- Wine-Devel - list.winehq.org

Re: [wine-test-devel] Patch - cabinet.dll FDI Conformance Test
by Dan Kegel March 11, 2005

March 11, 2005

Rizwan Kassim wrote: > Signed-off-by: Rizwan Kassim <rizwank(a)geekymedia.com> > Signed-off-by: Dan Kegel <dank(a)kegel.com> > Signed-off-by: Aleksandr Liber <aliber(a)ucla.edu> I didn't sign off on this version, for what it's worth. Rizwan, please only put the Signed-off-by header with my name on it if I've actually reviewed the patch being submitted. A few notes: --- #define _S_IREAD 0x0100 #define _S_IWRITE 0x0080 --- I told you earlier that's not a good idea; you should get those from a standard windows header. --- #ifndef WIN_ENV #include <msvcrt/fcntl.h> #else #include <fcntl.h> #endif --- I told you earlier that plain old <fcntl.h> should do, and it works fine here with msvc and winegcc. You shouldn't ever need to include <msvcrt/fcntl.h>. You decompress complex.cab but don't verify the results; that seems bad. Can you address those issues, please? - Dan -- Trying to get a job as a c++ developer? See http://kegel.com/academy/getting-hired.html

1 0

Re: Theming for Wine via the registry
by Oliver Stieber March 11, 2005

March 11, 2005

--- Scott Ritchie <scott(a)open-vote.org> wrote: > This thread on the forums caught my attention: > > http://www.ubuntuforums.org/showthread.php?p=86939&highlight=wine#post86939 > > 1) Is there a way we can make Wine respect the > current GTK themes? > Wasn't someone working on this a while ago? What's > the status? I think this falls under... http://www.winehq.org/site/fun_projects Integrate the Registry With Native Configurations Implement a bridge between the Windows registry and the GNOME/KDE configuration systems. If you set the wallpaper in a Windows app it should reflect on your real desktop. This is useful for programs like the WebShots desktop. This could be done by extending the wineexplorer workers: wanted status: to do updated: Dec 21, 2004 It's fairly easy to do, just copy the GTK system settings into the correct registry entries on startup and visa-versa. I have an old email from myself that tells me to take a look at .... hkey_users \console ... Control Panel (for desktop stuff) and Keyboard Layout Software\microsoft\windows\currentversion\print\printers and Software\microsoft\windows\currentversion\profileList.. Software\microsoft\windows\currentversion\profileGuid HKEY_Local_machine\software\Microsoft\Windows\CurrentVersion\SyncMgr\AutoSync\DOMAIN_UNAME HKEY_Local_machine\system\controlSetxxx\services\* So I expect that's a good place to start looking. details on reading-writing Gnome/Kde settings should be at www.freedesktop.org > > 2) If I want to hack around this by changing the > default registry for > the Ubuntu packages, what's the best way to do that? > > Thanks, > Scott Ritchie > > > Send instant messages to your online friends http://uk.messenger.yahoo.com

1 0

flexible-mmap breaks application
by Uwe Bonnes March 11, 2005

March 11, 2005

Hallo, from the XILINX Ise suite (version 6.3 tested), the GUI (_pn.exe) and the web update (webupdate.exe) crash with recent CVS when run on a 2.6 Kernel (Suse 9.2, 2.6.11.1-20050307183958-default, but also the default 2.6 distribution kernel and a Athlon/ Pentium M). Both programs run fine with 2.4 (meeding a step back to Suse 9.0, but the same wine executable, only tested on the Athlon). The problem was hard to pinpoint, but I suspected differences in HeapAlloc() return values to cause the problem, hinted by looking at the disassembly where the error happens. Clearly there is a difference between 2.4 and 2.6 kernels: While mmap2() in kernel 2.4 in map_view() returns values in the 0x4000 0000 range, kernel 2.6 returns values like 0xb7fe9000, 0x58710000, 0xa07f0000 and 0x77abf000. A check in RtlAllocateHeap() for return values above 0x80000000 also triggers with both programs when run on 2.6 and not with 2.4. Looking deeper, I found http://lwn.net/Articles/92281/ about flexible-mmap-update.patch and the wine-devel discussion from around new year about "Re: ptrace single-stepping change breaks Wine". Setting /proc/sys/vm/legacy_va_layout, like proposed by Ingo, lets the app finally run. Alexandre finally concluded the discussion by > Probably the easiest would be to have a way for an app to specify the > mmap range it wants. So instead of having the kernel try to guess from > brk and stack ulimit, both of which are meaningless for Win32 apps, we > could set the range from "end of win32 exe" to 0x7ff0000. This would > also avoid the need to preallocate everything above 0x80000000 that we > currently do and that plays havoc with address space limits. However nothing seems to have happened with regard to that problem until now. Could we revive that discussion? Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

4 7

Re: [DMUSIC/DINPUT] includes fixes
by Alexandre Julliard March 11, 2005

March 11, 2005

Raphael <fenix(a)club-internet.fr> writes: > --- dinput.h 5 Oct 2004 04:38:15 -0000 1.46 > +++ dinput.h 10 Mar 2005 22:13:43 -0000 > @@ -640,7 +640,9 @@ > DWORD dwData; > DWORD dwTimeStamp; > DWORD dwSequence; > +#if(DIRECTINPUT_VERSION >= 0x0800) > UINT_PTR uAppData; > +#endif As a rule we avoid that sort of thing in Wine (I know there are other such #ifs in that file but they should be removed). It makes no difference for binary compatibility, and it's very unlikely that an app would depend on it at the source level, so it's not worth the extra headaches. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: wine/dlls/gdi gdiobj.c
by Krzysztof Foltman March 11, 2005

March 11, 2005

Alexandre Julliard wrote: > Log message: > Allocate fonts and regions from the large heap too. While this is surely good, I guess it would be nice to have an option to restore the previous behaviour. It's much easier to find font leaks this way. Or is there an official way to determine how many GDI handle get leaked (if any) and how many of each type ? (WINEDEBUG=lotsaoptions wine app.exe | egrep complicatedgrepexpression doesn't count) Krzysztof

2 1

AppDB: Escaping / new-line problems
by Francois Gouget March 11, 2005

March 11, 2005

A lot of AppDB entries have broken escaping and some even seem to have only one line. That greatly reduces their readability. Not sure if this is because the formatting was lost over the years. Hopefully it can be fixed in one go for all entries. Otherwise we'll have a big AppDB janitorial cleanup task... Here are some examples: * Gepasi http://appdb.winehq.org/appview.php?appId=1509 Lots of backslashes, quotes are escaped. Appears to be all on one line. * StreetChaves http://appdb.winehq.org/appview.php?appId=1852 Quotes and single-quotes are escaped. All on one line * Neverwinter Nights http://appdb.winehq.org/appview.php?appId=870&versionId=1191 Single-quotes are escaped. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Linux: It is now safe to turn on your computer.

2 2

Re: Warcraft 3 - "no program start menu found"
by Paul Vriens March 11, 2005

March 11, 2005

On Fri, 2005-03-11 at 06:20, Tony Lambregts wrote: > Mike Hearn wrote: > > On Sun, 27 Feb 2005 14:53:51 -0700, Tony Lambregts wrote: > > > >>I had the same error but after some googling (for a good 20 minutes or > >>so) I found the solution: > >> > >>Adding > >> > >>[Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\Shell Folders\\] > >>"Programs"="c:\\windows\\Start Menu\\Programs" > >> > >>to my users.reg file in ~/.wine/ fixed the problem for me. > > > > > > So it's a bug in Wine then. We should add this to the default registry! > > I started a bug on this. > > http://bugs.winehq.org/show_bug.cgi?id=2765 > > I think this is the correct spot to put this. > > Change log: Add start menu entry to the default registry > > Files changed: tools/wine.inf > > ______________________________________________________________________ > Index: tools/wine.inf > =================================================================== > RCS file: /home/wine/wine/tools/wine.inf,v > retrieving revision 1.28 > diff -u -r1.28 wine.inf > --- tools/wine.inf 6 Jan 2005 20:42:10 -0000 1.28 > +++ tools/wine.inf 11 Mar 2005 05:11:40 -0000 > @@ -112,6 +112,7 @@ > HKLM,%CurrentVersion%\Setup,"SharedDir",,"%25%" > HKLM,Software\Microsoft\Windows NT\CurrentVersion\Fonts,,,"" > HKLM,Software\Microsoft\Windows NT\CurrentVersion\Hotfix\Q246009,"Installed",,"1" > +HKLM,Software\Microsoft\Windows\CurrentVersion\Explorer\Shell Folders, "Programs",,"c:\\windows\\Start Menu\\Programs" Should we hard-code "c:\\windows" ? Apart from that, when I look at my WinXP system the entries are in: [HKCU\Software\Microsoft\Windows\CurrentVersion\Explorer\Shell Folders] So there is probably more to it. Cheers, Paul.

1 0

Segfault. Conformance Test/Patch in Cabinet.dll
by Rizwan Kassim March 11, 2005

March 11, 2005

A follow up to the cabinet patch we released before hand. It works perfectly in windows/cl. When its in the wine tree, gcc doesn't like it as much. cabinet_fdi.c: In function `TestCreate': cabinet_fdi.c:263: warning: passing arg 3 of `FDICreate' from incompatible pointer type Running it = segfault. WINEDEBUG=+all wine dlls/cabinet/tests/cabinet_test.exe.so results in the attached debug trace, which doesn't help (me) much. Anyhow, point being, it works in Windows, dies in Wine, but I'm not sure that its a Wine error. Can anyone offer any suggestions/fixes? -- Thank you, Rizwan Kassim http://www.geekymedia.com - GPL code, How-To guides and more. "There's no such thing as non-lethal shaving damage with a sword!" -- Scott Martin Octal is the way God meant programmers to count - the thumbs are parity bits. "The first creature we encountered (and this is the god's honest truth) was a Duck with a +4 hotdog cart. For some reason, we never found out why, this Duck's sole purpose in life was to go out around smasing his cart into adventurers until they agreed to buy a hotdog from him. Once doing so, you were prompty told to roll a save vs poison (food poisoning) or die."

2 2

Reporting "Unknown EXE OS version" and a non-functioning program
by Pavel Troller March 11, 2005

March 11, 2005

Hi! Because my kids are obsessed by trains and everything about them, I've decided to download them a train simulation game. Because I refuse to use anything from M$ and I didn't find any train simulator for Linux, I've found and downloaded a free "miniature train simulator", available for download at http://jeanmichel.kerdal.free.fr/Train/train.zip . I've unzipped the package and tried to run the Install.exe file. But the only I get is patrol(a)arcus:~/train$ wine Install.exe fixme:ver:VERSION_GetLinkedDllVersion Unknown EXE OS version 4.0, please report !! fixme:ole:CoRegisterMessageFilter stub fixme:ole:OleLoadPictureEx (0x7f9bdd4c,326,0,{7bf80980-bf32-101a-8bbb-00aa00300cab},x=0,y=0,f=0,0x720dfb0c), partially implemented. fixme:ole:OLEPictureImpl_SaveAsFile (0x767bd768)->(0x767bffd0, 0, (nil)), hacked stub. The program then simply runs and waits fot the rain (or maybe train :-) ?) without any visible activity and should be killed. Normally I wouldn't report this because there is still a lot of programs which are incompatible with wine (oh these windows programmers... :-) ). But this one explicitly asked for reporting, so I'm just doing what it asked for :-). This is with a fresh wine CVS update, which reports itself as wine-20050310. With regards, Pavel Troller

1 0

winebrowser
by Steven Edwards March 11, 2005

March 11, 2005

Hi, This is a port of the ReactOS Internet Browser to Wine. I will not have time to mess with it so I figured I would send it here if anyone was interested. Its just a small wrapper around the IE control so if you don't have IE installed then it will use Wines shdocvw.dll and the Mozilla ActiveX Control. My plan was to convert it to C and then have it be the fallback case in the event you did not have a Unix browser listed in your registry. Thanks Steven __________________________________ Do you Yahoo!? Yahoo! Sports - Sign up for Fantasy Baseball. http://baseball.fantasysports.yahoo.com/

1 0