Odd display problem for 0.9.59/60

Open forum for end-user questions about Wine. Before asking questions, check out the Wiki as a first step.
Forum Rules
Locked
ianpwelch
Newbie
Newbie
Posts: 1
Joined: Sat Apr 26, 2008 1:09 pm

Odd display problem for 0.9.59/60

Post by ianpwelch »

I came across a problem after updating to 0.9.59 and it appears to have continued to 0.9.60.

I started with gutsy, ATI 9600 with the fglrx drivers, direct rendering working as with the games i wanted (CNC3, Champ Man...) in 0.9.58 however on upgrading to .9.59 software installs and winecfg both display properly, However on running the games i get a fuzzy display a bit like a tv half tuned in to something but with heavy interference, I cant see any of the game display just the fuzzyness.

So i did a clean install of hardy installed all

sudo apt-get build-dep wine wine-dev

re-installed the ati drivers so fglrxinfo returns;

display: :0.0 screen: 0
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: ATI RADEON 9600 Series
OpenGL version string: 2.1.7415 Release

reinstalled wine from the package manager (0.9.60) and i get the same problem.

However 0.9.58 which i have compiled for Cnc with the cursor and lan patches runs my games fine. I then tried compiling 0.9.60 in the same way (no errors found during the process) and again it shows the fuzzy screen. The strange thing is there are no error messages produced in the terminal that differ from the usual ones i get in the working wine.

How would i find out what the problem is and subsequently fix it?

Any help would be appreciated.

Regards

cpu: Intel(R) Pentium(R) 4 CPU 3.00GHz
cpu_ghz: 3.0
memory: 1519
videocard_manufacturer: ATI Technologies Inc.
videocard_type: ATI RADEON 9600 Series
videocard_ram: 256
agp_aperture_size: N/A
videocard_driver_version: 2.1.7415 Release
soundcard: Intel ICH5 with ALC655 at irq 2
soundcard_driver: ALSA Version 1.0.15
machine_bitness: 32
kernel: 2.6.24-16-generic
x_version:
distro: Debian lenny/sid
GUI version: 6.0.2
kitpz2
Newbie
Newbie
Posts: 3
Joined: Mon Apr 28, 2008 11:02 am

Post by kitpz2 »

I have similar problem. since wine 0.9.59 all games wich requires directx displays black screen (simcity 4, Gothic I, Gothic II, and few more).

My hardware:
AMD Turion 2Ghz

ATI Radeon XPress X1100
(display: :0.0 screen: 0
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: ATI Radeon Xpress Series
OpenGL version string: 2.1.7412 Release)

768MB ram
Ubuntu gutsy/hardy
Cladan
Level 1
Level 1
Posts: 8
Joined: Thu May 01, 2008 2:28 am

Post by Cladan »

the exact same problem:

Wine 0.9.60
Suse 10.3
ATI X1900
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: ATI Radeon Xpress Series
OpenGL version string: 2.1.7412 Release)
all seems to work just that the screen is black with static
Rico
Moderator
Moderator
Posts: 91
Joined: Sat Feb 23, 2008 12:10 pm

Re: Odd display problem for 0.9.59/60

Post by Rico »

ianpwelch wrote:I cant see any of the game display just the fuzzyness.
Could you attach a picture? Does it work with an earlier wine version? If so, which? If all your games run with the earlier version then you can run a regression test http://wiki.winehq.org/RegressionTesting .
Cladan
Level 1
Level 1
Posts: 8
Joined: Thu May 01, 2008 2:28 am

Re: Odd display problem for 0.9.59/60

Post by Cladan »

Rico wrote:
ianpwelch wrote:I cant see any of the game display just the fuzzyness.
Could you attach a picture? Does it work with an earlier wine version? If so, which? If all your games run with the earlier version then you can run a regression test http://wiki.winehq.org/RegressionTesting .
I can easly reproduce it with 3DMark, so am doing git between version 0.9.58 and 0.9.59. Compiling goes slow (am on 2nd compile, 132 left).

am new to all this so hope I do it right but the instruction page is pretty clear.

--
edit: might take a bit longer, getting error on 2nd compile run:
/home/akm/wine-git/dlls/mscms/mscms_main.c:49: undefined reference to `free_handle_tables'
collect2: ld returned 1 exit status
winegcc: ccache failed
make[2]: *** [mscms.dll.so] Error 2
make[2]: Leaving directory `/home/akm/wine-git/dlls/mscms'
make[1]: *** [mscms] Error 2
make[1]: Leaving directory `/home/akm/wine-git/dlls'
make: *** [dlls] Error 2
Cladan
Level 1
Level 1
Posts: 8
Joined: Thu May 01, 2008 2:28 am

Post by Cladan »

any GIT guru have an solution for the compile error ?

I know it can either be good or bad (the result of the 132 compile) so I went both ways, going bad way (so error is between 0 and 66) gives me another compile error when 8 left. going good (so error is bewteen 66 and 132) gives me another compile error when 16 left)

would be better if I could compile without guessing the answers to narrow it down :)
austin987
Wine Developer
Wine Developer
Posts: 2383
Joined: Fri Feb 22, 2008 8:19 pm

Odd display problem for 0.9.59/60

Post by austin987 »

On Thu, May 1, 2008 at 11:27 AM, Cladan <[email protected]> wrote:
any GIT guru have an solution for the compile error ?
What error?
Cladan
Level 1
Level 1
Posts: 8
Joined: Thu May 01, 2008 2:28 am

Re: Odd display problem for 0.9.59/60

Post by Cladan »

austin987 wrote:On Thu, May 1, 2008 at 11:27 AM, Cladan <[email protected]> wrote:
any GIT guru have an solution for the compile error ?
What error?
the above one, but let me re-cap:

git bisect start
git bisect good wine-0.9.58
git bisect bad wine-0.9.59

gives me a 265 mark

CC="ccache gcc" ./configure --verbose && make clean && make

works fine.

I test my application (3Dmark works fine as its directx) and get a bad result

git bisect bad

gives me a 132 mark

CC="ccache gcc" ./configure --verbose && make clean && make

returns below error.

/home/akm/wine-git/dlls/mscms/mscms_main.c:49: undefined reference to `free_handle_tables'
collect2: ld returned 1 exit status
winegcc: ccache failed
make[2]: *** [mscms.dll.so] Error 2
make[2]: Leaving directory `/home/akm/wine-git/dlls/mscms'
make[1]: *** [mscms] Error 2
make[1]: Leaving directory `/home/akm/wine-git/dlls'
make: *** [dlls] Error 2

So as I posted I can either go good or bad on this mark, if I go bad I get another one (diffrent one) once I narrow it down to 8 marks. If I go good I get another one (diffrent one) once I narrow it down to 16 marks.
austin987
Wine Developer
Wine Developer
Posts: 2383
Joined: Fri Feb 22, 2008 8:19 pm

Odd display problem for 0.9.59/60

Post by austin987 »

On Thu, May 1, 2008 at 11:57 AM, Cladan <[email protected]> wrote:

austin987 wrote:
On Thu, May 1, 2008 at 11:27 AM, Cladan <[email protected]> wrote:
any GIT guru have an solution for the compile error ?
What error?
the above one, but let me re-cap:
You never posted the actual error ;-).
git bisect start
git bisect good wine-0.9.58
git bisect bad wine-0.9.59

gives me a 265 mark

CC="ccache gcc" ./configure --verbose && make clean && make

works fine.

I test my application (3Dmark works fine as its directx) and get a bad result

git bisect bad

gives me a 132 mark

CC="ccache gcc" ./configure --verbose && make clean && make

returns below error.

/home/akm/wine-git/dlls/mscms/mscms_main.c:49: undefined reference to `free_handle_tables'
collect2: ld returned 1 exit status
winegcc: ccache failed
make[2]: *** [mscms.dll.so] Error 2
make[2]: Leaving directory `/home/akm/wine-git/dlls/mscms'
make[1]: *** [mscms] Error 2
make[1]: Leaving directory `/home/akm/wine-git/dlls'
make: *** [dlls] Error 2

So as I posted I can either go good or bad on this mark, if I go bad I get another one (diffrent one) once I narrow it down to 8 marks. If I go good I get another one (diffrent one) once I narrow it down to 16 marks.
This is a bit of a tricky situation...best thing to do is:
http://wiki.winehq.org/ReverseRegressionTesting
Cladan
Level 1
Level 1
Posts: 8
Joined: Thu May 01, 2008 2:28 am

Post by Cladan »

quick question:

if I do:
git bisect reset
CC="ccache gcc" ./configure --verbose && make clean && make

I am compiling the latest build right ?
James McKenzie

Odd display problem for 0.9.59/60

Post by James McKenzie »

Cladan wrote:
Rico wrote:
ianpwelch wrote:
I cant see any of the game display just the fuzzyness.
Could you attach a picture? Does it work with an earlier wine version? If so, which? If all your games run with the earlier version then you can run a regression test http://wiki.winehq.org/RegressionTesting .
I can easly reproduce it with 3DMark, so am doing git between version 0.9.58 and 0.9.59. Compiling goes slow (am on 2nd compile, 132 left).

am new to all this so hope I do it right but the instruction page is pretty clear.


Are you doing git bisects?

This may make it faster. Also, are you using ccache?

James McKenzie
austin987
Wine Developer
Wine Developer
Posts: 2383
Joined: Fri Feb 22, 2008 8:19 pm

Odd display problem for 0.9.59/60

Post by austin987 »

On Fri, May 2, 2008 at 12:25 AM, Cladan <[email protected]> wrote:
quick question:

if I do:
git bisect reset

CC="ccache gcc" ./configure --verbose && make clean && make

I am compiling the latest build right ?
No...that will give HEAD, which is the whatever you had git set to
before telling it to bisect.

$ git bisect reset ; git reset origin

will give you the latest git you have downloaded. Run:

$ git fetch ; git rebase origin

To update git from winehq
kitpz2
Newbie
Newbie
Posts: 3
Joined: Mon Apr 28, 2008 11:02 am

Post by kitpz2 »

Code: Select all

4640be8dc8347f4147cbf87c78b3cfa510fcccc3 is first bad commit
commit 4640be8dc8347f4147cbf87c78b3cfa510fcccc3
Author: Stefan Dösinger <[email protected]>
Date:   Sat Mar 22 14:31:52 2008 +0100

    wined3d: Fragment processing using GL_ATI_fragment_shader.
    
    This adds code for handling fixed function fragment processing with the
    GL_ATI_fragment_shader extension. This is a sort-of programmable
    interface for fragment processing at the level of shader model 1.4 in
    d3d. This code is of use on r200, r250 and r280 cards(radeon 8500 to
    9200) which do not support GL_ARB_fragment_program, but support pixel
    shader 1.4 on Windows. This code is somewhat a counterpart to the
    existing fragment processing code using GL_NV_register_combiners and
    GL_NV_texture_shader.

:040000 040000 1f98025b56ec69ff0ae262779ce95e40d9514c2c 8691da5c59cc3ecc2ebb39d225e3ed2d16a31111 M	dlls
That was printed after Regression Testing.
vitamin
Moderator
Moderator
Posts: 6605
Joined: Sat Feb 23, 2008 2:29 pm

Post by vitamin »

kitpz2 wrote:That was printed after Regression Testing.
So is it still doesn't work for you with wine-0.9.61?
kitpz2
Newbie
Newbie
Posts: 3
Joined: Mon Apr 28, 2008 11:02 am

Post by kitpz2 »

in 0.9.61 this problem vanished. Thanks :)
Locked