January 2006 - Wine-Devel - list.winehq.org

Need help in debugging a stack corruption
by Stefan Dösinger Jan. 7, 2006

Jan. 7, 2006

Hello, I am working on a DDraw rewrite with WineD3D, and I've hit a problem with Half-Life, which looks like a stack corruption. HL makes a ret from a function which leads into sensless code, and I wasn't able to debug it. That's what happens: From 0x00428a35 there's a call to a function at 0x00408b70. This function calls a lot of DDraw calls, which seem to work fine, but on return, it doesn't return to 0x00428a3a, as it should, but to 0x7fbcd888, which is valid, but contains only garbage code. I assumed that some call runs amok on the stack, and I've set tried setting breakpoints at 0x00428a35, 0x00408b70 and some other addresses. Oddly, this doesn't work(winedbg disables them when continuing), although I can access the code right after start. I can set a DebugBreak() in the last called function(IDirectDraw::FlipToGDISurface), but there it is too late. Are there any other ways to debug this? Does anyone know why setting the breakpoints fails? Well, that's my suspicion for the problem: (an any COM expert comment on this?) To create interfaces for IDirectDraw1 to 7, I created 4 lpVtbl structures for each version, and I cast most DD7 funtions info the older versions. Only where the type or the number of parameters is different, I use a wrapper function. At DD creation, I use whatever version was requested by the app for the new object. The old dd version used a number of macros to cast the various interfaces(defined in ddcomimpl.h). I don't use them. Might this cause the problem? Thanks for your help, Stefan

4 7

Re: Rein Klazes : user32: Separate menu bitmaps and strings.
by paul Jan. 7, 2006

Jan. 7, 2006

Alexandre Julliard wrote: > Module: wine > Branch: refs/heads/master > Commit: b34868dffa0df144ea9d6e1586a67cff70ae9c4e > URL: http://source.winehq.org/git/?p=wine.git;a=commit;h=b34868dffa0df144ea9d6e1… > > Author: Rein Klazes <wijn(a)wanadoo.nl> > Date: Tue Jan 3 13:23:28 2006 +0100 > > user32: Separate menu bitmaps and strings. > Store bitmaps and bitmaps always in separate fields in the internal > menu structure. Get rid of a lot of assumptions in the code that the > menu can have strings only when it is not some other type and that > bitmaps come in two flavors. > Add a lot of conformance tests, including some submitted by Jason > Edmeades. > > --- > > dlls/user/menu.c | 317 ++++++++++++++++---------- > dlls/user/tests/menu.c | 593 ++++++++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 785 insertions(+), 125 deletions(-) > > > > Regression see http://bugs.winehq.org/show_bug.cgi?id=4250 The set last error function added to GetMenuStringA and GetMenuStringW removed this check which seems to be still needed. if (!IS_STRING_ITEM(item->fType)) return 0; I have tried several times to make a patch of the changes required but I get a lot of white space changes. Paul R.

3 2

Microsoft Office on FreeBSD
by Tom Wickline Jan. 7, 2006

Jan. 7, 2006

Hello, I've put together a little Microsoft Office on BSD howto on the Wiki, If anyone here uses OpenBSD, NetBSD or DragonFlyBSD I would be interested to know if you can get Office 2000 to install and run with Wine out of the box or with my cheats. :D So others who read the howto in the future will know the status of there distro. http://wiki.winehq.org/Office-BSD?action=show Tom

1 0

Re: shell32: Reimplement a factored SHFileOperation [RESEND 2]
by Mike McCormack Jan. 7, 2006

Jan. 7, 2006

James Hawkins wrote: > Haven't received any comments yet. Is there anything wrong with this > patch? I know it's large, but it can't be sent in pieces. The new > implementation passes all the tests. > > Changelog: > * Reimplement a factored SHFileOperation. > > dlls/shell32/shlfileop.c | 1131 +++++++++++++++++++--------------------- > dlls/shell32/tests/shlfileop.c | 242 ++------- Hi James, That's a lot of great work! Try submit the tests first with todo_wine {} blocks around the stuff that fails now, then submit a further change with your main patch that shows with tests now pass. That should reduce the size of your patch a little, and show how messed up the old implementation of SHFileOperation is. thanks, Mike

1 0

Re: commit 15667b36e43a571f6bb8726b9dec4f1099d2b4f9 (_fullpath)
by Detlef Riekenberg Jan. 6, 2006

Jan. 6, 2006

Am Freitag, den 06.01.2006, 10:01 -0600 schrieb Aric Stewart: > ChangeLog: > Additional add a regression test for _fullpath > > dlls/msvcrt/tests/dir.c | 58 + + SetCurrentDirectory("C:\\Windows\\System\\"); + Please do not use a fixed Path. This will not work on many Systems: nt351: "c:\winnt351" nt4.0 + w2k: "c:\winnt" and every windows with a custom Install-Directory. + + ok(_fullpath(full,"test", MAX_PATH)>0,"_fullpath failed\n"); + ">0" is not correct for "char *". The Text in your Tests are useless. Please print error-values and received buffer-content. -- By By ... ... Detlef

1 0

Re: winspool: Printing-Environment-Support for GetPrinterDriverDirectory
by Detlef Riekenberg Jan. 6, 2006

Jan. 6, 2006

Am Dienstag, den 03.01.2006, 00:51 +0100 schrieb Detlef Riekenberg: What can I do with this Patch ? > Changelog: > - Add Function "validate_envW" to validate the user-supplied > Printing Environment (respect "Windows Version" - Setting) > - Change GetPrinterDriverDirectoryW on top of "validate_envW" > to Report the native Path for NT and Win9x > - 2 Tests are now working in Wine > > > Synced to current git-version with suggestions from Alexandre: > Use a struct to handle the environment-specific parts > > > "validate_envW" / printenv_t will be used by more Functions later. > (WINSPOOL_OpenDriverReg as Example) > -- By By ... ... Detlef

1 0

Re: shell32: fix folder icon index when read from registry
by Martin Fuchs Jan. 6, 2006

Jan. 6, 2006

Any thing wrong with this patch? On 31.12.2005 21:31:25 Martin Fuchs wrote: > Changelog: > fix folder icon index when read from registry > Index: folders.c > =================================================================== > RCS file: /home/wine/wine/dlls/shell32/folders.c,v > retrieving revision 1.66 > diff -u -p -d -r1.66 folders.c > --- folders.c 26 Aug 2005 10:05:35 -0000 1.66 > +++ folders.c 31 Dec 2005 20:29:30 -0000 > @@ -166,7 +166,7 @@ static HRESULT getIconLocationForFolder( > LPWSTR szIconFile, UINT cchMax, int *piIndex, UINT *pwFlags) > { > IExtractIconWImpl *This = (IExtractIconWImpl *)iface; > - DWORD dwNr; > + DWORD dwNr; /*TODO: change to int type */ > WCHAR wszPath[MAX_PATH]; > WCHAR wszCLSIDValue[CHARS_IN_GUID]; > static const WCHAR shellClassInfo[] = { > '.','S','h','e','l','l','C','l','a','s','s','I','n','f','o',0 }; > @@ -202,9 +202,13 @@ static HRESULT getIconLocationForFolder( > if (!HCR_GetDefaultIconW(folder, szIconFile, cchMax, &dwNr)) > { > lstrcpynW(szIconFile, swShell32Name, cchMax); > - dwNr = IDI_SHELL_FOLDER; > + dwNr = (DWORD)-IDI_SHELL_FOLDER; > } > - *piIndex = -((uFlags & GIL_OPENICON) ? dwNr + 1 : dwNr); > + > + if (uFlags & GIL_OPENICON) > + *piIndex = (int)dwNr<0? (int)dwNr-1: dwNr+1; > + else > + *piIndex = dwNr; > } > return S_OK; > }

2 1

Re: commctl32 datetime fields mad editable with keyboard
by Robert Shearman Jan. 6, 2006

Jan. 6, 2006

Siim Karus wrote: >@@ -962,6 +1037,33 @@ > DATETIME_IncreaseField (infoPtr, fieldNum, INT_MAX); > DATETIME_SendDateTimeChangeNotify (infoPtr); > break; >+ case VK_NUMPAD0: >+ case VK_NUMPAD1: >+ case VK_NUMPAD2: >+ case VK_NUMPAD3: >+ case VK_NUMPAD4: >+ case VK_NUMPAD5: >+ case VK_NUMPAD6: >+ case VK_NUMPAD7: >+ case VK_NUMPAD8: >+ case VK_NUMPAD9: >+ DATETIME_EditField (infoPtr, fieldNum, vkCode-VK_NUMPAD0); >+ DATETIME_SendDateTimeChangeNotify (infoPtr); >+ break; >+ case (WORD)'0': >+ case (WORD)'1': >+ case (WORD)'2': >+ case (WORD)'3': >+ case (WORD)'4': >+ case (WORD)'5': >+ case (WORD)'6': >+ case (WORD)'7': >+ case (WORD)'8': >+ case (WORD)'9': >+ DATETIME_EditField (infoPtr, fieldNum, vkCode-(WORD)('0')); >+ DATETIME_SendDateTimeChangeNotify (infoPtr); >+ break; >+ > case VK_LEFT: > do { > if (infoPtr->select == 0) { > > I think this would be better done in the processing for the WM_CHAR message instead. Then you wouldn't have the mess of having to handle the numpad keys as well. -- Rob Shearman

1 0

bug in dlls/wined3d/device.c?
by Daniel Skorka Jan. 6, 2006

Jan. 6, 2006

Hi, I have a question regarding the function IWineD3DDeviceImpl_ActiveRender at lines 5806-5807 (that is in wine 0.9.5). If I read [1] correctly, these two lines: PUSH2(GLX_DRAWABLE_TYPE, GLX_PBUFFER_BIT); PUSH2(GLX_X_RENDERABLE, TRUE); are illegal, as GLX_X_RENDERABLE must be FALSE when GLX_PBUFFER_BIT is set. Comments? Daniel [1] http://publib.boulder.ibm.com/infocenter/pseries/v5r3/index.jsp?topic=/com.…

1 0

WINE 0.9.3 and Photoshop 7 & CS
by Aleksey Jan. 6, 2006

Jan. 6, 2006

Yesterday I upgraded to WINE 0.9.3, and then tried running Photoshop 7. Photoshop gave me a hardware error. I thought, no problem, I'll just reinstall it, same result. Then I reconfigured WINE using winetools and tried again, same thing. Here is what I got in the Terminal when running Photoshop 7: fixme:actctx:QueryActCtxW stub! fixme:font:WineEngRemoveFontResourceEx :stub fixme:win:EnumDisplayDevicesW ((null),0,0x7fa7de2c,0x00000000), stub! fixme:win:EnumDisplayDevicesW (L"\\\\.\\DISPLAY1",0,0x7fa7de2c,0x00000000), stub! fixme:mscms:EnumColorProfilesA ( (nil), 0x7fa7e694, 0x7fa7e1ac, 0x7fa7e6f4, 0x7fa7e704 ) stub fixme:win:EnumDisplayDevicesW ((null),0,0x7fa7de34,0x00000000), stub! fixme:win:EnumDisplayDevicesW (L"\\\\.\\DISPLAY1",0,0x7fa7de34,0x00000000), stub! fixme:mscms:EnumColorProfilesA ( (nil), 0x7fa7e69c, 0x7fa7e1b4, 0x7fa7e6fc, 0x7fa7e70c ) stub err:shell:HCR_GetFolderAttributes HCR_GetFolderAttributes should be called for simple PIDL's only! fixme:font:WineEngRemoveFontResourceEx :stub fixme:font:WineEngRemoveFontResourceEx :stub Then I tried installing Photoshop CS, knowing that it worked fine with 0.9.2. Photoshop CS told me that certain required files was missing, but I forgot to get down the results from the Terminal. After that I downgraded to WINE 0.9.2 and both versions of Photoshop work great. A little side note: ImageReady 7 works great under 0.9.3, I did not test ImageReady CS. ImageReady 7 works great under 0.9.2 also, but ImageReady CS crashes. God Bless. - Aleksey

6 8