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

Broken FC5 packages - stay clear.
by Vitaliy Margolen 16 Jun '06

16 Jun '06

It seems that FC5 packager created a broken Wine distribution. Instead of being 1-3 packages it's 15 (from what user told me on #winehq. And it doesn't work at all: "Application tried to create a window, but no driver could be loaded." Which tells me it's compiled without X headers. I didn't even new Wine could compile without X, since we don't have tty driver. Vitaliy

9 23

Re: add stub for Direct3DShaderValidatorCreate9 (d3d9)
by Dmitry Timoshkov 16 Jun '06

16 Jun '06

"Louis. Lenders" <xerox_xerox2000(a)yahoo.co.uk> wrote: > It's not really possible to test this function on windows , but > i think this is quite a good guess for the function's prototype. > I've tested with quite some samples from www.codesampler.com. Other > implementations make the samples crash . > > (http://www.codesampler.com/source/dx9_cg_simple.zip is an example > of an app that calls this function) You're must be doing something not usual, but the .exe pointed out above doesn't contain Direct3DShaderValidatorCreate9 in its imports. -- Dmitry.

1 0

Re: Remove an assert in StgStreamImpl_Write()
by Robert Shearman 16 Jun '06

16 Jun '06

Francois Gouget wrote: > > Changelog: > > * dlls/ole32/stg_stream.c > > Aric Stewart <aric(a)codeweavers.com> > Remove an assert and instead just fail. > This allows iTunes to finish installing with the reboot. > A better fix would be to check the result of IStream_SetSize in this function, although I think the original problem may well be fixed after the STG_E_REVERTED checks. -- Rob Shearman

2 1

Re: Quartz: Improve a stub
by Robert Shearman 16 Jun '06

16 Jun '06

ivanleo(a)gmail.com wrote: > It looks like someone was implementing some stuff and left it half way > through, That was me. I didn't have any documentation on how the extensions processing was meant to work. > so the result is GetClassMediaFile was hard coded to fail > because process_extensions always fails. Err, no. Did you read the code? The fallback is the more common method of recognising a byte sequence at the start of the file. > This patch is needed by the > IAMMultiMediaStreamImpl_OpenFile which will follow. The two together > allow Crimson Skies to start and get to the menu instead of crashing. >--- winenew/dlls/quartz/filesource.c 2006-05-23 13:48:29.000000000 +0100 >+++ wine/dlls/quartz/filesource.c 2006-06-15 22:50:40.000000000 +0100 >@@ -65,7 +65,10 @@ > static HRESULT process_extensions(HKEY hkeyExtensions, LPCOLESTR pszFileName, GUID * majorType, GUID * minorType) > { > /* FIXME: implement */ >- return E_NOTIMPL; >+ >+ FIXME("(%p, %s, %p, %p), stub!\n", hkeyExtensions, debugstr_w(pszFileName), majorType, minorType); >+ >+ return S_OK; > } > > Great. Now majorType and minorType will be left uninitialised, causing weird and hard-to-debug problems later on. -- Rob Shearman

1 0

wined3d compile failure
by James Hawkins 16 Jun '06

16 Jun '06

Hi, I'm getting this compile failure as of today: make[2]: Entering directory `/home/jhawkins/wine/dlls/wined3d' gcc -c -I. -I. -I../../include -I../../include -I/usr/X11R6/include -D__WINESRC__ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o device.o device.c device.c: In function `IWineD3DDeviceImpl_SetRenderState': device.c:3833: `GL_FOG_COORD_SRC' undeclared (first use in this function) device.c:3833: (Each undeclared identifier is reported only once device.c:3833: for each function it appears in.) device.c:3873: `GL_FOG_COORD' undeclared (first use in this function) make[2]: *** [device.o] Error 1 make[2]: Leaving directory `/home/jhawkins/wine/dlls/wined3d' make[1]: *** [wined3d] Error 2 make[1]: Leaving directory `/home/jhawkins/wine/dlls' make: *** [dlls] Error 2 -- James Hawkins

4 4

how to debug a segfault in wine-preloader?
by Paul Davis 16 Jun '06

16 Jun '06

i am building ardour (http://ardour.org) as a win32 executable to allow us to continue using win32 DLL's as plugins. when i run the result with wine, i get a segfault long before control has been transferred to my code. it appears to be in wine-preloader (based on what "file core.NNNN" says). i cannot see how to debug this without endless printf's - wine-preloader appears to need wine to run before it, and gdb cannot cross exec boundaries. any hints on how to debug this sort of issue? btw, wine in general works fine, of course. --p

1 0

Re: Will pay $$ for fix of Magic: The Gathering Online bug
by Kyle Winkler 16 Jun '06

16 Jun '06

---------- Forwarded message ---------- From: Kyle Winkler <kylewinkler(a)gmail.com> Date: Jun 15, 2006 6:47 PM Subject: Re: Will pay $$ for fix of Magic: The Gathering Online bug To: EA Durbin <ead1234(a)hotmail.com> I'm currently using Kanotix 2005.04, 32 bit. However, according to my own experience and the info on wine's bug list (its bug 5404), this problem is universal to all distros. On 6/15/06, EA Durbin <ead1234(a)hotmail.com> wrote: > > > >On 6/14/06, Mike McCormack <mike(a)codeweavers.com> wrote: > >>hehe. If you're compiling Wine yourself, do I get $200 for telling you > >>to install the latest version of Fontforge, and make sure recent > >>freetype libraries are on your system? > > Which distro are you using, and is it 64 bit or 32 bit? > > I ran in to a similar problem on my 64 bit gentoo box. If you have 64 bit > linux make sure you copy freetype over to a directory containing your 32 > compatibility libraries. > > cp /usr/lib/libfreetype.* /emul/linux/x86/usr/lib/ > > ./configure --x-libraries=/emul/linux/x86/usr/lib ; make clean ; make > depend > ; make > > > > >

1 0

Will pay $$ for fix of Magic: The Gathering Online bug
by Kyle Winkler 15 Jun '06

15 Jun '06

Currently, Magic: the Gathering Online almost works, except that the fonts are either absent or illegible. I am willing the pay US $200 (via PayPal) for this bug to be fixed. If any of you are willing to give it a try, let me know. I will pay up front.

4 5

Re: secur32: Implement QueryContextAttributes. Add tests. cbMaxToken is 1904 not 2010.
by Kai Blin 15 Jun '06

15 Jun '06

* Kai Blin <kai.blin(a)gmail.com> [15/06/06, 20:50:47]: Please disregard this patch. I sent this out to wine-patches by accident. It still has some errors. > + case SECPKG_ATTR_LOCAL_CRED: > + TRACE("MSDN claims SECPKG_ATTR_LOCAL_CRED is obsolete\n"); > + /* fall through */ > + case SECPKG_ATTR_LOCAL_CERT_CONTEXT: > + FIXME("SECPKG_ATTR_LOCAL_CERT_CONTEXT.stub\n"); > + break; > + case SECPKG_ATTR_ROOT_STORE: > + FIXME("SECPKG_ATTR_ROOT_STORE stub\n"); > + break; Those only exist for SCHANNEL, not NTLM. > + case SECPKG_ATTR_SIZES: > + { > + PSecPkgContext_Sizes spcs = > + SECUR32_ALLOC(sizeof(SecPkgContext_Sizes)); > + spcs->cbMaxToken = NTLM_MAX_BUF; > + spcs->cbMaxSignature = 16; > + spcs->cbBlockSize = 1; > + spcs->cbSecurityTrailer = 16; > + pBuffer = spcs; > + return SEC_E_OK; > + } This doesn't seem to set pBuffer to spcs. What did I miss here? > + switch(sec_pkg_name) > + { > + case "NTLM": This doesn't work, of course. -- Kai Blin, (blin at gmx dot net) "Let us condemn to hellfire all those who disagree with us." -- militant religionists everywhere

2 1

Re: Re: [WineD3D] fix bug 4872
by fenix＠club-internet.fr 15 Jun '06

15 Jun '06

----Message d'origine---- >Date: Wed, 14 Jun 2006 22:03:58 -0600 >De: Vitaliy Margolen <wine-devel(a)kievinfo.com> >A: Raphael <fenix(a)club-internet.fr> >Copie à: wine-devel(a)winehq.org >Sujet: Re: [WineD3D] fix bug 4872 > >Wednesday, June 14, 2006, 6:51:21 PM, Raphael wrote: >> On Thursday 15 June 2006 02:34, Raphael wrote: >>> Hi, >>> >>> first patch since a long time :) >>> >>> Changelog: >>> - rename WineD3D_(Create|Release)FakeGLContext to WineD3D_(Create| >>> Release)FakeGLContextOnNeed >>> - and if a GL context is already active (and was not a fake context) >>> returns simili fake context with current GL context infos >>> - cleanup not more needed checks >>> >>> fix #4872: make IWineD3DImpl_CheckDeviceType working if a GL context is >>> already active >>> >>> Regards, >>> Raphael > >> Use this patch instead, i let some stupid (and buggy) debug code > >> Regards, >> Raphael > >I think I counted at least 5 indentation stiles in your patch. I understand >staying consistent with the code's style and using your own stile. But 5?! No i only see two problems, and made by use of xemacs (as i have broken my kate install). Anyway i plan to cleanup code in file in a next patch See http://marc.theaimsgroup.com/?l=wine-patches&m=115033272215006&w=2 >> } >You forgot LEAVE_GL(); NO, LEAVE_GL() is done on ReleaseFakeGLContextOnNeed. to permit between two calls to can make X/GLX calls safely >> + case WINED3DFMT_A4R4G4B4: >> + if (4 == rb && 4 == gb && 4 == bb && 4 == ab) return TRUE; >> + break; >Has nothing to do with context. ?? >> + case WINED3DFMT_A8: >> + if (8 == ab) return TRUE; >> + break; >Same here ?? >> if (Adapter >= IWineD3D_GetAdapterCount(iface)) { >> - TRACE("(%p) Failed: Atapter (%u) higher than supported adapters (%u) returning WINED3DERR_INVALIDCALL\n", This, Adapter, IWineD3D_GetAdapterCount(iface)); >> + TRACE_(d3d_caps)("[FAILED: Atapter (%u) higher than supported adapters (%u)]\n", Adapter, IWineD3D_GetAdapterCount(iface)); >> return WINED3DERR_INVALIDCALL; >> } >This one as well. And I haven't seen any traces so far that used square >brackets. Now we will! Have you ever seen D3D traces ? I only made same as IWineD3DImpl_CheckDeviceFormat (for coherency) see http://source.winehq.org/source/dlls/wined3d/directx.c#L1425 Anyway displaying [OK] or [FAILED] for checks results seems a good idea (and easy to find in logs) > >Vitaliy Margolen Regards, Raphael PS: before many complains, my ISP webmail is buggy, sorry :)

5 4