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

recent ole32/shdocvw "regression" crashes adobe installers
by James Hawkins Oct. 7, 2006

Oct. 7, 2006

Hi, A recent patch by Rob Shearman has caused two Adobe installers to crash when they try to load up a WebBrowser control to display a EULA: Commit: ccb634fdee50936c35880a6cffd70e0bcd946b5a URL: http://source.winehq.org/git/?p=wine.git;a=commit;h=ccb634fdee50936c35880a6… Author: Robert Shearman <rob at codeweavers.com> Date: Thu Aug 31 17:14:38 2006 +0100 ole32: Re-structure OleCreate to match to order of calls that native does. Get rid of hres1 since all failures are returned. Cleanup pUnk on failure and make sure to return NULL in ppvObj. We get the following backtrace: fixme:ole:OleCreate {8856f961-340a-11d0-a96b-00c04fd705a2} {00000112-0000-0000-c000-000000000046} semi-stub! fixme:shdocvw:PersistStorage_InitNew (0x1f62b0)->(0x10013378) wine: Unhandled page fault on read access to 0x00000000 at address 0x100047b0 (thread ... Backtrace: =>1 0x100047b0 in epic_eula (+0x47b0) (0x100047b0) fixme:dbghelp:elf_load_debug_info_from_map Alpha-support for Dwarf2 information for user32<elf> 2 0x7ebcf97a WINPROC_wrapper+0x1a in user32 (0x7ebcf97a) 3 0x7ebd0126 call_dialog_proc+0x66(hwnd=0x10034, msg=0x110, wp=0x0, lp=0x761958, result=0x33bbbc, arg=0x10009010) [/home/jhawkins/wine/dlls/user/winproc.c:473] in user32 (0x7ebd0126) 4 0x7ebd490d WINPROC_CallDlgProcW+0x4d(func=0xffff0020, hwnd=0x10034, msg=0x110, wParam=0x0, lParam=0x761958) [/home/jhawkins/wine/dlls/user/winproc.c:2339] in user32 (0x7ebd490d) 5 0x7eb7118b DefDlgProcW+0x5b(hwnd=0x10034, msg=0x110, wParam=0x0, lParam=0x761958) [/home/jhawkins/wine/dlls/user/defdlg.c:493] in user32 (0x7eb7118b) 6 0x7ebcf97a WINPROC_wrapper+0x1a in user32 (0x7ebcf97a) 7 0x7ebd1956 call_window_proc+0x66(hwnd=0x10034, msg=0x110, wp=0x0, lp=0x761958, result=0x33bcbc, arg=0x7eb71130) [/home/jhawkins/wine/dlls/user/winproc.c:452] in user32 (0x7ebd1956) 8 0x7ebd49fd CallWindowProcW+0x4d(func=0x7eb71130, hwnd=0x10034, msg=0x110, wParam=0x0, lParam=0x761958) [/home/jhawkins/wine/dlls/user/winproc.c:2252] in user32 (0x7ebd49fd) 9 0x7eba3197 call_window_proc+0x127(hwnd=<register not in topmost frame>, msg=0x110, wparam=0x0, lparam=0x761958, unicode=0x1, same_thread=0x1) [/home/jhawkins/wine/dlls/user/message.c:1531] in user32 (0x7eba3197) 10 0x7eba656c SendMessageTimeoutW+0x16c(hwnd=0x10034, msg=0x110, wparam=0x0, lparam=0x761958, flags=0x0, timeout=0x0, res_ptr=0x33bdd4) [/home/jhawkins/wine/dlls/user/message.c:2374] in user32 (0x7eba656c) 11 0x7eba65b7 SendMessageW+0x37(hwnd=0x10034, msg=0x110, wparam=0x0, lparam=0x761958) [/home/jhawkins/wine/dlls/user/message.c:2460] in user32 (0x7eba65b7) 12 0x7eb769a1 DIALOG_CreateIndirect+0xe91(hInst=0x340000, dlgTemplate=<register not in topmost frame>, owner=0x10024, dlgProc=0x10009010, param=0x761958, unicode=0x1, modal=0x1) [/home/jhawkins/wine/dlls/user/dialog.c:637] in user32 (0x7eb769a1) 13 0x7eb7732f DialogBoxParamW+0x5f(hInst=0x340000, name=0x65, owner=0x10024, dlgProc=0x10009010, param=0x761958) [/home/jhawkins/wine/dlls/user/dialog.c:798] in user32 (0x7eb7732f) 14 0x10003114 in epic_eula (+0x3114) (0x10003114) 15 0x100012ea in epic_eula (+0x12ea) (0x100012ea) 16 0x00000000 (0x00000000) 0x100047b0: movl 0x0(%eax),%eax epic_eula calls OleCreate. OleCreate calls shdocvw:PersistStorage_InitNew which returns E_NOTIMPL. The new version of OleCreate handles the error and sets pUnk to NULL and returns the error value. epic_eula doesn't check the return value for an error or check if pUnk is NULL. When the dialog loads up, epic_eula tries to use pUnk and we crash. Note that Rob's patch is correct and shouldn't be changed. Making PersistStorage_InitNew return S_OK fixes the problem, but I don't know if that is acceptable or not. The real solution is to implement PersistStorage. What are people's thoughts on this? Is it OK to return S_OK for now? Thanks, James Hawkins

2 1

Wine and HeidiSQL
by Ansgar Becker Oct. 7, 2006

Oct. 7, 2006

Hi there, I'm Ansgar Becker, the main-author of HeidiSQL. I saw that HeidiSQL is in your appdb! very cool ;) http://appdb.winehq.org/appview.php?iAppId=3326 However, some of the buttons on HeidiSQL's main-toolbar are twisted. I'm not a C programmer, so I'm a bit lost in Wine... but I can fix it in HeidiSQL. The only thing is, how to properly detect if I'm running under Wine? Any help would be appreciated! Greetings and thanks a lot! Ansgar

3 2

Re: oleaut32: Add 64-bit proxy/stub code for core COM interfaces
by Mike McCormack Oct. 6, 2006

Oct. 6, 2006

Ge van Geldorp wrote: > Changelog: > Ge van Geldorp <ge(a)gse.nl> > - Add 64-bit proxy/stub code for core COM interfaces It's unlikely that this patch will be accepted. It's better to fix widl to generate the 64bit code rather than add more midl generated code. Mike

1 0

ping about bug #1598
by Alex Villacís Lasso Oct. 6, 2006

Oct. 6, 2006

Has there been any discussion or progress on how to fix bug #1598 about Delphi/VisualBasic applications not having minimize/maximize decorations (http://bugs.winehq.org/show_bug.cgi?id=1598)? According to one of the comments, the window management code would not be touched before 0.9 . However wine is already in beta status and the bug remains unfixed. Should this be fixed in the wine side, or the metacity side? -- The following cryptic message was allegedly found in the inner edge of a Windows XP installation CD: 4F6E65204F5320746F2072756C65207468656D20616C6C2C204F6E65204F5320746F2066696E6420 7468656D2C0D0A4F6E65204F5320746F206272696E67207468656D20616C6C20616E6420696E2074 6865206461726B6E6573732062696E64207468656D2E0A It is rumored that only a true Unix Wizard can decypher this mysterious message, which supposedly encodes the true nature and purpose of the software.

1 0

Coverity scan database reset soon...
by Dan Kegel Oct. 6, 2006

Oct. 6, 2006

Coverity has to reset their database soon to recover from a database problem (the one that's keeping new scans from showing up). Somebody should probably scrape the Wine results from the Coverity database soon so we can avoid losing any annotations we care about... - Dan

3 7

Re: [PATCH] oleaut32: Add tmarshal conformance test
by James Hawkins Oct. 6, 2006

Oct. 6, 2006

On 10/6/06, Michal Okresa <michal.okresa(a)ifne.eu> wrote: > --- > dlls/oleaut32/tests/tmarshal.c | 1099 > ++++++++++++++++++++++++++++++++ > dlls/oleaut32/tests/tmarshal.idl | 159 +++++ > dlls/oleaut32/tests/tmarshal.rc | 30 + > dlls/oleaut32/tests/tmarshal_dispids.h | 34 + > 4 files changed, 1322 insertions(+), 0 deletions(-) > > diff --git a/dlls/oleaut32/tests/tmarshal.c b/dlls/oleaut32/tests/tmarshal.c > new file mode 100644 > index 0000000..be22443 > --- /dev/null > +++ b/dlls/oleaut32/tests/tmarshal.c > @@ -0,0 +1,1099 @@ > +/* > + * Copyright (C) 2005-2006 Robert Shearman for CodeWeavers > + * Um...are you sending this in for Rob Shearman? If so, you need to add a changelog with his name in it. -- James Hawkins

1 0

Re: WineD3D: Add WINED3DUSAGE_OVERLAY
by Vitaliy Margolen Oct. 6, 2006

Oct. 6, 2006

Stefan Dösinger wrote: > D3D9 does not have Overlay surfaces like ddraw does, so we need a > wined3d-private way to create ddraw overlays. I decided to add a new usage > flag outside of the flags defined by Direct3D and use it in wined3d. d3d8 and > d3d9 have to filter usage flags which are not valid in d3d8 and d3d9 to > prevent broken applications from accidentally passing WINED3DUSAGE_OVERLAY. > > D3D9 does not need overlays, because dynamic textures + drawing textured quads > can provide simmilar functionality. But it is hard to implement ddraw ovelays > using dynamic textures in ddraw.dll in an efficient way, so I decided to > implement overlays in the ddraw way in wined3d. > > > @@ -274,7 +274,7 @@ static HRESULT WINAPI IDirect3DDevice9I > > TRACE("(%p) : w(%d) h(%d) fmt(%d) surf@%p\n", This, Width, Height, Format, *ppSurface); > > - hrc = IWineD3DDevice_CreateSurface(This->WineD3DDevice, Width, Height, Format, Lockable, Discard, Level, &object->wineD3DSurface, Type, Usage, (WINED3DPOOL) Pool,MultiSample,MultisampleQuality,pSharedHandle,SURFACE_OPENGL,(IUnknown *)object); > + hrc = IWineD3DDevice_CreateSurface(This->WineD3DDevice, Width, Height, Format, Lockable, Discard, Level, &object->wineD3DSurface, Type, Usage & WINED3DUSAGE_MASK, (WINED3DPOOL) Pool,MultiSample,MultisampleQuality,pSharedHandle,SURFACE_OPENGL,(IUnknown *)object); > Should we split lines like this? It's doesn't fit even on my 1600x1200 display. > diff --git a/include/wine/wined3d_types.h b/include/wine/wined3d_types.h > index e70a814..278e7d7 100644 > --- a/include/wine/wined3d_types.h > +++ b/include/wine/wined3d_types.h > @@ -1053,6 +1053,7 @@ #define WINED3DUSAGE_DYNAMIC > #define WINED3DUSAGE_AUTOGENMIPMAP 0x00000400L > #define WINED3DUSAGE_DMAP 0x00004000L > #define WINED3DUSAGE_MASK 0x00004FFFL > +#define WINED3DUSAGE_OVERLAY 0x00010000L > I think it would be safer to make it 0x10000000L instead. Vitaliy

5 4

Adobe Photoshop idea
by Karsten Anderson Oct. 6, 2006

Oct. 6, 2006

I know a lot of people really want newer versions of photoshop (CS, CS2) to work with wine, myself included, so I opened up a feature request on Adobe's Photoshop forum, first with the intent to lobby for a Linux version, but then had the idea that maybe it would be more realistic to get Photoshop to work in Linux via Wine, possibly via having some sort of collaboration between the two dev groups. A reply I got on their messageboard is actually from a dev, and I suggested to him that they work with the Wine devs on this, as it seems much more realistic and possible than a Linux port. So I'm sending this to you guys as a heads up that this might be a good opportunity to get in contact with the Adobe people about getting some ideas on getting Photoshop working in Wine. Here's the link to the thread: http://www.adobeforums.com/cgi-bin/webx/.3bc1d4af/7 maybe someone could reply and say Hi at least and say to Chris (that's the guy) that you'd be receptive to some input from Adobe on this. I can only see this as a win-win situation for everyone, Linux users for getting Photoshop working, and Adobe for the admiration of the Linux community and the possibility of new revenue that might be available to them via new sales of the software. Sorry to drop in from out of nowhere on this, but I thought the opportunity seemed to good to not say anything. So long and thanks for wine, it's awesome :) Karsten

30 59

Re: [WINED3D 1/6] Move D3DSIO structure into the WINED3D namespace.
by Alexandre Julliard Oct. 6, 2006

Oct. 6, 2006

Ivan Gyurdiev <ivg231(a)gmail.com> writes: > Note: A new file was started for this purpose: wined3d_types.h. This > is _not_ the same as the one in the wine include folder - that one > should be for types which appear in the interface of wined3d, and must > therefore be exposed to other dlls. This one is private. I think the > dll interface should be strictly enforced, so private structures have > no place outside our folder. This will cause massive confusion. There is already a wined3d_private.h file, you should put that stuff in there. -- Alexandre Julliard julliard(a)winehq.org

4 5

Re: DDraw: Protect the ddraw object list
by Mike McCormack Oct. 6, 2006

Oct. 6, 2006

Stefan Dösinger wrote: > @@ -851,6 +853,13 @@ DllMain(HINSTANCE hInstDLL, > + /* Initialize the ddraw object list critical section unconditionally > + * Even if the app doesn't use multithreaded ddraw objects it can still > + * use multiple ddraw objects in multiple threads > + * Well, initialize it only once > + */ > + if(counter == 0) InitializeCriticalSection(&listCrit); > + Isn't "counter" redundant? When DllMain is called, the loader lock is held. Furthermore, DllMain( ... DLL_PROCESS_ATTACH) is only called once per process, so counter should only ever be 1 after a DLL_PROCESS_ATTACH and 0 after a DETACH. Mike

2 1