August 2008 - Wine-Devel - list.winehq.org

Re: Opengl32: add a WGL_ARB_create_context / OpenGL 3.0 test
by Chris Robinson Aug. 26, 2008

Aug. 26, 2008

On Tuesday 26 August 2008 01:17:18 pm Roderick Colenbrander wrote: > + int attribs[] = {WGL_CONTEXT_MAJOR_VERSION_ARB, 3, > WGL_CONTEXT_MINOR_VERSION_ARB, 0, 0}; > + HGLRC gl3Ctx = > pwglCreateContextAttribsARB(hdc, 0, attribs); > + ok(gl3Ctx != NULL, > "Creation of an OpenGL 3.0 context failed!\n"); I'm not sure it should be considered a test failure if GL3 isn't supported..

1 0

Re: [10/10] WineX11: add a run-total fps counter to the fps debug channel
by Detlef Riekenberg Aug. 26, 2008

Aug. 26, 2008

On Mo, 2008-08-25 at 09:32 -0500, Stefan Dösinger wrote: > @@ -3486,15 +3486,19 @@ BOOL X11DRV_SwapBuffers(X11DRV_PDEVICE > *physDev) > /* FPS support */ > if (TRACE_ON(fps)) > { > - static long prev_time, frames; > + static long prev_time, start_time; > + static unsigned long frames, frames_total; > > DWORD time = GetTickCount(); > frames++; > + frames_total++; > /* every 1.5 seconds */ > if (time - prev_time > 1500) { > - TRACE_(fps)("@ approx %.2ffps\n", 1000.0*frames/(time - > prev_time)); > + TRACE_(fps)("@ approx %.2ffps, total %.2ffps\n", > + 1000.0*frames/(time - prev_time), > 1000.0*frames_total/(time - start_time)); > prev_time = time; > frames = 0; > + if(start_time == 0) start_time = time; The first fps report is not correct (0 as prev_time is not the time, when the device was created) Much more Important is the last fps report. Thiat can be from the actual frame up to 1.5 seconds old. When a DirectX test run ~10 sec, the last fps line is one second old. Any change in the Driver, that increase or decrease the needed test-time between 9 sec and 10.5 sec is not detected. You need to make the counters globals (or move it in X11DRV_PDEVICE) and dump a last fps report, when the device is destroyed. I suggest to rename the variable "time", to make clear what it is and to avoid a mix-up with time(). Example: "last_swapbuffers_time" -- By by ... Detlef

2 1

Re: shlwapi/tests: test UrlUnEscape inplace behavior (1/2) (try 2)
by Alexandre Julliard Aug. 26, 2008

Aug. 26, 2008

"Dan Kegel" <dank(a)kegel.com> writes: > Add test showing UrlUnEscape should convert the url in place > when URL_UNESCAPE_INPLACE is set, even if the output buffer is NULL. > Passes on XP. > > Changes since originally posted by Louis Lenders as > http://www.winehq.org/pipermail/wine-patches/2008-August/059567.html > 1) added todo_wine around parts that fail in Wine > 2) he has requested that somebody else take it over, > so I am raising my hand. If you are resending a patch written by someone else you need to include a From: line with the complete attribution. Alternatively if you have made significant changes you need to add a "Based on a patch by" note in the log message. -- Alexandre Julliard julliard(a)winehq.org

2 3

Re: imagehlp(1/2): Add tests for ImageGetDigestStream
by Detlef Riekenberg Aug. 26, 2008

Aug. 26, 2008

On Mo, 2008-08-18 at 09:00 -0700, Juan Lang wrote: > dlls/imagehlp/tests/Makefile.in | 13 + > dlls/imagehlp/tests/main.c | 1222 > +++++++++++++++++++++++++++++++++++++++ > 2 files changed, 1235 insertions(+), 0 deletions(-) > +/* "Minimal" executable from compiling "int main(void) { return 0; }" > */ > +static const BYTE bin[] = { The Patch is really large. TCC creates smaller Images: http://bellard.org/tcc/ (But I did not test your Patch with that binary) -- By by ... Detlef

1 0

Re: Add TRACEs to NtCreateFile returns (second try)
by James Hawkins Aug. 26, 2008

Aug. 26, 2008

2008/8/24 Markus Hitter <mah(a)jump-ing.de>: > + if (!attr || !attr->ObjectName) + { + TRACE("returning STATUS_INVALID_PARAMETER\n"); + return STATUS_INVALID_PARAMETER; + } These are all very useless TRACES, except for possibly the returned handle value. You should be able to look at the traced parameters and see if there is an invalid parameter. As a side note of something I just noticed, the check for NULL attr will never be true because we'll crash in the TRACE when we dereference attr. -- James Hawkins

4 8

Re: [crypt32/tests] Use CryptAcquireContextA instead of CryptAcquireContextW
by Detlef Riekenberg Aug. 26, 2008

Aug. 26, 2008

On Mo, 2008-08-25 at 22:16 +0200, Paul Vriens wrote: > static BOOL (WINAPI * pCryptAcquireContextA) > (HCRYPTPROV *, LPCSTR, LPCSTR, DWORD, DWORD); > -static BOOL (WINAPI * pCryptAcquireContextW) > - (HCRYPTPROV *, LPCWSTR, LPCWSTR, DWORD, > DWORD); > @@ -47,7 +45,6 @@ static void init_function_pointers(void) > trace("GetProcAddress(%s) failed\n", #func); > > GET_PROC(hAdvapi32, CryptAcquireContextA) > - GET_PROC(hAdvapi32, CryptAcquireContextW) > @@ -2799,10 +2794,10 @@ static BOOL detect_nt(void) > signer.pCertInfo = &certInfo; > signer.HashAlgorithm.pszObjId = oid_rsa_md5; > > - ret = pCryptAcquireContextW(&signer.hCryptProv, cspNameW, NULL, > + ret = pCryptAcquireContextA(&signer.hCryptProv, cspNameA, NULL, > PROV_RSA_FULL, CRYPT_NEWKEYSET); > - ret = pCryptAcquireContextW(&signer.hCryptProv, cspNameW, > NULL, > + ret = pCryptAcquireContextA(&signer.hCryptProv, cspNameA, > NULL, > PROV_RSA_FULL, 0); > } > > @@ -2810,7 +2805,7 @@ static BOOL detect_nt(void) > > /* cleanup */ > CryptReleaseContext(signer.hCryptProv, 0); > - pCryptAcquireContextW(&signer.hCryptProv, cspNameW, NULL, > PROV_RSA_FULL, > + pCryptAcquireContextA(&signer.hCryptProv, cspNameA, NULL, > PROV_RSA_FULL, > CRYPT_DELETEKEYSET); This changes are wrong. Please compare your change with the name of the last Function. The code is expected to Fail on Win9x. This should work as first code in detect_nt(): + if (!pCryptAcquireContextW) return FALSE; Thanks -- By by ... Detlef

2 1

Re: regedit: Fix concat handling for ANSI files
by Alexandre Julliard Aug. 26, 2008

Aug. 26, 2008

Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> writes: > + do > + { > + *s = fgetc(in); > + } while(*s == ' ' || *s == '\t'); > + > + if(*s == EOF) > fprintf(stderr,"%s: ERROR - invalid continuation.\n", > getAppName()); You can't do that, EOF is an int not a char. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: shlwapi: [PATCH 3/4] Fix results for UrlApplyScheme
by Dmitry Timoshkov Aug. 26, 2008

Aug. 26, 2008

"Detlef Riekenberg" <wine.dev(a)web.de> wrote: > Subject: [PATCH 2/4] shlwapi: Fix results for UrlApplyScheme This subject doesn't match the e-mail subject line, and may confuse things. > in = HeapAlloc(GetProcessHeap(), 0, > - (2*INTERNET_MAX_URL_LENGTH) * sizeof(WCHAR)); > + (2*INTERNET_MAX_URL_LENGTH) * sizeof(WCHAR)); > out = in + INTERNET_MAX_URL_LENGTH; > > - MultiByteToWideChar(0, 0, pszIn, -1, in, INTERNET_MAX_URL_LENGTH); > + MultiByteToWideChar(CP_ACP, 0, pszIn, -1, in, INTERNET_MAX_URL_LENGTH); It would simplify things to allocate the buffer on the stack, since the buffer is small and its size is known. > +/************************************************************************* > + */ > + > static HRESULT URL_GuessScheme(LPCWSTR pszIn, LPWSTR pszOut, LPDWORD pcchOut) > { It's a usual convention to not leave an empty line after the API/function description comment, and adding an empty comment is not useful at all. -- Dmitry.

1 0

Re: comdlg32: Update Japanese translation
by Alexandre Julliard Aug. 26, 2008

Aug. 26, 2008

"Muneyuki Noguchi" <nogu.dev(a)gmail.com> writes: > +/****************************************************************/ > +/* Japanese neutral resources > +/****************************************************************/ > + > +LANGUAGE LANG_JAPANESE, SUBLANG_NEUTRAL You shouldn't need neutral resources for Japanese, there's only one sublanguage. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [2/3] d3dx9: Add d3dx9effect.h
by Alexandre Julliard Aug. 26, 2008

Aug. 26, 2008

Busquets Pérez Luis C. <luiscar2001(a)yahoo.es> writes: > + /*** ID3DXEffect ***/ > + STDMETHOD(GetPool)(THIS_ LPD3DXEFFECTPOOL* ppPool) PURE; > + STDMETHOD(SetTechnique)(THIS_ D3DXHANDLE hTechnique) PURE; > + STDMETHOD_(D3DXHANDLE, GetCurrentTechnique)(THIS) PURE; > + STDMETHOD(ValidateTechnique)(THIS_ D3DXHANDLE hTechnique) PURE; > + STDMETHOD(FindNextValidTechnique)(THIS_ D3DXHANDLE hTechnique, D3DXHANDLE *pTechnique) PURE; > + STDMETHOD_(BOOL, IsParameterUsed)(THIS_ D3DXHANDLE hParameter, D3DXHANDLE hTechnique) PURE; > + STDMETHOD(Begin)(THIS_ UINT *pPasses, DWORD Flags) PURE; > + STDMETHOD(Pass)(THIS_ UINT Pass) PURE; > + STDMETHOD(End)(THIS) PURE; > + STDMETHOD(GetDevice)(THIS_ LPDIRECT3DDEVICE9* ppDevice) PURE; > + STDMETHOD(OnLostDevice)(THIS) PURE; > + STDMETHOD(OnResetDevice)(THIS) PURE; > + STDMETHOD(CloneEffect)(THIS_ LPDIRECT3DDEVICE9 pDevice, LPD3DXEFFECT* ppEffect) PURE; This doesn't match the headers I have. Is the interface different between d3dx9 versions? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: How should I name the Traditional Chinese resource file ?
by Alexandre Julliard Aug. 26, 2008

Aug. 26, 2008

Hongbo Ni <hongbo(a)njstar.com> writes: > 1. name it Zh_tw.rc, (but Alexandre seems not like it) > 2. name it Tw.rc > 3. rename all Zh.rc to Cn.rc, then name traditional as Zh.rc > 4. just put both simplified and traditional Chinese in Zh.rc > (not good as they are different encoding). > > My question is how should I name it ? The best solution is 4, we try to use a single file per language. If encoding is an issue you can convert the file to utf-8. -- Alexandre Julliard julliard(a)winehq.org

2 2

Patchwatcher status: regenerates configure, notifies authors on failure, filtered patches mailing list
by Dan Kegel Aug. 26, 2008

Aug. 26, 2008

Patchwatcher now regenerates configure and makefiles after each patch, so it can handle patches that change configure.ac. I have enabled email notification to authors of patches that don't build. The script now forwards patches that pass all tests to the mailing list http://groups.google.com/group/wine-patches-filtered so potentially Alexandre could start looking only at that list (though it probably only has two nines of reliability at this point, as patchwatcher doesn't handle every case yet, e.g. it doesn't handle partial resends of patch series).

10 19

winetricks: add ie6 verb
by Steven Edwards Aug. 26, 2008

Aug. 26, 2008

This might be of use to someone wanting to compare Wine IE gecko rendering to IE6 without having to use a bigger hack like ies4linux. I am sending here rather than privately to Dan as I don't expect it to be merged to winetricks mainline Changelog: Add ie6 verb and support IE6 <rant> I doubt Dan will merge it (no disrespect intended, I am not trying to single him out) because it might be viewed as 'lacking sense taste' whatever that is supposed to mean, see http://www.winehq.org/pipermail/wine-devel/2008-July/067109.html regarding a script for Media player and codecs. I don't see what could be poor taste if Wine is going to have to reimplement this stuff anyway, to provide support for it in winetricks so developers can compare native verses builtin. All the better if some users find it of use too. I would see something like Office being added as crossing the line Dan has drawn for winetricks being a developer tool rather than a user tool but that is not really my point in any case. We've been accused in private by other that Bordeaux will not, or does not contribut back to Wine, leeches off of Winehq in a Transgaming like style, etc. Even though 90% of my patches and ideas (however small they are) have been rejected and even with IE being a nice part of the revenue stream, I've decide to give the hack back in all of its glory. We are committed to maintain NO "proprietary advantages" and seeing everything that runs on Bordeaux work out of the box on the next Wine stable release. We don't have the development staff or funding yet to fully ramp up and be as active as we would like to be but here is another token of our commitment. </rant> -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

2 2

ddraw error in Tibia (wine told me to post here)
by Michael Reiley Aug. 26, 2008

Aug. 26, 2008

There's some progress on the MMORPG Tibia since the last wine version, but now there's a new error. before you tell me "This list is for discussions regarding Wine development. It is not for bug reports or user support questions", I'd like to tell you "err:ddraw:IDirectDrawImpl_QueryInterface (0x13b5d0) You may want to contact wine-devel for help" Assuming wine knows what it's talking about, I'm going to post the error log here and see what all of you have to say. fuzzypig(a)fp-ubuntu-laptop:~/.wine/drive_c/Program Files/Tibia$ wine Tibia.exe engine 0 fixme:win:EnumDisplayDevicesW ((null),0,0x32e468,0x00000000), stub! err:ddraw:IDirectDrawImpl_QueryInterface (0x13b5d0) The App is requesting a D3D device, but a non-OpenGL surface type was choosen. Prepare for trouble! err:ddraw:IDirectDrawImpl_QueryInterface (0x13b5d0) You may want to contact wine-devel for help err:ddraw:IDirectDrawSurfaceImpl_QueryInterface No interface fixme:d3d:WineD3D_ChoosePixelFormat Add OpenGL context recreation support to SetDepthStencilSurface fixme:d3d:CreateContext >>>>>>>>>>>>>>>>> GL_INVALID_OPERATION (0x502) from glClearColor @ context.c / 514 fixme:d3d_shader:state_texfactor_arbfp >>>>>>>>>>>>>>>>> GL_INVALID_ENUM (0x500) from glProgramEnvParameter4fvARB(GL_FRAGMENT_PROGRAM_ARB, ARB_FFP_CONST_TFACTOR, col) @ arb_program_shader.c / 2327 fixme:d3d_shader:set_bumpmat_arbfp >>>>>>>>>>>>>>>>> GL_INVALID_ENUM (0x500) from glProgramEnvParameter4fvARB(GL_FRAGMENT_PROGRAM_ARB, ARB_FFP_CONST_BUMPMAT(stage), &mat[0][0]) @ arb_program_shader.c / 2361 fixme:d3d_shader:set_bumpmat_arbfp >>>>>>>>>>>>>>>>> GL_INVALID_ENUM (0x500) from glProgramEnvParameter4fvARB(GL_FRAGMENT_PROGRAM_ARB, ARB_FFP_CONST_BUMPMAT(stage), &mat[0][0]) @ arb_program_shader.c / 2361 fixme:d3d_shader:set_bumpmat_arbfp >>>>>>>>>>>>>>>>> GL_INVALID_ENUM (0x500) from glProgramEnvParameter4fvARB(GL_FRAGMENT_PROGRAM_ARB, ARB_FFP_CONST_BUMPMAT(stage), &mat[0][0]) @ arb_program_shader.c / 2361 fixme:d3d_shader:set_bumpmat_arbfp >>>>>>>>>>>>>>>>> GL_INVALID_ENUM (0x500) from glProgramEnvParameter4fvARB(GL_FRAGMENT_PROGRAM_ARB, ARB_FFP_CONST_BUMPMAT(stage), &mat[0][0]) @ arb_program_shader.c / 2361 fixme:d3d_shader:set_bumpmat_arbfp >>>>>>>>>>>>>>>>> GL_INVALID_ENUM (0x500) from glProgramEnvParameter4fvARB(GL_FRAGMENT_PROGRAM_ARB, ARB_FFP_CONST_BUMPMAT(stage), &mat[0][0]) @ arb_program_shader.c / 2361 Happy trails! ~Michael Reiley

1 0

Re: winex11.drv: Fix buffer overflow bug in X11DRV_KeyEvent() andX11DRV_ToUnicodeEx()
by Dmitry Timoshkov Aug. 25, 2008

Aug. 25, 2008

"Muneyuki Noguchi" <nogu.dev(a)gmail.com> wrote: > winex11.drv: Fix buffer overflow bug in X11DRV_KeyEvent() and > X11DRV_ToUnicodeEx() > + Str = (char *)malloc(64); > + if (Str == NULL) > + ERR("Failed to allocate memory!\n"); Please don't use malloc() in Wine, use win32 Heap*** APIs instead. Also, you need to properly handle memory allocation errors, not just print an ERR. -- Dmitry.

2 1

Using wine directdraw dll on windows?
by Joseph Garvin Aug. 25, 2008

Aug. 25, 2008

Intel's windows vista graphics drivers are pretty terrible, and some older directdraw games won't run at all with them. I know that there was talk before about using wine's opengl-based direct3d implementation on windows, so I was curious if it was at all possible to do the same with directdraw. Ironically Worms Armageddon runs better on my laptop under wine than it does on my gf's under Vista. Thanks, Joe

3 3

New steam survey statistics?
by Scott Ritchie Aug. 25, 2008

Aug. 25, 2008

When we last heard from Valve a year ago, Wine had about a .4% share of the steam user base. It would be interesting to see how this has changed since then - in that time, we've both released Wine 1.0 and likely seen Linux adoption increase a great deal. Linux had doubled its desktop marketshare the year preceding the last survey, I wouldn't be surprised to see Wine usage at around 1% on steam now. Thanks, Scott Ritchie

4 5

Re: cryptnet: Sign-compare warning fix (try 2)
by Juan Lang Aug. 25, 2008

Aug. 25, 2008

Hi Andy, + LONG last_error; ret = CryptGetObjectUrl(URL_OID_CERTIFICATE_CRL_DIST_POINT, rgpvContext[i], 0, NULL, &cbUrlArray, NULL, NULL, NULL); - if (!ret && GetLastError() == CRYPT_E_NOT_FOUND) + last_error = GetLastError(); + if (!ret && last_error == CRYPT_E_NOT_FOUND) I don't get it. How does using a temporary LONG rather than GetLastError directly remove the warning? Either the compiler was wrong about the first warning, or it's wrong about not warning about this form. I don't see the utility of this patch. Furthermore, any patches that remove warnings due to sign mismatches between GetLastError and an error in winerror.h reflect an error on Microsoft's part, not on Wine's, so I don't see the point of removing them. --Juan

2 4

Re: [3/3] d3dx9: Begin Implementation of ID3DXEffect and D3DXCreateEffectEx
by tony.wasserka＠freenet.de Aug. 25, 2008

Aug. 25, 2008

Hi, > +HRESULT WINAPI D3DXCreateEffectEx(LPDIRECT3DDEVICE9 pdevice, LPCVOID psrcdata, UINT srcdatalen, D3DXMACRO *pdefines, > + LPD3DXINCLUDE pinclude, LPCSTR pskipconstants, DWORD flags, LPD3DXEFFECTPOOL ppool, LPD3DXEFFECT * ppeffect, > + LPD3DXBUFFER * ppcompilationerrors) > +{ > + ID3DXEffectImpl *This; > + This=(ID3DXEffectImpl *)malloc(sizeof(ID3DXEffectImpl)); > + This->lpVtbl=&ID3DXEffect_Vtbl; > + This->device=pdevice; > + return D3D_OK; > +} > + Don't use malloc for windows functions, use HeapAlloc for that. Also, you should do some error checking to make sure that the memory was allocated successfully, or this code will segfault. You forgot to assign the ppeffect pointer to This, i.e. *ppeffect=This. you'll need to check that ppeffect isn't NULL first, of course. At least ID3DXBaseEffectImpl_AddRef/Release/QueryInterface (well, actually only Release) need to be implemented in the same patch as the interface leaks memory otherwise (one more unnecessary valgrind error). Last thing, you need to AddRef the device when creating another reference to it, and Release it on ID3DXBaseEffectImpl_Release. (Though a device pointer isn't even neccessary yet, so you could also leave it away) Best regards, Tony Heute schon ge"freeMail"t? Jetzt kostenlose E-Mail-Adresse sichern! http://email.freenet.de/dienste/emailoffice/produktuebersicht/basic/mail/in…

1 0

RE: [try 3][1/2]wined3d: Support clipplanes with GLSL shaders
by Stefan Dösinger Aug. 25, 2008

Aug. 25, 2008

+ if(gl_info->max_glsl_varyings > 32) + gl_info->max_glsl_varyings -= 4; + if(!apple && gl_info->gl_vendor == VENDOR_ATI && gl_info->max_glsl_varyings == 40) { + WARN("Hardcoded 32 varyings instead of 44 to avoid crashing the ATI driver\n"); + gl_info->max_glsl_varyings -= 4; + } This will give you 36 varyings, not 32. Is this intentional? > -----Original Message----- > From: wine-patches-bounces(a)winehq.org [mailto:wine-patches- > bounces(a)winehq.org] On Behalf Of Fabian Bieler > Sent: Monday, August 25, 2008 6:25 AM > To: wine-patches(a)winehq.org > Subject: [try 3][1/2]wined3d: Support clipplanes with GLSL shaders >

1 0

Re: d3dx9: Add stubs for D3DXCreateEffectxx and D3DXCreateEffectFromFilexx
by tony.wasserka＠freenet.de Aug. 24, 2008

Aug. 24, 2008

Hi, There were quite a number of minor issues in you code: In Makefile.in: > C_SRCS = \ > d3dx9_36_main.c \ >+ effect.c \ > font.c \ Use tabs here, please, as the intendation switches here on other computers. In effect.c: + * modify it under the terms of the GNU Lesser General Public + GetCurre* License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. I guess that was a paste at the wrong position ;-) > +HRESULT D3DXCreateEffectEx( You missed the WINAPI for that one. In d3dx9effect.h: I haven't looked too much into the whole effect stuff, but this thing looks like Unicode isn't supported here (only LPCSTRs are used). Are you sure about that? >+#endif //__D3DX9EFFECT_H__ This is C code, use /* */, please. As David already said, you should split the patch into smaller chunks. Also, you need to convert a D3DXCreateEffect call into a D3DXCreateEffectEx call, look at font.c to see how it's done for D3DXCreateFont. This applies to the D3DXCreateEffectFromFile calls, too. Best regards, Tony Unbegrenzter Speicher, Top-Spamschutz, 120 SMS und eigene E-MailDomain inkl. http://office.freenet.de/dienste/emailoffice/produktuebersicht/power/mail/i…

1 0

Re: d3dx9: Add stubs for D3DXCreateEffectxx and D3DXCreateEffectFromFilexx
by Henri Verbeet Aug. 24, 2008

Aug. 24, 2008

2008/8/24 Busquets Pérez Luis C. <luiscar2001(a)yahoo.es>: > Why are you adding stubs for these? I don't think having a stub implementation is useful here.

1 0

Re: d3dx9: Add stubs for D3DXCreateEffectxx and D3DXCreateEffectFromFilexx
by David Adam Aug. 24, 2008

Aug. 24, 2008

Hello, A few remarks: 1) this patch is too huge You need to split it in several patches. I advice you to do a patch for - d3dx9effect.h -an another one for d3dx9shader.h -an another one for D3DXCreateEffect -an another one for D3DXCreateEffectExW and D3DXCreateEffectExW 2) for the patch D3DXCreateEffect you need to stub not only this function but also, all the methods coming with this function David

1 0

Re: [10/10] WineD3D: Sort out some limit confusion
by Henri Verbeet Aug. 24, 2008

Aug. 24, 2008

2008/8/22 Stefan Dösinger <stefan(a)codeweavers.com>: > + glGetIntegerv(GL_MAX_TEXTURE_COORDS, &gl_max); > + gl_info->max_texture_coords = min(MAX_TEXTURES, gl_max); > + TRACE_(d3d_caps)("Max texture coords: %d\n", gl_info->max_texture_coords); I don't know if this is intentional or not, but did you mean to use gl_info->max_textures here? > @@ -3172,7 +3172,7 @@ static void loadTexCoords(IWineD3DStateBlockImpl *stateblock, WineDirect3DVertex > return; > } > > - for (textureNo = 0; textureNo < GL_LIMITS(texture_stages); ++textureNo) { > + for (textureNo = 0; textureNo < GL_LIMITS(texture_coords); ++textureNo) { > int coordIdx = stateblock->textureState[textureNo][WINED3DTSS_TEXCOORDINDEX]; > > mapped_stage = stateblock->wineD3DDevice->texUnitMap[textureNo]; Shouldn't that be testing mapped_stage against GL_LIMITS(texture_coords) instead? > - if (GL_SUPPORT(NV_REGISTER_COMBINERS)) { > - /* The number of the mapped stages increases monotonically, so it's fine to use the last used one */ > - for (textureNo = mapped_stage + 1; textureNo < GL_LIMITS(textures); ++textureNo) { > - GL_EXTCALL(glMultiTexCoord4fARB(GL_TEXTURE0_ARB + textureNo, 0, 0, 0, 1)); > - } > - } I'm probably missing something here, but why is it safe to remove that? > @@ -420,7 +420,7 @@ static void drawStridedSlow(IWineD3DDevice *iface, WineDirect3DVertexStridedData > } > > /* Texture coords --------------------------- */ > - for (textureNo = 0; textureNo < GL_LIMITS(texture_stages); ++textureNo) { > + for (textureNo = 0; textureNo < GL_LIMITS(texture_coords); ++textureNo) { > > if (!GL_SUPPORT(ARB_MULTITEXTURE) && textureNo > 0) { > FIXME("Program using multiple concurrent textures which this opengl implementation doesn't support\n"); Similar to above, shouldn't this test texture_idx against GL_LIMITS(texture_coords) instead of textureNo? In the bigger picture, these various limits are pretty tricky to get right, I wonder if it might make more sense to handle most of this stuff in the tex unit map instead. That way we could be sure that if a stage is mapped to a sampler that sampler is safe to use.

3 7

d3dx9: Effects
by Busquets Pérez Luis C. Aug. 24, 2008

Aug. 24, 2008

What is wrong with the attached patch? __________________________________________________ Correo Yahoo! Espacio para todos tus mensajes, antivirus y antispam ¡gratis! Regístrate ya - http://correo.yahoo.es

1 0