- Wine-Devel - list.winehq.org

Re: Fix Win95 advapi crypt test - dlls/advapi32/tests/crypt.c
by Alexandre Julliard March 22, 2005

March 22, 2005

Jakob Eriksson <jakov(a)vmlinux.org> writes: > --- dlls/advapi32/tests/crypt.c 15 Mar 2005 19:32:38 -0000 1.19 > +++ dlls/advapi32/tests/crypt.c 22 Mar 2005 15:31:55 -0000 > @@ -138,7 +138,10 @@ > hProv = 0; > SetLastError(0xdeadbeef); > result = pCryptAcquireContextA(&hProv, szKeySet, szRsaBaseProv, PROV_RSA_FULL, 0); > - ok(result && (GetLastError() == ERROR_ENVVAR_NOT_FOUND || GetLastError() == ERROR_SUCCESS || GetLastError() == ERROR_RING2_STACK_IN_USE || GetLastError() == NTE_FAIL), "%d/%ld\n", result, GetLastError()); > + ok(result && (GetLastError() == ERROR_ENVVAR_NOT_FOUND || GetLastError() == ERROR_SUCCESS || > + GetLastError() == ERROR_RING2_STACK_IN_USE || GetLastError() == NTE_FAIL || > + GetLastError() == ERROR_NOT_LOGGED_ON), > + "%d/%ld\n", result, GetLastError()); When there are more than 2 or 3 possible last error returns, we should simply get rid of the test. It's very unlikely that any app would rely on the value if it's different on every Windows install. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: KERNEL: PeekNamedPipe must check for zero-length buffer as well as for NULL buffer
by Jakob Eriksson March 22, 2005

March 22, 2005

Alex Villacis Lasso wrote: > > Changelog: > * PeekNamedPipe now checks both for a NULL buffer and a zero-length > buffer before trying to recv() from the pipe Could you write a regression test? regards, Jakob

1 0

Re: NTDLL: teach wine about Mingw32 (RESUBMIT)
by Steven Edwards March 22, 2005

March 22, 2005

Hi, --- "C. Scott Ananian" <cscott(a)cscott.net> wrote: > This patch seems to have gotten lost: it was sent last Thursday but hasn't > yet been applied or rejected (as far as I can tell). It's even got a > testimonial from a wine developer: Andreas Mohr says, "Nice patch." =) I don't understand why Mingw does not report the same value as the Windows compiler here. Do you know lcc or and other free compiler based application need this? If not maybe it should be fixed in mingw. Thanks Steven __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs

2 1

Re: registry TODO
by Filip Navara March 22, 2005

March 22, 2005

Boaz Harrosh wrote: > Thomas Weidenmueller wrote: > >>> How is ReactOS's current (future) system? >> >> See above > > I was Just asking what ReactOS is doing, not that they need change. > ReactOS is different in this respect, I agree. > Is ReactOS's format the same as Windows. or do you have your own > Binary format? We use the a Windows registry format, but since some field haven't been reversed yet it is not 100% compatible... ;-) The registry files are mapped as sections in memory along with some other control data stored in non-paged pool. - Filip

1 0

Re: registry TODO
by Jonathan Wilson March 22, 2005

March 22, 2005

> Is ReactOS's format the same as Windows. or do you have your own Binary The answer to "is the reactos registry format the same as windows" depends on which version of windows you are comparing it to. Windows 95/98/ME used system.dat and user.dat to store the registry. NT kernels (NT4, 2000, XP etc) use "registry hives" (IIRC ReactOS uses hives too, no idea if they are the same format as NT/2K/XP or not)

1 0

Re: [WINEALSA] direct sound fallback patch
by Robert Reif March 22, 2005

March 22, 2005

Robert Reif wrote: > Don't fail when opening a device in direct sound mode. > Rather, try alternate formats first before failing. > > With the wave api, requesting a format that the > hardware can't do should fail so the wave mapper can > fix it up. However, the direct sound api does not > expect a failure (within reason) if the format isn't > supported in hardware but expects the closest > format that the hardware does support. > > This gets the direct sound regression tests working > better with the ALSA "hw" device and my lame Intel > builtin sound card. Use this patch instead. Fixes a format bug. Index: dlls/winmm/winealsa/audio.c =================================================================== RCS file: /home/wine/wine/dlls/winmm/winealsa/audio.c,v retrieving revision 1.71 diff -u -p -r1.71 audio.c --- dlls/winmm/winealsa/audio.c 21 Mar 2005 12:32:48 -0000 1.71 +++ dlls/winmm/winealsa/audio.c 21 Mar 2005 18:02:05 -0000 @@ -315,6 +315,22 @@ static const char * getMessage(UINT msg) return unknown; } +static const char * getFormat(WORD wFormatTag) +{ + static char unknown[32]; +#define FMT_TO_STR(x) case x: return #x + switch(wFormatTag) { + FMT_TO_STR(WAVE_FORMAT_PCM); + FMT_TO_STR(WAVE_FORMAT_EXTENSIBLE); + FMT_TO_STR(WAVE_FORMAT_MULAW); + FMT_TO_STR(WAVE_FORMAT_ALAW); + FMT_TO_STR(WAVE_FORMAT_ADPCM); + } +#undef FMT_TO_STR + sprintf(unknown, "UNKNOWN(0x%04x)", wFormatTag); + return unknown; +} + static DWORD bytes_to_mmtime(LPMMTIME lpTime, DWORD position, WAVEFORMATPCMEX* format) { @@ -1758,6 +1774,12 @@ static DWORD wodOpen(WORD wDevID, LPWAVE memcpy(&wwo->waveDesc, lpDesc, sizeof(WAVEOPENDESC)); copy_format(lpDesc->lpFormat, &wwo->format); + TRACE("Requested this format: %ldx%dx%d %s\n", + wwo->format.Format.nSamplesPerSec, + wwo->format.Format.wBitsPerSample, + wwo->format.Format.nChannels, + getFormat(wwo->format.Format.wFormatTag)); + if (wwo->format.Format.wBitsPerSample == 0) { WARN("Resetting zeroed wBitsPerSample\n"); wwo->format.Format.wBitsPerSample = 8 * @@ -1789,7 +1811,19 @@ static DWORD wodOpen(WORD wDevID, LPWAVE else wwo->write = snd_pcm_mmap_writei; - EXIT_ON_ERROR( snd_pcm_hw_params_set_channels(pcm, hw_params, wwo->format.Format.nChannels), MMSYSERR_INVALPARAM, "unable to set required channels"); + if ((err = snd_pcm_hw_params_set_channels(pcm, hw_params, wwo->format.Format.nChannels)) < 0) { + WARN("unable to set required channels: %d\n", wwo->format.Format.nChannels); + if (dwFlags & WAVE_DIRECTSOUND) { + if (wwo->format.Format.nChannels > 2) + wwo->format.Format.nChannels = 2; + else if (wwo->format.Format.nChannels == 2) + wwo->format.Format.nChannels = 1; + else if (wwo->format.Format.nChannels == 1) + wwo->format.Format.nChannels = 2; + WARN("changed number of channels from %d to %d\n", lpDesc->lpFormat->nChannels, wwo->format.Format.nChannels); + } + EXIT_ON_ERROR( snd_pcm_hw_params_set_channels(pcm, hw_params, wwo->format.Format.nChannels ), MMSYSERR_INVALPARAM, "unable to set required channels" ); + } if ((wwo->format.Format.wFormatTag == WAVE_FORMAT_PCM) || ((wwo->format.Format.wFormatTag == WAVE_FORMAT_EXTENSIBLE) && @@ -1819,7 +1853,24 @@ static DWORD wodOpen(WORD wDevID, LPWAVE return WAVERR_BADFORMAT; } - EXIT_ON_ERROR( snd_pcm_hw_params_set_format(pcm, hw_params, format), MMSYSERR_INVALPARAM, "unable to set required format"); + if ((err = snd_pcm_hw_params_set_format(pcm, hw_params, format)) < 0) { + WARN("unable to set required format: %s\n", snd_pcm_format_name(format)); + if (dwFlags & WAVE_DIRECTSOUND) { + if ((wwo->format.Format.wFormatTag == WAVE_FORMAT_PCM) || + ((wwo->format.Format.wFormatTag == WAVE_FORMAT_EXTENSIBLE) && + IsEqualGUID(&wwo->format.SubFormat, &KSDATAFORMAT_SUBTYPE_PCM))) { + if (wwo->format.Format.wBitsPerSample != 16) { + wwo->format.Format.wBitsPerSample = 16; + format = SND_PCM_FORMAT_S16_LE; + } else { + wwo->format.Format.wBitsPerSample = 8; + format = SND_PCM_FORMAT_U8; + } + WARN("changed bits per sample from %d to %d\n", lpDesc->lpFormat->wBitsPerSample, wwo->format.Format.wBitsPerSample); + } + } + EXIT_ON_ERROR( snd_pcm_hw_params_set_format(pcm, hw_params, format), MMSYSERR_INVALPARAM, "unable to set required format" ); + } rate = wwo->format.Format.nSamplesPerSec; dir=0; @@ -1830,9 +1881,14 @@ static DWORD wodOpen(WORD wDevID, LPWAVE return WAVERR_BADFORMAT; } if (rate != wwo->format.Format.nSamplesPerSec) { - ERR("Rate doesn't match (requested %ld Hz, got %d Hz)\n", wwo->format.Format.nSamplesPerSec, rate); - snd_pcm_close(pcm); - return WAVERR_BADFORMAT; + if (dwFlags & WAVE_DIRECTSOUND) { + WARN("changed sample rate from %ld Hz to %d Hz\n", wwo->format.Format.nSamplesPerSec, rate); + wwo->format.Format.nSamplesPerSec = rate; + } else { + ERR("Rate doesn't match (requested %ld Hz, got %d Hz)\n", wwo->format.Format.nSamplesPerSec, rate); + snd_pcm_close(pcm); + return WAVERR_BADFORMAT; + } } dir=0; EXIT_ON_ERROR( snd_pcm_hw_params_set_buffer_time_near(pcm, hw_params, &buffer_time, &dir), MMSYSERR_INVALPARAM, "unable to set buffer time");

1 2

Re: msde on wine
by Dan Kegel March 22, 2005

March 22, 2005

Damiano Venturin wrote: > I'm interested in running msde on wine. I'm just looking around to see > if someone reached the goal. I saw you mail at > http://www.winehq.com/hypermail/wine-devel/2003/02/0583.html > > I just want to know if u was able to run it. I haven't tried since then. In the two years since then, MSI support on Wine has come a long ways, perhaps it will install now. Give it a try and report your experiences to the wine-devel mailing list, perhaps someone there can help you past the inevitable problems. - Dan -- Trying to get a job as a c++ developer? See http://kegel.com/academy/getting-hired.html

1 0

registry TODO
by James Hawkins March 22, 2005

March 22, 2005

Alexandre, Recently you removed the loading of the Windows registry on startup and stated that > Nobody's working on it, so it won't be supported until someone cares > enough to do it. I encouraged a few people to start working on it but > nobody did, so taking out the existing support is a way to provide > more encouragement. If that's not enough then the feature simply won't > be supported anymore. What needs to be done concerning this TODO? Looking through Bugzilla I found bug #422: "Currently Wine loads the complete registry into memory at startup. This is relatively ok when the registry is small such as in fake_windows configurations, but when importing the registry fro a Windows partition this can result in the wineserver allocating more than 30MB. This slows down startup times very significantly and wastes memory even if most of it is going to be swapped out. One of the reasons why Wine puts all the registry into memory on startup is that the current registry file format does not lend itself well to random accesses (especially writes). But there is no fundamental reason why Wine needs to read all the registry on startup. Hence this proposal to: * change the registry file format so that Wine can easily make random accesses. A 'database' format may be appropriate here, maybe based on the Berkeley database Library (libdb2/libdb3) or the GNU dbm library (libgbm1, but check license issues). * put in place a mechanism to update these files from the Windows registry files when appropriate (it seems impractical to access the Windows files directly) * modify the registry routines to retrieve only the data required by the application and cache it in memory" Is this a valid description of what needs to be done? -- James Hawkins

3 2

Regression in file tests
by Jakob Eriksson March 22, 2005

March 22, 2005

Between http://cvs.winehq.org/cvsweb/wine/dlls/kernel/tests/file.c#rev1.46 http://test.winehq.org/data/200502231000/ and http://test.winehq.org/data/200503041000/ http://cvs.winehq.org/cvsweb/wine/dlls/kernel/tests/file.c#rev1.48 we have more about 50 more failed test results. What happened? regards, Jakob

1 1

Re: Eliminate casts of the return value of HeapAlloc
by Francois Gouget March 22, 2005

March 22, 2005

On Mon, 21 Mar 2005, Alexandre Julliard wrote: [...] > Log message: > Mike McCormack <mike(a)codeweavers.com> > Eliminate casts of the return value of HeapAlloc. This could be turned into a pretty easy janitorial the starting point of which would be the output of this command: find . -name "*.[ch]" | xargs egrep "\) *Heap(Re)?Alloc" (344 lines) There are a couple of places where the output of HeapAlloc is being cast to a DWORD but except for those it seems pretty straightforward. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ La terre est une b\xEAta...

3 2