https://bugs.freedesktop.org/show_bug.cgi?id=37476
Summary: [wine] Devil May Cry 4: TXD tgsi opcode unsupported / translation from TGSI failed / missing vertex shader Product: Mesa Version: git Platform: Other URL: http://www.gamershell.com/download_27216.shtml OS/Version: All Status: NEW Severity: normal Priority: medium Component: Drivers/Gallium/r600 AssignedTo: dri-devel@lists.freedesktop.org ReportedBy: sa@whiz.se
The demo of the game Devil May Cry 4 (running in Wine) isn't rendering correctly, quite a few objects are missing. The following errors are printed on the terminal:
EE r600_shader.c:829 tgsi_unsupported - TXD tgsi opcode unsupported EE r600_shader.c:118 r600_pipe_shader_create - translation from TGSI failed ! EE r600_state_common.c:564 r600_draw_vbo - missing vertex shader EE r600_shader.c:829 tgsi_unsupported - TXD tgsi opcode unsupported EE r600_shader.c:118 r600_pipe_shader_create - translation from TGSI failed ! EE r600_state_common.c:564 r600_draw_vbo - missing vertex shader
I used apitrace to capture the game, even though the trace ends in a segfault it's enough to reproduce the above errors. http://dl.dropbox.com/u/28577999/dmc4.trace.7z (93M)
(To run the game in Wine native versions of WMASF.DLL and WMVCORE.DLL needs to be used)
System environment: -- system architecture: 32-bit -- Linux distribution: Debian unstable -- GPU: REDWOOD -- Model: XFX Radeon HD 5670 1GB -- Display connector: DVI -- xf86-video-ati: 6.14.1 -- xserver: 1.10.1 -- mesa: git-fd6f2d6 -- drm: 2.4.25 -- kernel: 2.6.39
https://bugs.freedesktop.org/show_bug.cgi?id=37476
Mike Kaplinskiy mike.kaplinskiy@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Platform|Other |x86 (IA32) AssignedTo|dri-devel@lists.freedesktop |mesa-dev@lists.freedesktop. |.org |org CC| |mike.kaplinskiy@gmail.com
--- Comment #1 from Mike Kaplinskiy mike.kaplinskiy@gmail.com 2011-06-05 22:32:34 PDT --- [changing assignee as this is clearly a mesa bug]
piglit arb_shader_texture_lod-texgrad shows the problem (using texture2DGradARB). Currently running it may also get you a fortify crash due to TXD having 4 inputs (the max in the context structure is 3).
I'm attaching a patch that has an implementation. Unfortunately I can't seem to get it to work entirely correctly. The piglit test looks almost correct but has several artifacts in random places. If someone could point me in the right direction (as I'm completely clueless), I would be happy to finish it. In case it matters, piglit does not show any new failures from this patch.
On Sun, Jun 05, 2011 at 10:32:34PM -0700, bugzilla-daemon@freedesktop.org wrote:
I'm attaching a patch that has an implementation. Unfortunately I can't seem to get it to work entirely correctly. The piglit test looks almost correct but has several artifacts in random places. If someone could point me in the right direction (as I'm completely clueless), I would be happy to finish it. In case it matters, piglit does not show any new failures from this patch.
Did you make any progress? This also affects EVE Online btw.
-David
On Tue, 2011-06-14 at 19:09 +0200, David Lamparter wrote:
On Sun, Jun 05, 2011 at 10:32:34PM -0700, bugzilla-daemon@freedesktop.org wrote:
I'm attaching a patch that has an implementation. Unfortunately I can't seem to get it to work entirely correctly. The piglit test looks almost correct but has several artifacts in random places. If someone could point me in the right direction (as I'm completely clueless), I would be happy to finish it. In case it matters, piglit does not show any new failures from this patch.
Did you make any progress? This also affects EVE Online btw.
AFAIK, you need to use the web interface to comment on bugs. Otherwise Mike will miss this (unless he follows dri-devel): https://bugs.freedesktop.org/show_bug.cgi?id=37476
On Wed, Jun 15, 2011 at 4:09 AM, Sven Arvidsson sa@whiz.se wrote:
On Tue, 2011-06-14 at 19:09 +0200, David Lamparter wrote:
On Sun, Jun 05, 2011 at 10:32:34PM -0700, bugzilla-daemon@freedesktop.org wrote:
I'm attaching a patch that has an implementation. Unfortunately I can't seem to get it to work entirely correctly. The piglit test looks almost correct but has several artifacts in random places. If someone could point me in the right direction (as I'm completely clueless), I would be happy to finish it. In case it matters, piglit does not show any new failures from this patch.
Did you make any progress? This also affects EVE Online btw.
AFAIK, you need to use the web interface to comment on bugs. Otherwise Mike will miss this (unless he follows dri-devel): https://bugs.freedesktop.org/show_bug.cgi?id=37476
Should all be fixed now in master.
Dave.
dri-devel@lists.freedesktop.org