- Wine-Devel - list.winehq.org

winmm:wave testing
by Jakob Eriksson Nov. 4, 2004

Nov. 4, 2004

How should this be handled, is this a buggy driver? http://test.winehq.org/data/200411021000/2000_BennyGoemans/winmm:wave.txt wave.c:998:found 1 WaveOut devices wave.c:664: 0: "SB PCI" (\\?\pci#ven_1274&dev_1371&subsys_13711274&rev_08#3&61aaa01&0&68#{6994ad04-93ef-11d0-a3cc-00a0c9223196}\wave) 5.0 (1:100) wave.c:667: channels=65535 formats=00fff support=002c wave.c:669: WAVECAPS_VOLUME WAVECAPS_LRVOLUME WAVECAPS_SAMPLEACCURATE wave.c:524:Playing 1 second silence at 22050x 8x1 WAVE_FORMAT_PCM wave.c:420:waveOutGetPosition(0): TIME_MS not supported, returned TIME_SAMPLES wave.c:443:waveOutGetPosition(0): TIME_SMPTE not supported, returned TIME_SAMPLES wave.c:549: Test failed: The sound played for 1342 ms instead of 1000 ms

1 0

Wine dll testing state of affairs
by Jakob Eriksson Nov. 4, 2004

Nov. 4, 2004

I looked around in the Wine dlls tree to see how many dlls had a "tests" directory in them. Quite a few had, but these did not: Just thought it could be interesting to see. . ./advpack ./amstream ./atl ./avicap32 ./avifil32 ./cabinet ./capi2032 ./cards ./cfgmgr32 ./comcat ./commdlg ./crtdll ./crypt32 ./ctl3d ./d3d8 ./d3d9 ./d3dim ./d3drm ./d3dx8 ./d3dxof ./dbghelp ./dciman32 ./devenum ./dinput ./dinput8 ./dmband ./dmcompos ./dmime ./dmloader ./dmscript ./dmstyle ./dmsynth ./dmusic ./dmusic32 ./dplay ./dplayx ./dpnet ./dpnhpast ./dswave ./dxdiagn ./dxerr8 ./dxerr9 ./dxguid ./glu32 ./glut32 ./hhctrl.ocx ./iccvid ./icmp ./ifsmgr.vxd ./imagehlp ./imm32 ./itss ./lzexpand ./mmdevldr.vxd ./monodebg.vxd ./mpr ./msdmo ./mshtml ./msi ./msimg32 ./msisys ./msnet32 ./msrle32 ./msvcrt20 ./msvcrt40 ./msvidc32 ./msvideo ./mswsock ./newdev ./odbc32 ./oleacc ./olecli ./oledlg ./olepro32 ./olesvr ./opengl32 ./qcap ./rasapi32 ./richedit ./secur32 ./serialui ./setupapi ./shdocvw ./shfolder ./snmpapi ./sti ./strmiids ./tapi32 ./ttydrv ./twain ./unicows ./url ./uuid ./uxtheme ./vdhcp.vxd ./vdmdbg ./vmm.vxd ./vnbt.vxd ./vnetbios.vxd ./vtdapi.vxd ./vwin32.vxd ./win32s ./winaspi ./wined3d ./winedos ./wineps ./winnls ./wintab32 ./wintrust ./wow32 ./wsock32 ./x11drv These have tests: ./advapi32/tests ./comctl32/tests ./ddraw/tests ./dsound/tests ./gdi/tests ./iphlpapi/tests ./kernel/tests ./mapi32/tests ./mlang/tests ./msacm/tests ./mscms/tests ./msvcrt/tests ./msvcrtd/tests ./netapi32/tests ./ntdll/tests ./ole32/tests ./oleaut32/tests ./psapi/tests ./quartz/tests ./rpcrt4/tests ./rsabase/tests ./shell32/tests ./shlwapi/tests ./urlmon/tests ./user/tests ./version/tests ./wininet/tests ./winmm/tests ./winsock/tests ./winspool/tests

1 0

Update region patch
by Alexandre Julliard Nov. 4, 2004

Nov. 4, 2004

Here's the first step of moving the update region management to the server. The refresh on window moves is not optimized yet, so it will probably flicker quite a bit, but things should still be painted properly in the end. If you feel like giving this a try, I'm interested in hearing about any case where things don't repaint correctly. -- Alexandre Julliard julliard(a)winehq.org

3 3

Re: Some conformance tests for msvcrt printf functions
by Aneurin Price Nov. 4, 2004

Nov. 4, 2004

Is there any particular reason this patch hasn't been committed? Something wrong or just overlooked?

2 1

Access Tokens and Threads
by Elad Lahav Nov. 3, 2004

Nov. 3, 2004

I've been looking into the annoying "Set ThreadImpersonationToken handle" messages. In order to fix this issue, I need to know a few things: 1. Does Wine maintain access tokens at all? It seems that most of the relevant functions (e.g., NtQueryInformationThread and NtSetInformationThread) simply ignore it. 2. It looks like the thread's token is a part of the so-called "security descriptor" member of the thread's SECURITY_ATTRIBUTES structure. Does anyone know what structure this descriptor points to? Elad

2 1

source level debugging with WineLib?
by Geoff K. Hart Nov. 3, 2004

Nov. 3, 2004

Hello, We just started using WineLib, and have very basic question: which -g??? option do you use to do source level debugging with winedbg? I searched the documentation and mailing list for any step-by-step about this, but found nothing. So either the answer is so obvious that I just can't see it - or it doesn't work? By default, winemaker seems configure things for no debug - so I edited one Makefile manually, and added -g. Winedbg can attach to the process, and can even show the stack - but didn't know much beyond that. I tried all the -g variants that I could find, but only -gstabs and -gstabs+ seemed to help. They allowed the debugger to find the correct place in the source, and "whatis" seemed to work - however, if I tried to dereference something, it complained about "no type ...", like this: Wine-dbg>info process pid threads parent executable (all id:s are in hex) 00000008 3 00000000 'Z:\home\s\src\SysSrv \DataDistributionService\DDSServer2\ddsserver2.exe' Wine-dbg>attach 8 In 32 bit mode. 0x546387a2: ret Wine-dbg>bt Backtrace: =>1 0x546387a2 (0x10252c50) 2 0x48714a7e NTDLL_wait_for_multiple_objects+0x12e in ntdll (0x10252cf8) 3 0x48712fec vm86_return_end+0xc55 in ntdll (0x10252d1c) 4 0x1000b808 (0x10352d64) 5 0x48714a7e NTDLL_wait_for_multiple_objects+0x12e in ntdll (0x10352e0c) 6 0x48714afc NtWaitForMultipleObjects+0x6c in ntdll (0x10352e30) 7 0x61e57a95 WaitForMultipleObjectsEx+0xb5 in kernel32 (0x10352f60) 8 0x61e5797b WaitForSingleObjectEx+0x3b in kernel32 (0x10352f80) 9 0x1002878afixme:dbghelp:elf_new_wine_thunks Duplicate in ddsserver2<elf>: _ZZ17BroadCastIdentityPvmmE22lRepCntrClientDeletion<10037720-00000004> lRepCntrClientDeletion<10037720-00000000> fixme:dbghelp:elf_new_wine_thunks Duplicate in ddsserver2<elf>: _ZZ17BroadCastIdentityPvmmE20lRepKeepAliveCounter<10037724-00000004> lRepKeepAliveCounter<10037724-00000000> fixme:dbghelp:elf_new_wine_thunks Duplicate in ddsserver2<elf>: _ZZ17BroadCastIdentityPvmmE17lRepCntrBroadcast<1003771c-00000004> lRepCntrBroadcast<1003771c-00000000> _Z13DDSServerMainPv+0x88(lpParameter=0x44c47098) [/home/s/src/SysSrv/ DataDistributionService/DDSServer2/AcceptSocket.cpp:220] in ddsserver2 (0x10353150) 10 0x61e5eeda (0x10353228) 11 0x48716523 (0x10353a78) 12 0x1000508d (0x10353b0c) 13 0x660d255a (0x00000000) Wine-dbg>up 8 220 dwRetVal = ::WaitForSingleObjectEx (pServer- >m_hStopEvent, INFINITE, TRUE); Wine-dbg>p pServer 0x44c47098 Wine-dbg>p *pServer No type or type mismatch Wine-dbg>whatis pServer struct DDSServer* Wine-dbg> Any help? Geoff

5 4

Re: InternetGetConnectedStateExA
by Alexandre Julliard Nov. 3, 2004

Nov. 3, 2004

Aric Stewart <aric(a)codeweavers.com> writes: > +BOOL WINAPI InternetGetConnectedStateExA(LPDWORD lpdwStatus, > + LPSTR lpszConnectionName, DWORD dwNameLen, DWORD dwReserved) > +{ > + INT len; > + LPWSTR lpwszConnectionName = NULL; > + BOOL rc; > + > + TRACE("(%p, %s, %ld, 0x%08lx)\n", lpdwStatus, debugstr_a(lpszConnectionName), dwNameLen, dwReserved); > + > + if (lpszConnectionName) > + { > + len = MultiByteToWideChar(CP_ACP, 0, lpszConnectionName, -1, NULL, 0); > + lpwszConnectionName= HeapAlloc(GetProcessHeap(), 0, len*sizeof(WCHAR)); > + MultiByteToWideChar(CP_ACP, 0, lpszConnectionName, -1, > + lpwszConnectionName, len); > + } > + > + rc = InternetGetConnectedStateExW(lpdwStatus,lpwszConnectionName, dwNameLen, lpszConnectionName is an output parameter, it should be converted on return from the W call. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: WINSOCK: Implement WSAAddressToStringA/W
by Alexandre Julliard Nov. 3, 2004

Nov. 3, 2004

Hans Leidekker <hans(a)it.vu.nl> writes: > + if (stringW) > + { > + ret = WSAAddressToStringW( sockaddr, len, NULL, stringW, &lenstrW ); > + *lenstr = lenstrW / sizeof(WCHAR); To be really correct you have to use WideCharToMultiByte to compute the ASCII length. Though in this case it would probably be easier to just duplicate the code and do a simple sprintf in the A function instead of trying to call the W function. -- Alexandre Julliard julliard(a)winehq.org

1 0

compiling wine with msvc
by James Hawkins Nov. 3, 2004

Nov. 3, 2004

I'm trying to get the wine tests to compile under msvc, so I ran msvcmaker in the wine source under cygwin (which was successful). Then I opened the winetest project and did a 'rebuild all'. When it tries to compile advapi32_test/security.c I get an warning and 102 errors: C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\rpc.h(89) : warning C4005: 'RPCRTAPI' : macro redefinition ..\..\..\include\rpc.h(45) : see previous definition of 'RPCRTAPI' :\Program Files\Microsoft Visual Studio\VC98\INCLUDE\rpcnsi.h(37) : error C2143: syntax error : missing ')' before 'type' C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\rpcnsi.h(37) : error C2491: 'RpcNsBindingExportA' : definition of dllimport function not allowed ......etc I'm pretty sure the problem is that the the msvc's rpc.h and rpcnsi.h files are being included when they shouldn't. Is that about right? If so, what could be a possible solution? BTW I'm compiling with MSVC++ 6.0. wine cvs from 11-01-2004. -- James Hawkins

3 6

Re: Fix for metafile test
by Dmitry Timoshkov Nov. 3, 2004

Nov. 3, 2004

"Jakob Eriksson" <jakov(a)vmlinux.org> wrote: > Seems the test fails in non-interactive mode without this patch. No, the problem seems to be related to a window's visibility. Windows does not play EMF records if the DC was created for a not visible window. A lot of other tests fail due to exactly that problem: user32/win.c and user32/msg.c are the examples. The question is what makes the tests create invisible windows? Does winetest run as a service on a not visible desktop? Anything else? -- Dmitry.

4 5