[Bug 19002] New: Regression in postal 2 with wine 1.1.24

http://bugs.winehq.org/show_bug.cgi?id=19002

Summary: Regression in postal 2 with wine 1.1.24
           Product: Wine
           Version: 1.1.24
          Platform: PC
        OS/Version: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: directx-d3d
        AssignedTo: [e-mail]
        ReportedBy: [e-mail]
                CC: [e-mail]

Todays commits caused a regression in Postal2 game.

Regression test prints:

b2f09fd20421d0a5e179b42332ca63bc5ac17d8a is first bad commit
commit b2f09fd20421d0a5e179b42332ca63bc5ac17d8a
Author: Stefan Doesinger 
Date:   Wed Jun 17 22:51:34 2009 +0200

wined3d: Emulate clipplanes in ARB if the NV extensions are not available.

This patch tries to find a free texture coordinate to load up to 4 clip
    coordinates into the pixel shader, and uses KIL to throw away fragments
    that are cut by a clipplane. If no free texture coordinate is found,
    clipping is not done. If more than 4 clipplanes are used, only the first
    4 are actually enabled. That should be pretty rare though.

:040000 040000 451683035b96e5b0e1bd85b72aeecb1c4746a75b
779b982f8d4d876f0f782b550af91557e6f3ac64 M    dlls

My box runs Ubuntu 9.04.
I have NVIDIA 6600GT with 180.44 driver

[Bug 19002] Regression in postal 2 with wine 1.1.24

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #1 from Matej Spindler   2009-06-19 13:00:32 ---
Screenshots of the game:

Loading menu:
http://www.shrani.si/f/2i/Ny/1Vluu84x/postal2regression1.png

In game:
http://www.shrani.si/f/2m/10f/1bZVZzKh/postal2regression2.png

[Bug 19002] Regression in postal 2 with wine 1.1.24

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #2 from Matej Spindler   2009-06-19 15:37:22 ---
Same thing with Lineage 2 Gracia Final Client !

http://www.shrani.si/f/3v/da/4HQ92Q3w/lineagewinebug.png

[Bug 19002] Regression in postal 2 with wine 1.1.24

http://bugs.winehq.org/show_bug.cgi?id=19002

Stefan Dösinger  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #3 from Stefan Dösinger   2009-06-19 17:20:49 ---
Are you using GLSL or ARB shaders?

It seems a bit odd that this causes a regression on nvidia hardware since the
patch should not become active on cards that do not support
GL_NV_vertex_program2_option.

Are there downloadable demos of this game?

[Bug 19002] Regression in postal 2 with wine 1.1.24

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #4 from Matej Spindler   2009-06-19 17:58:52 ---
There is a download for Postal 2 Demo at:
http://www.gamershell.com/download_1851.shtml
or
http://games.softpedia.com/progDownload/Postal-DEMO-Download-99.html

Demo has the same problem. (Set MouseWarpOverride = force to avoid mouse
problems)

GLSL or ARB ?

I guess GLSL since i didn't change anything related do d3d in registry.

According to glxinfo my card does have GL_NV_vertex_program2_option.

[Bug 19002] Regression in postal 2 with wine 1.1.24

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #5 from Stefan Dösinger   2009-06-20 07:40:17 ---
This patch only changes arb_program_shader.c, so it cannot affect GLSL. See if
you have a HKEY_CURRENT_USER/Software/Wine/Direct3D/UseGLSL key. If it doesn't
exist, or is set to anything other than "disabled", you're using GLSL, and the
regression test is most likely wrong. If it is set to "disabled", you're using
ARB shaders, and the regression test may be correct.

Note that even if you use ARB and switching to GLSL fixes the problem, the bug
still remains valid - ARB should work as well.

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

Vitaliy Margolen  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |regression
            Summary|Regression in postal 2 with |Postal 2 - graphics
                   |wine 1.1.24                 |problems

--- Comment #6 from Vitaliy Margolen   2009-06-20 12:28:43 ---
What sort of regression are you talking about? Specify _what_ is not working.

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #7 from Matej Spindler   2009-06-20 13:38:37 ---

(In reply to comment #6)
> What sort of regression are you talking about? Specify _what_ is not working.

Well see the screen shots. There are transparent textures all over the place!

So I did some testing.

I tested 1.1.24 deb from WienHQ repo. Same story. So probably nothing wrong
with my build tree.

I found another game with problems. Killing floor. Like with Postal and Lineage
this game is using unreal engine.

Changing UseGLSL doesn't produce any changes. I tested  with new wine prefix
(no DirectX key in reg), with UseGLSL="disabled" and with UseGLSL="asdf". Same
results at every run.

Is there a way to find out what is wine using (GLSL or ARB) during runtime ? Is
there any chance my card is ignoring GLSL and falling back to ARB for whatever
reason?

I did a "manual" regression test and firs checked out:

commit 7baf7c65184fd112938475804b7074abe11d234e
Author: Stefan Doesinger 
Date:   Mon Jun 8 15:35:42 2009 -0700

wined3d: Implement vertex texture fetch in ARB.

Did a "./configure && make" and all three games worked fine. Then I checked out
and compiled:

commit b2f09fd20421d0a5e179b42332ca63bc5ac17d8a
Author: Stefan Doesinger 
Date:   Wed Jun 17 22:51:34 2009 +0200

wined3d: Emulate clipplanes in ARB if the NV extensions are not available.

All three games now have problems. So problems do start with this commit.

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #8 from Henri Verbeet   2009-06-20 13:45:55 ---
Created an attachment (id=21913)
 --> (http://bugs.winehq.org/attachment.cgi?id=21913)
patch?

No idea if this is at all related, but does this patch make any difference?

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

Henri Verbeet  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #21913|0                           |1
           is patch|                            |

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #9 from Matej Spindler   2009-06-20 14:33:16 ---
(In reply to comment #8)
> Created an attachment (id=21913)

--> (http://bugs.winehq.org/attachment.cgi?id=21913) [details]
> patch?
>
> No idea if this is at all related, but does this patch make any difference?

No difference.

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #10 from Stefan Dösinger   2009-06-21 04:03:09 ---
Can you search for

if(ffp_clip_emul(stateblock) && device->vs_clipping) shader_addline(&buffer,
"KIL fragment.texcoord[7];\n");

in gen_arbfp_ffp_shader(), arb_program_shader.c, and replace it with

if(ffp_clip_emul(stateblock) && !device->vs_clipping) shader_addline(&buffer,
"KIL fragment.texcoord[7];\n");

Ie, negate device->vs_clipping. It should be in line 5173. (Sorry, no patch
atm, my wine tree is quite different)

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #11 from Matej Spindler   2009-06-21 04:42:29 ---
Created an attachment (id=21924)
 --> (http://bugs.winehq.org/attachment.cgi?id=21924)
negate device->vs_clipping

That worked!  All three games now render correctly!

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

--- Comment #12 from Stefan Dösinger   2009-06-21 06:01:08 ---
Ok, I'll send a patch.

The added negation is correct by the way. device->vs_clipping is true if the GL
vertex shader implementation supports clipping, for example with GLSL, or
ARB+NVvp2. In that case the vertex shader doesn't set the clipping texcoord,
and random vertices are clipped. I don't know why I forgot to negate this in
the first place, or why I didn't notice the problem. I did test GLSL, but maybe
I was just "lucky" with my random fragment.texcoord contents.

A similar consideration applies to the fixed function pipeline. The KIL should
not be used in this case because (a) clipping works the normal way anyway, and
(b) the texcoord is not initialized.

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

Ken Sharp  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #13 from Ken Sharp   2009-06-22 17:28:33 ---
*** Bug 19010 has been marked as a duplicate of this bug. ***

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

Casey Jones  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #14 from Casey Jones   2009-06-22 17:32:59 ---
I can confirm that Stefan's patch works too.
Also using an Nvidia card.

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

Matej Spindler  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |FIXED

--- Comment #15 from Matej Spindler   2009-06-23 16:34:14 ---
Today commits fixed the problem!

[Bug 19002] Postal 2 - graphics problems

http://bugs.winehq.org/show_bug.cgi?id=19002

Alexandre Julliard  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #16 from Alexandre Julliard   2009-07-03 12:27:05 ---
Closing bugs fixed in 1.1.25.