August 2007 - Wine-Devel - list.winehq.org

dinput compile error
by Robert Reif Aug. 20, 2007

Aug. 20, 2007

make[2]: Entering directory `/home/wine/wine/dlls/dinput' gcc -c -I. -I. -I../../include -I../../include -D__WINESRC__ -D_REENTRANT -fPIC -Wall -pipe -fno-strict-aliasing -Wwrite-strings -Wpointer-arith -g -O2 -o joystick_linuxinput.o joystick_linuxinput.c joystick_linuxinput.c:153: field `axes' has incomplete type joystick_linuxinput.c:154: confused by earlier errors, bailing out make[2]: *** [joystick_linuxinput.o] Error 1 make[2]: Leaving directory `/home/wine/wine/dlls/dinput' It seems struct input_absinfo is not defined.

3 3

Re: [4/5] D3D9: Some caps fixes
by H. Verbeet Aug. 20, 2007

Aug. 20, 2007

On 20/08/07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > /* Some functionality is implemented in d3d9.dll, not wined3d.dll. Add the needed caps */ > pCaps->DevCaps2 |= D3DDEVCAPS2_CAN_STRETCHRECT_FROM_TEXTURES; You should move that to filter_caps() as well, or at the very least add it to IDirect3DDevice9Impl_GetDeviceCaps().

2 1

Re: winefile: Save font settings for registry
by Michael Stefaniuc Aug. 20, 2007

Aug. 20, 2007

Ivan Sinitsin wrote: > this is corrected patch for winefile. You want to use the explicit RegQueryValueExW and RegSetValueExW functions. bye michael > > This patch adds an opportunity to keep adjustments of a font between sessions, > if menu "Save settings on exit" is checked. > > > Changelog: > this patch add opportunity to save font option to the regestry, and load font > option from regestry in the winefile program > > > > ------------------------------------------------------------------------ > > Index: programs/winefile/winefile.c > =================================================================== > RCS file: /home/wine/wine/programs/winefile/winefile.c,v > retrieving revision 1.75 > diff -u -p -u -r1.75 winefile.c > --- programs/winefile/winefile.c 25 Oct 2006 18:31:13 -0000 1.75 > +++ programs/winefile/winefile.c 14 Jun 2007 07:22:19 -0000 > @@ -66,6 +66,20 @@ static const WCHAR reg_start_x[] = { 's' > static const WCHAR reg_start_y[] = { 's','t','a','r','t','Y','\0'}; > static const WCHAR reg_width[] = { 'w','i','d','t','h','\0'}; > static const WCHAR reg_height[] = { 'h','e','i','g','h','t','\0'}; > +static const WCHAR reg_font_height[] = { 'f','o','n','t','_','h','e','i','g','h','t','\0'}; > +static const WCHAR reg_font_width[] = { 'f','o','n','t','_','w','i','d','t','h','\0'}; > +static const WCHAR reg_font_escapement[] = { 'f','o','n','t','_','E','s','c','a','p','e','m','e','n','t','\0'}; > +static const WCHAR reg_font_orientation[] = { 'f','o','n','t','_','O','r','i','e','n','t','a','t','i','o','n','\0'}; > +static const WCHAR reg_font_weight[] = { 'f','o','n','t','_','W','e','i','g','h','t','\0'}; > +static const WCHAR reg_font_italic[] = { 'f','o','n','t','_','I','t','a','l','i','c','\0'}; > +static const WCHAR reg_font_underline[] = { 'f','o','n','t','_','U','n','d','e','r','l','i','n','e','\0'}; > +static const WCHAR reg_font_strikeout[] = { 'f','o','n','t','_','S','t','r','i','k','e','O','u','t','\0'}; > +static const WCHAR reg_font_charset[] = { 'f','o','n','t','_','C','h','a','r','S','e','t','\0'}; > +static const WCHAR reg_font_outprecision[] = { 'f','o','n','t','_','O','u','t','P','r','e','c','i','s','i','o','n','\0'}; > +static const WCHAR reg_font_clipprecision[] = { 'f','o','n','t','_','C','l','i','p','P','r','e','c','i','s','i','o','n','\0'}; > +static const WCHAR reg_font_quality[] = { 'f','o','n','t','_','Q','u','a','l','i','t','y','\0'}; > +static const WCHAR reg_font_pitchandfamily[] = { 'f','o','n','t','_','P','i','t','c','h','A','n','d','F','a','m','i','l','y','\0'}; > +static const WCHAR reg_font_facename[] = { 'f','o','n','t','_','F','a','c','e','N','a','m','e','\0'}; > > enum ENTRY_TYPE { > ET_WINDOWS, > @@ -1588,15 +1602,20 @@ static void get_path(Entry* dir, PTSTR p > > static windowOptions load_registry_settings(void) > { > - DWORD size; > + DWORD size, sizeW; > DWORD type; > HKEY hKey; > windowOptions opts; > + LOGFONT logfont; > + HFONT hFont; > > RegOpenKeyEx( HKEY_CURRENT_USER, registry_key, > 0, KEY_QUERY_VALUE, &hKey ); > > + hFont=GetStockObject(DEFAULT_GUI_FONT); > + GetObject(hFont,sizeof(logfont),&logfont); > size = sizeof(DWORD); > + sizeW = sizeof(logfont.lfFaceName); > > if( RegQueryValueEx( hKey, reg_start_x, NULL, &type, > (LPBYTE) &opts.start_x, &size ) != ERROR_SUCCESS ) > @@ -1614,8 +1633,51 @@ static windowOptions load_registry_setti > (LPBYTE) &opts.height, &size ) != ERROR_SUCCESS ) > opts.height = CW_USEDEFAULT; > > + RegQueryValueEx( hKey, reg_font_height, NULL, &type, > + (LPBYTE) &logfont.lfHeight, &size ); > + > + RegQueryValueEx( hKey, reg_font_width, NULL, &type, > + (LPBYTE) &logfont.lfWidth, &size ); > + > + RegQueryValueEx( hKey, reg_font_escapement, NULL, &type, > + (LPBYTE) &logfont.lfEscapement, &size ); > + > + RegQueryValueEx( hKey, reg_font_orientation, NULL, &type, > + (LPBYTE) &logfont.lfOrientation, &size ); > + > + RegQueryValueEx( hKey, reg_font_weight, NULL, &type, > + (LPBYTE) &logfont.lfWeight, &size ); > + > + RegQueryValueEx( hKey, reg_font_italic, NULL, &type, > + (LPBYTE) &logfont.lfItalic, &size ); > + > + RegQueryValueEx( hKey, reg_font_underline, NULL, &type, > + (LPBYTE) &logfont.lfUnderline, &size ); > + > + RegQueryValueEx( hKey, reg_font_strikeout, NULL, &type, > + (LPBYTE) &logfont.lfStrikeOut, &size ); > + > + RegQueryValueEx( hKey, reg_font_charset, NULL, &type, > + (LPBYTE) &logfont.lfCharSet, &size ); > + > + RegQueryValueEx( hKey, reg_font_outprecision, NULL, &type, > + (LPBYTE) &logfont.lfOutPrecision, &size ); > + > + RegQueryValueEx( hKey, reg_font_clipprecision, NULL, &type, > + (LPBYTE) &logfont.lfClipPrecision, &size ); > + > + RegQueryValueEx( hKey, reg_font_quality, NULL, &type, > + (LPBYTE) &logfont.lfQuality, &size ); > + > + RegQueryValueEx( hKey, reg_font_pitchandfamily, NULL, &type, > + (LPBYTE) &logfont.lfPitchAndFamily, &size ); > + > + RegQueryValueEx( hKey, reg_font_facename, NULL, &type, > + (LPBYTE) &logfont.lfFaceName, &sizeW); > + > RegCloseKey( hKey ); > > + Globals.hfont = CreateFontIndirect(&logfont); > return opts; > } > > @@ -1624,11 +1686,13 @@ static void save_registry_settings(void) > WINDOWINFO wi; > HKEY hKey; > INT width, height; > + LOGFONT logfont; > > wi.cbSize = sizeof( WINDOWINFO ); > GetWindowInfo(Globals.hMainWnd, &wi); > width = wi.rcWindow.right - wi.rcWindow.left; > height = wi.rcWindow.bottom - wi.rcWindow.top; > + GetObject(Globals.hfont, sizeof(logfont), &logfont); > > if ( RegOpenKeyEx( HKEY_CURRENT_USER, registry_key, > 0, KEY_SET_VALUE, &hKey ) != ERROR_SUCCESS ) > @@ -1651,6 +1715,34 @@ static void save_registry_settings(void) > (LPBYTE) &width, sizeof(DWORD) ); > RegSetValueEx( hKey, reg_height, 0, REG_DWORD, > (LPBYTE) &height, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_height, 0, REG_DWORD, > + (LPBYTE) &logfont.lfHeight, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_width, 0, REG_DWORD, > + (LPBYTE) &logfont.lfWidth, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_escapement, 0, REG_DWORD, > + (LPBYTE) &logfont.lfEscapement, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_orientation, 0, REG_DWORD, > + (LPBYTE) &logfont.lfOrientation, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_weight, 0, REG_DWORD, > + (LPBYTE) &logfont.lfWeight, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_italic, 0, REG_DWORD, > + (LPBYTE) &logfont.lfItalic, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_underline, 0, REG_DWORD, > + (LPBYTE) &logfont.lfUnderline, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_strikeout, 0, REG_DWORD, > + (LPBYTE) &logfont.lfStrikeOut, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_charset, 0, REG_DWORD, > + (LPBYTE) &logfont.lfCharSet, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_outprecision, 0, REG_DWORD, > + (LPBYTE) &logfont.lfOutPrecision, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_clipprecision, 0, REG_DWORD, > + (LPBYTE) &logfont.lfClipPrecision, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_quality, 0, REG_DWORD, > + (LPBYTE) &logfont.lfQuality, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_pitchandfamily, 0, REG_DWORD, > + (LPBYTE) &logfont.lfPitchAndFamily, sizeof(DWORD) ); > + RegSetValueEx( hKey, reg_font_facename, 0, REG_SZ, > + (LPBYTE) &logfont.lfFaceName,(strlenW(&logfont.lfFaceName) + 1)*sizeof(WCHAR)); > > /* TODO: Save more settings here (List vs. Detailed View, etc.) */ > RegCloseKey( hKey ); > > > ------------------------------------------------------------------------ > > -- Michael Stefaniuc Tel.: +49-711-96437-199 Sr. Network Engineer Fax.: +49-711-96437-111

1 0

Re: comctl resource diff
by Maarten Lankhorst Aug. 20, 2007

Aug. 20, 2007

Dj Apal [GR] schreef: > Hope this is the one you want. PLease commit it! > > TIA! > ------------------------------------------------------------------------ > > --- ./dlls/comctl32/rsrc.rc 2007-08-20 19:52:26.000000000 +0300 > +++ ./dlls/comctl32/rsrc1.rc 2007-08-20 19:50:03.000000000 +0300 > @@ -1196,6 +1196,7 @@ IDI_TT_ERROR_SM ICON LOADONCALL DISCARDA > #include "comctl_Cn.rc" > #include "comctl_Cs.rc" > #include "comctl_De.rc" > +#include "comctl_El.rc" > #include "comctl_En.rc" > #include "comctl_Eo.rc" > #include "comctl_Es.rc" > I'm missing a comctl_El.rc here?

1 0

Fixing wrong prototypes in a header
by Juan Lang Aug. 20, 2007

Aug. 20, 2007

Some of the function prototypes in wintrust.h have the wrong calling convention. The trouble is, PSDK gets them wrong too. So calling the functions, as declared in the PSDK header, results in a crash in Windows. Casting them to function pointers with the correct calling convention succeeds. Should I leave the prototypes incorrect in wintrust.h to match the PSDK, or should I fix them to make them more convenient for me to use? Thanks, --Juan

2 2

Re: winmm: Change default driver order to be alsa,coreaudio,oss - try2
by Vitaliy Margolen Aug. 20, 2007

Aug. 20, 2007

Maarten Lankhorst wrote: > Forgot to remove the strlen check. > Isn't this a bit premature? ALSA still does not support proper sound capture, working mixer, and not stable enough on some configurations. Vitaliy

2 8

Re: setupapi: if needed create directories for fake dlls
by Alexandre Julliard Aug. 20, 2007

Aug. 20, 2007

"Mikolaj Zalewski" <mikolajz(a)google.com> writes: > @@ -302,9 +302,28 @@ BOOL create_fake_dll( const WCHAR *name, > else > { > h = CreateFileW( name, GENERIC_WRITE, 0, NULL, CREATE_NEW, 0, NULL ); > + > + if (h == INVALID_HANDLE_VALUE && GetLastError() == ERROR_PATH_NOT_FOUND) > + { > + WCHAR *pathel; > + > + /* create the directory/directories */ > + pathel = strchrW(name, '\\'); > + while (pathel != NULL) > + { > + *pathel = 0; You shouldn't write to the name, it's a const pointer, even if strchrW hides the warning. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] handle builtin regedit.exe.so
by Alexandre Julliard Aug. 20, 2007

Aug. 20, 2007

Marcus Meissner <marcus(a)jet.franken.de> writes: > On my system it tries to load /usr/lib/wine/regedit.exe.so > (via symlink) and this fails. We stopped using symlinks a long time ago, it sounds like your .wine dir is out of date. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: comdlg32: printdlg: Fix number display in PrintDlgW()
by Alexandre Julliard Aug. 20, 2007

Aug. 20, 2007

Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> writes: > vsprintfW and friends do not support float numbers, so we have to format an > initial string with printf() and then print it to a wide string with > vsprintfW()'s %hs format. For a simply conversion to Unicode MultiByteToWideChar is a better choice. -- Alexandre Julliard julliard(a)winehq.org

1 0

Dyne:bolic binary module
by Robert Shoemaker Aug. 20, 2007

Aug. 20, 2007

I was wondering what the steps would be to put my binary build for the i686 Dyne:bolic distro on the releases page, it was no small task getting it to compile.

2 1

Direct3D 10 design considerations
by Stefan Dösinger Aug. 20, 2007

Aug. 20, 2007

Hi, In the past days there was a bit of discussion about the implementation design of our future d3d10 implementation. The major question was wether we should make a clean start over with opengl 3.0, or extend wined3d for d3d10. A few facts: *) D3D10 is a clean restart without any backwards compatiblity, so is gl3. The fixed function pipeline was removed in both. *) The D3D10 interface is a present. Many of the nasty things in d3d9 have gone. SRGB textures are a pixel format property now. Sampler states are a texture property, not a device setting. Offscreen-rendering is more fbo like. *) WineD3D contains a huge set of functionality of d3d10 already. D3D10 is an evolutionary step, not a revolution. I estimate that 80% of the functionality that is of our interest already existed in d3d9. Code reuse, especially of the shader code and texture management, is desireable. *) GL3 isn't out yet. When will we have a gl3 implementation on all platforms and drivers? *) If we split, we do not have to take care for old opengl implementations and missing extensions. For d3d10 we could just require gl3, period. Of course that will only work satisfiably for the user if there are gl3 implementations everywhere. So we have two ways: 1) Do the same Microsoft did, and do a fresh start. Some code, like the shader code, could be copied over and forked. This is propably the fastest way, but it leaves us with 2 different d3d implementation. The d3d10 implementation on the other hand would be free of old junk. It would make it hard to run d3d10 on gl2+extensions, and even more harder to implement d3d9- on gl3. 2) Extend wined3d. We could use the inheritance scheme like in old ddraw, where the non-rendering functionality was implemented in a base object, and the renderer specific things(DGA / User / 3D) was implemented in objects derived from the base objects. This was a nice thing, but it made the code next to unreadable IMHO because of a huge number of objects and unintuitive interactions(DGA code calling User stuff, common code calling callbacks that are set in 5 places in the code, 5 different surface types). WineD3D already has the attachment points for such an abstraction. We have COM vtables and state management callbacks. Implementing d3d10 properly in wined3d is most likely more work than a restart is, but it would allow us to reuse some code. What we would have to do is this: *) Make sampler states work like in opengl. D3D7-9 takes care for updating the per texture states based on the device states. This will require some code to be moved from wined3d to d3d7-9 and duplicated in the 3 libs, most notable sampler state tracking in stateblocks. *) Move the srgb flag -> format matching from wined3d to d3d9. Before this can be done, wined3d has to be changed from D3DFORMAT to DXGI_FORMAT style. *) Clearly separate gl-specific code from non-gl-specific code in surface, texture, cubetexture, volumetexture, indexbuffer and vertexbuffer. Then split them in a base object and a derived gl2 object. Add a gl3 object implementing the gl-specific stuff in gl3. Make sure there is no gl specific buffer/texture stuff in the device(like CreateVBO & *) Add a 2nd state table and callback functions for gl3. GL3 will have different state linkings. Most of the states can stay unimplemented unless we want to run d3d9 with gl3. D3D10 is much simpler here because the fixed function pipeline was removed. *) Do something about the context manager. Multithreading has changed in d3d10, and it may change in gl3. Offscreen rendering in d3d10 can require fbos(no point in d3d10 offscreen rendering without them), this has much less hassles than the back buffer or pbuffer orm. Context management will most likely be much simpler in d3d10+gl3. Andras tells me that he has the d3d10 stubs(no rendering yet) done now, and he is writing some tests currently, then he will send stubs and tests in. Which way do you think should we go?

5 9

Re: dsound: Handle failing of IDsDriver_CreateSoundBuffer better
by Vitaliy Margolen Aug. 20, 2007

Aug. 20, 2007

Maarten Lankhorst wrote: > There, now hush. :p > > Does 'Hardware Acceleration to Emulation' silently, without requiring of > the user to do something. > > > - ERR("Could not map sound device for direct access (%s)\n", strerror(errno)); > - ERR("Please run winecfg, open \"Audio\" page and set\n" > - "\"Hardware Acceleration\" to \"Emulation\".\n"); > + WARN("Could not map sound device for direct access (%s)\n", strerror(errno)); > return DSERR_GENERIC; And this really fixes the bug, doesn't just hide the error message? If it fixed, then there should be no messages at all. I really do not want to have people opening bugs because something crashes/sound doesn't work, etc. Vitaliy

2 1

Re: Greek resource file for comctl final
by Paul Vriens Aug. 19, 2007

Aug. 19, 2007

Dj Apal [GR] wrote: > OK changed the copyright and added the diff file for the rsrc.rc file so as > to be included. > > Thnx for noticing that! > > I think that we are OK now. Sorry for all this mess but it was my first time > posting a patch! > > > > ------------------------------------------------------------------------ > > Your 'greek_comctl' patch is in 'diff -up' format. The 'rsrc.diff' patch should also be in that format (and done from the root of your Wine source directory). Cheers, Paul.

1 0

choppy anim
by Skrzynka Aug. 18, 2007

Aug. 18, 2007

Hi, I'm trying to get some game (Settlers 3) to work on Wine (updated sources recently from git). If possible, I'd like to ask for some clues for a start: 1) the issue is choppy animation - the frame rate is abt 1/sec while the cpu load is low. However there are random periods when there seem to be about few frames per second (and cpu is laoded 100%). I tried measuring the time between call and return of DLL functions in relay.c. The only thing that looked suspiciously was WaitForSingle|MultipleObjects having times like 200-500ms. I also tried adding some delay to WaitForVerticalBlank in ddraw.c but also to no effect. Where should I start looking for the problem - some events or callbacks code? which? 2) when I run it from within winedbg i get lots of exceptions (no matter if app was started from winedbg or it was attached later) and no matter what combination of pass/continue I do, the app crashed with some msg box about exception or structured exception. Any pointers to docs on this? Any suggestions will help greatly. Thanks & regards Piotr -- ----- AlphaNet - najtaniej w sieci! -------- Odnowienia domen w rewelacyjnych cenach! .pl - 65 zl, .com.pl - 50 zl, reg - 20 zl http://www.domeny.alpha.pl --------------------------------------------

1 0

Re: #winehq admin abuse
by Abel Chiaro Aug. 18, 2007

Aug. 18, 2007

Tom Wickline <twickline(a)gmail.com> wrote: > I gladly take the side of vitamin on this issue. I believe he is a > valuable member of the FOSS community, he has contributed many patches > to this and other OSS projects. And in HIS free time he volunteers on > the #winehq IRC channel to help people who want to accept his help... > He tried to help YOU and YOU refused to try what he ask YOU to do!! > YOU kept asking the same stupid *shit* over and over and over until > you got on his last nerve. He cant help someone who doesn't want to > help them self. > > Also, it doesn't look to me like he cussed anyone out! but I guess I > just cussed YOU as I used the poo word in this reply. :D I beg your pardon if I got this wrong, but I seems to me that Jonathan is *Pie-rate* rather than carretto. While carretto did ask the same thing over and over until vitamin gave him the download link, Pie-rate only asked once (timestamp: 09:43:50 PM), to which vitamin replied with his "oh well". I don't want to take any sides here — especially because I refrain from going to support channels exactly to avoid such situations —, but I have to say that vitamin's behavior in that discussion is definitely less than polite. However, I do think that Jonathan made the issue a bit bigger than it actually is. Ah, yes, regarding cussing around, I recall reading vitamin saying "F***" to Pie-rate twice, all the while not cussing *carretto* out. Anyway, being civil does no harm. My 2¢. - Pedro.

5 5

Re: Greek resource file for comctl final
by Maarten Lankhorst Aug. 18, 2007

Aug. 18, 2007

Dj Apal [GR] schreef: > > OK changed the copyright and added the diff file for the rsrc.rc file > so as to be included. > > Thnx for noticing that! > > I think that we are OK now. Sorry for all this mess but it was my > first time posting a patch! Just curious, are files supposed to have dos line terminators? It seems to me like your file has it. Cheers, Maarten

1 0

re: [patch] Garbage code in text string display
by Dan Kegel Aug. 18, 2007

Aug. 18, 2007

Xin Guan wrote: > This patch solves the problem that many MBCS programs display garbage code > for text strings. Because most windows programmers do not handle charsets > properly, we have to force DEFAULT_CHARSET in here. Can you give a simple example of such a program (preferably downloadable)? Can you add a conformance test showing that your patch is correct? - Dan

1 0

Re: Greek resource patch for comctl [third attempt, last I hope!!]
by Paul Vriens Aug. 18, 2007

Aug. 18, 2007

Dj Apal [GR] wrote: > Hope this is what u want us to send! ;) used --> diff -up /dev/null ... etc etc > > > > ------------------------------------------------------------------------ > > Hi, Just a few remarks. The copyright says 'Eric Kohl' is that correct? You also need to add this resource file to 'rsrc.rc' in that same directory otherwise it will never be picked up. Cheers, Paul.

1 0

Re: bugs 9062,9325
by Stefan Dösinger Aug. 17, 2007

Aug. 17, 2007

Am Freitag, 17. August 2007 15:39 schrieb Martin Vanek: > I have posted some minor bugs in combobox a week ago (ID=9062, with a > source text example and pictures documenting what's wrong). I am not an expert on user32 stuff, but generally you should send one patch per mail. If you have 2 patches, just send 2 mails. If you have written a test application, you may want to include it in dlls/user32/tests/ to proove the correctness of your patches and prevent the issue from coming back.

1 0

Re: [PATCH 3/6] winealsa: Implement IDsCaptureDriverImpl_Open
by Tomas Carnecky Aug. 17, 2007

Aug. 17, 2007

Maarten Lankhorst wrote: > Sanity check to see if device can be opened or not Is alloca() not allowed in wine? (asking because you allocate hw_params on the stack, but it's freed if the function exits, so it's only used inside this one function). tom

2 2

Re: check_sharing disregards requested access
by Alexandre Julliard Aug. 17, 2007

Aug. 17, 2007

"James Keane" <james.keane(a)gmail.com> writes: > check_access will now fail because the second caller is not honoring > FILE_SHARE_WRITE. but that shouldn't be neccesary as they are not > planning on writing anyways. The second caller won't, but the first one will, so you need FILE_SHARE_WRITE access. -- Alexandre Julliard julliard(a)winehq.org

1 0

re: A script for automatic regression testing
by Dan Kegel Aug. 17, 2007

Aug. 17, 2007

Martin wrote: > Regarding your script interface, I think it should accept these parameters: > * GIT_ID_1 - bisection start > * GIT_ID_2 - bisection end Sure, and/or dates and/or releases. * PNGs - list of PNG files which describe "bug hit" cases; No. The script should not invoke autohotkey, either. IMHO, it should instead factor out any knowledge about how to test for the bug into a user-supplied shell script. The script would be run once Wine is built, would check for the specific bug being searched for, and its exit status would be zero for success (no bug) and nonzero for failure (the bug was found). So using Mikolaj's script would require you to write a tiny shell script that invoked autohotkey on a script, or maybe just ran an installer and checked for the existance of a file, or something to do with PNG's, whatever you want. http://delta.tigris.org works this way; it calls the shell script "the oracle". - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

1 0

Legal issues for contribution code into Wine
by Dmitry Morovkov Aug. 17, 2007

Aug. 17, 2007

Hello, I'm a (mostly windows, but with some linux skills) programmer from russia who is interested in helping developing Wine, but due to all the mess I read about the legal issues I want to ask first. What is Wine's position regarding people who had seen the windows source code? I tried to search for this topic, and, found this: http://www.gnu.org/prep/standards/html_node/Reading-Non_002dFree-Code.html After reading this, it looks like it's ok if someone have seen the windows source code, and provided he is not going to directly copy unmodified source code (what a stupid idea!), nor refer to it, he would be considered as a good/trusted person? Please clarify this to me. Regards, Dmitry Morovkov.

3 2

Re: Issue(s) when running winetest on Windows
by Kirill K. Smirnov Aug. 17, 2007

Aug. 17, 2007

>file: 2 tests executed (0 marked as todo, 0 failures), 0 skipped. >file: 373 tests executed (0 marked as todo, 0 failures), 0 skipped. These lines are displayed via printf (include/wine/test.h: 391). > msvcrt:file done (0) And this via xprintf (programs/winetest/main.c: 420), which is wrapper over write syscall. If you need these line be output in appropriate way, try to add fflush(stdout) in proper place (into run_test function in include/wine/test.h, I suppose). -- Kirill

4 6

About WritePrivateProfileString and file operations
by Zhongli Xu Aug. 16, 2007

Aug. 16, 2007

Hi all, I run some tests about mixed using WritePrivateProfileString and stand file operations Codes look like this: GetPrivateProfileString(keyname, value); fopen(); fprintf("# comments"); fclose(); WritePrivateProfileString(keyname, newvalue); What I found in the file is that either WritePrivateProfileString operation succeeded(newvalue has been written) or fprintf succeeded("#comments" shows up in the file). On windows hosts, there is no such problem. Any suggestion or idea is welcomed. Thanks. Milo

3 2

RFC: root cert tool
by Juan Lang Aug. 16, 2007

Aug. 16, 2007

Since there wasn't a clear consensus about how to get CA certificates into the registry, I decided to do what Mono does: punt. So I've written a tool that can load certificates from a file or from a URL and stick them in the registry. By default it assumes you want to download them from Mozilla's CVS front-end, and does so. A patch that adds it is attached. Comments? Thanks, --Juan

6 10

Re: secur32: add stub for LsaLookupAuthenticationPackage
by Kai Blin Aug. 16, 2007

Aug. 16, 2007

On Thursday 16 August 2007 21:52:01 Louis. Lenders wrote: > Hi, this fixes a crash in TrueImage installer (bug > http://bugs.winehq.org/show_bug.cgi?id=3572) Personally I'd prefer if the LSA functions could go into a seperate source file. Cheers, Kai -- Kai Blin WorldForge developer http://www.worldforge.org/ Wine developer http://wiki.winehq.org/KaiBlin Samba team member http://www.samba.org/samba/team/ -- Will code for cotton.

1 0

#winehq admin abuse
by Jonathan Challinger Aug. 16, 2007

Aug. 16, 2007

Below is a log of #winehq, from 9:44 to 10:09 on Wednesday August 15th: (09:43:50 PM) Pie-rate: my brother's girlfriend's WoW install is crashing (locking up, stops responding) randomly. she will install windows tomorrow if it doesn't get fixed. i don't like windows and i don't want to deal with it. the message it crashes with is: (09:43:50 PM) Pie-rate: err:ntdll:RtlpWaitForCriticalSection section 0x509da8c "?" wait timed out in thread 0047, blocked by 003e, retrying (60 sec) (09:43:50 PM) Pie-rate: this is with an install copied over from my computer, which i KNOW works flawlessly, and wine 0.9.41 (09:43:54 PM) carretto: vitamin: what? (09:44:10 PM) carretto: vitamin: _period? (09:44:15 PM) vitamin: <vitamin> carretto, you have to install the game on Wine (09:44:34 PM) vitamin: Pie-rate, oh well (09:44:43 PM) Pie-rate: vitamin: "oh well?" (09:44:46 PM) Pie-rate: vitamin: no magic fix? (09:44:55 PM) carretto: vitamin: wine <game install>? (09:45:11 PM) vitamin: carretto, something like that (09:45:33 PM) otaku [n=otaku(a)c-24-15-187-230.hsd1.il.comcast.net] entered the room. (09:46:53 PM) carretto: vitamin: wine: Call from 0x40136e to unimplemented function MFC42.DLL.6929, aborting (09:47:10 PM) vitamin: carretto, get the newer version of that dll (09:49:56 PM) carretto: vitamin: i donwload MFC42.dll on dll.files and past to wine system folder (09:50:04 PM) vitamin: carretto, get the newer version of that dll (09:50:14 PM) carretto: here i get the newer version? (09:50:21 PM) vitamin: from windows (09:50:50 PM) carretto: vitamin: i download version v. 4.21.71 (09:51:32 PM) shingo2 [n=shingo(a)p5487EFF1.dip.t-dialin.net] entered the room. (09:52:27 PM) carretto: vitamin: where i get the new version? (09:54:48 PM) prgrmr [n=prgrmr(a)bzq-88-154-138-152.red.bezeqint.net] entered the room. (09:55:37 PM) _3 [n=lol(a)89-178-108-192.broadband.corbina.ru] entered the room. (09:57:20 PM) carretto: vitamin: can help me? (09:57:38 PM) vitamin: carretto, I already answered all your questions (09:57:50 PM) Pie-rate: vitamin: Read the topic. "End user/tech support channel for Wine." A derogatory-sounding "oh well" is nothing more than flamebait. Either say nothing, or help. That's what I try to do, and that's what's in the best interest of the project. I believe you are a channel admin, correct? Saying "oh well," without even trying to help is rude, but saying "oh well" without even trying to help, as a channel admin? That's like inviting people to an installfest, and then when they ask for help installing something saying "OH WELL" and spitting in their faces. (09:57:59 PM) vitamin: carretto, if you do not undestand answers, then RTFM, Google, etc (09:58:56 PM) carretto: vitamin: where i downlaod a newer version? (09:59:01 PM) vitamin: Pie-rate, Did I offered help? DId you came here looking for help? Or state that fact that you don't F** care and will switch to windows? (09:59:14 PM) vitamin: carretto, Last time: <vitamin> from windows (09:59:35 PM) Pie-rate: vitamin: i'm not the one switching to windows. (09:59:39 PM) vitamin: carretto, I do not f*** care where you download the file. You using _OLD_ version. you need _NEWER_ version of that file! (09:59:55 PM) Pie-rate: vitamin: bit testy, today? (09:59:59 PM) vitamin: carretto, here is the LATEST http://support.microsoft.com/dllhelp/default.aspx?l=55&fid=198511 (10:00:19 PM) vitamin: Pie-rate, STFU if you have nothing to say (10:00:46 PM) Pie-rate: vitamin: same to you. you are unnecessarily cussing out people in a help channel that you administrate. (10:01:33 PM) mode (+o vitamin ) by ChanServ (10:01:33 PM) mode (+b %*!*(a)cust-69-19-214-183.static.o1.com ) by vitamin (10:01:36 PM) mode (-o vitamin ) by ChanServ (10:05:08 PM) Triskelios: vitamin, you really should try to be civil (10:07:08 PM) wckdkl0wn: how would i run 2 copies of diablo 2 in wine? (2 copies in windows) (10:08:15 PM) vitamin: wckdkl0wn, use separate WINEPREFIX for both (10:08:25 PM) mardum [n=mardum(a)71-81-73-42.dhcp.stls.mo.charter.com] entered the room. (10:09:17 PM) vitamin: Triskelios, thanks for useless comment (10:09:35 PM) Triskelios: you're welcome ---------------------------end----------------------------- Vitamin is a disgrace to the OSS community. It is "RTFM and google it" elitists like him who scare people away from switching to open source applications. The fact that he was made an admin is inexcusable. Cussing out everyone asking for help is not behavior an admin should exhibit. I reported this incident in #winehackers, only to be told that there are no standards for wine admins. I suggest you get some standards, this is pathetic. I think my analogy is the best way to describe exactly WHY this is pathetic: "Saying "oh well," without even trying to help is rude, but saying "oh well" without even trying to help, as a channel admin? That's like inviting people to an installfest, and then when they ask for help installing something saying "OH WELL" and spitting in their faces." No, the people in the channel aren't being paid anything to help me or anyone else. They can choose not to help me. But then, what is the point of having the channel at all? This kind of abuse is completely against the spirit of open source and it is despicable that an admin is exhibiting this behavior. You may ask why I'm posting this to the development list. I believe it concerns everyone involved in the project when you have an admin cussing out anyone and everyone in the help channel, compromising its purpose. I suggest you develop some standards for your admins.

4 4

Re: dsound: Use wine dxguid rather then mingw dxguid
by Roderick Colenbrander Aug. 16, 2007

Aug. 16, 2007

On Thursday 16 August 2007 12:06, Maarten Lankhorst wrote: > Makes it cross-compilable to produce a windows binary. I don't think this is the right way. It appears that native dxguid doesn't define some of the uuids needed for directdraw / directsound. You need to check where the missing guids are defined on windows. In case of ddraw IDirectDraw3 stuff is in ddrawex.h whereas we define it in ddraw.h. Something similar could be the case for directsound. Roderick

3 2

Fwd: #winehq admin abuse
by James Keane Aug. 16, 2007

Aug. 16, 2007

> (09:44:15 PM) vitamin: <vitamin> carretto, you have to install the game > on Wine > (09:44:34 PM) vitamin: Pie-rate, oh well I think that vitamin provided a valid answer to your question that would have solved your problem. Why are you getting so worked up that he was a little rude? He is not our employee we can not reprimand him for his actions and neither can you. I think that he did try to help you given the information you provided, and maybe that is the problem with OSS, but if you don't like it go buy cedega. I am sure they have great technical support.

1 0

Poll: where do you have certs installed?
by Juan Lang Aug. 16, 2007

Aug. 16, 2007

Could some of you check where you have, say, OpenSSL's CA certificates installed, and email me what distro you're running, and the path? E.g., I'm running Goobuntu, and I have them installed in /etc/ssl/certs/ca-certificates.crt. Thanks, --Juan

12 13

Re: kernel32:set username (try 4)
by Paul Vriens Aug. 16, 2007

Aug. 16, 2007

Louis. Lenders wrote: > fix bug http://bugs.winehq.org/show_bug.cgi?id=3794 > > > --------------------------------- > Yahoo! Answers - Get better answers from someone who knows. Tryit now. > > > ------------------------------------------------------------------------ > > Aren't the curly brackets strangely placed? The rest of the file don't indent the curly brackets. Cheers, Paul.

1 0

Re: A script for automatic regression testing
by martin pilka Aug. 16, 2007

Aug. 16, 2007

Mikolaj Zalewski wrote: > I now start to understand the structure of the tests. Do you know if > there is a known regression in current Wine that I could use to try to > do some scripting? If not I could use wine-0.9.43 as a reference with > it's Win16 regression that affects the pptviewer and excelviewer > installers. Unfortunately, I am not aware of any regression in current Wine. > PS: are you doing some testing of Photoshop? Yes, we are testing it on 4 distros each night, see i.e. http://www.cxtest.org/raw-data?id_result=635012 search for photoshop word Regarding your script interface, I think it should accept these parameters: * GIT_ID_1 - bisection start * GIT_ID_2 - bisection end * PNGs - list of PNG files which describe "bug hit" cases; usually just 1 screenshot with the meaning "if CxTest script ended with this screenshot, it means it hit the bug we are finding regression patch for" How does it sound to you? Martin

1 0

Re: #winehq admin abuse
by Patrick Leamon Aug. 16, 2007

Aug. 16, 2007

I think the poster was 'Pie-rate' in the chat logs. Pie-rate - you over-reacted to 'oh well' and you're next message was flame bait. Vitamin - you took the bait. I wouldn't have expected that from an admin in a help channel. Take a break if you feel the need to shout/swear at people. Or tell them you can't help them. Tom - I'd feel cussed if I got told to 'STFU'. Which I probably will now as I've never patched wine so my opinion counts for nothing. cheers, pat. ____________________________________________________________________________________ Feel safe with award winning spam protection on Yahoo!7 Mail. http://mail.yahoo.com.au

1 0

re: Fixing wrong prototypes in a header
by Dan Kegel Aug. 15, 2007

Aug. 15, 2007

I'd say leave them wrong in our headers, but report the problem to Microsoft, and fix our headers when they fix theirs... - Dan

1 0

A script for automatic regression testing
by Mikolaj Zalewski Aug. 15, 2007

Aug. 15, 2007

I wrote a small script that automates regression testing. It requires an Autohotkey script that tests the program and creates a file C:\Test-failed.txt if the test failed. The bash script will then do the regression testing. It starts with copying the source tree and wineprefix so it should be possible to do some normal work during compilations. At the beginning of the script there are some variables that needs to be configured before running it. The wineprefix has to have Autohotkey installed and associated with *.ahk. The attached script's settings will find the regression for bug #9249. If anyone is interested in it I'd be interested in some feedback. Mikołaj Zalewski

4 4

Re: [PATCH] Fix Bug in DeadZone calculation
by Vitaliy Margolen Aug. 15, 2007

Aug. 15, 2007

Artur Szymiec wrote: > Here is attached patch > for joystick_linuxinput.c where bug in dead zone > calculation make joystick unusable. After correction > tested in two games and works properly. > > Best Regards > Artur Szymiec > Thanks for spotting the problem. Unfortunately your patch has few problems too. I'll sending a better patch. Vitaliy.

4 5

[wininet/tests] Replace import of ws2_32 by wsock32 to please win95
by Paul Vriens Aug. 15, 2007

Aug. 15, 2007

Hi, Currently we import ws2_32 for our wininet tests. ws2_32 is however not available on win95 and thus presents users of winetest with a pop-up. A crash of the tests I could live with but a pop-up gets in the way of automatic testing (if ever implemented again). I could do LoadLibrary/GetProcAddress to overcome that but a much easier solution seems to be to import wsock32 instead of ws2_32. Comments? Cheers, Paul.

1 0

Problems with focus losing on virtual desktops
by Patrick Leamon Aug. 15, 2007

Aug. 15, 2007

There seems to be a bug for this now: http://bugs.winehq.org/show_bug.cgi?id=9320 ____________________________________________________________________________________ Sick of deleting your inbox? Yahoo!7 Mail has free unlimited storage. http://au.docs.yahoo.com/mail/unlimitedstorage.html

1 0

AMD64 package broken on Ubuntu Gutsy
by Scott Ritchie Aug. 15, 2007

Aug. 15, 2007

The Wine package in Ubuntu Gutsy is currently failing to build on the AMD 64 arch, and I'm not sure why. I don't have a gutsy 64 system to play around with at the moment, but you can see the package error log here: http://launchpadlibrarian.net/8598218/buildlog_ubuntu-gutsy-amd64.wine_0.9.… I suspect one of the 32 bit libraries that we have to link by hand got renamed or upgraded, but I can't figure out which one from the log there. Thanks, Scott Ritchie

2 2

WINEDEBUG output issues
by Paul Vriens Aug. 15, 2007

Aug. 15, 2007

Hi, I was trying to read a '+all,+relay,-syslevel,-gdi,-font' trace but several lines show something like: 000d:Cget_window_property000d:Call0x000d:, atom=000d:Call usePropertySheetInfo000d000000d:Call user32.CharNextA(0 {000d:Call user32.CharNextA(0040a04d "les\\Nmap") ret=0040551c The above is output from the same thread. I also have: 0009:trace:heap:RtlAllocateHeap (0x110000,00000002,0000014000d:Call user32.CharNe0009:trace:dc:GetDCState (0x474): retu000d:Ret user32.CharNextA() retval=004093c6 ret=0040551c Which mixes threads. Debug traces can already be hard to read but this is even more confusing. Any idea what the cause is (or better, the solution)? Cheers, Paul.

1 0

Re: [2/5] WineD3D: Optimize float constants in stateblocks
by Alexandre Julliard Aug. 14, 2007

Aug. 14, 2007

Stefan Dösinger <stefan(a)codeweavers.com> writes: > From f49fd192ed2747663bdb0445dff9d602519fbc95 Mon Sep 17 00:00:00 2001 > From: Stefan Doesinger <stefan(a)codeweavers.com> > Date: Fri, 3 Aug 2007 20:26:29 +0200 > Subject: [PATCH] WineD3D: Optimize float constants in stateblocks > > --- > dlls/wined3d/device.c | 24 +++++++++++++- > dlls/wined3d/stateblock.c | 68 +++++++++++++++++++++------------------- > dlls/wined3d/wined3d_private.h | 4 ++ > 3 files changed, 62 insertions(+), 34 deletions(-) This one is broken: ../../../tools/runtest -q -P wine -M d3d8.dll -T ../../.. -p d3d8_test.exe.so device.c && touch device.ok wine: Unhandled page fault on write access to 0x000000f6 at address 0x61aad11c (thread 0039), starting debugger... WineDbg starting on pid 003a Unhandled exception: page fault on write access to 0x000000f6 in 32-bit code (0x61aad11c). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b EIP:61aad11c ESP:0033fa90 EBP:0033fac8 EFLAGS:00010202( - 00 - -RI1) EAX:000000f6 EBX:61b3ae04 ECX:00000001 EDX:00000000 ESI:001bd4d0 EDI:001c6f78 Stack dump: 0x0033fa90: 001bd4d0 001bd4e4 00000001 00000000 0x0033faa0: 00000000 00000000 00000000 602d53b1 0x0033fab0: 00000000 00000000 0000002b 617da640 0x0033fac0: 617dad24 001bd4b8 0033faf8 617caebf 0x0033fad0: 00123d30 00000001 001bd4c0 001bd4b8 0x0033fae0: 606e97df 0000005c 617cae3b 60715808 Backtrace: =>1 0x61aad11c IWineD3DDeviceImpl_CreateStateBlock+0x3dc(iface=0x123d30, Type=0x1, ppStateBlock=0x1bd4c0, parent=0x1bd4b8) [/home/julliard/wine/wine/dlls/wined3d/device.c:478] in wined3d (0x0033fac8) 2 0x617caebf IDirect3DDevice8Impl_CreateStateBlock+0x8f(iface=0x121780, Type=0x1, pToken=0x33fe08) [/home/julliard/wine/wine/dlls/d3d8/device.c:1177] in d3d8 (0x0033faf8) 3 0x606e1260 func_device+0x2190() [/home/julliard/wine/wine/dlls/d3d8/tests/device.c:528] in d3d8_test (0x0033fe58) 4 0x606e8818 run_test+0x128(name=0x1103c5) [/home/julliard/wine/wine/dlls/d3d8/tests/../../../include/wine/test.h:389] in d3d8_test (0x0033fea8) 5 0x606e8ead main+0x14d(argc=<register ECX not in topmost frame>, argv=<register ECX not in topmost frame>) [/home/julliard/wine/wine/dlls/d3d8/tests/../../../include/wine/test.h:437] in d3d8_test (0x0033fed8) 6 0x606e958b __wine_spec_exe_entry+0x5b(peb=0x7ffdf000) [/home/julliard/wine/wine/dlls/winecrt0/exe_entry.c:36] in d3d8_test (0x0033ff08) 7 0x6041352e start_process+0xee(arg=0x0) [/home/julliard/wine/wine/dlls/kernel32/process.c:820] in kernel32 (0x0033ffe8) 8 0x60027997 wine_switch_to_stack+0x17() in libwine.so.1 (0x00000000) 0x61aad11c IWineD3DDeviceImpl_CreateStateBlock+0x3dc [/home/julliard/wine/wine/dlls/wined3d/device.c:478] in wined3d: movl %edx,0x0(%eax,%edx,4) 478 object->contained_vs_consts_f[j] = j; -- Alexandre Julliard julliard(a)winehq.org

2 1

Partial phase out of *winehq.com addresses
by Duane Clark Aug. 14, 2007

Aug. 14, 2007

As mentioned about a month ago, we were planning to phase out the winehq.com email addresses, and use winehq.org exclusively. It looks like the few holdouts have converted over for awhile now, so I added a filter to the mailman program to silently drop anything mailed to winehq.com. If Jeremy wants to set the MTA (or whatever) to reject those addresses, that might be better. Then anyone accidently mailing to the wrong address would (presumably) get a delivery error message. And it would take a small portion of the load off mailman.

2 1

Re: Font problem + patch solution
by Abel Chiaro Aug. 14, 2007

Aug. 14, 2007

Alexandre Julliard wrote: > penna(a)bb.com.br writes: > > > There is a patch proposed at the bug #7571 (http://bugs.winehq.org/show_bug.cgi?id=7571) > > It works very well and I did not experiment any regression. Maybe > > it is not the ideal patch and if it is not, maybe we can provide a fix for > > that. However, at least it fixes the font problem and did not introduce > > any undesirable efects. > > Is that possible to apply that patch at the main tree? Comments > > are also welcome! > > You should provide test cases that show that your patch is correct. I used Dmitry's tests [1] for this patch. [1] http://www.winehq.org/pipermail/wine-patches/2007-March/036706.html Shouldn't that be enough, I'll look into it again. - Pedro. -- "Caritas omnia vincit."

2 3

Re: WGL: GDI support is not advertised with double buffering
by Roderick Colenbrander Aug. 14, 2007

Aug. 14, 2007

Wouldn't it be cleaner to cache the double buffering property in the WineGLPIxelFormat and set the gdi cap in there? That would skip another unneeded function. Roderick

1 1

Re: user32: LB_SETITEMDATA should return 1 on success, not 0
by Alexandre Julliard Aug. 14, 2007

Aug. 14, 2007

Peter Verthez <peter.verthez(a)advalvas.be> writes: > diff --git a/include/winuser.h b/include/winuser.h > index 4297a8d..97363e0 100644 > --- a/include/winuser.h > +++ b/include/winuser.h > @@ -2428,6 +2428,7 @@ typedef struct tagSCROLLBARINFO > > /* Listbox message return values */ > #define LB_OKAY 0 > +#define LB_SETITEMDATA_OKAY 1 /* undoc. OK return for LB_SETITEMDATA */ Please don't invent constants that don't exist on Windows. In this case TRUE would work just as well. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: [d3d8/tests] Make tests run again on win95
by James Hawkins Aug. 14, 2007

Aug. 14, 2007

On 8/14/07, Paul Vriens <paul.vriens.wine(a)gmail.com> wrote: > Hi, > > Again only to avoid a pop-up. Didn't fix/test the actual tests. > > Changelog > Make tests run again on win95 > Please change the subject of your emails to match the dll that you're patching. As it stands, it looks like you're patching advapi32 5 times, and we have to dig into the patch to see what dll you're actually changing. -- James Hawkins

2 2

Font problem + patch solution
by penna＠bb.com.br Aug. 14, 2007

Aug. 14, 2007

Hi all, Since wine 0.9.31 there is a font problem (at least latin charset). There is a patch proposed at the bug #7571 (http://bugs.winehq.org/show_bug.cgi?id=7571) It works very well and I did not experiment any regression. Maybe it is not the ideal patch and if it is not, maybe we can provide a fix for that. However, at least it fixes the font problem and did not introduce any undesirable efects. Is that possible to apply that patch at the main tree? Comments are also welcome! Thank you in advance.

2 1

Problems with focus losing on virtual desktops
by Christian Authmann Aug. 14, 2007

Aug. 14, 2007

Hi, after installing 0.9.42 I noticed some odd behavior when playing Diablo 2. I did a git bisect and nailed it down to this patch: http://source.winehq.org/git/wine.git/?a=commit;h=d836a5062141dd42293ed044d… winex11drv: Correctly react to focus loss away from Wine. I noticed the following changes: a) Using diablo 2's windowed mode: d2 will realize when it's losing focus again (as it did somewhere around 0.9.2x). As far as I can tell, this works reliable. (That's not a good thing for diablo 2, since d2 will minimize on losing focus and cannot be brought back to life afterwards. But that's a different bug.) >From now on, please read "Diablo 2 minimizes" as "Diablo 2 thinks it lost focus". b) Using d2 in fullscreen mode inside wine's virtual desktop: d2 will sometimes think it lost focus, even if it didn't. Moving the wine window around with hotkey+drag: d2 minimizes. Moving the window by the title bar: nothing happens. (using kde/kwin btw) Sometimes just clicking inside the window while playing will cause d2 to minimize, although I haven't found a reproducible pattern (i.e. do I have to press certain keys, do I have to double-click, etc). Activating a different window outside the virtual desktop? d2 stays open. Or opening another application via hotkey, or pulling down yakuake via hotkey. It stays open. But, if I reactivate the virtual desktop window afterwards: d2 minimizes. Ouch. c) virtual desktop again: when holding ALT to show the dropped items, then left clicking somewhere, d2 thinks that ALT has been released. This makes it quite hard to pick up items. This problem does not occur when using regular windowed mode or when changing the hotkey to something different - as far as I can tell only ALT in virtual desktop mode is affected. Note that I don't have any hotkeys on Alt (or Alt+click), I bound window dragging to the otherwised unused windows key. While the patch does as intended on regular windows, it's effects on virtual desktop windows can only be described as weird. I don't know how virtual desktops are supposed to work. At least for d2, it'd be useful if one application inside the virtual desktop always keeps focused (or rather, is told it'd still be focused). IMHO that's the purpose of virtual desktops: banning the apps inside a container where they shouldn't be influenced by outside stuff, including activating a different window. Not losing focus is also more sensible when virtual desktops are used as a way to ban fullscreen-apps into windows: most fullscreen applications don't expect to lose focus and behave weird. However, even if you think of virtual desktops simply as a means to group windows, it's currently broken: losing focus on reactivation is a little late. I have compiled the latest git version without this patch, and virtual desktop mode works fine again, so it's really down to this single line of code. I'm not far enough into the code to know any possible solution, but for immediate results a simple if (inside a virtual desktop) do old behaviour else do new behaviour should do the trick until a better solution is found. greetings, Christian

3 5

Re: kernel: add execution permission to DOS memory area
by Alexandre Julliard Aug. 14, 2007

Aug. 14, 2007

Mikolaj Zalewski <mikolaj(a)zalewski.pl> writes: > Execution permission is needed because we store some (mostly > DPMI-related) code in UMB (in winedos/himem.c) that is executed for > each Win16 app. The user may put even more code there. Ntvdm.exe also > makes the DOS memory executable (except for some parts of the UMB) winedos is supposed to make that memory executable when needed. The executable protection was removed from kernel32 as it causes failures on platforms that don't allow write+exec. -- Alexandre Julliard julliard(a)winehq.org

1 0