- Wine-Devel - list.winehq.org

Re: add proper support for native NT drivers and programs to the build system
by Alexandre Julliard Sept. 7, 2005

Sept. 7, 2005

Ivan Leo Puoti <ivanleo(a)gmail.com> writes: > @@ -469,8 +469,10 @@ > /* get the default entry point for a given spec file */ > static const char *get_default_entry_point( const DLLSPEC *spec ) > { > + if (spec->subsystem == IMAGE_SUBSYSTEM_NATIVE && spec->characteristics & IMAGE_FILE_DLL) > + return "__wine_spec_drv_entry"; > + if (spec->subsystem == IMAGE_SUBSYSTEM_NATIVE) return "__wine_spec_exe_nentry"; Native drivers are not supposed to be dlls. > @@ -322,9 +322,15 @@ > ORDDEF *odp = &spec->entry_points[i]; > if (odp->type != TYPE_STUB) continue; > fprintf( outfile, "#ifdef __GNUC__\n" ); > - fprintf( outfile, "extern void __wine_spec_unimplemented_stub( const char *module, const char *func ) __attribute__((noreturn));\n" ); > + if(spec->subsystem == IMAGE_SUBSYSTEM_NATIVE) > + fprintf( outfile, "extern void __wine_spec_nunimplemented_stub( const char *module, const char *func ) __attribute__((noreturn));\n" ); > + else > + fprintf( outfile, "extern void __wine_spec_unimplemented_stub( const char *module, const char *func ) __attribute__((noreturn));\n" ); > fprintf( outfile, "#else\n" ); > - fprintf( outfile, "extern void __wine_spec_unimplemented_stub( const char *module, const char *func );\n" ); > + if(spec->subsystem == IMAGE_SUBSYSTEM_NATIVE) > + fprintf( outfile, "extern void __wine_spec_nunimplemented_stub( const char *module, const char *func ) __attribute__((noreturn));\n" ); > + else > + fprintf( outfile, "extern void __wine_spec_unimplemented_stub( const char *module, const char *func );\n" ); You should provide your own handling of stubs inside ntoskrnl, like ntdll does. -- Alexandre Julliard julliard(a)winehq.org

2 2

Object manager case sensitivity issue
by Vitaliy Margolen Sept. 7, 2005

Sept. 7, 2005

I want to make sure that this problem is well understood. Also to make it clear that Wine differs from windows in this really important aspect. For more details, you can refer to the following thread: http://www.winehq.org/hypermail/wine-devel/2005/04/0426.html It needs to be fixed at least on the server level (I have sent a patch for one of the possible solutions). Then we can get ntdll to follow the suite. For a test see attached program. You should be able to run it on windows with administrative privileges. As you can see, this is something very dangerous yet perfectly legal, at least on win2k sp4. -- Best regards, Vitaliy

1 0

Strange problem with writing files from wine
by Pavel Troller Sept. 7, 2005

Sept. 7, 2005

Hi! I have problems with writing files from wine (today's CVS). Typical example is word95, which always writes something like: Cannot access folder "F:\filename.doc ": unaccessible or protected by password (in my local language, so it may be inaccurate). There is a small problem with this strange indentation (that the second double quote is on a next line, but it is not fatal. More fatal is that this dialog even occurs and the file cannot be written. I've created a tracefile with +file and I think I've found the problem. At the moment of attempt to write (in this example, F:\brumbal.doc), there appears: trace:file:CreateFileW L"\\\\.\\F:" GENERIC_READ FILE_SHARE_READ FILE_SHARE_WRITE creation 3 attributes 0x0 trace:file:RtlDosPathNameToNtPathName_U (L"\\\\.\\F:",0x7fc8975c,(nil),(nil)) trace:file:RtlGetFullPathName_U (L"\\\\.\\F:" 520 0x7fc89508 (nil)) trace:file:get_dos_device L"F:" -> "/dev/hdb1" warn:file:CreateFileW Unable to create file L"\\\\.\\F:" (status c0000022) trace:file:CreateFileW returning 0xffffffff trace:file:RtlDosPathNameToNtPathName_U (L"F:\\",0x7fc89744,(nil),(nil)) trace:file:RtlGetFullPathName_U (L"F:\\" 520 0x7fc89508 (nil)) trace:file:wine_nt_to_unix_file_name L"\\??\\F:\\" -> "/home/patrol/.wine/dosdevices/f:/" trace:file:GetFileAttributesW L"F:\\brumbal.doc\r\n" trace:file:RtlDosPathNameToNtPathName_U (L"F:\\brumbal.doc\r\n",0x7fc8aae4,(nil),(nil)) trace:file:RtlGetFullPathName_U (L"F:\\brumbal.doc\r\n" 520 0x7fc8a8b0 (nil)) trace:file:CreateFileW L"\\\\.\\F:" GENERIC_READ FILE_SHARE_READ FILE_SHARE_WRITE creation 3 attributes 0x0 trace:file:RtlDosPathNameToNtPathName_U (L"\\\\.\\F:",0x7fc8975c,(nil),(nil)) trace:file:RtlGetFullPathName_U (L"\\\\.\\F:" 520 0x7fc89508 (nil)) trace:file:get_dos_device L"F:" -> "/dev/hdb1" warn:file:CreateFileW Unable to create file L"\\\\.\\F:" (status c0000022) trace:file:CreateFileW returning 0xffffffff trace:file:RtlDosPathNameToNtPathName_U (L"F:\\",0x7fc89744,(nil),(nil)) trace:file:RtlGetFullPathName_U (L"F:\\" 520 0x7fc89508 (nil)) trace:file:wine_nt_to_unix_file_name L"\\??\\F:\\" -> "/home/patrol/.wine/dosdevices/f:/" patrol(a)arcus:/opt/samba-dir/Program Files/Microsoft Office/Office$ Why the hell wine is trying to access /dev/hdb1 ? Yes, my home directory is on it, but I think that such a form of access is not good :-). In my .wine/dosdevices, there is clearly a line lrwxrwxrwx 1 patrol users 12 Sep 7 06:50 f: -> /home/patrol Or am I misinterpreting the log and the failure is caused by something else ? Otherwise wine is working near to perfect now. With regards, Pavel Troller

1 0

Path for Testfiles (MSI: Test creating a package)
by Detlef Riekenberg Sept. 7, 2005

Sept. 7, 2005

Hi After reading reading this commit, .... > ChangeSet ID: 20009 > Added files: > dlls/msi/tests : package.c > Test creating a package. > Patch: http://cvs.winehq.org/patch.py?id=20009 ... i Suggest that: - all directories and files for tests should be created in "%TEMP%\winetest" - we modify "wine/tests.h" to get this path and check for path and name overflows (MAX_PATH and nesting level). "c:" could be readonly or absent/hidden Any comments please. -- By By ... ... Detlef

2 1

DirectX status update....
by Oliver Stieber Sept. 6, 2005

Sept. 6, 2005

I though it was about time I gave a status update on DirectX9/wined3d, so here goes. Short of a couple of bug fixes (one to shaders, and another that causes some games to crash) and a couple of patches that need resending (support for non-power2 textures in warhammer 40k) wined3d and d3d9 is functionally as complete as the version on http://directxwine.sourceforge.net/, that means that as soon as pixel shaders are coppied across from d3d8 it's more or less the end of adding features to wined3d and the start of fixing the remaining bugs and getting the performance up to spec. Anyhow, as soon as I've sent in those patches I'll provide another status update with a bit more information about moving d3d8 over to wined3d and the improvements that have been made. Oliver. ___________________________________________________________ Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com

3 2

Re: URLMON: Added file protocol implementation
by Stefan Leichter Sept. 6, 2005

Sept. 6, 2005

Am Samstag, 3. September 2005 21:22 schrieb Jacek Caban: > Changelog: > Added file protocol implementation Hello, this patch breaks the cross compiling of the tests with mingw. Which is the correct location of the class id? Bye Stefan make[3]: Entering directory `/usr/src/wine/wine-mingw/dlls/urlmon/tests' i386-mingw32-gcc generated.cross.o misc.cross.o protocol.cross.o url.cross.o testlist.cross.o -o urlmon_crosstest.exe -lurlmon -luser32 -lkernel32 -lole32 -luuid /usr/local/lib/gcc/i386-mingw32/3.4.2/../../../../i386-mingw32/lib/ libuuid.a(uuid.o):/usr/src/packages/BUILD/mingw-3.7/w32api-3.2/lib/uuid.c: (.rdata+0x1340): multiple definition of `_CLSID_FileProtocol' protocol.cross.o:/usr/src/wine/wine-mingw/dlls/urlmon/tests/../../../../wine/ dlls/urlmon/tests/protocol.c:63: first defined here collect2: ld returned 1 exit status make[3]: *** [urlmon_crosstest.exe] Fehler 1 make[3]: Leaving directory `/usr/src/wine/wine-mingw/dlls/urlmon/tests'

1 0

Re: Wined3d: added A2R10G10B10 and D3DFMT_D24FS8
by Oliver Stieber Sept. 6, 2005

Sept. 6, 2005

--- Karsten Elfenbein <kelfe(a)gmx.de> wrote: Can you put together a patch to add A2R10G10B10 and D3DFMT_D24FS8 to utils.c as well please. D3DFmt2GLIntFmt needs to return GL_RGBA, D3DFmt2GLFmt needs to return GL_RGBA, D3DFmt2GLType needs to return GL_UNSIGNED_INT_2_10_10_10_REV, D3DFmtGetBpp needs to return 4, D3DFmtMakeGlCfg needs to PUSH2(GLX_ALPHA_SIZE, 2); PUSH2(GLX_RED_SIZE, 10); PUSH2(GLX_GREEN_SIZE, 10); PUSH2(GLX_BLUE_SIZE, 10); Also in surface.c IWineD3DSurfaceImpl_UnlockRect needs to use glDrawPixels(This->lockedRect.right - This->lockedRect.left, (This->lockedRect.bottom - This->lockedRect.top) - 1, GL_BGRA, GL_UNSIGNED_INT_2_10_10_10_REV, This->resource.allocatedMemory); when unlocking the back buffer Thanks, Oliver. > Changelog: > * added A2R10G10B10 and D3DFMT_D24FS8 modes > > > ___________________________________________________________ Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com

2 2

Re: fonts: don't call LineTo with the gdi lock
by Alexandre Julliard Sept. 6, 2005

Sept. 6, 2005

Huw D M Davies <h.davies1(a)physics.ox.ac.uk> writes: > On Sun, Aug 28, 2005 at 12:04:41PM +0100, Huw D M Davies wrote: >> Huw Davies <huw(a)codeweavers.com> >> Don't call LineTo while holding the gdi lock > > Is this one better? I think it would be better to release the DC earlier on and then just do normal GDI calls. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: MSI: test creating a package
by Andreas Mohr Sept. 6, 2005

Sept. 6, 2005

Hi, On Tue, Sep 06, 2005 at 03:21:55PM +0900, Mike McCormack wrote: > ChangeLog: > <aric(a)codeweavers.com> > <mike(a)codeweavers.com> > * test creating a package + res = MsiCloseHandle( suminfo); + ok( res == ERROR_SUCCESS , "Failed to close suminfo" ); + + + res = MsiCloseHandle( hPackage); + ok( res == ERROR_SUCCESS , "Failed to close package" ); +} Hmm, something seems to be missing here... Andreas

2 2

Messages from 2030 (Laxdragon please read)
by Ivan Leo Puoti Sept. 6, 2005

Sept. 6, 2005

http://www.winehq.com/pipermail/wine-devel/2030-March/000355.html Well, now we all know wineserver will still be around in 2030... Seriously, this should be fixed (not wineserver, the list archive) Ivan.

1 0