April 2010 - Wine-devel - list.winehq.org

Re: About Wine Test Bot
by Greg Geldorp 07 Apr '10

07 Apr '10

Hi Alexandre, > Ok so I could not run any test until compiling on cygwin works. > But the idea was to be able to remotly compile wine with some of our patches > applied. > > I was told "you can't do that, it will break minGW". > The WINEBUILD VM exists but we can't use it (or tell me how to do that) The WINEBUILD VM is only used to cross-compile tests in case you submit a patch and not a test executable to the bot. Running tests on Wine is outside the scope of WineTestBot (perhaps it should be called WinTestBot...), it is aimed at making sure that the tests themselves are correct (by checking that they don't fail on Windows). Ge van Geldorp.

1 0

About Wine Test Bot
by Alexandre Goujon 07 Apr '10

07 Apr '10

Hi, I have a suggestion about wineTestBot. It's maybe the wrong place but I'm quite sure interested people read wine-devel. I think it would be great if we could not only run tests (as it already does) but also check if a patch does not break wine (on minGW, on cygwin or on MAC for instance). Thus, we could choose a kind of WINEBUILD VM to tell wineTestBot to only compile wine with our patch. Currently, I can't compile on Linux and cygwin is very slow. So it would make my life easier if I could rely on wineTestBot... Comments are welcome Thanks

3 5

winmm: Improve MCI A/W mapping slightly.
by Joerg-Cyril.Hoehle＠t-systems.com 07 Apr '10

07 Apr '10

Hi, I forgot to mention that this patch was rebased against 1.1.42, supersedes the former "Fix A/W mapping for 64bit Wine" patch and is not dependent on prior 64bit patches. Currently, I can't think of a fully general solution to the A/W mapping problem. In theory, an unknown driver with new unknown commands and keywords could be registered. We have little idea how large the structure is and which flags controls the validity of the nth slot, e.g. when to call strdupAtoW(). One could think about parsing the command table to obtain this information. However, a) I see no guarantee that there's a unique MCI_STRING value for a given bit that is set in the flags parameter. b) This does not help with baroque commands like SYSINFO which sometimes put a string, sometimes a DWORD into the result buffer or which are documented to count buffer sizes in bytes, unlike other commands. This latter example reveals that calling MCI_Cleanup() before unwrapping is unfortunate, as valuable type information is lost. It is currently encapsulated inside mciSendCommandW. Open questions: - Does Wine actually support hooking in a new native MCI driver? (beside DLL overrides) - Why does Wine copy dwCallback only when MCI_NOTIFY is set? It is not wrong, but why care? To prevent a Valgrind message when that slot was not filled by the caller? BTW, my (unpublished) testing reveals that unlike what MSDN says, dwRetSize is counted in characters in MCI_SYSINFO_PARMSW, not bytes (at least on w2k with SYSINFO_NAME [OPEN]). But that shall be the subject of another patch. Regards, Jörg Höhle

1 0

Re: Participation in GSoC
by Roderick Colenbrander 07 Apr '10

07 Apr '10

Hi Alexander, Good to see you back in GSoc. I can't comment on IE / control panel as I don't know much about it but I will comment to another small part. > As for my own project ideas, I think it would have been interesting to > implement IPX over TCP in winsock, since most of the games that use IPX > require root access for it to work. This feature could perhaps be enabled in > winecfg or the Control Panel. I don't have any experience with networkking, > though, so I'm concerned that this might be a bit amibitous. It should be possible to implement IPX over TCP but when you do this, the other side also needs to wrap IPX in a TCP packet. (there are various user space tools I believe which can do this). IPX can be used without root permissions if you use posix file capabilities to enable raw sockets on wineserver. I experimented with this a long time ago and I wasn't able to get it working at that time but this should somehow work. Roderick

1 0

Re: [PATCH 2/5] [Server]: correctly set the DR0..7 registers on Linux for x86_64 architecture
by Alexandre Julliard 07 Apr '10

07 Apr '10

Eric Pouech <eric.pouech(a)orange.fr> writes: > @@ -549,12 +549,21 @@ void get_thread_context( struct thread *thread, context_t *context, unsigned int > goto done; > } > } > +#ifdef __i386__ > context->debug.i386_regs.dr0 = data[0]; > context->debug.i386_regs.dr1 = data[1]; > context->debug.i386_regs.dr2 = data[2]; > context->debug.i386_regs.dr3 = data[3]; > context->debug.i386_regs.dr6 = data[6]; > context->debug.i386_regs.dr7 = data[7]; > +#else /* __x86_64__ */ > + context->debug.x86_64_regs.dr0 = data[0]; > + context->debug.x86_64_regs.dr1 = data[1]; > + context->debug.x86_64_regs.dr2 = data[2]; > + context->debug.x86_64_regs.dr3 = data[3]; > + context->debug.x86_64_regs.dr6 = data[6]; > + context->debug.x86_64_regs.dr7 = data[7]; > +#endif You can't use ifdefs, this needs to depend on the CPU type of the client process. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: ddraw/tests: Added todo_wine test to verify resizing of fullscreen windows
by Henri Verbeet 07 Apr '10

07 Apr '10

On 6 April 2010 23:14, Johan Gill <johan.gill(a)gmail.com> wrote: > +»······»·······if (!EqualRect(&scrn, &orig_rect)) { > +»······»······· /* Check that the client rect was resized */ > +»······»······· rc = GetClientRect(hwnd, &test); > +»······»······· ok(rc!=0, "GetClientRect returned %x\n", rc); > +»······»······· rc = EqualRect(&scrn, &test); > +»······»······· todo_wine ok(rc!=0, "Fullscreen window has wrong size\n"); > +»······»·······} Please don't mix tabs and spaces, just use 4 space indent.

1 0

Re: [PATCH] user32: honor bfOffBits in BITMAP_Load
by Dmitry Timoshkov 07 Apr '10

07 Apr '10

Wolfram Sang <wolfram(a)the-dreams.de> wrote: > + offbits = bmfh->bfOffBits - sizeof(BITMAPFILEHEADER); > + bits = (char *)info + (offbits ?: size); The purpose of offbits is not clear. Besides "?:" is absolutely not readable, putting an explicit 0 there would help a bit. Adding a test case wouldn't hurt either. -- Dmitry.

3 3

Re: winmm: Rewrite time.c using ntdll timers
by Steven Edwards 06 Apr '10

06 Apr '10

On Tue, Apr 6, 2010 at 6:08 PM, Maarten Lankhorst <m.b.lankhorst(a)gmail.com> wrote: > --- > Should fix bug 22252 as a side effect. It should also make it possible to > run our winmm on windows for wave playback because the timers are now using > win32 apis. :) Thanks, I've been waiting for something like this since 0383e4e4990e414a7dcaba51e6b81bf60499944a ;) -- 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

1 0

Wine FIXME Report March 2010
by Michael Stefaniuc 06 Apr '10

06 Apr '10

Wine FIXME Report March 2010 ============================ The data for this report comprise the emails from March 2010 archives of the mailing lists wine-{bugs,devel,users} as well as the bugzilla attachments referenced by the wine-bugs emails. Although WARN messages were collected too they aren't handled in this report. File parsed: 6867 (emails: 6282, bugzilla attachments: 585) Files (logs) with FIXME/ERR/WARN: 517 Type | Total | Unique ------+---------+------- fixme | 191778 | 22303 err | 101542 | 3632 warn | 6125 | 1199 Most Frequent Messages By Logs ------------------------------ Messages that are present in most Wine output logs collected. FIXME: Logs | Lines | Mean | Channel:Function Text ------+-------+--------+---------------------- 68 | 200 | 2.94 | d3d:WineD3D_ChoosePixelFormat Add OpenGL context recreation support to SetDepthStencilSurface 62 | 82 | 1.32 | system:SetProcessDPIAware stub! 51 | 92 | 1.80 | iphlpapi:GetAdaptersAddresses no support for IPv6 addresses 37 | 62 | 1.68 | urlmon:URLMoniker_BindToObject use running object table 35 | 36 | 1.03 | sync:CreateMemoryResourceNotification (0) stub 33 | 57 | 1.73 | shdocvw:ClOleCommandTarget_QueryStatus command_0: 27, 0x0 31 | 43 | 1.39 | resource:GetGuiResources (0xffffffff,0): stub 31 | 72 | 2.32 | shdocvw:PropertyNotifySink_OnChanged unimplemented dispid 1005 28 | 66 | 2.36 | d3d9:Direct3DShaderValidatorCreate9 stub 24 | 85 | 3.54 | shdocvw:BindStatusCallback_OnProgress status code 11 ERR: Logs | Lines | Mean | Channel:Function Text ------+-------+--------+---------------------- 31 | 31 | 1.00 | ole:CoGetContextToken apartment not initialised 24 | 35 | 1.46 | ole:CoGetClassObject class {9a5ea990-3034-4d6f-9128-01f3c61022bc} not registered 24 | 35 | 1.46 | ole:CoGetClassObject no class object {9a5ea990-3034-4d6f-9128-01f3c61022bc} could be created for context 0x1 22 | 44 | 2.00 | ole:CoInitializeEx Attempt to change threading model of this apartment from multi-threaded to apartment threaded 21 | 55 | 2.62 | alsa:ALSA_CheckSetVolume Could not find 'PCM Playback Volume' element 20 | 220 | 11.00 | ole:ClientIdentity_QueryMultipleInterfaces IRemUnknown_RemQueryInterface failed with error 0x80004002 15 | 15 | 1.00 | ole:CoInitializeEx Attempt to change threading model of this apartment from apartment threaded to multi-threaded 14 | 49 | 3.50 | ole:CoUninitialize Mismatched CoUninitialize 13 | 13 | 1.00 | ole:apartment_getclassobject DllGetClassObject returned error 0x80040111 10 | 15 | 1.50 | eventlog:ReportEventW L"0" Most Frequent Messages By Mean Lines Per Logs --------------------------------------------- This are the messages which show up in at least 1% of the logs and have the highest mean lines per logs ratio. FIXME: Logs | Lines | Mean | Channel:Function Text ------+-------+--------+---------------------- 15 | 2307 | 153.80 | advapi:SetSecurityInfo stub 11 | 1655 | 150.45 | d3d:debug_d3dformat Unrecognized 909200449 (as fourcc: AL16) WINED3DFORMAT! 11 | 1655 | 150.45 | d3d:debug_d3dformat Unrecognized 909201952 (as fourcc: R16) WINED3DFORMAT! 11 | 1655 | 150.45 | d3d:getFormatDescEntry Can't find format unrecognized(909200449) in the format lookup table 11 | 1655 | 150.45 | d3d:getFormatDescEntry Can't find format unrecognized(909201952) in the format lookup table 14 | 1706 | 121.86 | gdiplus:GdipCreateHalftonePalette 7 | 715 | 102.14 | d3d:buffer_PreLoad Too many full buffer conversions, stopping converting 13 | 565 | 43.46 | gdiplus:GdipGetFamilyName No support for handling of multiple languages! 21 | 423 | 20.14 | wininet:InternetLockRequestFile STUB 7 | 110 | 15.71 | shell:URL_ParseUrl failed to parse L"cdbxpp.resources" ERR: Logs | Lines | Mean | Channel:Function Text ------+-------+--------+---------------------- 7 | 7295 |1042.14 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R32G32B32A32_FLOAT 7 | 7295 |1042.14 | d3d:getColorBits Unsupported format: WINED3DFMT_R32G32B32A32_FLOAT 6 | 347 | 57.83 | d3d_surface:surface_set_texture_name Surface has SFLAG_INTEXTURE set, but no texture name 7 | 211 | 30.14 | ddraw:PixelFormat_WineD3DtoDD Can't translate this Pixelformat 58 7 | 211 | 30.14 | ddraw:PixelFormat_WineD3DtoDD Can't translate this Pixelformat 64 20 | 220 | 11.00 | ole:ClientIdentity_QueryMultipleInterfaces IRemUnknown_RemQueryInterface failed with error 0x80004002 8 | 66 | 8.25 | ole:CoCreateInstance apartment not initialised 7 | 48 | 6.86 | richedit:ReadStyleSheet skipping optional destination 7 | 28 | 4.00 | imm:ImmAssociateContextEx Unknown dwFlags 0x0 14 | 49 | 3.50 | ole:CoUninitialize Mismatched CoUninitialize Functions With Most Messages By Mean Lines Per Logs --------------------------------------------------- This are the functions which show up in at least 1% of the logs and have the highest mean lines per logs ratio. FIXME: Logs | Lines | Mean | Channel:Function ------+-------+--------+----------------- 7 | 6082 | 868.86 | d3d_draw:drawStridedFast 80 | 19784 | 247.30 | d3d_surface:IWineD3DVolumeImpl_LockBox 15 | 2307 | 153.80 | advapi:SetSecurityInfo 117 | 16272 | 139.08 | d3d:context_check_fbo_status 14 | 1706 | 121.86 | gdiplus:GdipCreateHalftonePalette 6 | 464 | 77.33 | d3d:IWineD3DDeviceImpl_ClearSurface 6 | 352 | 58.67 | shell:IDropTargetHelper_fnDragOver 57 | 2695 | 47.28 | dsound:IDirectSoundBufferImpl_AcquireResources 14 | 566 | 40.43 | gdiplus:GdipGetFamilyName 22 | 847 | 38.50 | d3d:buffer_PreLoad ERR: Logs | Lines | Mean | Channel:Function ------+-------+--------+----------------- 29 | 46357 |1598.52 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt 29 | 46357 |1598.52 | d3d:getColorBits 7 | 349 | 49.86 | d3d_surface:surface_set_texture_name 14 | 422 | 30.14 | ddraw:PixelFormat_WineD3DtoDD 21 | 222 | 10.57 | ole:ClientIdentity_QueryMultipleInterfaces 8 | 66 | 8.25 | ole:CoCreateInstance 10 | 71 | 7.10 | richedit:ReadStyleSheet 16 | 89 | 5.56 | ole:ITypeInfo_fnInvoke 47 | 245 | 5.21 | ole:marshal_object 7 | 28 | 4.00 | imm:ImmAssociateContextEx Messages With The Highest Repetition Count ------------------------------------------ The messages with the highest repetition count in a single log file. FIXME: Logs | Lines | Mean | Channel:Function Text ------+-------+--------+---------------------- 1 | 76325 |76325.0 | d3d7:IDirect3DDeviceImpl_3_SetLightState D3DLIGHTSTATE_MATERIAL called with NULL material !!! 1 | 17648 |17648.0 | d3d_surface:IWineD3DVolumeImpl_LockBox (0x129fc038) : pBox=0xbff040 stub 1 | 4256 | 4256.0 | d3d_texture:basetexture_generate_mipmaps iface 0x4b577a0 stub! 1 | 2977 | 2977.0 | d3d:debug_d3dusagequery Unrecognized usage query flag(s) 0x1 1 | 2717 | 2717.0 | d3d:context_check_fbo_status Color attachment 0: (0x38eebdc0) WINED3DFMT_B8G8R8A8_UNORM 1280x1024 1 | 2717 | 2717.0 | d3d:context_check_fbo_status Depth attachment: (0x1a6668) WINED3DFMT_D24_UNORM_S8_UINT 1280x1024 1 | 2717 | 2717.0 | d3d:context_check_fbo_status FBO status GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT (0x8cd6) 1 | 2552 | 2552.0 | d3d_draw:drawStridedFast >>>>>>>>>>>>>>>>> GL_INVALID_FRAMEBUFFER_OPERATION (0x506) from glDrawElements @ drawprim.c / 46 1 | 2129 | 2129.0 | d3d_surface:IWineD3DBaseSurfaceImpl_Blt Filters not supported in software blit 4 | 2074 | 518.5 | d3d:context_check_fbo_status FBO status GL_FRAMEBUFFER_UNSUPPORTED (0x8cdd) ERR: Logs | Lines | Mean | Channel:Function Text ------+-------+--------+---------------------- 1 | 2990 | 2990.0 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R16G16B16A16_UNORM 1 | 2990 | 2990.0 | d3d:getColorBits Unsupported format: WINED3DFMT_R16G16B16A16_UNORM 4 | 2950 | 737.5 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R32G32_FLOAT 4 | 2950 | 737.5 | d3d:getColorBits Unsupported format: WINED3DFMT_R32G32_FLOAT 4 | 2880 | 720.0 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R16G16B16A16_FLOAT 4 | 2880 | 720.0 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R16G16_FLOAT 4 | 2880 | 720.0 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R16_FLOAT 7 | 2880 | 411.4 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R32G32B32A32_FLOAT 4 | 2880 | 720.0 | d3d:IWineD3DImpl_IsPixelFormatCompatibleWithRenderFmt Unable to check compatibility for Format=WINED3DFMT_R32_FLOAT 4 | 2880 | 720.0 | d3d:getColorBits Unsupported format: WINED3DFMT_R16G16B16A16_FLOAT Further Info: ============= The sqlite3 database and the perl script to output the above report can be downloaded from http://people.redhat.com/mstefani/winefixme/reports/201003/ bye michael

1 0

re: user32/dialog.c: GetNextDlgTabItem returns hwndDlg if next item not found
by Dan Kegel 06 Apr '10

06 Apr '10

"Mikhail Maroukhine" <mikolg at connectivegames.com> wrote: > user32/dialog.c: GetNextDlgTabItem returns hwndDlg if next item not found > Behavior checked on Windows XP Welcome to Wine development! Poker playing penguins salute you. As Dmitry wrote, adding a test case would be a good idea; see http://wiki.winehq.org/WritingConformanceTests - Dan

2 1