- Wine-Devel - list.winehq.org

Turning off the exception dialog
by Robert Lunnon June 8, 2005

June 8, 2005

The latest build of wine I am getting a dialog telling me about an exception rather than wine just dumping core. gdb can't catch the point of the exception allowing me to debug the problem. From memory there is a way to turn off this "feature" no ? Bob

1 0

[Resent]Request: Detect a broken pipe
by Uwe Bonnes June 8, 2005

June 8, 2005

Hallo, Alexandre just added my testcase to wine-patches, fingerpointing at a todo_wine() with wine not seeing a broken pipe. This still keeps Altera Quartus from starting up. I tried to fix for myself for some time , but to no avail. Any help/patches welcome. Thanks -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

1 0

Problems in bitblt (in x11drv)
by David Albrecht June 7, 2005

June 7, 2005

I'm new to wine and relatively new to windows programming being mostly a unix hack. Just for yucks I thought I'd get a relatively simple app emu48 to work. Turns out emu48 is not so simple (multi-threaded, etc.). First task towards functionality is to get the boot up display to look right and a BitBlt didn't seem to be functioning correctly. Much investigation later it turns out that in the file dib.c: /* if the source bitmap is 8bpp or less, we're supposed to use the * DC's palette for color conversion (not the DIB color table) */ is NOT true at least not for my case where the DC containing the bitmap has one palette and the colormap for the bitmap is not the same. If I disable the hPalette tests as I did below and always use the DIB colormap it boots up with the correct display otherwise the color mapping doesn't match the DIB bits and I get garbage. if (dib.dsBm.bmBitsPixel <= 8) { HPALETTE hPalette = GetCurrentObject( physDevSrc->hdc, OBJ_PAL ); if (/*!hPalette || (hPalette == GetStockObject(DEFAULT_PALETTE))*/1) { /* HACK: no palette has been set in the source DC, * use the DIB colormap instead - this is necessary in some * cases since we need to do depth conversion in some places * where real Windows can just copy data straight over */ colorMap = physBitmap->colorMap; nColorMap = physBitmap->nColorMap; } else { colorMap = X11DRV_DIB_BuildColorMap( physDevSrc, (WORD)-1, dib.dsBm.bmBitsPixel, (BITMAPINFO*)&dib.dsBmih, &nColorMap ); if (colorMap) aColorMap = TRUE; } } Not knowing the full ramifications of this code I thought I'd pass this along so that someone more familiar with it can reassess it. The DC and bitmap are created thusly: #if defined GRAYSCALE #define NOCOLORS 8 #else #define NOCOLORS 2 #endif static struct { BITMAPINFOHEADER Lcd_bmih; RGBQUAD bmiColors[NOCOLORS]; } bmiLcd = { {0x28,0/*x*/,0/*y*/,1,8,BI_RGB,0,0,0,NOCOLORS,0} }; VOID CreateLcdBitmap(VOID) { // create LCD bitmap _ASSERT(nLcdDoubled == 1 || nLcdDoubled == 2 || nLcdDoubled == 4); bmiLcd.Lcd_bmih.biWidth = LCD_ROW * nLcdDoubled; bmiLcd.Lcd_bmih.biHeight = -64 * nLcdDoubled; _ASSERT(hLcdDC == NULL); hLcdDC = CreateCompatibleDC(hWindowDC); _ASSERT(hLcdDC != NULL); hLcdBitmap = CreateDIBSection(hLcdDC, (BITMAPINFO*)&bmiLcd,DIB_RGB_COLORS, (LPVOID*)&pbyLcd, NULL, 0); _ASSERT(hLcdBitmap != NULL); hOldLcdBitmap = SelectObject(hLcdDC, hLcdBitmap); _ASSERT(hPalette != NULL); SelectPalette(hLcdDC, hPalette, FALSE); // set palette for LCD DC RealizePalette(hLcdDC); // realize palette BuildPattern(); // build Nibble -> DIB mask pattern UpdateContrast(Chipset.contrast); return; } Note that update contrast doe a SetDIBColorTable: VOID UpdateContrast(BYTE byContrast) { #if defined GRAYSCALE INT i; RGBQUAD c = *(RGBQUAD*)&dwKMLColor[byContrast]; // pixel on color RGBQUAD b = *(RGBQUAD*)&dwKMLColor[byContrast+32]; // pixel off color if ((Chipset.IORam[BITOFFSET] & DON) == 0 || I == *(DWORD*)&b) b = *(RGBQUAD*)&dwKMLColor[0]; // fill color palette of bitmap for (i = 0; i < ARRAYSIZEOF(bmiLcd.bmiColors); ++i) { const INT nCAdj[] = { 0, 1, 1, 2, 1, 2, 2, 3 }; _ASSERT(i < ARRAYSIZEOF(bmiLcd.bmiColors)); _ASSERT(i < ARRAYSIZEOF(nCAdj)); bmiLcd.bmiColors[i] = b; bmiLcd.bmiColors[i].rgbRed += (((INT) c.rgbRed - (INT) b.rgbRed) * nCAdj[i] / 3); bmiLcd.bmiColors[i].rgbGreen += (((INT) c.rgbGreen - (INT) b.rgbGreen) * nCAdj[i] / 3); bmiLcd.bmiColors[i].rgbBlue += (((INT) c.rgbBlue - (INT) b.rgbBlue) * nCAdj[i] / 3); } #else bmiLcd.bmiColors[1] = *(RGBQUAD*)&dwKMLColor[byContrast]; // pixel on color bmiLcd.bmiColors[0] = *(RGBQUAD*)&dwKMLColor[byContrast+32]; // pixel off color if ((Chipset.IORam[BITOFFSET] & DON) == 0 || I == *(DWORD*)&bmiLcd.bmiColors[0]) bmiLcd.bmiColors[0] = *(RGBQUAD*)&dwKMLColor[0]; #endif // update palette information _ASSERT(hLcdDC); SetDIBColorTable(hLcdDC,0,ARRAYSIZEOF(bmiLcd.bmiColors),bmiLcd.bmiColors); return; } The BitBlt: BitBlt(hWindowDC, nLcdX, nLcdY, 131*nLcdDoubled, nLines*nLcdDoubled, hLcdDC, Chipset.boffset*nLcdDoubled, 0, SRCCOPY); As emu48 still doesn't work I'll probably have more things I find in likely much different areas later. Dave

1 0

CPU Emulation
by Robert Lunnon June 7, 2005

June 7, 2005

Has any consideration been given to hooking a CPU emulation into wine to allow non x86 derivative CPUs execute native Windows binaries. For example it aught to be possible to hook qmu into the wine loader. Would this make a good Google Summer of Code project ? Wine could become Wine Is Now an Emulator Bob

4 3

Re: Fix our waveout tests by adjusting for underrun conditions
by Duane Clark June 7, 2005

June 7, 2005

Jeremy White wrote: > On my VIA8237 sound card at work, the Wave tests for winmm > have been failing for me with Alsa. > > > Now, this patch Works For Me (TM), on both of my systems, > but turning off the xrun mode seems like a big step to me; > I'd appreciate testing by any concerned parties. This seems to cause severe problems for VirtualDub: http://virtualdub.sourceforge.net/ Here is the movie I am using: http://ftp.pub.cri74.org/pub/win9x/video/SampleMovie/ If you open and run the movie, it plays ok, including sound. But try clicking stop in the middle of playing. Or when the movie gets to the end, everything is stuck; it appears the program cannot tell that the movie has ended. This is on current CVS, with winealsa (I think, at least that is what is set in my config file), and RH9.

1 0

Re: wine-systray problem
by Andreas Mohr June 7, 2005

June 7, 2005

[Darn, forgot group reply] Hi, On Tue, Jun 07, 2005 at 10:15:02PM +0600, Pashok wrote: > Andreas Mohr wrote: > > > >Maybe openbox-3.2 doesn't (fully?) support systray desktop guidelines > >(see freedesktop.org?) and thus Wine falls back to using a popup window > >instead? > > Maybe, but Openbox "was written first to comply with standards and to > work properly".... Hmm, then it should have worked if that fully was the case. > So i can't change this? I must use xfce??? I like openbox.... :( You are using Linux. You can change *EVERYTHING*. > -- > Sorry, im still using windows... Oh wait, you aren't... ;) If systray integration doesn't work despite openbox advertising standards compliance, then use some relevant Wine debug channel (it's probably dlls/shell32/systray.c, so it's WINEDEBUG=+shell) to find out what it tries to do and what doesn't work there. Andreas Mohr

1 0

List settings
by Robert Lunnon June 7, 2005

June 7, 2005

It appears the mailing list isn't quite right. When I reply-all I seem to get the list twice, once as recipient and once as CC, can this be fixed ? This seems to be affecting more than just me. Bob

5 6

Re: ViaVoice (98) in Wine
by René Rebe June 7, 2005

June 7, 2005

Hi, On Tuesday 07 June 2005 02:37, you wrote: > >No problem - a log with my normal setup, and one without the module for > > the 10 channel card loaded attached (I wanted to see for myself if that > > makes a significant difference in the behaviour of Wine). > > > >Yours, > > OSS thinks your sound card doesn't support volume control on the input. > Not much you can do about that. I assume this is OSS on ALSA? Then it's > an ALSA OSS emulation issue (assuming the card really does support it). > The peak meter issue is not a problem because most drivers don't support > it. > > Does the software really refuse to work because of no record volume > control? That shouldn't be a fatal error. Maybe - since ViaVoice is quite agressive to tweak the microphone gain ... Have you seen the invalid handle in the log? trace:msacm:acmDriverDetailsA ((nil), 0x504840, 00000000) trace:msacm:acmDriverDetailsW ((nil), 0x77a7eebc, 00000000) trace:msacm:acmDriverOpen (0x77a7e3e4, (nil), 00000000) warn:msacm:acmDriverOpen invalid handle trace:msacm:acmDriverOpen (0x504c94, (nil), 00000000) warn:msacm:acmDriverOpen invalid handle Maybe via voice fails due to that ... at least it matches what I found in the speech engines log (oh - I mistyped the wine list address of the mail containing this information, yesterday - resent it now): AudMME: Set Audio Input --------------------------------- [Jun 06 12:29:1 2005] Using record device # 0 which is a Analog Devices AD1980 AUDMME will use device # 0 for recording AUDMME will use device # 0 for playback PCM Device WAV Audio Adapter Default Record Device WAV Audio Default Playback Device WAV Audio Adapter acmDriverDetails failed, rc = 5 acmDriverOpen failed, rc = 5 acmDriverDetails failed, rc = 5 acmDriverOpen failed, rc = 5 Alphabet/Language/Codepage: R Gr_GR 850 a returncode of 5 seems to be: #define MMSYSERR_INVALHANDLE (MMSYSERR_BASE + 5) Yours, -- René Rebe - Rubensstr. 64 - 12157 Berlin (Europe / Germany) http://www.exactcode.de | http://www.t2-project.org +49 (0)30 255 897 45

2 1

ViaVoice (98) in Wine
by René Rebe June 7, 2005

June 7, 2005

Hi all, over the weekend I tried to get ViaVoice (98) running in Wine (20050524). The installation goes smooth - even the embedded videos how to handle the microsfone, etc. are displayed ,-) However after the installation some "audio setup" tool is run to adjust the mixer level and maybe more like capture channel. This tools does just print an error that it can not find "mixer elements / devices" (translation from the german text I had on screen). I tried a few install and wine config variations, including switching from OSS to ALSA - however without success. I then tried to copy my trained profile from a real Windows box in order to skip this audio setup stuff. However the real applications then also complain about no mixer found. Capturing from my mic with some "proof of concept" and tiny audio recorder as shipped with ViaVoice (snwave.exe IIRC) did work. Injecting all the windows partition content into the wine folder did also not help. Are there some holes in some kind of mixer interface implementation? Is there some kind of tracing I can enable in order to take a look what kind of call did fail? Any help appreciated. Yours, -- René Rebe - Rubensstr. 64 - 12157 Berlin (Europe / Germany) http://www.exactcode.de | http://www.t2-project.org +49 (0)30 255 897 45

4 8

SetTextJustification() fix
by Keith Dunwoody June 6, 2005

June 6, 2005

Hi, A little while ago I sent a fix for bug #50 (PrgWin95: Text justification needs beefing up), in which SetTextJustification() wasn't spacing text lines out properly. I was wondering why my patch was rejected. Thanks, -- Keith __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 3