October 2006 - Wine-Devel - list.winehq.org

Re: Regression in patch wined3d: Implement D3DSIO_MOVA in ARB backend.

by Tom Wickline

On 10/28/06, Mirek <thunder.m(a)czela.net> wrote: > Yes I did that this way. I tried it three times as it is now in CVS > (3DMark 2003 and Flatout 2 not working) and three times without this > patch (only) (3DMark 2003 and Flatout 2 works just perfect). > > Mirek > I see the same thing as Mirek, 3dmark03 is totally broken now. Tom

Oct. 29, 2006

Re: [SM-Spell-Submit] Wine release 0.9.24

by David Kowis

Alexandre Julliard wrote: > This is release 0.9.24 of Wine, a free implementation of Windows on Unix. > > What's new in this release: > - Support for multiple monitors using Xinerama. W00t! Now if I could get it to build in amd64 :( > - Various MSI fixes and improvements. > - A ton of memory leaks fixed. > - Many common controls fixes. > - Lots of bug fixes. > -- David Kowis Cauldron Component Lead SourceMage GNU/Linux -- www.sourcemage.org Progress isn't made by early risers. It's made by lazy men trying to find easier ways to do something. - Robert Heinlein

Oct. 29, 2006

REGRESSION: 0.9.24 crashes on regsvr32 msvbvm60.dll, temporary workaround attached

by a_villacis＠palosanto.com

Somewhere between 0.9.23 and 0.9.24, a commit introduced a bug in oleaut32 which results in a nasty crash when parsing some MSFT typelibs. Notably, regsvr32 msvbvm60.dll" will crash on this bug, although I experienced it with DBGRID32.OCX too. I have traced the point of the crash to typelib.c:1814 if ( pFuncRec->FKCCIC & 0x2000 ) { (*pptfd)->Entry = SysAllocString((WCHAR*)pFuncRec->OptAttr[2]); } Apparently, 0x2000 as a flag in FKCCIC indicates that pFuncRec->OptAttr[2] is a pointer to some string. If what little understanding I have of typelib loading is correct, these typelibs are read from DLL resources on disk. Therefore, I fail to grasp how they can possibly refer to valid memory locations. I have this attached patch that temporarily plugs the crash, but I don't think the code is right in the first place. Or am I missing some crucial fact about typelib parsing? >From compare with version 1.266: if ( pFuncRec->FKCCIC & 0x2000 ) { (*pptfd)->Entry = (WCHAR*) pFuncRec->OptAttr[2] ; } I would dare to say that it has always been wrong, only now is crashing because the code now attempts to strdup() it with SysAllocString. Alex Villacís Lasso

Oct. 29, 2006

Re: Regression in patch wined3d: Implement D3DSIO_MOVA in ARB backend.

by Mirek

Yes I did that this way. I tried it three times as it is now in CVS (3DMark 2003 and Flatout 2 not working) and three times without this patch (only) (3DMark 2003 and Flatout 2 works just perfect). Mirek Ivan Gyurdiev napsal(a): > Mirek wrote: > >> I tried it three times with same result. Can someone else test if >> 3DMark 2003 is working with latest wine? > > > You shouldn't test if it's working now - test if it's working before the > MOVA patch, and if it's not working after the MOVA patch [ no other > patches in between ]. > > > >

Oct. 28, 2006

Re: Regression in patch wined3d: Implement D3DSIO_MOVA in ARB backend.

by Ivan Gyurdiev

Mirek wrote: > I tried it three times with same result. Can someone else test if > 3DMark 2003 is working with latest wine? You shouldn't test if it's working now - test if it's working before the MOVA patch, and if it's not working after the MOVA patch [ no other patches in between ].

Oct. 28, 2006

Re: Regression in patch wined3d: Implement D3DSIO_MOVA in ARB backend.

by Mirek

I tried it three times with same result. Can someone else test if 3DMark 2003 is working with latest wine? Mirek Ivan Gyurdiev napsal(a): > The MOVA instruction is a 2.0 instruction, which cannot be fully > implemented in ARB. > However, the patch shouldn't have caused a regression - if you're seeing > a black screen with GLSL also, that means another patch is responsible > for this. > > >

Oct. 28, 2006

Re: Regression in patch wined3d: Implement D3DSIO_MOVA in ARB backend.

by Ivan Gyurdiev

The MOVA instruction is a 2.0 instruction, which cannot be fully implemented in ARB. However, the patch shouldn't have caused a regression - if you're seeing a black screen with GLSL also, that means another patch is responsible for this.

Oct. 28, 2006

Re: dbghelp: Cast-qual warnings fix (1 of 2)

by Michael [Plouj] Ploujnikov

On 10/28/06, Michael [Plouj] Ploujnikov <ploujj(a)gmail.com> wrote: > On 10/28/06, Andrew Talbot <Andrew.Talbot(a)talbotville.com> wrote: > > Changelog: > > dbghelp: Cast-qual warnings fix. > > > > diff -urN a/dlls/dbghelp/path.c b/dlls/dbghelp/path.c > > --- a/dlls/dbghelp/path.c 2006-07-13 16:13:33.000000000 +0100 > > +++ b/dlls/dbghelp/path.c 2006-10-28 16:06:46.000000000 +0100 > > @@ -349,6 +349,7 @@ > > struct process* pcs = process_find_by_handle(hProcess); > > char tmp[MAX_PATH]; > > char* ptr; > > + char* buf; > > const char* filename; > > const char* searchPath = inSearchPath; > > > > @@ -360,7 +361,6 @@ > > if (!searchPath) > > { > > unsigned len = WideCharToMultiByte(CP_ACP, 0, pcs->search_path, -1, NULL, 0, NULL, NULL); > > - char* buf; > > > > searchPath = buf = HeapAlloc(GetProcessHeap(), 0, len); > > From first observation the above line seems to be missing a + before it. Never mind. -- () ASCII Ribbon Campaign /\ - against HTML mail & vCards

Oct. 28, 2006

Re: Regression in patch wined3d: Implement D3DSIO_MOVA in ARB backend.

by Mirek

Hi, screen is just black in those two apps, other apps (3DMark 2001, 2005, 2006, NFS MW, Oblivion, TRL, W3, WoW, Prey, Starcraft, Diablo 2, GTA SA, HL 2 EO, HL, NFS 3, NFS 4, Turok, Ignition, Carmageddon 2, Carmageddon 3, Pandemonium 2) works without any problem. Here is screenshot from 3DMark 2003 with GLSL enabled and disabled: http://www.czela.net/pub/Czela_Debian_2.5/mirek/wine/regression/3DMark2003_… http://www.czela.net/pub/Czela_Debian_2.5/mirek/wine/regression/3DMark2003_… My software: Debian Sid, kernel 2.6.18.1 SMP, nVidia 9625 drivers. My hardware: Core 2 Extreme, GF 6800 GS, 1GB RAM Do you need anything else? Mirek Stefan Dösinger napsal(a): > Am Samstag 28 Oktober 2006 10:35 schrieb Mirek: > >>Hi, i just installed latest version of wine and now 3DMark 2003 and >>Flatout 2 is not working anymore, i look at the patches commited >>yesterday and found this patch which is causing this regression. > > In which way do they not work any more? Is there any debug output, or can you > provide screenshots? > > Can you test ARB and glsl shaders? > > > ------------------------------------------------------------------------ > >

Oct. 28, 2006

Re: dbghelp: Cast-qual warnings fix (1 of 2)

by Michael [Plouj] Ploujnikov

On 10/28/06, Andrew Talbot <Andrew.Talbot(a)talbotville.com> wrote: > Changelog: > dbghelp: Cast-qual warnings fix. > > diff -urN a/dlls/dbghelp/path.c b/dlls/dbghelp/path.c > --- a/dlls/dbghelp/path.c 2006-07-13 16:13:33.000000000 +0100 > +++ b/dlls/dbghelp/path.c 2006-10-28 16:06:46.000000000 +0100 > @@ -349,6 +349,7 @@ > struct process* pcs = process_find_by_handle(hProcess); > char tmp[MAX_PATH]; > char* ptr; > + char* buf; > const char* filename; > const char* searchPath = inSearchPath; > > @@ -360,7 +361,6 @@ > if (!searchPath) > { > unsigned len = WideCharToMultiByte(CP_ACP, 0, pcs->search_path, -1, NULL, 0, NULL, NULL); > - char* buf; > > searchPath = buf = HeapAlloc(GetProcessHeap(), 0, len); >From first observation the above line seems to be missing a + before it. > if (!searchPath) return FALSE; > @@ -382,7 +382,7 @@ > { > strcpy(buffer, full_path); > if (searchPath != inSearchPath) > - HeapFree(GetProcessHeap(), 0, (char*)searchPath); > + HeapFree(GetProcessHeap(), 0, buf); > return TRUE; > } > > @@ -404,11 +404,11 @@ > { > strcpy(buffer, tmp); > if (searchPath != inSearchPath) > - HeapFree(GetProcessHeap(), 0, (char*)searchPath); > + HeapFree(GetProcessHeap(), 0, buf); > return TRUE; > } > } > if (searchPath != inSearchPath) > - HeapFree(GetProcessHeap(), 0, (char*)searchPath); > + HeapFree(GetProcessHeap(), 0, buf); > return FALSE; > } > > > -- () ASCII Ribbon Campaign /\ - against HTML mail & vCards

Oct. 28, 2006