November 2006 - Wine-devel - list.winehq.org

Re: winmm/wineoss: Cast-qual warning fix
by Marcus Meissner 24 Nov '06

24 Nov '06

On Thu, Nov 23, 2006 at 08:29:00PM +0000, Andrew Talbot wrote: > Changelog: > winmm/wineoss: Cast-qual warning fix. > > diff -urN a/dlls/winmm/wineoss/dscapture.c b/dlls/winmm/wineoss/dscapture.c > --- a/dlls/winmm/wineoss/dscapture.c 2006-11-13 17:34:40.000000000 +0000 > +++ b/dlls/winmm/wineoss/dscapture.c 2006-11-23 20:15:22.000000000 +0000 > @@ -853,7 +853,7 @@ > { > int i; > for (i = 0; i < length; i++) > - ((char *)dst)[i] = ((char *)src)[i]; > + ((char *)dst)[i] = ((const char *)src)[i]; > return dst; > } I really wonder why we do have our own memcpy() here and why there is a strange /* FIXME: using memcpy can cause strange crashes so use this fake one */ comment above it. Ciao, Marcus

2 1

Re: wine crashes on every win32 application (null ptr access early); winedbg adventures
by Saulius Menkevičius 24 Nov '06

24 Nov '06

alexis.deruelle wrote: > Try removing nvidia drivers if installed (see bug > http://bugs.winehq.org/show_bug.cgi?id=6735) > > Cheers, > > -- > Alexis > > > Accédez au courrier électronique de La Poste > sur www.laposte.net ou sur 3615 LAPOSTENET (0,34€ TTC /mn) > 1 Giga de stockage gratuit – Antispam et antivirus intégrés > > > > yep, that will be it.. thanks

1 0

[kernel32/console.c] A really very small fix.
by Anatoly 24 Nov '06

24 Nov '06

Hi! This fix change CodePage parameter in MultiByteToWideChar() type function from CP_ACP to return of GetConsoleOutputCP(). Now unicode symbols correct outs in wineconsole application (ex: Far). Changelog: Change codepage parameter in ReadConsoleA function. Change codepage parameter in ScrollConsoleScreenBufferA function. Index: dlls/kernel32/console.c =================================================================== RCS file: /home/wine/wine/dlls/kernel32/console.c,v retrieving revision 1.2 diff -u -p -u -r1.2 console.c --- dlls/kernel32/console.c 13 Oct 2006 10:27:18 -0000 1.2 +++ dlls/kernel32/console.c 24 Nov 2006 12:56:13 -0000 @@ -1200,7 +1199,7 @@ BOOL WINAPI ReadConsoleA(HANDLE hConsole BOOL ret; if ((ret = ReadConsoleW(hConsoleInput, ptr, nNumberOfCharsToRead, &ncr, NULL))) - ncr = WideCharToMultiByte(CP_ACP, 0, ptr, ncr, lpBuffer, nNumberOfCharsToRead, NULL, NULL); + ncr = WideCharToMultiByte(GetConsoleOutputCP(), 0, ptr, ncr, lpBuffer, nNumberOfCharsToRead, NULL, NULL); if (lpNumberOfCharsRead) *lpNumberOfCharsRead = ncr; HeapFree(GetProcessHeap(), 0, ptr); @@ -2258,7 +2257,7 @@ BOOL WINAPI ScrollConsoleScreenBufferA(H CHAR_INFO ciw; ciw.Attributes = lpFill->Attributes; - MultiByteToWideChar(CP_ACP, 0, &lpFill->Char.AsciiChar, 1, &ciw.Char.UnicodeChar, 1); + MultiByteToWideChar(GetConsoleOutputCP(), 0, &lpFill->Char.AsciiChar, 1, &ciw.Char.UnicodeChar, 1); return ScrollConsoleScreenBufferW(hConsoleOutput, lpScrollRect, lpClipRect, dwDestOrigin, &ciw);

1 0

oleaut32: preliminary support for VT_DISPATCH as parameters to VarAdd (and other 2-operand operations) (RESEND)
by Alex Villacís Lasso 24 Nov '06

24 Nov '06

This is a preliminary patch to add support for VT_DISPATCH as parameters to VarAdd and other arithmetic operations with variants. The tests still "pass" in the sense that null VT_DISPATCH as issued by the tests are still rejected. I post this in the hope that it will be reviewed while I prepare a test case. Changelog: * Add support for receiving of VT_DISPATCH parameters in several operator functions by fetching the result value, then reissuing the same operation on result data. -- The following cryptic message was allegedly found in the inner edge of a Windows XP installation CD: 4F6E65204F5320746F2072756C65207468656D20616C6C2C204F6E65204F5320746F2066696E6420 7468656D2C0D0A4F6E65204F5320746F206272696E67207468656D20616C6C20616E6420696E2074 6865206461726B6E6573732062696E64207468656D2E0A It is rumored that only a true Unix Wizard can decypher this mysterious message, which supposedly encodes the true nature and purpose of the software.

2 1

[winex11.drv/keyboard.c] :Some fix and new function.
by Anatoly 24 Nov '06

24 Nov '06

Hi! This patch adds in /dlls/winex11.drv/ new module (ucs2keysym.c). Keyboard.c needs this module (function ucs2ketsym) for convert Unicode values to KeySym (See patch). Now in function X11DRV_VkKeyScanEx have removed FIXME,because function ucs2keysym returns correct keysym for wchar/ Changelog: Add function usc2keysym to convert from WCHAR values to Keysym. And now we have correct keycode for wchar values.

1 0

RFC: OpenAL Winmm driver and OpenAL32.dll thunk -- 2
by Nick Burns 24 Nov '06

24 Nov '06

(due to alut changes -- im asking for comments again -- if no problems -- ill try and submit) ------ Very cool audio worked for you as well... I too noticed the stereo effects in ut2004 and was taken aback -- first it was like -- its only on the right -- the the sound zipped over (never heard it like that before) good thing i had my airsick-bag handy. -- Ok those build failures are not totally unexpected -- depending on yer impl of openal -- they may have broken binary compat (i would 'nm' (list exported symbols of) your openal lib) and look for alut*. Very surprising that those functions are missing thou -- they are supposed to be there... freealut will fix the binary compat issues -- so its not the biggest problem -- what i could do is be safer with alut and have the configure test for each alut function that SHOULD be in the openal lib. OR better yet I can just IMPL the lame (OLD DEPRECATED) alut functions on top of openal inside openal32.dll -- this is probably the most portable and best solution... (I would just copy the impl from freealut) Looking at freealut... It is doable -- but it quite a bit of extra code... OR I can drop alut support altogether so the possibilities... 1 - require freealut 2 - impl alut functions on openal (using freealut code 5 functions -> 7 files) 3 - stub alut -- then use freealut to impl these (never used dead deprecated) functions 4 - remove alut I have choosen 3 -- attaching my patch with those changes in (and removed some meaningless comments) These stubs will FIXME -- if hit. The stubbing is controlled with 'USE_BUILTIN_ALUT' in openal.c --- btw if you dont have openal it will still build -- but spew at you if it is used -- that openal is not supported... - Nick >From: Stefan Dösinger <stefandoesinger(a)gmx.at> >To: wine-devel(a)winehq.org >CC: Nick Burns <adger44(a)hotmail.com> >Subject: Re: RFC: OpenAL Winmm driver and OpenAL32.dll thunk was >Re:OpenALand DirectSound >Date: Thu, 23 Nov 2006 21:56:52 +0100 > >Am Mittwoch 22 November 2006 11:44 schrieb Nick Burns: > > Good luck > > And may the Sound be with you... >Ok, after cleaning up the d3d mess in my tree a bit I gave your patch a >try: > >It applied fine, found AL/al.h after creating configure with autoconf, but >compilation failed in dlls/openal: > >openal.o: In function `wine_alutLoadWAVMemory': >/usr/src/wine/dlls/openal32/openal.c:327: undefined reference to >`alutLoadWAVMemory' >openal.o: In function `wine_alutLoadWAVFile': >/usr/src/wine/dlls/openal32/openal.c:327: undefined reference to >`alutLoadWAVFile' >openal.o: In function `wine_alutUnloadWAV': >/usr/src/wine/dlls/openal32/openal.c:326: undefined reference to >`alutUnloadWAV' >openal.o: In function `wine_alutExit': >/usr/src/wine/dlls/openal32/openal.c:326: undefined reference to `alutExit' >openal.o: In function `wine_alutInit': >/usr/src/wine/dlls/openal32/openal.c:326: undefined reference to `alutInit' >collect2: ld returned 1 exit status >winegcc: gcc failed. > >turned out that I needed the freealut package(gentoo). Stangely wine didn't >link against that lib automatically, I had to specify -lalut manually. > >Once the thing was compiled it worked perfectly with Jedi Academy :-) . >Wine >loaded openal32.dll builtin, and I had sound although I had no sound driver >selected in winecfg, which shows that the whole api rollercoaster is >bypassed. I had to configure the game to use openal first. > >The sound appeared strange at first, but after using headphones instead of >my >broken notebook speakers it turned out that those were just stereo effects >I >never heard before :-) . I also used to wonder what that promiently >marked "EAX Sound" option was that used to crash wine: Apparently exactly >that openal option I enabled first in the config files :-) > >/me joins the dark side of the sound > >Great job! >Stefan

1 0

RFC: OpenAL Winmm driver and OpenAL32.dll thunk was Re: OpenAL and DirectSound
by Nick Burns 23 Nov '06

23 Nov '06

Attached is my diff/patch for both an openal winmm driver and an openal32.dll thunk I would like to get these into wine -- please comment and stuff... ----Rant follows---- I seem to be making a habit of responding to myself... (this antihistamine is kicking in) Sometimes in life you feel retarded -- at least this time I have an excuse (I am sick) Anyhow -- I took all sorts of care to make sure that my OpenAL thunk was CDECL -- I even made sure that my spec file was perfect... But, I for some reason (unknown to even the person who wrote this code -- me) I used WINAPI (stdcall) in all the actual impl of the OpenAL thunk This explains why 'WINEDEBUG=relay' worked just fine -- but it failed in odd ways without the relay. I fixed that in my thunk -- and all seems well Tested with UT2004Demo (uses openal) hello_world.exe (al example code) playfile.exe (al example code) On Mac Book Pro running Mac OSX 10.4.8 and tot wine ut2004 notes -- in System (this can be done in the gui -- uncheck (yes uncheck -- the gui seems backwards) 'use system driver' under audio) mv DefOpenAL32.dll DefOpenAL32.dll_ UT2004.ini -- change [ALAudio.ALAudioSubsystem] UseDefaultDriver=False then run WINEDEBUG=trace+openal wine UT2004.exe Also I did play ut2004 -- menus and 2 bot matches sounded fine (though the mouse seems sluggish...) (not tested in linux) - Nick >From: "Nick Burns" <adger44(a)hotmail.com> >To: wine-devel(a)winehq.org >Subject: Re: OpenAL and DirectSound -- openal32.dll first pass >Date: Mon, 20 Nov 2006 20:43:05 -0800 > >Sorry for the double post -- was unaware that 'wine.dev(a)web.de' was the >same thing... > >I was sick last night and could not sleep .: I wrote up an openal32.dll >thunk. > >I will be testing with ut2004/and the demos (hello_world and playfile) very >soon... > >I get sound with tracing but not without -- when it works ill submit a >patch... > >- Nick > >>From: "Nick Burns" <adger44(a)hotmail.com> >>To: wine.dev(a)web.de, stefandoesinger(a)gmx.at >>CC: wine-devel(a)winehq.org >>Subject: Re: OpenAL and DirectSound >>Date: Mon, 20 Nov 2006 00:29:42 -0800 >> >>I would really like to get my openal driver into wine... >>But we should also have an openal32.dll >> >>For example -- on my Mac Book Pro running Mac OSX 10.4.8 and tot wine >>If a windows application uses openal it will go from openal(windows) -> >>dsound HEL(wine) -> openal(osx) >>This means the audio data has soo much fun traveling from api to api -- >>kinda like a code rollercoaster... >> >>Anyhow -- I will resubmit my openal patch -- maybe if I get some time I >>will try to make a thunk (or I could help ... or work more)... >> >>Also I would like help with my openal patch (get hw accel working for >>dsound and dsound3d) >> >>- Nick >> >>>From: Detlef Riekenberg <wine.dev(a)web.de> >>>To: Stefan Dösinger <stefandoesinger(a)gmx.at> >>>CC: wine-devel(a)winehq.org >>>Subject: Re: OpenAL and DirectSound >>>Date: Fri, 17 Nov 2006 20:06:01 +0100 >>> >>>On Fr, 2006-11-17 at 17:31 +0100, Stefan Dösinger wrote: >>> >>> > Just stumbled uppon some interesting thing about openal and >>>DirectSound 3D on >>> > vista: http://www.openal.org/openal_vista.html >>> > >>> > It sounds like openal will be the old way to get hw accelerated 3d >>>sound on >>> > vista, and creative is writing a dsound -> openal wrapper. Maybe it >>>gets a >>> > proper license so we can include it in wine :-) >>> > >>> > Maybe we should write an openal thunk for wine? Many games use openal, >>>and it >>> > seems that on wine openal uses directsound. That might work more >>>better and >>> > faster if we had a openal.dll which directly called linux openal :-) >>> > >>> > Just dropping the suggestions :-) >>> >>>There was something a while ago... >>> >>> >>>Google-Result for "wineopenal": >>>http://www.winehq.com/pipermail/wine-devel/2006-June/048273.html >>> >>> >>>-- >>> >>>By by ... Detlef >>> >> >> >> >> > > > >

3 10

Wine internship available
by Dan Kegel 23 Nov '06

23 Nov '06

James Hawkins is wrapping up his internship with me next month, and boy will I miss him! (See http://cia.navi.cx/stats/author/James%20Hawkins%20%3Ctruiken(a)gmail.com%3E for one reason why!) If you know any students who are good programmers who would like to get paid to work on Wine for a term, please send them to http://www.kegel.com/academy/ Thanks! - Dan

1 0

Suggestions on your FAQ
by M.D. van Es 23 Nov '06

23 Nov '06

Hello, Searching your website http://www.winehq.org for information on release 1.0of Wine, I read your FAQ and found some items that appear to be out of date. Hopefully, you'll find the following remarks helpful. 1. - following the link: http://www.winehq.com/site/docs/wine-faq/index#ARE-HERE-ANY-ALTERNATIVES-TO… - that is paragraph 2.3 - here is nothed that both VMWare and Win4Lin cost money. To my knowledge, some months ago, VMWare has released some free software, first the Viewer and later on the VMWare Server, although it remains closed source and in my opinion cannot be compared to Wine in that it does emulate. 2. - following the link: http://www.winehq.com/site/docs/wine-faq/index#WHEN-WILL-WINE-BE-FINISHED - that is paragraph 2.12 - The current release being referred to in the last line is Wine 0.9 to be released around September 30th 2005. I'm sure this is not valid. In fact, I read you were working on release 1.0 for the end of 2006. http://tweakers.net/nieuws/44607/Plannen-voor-Wine-1.0-gepresenteerd.html(D…. Succes with your work, I'm enjoying it every day. Regards, Martin van Es

1 0

Problem with optary.h
by Francois Gouget 23 Nov '06

23 Nov '06

I have done a test compilation of the Wine conformance tests on Windows and found that dlls/mshtml/tests/misc.c fails to compile because it depends on the optary.h header. The problem is that this header is not part of the Platform SDK or of the DirectX SDK. Apparently it is only part of the IE6 SDK, which seems to be deprecated now. At least that's the impression I got from: http://msdn.microsoft.com/archive/default.asp?url=/archive/en-us/samples/in… This raises the following questions: * Should optary.h really be in Wine's global include directory? * How should one go about compiling misc.c with the Windows headers? Should we add the IE SDK as a requirement for compiling the Wine conformance tests? * Is optary.h still present in the IE 7 SDK (or equivalent)? -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Linux: the choice of a GNU generation

1 0