April 2004 - wine-devel

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar 12 Apr '04

12 Apr '04

OK. I will resend the fix after incorporating your feedback. Thanks Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry@baikal.ru] > Sent: Monday, April 12, 2004 6:34 PM > To: Santosh Siddheshwar > Cc: wine-devel(a)winehq.com > Subject: Re: Handling dialog messages for non-template based dialogs > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > Change Log: > > windows/defdlg.c - Added code for allocating DIALOGINFO structure and > > storing it as an attribute of the window > > windows/dialog.c - Removed code for allocating DIALOGINFO structure and > > moved initialization of DIALOGINFO structure > > to after CreateWindowEx has been called. > > Very nice! I'm glad to see that you finally have fixed that. However, > I see one minor point, which needs to be addressed as well: instead > of duplicating WM_CREATE code in both DefDlgProcA and W just move it > to DEFDLG_Proc. That way you will also fix 16-bit dialog creation code > by adding the same case into DefDlgProc16. > > Thanks for your great work! > > -- > Dmitry. ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

1 0

Re: Handling dialog messages for non-template based dialogs
by Dmitry Timoshkov 12 Apr '04

12 Apr '04

"Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > Change Log: > windows/defdlg.c - Added code for allocating DIALOGINFO structure and > storing it as an attribute of the window > windows/dialog.c - Removed code for allocating DIALOGINFO structure and > moved initialization of DIALOGINFO structure > to after CreateWindowEx has been called. Very nice! I'm glad to see that you finally have fixed that. However, I see one minor point, which needs to be addressed as well: instead of duplicating WM_CREATE code in both DefDlgProcA and W just move it to DEFDLG_Proc. That way you will also fix 16-bit dialog creation code by adding the same case into DefDlgProc16. Thanks for your great work! -- Dmitry.

1 0

Re: shell32: open desktop.ini to fetch icon
by Dmitry Timoshkov 12 Apr '04

12 Apr '04

"Juan Lang" <juan_lang(a)yahoo.com> wrote: > ChangeLog: if a folder contains a desktop.ini, check > it for the icon location before using the default > folder icon It would be much better if you could mark all string constants as such using 'const' key word in all your patches (including already committed, submitted and future ones). Taking into account Wine size that should considerably improve overall virtual memory usage. -- Dmitry.

1 0

Need to disable DTR on serial port
by Geoffrey Hausheer 12 Apr '04

12 Apr '04

In an ongoing effort to get TzMax working, I've been fighting to get it to talk to the device through the COM port. The problem is that the device (a wrist-watch...the program is used to download heart-rate data from said watch) is very difficult to comunicate with. In theory it is quite simple: send a 'ping' command, wait for a response, then send the 'receive' command, and read the data. In practice, getting the watch to acknowledge the initial ping is challenging. You need to repetitively send the 'ping' at odd intervals until the watch responds. Once it does, it will respond to commands readily. In any case, it turns out that one of the many requirements to get the watch to respond is to have DTR disabled (done by setting fDtrControl = DTR_CONTROL_DISABLE). But apparently the default for linux is DTR_CONTROL_ENABLE, and Wine will ignore any requests to change it (DTR_CONTROL_HANDSHAKE being unimplemented as well) I assume that driving the DTR line low is the trigger that tells the watch to listen to the Rx line. I am not sure if this is a Wine or Linux limitation, (as I have very limited knowledge of serial protocols), so I was wondering if anyone had any ideas for a suibtable workaround? Thanks, .Geoff

3 4

None
by None 12 Apr '04

12 Apr '04

by Pony.ns.zel.ru with ESMTP id i38KgMrt059607 for <wine-devel(a)winehq.org.KAV>; Fri, 9 Apr 2004 00:42:22 +0400 (MSD) (envelope-from deepfire(a)sic-elvis.zel.ru) Received: from mailserver sic-elvis by Pony.ns.zel.ru with ESMTP id i38KgMrt059598 for <wine-devel(a)winehq.org>; Fri, 9 Apr 2004 00:42:22 +0400 (MSD) (envelope-from deepfire(a)sic-elvis.zel.ru) Date: Thu, 08 Apr 2004 23:43:44 +0400 Message-ID: <87wu4qdzm7.wl(a)canopus.ns.zel.ru> From: Samium Gromoff <deepfire(a)sic-elvis.zel.ru> To: wine-devel(a)winehq.org Subject: Re: ANNOUNCE: CVS to arch gateway User-Agent: Wanderlust/2.10.1 (Watching The Wheels) SEMI/1.14.6 (Maruoka) FLIM/1.14.6 (Marutamachi) APEL/10.6 Emacs/21.3 (i386-pc-linux-gnu) MULE/5.0 (SAKAKI) MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Sender: wine-devel-admin(a)winehq.org Errors-To: wine-devel-admin(a)winehq.org X-BeenThere: wine-devel(a)winehq.org X-Mailman-Version: 2.0.13 Precedence: bulk List-Unsubscribe: <http://www.winehq.org/mailman/listinfo/wine-devel>, <mailto:wine-devel-request@winehq.org?subject=unsubscribe> List-Id: Wine Developer's List <wine-devel.winehq.org> List-Post: <mailto:wine-devel@winehq.org> List-Help: <mailto:wine-devel-request@winehq.org?subject=help> List-Subscribe: <http://www.winehq.org/mailman/listinfo/wine-devel>, <mailto:wine-devel-request@winehq.org?subject=subscribe> List-Archive: <http://www.winehq.org/hypermail/wine-devel/> Perphaps it might be interesting, that there exists a gtk2 frontend to tla, called tlator. Well, it might sound like a shameless plug, because that`s me who actually wrote it, but the perphaps the main incentive behind this mail is to widen arch acceptance among people, so who knows... ;-) The tlator arch repository is located at James Blackwell`s sourcecontrol crossroad-like place: http://sourcecontrol.net/~sgromoff/%{archives}/deepfire@betelheise.deep.net… regards, Samium Gromoff

1 0

Re: win32 build error log
by Ivan Leo Murray-Smith 11 Apr '04

11 Apr '04

> Patch is in Wine CVS. w32 packages are built against the release source, not the CVS. These fixes will affect next months build. > Patch is in MinGW CVS. Will this be ready for next month, or should I install mingw from CVS? Ivan.

1 0

Wine menu woes
by Geoffrey Hausheer 11 Apr '04

11 Apr '04

I have an application that seems to work very well, except that the menus don't work (in wine, they are fine in Windows) The app is located here (free download): http://www.acumeninc.com/download/tzmax50v3.21.exe After some debugging, I found that the app creates a menu, then destroys it and recreates it using different titles. Something like: hMenu = CreateMenu(); hSub1 = CreatePopupMenu(); ... hSub2 = CreatePopupMenu(); ... AppendMenu(hMenu, MF_STRING | MF_POPUP, hSub1, "&File"); AppendMenu(hMenu, MF_STRING | MF_POPUP, hSub2, "&Other"); ... SetMenu(hWnd,hMenu); while(GetMenuItemCount(hMenu)) RemoveMenu(hMenu,0, 0x400); //Note: this just detaches the menus hOldMenu = hMenu; DestroyMenu(hMenu); hMenu=CreateMenu(); AppendMenu(hMenu, MF_STRING | MF_POPUP, hSub1, "[&File]"); AppendMenu(hMenu, MF_STRING | MF_POPUP, hSub2, "[&Other]"); if(hMenu != hOldMenu) SetMenu(hWnd, hMenu); That last line is where wine and Windows differ. In windows, the menuid is always different, so hMenu never equals hOldMenu. In wine, we reuse the menuid, so hMenu always equals hOldMenu. Since SetMenu didn't get called after the last createMenu, the menus don't work. I have no idea why the app implements its menus this way, but I removed the 'USER_HEAP_FREE( hMenu );' in DestroyMenu() and the menus work fine, as expected (obviously this isn't a fix, just a validation that the code behaves this way). I've included a test app that does the above if anyone is interested. As a secondary issue, for some reason the 'Exit' item executes the ID_DO_SOMETHING code instead of the ID_FILE_EXIT code. It doesn't do that in windows (it correctly exits instead). I don't think it is related, and i haven't looked into it, but I thought I'd mention it. Thanks, .Geoff

4 3

Re: win32 build error log
by Filip Navara 11 Apr '04

11 Apr '04

>> i586-mingw32msvc/bin/ld: cannot find -ldxerr9 > > MinGW lacks up to date DirectX headers and import libs > but Filip Navara has been working on this. His work just > needs to be integrated into MinGW. Meanwhile I will add > a dxerr8 and dxerr9 to my MinGW rpm packages so we have > a means of cross compiling our tests. My work was recently merged into MinGW CVS. [Danny Smith immediately did some changes in dxerr libraries that I think aren't right, but haven't tested it. So it's possible that the versions in MinGW CVS will not work anyway.] - Filip

1 0

win32 build error log
by Ivan Leo Murray-Smith 11 Apr '04

11 Apr '04

The error log from the win32 build of wine-20040408 Ivan.

2 1

Re: exec-shield workaround take 2
by Dan Kegel 11 Apr '04

11 Apr '04

Mike Hearn wrote: > It seems it's turned from being an normal (albiet static) app which > reserves the areas needed then boots wine, into a reimplementation of > ld-linux.so? This isn't all bad. Consider: we could add the ability to load PE executables and shared libraries into this private ld-linux.so. At that point, the need for winelib executables to be shared libraries goes away, doesn't it? And that would make everybody very, very happy, as it would solve the problem that made mono and everybody else fork wine just so they could load Windows DLLs into native linux apps. Or am I confused? Wouldn't be the first time... - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

2 1

programs/winecfg/main.c breakage
by Gerald Pfeifer 11 Apr '04

11 Apr '04

I'm afraid the following change to programs/winecfg/main.c date: 2004/03/16 01:32:36; author: julliard; state: Exp; lines: +2 -0 Need to include config.h. broke FreeBSD (and others): /sw/gcc-3.3.2/bin/gcc -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o main.o main.c In file included from main.c:35: /usr/include/unistd.h:189: error: conflicting types for `select' ../../include/winsock.h:975: error: previous declaration of `select' Gerald -- Gerald Pfeifer (Jerry) gerald(a)pfeifer.com http://www.pfeifer.com/gerald/

1 1

Abi Word ported to Winelib!
by Dimitrie O. Paun 11 Apr '04

11 Apr '04

Hi folks, In the past couple days I've entertained myself with porting Abi Word to Winelib. This was possible because now Abi Word has a nice MinGW build system. A total of 5 trivial patches were indeed needed for Wine: http://www.winehq.org/hypermail/wine-patches/2004/04/0122.html http://www.winehq.org/hypermail/wine-patches/2004/04/0123.html http://www.winehq.org/hypermail/wine-patches/2004/04/0124.html http://www.winehq.org/hypermail/wine-patches/2004/04/0147.html http://www.winehq.org/hypermail/wine-patches/2004/04/0155.html Notice how little we had to fix! To encourage others to port stuff to Winelib, I'll inline the hacks I did to port Abi Word. These we'll need to be cleaned up for integration into the Abi Word tree, but again, please note how little we had to change. (I've also attached a screeshot of the resulting Abi Word in action). diff -ru abiword-2.0.6/abi/src/config/platforms/mingw32.mk abiword-2.0.6-dimi/abi/src/config/platforms/mingw32.mk --- abiword-2.0.6/abi/src/config/platforms/mingw32.mk 2003-04-22 12:03:54.000000000 -0400 +++ abiword-2.0.6-dimi/abi/src/config/platforms/mingw32.mk 2004-04-09 19:50:33.000000000 -0400 @@ -28,7 +28,7 @@ ## PLATFORM_FLAGS = PORT_FLAGS = -OS_CFLAGS = -mwindows -DDM_SPECVERSION=0x0320 +OS_CFLAGS = -mno-cygwin -mwindows -DDM_SPECVERSION=0x0320 # Earnie Boyd has added DM_SPECVERSION macro (of this value) to MinGW # CVS tree on 04/14/03 so this is only for versions 2.0.0-3 and earlier ## @@ -49,8 +49,8 @@ IA64_ARCH_FLAGS = # Define tools -CC = gcc -CCC = g++ +CC = winegcc +CCC = wineg++ RANLIB = ranlib # Suffixes @@ -155,10 +155,10 @@ ABI_REQUIRE_PEER_ICONV = 1 # Define tools - resource compiler -RC = windres +RC = wrc # Suffixes -EXE_SUFFIX = .exe +#EXE_SUFFIX = .exe # Compiler flags # requires the commctrl.dll from ie4.0 or greater diff -ru abiword-2.0.6/abi/src/wp/ap/win/ap_Win32App.cpp abiword-2.0.6-dimi/abi/src/wp/ap/win/ap_Win32App.cpp --- abiword-2.0.6/abi/src/wp/ap/win/ap_Win32App.cpp 2004-01-17 07:33:11.000000000 -0500 +++ abiword-2.0.6-dimi/abi/src/wp/ap/win/ap_Win32App.cpp 2004-04-09 20:55:03.000000000 -0400 @@ -35,7 +35,6 @@ #include <string.h> #include <io.h> #include <fcntl.h> -#include <iostream.h> #include <ole2.h> #include "ut_debugmsg.h" @@ -1296,7 +1295,7 @@ // -#if !defined(__MINGW32__) && !defined(DEBUG) +#if defined(WIN32) && !defined(__MWERKS__) && !defined(__WINE__) __try #endif { @@ -1364,7 +1363,7 @@ // If an exception happens, with "catch" the block // and then the save it into disk // -#if !defined(__MINGW32__) && !defined(DEBUG) +#if defined(WIN32) && !defined(__MWERKS__) && !defined(__WINE__) __except (1) { AP_Win32App *pApp = (AP_Win32App *) XAP_App::getApp(); diff -ru abiword-2.0.6/abi/src/wp/main/win/Makefile abiword-2.0.6-dimi/abi/src/wp/main/win/Makefile --- abiword-2.0.6/abi/src/wp/main/win/Makefile 2003-07-23 23:31:36.000000000 -0400 +++ abiword-2.0.6-dimi/abi/src/wp/main/win/Makefile 2004-04-09 18:18:39.000000000 -0400 @@ -91,10 +91,10 @@ ifeq ($(OS_NAME), MINGW32) $(PROGRAM): $(OBJS) $(EXTRA_LIBDEP) @$(MAKE_OBJDIR) - @echo Linking dynamic $(PROGRAM) - $(CCC) -o $(BINDIR)/libAbiWord.dll $(SHARED_LDFLAGS) $(CFLAGS) $(OBJDIR)/dummy.o $(LDFLAGS) $(EXTRA_LIBS) $(OS_LIBS) $(XLDFLAGS) - @mv libAbiWord.dll.a $(LIBDIR) - $(CCC) -o $@ $(CFLAGS) $(OBJDIR)/Win32Main.o $(OBJDIR)/Win32Mainrc.o $(LDFLAGS) -L$(LIBDIR) -lAbiWord.dll $(OS_LIBS) $(XLDFLAGS) + #@echo Linking dynamic $(PROGRAM) + #$(CCC) -o $(BINDIR)/libAbiWord.dll $(SHARED_LDFLAGS) $(CFLAGS) $(OBJDIR)/dummy.o $(LDFLAGS) $(EXTRA_LIBS) $(OS_LIBS) $(XLDFLAGS) + #@mv libAbiWord.dll.a $(LIBDIR) + $(CCC) -o $@ $(CFLAGS) $(OBJDIR)/Win32Main.o $(OBJDIR)/Win32Mainrc.o $(LDFLAGS) $(OBJDIR)/dummy.o $(EXTRA_LIBS) $(OS_LIBS) $(XLDFLAGS) else $(PROGRAM): $(OBJS) $(EXTRA_LIBDEP) @$(MAKE_OBJDIR) diff -ru abiword-2.0.6/wv/config.h.in abiword-2.0.6-dimi/wv/config.h.in --- abiword-2.0.6/wv/config.h.in 2003-06-21 11:58:55.000000000 -0400 +++ abiword-2.0.6-dimi/wv/config.h.in 2004-04-08 00:25:53.000000000 -0400 @@ -164,7 +164,7 @@ #endif /* redefs of things that are either in glibc or we have to include them ourselves */ -#if defined(WIN32) && !defined(__MWERKS__) +#if defined(WIN32) && !defined(__MWERKS__) && !defined(__WINE__) #define strcasecmp(s1,s2) stricmp(s1,s2) #else #if !defined(__GLIBC__) || (__GLIBC__ < 2) || (__GLIBC__ == 2 && __GLIBC_MINOR__ < 2) diff -ru abiword-2.0.6/wv/config.h.msvc abiword-2.0.6-dimi/wv/config.h.msvc --- abiword-2.0.6/wv/config.h.msvc 2002-11-10 21:00:49.000000000 -0500 +++ abiword-2.0.6-dimi/wv/config.h.msvc 2004-04-08 00:23:58.000000000 -0400 @@ -169,7 +169,7 @@ #endif /* redefs of things that are either in glibc or we have to include them ourselves */ -#if defined(WIN32) && !defined(__MWERKS__) +#if defined(WIN32) && !defined(__MWERKS__) && !defined(__WINE__) #define strcasecmp(s1,s2) stricmp(s1,s2) #else #if !defined(__GLIBC__) || (__GLIBC__ < 2) || (__GLIBC__ == 2 && __GLIBC_MINOR__ < 2) diff -ru abiword-2.0.6/wv/configure.ac abiword-2.0.6-dimi/wv/configure.ac --- abiword-2.0.6/wv/configure.ac 2003-09-15 06:03:44.000000000 -0400 +++ abiword-2.0.6-dimi/wv/configure.ac 2004-04-08 00:21:32.000000000 -0400 @@ -701,7 +701,7 @@ #endif -- Dimi.

2 2

exec-shield workaround take 2
by Mike McCormack 11 Apr '04

11 Apr '04

Hi, This patch works around the exec-shield problems with Fedora Core. I'd be greatful if any people using Fedora core could test it out and see how it works. It should be enough to apply it to the latest CVS tip and recompile. All comments and flames accepted... /me wears his asbestos suit. Mike ChangeLog: * workaround exec-shield by reserving memory before run ld-linux.so.2 Index: dlls/kernel/process.c =================================================================== RCS file: /home/wine/wine/dlls/kernel/process.c,v retrieving revision 1.50 diff -u -r1.50 process.c --- dlls/kernel/process.c 25 Mar 2004 23:41:04 -0000 1.50 +++ dlls/kernel/process.c 10 Apr 2004 12:01:01 -0000 @@ -1055,7 +1055,8 @@ * * Build the environment of a new child process. */ -static char **build_envp( const WCHAR *envW, const WCHAR *extra_envW ) +static char **build_envp( const WCHAR *envW, const WCHAR *extra_envW, + const char *workdir ) { const WCHAR *p; char **envp; @@ -1078,7 +1079,7 @@ if (!(env = malloc( length ))) return NULL; WideCharToMultiByte( CP_UNIXCP, 0, envW, p - envW, env, length, NULL, NULL ); - count += 4; + count += 5; if ((envp = malloc( count * sizeof(*envp) ))) { @@ -1093,6 +1094,10 @@ if ((p = getenv("TMP"))) *envptr++ = alloc_env_string( "TMP=", p ); if ((p = getenv("HOME"))) *envptr++ = alloc_env_string( "HOME=", p ); if ((p = getenv("WINEPREFIX"))) *envptr++ = alloc_env_string( "WINEPREFIX=", p ); + + /* record the unix working directory */ + if (workdir) *envptr++ = alloc_env_string( "WINEWORKDIR=", workdir ); + /* now put the Windows environment strings */ for (p = env; *p; p += strlen(p) + 1) { @@ -1131,7 +1136,7 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmdline, 0 ); - char **envp = build_envp( env, NULL ); + char **envp = build_envp( env, NULL, NULL ); close( fd[0] ); /* Reset signals that we previously set to SIG_IGN */ @@ -1263,7 +1268,16 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmd_line, 1 ); - char **envp = build_envp( env, extra_env ); + char **envp; + + /* preseve the current directory, the preloader may forget it */ + if( !unixdir ) + { + char *dir = HeapAlloc( GetProcessHeap(), 0, MAX_PATH ); + getcwd( dir, MAX_PATH ); + unixdir = dir; + } + envp = build_envp( env, extra_env, unixdir ); close( startfd[1] ); close( execfd[0] ); @@ -1275,8 +1289,6 @@ /* Reset signals that we previously set to SIG_IGN */ signal( SIGPIPE, SIG_DFL ); signal( SIGCHLD, SIG_DFL ); - - if (unixdir) chdir(unixdir); if (argv && envp) { Index: libs/wine/config.c =================================================================== RCS file: /home/wine/wine/libs/wine/config.c,v retrieving revision 1.3 diff -u -r1.3 config.c --- libs/wine/config.c 11 Nov 2003 22:21:29 -0000 1.3 +++ libs/wine/config.c 10 Apr 2004 12:01:01 -0000 @@ -243,6 +243,23 @@ return user_name; } +/* run wine binary in it's own directory, so it can find it's preloader */ +int wine_cwd_execve( const char *name, char **argv, char **envp ) +{ + char *dir, *p; + + dir = strdup( name ); + p = strrchr( dir, '/' ); + if( p ) + { + *p++ = 0; + chdir( dir ); + name = p; + } + + return execve( name, argv, envp ); +} + /* exec a wine internal binary (either the wine loader or the wine server) */ /* if name is null, default to the name of the current binary */ void wine_exec_wine_binary( const char *name, char **argv, char **envp ) @@ -257,7 +274,7 @@ argv[0] = xmalloc( sizeof(BINDIR "/") + strlen(name) ); strcpy( argv[0], BINDIR "/" ); strcat( argv[0], name ); - execve( argv[0], argv, envp ); + wine_cwd_execve( argv[0], argv, envp ); free( argv[0] ); /* now try the path of argv0 of the current binary */ @@ -266,7 +283,7 @@ argv[0] = xmalloc( strlen(argv0_path) + strlen(name) + 1 ); strcpy( argv[0], argv0_path ); strcat( argv[0], name ); - execve( argv[0], argv, envp ); + wine_cwd_execve( argv[0], argv, envp ); free( argv[0] ); } @@ -283,7 +300,7 @@ memcpy( argv[0], pos, ptr - pos ); strcpy( argv[0] + (ptr - pos), "/" ); strcat( argv[0] + (ptr - pos), name ); - execve( argv[0], argv, envp ); + wine_cwd_execve( argv[0], argv, envp ); pos = ptr; } free( argv[0] ); Index: loader/Makefile.in =================================================================== RCS file: /home/wine/wine/loader/Makefile.in,v retrieving revision 1.14 diff -u -r1.14 Makefile.in --- loader/Makefile.in 22 Nov 2003 00:08:26 -0000 1.14 +++ loader/Makefile.in 10 Apr 2004 12:01:01 -0000 @@ -23,20 +23,30 @@ LIBPTHREAD = @LIBPTHREAD@ LDEXECFLAGS = @LDEXECFLAGS@ -wine-glibc: glibc.o Makefile.in +WINELDEXECFLAGS = $(LDEXECFLAGS),--dynamic-linker=$(WINELDSO) + +WINELDSO=ld-winepreload.so + +$(WINELDSO): preload.o Makefile.in + $(CC) -nostartfiles -lc -static -Wl,"-Ttext=0x7ff81000" -o $@ $< + +preload.c: preload.h + +wine-glibc: glibc.o Makefile.in $(WINELDSO) $(CC) -o $@ $(LDEXECFLAGS) glibc.o $(LIBWINE) $(LIBPORT) $(LIBPTHREAD) $(EXTRALIBS) $(LDFLAGS) -wine-kthread: $(KTHREAD_OBJS) Makefile.in - $(CC) -o $@ $(LDEXECFLAGS) $(KTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(EXTRALIBS) $(LDFLAGS) +wine-kthread: $(KTHREAD_OBJS) Makefile.in $(WINELDSO) + $(CC) -o $@ $(WINELDEXECFLAGS) $(KTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(EXTRALIBS) $(LDFLAGS) -wine-pthread: $(PTHREAD_OBJS) Makefile.in - $(CC) -o $@ $(LDEXECFLAGS) $(PTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(LIBPTHREAD) $(EXTRALIBS) $(LDFLAGS) +wine-pthread: $(PTHREAD_OBJS) Makefile.in $(WINELDSO) + $(CC) -o $@ $(WINELDEXECFLAGS) $(PTHREAD_OBJS) $(LIBWINE) $(LIBPORT) $(LIBPTHREAD) $(EXTRALIBS) $(LDFLAGS) $(MODULE): $(MAIN_BINARY) $(RM) $(MODULE) && $(LN_S) $(MAIN_BINARY) $(MODULE) install:: $(WINE_BINARIES) $(MKINSTALLDIRS) $(bindir) + $(INSTALL_PROGRAM) $(WINELDSO) $(bindir); \ for f in $(WINE_BINARIES); do \ if [ "$(MAIN_BINARY)" = "$$f" ]; \ then $(INSTALL_PROGRAM) $$f $(bindir)/$(MODULE); \ @@ -48,6 +58,6 @@ $(RM) $(WINE_BINARIES:%=$(bindir)/%) $(bindir)/$(MODULE) clean:: - $(RM) $(WINE_BINARIES) $(MODULE) + $(RM) $(WINE_BINARIES) $(MODULE) $(WINELDSO) ### Dependencies: Index: loader/glibc.c =================================================================== RCS file: /home/wine/wine/loader/glibc.c,v retrieving revision 1.3 diff -u -r1.3 glibc.c --- loader/glibc.c 20 Feb 2004 20:19:23 -0000 1.3 +++ loader/glibc.c 10 Apr 2004 12:01:01 -0000 @@ -75,6 +75,17 @@ { const char *loader = getenv( "WINELOADER" ); const char *threads = get_threading(); + char cwd[MAX_PATH], *cwd_var; + + if( !getcwd( cwd, sizeof cwd ) ) + { + fprintf( stderr, "wine: could get current directory\n" ); + exit(1); + } + cwd_var = xmalloc( sizeof("WINEWORKDIR=") + strlen(cwd) ); + strcpy( cwd_var, "WINEWORKDIR=" ); + strcat( cwd_var, cwd ); + putenv( cwd_var ); if (loader) { @@ -86,6 +97,9 @@ new_name = xmalloc( (path - loader) + strlen(threads) + 1 ); memcpy( new_name, loader, path - loader ); + new_name[ path - loader ] = 0; + if( new_name[0] ) + chdir( new_name ); strcpy( new_name + (path - loader), threads ); /* update WINELOADER with the new name */ Index: loader/main.c =================================================================== RCS file: /home/wine/wine/loader/main.c,v retrieving revision 1.90 diff -u -r1.90 main.c --- loader/main.c 20 Jan 2004 00:28:01 -0000 1.90 +++ loader/main.c 10 Apr 2004 12:01:01 -0000 @@ -20,6 +20,7 @@ #include <stdio.h> #include <stdlib.h> +#include <unistd.h> #include "wine/library.h" /********************************************************************** @@ -27,12 +28,13 @@ */ int main( int argc, char *argv[] ) { - char error[1024]; + char error[1024], *dir; -#if 0 - static char pe_load[256*1024*1024] __attribute__((aligned(4096))); - wine_set_pe_load_area( pe_load, sizeof(pe_load) ); -#endif + wine_set_pe_load_area((void*)0x00400000, 0x01000000); + + dir = getenv( "WINEWORKDIR" ); + if( dir ) + chdir( dir ); wine_init( argc, argv, error, sizeof(error) ); fprintf( stderr, "wine: failed to initialize: %s\n", error ); --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.c 2004-04-10 21:43:23.000000000 +0900 @@ -0,0 +1,570 @@ +/* + mini loader for ld.so - x86 Linux 2.4.x+ kernels only + + Copyright 2004 Mike McCormack for Codeweavers + + Copyright (C) 1995,96,97,98,99,2000,2001,2002 Free Software Foundation, Inc. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. + */ + +/* + * Design notes + * + * The goal of this program is to be a workaround for exec-shield, as used + * by the Linux kernel distributed with Fedora Core and other distros. + * + * To do this, we implement our own shared object loader that reserves memory + * that is important to Wine, and then loads the the standard shared object + * loader (ELF interpreter), usually ld-linux.so.2 + * + * As Wine is i386 specific, we are not too worried about portability issues, + * but will do our best to keep the system dependencies (syscalls) seperate to + * the rest (ELF loader and printf) part of the program. + * + * This program will be invoked as follows: + * + * wld <full path to ELF executable to be loaded> <args...> + * + * We will try to set up the stack an memory area so that the program that + * loads after us (eg. the wine binary) never knows we were here, except that + * areas of memory it needs are already magically reserved. + * + * The following memory areas are important to Wine: + * 0x00000000 - the DOS area + * 0x00400000 - the standard Win32 load address + * 0x65430000 - the shared heap at + * + * There are a few things that will be different after this program passes + * control of execution to the loaded binary: + * 1) The _ variable in the environment (will be the name of this exe) + * 2) The brk area returned by brk(2) + * 3) This program will still be mapped into memory. + * 4) The command line kept by the kernel in kernel memory (/proc/self/cmdline) + * + */ + +/* + * References (things I consulted to understand how ELF loading works): + * + * glibc 2.3.2 elf/dl-load.c + * http://www.gnu.org/directory/glibc.html + * + * Linux 2.6.4 fs/binfmt_elf.c + * ftp://ftp.kernel.org/pub/linux/kernel/v2.6/linux-2.6.4.tar.bz2 + * + * Userland exec, by <grugq(a)hcunix.net> + * http://cert.uni-stuttgart.de/archive/bugtraq/2004/01/msg00002.html + * + * The ELF specification: + * http://www.linuxbase.org/spec/booksets/LSB-Embedded/LSB-Embedded/book387.ht… + */ + +#include "preload.h" +#include <elf.h> + +/* debugging */ +#undef DUMP_SEGMENTS +#undef DUMP_ARG +#undef DUMP_AUX_INFO + +unsigned int GL(dl_pagesize) = 0x0000; + +struct link_map { + ElfW(Addr) l_addr; + ElfW(Dyn) *l_ld; + const ElfW(Phdr) *l_phdr; + ElfW(Addr) l_entry; + ElfW(Half) l_ldnum; + ElfW(Half) l_phnum; + ElfW(Addr) l_map_start, l_map_end; +}; + +/* + * wld_printf - just the basics + * + * %x prints a hex number + * %s prints a string + */ +void wld_sprintf(char *str, char *fmt, int *args ) +{ + char *p = fmt; + + while( *p ) + { + if( *p == '%' ) + { + p++; + if( *p == 'x' ) + { + int ch, i, x = *++args; + for(i=7; i>=0; i--) + { + ch = (x>>(i*4))&0xf; + ch += '0'; + if(ch>'9') + ch+=('A'-10-'0'); + *str++ = ch; + } + } + else if( *p == 's' ) + { + char *s = (char*)*++args; + while(*s) + *str++ = *s++; + } + else if( *p == 0 ) + break; + p++; + } + *str++ = *p++; + } + *str = 0; +} + +int wld_strlen(char *str) +{ + int n = 0; + while( str[n] ) + n++; + return n; +} + +void wld_printf(char *fmt, ... ) +{ + char buffer[100]; + int n; + + buffer[0]=0; + wld_sprintf(buffer, fmt, (int*)&fmt ); + n = wld_strlen( buffer ); + write(2, buffer, n); +} + +void wld_strncpy(char *dest, char *src, int count ) +{ + while( (*dest++ = *src++) && --count ) + ; + *dest = 0; +} + +void wld_memset( unsigned char *buf, int val, size_t len ) +{ + while( len-- ) + *buf++=val; +} + +/* + * Dump interesting bits of the ELF auxv_t structure that is passed + * as the 4th parameter to the _start function + */ +void wld_dump_auxilary( ElfW(auxv_t) *av ) +{ + for ( ; av->a_type != AT_NULL; av++) + switch (av->a_type) + { + case AT_PAGESZ: + GL(dl_pagesize) = av->a_un.a_val; +#ifndef DUMP_AUX_INFO + break; +#else + wld_printf("AT_PAGESZ = %x\n",av->a_un.a_val); + break; + case AT_PHDR: + wld_printf("AT_PHDR = %x\n",av->a_un.a_ptr); + break; + case AT_PHNUM: + wld_printf("AT_PHNUM = %x\n",av->a_un.a_val); + break; + case AT_ENTRY: + wld_printf("AT_ENTRY = %x\n",av->a_un.a_val); + break; + case AT_BASE: + wld_printf("AT_BASE = %x\n",av->a_un.a_val); + break; +#endif + } +} + +/* + * wld_set_auxilary + * + * Set a field of the auxillary structure + */ +void wld_set_auxilary( ElfW(auxv_t) *av, int type, long int val ) +{ + for ( ; av->a_type != AT_NULL; av++) + if( av->a_type == type ) + av->a_un.a_val = val; +} + +/* + * wld_map_so + * + * modelled after _dl_map_object_from_fd() from glibc-2.3.1/elf/dl-load.c + * + * This function maps the segments from an ELF object, and optionally + * stores information about the mapping into the auxv_t structure. + */ +void* wld_map_so( int fd, void* buf, size_t buflen, ElfW(auxv_t)* av, struct link_map *l) +{ + ElfW(Ehdr) *header = buf; + ElfW(Phdr) *phdr, *ph; + /* Scan the program header table, collecting its load commands. */ + struct loadcmd + { + ElfW(Addr) mapstart, mapend, dataend, allocend; + off_t mapoff; + int prot; + } loadcmds[header->e_phnum], *c; + size_t nloadcmds = 0, maplength; + + header = buf; + phdr = (void*) (((unsigned char*)buf) + header->e_phoff); + + if( header->e_ident[0] != 0x7f && + header->e_ident[0] != 'E' && + header->e_ident[0] != 'L' && + header->e_ident[0] != 'F' ) + { + wld_printf("Not an ELF binary... don't know how to load it\n"); + return NULL; + } + + if( header->e_machine != EM_386 ) + { + wld_printf("Not an i386 ELF binary... don't know how to load it\n"); + return NULL; + } + + maplength = header->e_phnum * sizeof (ElfW(Phdr)); + if (header->e_phoff + maplength > (size_t) buflen) + { + wld_printf("oops... not enough space for ELF headers\n"); + exit(1); + } + + l->l_ld = 0; + l->l_addr = 0; + l->l_phnum = header->e_phnum; + l->l_entry = header->e_entry; + + for (ph = phdr; ph < &phdr[l->l_phnum]; ++ph) + { + +#ifdef DUMP_SEGMENTS + wld_printf( "ph = %x\n", ph ); + wld_printf( " p_type = %x\n", ph->p_type ); + wld_printf( " p_flags = %x\n", ph->p_flags ); + wld_printf( " p_offset = %x\n", ph->p_offset ); + wld_printf( " p_vaddr = %x\n", ph->p_vaddr ); + wld_printf( " p_paddr = %x\n", ph->p_paddr ); + wld_printf( " p_filesz = %x\n", ph->p_filesz ); + wld_printf( " p_memsz = %x\n", ph->p_memsz ); + wld_printf( " p_align = %x\n", ph->p_align ); +#endif + + switch (ph->p_type) + { + /* These entries tell us where to find things once the file's + segments are mapped in. We record the addresses it says + verbatim, and later correct for the run-time load address. */ + case PT_DYNAMIC: + l->l_ld = (void *) ph->p_vaddr; + l->l_ldnum = ph->p_memsz / sizeof (Elf32_Dyn); + break; + + case PT_PHDR: + l->l_phdr = (void *) ph->p_vaddr; + break; + + case PT_TLS: + /* + * We don't need to set anything up because we're + * emulating the kernel, not ld-linux.so.2 + * The ELF loader will set up the TLS data itself. + */ + break; + + case PT_LOAD: + { + if ((ph->p_align & (GL(dl_pagesize) - 1)) != 0) + { + wld_printf("ELF load command alignment not page-aligned"); + return NULL; + } + if (((ph->p_vaddr - ph->p_offset) & (ph->p_align - 1)) != 0) + { + wld_printf("ELF load command address/offset not properly aligned"); + return NULL; + } + + c = &loadcmds[nloadcmds++]; + c->mapstart = ph->p_vaddr & ~(ph->p_align - 1); + c->mapend = ((ph->p_vaddr + ph->p_filesz + GL(dl_pagesize) - 1) + & ~(GL(dl_pagesize) - 1)); + c->dataend = ph->p_vaddr + ph->p_filesz; + c->allocend = ph->p_vaddr + ph->p_memsz; + c->mapoff = ph->p_offset & ~(ph->p_align - 1); + + c->prot = 0; + if (ph->p_flags & PF_R) + c->prot |= PROT_READ; + if (ph->p_flags & PF_W) + c->prot |= PROT_WRITE; + if (ph->p_flags & PF_X) + c->prot |= PROT_EXEC; + } + break; + + case PT_INTERP: + break; + + case PT_SHLIB: + case PT_NOTE: + case PT_GNU_EH_FRAME: + case 0x6474e551: /* FIXME: what is this? */ + break; + + default: + wld_printf("warning: unknown header type %x\n", ph->p_type); + break; + } + } + + /* Now process the load commands and map segments into memory. */ + c = loadcmds; + + /* Length of the sections to be loaded. */ + maplength = loadcmds[nloadcmds - 1].allocend - c->mapstart; + + if( header->e_type == ET_DYN ) + { + ElfW(Addr) mappref; + mappref = (ELF_PREFERRED_ADDRESS (loader, maplength, c->mapstart) + - MAP_BASE_ADDR (l)); + + /* Remember which part of the address space this object uses. */ + l->l_map_start = (ElfW(Addr)) mmap ((void *) mappref, maplength, + c->prot, MAP_COPY | MAP_FILE, + fd, c->mapoff); + /* wld_printf("set : offset = %x\n", c->mapoff); */ + /* wld_printf("l->l_map_start = %x\n", l->l_map_start); */ + + l->l_map_end = l->l_map_start + maplength; + l->l_addr = l->l_map_start - c->mapstart; + + mprotect ((caddr_t) (l->l_addr + c->mapend), + loadcmds[nloadcmds - 1].allocend - c->mapend, + PROT_NONE); + goto postmap; + } + else + { + ELF_FIXED_ADDRESS (loader, c->mapstart); + } + + /* Remember which part of the address space this object uses. */ + l->l_map_start = c->mapstart + l->l_addr; + l->l_map_end = l->l_map_start + maplength; + + while (c < &loadcmds[nloadcmds]) + { + if (c->mapend > c->mapstart) + /* Map the segment contents from the file. */ + mmap ((void *) (l->l_addr + c->mapstart), + c->mapend - c->mapstart, c->prot, + MAP_FIXED | MAP_COPY | MAP_FILE, fd, c->mapoff); + + postmap: + if (l->l_phdr == 0 + && (ElfW(Off)) c->mapoff <= header->e_phoff + && ((size_t) (c->mapend - c->mapstart + c->mapoff) + >= header->e_phoff + header->e_phnum * sizeof (ElfW(Phdr)))) + /* Found the program header in this segment. */ + l->l_phdr = (void *) (c->mapstart + header->e_phoff - c->mapoff); + + if (c->allocend > c->dataend) + { + /* Extra zero pages should appear at the end of this segment, + after the data mapped from the file. */ + ElfW(Addr) zero, zeroend, zeropage; + + zero = l->l_addr + c->dataend; + zeroend = l->l_addr + c->allocend; + zeropage = ((zero + GL(dl_pagesize) - 1) + & ~(GL(dl_pagesize) - 1)); + + /* + * This is different to the dl-load load... + * ld-linux.so.2 relies on the whole page being zero'ed + */ + zeroend = (zeroend +GL(dl_pagesize) - 1) & ~(GL(dl_pagesize) - 1); + + if (zeroend < zeropage) + { + /* All the extra data is in the last page of the segment. + We can just zero it. */ + zeropage = zeroend; + } + + if (zeropage > zero) + { + /* Zero the final part of the last page of the segment. */ + if ((c->prot & PROT_WRITE) == 0) + { + /* Dag nab it. */ + mprotect ((caddr_t) (zero & ~(GL(dl_pagesize) - 1)), + GL(dl_pagesize), + c->prot|PROT_WRITE); + } + memset ((void *) zero, '\0', zeropage - zero); + if ((c->prot & PROT_WRITE) == 0) + mprotect ((caddr_t) (zero & ~(GL(dl_pagesize) - 1)), + GL(dl_pagesize), c->prot); + } + + if (zeroend > zeropage) + { + /* Map the remaining zero pages in from the zero fill FD. */ + caddr_t mapat; + mapat = mmap ((caddr_t) zeropage, zeroend - zeropage, + c->prot, MAP_ANON|MAP_PRIVATE|MAP_FIXED, + ANONFD, 0); + } + } + + ++c; + } + + if (l->l_phdr == NULL) + { + wld_printf("no program header\n"); + exit(1); + } + + (ElfW(Addr)) l->l_phdr += l->l_addr; + (ElfW(Addr)) l->l_entry += l->l_addr; + + return (void*)l->l_entry; +} + +/* + * wld_reserve_wine_memory + * + * This is the reason we are here. + * We want to reserve address space so that the kernel doesn't + * map stuff there. We do this by getting in first before + * anything else is mapped (except this program) and reserve + * the memory we want. + */ +void wld_reserve_wine_memory() +{ + mmap((void*)0x00400000, 0x01000000, PROT_NONE, + MAP_FIXED|MAP_PRIVATE|MAP_ANON, -1, 0); +} + +/* + * wld_start + * + * Repeat the actions the kernel would do when loading a dynamically linked .so + * Load the ELF interperter ld-linux.so and then binary itself. + * + * Note, we assume that the binary is a dynamically linked ELF shared object + * and that ld-linux.so is its ELF interpreter. + */ +void* wld_start( int argc ) +{ + int fd; + unsigned char buf[0x800], interp[0x100]; + void *start; + char **argv = (char**) &argc, **p; + struct link_map map; + ElfW(auxv_t)* av; + + memset( &map, 0, sizeof map ); + + /* skip over the parameters */ + p = ++argv; +#ifdef DUMP_ARG + { + int i; + for( i = 0; i<argc; i++ ) + wld_printf("argv[%x] = %s\n", i, *p++); + if(*p++) + wld_printf("argv[argc] not NULL!!\n"); + } +#else + p+=(argc+1); +#endif + + /* skip over the environment */ + while( *p++ ) + ; + + av = (ElfW(auxv_t)*) p; + wld_dump_auxilary( av ); + + /* reserve memory that Wine needs */ + wld_reserve_wine_memory(); + + /* load the original dynamic loader itself */ + fd = open( LDSO_NAME, O_RDONLY, 0); + if( fd < 0 ) + { + wld_printf( "error: fd = %x\n", fd); + exit(1); + } + read( fd, buf, sizeof buf ); + interp[0] = 0; + start = wld_map_so( fd, buf, sizeof buf, av, &map ); + if( !start ) + exit(0); + close( fd ); + + wld_set_auxilary( av, AT_BASE, map.l_addr ); + +#ifdef DUMP_AUX_INFO + wld_printf("New auxillary info:\n"); + wld_dump_auxilary( av ); + wld_printf("jumping to %x\n", start); +#endif + + return start; +} + +/* + * The _start function is the entry and exit point of this program + * + * It calls wld_start, passing a pointer to the args it receives + * then jumps to the address wld_start returns after removing the + * first argv[] value, and decrementing argc + */ +void _start(void); +__asm ( + ".align 4\n" + "\t.global _start\n" + "\t.type _start,@function\n" +"_start:\n" + "\tcall wld_start\n" + "\tpush %eax\n" + "\txor %eax,%eax\n" + "\txor %ebx,%ebx\n" + "\txor %ecx,%ecx\n" + "\txor %edx,%edx\n" + "\tret\n" +); + --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.h 2004-04-10 21:43:23.000000000 +0900 @@ -0,0 +1,51 @@ +/* + mini loader for ld.so - x86 Linux 2.4.x+ kernels only + + Copyright 2004 Mike McCormack for Codeweavers + + Copyright (C) 1995,96,97,98,99,2000,2001,2002 Free Software Foundation, Inc. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. + */ + +#define LDSO_NAME "/lib/ld-linux.so.2" + +/* ELF definitions */ +#define ELF_PREFERRED_ADDRESS(loader, maplength, mapstartpref) (mapstartpref) +#define ELF_FIXED_ADDRESS(loader, mapstart) ((void) 0) + +#define __ELF_NATIVE_CLASS 32 +#define MAP_BASE_ADDR(l) 0 +#define MAP_COPY MAP_PRIVATE +#define ANONFD -1 + +#define ElfW(type) _ElfW (Elf, __ELF_NATIVE_CLASS, type) +#define _ElfW(e,w,t) _ElfW_1 (e, w, _##t) +#define _ElfW_1(e,w,t) e##w##t + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> +#include <sys/mman.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <fcntl.h> + +#define GL(x) x + +extern unsigned int GL(dl_pagesize); +

2 2

Re: Regression
by Ivan Leo Murray-Smith 10 Apr '04

10 Apr '04

> Updated to recent (GMT 21:07) CVS, still the same behavior. Apparently the '> fix' didn't really fix all of it, or maybe something else is broken? No, the regression was w a problem with the initial decompression of files, I think it's probably a different problem. Also, downgrading to wine-20030508 didn't help, so this bug is quite probably not related to the regression in any way. I'm not sure what kind of log would be useful. Ivan.

1 0

Re: spec file update [corrected]
by Dmitry Timoshkov 10 Apr '04

10 Apr '04

"Hans Leidekker" <hans(a)it.vu.nl> wrote: > Changelog: > Forward MsiLoadString* to user32.LoadString*. > Forward MsiMessageBox* to user32.MessageBox*. Are you sure that you correctly have figured out number of arguments for that APIs? I can't confirm that you are right. Also you had to remove old stub implementations for replaced entry points. -- Dmitry.

2 1

Re: winehq & 20040408
by Brian Vincent 10 Apr '04

10 Apr '04

> winehq still shows: > > Latest Release: > > Wine 20040309 (diff:s) > released: 2004/03/09 > > Nothing under "Latest News" either. I just updated Latest News, but I didn't update wine_release.template. I didn't want to step on Newman's toes. -Brian

1 0

winehq & 20040408
by Rein Klazes 10 Apr '04

10 Apr '04

Hi, winehq still shows: Latest Release: Wine 20040309 (diff:s) released: 2004/03/09 Nothing under "Latest News" either. Can somebody fix this? Rein. -- Rein Klazes rklazes(a)xs4all.nl

1 0

wine native linux executable /
by Arne Gellhaus 10 Apr '04

10 Apr '04

Hi, I am new to wine and i am trying to get some things done. 1. Is it possible to compile a native linux executable with winelib and use winapi functions like LoadLibrary() and GetProcAddress()? I made my buildsystem with winemaker. 2. I tried to use Common C++ [1] in my winelib application. But because of winelib/wineg++ (?) it assumes that it is running on windows and so the compilation failed. Is it possible to use such portable frameworks in a winelib application like in any other linux application? Thanks for helping me out, Arne Gellhaus [1] http://www.gnu.org/software/commoncpp/

6 9

Re: Security bug in ShellExecuteEx()
by Michael Stefaniuc 09 Apr '04

09 Apr '04

Hello, On Fri, Apr 09, 2004 at 02:14:10AM +0200, Francois Gouget wrote: > Index: dlls/shell32/shlexec.c > =================================================================== > RCS file: /var/cvs/wine/dlls/shell32/shlexec.c,v > retrieving revision 1.40 > diff -u -r1.40 shlexec.c > --- a/dlls/shell32/shlexec.c 7 Apr 2004 03:49:51 -0000 1.40 > +++ b/dlls/shell32/shlexec.c 8 Apr 2004 23:12:53 -0000 > @@ -1114,10 +1126,6 @@ > strcatW(wszApplicationName, wszCommandline); > } > > - retval = execfunc(wszApplicationName, NULL, FALSE, &sei_tmp, sei); > - if (retval > 32) > - return TRUE; > - > /* Else, try to find the executable */ ^^^^^ You may want to remove this because it refers to the just removed if. > wcmd[0] = '\0'; > retval = SHELL_FindExecutable(sei_tmp.lpDirectory, lpFile, sei_tmp.lpVerb, wcmd, 1024, lpstrProtocol, &env, sei_tmp.lpIDList, sei_tmp.lpParameters); bye michael -- Michael Stefaniuc Tel.: +49-711-96437-199 System Administration Fax.: +49-711-96437-111 Red Hat GmbH Email: mstefani(a)redhat.com Hauptstaetterstr. 58 http://www.redhat.de/ D-70178 Stuttgart

3 2

Wine-20040408 Compile Bug
by greensh＠knology.net 09 Apr '04

09 Apr '04

Had a problem compiling this latest version on my Dell Latitude laptop... here is the exact make error; gcc -c -I. -I. -I../../include -I../../include -I/usr/X11R6/include -D__WINESRC__ -DBINDIR="\"/usr/local/bin\"" -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o keyboard.o keyboard.c keyboard.c: In function `X11DRV_ToUnicodeEx': keyboard.c:2128: error: `XK_EuroSign' undeclared (first use in this function) keyboard.c:2128: error: (Each undeclared identifier is reported only once keyboard.c:2128: error: for each function it appears in.) make[2]: *** [keyboard.o] Error 1 make[2]: Leaving directory `/data/0MD2/wine-20040408/dlls/x11drv' make[1]: *** [x11drv] Error 2 make[1]: Leaving directory `/data/0MD2/wine-20040408/dlls' make: *** [dlls] Error 2 Compilation failed, aborting install. Looks like it's wrong in detecting the keyboard. A fix was to comment out the if statement entirely; < if (keysym == XK_EuroSign) < { < bufW[0] = 0x20AC; < ret = 1; < goto found; < } --- > /* > if (keysym == XK_EuroSign) > { > bufW[0] = 0x20AC; > ret = 1; > goto found; > } > /* Then it compiled/installed fine. Shaun

1 0

abnt2 keyboard fix
by Rafael Ávila de Espíndola 09 Apr '04

09 Apr '04

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have realised that the tables main_key_scan_abnt_qwerty and main_key_vkey_abnt_qwerty are both wrong. But one error compensates the other. For example, when I press x the message "Translating key VK_Z..." is displayed and an 'x' appears on the screen. The other problem with the abnt2 keyboard is that wine assumes that: /* Only the Keypad keys 0-9 and . send different keysyms * depending on the NumLock state */ This is not true for the abnt2 keyboard. The comma may also work as a delete. The attached patch solves both problems. I have tested the patch with a abnt2 and a us-int keyboard. Both work correctly. One last comment: as far as I know, all abnt2 keyboards have an "alt gr" key. How may I contact Mauro Carvalho Chehab and Raul Gomes Fernandes? Thanks Rafael Ávila de Espíndola -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAdrYZLlrfGJ8JUHwRAkxbAKCdrcsZ52jBWNtGqggmEzCKevhCgwCfclmx 0uNw50qMBFAp5HzZdIBtCjY= =NdP/ -----END PGP SIGNATURE-----

1 0

Serial Port Problem
by TN 09 Apr '04

09 Apr '04

Hi I already posted this on gmane.comp.emulators.wine.user but didn't get a response, maybe I have more luck here. I'm running wine-20040309 on a SuSE 8.2 system and trying to get Keil µVision2 Evaluation Version (Integrated Development Environment for Microcontrollers) working. So far, I've been pretty successful - most of it works. However, connecting a µC-Board at the serial port fails. The traces I get are as follows: fixme:comm:GetCommProperties (0xa0 0x408bf6f0 ) trace:comm:SetCommTimeouts (0xa0,0x42be02f8) trace:comm:SetCommMask handle 0xa0, mask 1 trace:comm:GetCommState handle 0xa0, ptr 0x42be02d8 trace:comm:GetCommState OK trace:comm:GetCommState bytesize 8 baudrate 9600 fParity 0 Parity 0 stopbits 1 trace:comm:GetCommState IXON ~IXOFF trace:comm:GetCommState CRTSCTS trace:comm:SetCommState handle 0xa0, ptr 0x42be02d8 trace:comm:SetCommState bytesize 8 baudrate 9600 fParity 0 Parity 0 stopbits 1 trace:comm:SetCommState IXON ~IXOFF trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:WaitCommEvent (0xa0 0x42f00180 0x42be0328 ) trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:COMM_WaitCommEventService iosb 0x42be0328 trace:comm:GetCommMask handle 0xa0, mask 0x42f00184 trace:comm:ClearCommError handle 0xa0 cbInQue = 1 cbOutQue = 0 err:file:GetOverlappedResult PENDING status after waiting! trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:PurgeComm handle 0xa0, flags f trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:ClearCommError handle 0xa0 cbInQue = 0 cbOutQue = 0 trace:comm:WaitCommEvent (0xa0 0x42f00180 0x42be0328 ) Any help would be greatly appreciated. Thanks. Regards, Thomas

2 1

inheriting exec-shield avoidance
by Peter Riocreux 09 Apr '04

09 Apr '04

I am intermittently trying to get Wine to play nicely with a big EDA tool, and it is doing not too badly today with the 20040309 snapshot on Fedora Core 1. The thing that I think is stopping it working is that whatever is done to stack-shield by the prepending "setarch i386" on invocation is not inherited by the .exe that the program calls. Net result is that I get the "..... security-patched kernel ?" error message from the child process instead of the parent. Progress of a sort. Is there anything, short of turning off exec-shield entirely, that can be done about this? The full set of messages it emits (with program identities obscured to avoid identifying the much-cheaper-when-bought-as-windows-binary application) is below. The printer message mentions an extant queue. ----8<--------8<--------8<--------8<--------8<--------8<---- err:dosmem:setup_dos_mem Cannot use first megabyte for DOS address space, please report err:dosmem:setup_dos_mem Cannot use first megabyte for DOS address space, please report err:dosmem:setup_dos_mem Cannot use first megabyte for DOS address space, please report fixme:console:SetConsoleCtrlHandler (0x41a8af,1) - no error checking or testing yet err:comm:GetCommState tcgetattr error 'Invalid argument' fixme:commdlg:PRINTDLG_OpenDefaultPrinter Could not open printer pica_duplex?! fixme:console:SetConsoleCtrlHandler (0x7800f03e,1) - no error checking or testing yet err:virtual:map_image Standard load address for a Win32 program (0x00400000) not available - security-patched kernel ? wine: could not load L"G:\\path\\to\\my\\child.exe" as Win32 binary ----8<--------8<--------8<--------8<--------8<--------8<---- Peter

4 8

Re: CryptoAPI
by dataphile 09 Apr '04

09 Apr '04

I read in one of the list archives that someone was working on implementing CryptoAPI functionality via libgcrypt on the backend. Anyone have any more info on this?

1 0

Re: wine/dlls/msvcrt math.c
by Hans Leidekker 09 Apr '04

09 Apr '04

On Thursday 08 April 2004 21:49, Alexandre Julliard wrote: > Modified files: > dlls/msvcrt : math.c > > Log message: > Portability fix. > > Patch: http://cvs.winehq.org/patch.py?id=11915 This isn't needed anymore. MinGW has accepted my patch that adds the defines to their header. I think that's the proper fix: they export fpclass() so they should export the corresponding defines for it's return values too. -Hans

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel April 2004