe there's someone out there with a Geforce Go 7300 who could test?
http://bugs.winehq.org/show_bug.cgi?id=29245
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #13 from joaopa jeremielapuree@yahoo.fr 2012-02-16 11:13:50 CST --- Works fine here with Nvidia 7600Go and driver 290.20.
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #14 from joaopa jeremielapuree@yahoo.fr 2012-02-16 11:48:08 CST --- Obviously, I meant driver 295.20
http://bugs.winehq.org/show_bug.cgi?id=29245
--- Comment #15 from Mark K markk@clara.co.uk 2012-03-05 15:02:02 CST --- Testing with Nvidia driver 295.20 this problem seems to have gone. So it was most likely a driver bug.
http://bugs.winehq.org/show_bug.cgi?id=29245
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED CC| |adys.wh@gmail.com Resolution| |UPSTREAM
--- Comment #16 from Jerome Leclanche adys.wh@gmail.com 2012-03-05 15:46:38 CST --- Nvidia bug then.
http://bugs.winehq.org/show_bug.cgi?id=29245
Jerome Leclanche adys.wh@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #17 from Jerome Leclanche adys.wh@gmail.com 2012-03-05 15:46:44 CST --- Closing
http://bugs.winehq.org/show_bug.cgi?id=29245
Vitaliy Margolen vitaliy-bugzilla@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords|regression | Regression SHA1|94ae743ea668e49d40ae4e2dc5f | |e1f5d9be018cb |
--- Comment #18 from Vitaliy Margolen vitaliy-bugzilla@kievinfo.com 2012-03-05 20:55:21 CST --- So not a regression