wine users - Nov 2007 - Should I report my problem in Bugzilla? Outcast problem.

Hello everybody!

I'm trying to play Outast with wine 0.9.48 and wine fails with :

Can't find master volume control
fixme:win:EnumDisplayDevicesW ((null),0,0x34f820,0x00000000), stub!
fixme:d3d:IWineD3DDeviceImpl_Release (0x1352c0) Device released with 
resources still bound, acceptable but unexpected
fixme:d3d:dumpResources Leftover resource 0x134748 with type 
1,WINED3DRTYPE_SURFACE

Nevertheless, this game was working with wine some years ago as I saw in 
the application database :
http://appdb.winehq.org/appview.php?iAppId=2612

So my question is : where should I report my problem? Should I report it 
in Bugzilla even if I don't have coding skills and I don't know anything
about wine? :p

Besides, I see these lines in other bugs so I wonder if it's necessary 
to report it...

On Friday November 2 2007 14:19, R?mi G. wrote:
> Hello everybody!
>
> I'm trying to play Outast with wine 0.9.48 and wine fails with :
>
> Can't find master volume control
> fixme:win:EnumDisplayDevicesW ((null),0,0x34f820,0x00000000), stub!
> fixme:d3d:IWineD3DDeviceImpl_Release (0x1352c0) Device released with
> resources still bound, acceptable but unexpected
> fixme:d3d:dumpResources Leftover resource 0x134748 with type
> 1,WINED3DRTYPE_SURFACE
>
> Nevertheless, this game was working with wine some years ago as I saw in
> the application database :
> http://appdb.winehq.org/appview.php?iAppId=2612
	You may help to fix the bug if you find the patch that broke the game. See 
http://wiki.winehq.org/RegressionTesting . But before trying to start 
regression test please report the bug first (maybe someone else will find 
quick solution or suggest something useful).
> So my question is : where should I report my problem? Should I report it
> in Bugzilla even if I don't have coding skills and I don't know
anything
> about wine? :p
	Yes. In fact who have coding skills and enough time can fix problem 
themselves - they don't need to report bug; they just can send a fix
instead.
Of course, most users aren't programmers, have little free time, etc. but 
they can find bugs - and this is why bugzilla exists.
> Besides, I see these lines in other bugs so I wonder if it's necessary
> to report it...
	Rule of thumb: if some lines in terminal output in other bugs looks similar 
this doesn't mean that they are duplicates. If you don't know - simply
report
a bug.

I still have many difficulties with git...

I've first tried to install Wine 0.9 with the patches you give me :

*********
git-reset wine-0.9
gi-checkout -f
patch -p1 < 0.9-wrc-parser.diff
patch -p1 < 0.9-kernel-time.diff
patch -p1 < 0.9-gdi-freetype-old.diff
./configure && make depend && make
*********

And I got a new error :

*********
preloader.o: In function `map_so_lib':
/home/remjg/wine-git/loader/preloader.c:734: undefined reference to 
`__stack_chk_fail'
preloader.o: In function `wld_printf':
/home/remjg/wine-git/loader/preloader.c:360: undefined reference to 
`__stack_chk_fail'
collect2: ld a retourn? 1 code d'?tat d'ex?cution
make[1]: *** [wine-preloader] Erreur 1
make[1]: quittant le r?pertoire ? /home/remjg/wine-git/loader ?
make: *** [loader] Erreur 2
*********

I've also tried to compile Wine 0.9.10 and, as you write it in your 
reverse regression testing guide, it works if I apply the patch you give 
(but I can't run it because of a segmentation fault... not important in 
my case).
Therefore, I tried a regression test to find a solution to my error :

*********
remjg at remjg-laptop:~/wine-git$ git-bisect reset
Already on branch "master"
remjg at remjg-laptop:~/wine-git$ git-bisect start
remjg at remjg-laptop:~/wine-git$ git-bisect good wine-0.9
remjg at remjg-laptop:~/wine-git$ git-bisect bad wine-0.9.10
Bisecting: 1306 revisions left to test after this
fatal: Untracked working tree file 'dlls/advapi32/tests/.gitignore' 
would be overwritten by merge.
*********

Then, I follow your instructions but bitsecting seems not to work 
(because of the last error? "fatal:Untrac...").
I don't know what to do.. And all Git documentation is in english so 
it's difficult for me :s
Any ideas?

On Saturday November 10 2007 12:22, R?mi G. wrote:
> Yes, I enjoy solving "strange" problems and I would like to do it
> myself. This show me how developpers work and it's interesting for me
	That's great :) . By the way thanks for all your questions - they motivated
me to write the reverse regression testing guide :) .
> So perhaps you should add in your reverse regression guide the too
> following lines at the beginning of Step 1 :
	This problem may happen at other steps too so it isn't specific to the
first
step. I've added Troubleshooting section to the guide on the wiki.
> In fact, I don't know which version I'm compiling so I
> can't guess which patch I should use to correct "freetype"
problem... So
> how can I see the version I'm going to compile?
cat VERSION
> First, should I remove freetype.c.rej?
	freetype.c.rej doesn't participate in build process so its (non)existence
is
irrelevant. In other words: you don't need to remove it.
> Then, should I reapply the patch to remove changes or apply the other
> patch?
	As stated in http://wiki.winehq.org/ReverseRegressionTesting just 
run "git-checkouf -f" in case of mistake and try to apply patch(es)
again.
	It is possible to automatically choose right versions of patches (by writing 
shell script) but personally I do this manually.
> If apply again 0.9-gdi-freetype-old.diff :
> ...
> As I don't understand very well, I don't know what I should
answer...
>
> If I apply the other patch 0.9-gdi-freetype.diff :
> ...
> Skipping patch.
> 6 out of 6 hunks ignored -- saving rejects to file dlls/gdi/freetype.c.rej
> ##########
> I don't understand too...
	Most likely you tried to apply wrong version of the patch and forgot to 
run "git-checkouf -f" before trying to apply right version of the
patch.

	Thank you for using WINE.

On Sunday November 11 2007 13:59, R?mi G. wrote:
> > Or get the fix from commit 7cd9a8f8dc9536393e4e6042ae18bfb89b8671e5.
> > apply the patch with:
>
> Thank you, this patch resolves my last error and I can compile Wine 0.9 :)
>
> > Most likely you tried to apply wrong version of the patch and forgot
to
> > run "git-checkouf -f" before trying to apply right version
of the patch.
>
> In fact I discovered that neither 0.9-gdi-freetype.diff nor
> 0.9-gdi-freetype-old.diff was working with some versions...
	That's OK if you can compile successfully.
> but it's  no 
> yet a problem as I'm able now to compile wine 0.9 ;)
>
> Then I fixed the segementation fault on wine startup by applying the
> commit : a68d293bae6cec6deafafb94972c9277791256b5.
> I had to modify line 69 to avoid a problem when patching (I don't know
> how I did but it works... see attachement "segfault.diff").
>
> So, to compile wine-0.9, I need to execute :
>
> patch -p1 < ~/0.9-wrc-parser.diff
> patch -p1 < ~/0.9-kernel-time.diff
> patch -p1 < ~/0.9-gdi-freetype-old.diff
> patch -p1 < ~/segfault.diff
> git show 7cd9a8f8dc9536393e4e6042ae18bfb89b8671e5 | patch -p1
>
>
> And in spite of the fact that it was said in the appdb that Outcast runs
> with Wine 0.9, I didn't success :-(
>
> 1) If I run Outcast.exe, the game asks me for CD2 (I need to find a
> crack...) and wine displays this error in console :
> err:reg:SCSI_getprocentry SCSI type line scan count error
> But, I can see an animation so it's encouraging.
>
> 2) If I run directly loader.exe, wine crashes with the output :
> fixme:ddraw:Main_DirectDraw_SetCooperativeLevel
> (0x7fdf56d0)->((nil),00000008)
>
> I got the same errors with the outcast demo while runing
> Outcast_demo.exe or loader.exe (but of course wine don't ask me the
> CD2!). I've tried different loader.exe versions.
	If I understand everything correctly, you can run the game (at least Demo), 
it doesn't ask CD2, shows an animation and then crash. Full version
doesn't
crash but ask for CD2. Is this correct?
> With wine 0.9.49, I got :
> Can't find master volume control
> fixme:win:EnumDisplayDevicesW ((null),0,0x34f820,0x00000000), stub!
> fixme:d3d:IWineD3DDeviceImpl_Release (0x1352c0) Device released with
> resources still bound, acceptable but unexpected
> fixme:d3d:dumpResources Leftover resource 0x134748 with type
> 1,WINED3DRTYPE_SURFACE
>
>
> Perhaps should I try a regression test to find "when" theses
errors appear?
	As far as I understand with 0.9.49 you even cannot see the introductory 
animation it just crashes immediately? If so you can try to find the patch 
that broke the game' start up - *maybe* with new one WINE version it will 
work if you revert bad commit: it is possible that it doesn't work with 0.9 
because of additional non-obvious incompatibilities with new libraries.
	But even if this isn't the case finding bad commit that prevent start up of
the game will be helpful anyway because will help to resolve at least a part 
of the bug that prevent Outcase from running.
	If my understanding of your words wasn't correct then please describe 
behavior of the game in each case with more details.
	Thank you for using WINE.