https://bugs.winehq.org/show_bug.cgi?id=37728
Bug ID: 37728 Summary: [regression][workaround] Misplaced mountains in Elevated because of changes in shader generator Product: Wine Version: 1.7.33 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs@winehq.org Reporter: debuser1@mt2014.com Distribution: ---
Created attachment 50226 --> https://bugs.winehq.org/attachment.cgi?id=50226 elevated_1024x768.exe
Steps to reproduce: 1. Install `winetricks winxp vd=1024x768 d3dx9` in empty wineprefix (d3dx9_33 is enough, but winetricks doesn't have a term for it) 2. Run elevated_1024x768.exe ( http://www.pouet.net/prod.php?which=52938 ) 3. Get several misplaced mountains, white holes in textures, camera going through walls sometimes (see attached screenshots)
Regression since wine-1.5.27-205-g667f422: commit 667f4225b5cc25c5e4f2f725d55a7c011e12c388 wined3d: Put local constants into the shader code again.
Notes: May be specific to my videocard (NVidia 8600 GTS), or driver version (295.59), don't have a newer card around to test. Patches for bug 37723 are required to make Elevated working
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #1 from debuser debuser1@mt2014.com --- Created attachment 50227 --> https://bugs.winehq.org/attachment.cgi?id=50227 Expected look in first seconds
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #2 from debuser debuser1@mt2014.com --- Created attachment 50228 --> https://bugs.winehq.org/attachment.cgi?id=50228 Actual buggy look in first seconds
https://bugs.winehq.org/show_bug.cgi?id=37728
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression CC| |stefan@codeweavers.com Component|-unknown |directx-d3d Summary|[regression][workaround] |Elevated: Misplaced |Misplaced mountains in |mountains because of |Elevated because of changes |changes in shader generator |in shader generator |
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #3 from debuser debuser1@mt2014.com --- Created attachment 50229 --> https://bugs.winehq.org/attachment.cgi?id=50229 Workaround hack
With this workaround applied Elevated looks exactly as it's supposed to.
https://bugs.winehq.org/show_bug.cgi?id=37728
Sebastian Lackner sebastian@fds-team.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sebastian@fds-team.de
https://bugs.winehq.org/show_bug.cgi?id=37728
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Regression SHA1| |667f4225b5cc25c5e4f2f725d55 | |a7c011e12c388
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #4 from Stefan Dösinger stefan@codeweavers.com --- Created attachment 50323 --> https://bugs.winehq.org/attachment.cgi?id=50323 Affected shader
I have isolated the misbehaving shader and attached the GLSL code produced by Wine. The constant that triggers the bad behavior is vs_lc3.
I suspect the Nvidia driver is over-optimizing something based on vs_lc3.w = 1.0, but I haven't looked close enough at the shader code yet.
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #5 from debuser debuser1@mt2014.com --- I don't have non-nvidia cards to test, but I asked in #winehq and someone with ATI card said that elevated looks broken in wine-1.7.34 (which has bug 37723 fixed, and doesn't have bug 38477 introduced), but it looks correct in wine-1.3.23 (last pre-1.5.27 version that doesn't have bug 37723 and is able to run elevated).
Maybe it's not specific to Nvidia after all? Is it possible to write some test printing some debug information to run on different hardware and check if it behaves the same?
https://bugs.winehq.org/show_bug.cgi?id=37728
Sergey Isakov isakov-sl@bk.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |isakov-sl@bk.ru
--- Comment #6 from Sergey Isakov isakov-sl@bk.ru --- (In reply to debuser from comment #5)
I don't have non-nvidia cards to test, but I asked in #winehq and someone with ATI card said that elevated looks broken in wine-1.7.34 (which has bug 37723 fixed, and doesn't have bug 38477 introduced), but it looks correct in wine-1.3.23 (last pre-1.5.27 version that doesn't have bug 37723 and is able to run elevated).
Maybe it's not specific to Nvidia after all? Is it possible to write some test printing some debug information to run on different hardware and check if it behaves the same?
Tested on AMD Radeon HD6670 MacOSX 10.9.5 wine-1.7.42-50-g0d2817b Black screen with live mouse and crash.
fixme:ole:RemUnknown_QueryInterface No interface for iid {00000019-0000-0000-c000-000000000046} fixme:win:EnumDisplayDevicesW ((null),0,0x1464f9e8,0x00000000), stub! fixme:d3d:swapchain_init The application requested more than one back buffer, this is not properly supported. Please configure the application to use double buffering (1 back buffer) if possible. wine: Unhandled page fault on read access to 0x44915fa2 at address 0x44915fa2 (thread 0025), starting debugger...
https://bugs.winehq.org/show_bug.cgi?id=37728
Józef Kucia joseph.kucia@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download URL| |http://www.pouet.net/prod.p | |hp?which=52938 CC| |joseph.kucia@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=37728
Stefan Dösinger stefandoesinger@gmx.at changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|stefan@codeweavers.com |stefandoesinger@gmx.at
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #7 from Sergey Isakov isakov-sl@bk.ru --- I see same regression and same workaround as in bug 35207. Is it a dupe?
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #8 from Sergey Isakov isakov-sl@bk.ru --- Now the demo works with me. wine-1.9.2-132 native d3dx9_33.dll I found no geometry issues.
If I set d3dx9_33 to be builtin then the application crashes: ~~~~ Unhandled exception: page fault on read access to 0x4765dfa2 in 32-bit code (0x4765dfa2). Register dump: CS:001b SS:0023 DS:0023 ES:0023 FS:1007 GS:000f EIP:4765dfa2 ESP:1464fe84 EBP:00000000 EFLAGS:00010246( R- -- I Z- -P- ) EAX:00000000 EBX:004000d8 ECX:0016dc60 EDX:0016dc64 ESI:00870104 EDI:004300d8 Stack dump: 0x1464fe84: 00420851 00420caf 00000da4 00000000 0x1464fe94: 00000000 00420a40 00420a11 00000000 0x1464fea4: 00870104 00000000 00000000 004300d8 0x1464feb4: 00420a40 00000000 1464fed0 004000d8 0x1464fec4: 0016dc64 0016dc60 00000000 004200c8 0x1464fed4: 7b84fafc 7ffdf000 00000000 7b852491 0200: sel=1007 base=7ffc0000 limit=00000fff 32-bit rw- Backtrace: =>0 0x4765dfa2 .L__wine_spec_forwards+0x1aa() in d3dx9_33 (0x00000000) 0x4765dfa2 .L__wine_spec_forwards+0x1aa in d3dx9_33: xorl %fs:0x39(%eax,%edi,2),%esp Modules: Module Address Debug info Name (229 modules) PE 400000-12c80000 Deferred elevated_1024x768 PE 40001000-401b6000 Deferred libwine.1.0.dylib ~~~~ This is strange for me as I set d3dx9_36.dll native. As far as I understand calls to d3dx9_33 should be redirected to d3dx9_36.dll? It seems not happen.
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #9 from Matteo Bruni matteo.mystral@gmail.com --- (In reply to Sergey Isakov from comment #8)
Now the demo works with me.
This bug only affects Nvidia binary drivers.
If I set d3dx9_33 to be builtin then the application crashes: This is strange for me as I set d3dx9_36.dll native. As far as I understand calls to d3dx9_33 should be redirected to d3dx9_36.dll? It seems not happen.
That's because of bug 21817.
https://bugs.winehq.org/show_bug.cgi?id=37728
--- Comment #10 from Sergey Isakov isakov-sl@bk.ru --- Confirm the bug with Nvidia on MacOSX. The patch improves the picture while still not perfect may be because Geforce7300 is too old.
https://bugs.winehq.org/show_bug.cgi?id=37728
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #11 from winetest@luukku.com --- wine: Bad EXE format??
wine 2.0 series.
https://bugs.winehq.org/show_bug.cgi?id=37728
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |DUPLICATE
--- Comment #12 from Henri Verbeet hverbeet@gmail.com --- Duplicate of bug 35207, based on comment 3.
*** This bug has been marked as a duplicate of bug 35207 ***
https://bugs.winehq.org/show_bug.cgi?id=37728
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED CC| |nerv@dawncrow.de
--- Comment #13 from André H. nerv@dawncrow.de --- closing dup