http://bugs.winehq.org/show_bug.cgi?id=22482
Summary: Baldur's Gate 2: starting up fails Product: Wine Version: 1.1.43 Platform: x86-64 URL: http://www.brothersoft.com/games/baldur-s-gate-ii-shad ows-of-amn.html OS/Version: Linux Status: NEW Keywords: download Severity: normal Priority: P2 Component: gdi32 AssignedTo: wine-bugs@winehq.org ReportedBy: austinenglish@gmail.com
Created an attachment (id=27543) --> (http://bugs.winehq.org/attachment.cgi?id=27543) terminal output
Terminal gets filled with:
err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash err:d3d_surface:IWineD3DBaseSurfaceImpl_CreateDIBSection CreateDIBSection failed! err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #1 from Austin English austinenglish@gmail.com 2010-04-24 22:44:40 --- Created an attachment (id=27544) --> (http://bugs.winehq.org/attachment.cgi?id=27544) +gdi,+d3d_surface trace
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #2 from Austin English austinenglish@gmail.com 2010-04-24 23:17:12 --- Setting DirectDrawRenderer to opengl fixes the leak, but then it renders improperly (the cursor leaves a black square over the image, which stays on the image after the cursor is moved).
Terminal is then filled with: fixme:d3d_surface:surface_download_data Read back converted textures unsupported, format=WINED3DFMT_B5G6R5_UNORM
http://bugs.winehq.org/show_bug.cgi?id=22482
Roderick Colenbrander thunderbird2k@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |thunderbird2k@gmail.com
--- Comment #3 from Roderick Colenbrander thunderbird2k@gmail.com 2010-04-25 12:13:46 --- I think this game worked fine before. The error is quite bad and likely due to a bug in wined3d (but perhaps it is elsewhere). If you can perform a regression test.
http://bugs.winehq.org/show_bug.cgi?id=22482
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |wylda@volny.cz
--- Comment #4 from Wylda wylda@volny.cz 2010-04-25 14:32:38 ---
Hi Austin, that log you pasted here, has nothing to do with the problem, thus please reset the component or set to directx-d3d.
1. Confirming, please consider KEYWORDS: +REGRESSION
2. I did a regression test between 1.1.2 and 1.1.3:
commit 34b37fe2411d83f1a6f3e11a7a1e4d28bd9b36ba Author: Stefan Dösinger stefan@codeweavers.com Date: Tue Aug 5 14:24:10 2008 -0500
d3d: Remove IWineD3DDevice::SetHwnd.
:040000 040000 70726cbb30d4e915779d7f0897cd44227f549e3e afda0b9ddba310280afcbe82a78ae4e2212f1801 M dlls :040000 040000 405e4687c093ec5c082507ce0195a7958799b4bf fc8ba10a277dc8b364967673fb81eb08f0844897 M include
3. No other bug report suffers from this commit.
4. Revert of this patch after git checkout makes that problem go away.
5. Adding author of this patch to CC.
--private keyword: bisected
http://bugs.winehq.org/show_bug.cgi?id=22482
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefan@codeweavers.com
http://bugs.winehq.org/show_bug.cgi?id=22482
Vitaliy Margolen vitaliy@kievinfo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Component|gdi32 |directx-ddraw
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #5 from Austin English austinenglish@gmail.com 2010-04-25 15:54:00 --- Thanks Wylda. I tested a couple wine versions, but they were all broken. Guess I didn't go back far enough..
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #6 from Wylda wylda@volny.cz 2010-04-25 16:02:13 --- (In reply to comment #5)
Thanks Wylda. I tested a couple wine versions, but they were all broken. Guess I didn't go back far enough..
Pure luck ;) I tested a lot of recent wines than jump way back to 0.9.49, all failed. So i decided to give that up. Don't know why, i gave it another shot with 0.9.60 and it run...
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #7 from Austin English austinenglish@gmail.com 2010-04-25 16:09:22 --- Created an attachment (id=27556) --> (http://bugs.winehq.org/attachment.cgi?id=27556) +d3d,+gdi,+d3d_surface trace
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #8 from Austin English austinenglish@gmail.com 2010-04-25 16:48:01 --- Enabling 3d acceleration fixes this issue as well (with ddr=opengl or gdi).
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #9 from Wylda wylda@volny.cz 2010-06-19 08:10:15 ---
This regression (bisected) is still present in wine-1.2-rc4.
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #10 from Wylda wylda@volny.cz 2010-09-19 00:48:45 CDT ---
Hi Austin, it seems, that situation change for me with new nVidia. It looks like, that the regression test is valid only for nVidia 8600GT.
With GT 240 the game runs when memory is set to 67MB and lower in registry even under wine-1.3.3. Can you confirm?
Note: don't forget to turn off 3D acceleration in BGConfig.exe before testing, otherwise you are testing under OpenGL (i guess).
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #11 from Roderick Colenbrander thunderbird2k@gmail.com 2010-09-19 13:10:31 CDT --- As of Wine 1.3.3 it is a little easier to fake other cards. If you override the pci device id + pci vendor id, it will now also select the proper amount of video memory for that different card. To 'turn' your 240gt into a 8600gt, set the pci vendor id to 0x10de (it is a dword) and the device id to 0x402. (So adjust Direct3d->VideoPciDeviceID and Direct3D->VideoPciVendorID in the registry)
http://bugs.winehq.org/show_bug.cgi?id=22482
Jim Scott jimdscott@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jimdscott@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=22482
--- Comment #12 from Jim Scott jimdscott@gmail.com 2010-12-28 10:18:22 CST --- (In reply to comment #10)
Hi Austin, it seems, that situation change for me with new nVidia. It looks like, that the regression test is valid only for nVidia 8600GT.
With GT 240 the game runs when memory is set to 67MB and lower in registry even under wine-1.3.3. Can you confirm?
Note: don't forget to turn off 3D acceleration in BGConfig.exe before testing, otherwise you are testing under OpenGL (i guess).
I am having the same issue: Your graphics card: GeForce 8200/PCI/SSE2/3DNOW! Your distribution: Ubuntu Your distribution version: 10.10 Current Wine version: wine-1.3.10 Wine version used by PlayOnLinux: wine-1.3.10
Sorry, I'm not sure what other information might be useful for resolving the issue. I can only confirm that I am seeing the same issue. I've tried the current 1.2 and 1.3 versions of Wine available for Ubuntu. When I set my DirectDrawRenderer to OpenGL, the game launches, but I have black boxes following my cursor.
I saw the suggestion that enabling 3D acceleration would resolve the issue. I tried to enable 3D acceleration in BGConfig.exe, but enabling 3D acceleration failed silently in the test step.
If there is any information I can provide that would help address the bug, I would be happy to try to find it. Sorry again, I just don't know what information would be useful.
http://bugs.winehq.org/show_bug.cgi?id=22482
Wylda wylda@volny.cz changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |huw@codeweavers.com
--- Comment #13 from Wylda wylda@volny.cz 2011-07-03 03:10:13 CDT ---
With GT 240 the game runs when memory is set to 67MB and lower in registry even under wine-1.3.3. Can you confirm?
I noticed, that with 1.3.23 i have to further reduce the memory in registry to 53MB to avoid a flood of:
err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash
The "regression" is between 1.3.17 and 1.3.18. It is caused by:
commit f33bbbfdd7f56701e91e96945f27b258fe73c6d4 Author: Huw Davies huw@codeweavers.com Date: Wed Apr 13 14:57:44 2011 +0100
gdi32: Keep track of the combined clip region.
:040000 040000 2b1402332cf2fd8d5b1733e001aef2b4890b2dda b2216e728592412dd2943c204572087633454e2a M dlls
Maybe the commit could be hint what's going wrong here.
Note: Tested with 3D acceleration turned off in BGConfig.exe.
http://bugs.winehq.org/show_bug.cgi?id=22482
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|huw@codeweavers.com |
--- Comment #14 from Austin English austinenglish@gmail.com 2011-07-03 04:26:00 CDT --- (In reply to comment #13)
With GT 240 the game runs when memory is set to 67MB and lower in registry even under wine-1.3.3. Can you confirm?
I noticed, that with 1.3.23 i have to further reduce the memory in registry to 53MB to avoid a flood of:
err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash
The "regression" is between 1.3.17 and 1.3.18. It is caused by:
commit f33bbbfdd7f56701e91e96945f27b258fe73c6d4 Author: Huw Davies huw@codeweavers.com Date: Wed Apr 13 14:57:44 2011 +0100
gdi32: Keep track of the combined clip region.
:040000 040000 2b1402332cf2fd8d5b1733e001aef2b4890b2dda b2216e728592412dd2943c204572087633454e2a M dlls
Maybe the commit could be hint what's going wrong here.
Note: Tested with 3D acceleration turned off in BGConfig.exe.
New bug, please.
http://bugs.winehq.org/show_bug.cgi?id=22482
Alex Bradbury asb@asbradbury.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |asb@asbradbury.org
--- Comment #15 from Alex Bradbury asb@asbradbury.org 2011-07-03 06:22:33 CDT --- This bug seems very similar to bug #5535 for Planescape Torment (also an Infinity Engine game). See the latest comments from Stefan, comment 72 and 74 on that bug. It sounds like setting the DirectDrawRenderer to opengl should act as a workaround to the gdi handle problem, and you should open a new bug for the graphical glitches.
http://bugs.winehq.org/show_bug.cgi?id=22482
Huw Davies huw@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |huw@codeweavers.com
--- Comment #16 from Huw Davies huw@codeweavers.com 2011-07-05 05:55:15 CDT --- (In reply to comment #14)
(In reply to comment #13)
With GT 240 the game runs when memory is set to 67MB and lower in registry even under wine-1.3.3. Can you confirm?
I noticed, that with 1.3.23 i have to further reduce the memory in registry to 53MB to avoid a flood of:
err:gdi:alloc_gdi_handle out of GDI object handles, expect a crash
The "regression" is between 1.3.17 and 1.3.18. It is caused by:
commit f33bbbfdd7f56701e91e96945f27b258fe73c6d4 Author: Huw Davies huw@codeweavers.com Date: Wed Apr 13 14:57:44 2011 +0100
gdi32: Keep track of the combined clip region.
:040000 040000 2b1402332cf2fd8d5b1733e001aef2b4890b2dda b2216e728592412dd2943c204572087633454e2a M dlls
Maybe the commit could be hint what's going wrong here.
Note: Tested with 3D acceleration turned off in BGConfig.exe.
New bug, please.
I don't think there's a handle leak introduced with this patch. We're just using up handles faster now, since every dib section dc gets an extra HRGN.
http://bugs.winehq.org/show_bug.cgi?id=22482
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |34b37fe2411d83f1a6f3e11a7a1 | |e4d28bd9b36ba
http://bugs.winehq.org/show_bug.cgi?id=22482
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |DUPLICATE
--- Comment #17 from Stefan Dösinger stefan@codeweavers.com 2011-11-13 16:32:05 CST --- I think we can close this bug as a duplicate of bug 5535. The regression SHA1 may be different, but the problem is the same(running out of GDI handles because the game tries to fill the video memory with surfaces), the workarounds are the same and both games use the same engine.
I'm not sure why this game regressed later than 5535, but I suspect that it was pretty close to running out of handles before and the patch brought up by the regression test tipped it over.
*** This bug has been marked as a duplicate of bug 5535 ***
http://bugs.winehq.org/show_bug.cgi?id=22482
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #18 from Austin English austinenglish@gmail.com 2011-11-14 14:25:43 CST --- Closing.
http://bugs.winehq.org/show_bug.cgi?id=22482
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|directx-ddraw |directx-d3d