January 2003 - Wine-Devel - list.winehq.org

Re: Packagers Guide Update #5
by Andreas Mohr Jan. 25, 2003

Jan. 25, 2003

On Thu, Jan 23, 2003 at 11:42:31PM -0500, Tom Wickline wrote: > Changelog > > Update the Sample config IMHO this doesn't even have *any* business in the Packaging Guide. The guide should simply contain a pointer to the REAL sample config file, otherwise this exactly leads to the permanent updating mess which you just demonstrated ;-) -- The Declaration of Software Freedom: http://freedevelopers.net/freedomdec/index.php

4 3

Re: PATCH: glibc 2.3.x and errno
by Alexandre Julliard Jan. 25, 2003

Jan. 25, 2003

Marcus Meissner <meissner(a)suse.de> writes: > The afore mentioned glibc guru did not have any ideas either. > > There however is a happy but messy end, which appeared to me > yesterday: > Overwrite those 2 functions with a jump to our implementations. I'm afraid that won't be enough. When using thread-local storage, glibc doesn't even call __errno_location any more, it directly stores errno into the thread storage using %gs. It seems the only solution is to make Wine threads work on top of libc threads, but that will be messy. -- Alexandre Julliard julliard(a)winehq.com

4 3

Re: PATCH: glibc 2.3.x and errno
by Andreas Mohr Jan. 24, 2003

Jan. 24, 2003

Hi ! On Fri, Jan 24, 2003 at 05:20:12PM +0100, Marcus Meissner wrote: > Changelog: > Since glibc 2.3.cvs no longer exports __errno_location, we need > to overwrite the implementation in glibc/libpthread directly. Excuse me, but somehow I think this is p*ss poor. (and yes, I'm now marking you up for the dark side ;-) I mean, both Wine and glibc are successful (?) OSS projects, so they should be able to come up with something much better than this terribly embarassing solution (after all everybody knew that OSS development was a "superiour" approach, didn't they ? ;-). I for one would feel much better if we simply rejected that particular "broken" glibc version and supported a *new* glibc method of properly interfacing errno things in a newer glibc version... (maybe have some "advanced" setting in glibc that enables all sorts of funky interfacing capabilities in case a program needs it) After all if Wine needs this errno support, then there's probably a need for it, so it's glibc's bloody obligation to make sure there's proper support IMHO. BTW: why did they even choose to abandon a public errno_location ? Anyway, thanks for tackling this severe issue ! Andreas Mohr P.S.: Do you really think that Alexandre would commit it like that ? ;-)

2 1

dlls/d3d8/device.c: statement with no effect
by Gerald Pfeifer Jan. 24, 2003

Jan. 24, 2003

In dlls/d3d8/device.c I am getting the following warning (on FreeBSD): /usr/bin/gcc -c -I. -I. -I../../include -I../../include -I/usr/X11R6/include -g -O2 -Wall -Wpointer-arith -mpreferred-stack-boundary=2 -gstabs+ -fPIC -D__WINESRC__ -D_REENTRANT -o device.o device.c device.c: In function `DrawPrimitiveI': device.c:475: warning: statement with no effect Could someone please have a look? (If fixed about half a dozen portability problems last week, but this one I'm having a hard time with.) Gerald -- Gerald "Jerry" pfeifer(a)dbai.tuwien.ac.at http://www.pfeifer.com/gerald/

2 1

Re: Patch for crash caused by XIM and make XIM functions properly
by Dmitry Timoshkov Jan. 24, 2003

Jan. 24, 2003

"liu spider" <liuspider(a)yahoo.com> wrote: > @@ -1000,8 +1100,11 @@ > > wine_tsx11_lock(); > if (xic) > - ascii_chars = XmbLookupString(xic, event, Str, sizeof(Str), &keysym, NULL); > - else > + { > + if(XIM_KeyEvent(GetFocus(), xic, event)) > + return; > + } > + > ascii_chars = XLookupString(event, Str, sizeof(Str), &keysym, NULL); > wine_tsx11_unlock(); I can't comment on other parts, but in the above chunk the return after a successful XIM_KeyEvent() leaves X11 lock held. -- Dmitry.

1 0

Packaging Guide Update #6
by Tom Wickline Jan. 24, 2003

Jan. 24, 2003

Francois pointed out that I should fix <wineconf> as sgml will pick it up as a tag. Also replaced the <para> tag with <programlisting> to prevent the text from bunching together. Changelog Fix <wineconf> tag replace <para> tag with <programlisting> Tom Index: documentation/packaging.sgml =================================================================== RCS file: /opt/cvs-commit/wine/documentation/packaging.sgml,v retrieving revision 1.12 diff -u -r1.12 packaging.sgml --- documentation/packaging.sgml 24 Jan 2003 00:49:05 -0000 1.12 +++ documentation/packaging.sgml 24 Jan 2003 08:47:40 -0000 @@ -1442,7 +1442,9 @@ This is the sample config file provided with wine 20030115. You can edit this file to best suit the needs of your intended user. </para> - <para> + + <programlisting> + WINE REGISTRY Version 2 ;; All keys relative to \\Machine\\Software\\Wine\\Wine\\Config @@ -1512,6 +1514,8 @@ ;"ShowDotFiles" = "1" "ShellLinker" = "wineshelllink" +# <wineconf> + [Version] ; Windows version to imitate (win95,win98,winme,nt351,nt40,win2k,winxp,win20,win30,win31) ;"Windows" = "win98" @@ -1720,6 +1724,20 @@ ;"UseDnsComputerName" = "N" ;; sample AppDefaults entries + +; 3 InstallShield versions who like to put their full screen window in front, +; without any chance to switch to another X11 application. +; So just catch them in a desktop window. + +[AppDefaults\\_INS5576._MP\\x11drv] +"Desktop" = "640x480" + +[AppDefaults\\_INS5176._MP\\x11drv] +"Desktop" = "640x480" + +[AppDefaults\\_INS0466._MP\\x11drv] +"Desktop" = "640x480" + ;[AppDefaults\\iexplore.exe\\DllOverrides] ;"shlwapi" = "native" ;"rpcrt4" = "native" @@ -1742,7 +1760,9 @@ ;[AppDefaults\\pickygame.exe\\dsound] ;"EmulDriver" = "N" - </para> +# <wineconf> + + </programlisting> </sect1> </chapter>

1 0

Re: Packagers Guide Update #5
by Francois Gouget Jan. 24, 2003

Jan. 24, 2003

Hi Tom, On Thu, 23 Jan 2003, Tom Wickline wrote: [...] > Index: documentation/packaging.sgml > =================================================================== > RCS file: /opt/cvs-commit/wine/documentation/packaging.sgml,v > retrieving revision 1.12 > diff -u -r1.12 packaging.sgml > --- documentation/packaging.sgml 24 Jan 2003 00:49:05 -0000 1.12 > +++ documentation/packaging.sgml 24 Jan 2003 04:24:53 -0000 > @@ -1512,6 +1512,8 @@ > ;"ShowDotFiles" = "1" > "ShellLinker" = "wineshelllink" > > +# <wineconf> Hmmm, anything of the form <xxx> is a tag sgml so the above will not work very well. You should use: # <wineconf> I noticed another problem. This section is in a simple <para> tag. That means that linefeeds are not significant and everything can be bunched together. I think what's needed is to replace the <para> tags with <screen> tags. With these two changes it should work great. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Dieu dit: "M-x Lumière". Et la lumière fut.

2 1

Whither fnt2bdf?
by Dan Kegel Jan. 24, 2003

Jan. 24, 2003

What happened to fnt2bdf? It was removed as of http://cvs.winehq.com/cvsweb/wine/tools/Makefile.in.diff?r1=1.20&r2=1.21 but the comment for 1.21 doesn't seem to mention this removal http://cvs.winehq.com/cvsweb/wine/tools/Makefile.in Might this be a mistake? - Dan -- Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045

4 7

Doccumetation issue: "The Service Thread" Does it exist?
by Robert North Jan. 24, 2003

Jan. 24, 2003

This concerns the documentation on thw www.winehq.com website, if it's out of date, then this post may be irrelevant. In the "Wine Developer's Guide" In Section 8.1.4, "The Service thread" Says there exists a service thread, that is implemented in: scheduler/services.c. It also says this thread is used to support the X11 driver's event loop. Firstly, I cannot find the "scheduler/services.c" file. Secondly from my reading of the x11drv code, the X11 event loop is serviced for every call of to X11DRV_MsgWaitForMultipleObjectsEx, which is inturn called by MsgWaitForMultipleObjectsEx, and GetMessage. Therefore, whichever thread is active will service the X11 event queue. It also appears, because each thread has it's own XDisplay object that messages created by a X11 window will only be picked up in the thread that created the window. Am I correct, or have I missed something fundamental? Bye -Rob.

3 3

RtlpWaitForCriticalSection?
by Mike Hearn Jan. 23, 2003

Jan. 23, 2003

Using the latest Wine release, I see this error repeated over and over: err:ntdll:RtlpWaitForCriticalSection section 0x70460350 "?" wait timed out, retrying (60 sec) tid=08436988 when clicking on the CML demo at Adobes SVG demo site. Google indicates that this error can be caused by all sorts of things, one guy said it was due to Wine being a lot slower than Windows. So, could anybody write an explanation of what this error means? thanks -mike -- Mike Hearn <m.hearn(a)signal.qinetiq.com> QinetiQ - Malvern Technology Center

5 14

Windows API database
by David Miller Jan. 23, 2003

Jan. 23, 2003

I would be interested in knowing more about the goals of this project. Don't consider me officially working on such a thing just yet. I have no real perl experience, but would like to learn. First I'd like to know more about what I'm up against. 1. Is there a specific set of dll's and exe's we want included in the database or does it need to scan and gather a list of all dll and exe files? That would suggest a clean OS install is necessary. 2. What information should be in the database? (OS ver, name of dll or exe, name of imported dll, APIs and whether they are imported or exported, etc.) 3. Will a single MySQL table do or does it need to be a relational database with multiple tables? 4. Anything else you can tell me about this task?

3 3

Re: dos4gw stub exec failed
by Jukka Heinonen Jan. 23, 2003

Jan. 23, 2003

Enrico Horn wrote: > I have some dos games that want to exec dos4gw.exe when starting. > but this seems to fail. > I hope i read the debug messages correctly and the comment from > file:Win32HandleToDosFileHandle is still valid. > what happens is dos:int21 wants to open file with the dos handle and > then read from it. > reading the file calls the function file:ReadFile with the > converted win32 file handle which is invalid if the above mentioned > comment is correct. > > Any ideas how to solve this or am I completely off-track here. Well, I don't actually know what is wrong with dos4gw. It is probably something subtle with int21 functions. However, why it hangs is something I haven't checked out. You could try running those programs using DOS32A (DOS/32 Advanced) extender. It seems to work quite well. I just posted a patch that is required to get its installer working, though, so you might have some trouble unpacking it... If you get those games to work under dos32a you will probably find out that Wine SVGA support is lacking. That is something I have considered to fix if I can find the time to do it. -- Jukka Heinonen <http://www.iki.fi/jhei/>

3 2

Sacrifice demo & sound.dll
by Christian Costa Jan. 23, 2003

Jan. 23, 2003

Hi all! I've tried the Sacrifice demo and it complains about sound.dll. Indeed it uses its own 32-bit sound.dll which lies in its directory. Apparently the dll seems to conflict with our 16-bit windows sound.dll I've joined a small log. I tried to add a "sound" entry in the [DllOverrides] section in the config file to force the native to be loaded first but that does not help. Any idea ? Thanks in advance. Christian. trace:module:GetModuleFileNameA F:\Sacrifice\Sacrifice.exe trace:module:MODULE_GetLoadOrder looking for F:\Sacrifice\sound.dll trace:module:GetModuleFileNameA F:\Sacrifice\Sacrifice.exe trace:module:open_app_key searching 'F:\Sacrifice\sound' in AppDefaults\Sacrifice.exe\DllOverrides trace:module:MODULE_GetLoadOrder got hardcoded default "b,n" for "F:\\Sacrifice\\sound.dll" trace:module:MODULE_LoadLibraryExA Trying built-in 'F:\Sacrifice\sound.dll' err:module:BUILTIN32_LoadLibraryExA loaded .so but dll sound.dll still not found - 16-bit dll or version conflict. warn:module:MODULE_LoadLibraryExA Loading of builtin DLL F:\Sacrifice\sound.dll failed (error 193). warn:module:MODULE_LoadLibraryExA Failed to load module 'F:\Sacrifice\sound.dll'; error=193

2 1

debugger detection in newbin.
by Rein Klazes Jan. 23, 2003

Jan. 23, 2003

Hi, The latest version of newsbin 4.1B5 refuses to run, displaying "debugger or monitoring tool detected". The detection code is very simple, immedeately at the program entry point 0x516000 it does (intel syntax): | Disassembly of 0x00516000 | 0x51600D: 64A023000000 mov al,fs:[0x23] | 0x516013: EB03 jmp 0x516018 | ;*************************************************** | 0x516018: 84C0 test al,al | 0x51601A: EB03 jmp 0x51601f | ;*************************************************** | 0x51601F: 7567 jnz 0x516088 This jump is taken and leads immedeatly to the messagebox displaying the message above. Any idea's and/or explanation? Rein. -- Rein Klazes rklazes(a)xs4all.nl

3 3

Re: Wine kernel acceleration module?
by Brian Vincent (C) Jan. 23, 2003

Jan. 23, 2003

This issue has been discussed several times. If you dig in the WWN archives you'll find mentions of it going back a few years. Much more interesting was this post: http://www.winehq.com/hypermail/wine-devel/2002/12/0550.html The approach is a shared memory wineserver. --------------- Brian Vincent Copper Mountain Telecom vincentb(a)coppercolorado.com

8 15

Re: more HEAP_strdupAtoW removals
by Rolf Kalbermatter Jan. 23, 2003

Jan. 23, 2003

Matthew Davison <m.davison(a)virgin.net> wrote: >On Thu, 2003-01-23 at 01:36, Dmitry Timoshkov wrote: >> > @@ -760,24 +774,26 @@ >> > BOOL WINAPI OpenPrinterA(LPSTR lpPrinterName,HANDLE *phPrinter, >> > LPPRINTER_DEFAULTSA pDefault) >> > { >> > - LPWSTR lpPrinterNameW = HEAP_strdupAtoW(GetProcessHeap(),0,lpPrinterName); >> > + UNICODE_STRING lpPrinterNameW; >> > + UNICODE_STRING usBuffer; >> > + RtlCreateUnicodeStringFromAsciiz(&lpPrinterNameW,lpPrinterName); >> > PRINTER_DEFAULTSW DefaultW, *pDefaultW = NULL; >> > BOOL ret; > >> Are you sure that mixed data declarations/function call will be accepted >> by any C compiler out there? I doubt that even gcc will accept this. >well, im pretty sure gcc takes this, but ill change it anyway. >thanks. And I can tell you that Visual C is happy to compile this if the file has a .cpp ending but will bark for .c files ;-(. Rolf Kalbermatter

1 0

LockFile support
by David D. Hagood Jan. 23, 2003

Jan. 23, 2003

I have an old, nasty application that uses the Paradox database engine. This app attempts to lock the database files using LockFile. Now, it USED to work under Wine. It no longer does. The program coughs up an error "No access to directory", and stderr shows a bunch of "LockFile not supported in server" errors. Did the behavior of LockFile change in the past couple of months? Are there plans to add LockFile support into Wine?

2 1

Re: more HEAP_strdupAtoW removals.
by Dmitry Timoshkov Jan. 23, 2003

Jan. 23, 2003

"Matthew Davison" <m.davison(a)virgin.net> wrote: > Changelog > Remove some more calls to HEAP_strdupWtoA First of all, thanks for doing this! > @@ -554,7 +555,11 @@ > lpwstr = NULL; > for(i=0; i<dwNumServiceArgs; i++) > - lpwstr[i]=HEAP_strdupAtoW(GetProcessHeap(), 0, lpServiceArgVectors[i]); > + { > + RtlCreateUnicodeStringFromAsciiz (&usBuffer,lpServiceArgVectors[i]); > + lpwstr[i]=usBuffer.Buffer; > + RtlFreeUnicodeString(&usBuffer); > + } Of course this is a typo and last RtlFreeUnicodeString should be removed. > @@ -760,24 +774,26 @@ > BOOL WINAPI OpenPrinterA(LPSTR lpPrinterName,HANDLE *phPrinter, > LPPRINTER_DEFAULTSA pDefault) > { > - LPWSTR lpPrinterNameW = HEAP_strdupAtoW(GetProcessHeap(),0,lpPrinterName); > + UNICODE_STRING lpPrinterNameW; > + UNICODE_STRING usBuffer; > + RtlCreateUnicodeStringFromAsciiz(&lpPrinterNameW,lpPrinterName); > PRINTER_DEFAULTSW DefaultW, *pDefaultW = NULL; > BOOL ret; Are you sure that mixed data declarations/function call will be accepted by any C compiler out there? I doubt that even gcc will accept this. -- Dmitry.

2 1

Installshield 6 crash: ole trouble
by Dan Kegel Jan. 23, 2003

Jan. 23, 2003

So, how's that Installshield 6 support coming along? :-) The company I work for uses Installshield to package up its windows stuff. Today I tried running its installer under cvs Wine, and shortly after it started running IKernel.exe, it dropped me into the wine debugger. Copying and pasting from the wine debugger window is beyond me; it seems totally broken. However, it looks like it was complaining of a null pointer read. Here are the last few interesting lines of the wine log: trace:ole:ITypeInfo_fnGetTypeAttr (0x40298e58) trace:ole:ITypeInfo_fnGetRefTypeOfImplType (0x40298e58) index 0 trace:ole:dump_TypeInfo 0x40298e58 ref=2 trace:ole:dump_TypeInfo attr:{91814ec1-b5f0-11d2-80b9-00104b1f6cea} trace:ole:dump_TypeInfo kind:TKIND_INTERFACE trace:ole:dump_TypeInfo fct:13 var:0 impl:1 trace:ole:dump_TypeInfo parent tlb:0x4026c9c8 index in TLB:28 trace:ole:dump_TypeInfo L"ISetupCABFile" (null) trace:ole:ITypeInfo_fnGetRefTypeOfImplType -- 0x00000000 trace:ole:ITypeInfo_fnGetRefTypeInfo typeinfo in imported typelib that isn't already loaded 080761c8:Call ntdll.RtlUnicodeToMultiByteSize(40791ac8,40791ae4 L"stdole32.tlb",0000001a) ret=409e78c8 080761c8:Call kernel32.MultiByteToWideChar(00000000,00000001,40791cf8 "stdole32.tlb",0000000d,00000000,00000000) ret=40b9e285 080761c8:Call kernel32.MultiByteToWideChar(00000000,00000001,40791cf8 "stdole32.tlb",0000000d,4028bfdc,0000000d) ret=40b9e2a8 trace:ole:LoadTypeLib trace:ole:LoadTypeLibEx (L"stdole32.tlb",0,0x40791f2c) 080761c8:Call kernel32.SearchPathW(00000000,4028bfdc L"stdole32.tlb",00000000,00000105,40791cac,00000000) ret=40b9e41a trace:ole:LoadRegTypeLib (IID: {00020430-0000-0000-c000-000000000046}) load FAILED ((nil)) trace:ole:LoadTypeLib trace:ole:LoadTypeLibEx (L"C:\\WINNT\\System32\\StdOle32.tlb",0,0x40791f2c) trace:ole:ITypeInfo_fnGetRefTypeInfo (0x40298e58) hreftype 0x0000 loaded FAILURE (0x13) fixme:ole:_get_funcdesc Did not find a typeinfo for reftype 0? fixme:ole:PSFacBuf_CreateProxy GetFuncDesc 80004005 should not fail here. fixme:ole:StdMarshalImpl_UnmarshalInterface Failed to create a proxy for {91814ec1-b5f0-11d2-80b9-00104b1f6cea} fixme:ole:CoUnmarshalInterface Failed to Unmarshal the interface, 80004005? 080761c8:Ret ole32.CoUnmarshalInterface() retval=80004005 ret=40b9a2fd fixme:ole:_unmarshal_interface Marshaling interface {91814ec1-b5f0-11d2-80b9-00104b1f6cea} failed with 80004005 fixme:ole:deserialize_param failed to stuballoc in TKIND_RECORD. trace:ole:ITypeInfo_fnRelease (0x40279798)->(1) fixme:ole:xCall Failed to unmarshall param, hres 80004005 wine: Unhandled exception, starting debugger... Any suggestions? I am a babe in the woods when it comes to the wine debugger, or for that matter, ole. - Dan -- Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045

10 19

Re: PATCH: Adds my name to the AUTHORS list
by Shachar Shemesh Jan. 23, 2003

Jan. 23, 2003

Waldeck Schutzer wrote: > Now that my patches were incorporated to the tree, it would make me > feel happier if my name was listed too :-) > > Best, > Waldeck I think it's time Alexander regenerates the entire Authors list. People want the credit, Alexander. There is a script in "tools" that does something, but I think it only alphabetizes the list, it does not actually add names. Shachar

5 4

[Fwd: dead keys treatment using XmbLookupString]
by Mauro Carvalho Chehab Jan. 23, 2003

Jan. 23, 2003

Rikki, Please try to apply the following patch against the new CVS version (on /dlls/x11drv directory). This patch will allow keyboard.c to use your locale setting by the usage of xkb extension on X11R6. Please notice that this requires that you *don't have* xkbdisable on your XF86Config file (or XF86Config-4). -------- Original Message -------- Subject: dead keys treatment using XmbLookupString Date: Wed, 08 Jan 2003 12:48:06 -0200 From: Mauro Carvalho Chehab <mchehab(a)brturbo.com> To: wine-patches(a)winehq.com *Modifyied files*: keyboard.c window.c desktop.c *CVS Version*: 20021225 *Change Log: *Add dead key support using XmbLookupString on ToUnicode function, if *xkb* is available otherwise It falls back to the old code. The first Window will be used to define the XIC used by the functions. * Author*: Mauro Carvalho Chehab <mchehab(a)brturbo.com> ______________________________________________________________________________________________ --- keyboard.c 8 Jan 2003 11:24:51 -0000 1.1 +++ keyboard.c 8 Jan 2003 14:30:53 -0000 @@ -1389,6 +1389,44 @@ wine_tsx11_unlock(); } +#ifdef HAVE_XKB +static XIM xim = (XIM) NULL; +static XIC xic = (XIC) NULL; +#endif +/******************************************************************* + * KeyboardWindow (X11DRV.@) + * Called from desktop.c and window.c after Window creating + * This routine allows usage of XmbLookupString by ToUnicode. + * Only the first opened window will be used. It means that + * all windows will share the same locale. + * + */ +void X11DRV_KeyboardWindow (Display *display, Window w) +{ +#ifdef HAVE_XKB + char *p; + + if (xic) return; /* Only the first window will be used */ + + if ((p = XSetLocaleModifiers("")) != NULL && *p) + xim = XOpenIM(display, NULL, NULL, NULL); + if (!xim && (p = XSetLocaleModifiers("@im=none")) != NULL && *p) + xim = XOpenIM(display, NULL, NULL, NULL); + if (!xim) ERR_(key)( "Failed to open input method.\n"); + else { + xic = XCreateIC(xim, + XNInputStyle, XIMPreeditNothing | XIMStatusNothing, + XNClientWindow, w, + XNFocusWindow, w, + NULL); + if (!xic) { + ERR_(key)( "Failed to create input context.\n"); + XCloseIM(xim); + xim=NULL; + } + } +#endif +} /*********************************************************************** * X11DRV_MappingNotify @@ -1741,6 +1779,7 @@ * FIXME : should do the above (return 2 for non matching deadchar+char combinations) * */ + INT X11DRV_ToUnicode(UINT virtKey, UINT scanCode, LPBYTE lpKeyState, LPWSTR bufW, int bufW_size, UINT flags) { @@ -1756,6 +1795,9 @@ TRACE("Key UP, doing nothing\n" ); return 0; } + memset (&e,0,sizeof(e)); + + e.type = KeyPress; e.display = display; e.keycode = 0; e.state = 0; @@ -1814,7 +1856,15 @@ } else TRACE("Found keycode %d (0x%2X)\n",e.keycode,e.keycode); - ret = XLookupString(&e, (LPVOID)lpChar, 2, &keysym, NULL); +#ifdef HAVE_XKB + if (xic) { + ret = XmbLookupString(xic,&e, (LPVOID)lpChar, sizeof(lpChar), &keysym, NULL); + } else { + ret = XLookupString(&e, (LPVOID)lpChar, sizeof(lpChar), &keysym, NULL); + } +#else + ret = XLookupString(&e, (LPVOID)lpChar, sizeof(lpChar), &keysym, NULL); +#endif wine_tsx11_unlock(); if (ret == 0) --- window.c 8 Jan 2003 11:24:51 -0000 1.1 +++ window.c 8 Jan 2003 14:30:53 -0000 @@ -43,6 +43,7 @@ WINE_DEFAULT_DEBUG_CHANNEL(x11drv); extern Pixmap X11DRV_BITMAP_Pixmap( HBITMAP ); +extern X11DRV_KeyboardWindow (Display *display, Window w); #define HAS_DLGFRAME(style,exStyle) \ (((exStyle) & WS_EX_DLGMODALFRAME) || \ @@ -199,6 +200,7 @@ CWEventMask | CWBitGravity | CWBackingStore | CWColormap, &attr ); XSaveContext( display, data->icon_window, winContext, (char *)win->hwndSelf ); wine_tsx11_unlock(); + if (data->client_window) X11DRV_KeyboardWindow (display, data->whole_window); TRACE( "created %lx\n", data->icon_window ); SetPropA( win->hwndSelf, icon_window_atom, (HANDLE)data->icon_window ); @@ -686,6 +688,7 @@ wine_tsx11_unlock(); return 0; } + X11DRV_KeyboardWindow (display, data->whole_window); /* non-maximized child must be at bottom of Z order */ if ((win->dwStyle & (WS_CHILD|WS_MAXIMIZE)) == WS_CHILD) @@ -732,6 +735,7 @@ CWEventMask | CWBitGravity | CWBackingStore, &attr ); if (data->client_window && is_client_window_mapped( win )) XMapWindow( display, data->client_window ); + if (data->client_window) X11DRV_KeyboardWindow (display, data->whole_window); wine_tsx11_unlock(); return data->client_window; } --- desktop.c 8 Jan 2003 11:24:51 -0000 1.1 +++ desktop.c 8 Jan 2003 14:30:53 -0000 @@ -30,6 +30,7 @@ WINE_DEFAULT_DEBUG_CHANNEL(x11drv); +extern X11DRV_KeyboardWindow (Display *display, Window w); /* desktop window procedure */ static LRESULT WINAPI desktop_winproc( HWND hwnd, UINT message, WPARAM wParam, LPARAM lParam ) @@ -141,6 +142,8 @@ win = XCreateWindow( display, DefaultRootWindow(display), x, y, width, height, 0, screen_depth, InputOutput, visual, CWBackPixel | CWEventMask | CWCursor | CWColormap, &win_attr ); + + if (win) X11DRV_KeyboardWindow (display, win); /* Set window manager properties */ size_hints = XAllocSizeHints();

2 5

dos4gw stub exec failed
by Enrico Horn Jan. 23, 2003

Jan. 23, 2003

Hi I have some dos games that want to exec dos4gw.exe when starting. but this seems to fail. I hope i read the debug messages correctly and the comment from file:Win32HandleToDosFileHandle is still valid. what happens is dos:int21 wants to open file with the dos handle and then read from it. reading the file calls the function file:ReadFile with the converted win32 file handle which is invalid if the above mentioned comment is correct. Any ideas how to solve this or am I completely off-track here. Ciao Enrico farmboy1(a)subdimension.com log extract: Call DOS int 0x21 ret=00b7:15d0 eax=00003d00 ebx=000009f0 ecx=0000106a edx=00000a76 esi=00000a76 edi=0000ffff ebp=000009ce esp=000009ca ds=0319 es=0319 fs=0000 gs=0000 flags=000b0246 trace:int21:DOSVM_Int21Handler AX=3d00 BX=09f0 CX=106a DX=0a76 SI=0a76 DI=ffff DS=0319 ES=0319 EFL=000b0246 080eef28:Call kernel32.INT_Int21Handler(416d2c54) ret=4149829e trace:int21:INT_Int21Handler OPEN mode 0x00 C:\CCS\dos4gw.exe trace:file:_lopen ('C:\CCS\dos4gw.exe',0000) trace:file:CreateFileW L"C:\\CCS\\dos4gw.exe" GENERIC_READ FILE_SHARE_READ FILE_SHARE_WRITE OPEN_EXISTING attributes 0x0 trace:dosfs:DOSFS_GetFullName L"C:\\CCS\\dos4gw.exe" (last=1) trace:dosfs:DOSFS_FindUnixName /home/windows,L"CCS\\dos4gw.exe" trace:dosfs:DOSFS_ToDosFCBFormat (L"CCS\\dos4gw.exe", 0x416d19bc) trace:dosfs:DOSFS_OpenDir "/home/windows" trace:dosfs:DOSFS_ReadDir Read: long_name: L".", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"..", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"windows", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"Program Files", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"Temp", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"Angel", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"Buffy", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SFV32w.GID", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SFV32w.hlp", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"dcm95cfg.exe", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"dcom98.exe", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"quicksfv.exe", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"quicksfv.ini", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"sfv32w.exe", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"Spiele", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"wingtk-0.2.exe", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"dxdiag.exe", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"CCS", short_name: (null) trace:dosfs:DOSFS_FindUnixName (/home/windows,L"CCS\\dos4gw.exe") -> L"CCS" (L"CCS") trace:dosfs:DOSFS_FindUnixName /home/windows/CCS,L"dos4gw.exe" trace:dosfs:DOSFS_ToDosFCBFormat (L"dos4gw.exe", 0x416d19bc) trace:dosfs:DOSFS_OpenDir "/home/windows/CCS" trace:dosfs:DOSFS_ReadDir Read: long_name: L".", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"..", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SAMPLE.AD", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"OBJECT.DAT", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"CCSSTR.DAT", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"ADRV688.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"JAMMER.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"PROAUDIO.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"RAP10.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SB16.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SBLASTER.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SBPRO.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SNDSCAPE.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"SNDSYS.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"ULTRA.DIG", short_name: (null) trace:dosfs:DOSFS_ReadDir Read: long_name: L"DOS4GW.EXE", short_name: (null) trace:dosfs:DOSFS_FindUnixName (/home/windows/CCS,L"dos4gw.exe") -> L"DOS4GW.EXE" (L"DOS4GW.EXE") trace:dosfs:DOSFS_GetFullName returning /home/windows/CCS/DOS4GW.EXE = L"C:\\CCS\\DOS4GW.EXE" trace:dosfs:GetDriveTypeW (L"C:\\CCS\\DOS4GW.EXE") trace:file:CreateFileW returning 0x68 trace:file:Win32HandleToDosFileHandle Got 5 for h32 0x68 080eef28:Ret kernel32.INT_Int21Handler() retval=00000005 ret=4149829e trace:int21:DOSVM_Int21Handler returning: AX=0005 BX=09f0 CX=106a DX=0a76 SI=0a76 DI=ffff DS=0319 ES=0319 EFL=000b0246 Ret DOS int 0x21 ret=00b7:15d0 eax=00000005 ebx=000009f0 ecx=0000106a edx=00000a76 esi=00000a76 edi=0000ffff ebp=000009ce esp=000009ca ds=0319 es=0319 fs=0000 gs=0000 flags=000b0246 Call DOS int 0x21 ret=00b7:1f9f eax=00004400 ebx=00000005 ecx=00000005 edx=00000005 esi=0000038a edi=00000005 ebp=000009ce esp=000009ba ds=0319 es=0319 fs=0000 gs=0000 flags=000b0202 trace:int21:DOSVM_Int21Handler AX=4400 BX=0005 CX=0005 DX=0005 SI=038a DI=0005 DS=0319 ES=0319 EFL=000b0202 080eef28:Call kernel32.DosFileHandleToWin32Handle(00000005) ret=414965a0 080eef28:Ret kernel32.DosFileHandleToWin32Handle() retval=00000068 ret=414965a0 080eef28:Call kernel32.DOSFS_GetDeviceByHandle(00000068) ret=414965a6 080eef28:Ret kernel32.DOSFS_GetDeviceByHandle() retval=00000000 ret=414965a6 080eef28:Call kernel32.INT_Int21Handler(416d2c54) ret=41496676 trace:int21:ioctlGetDeviceInfo (5) 080eef28:Ret kernel32.INT_Int21Handler() retval=00000942 ret=41496676 trace:int21:DOSVM_Int21Handler returning: AX=4400 BX=0005 CX=0005 DX=0942 SI=038a DI=0005 DS=0319 ES=0319 EFL=000b0202 Ret DOS int 0x21 ret=00b7:1f9f eax=00004400 ebx=00000005 ecx=00000005 edx=00000942 esi=0000038a edi=00000005 ebp=000009ce esp=000009ba ds=0319 es=0319 fs=0000 gs=0000 flags=000b0202 Call DOS int 0x21 ret=00b7:1f9f eax=00004400 ebx=00000005 ecx=00000005 edx=00004000 esi=00000a76 edi=00000005 ebp=000009ce esp=000009c2 ds=0319 es=0319 fs=0000 gs=0000 flags=000b0246 trace:int21:DOSVM_Int21Handler AX=4400 BX=0005 CX=0005 DX=4000 SI=0a76 DI=0005 DS=0319 ES=0319 EFL=000b0246 080eef28:Call kernel32.DosFileHandleToWin32Handle(00000005) ret=414965a0 080eef28:Ret kernel32.DosFileHandleToWin32Handle() retval=00000068 ret=414965a0 080eef28:Call kernel32.DOSFS_GetDeviceByHandle(00000068) ret=414965a6 080eef28:Ret kernel32.DOSFS_GetDeviceByHandle() retval=00000000 ret=414965a6 080eef28:Call kernel32.INT_Int21Handler(416d2c54) ret=41496676 trace:int21:ioctlGetDeviceInfo (5) 080eef28:Ret kernel32.INT_Int21Handler() retval=00000942 ret=41496676 trace:int21:DOSVM_Int21Handler returning: AX=4400 BX=0005 CX=0005 DX=0942 SI=0a76 DI=0005 DS=0319 ES=0319 EFL=000b0246 Ret DOS int 0x21 ret=00b7:1f9f eax=00004400 ebx=00000005 ecx=00000005 edx=00000942 esi=00000a76 edi=00000005 ebp=000009ce esp=000009c2 ds=0319 es=0319 fs=0000 gs=0000 flags=000b0246 Call DOS int 0x21 ret=00b7:17a7 eax=00003f41 ebx=00000005 ecx=0000001c edx=00000ac6 esi=00000005 edi=0000041e ebp=000009e6 esp=000009d8 ds=0319 es=0319 fs=0000 gs=0000 flags=000b0202 trace:int21:DOSVM_Int21Handler AX=3f41 BX=0005 CX=001c DX=0ac6 SI=0005 DI=041e DS=0319 ES=0319 EFL=000b0202 080eef28:Call kernel32.INT_Int21Handler(416d2c54) ret=4149829e trace:int21:INT_Int21Handler READ from 5 to 0319:0AC6 for 28 byte trace:file:ReadFile 0x68 0x3c56 28 0x416d26dc (nil) 080eef28:Ret kernel32.INT_Int21Handler() retval=0000001c ret=4149829e trace:int21:DOSVM_Int21Handler returning: AX=001c BX=0005 CX=001c DX=0ac6 SI=0005 DI=041e DS=0319 ES=0319 EFL=000b0202 Ret DOS int 0x21 ret=00b7:17a7 eax=0000001c ebx=00000005 ecx=0000001c edx=00000ac6 esi=00000005 edi=0000041e ebp=000009e6 esp=000009d8 ds=0319 es=0319 fs=0000 gs=0000 flags=000b0202 Call DOS int 0x21 ret=00b7:1849 eax=00003e05 ebx=00000005 ecx=00000005 edx=00000f20 esi=00000005 edi=00000f0c ebp=00000a6c esp=000009e8 ds=0319 es=0319 fs=0000 gs=0000 flags=000b0206 trace:int21:DOSVM_Int21Handler AX=3e05 BX=0005 CX=0005 DX=0f20 SI=0005 DI=0f0c DS=0319 ES=0319 EFL=000b0206 080eef28:Call kernel32.INT_Int21Handler(416d2c54) ret=4149829e trace:int21:INT_Int21Handler CLOSE handle 5 trace:file:_lclose16 5 (handle32=0x68) 080eef28:Ret kernel32.INT_Int21Handler() retval=00000000 ret=4149829e trace:int21:DOSVM_Int21Handler returning: AX=0000 BX=0005 CX=0005 DX=0f20 SI=0005 DI=0f0c DS=0319 ES=0319 EFL=000b0206 Ret DOS int 0x21 ret=00b7:1849 eax=00000000 ebx=00000005 ecx=00000005 edx=00000f20 esi=00000005 edi=00000f0c ebp=00000a6c esp=000009e8 ds=0319 es=0319 fs=0000 gs=0000 flags=000b0206

3 3

stupidity: a slighlty less horrible wine tree diff gizmo
by Greg Turner Jan. 23, 2003

Jan. 23, 2003

Still up to the same old stupid bash tricks... but implemented slightly better. This will still have to be manually maintained to remain useful as wine's directory structure evolves, just like the old "version" of this awfulness. Use at your own (very real) risk as these are almost untested. License is bugroff. This version has better provisions for in-source compilation, runs in only one pass, and is cwd/location independant. It should also be a lot easier to maintain than the previous try winetreediff_excl (a text file listing shell-glob patterns to ignore): >> CVS *.o *.a *.s Makefile *.spec.c *.dll.dbg.c *.exe.dbg.c *.drv.dbg.c *.acm.dbg.c *.ocx.dbg.c *.so *.def *.glue.c *.ok *.mc.rc testlist.c aclocal.m4 *.res gmtdir Makeprog.rules Make.rules *.swp config.log config.status lex.yy.c y.tab.c y.tab.h *.so.?.? wineclipsrv doall doconfig wine.man autom4te.cache mydiffs gmt-cfg-vars Makedll.rules Maketest.rules tags delme.diff delme2.diff wine.texinfo <<EOF winetreediff (a script to filter out impossible-to-glob patterns and perform the real processing. Note the hard-coded path to winetreediff_excl. Change it to match where you put it. Needs the x bit, of course.) >> #!/bin/sh noslash=$( echo "$1" | sed s/\\/$// ) noslashquot=$( echo "$noslash" | sed s/\\//\\\\\\//g | sed s/\\./\\\\\\./g ) # echo "noslash=$noslash noslashquot=$noslashquot" rm -f delme.diff delme2.diff touch delme.diff delme2.diff diff -ur --minimal --exclude-from=/home/greg/bin/winetreediff_excl $noslash . | \ grep -v '^Only in.*wine$' | \ grep -v '^Only in .*dlls/rpcrt4/tests\: stringinput_server$' | \ grep -v '^Only in .*include\: config.h$' | \ grep -v '^Only in .*programs/avitools\: aviinfo$' | \ grep -v '^Only in .*programs/avitools\: aviplay$' | \ grep -v '^Only in .*programs/avitools\: icinfo$' | \ grep -v '^Only in .*programs/clock\: clock$' | \ grep -v '^Only in .*programs/cmdlgtst\: cmdlgtst$' | \ grep -v '^Only in .*programs/control\: control$' | \ grep -v '^Only in .*programs/expand\: expand$' | \ grep -v '^Only in .*programs/notepad\: notepad$' | \ grep -v '^Only in .*programs/osversioncheck\: osversioncheck$' | \ grep -v '^Only in .*programs/progman\: progman$' | \ grep -v '^Only in .*programs/regapi\: regapi$' | \ grep -v '^Only in .*programs/regedit\: regedit$' | \ grep -v '^Only in .*programs/regsvr32\: regsvr32$' | \ grep -v '^Only in .*programs/regtest\: regtest$' | \ grep -v '^Only in .*programs/rpcss\: rpcss$' | \ grep -v '^Only in .*programs/rundll32\: rundll32$' | \ grep -v '^Only in .*programs/uninstaller\: uninstaller$' | \ grep -v '^Only in .*programs/view\: view$' | \ grep -v '^Only in .*programs/wcmd\: wcmd$' | \ grep -v '^Only in .*programs\: wineapploader$' | \ grep -v '^Only in .*programs/wineboot\: wineboot$' | \ grep -v '^Only in .*programs/wineconsole\: wineconsole$' | \ grep -v '^Only in .*programs/winedbg\: winedbg$' | \ grep -v '^Only in .*programs/winefile\: winefile$' | \ grep -v '^Only in .*programs\: winelauncher$' | \ grep -v '^Only in .*programs/winemine\: winemine$' | \ grep -v '^Only in .*programs/winepath\: winepath$' | \ grep -v '^Only in .*programs/winhelp\: hlp2sgml$' | \ grep -v '^Only in .*programs/winhelp\: winhelp$' | \ grep -v '^Only in .*programs/winver\: winver$' | \ grep -v '^Only in .*server\: wineserver$' | \ grep -v '^Only in .*stringinput_server\: stringinput_server$' | \ grep -v '^Only in .*tools\: bin2res$' | \ grep -v '^Only in .*tools\: fnt2bdf$' | \ grep -v '^Only in .*tools\: makedep$' | \ grep -v '^Only in .*tools\: mingwrap$' | \ grep -v '^Only in .*tools/widl\: widl$' | \ grep -v '^Only in .*tools/winebuild\: winebuild$' | \ grep -v '^Only in .*tools/winebuild\: winebuild.man$' | \ grep -v '^Only in .*tools/winedump\: winedump$' | \ grep -v '^Only in .*tools\: winegcc$' | \ grep -v '^Only in .*tools\: winewrap$' | \ grep -v '^Only in .*tools/wmc\: wmc$' | \ grep -v '^Only in .*tools/wrc\: wrc$' | \ ( tee delme.diff | grep -v '^Only in .*\:' ) cat delme.diff | \ grep '^Only in .*\:' | \ sed s/^Only\ in\ // | \ sed s/\:\ /\\// | \ ( tee delme2.diff | grep "^\.\/" ) | \ sed "s/^\.\//diff\ -u\ \/dev\/null\ \.\//" | \ sed "s/$/;/" | \ bash -C - cat delme2.diff | \ grep -v "^\.\/" | \ sed "s/^$noslashquot\//diff\ -u\ \.\//" | \ sed "s/$/\ \/dev\/null;/" | \ ( cd $noslash && bash -C - ) rm -f delme.diff delme2.diff <<EOF still hardly a work of art. but ought to generate diffs between trees ready for submission to wine-devel, no cvs needed. For example, I use this as follows to update my horribly out-of-sync wine tree with the latest changes from cvs: .../wine> cd ../wine.test .../wine.test> winetreediff ../wine.vanilla | wc -l 0 .../wine> # great, these trees are identical w.r.t. winetreediff. .../wine> cd ../wine.vanilla .../wine.vanilla> cvs -qz4 up -dPA [cvs does something] .../wine.vanilla> # generate the difference between "../wine.test" and "." .../wine.vanilla> winetreediff ../wine.test > ../wine/mydiffs/cvsupdate.diff .../wine.vanilla> cd ../wine.test .../wine.test> patch -p0 < ../wine/mydiffs/cvsupdate.diff [ goes fine ] .../wine.test> cd ../wine Now I have a nice diff expressing the difference caused by the CVS update. If I am concerned about it, I could run .../wine> patch -p0 --dry-run < mydiffs/cvsupdate.diff To see what the damage will be, otherwise, just .../wine> patch -p0 < mydiffs/cvsupdate.diff would bring me up-to-date. LIMITATIONS: o Dangerous if used incautiously. o Breaks if wine changes. o Probably Buggy. o Only supports "." and "/" characters in directory argument. o Creates and destroys temp files named delme.diff and delme2.diff. (Should at least use something more obscure.) o Prefers (requires?) a matching directory structure between the two trees. -- gmt

1 0

Re: PATCH: kernel / errno testing
by Dan Kegel Jan. 22, 2003

Jan. 22, 2003

Marcus Meissner wrote: > Hi, > > glibc 2.3.current CVS is throwing a huge stone in our direction, > it no longer allows overload of __errno_location. > > To visualize this problem I have added this testcase. ... > sched_yield(); /* will not change errno */ BTW sched_yield() is allowed to be a no-op; would nanosleep() or something like that be a better choice? - Dan -- Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045

3 2

InstallShield 6: fixme:ole:_StubReaderThread Failed with hres 80004005
by Dan Kegel Jan. 22, 2003

Jan. 22, 2003

Whenever I run an InstallShield 6 installer, and it fails early on for any reason (e.g. a file was missing, or the "object reference not set" bug), the error messagebox "Error installing iKernel.exe: (0x1400)" appears. The user clicks 'OK', and the program terminates (after a few seconds). So far so good. But on stderr I see fixme:ole:_xread Read only 0 of 4 bytes from 0x58. fixme:ole:_StubReaderThread Failed with hres 80004005 and I note the presence of one or two stray wine processes and a wineserver: $ ps augxw | grep wine dank 947 14.9 0.3 3172 1236 ? S 11:58 0:10 wineserver dank 949 0.5 7.5 64400 29028 pts/0 S 11:58 0:00 /usr/local/bin/wine C:\PROG~FBU\COMM~CP1\INST~JM1\engine\6\INTE~MEX\I dank 957 0.0 7.5 64400 29028 pts/0 S 11:59 0:00 /usr/local/bin/wine C:\PROG~FBU\COMM~CP1\INST~JM1\engine\6\INTE~MEX\I Subsequent runs of the installer produce the error "error installing iKernel.exe: (0x1400)" probably because iKernel.exe is locked, and can't be overwritten. Killing all wine processes, e.g. killall wine lets the next run get past the 0x1400 error. I'm not stuck or even terribly bothered by this, but I thought I'd document it here in case anyone's interested. - Dan -- Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045

2 1