April 2004 - Wine-Devel - list.winehq.org

Re: help fixing up SetConsoleCtrlHandler()
by Juan Lang April 14, 2004

April 14, 2004

Hi Peter, Peter wrote: > 1) If I am using SetLastError, it is not obvious to > me what error values I should be using there. They should be a Win32 error message, the ones defined in winerror.h that aren't HRESULTs (so, lowish values, less then 32000 or so, not 0x80070002-ish). > 2) Can I assume that > Rtl{Enter,Leave}CriticalSection(), > HeapFree(), ERR() and WARN() cannot fail (the latter > two are presumably macros). Mostly yes. If your critical section hasn't been initialized or has been deleted, {Enter,Leave} CriticalSection can hang or otherwise do weird things, but that won't happen "normally". > Should I, either in general, or in this case in > particular, detect which Windows version is being > used and behave appropriately No--write it assuming the less restrictive case. A common way to get partway there is to handle the more restrictive (easier) case, and issue a FIXME and return an error if the caller passes NULL. --Juan __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html

1 0

Latest Wine with Corel Office 9
by David Lee Lambert April 14, 2004

April 14, 2004

I just got around to patching and compiling the '0408 release of Wine, and I notice a few improvements. I tried it out with Corel PerfectOffice 9, of which my family has several copies. Here are my results: WordPerfect hung at the splash screen. Presentations loaded up, but the menu was empty except for "File", and the "Open" item didn't appear to do anything, so I used "Exit" successfully. Quattro Pro quit after a few seconds of the splash screen, giving about a screen of errors: I await thy bidding $ wine qpw.exe err:keyboard:X11DRV_ToUnicodeEx Please report: no char for keysym 0000 (No Name) : err:keyboard:X11DRV_ToUnicodeEx (virtKey=0,scanCode=0,keycode=8,state=0) err:keyboard:X11DRV_ToUnicodeEx Please report: no char for keysym 0000 (No Name) : err:keyboard:X11DRV_ToUnicodeEx (virtKey=0,scanCode=0,keycode=8,state=1) err:keyboard:X11DRV_ToUnicodeEx Please report: no char for keysym FE20 (ISO_Left_Tab) : err:keyboard:X11DRV_ToUnicodeEx (virtKey=9,scanCode=F,keycode=17,state=1) err:keyboard:X11DRV_ToUnicodeEx Please report: no char for keysym FEF9 (Pointer_EnableKeys) : err:keyboard:X11DRV_ToUnicodeEx (virtKey=90,scanCode=45,keycode=4D,state=1) err:keyboard:X11DRV_ToUnicodeEx Please report: no char for keysym 0000 (No Name) : err:keyboard:X11DRV_ToUnicodeEx (virtKey=FC,scanCode=0,keycode=0,state=0) err:keyboard:X11DRV_ToUnicodeEx Please report: no char for keysym 0000 (No Name) : err:keyboard:X11DRV_ToUnicodeEx (virtKey=FC,scanCode=0,keycode=0,state=1) fixme:ole:CoRegisterMessageFilter stub fixme:tab:TAB_WindowProc Unimplemented msg TCM_SETEXTENDEDSTYLE fixme:accel:CreateAcceleratorTableA should check that the accelerator descriptions are valid, return NULL and SetLastError() if not. fixme:ole:CoCreateInstance no classfactory created for CLSID {eab38d25-26dc-11d2-98c0-00104b24170b}, hres is 0x80040154 fixme:storage:StgCreateDocfile Transacted mode not implemented. err:seh:setup_exception stack overflow 0 bytes in thread 0034 eip 401bfc99 esp 40851000 stack 0x40850000-0x40950000 I thought I ought to report the "Please Report" messages; WordPerfect gave similar ones. Is anyone planning to implement CreateAccelleratorTable and related functions any time soon? If not, is there any reason I shouldn't try it? -- resume at http://www.cse.msu.edu/~lamber45/resume.htm PGP key at http://www.cse.msu.edu/~lamber45/newmail.htm#GPGKey

1 0

exec-shield workaround take 4
by Mike McCormack April 14, 2004

April 14, 2004

Hi again, Here's a fix for the bug that Peter found. The calculation of the reserve area was correct, however the environment was not being cleared correctly, so memory was being reserved for the wrong .exe file. Anybody have a FreeBSD machine to test this on? Mike ----------------------------------------------------------------------- Hi, Here's a new patch for exec-shield. Changes against the last patch: * reserve only the area required by the PE EXE if launched from CreateProcess (else just a default slab) * pass the reserved area to wine, so wine knows about it * unmap the preloader after it is no longer used Again, comments, improvements, flames and test reports are all welcome! Mike ----------------------------------------------------------------------- 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 1.51 diff -u -r1.51 process.c --- dlls/kernel/process.c 12 Apr 2004 22:12:43 -0000 1.51 +++ dlls/kernel/process.c 14 Apr 2004 08:45:09 -0000 @@ -1059,7 +1059,8 @@ * * 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 char *unixpath ) { const WCHAR *p; char **envp; @@ -1082,7 +1083,7 @@ if (!(env = malloc( length ))) return NULL; WideCharToMultiByte( CP_UNIXCP, 0, envW, p - envW, env, length, NULL, NULL ); - count += 4; + count += 6; if ((envp = malloc( count * sizeof(*envp) ))) { @@ -1097,6 +1098,11 @@ 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 ); + if (unixpath) *envptr++ = alloc_env_string( "WINEPEEXEPATH=", unixpath ); + /* now put the Windows environment strings */ for (p = env; *p; p += strlen(p) + 1) { @@ -1105,6 +1111,8 @@ if (is_special_env_var( p )) /* prefix it with "WINE" */ *envptr++ = alloc_env_string( "WINE", p ); else if (strncmp( p, "HOME=", 5 ) && + strncmp( p, "WINEWORKDIR=", 12 ) && + strncmp( p, "WINEPEEXEPATH=", 14 ) && strncmp( p, "WINEPREFIX=", 11 )) *envptr++ = p; } *envptr = 0; @@ -1135,7 +1143,7 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmdline, 0 ); - char **envp = build_envp( env, NULL ); + char **envp = build_envp( env, NULL, NULL, NULL ); close( fd[0] ); /* Reset signals that we previously set to SIG_IGN */ @@ -1170,14 +1178,8 @@ RTL_USER_PROCESS_PARAMETERS *params; UNICODE_STRING image_str, cmdline_str, desktop, title; NTSTATUS status; - WCHAR buffer[MAX_PATH]; - - if(!GetLongPathNameW( filename, buffer, MAX_PATH )) - lstrcpynW( buffer, filename, MAX_PATH ); - if(!GetFullPathNameW( buffer, MAX_PATH, buffer, NULL )) - lstrcpynW( buffer, filename, MAX_PATH ); - RtlInitUnicodeString( &image_str, buffer ); + RtlInitUnicodeString( &image_str, filename ); RtlInitUnicodeString( &cmdline_str, cmdline ); if (startup->lpDesktop) RtlInitUnicodeString( &desktop, startup->lpDesktop ); if (startup->lpTitle) RtlInitUnicodeString( &title, startup->lpTitle ); @@ -1229,6 +1231,7 @@ pid_t pid; int err; char dummy = 0; + WCHAR buffer[MAX_PATH]; if (!env) { @@ -1236,7 +1239,12 @@ extra_env = DRIVE_BuildEnv(); } - if (!(params = create_user_params( filename, cmd_line, startup ))) + if(!GetLongPathNameW( filename, buffer, MAX_PATH )) + lstrcpynW( buffer, filename, MAX_PATH ); + if(!GetFullPathNameW( buffer, MAX_PATH, buffer, NULL )) + lstrcpynW( buffer, filename, MAX_PATH ); + + if (!(params = create_user_params( buffer, cmd_line, startup ))) { if (extra_env) HeapFree( GetProcessHeap(), 0, extra_env ); return FALSE; @@ -1267,7 +1275,20 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmd_line, 1 ); - char **envp = build_envp( env, extra_env ); + char **envp; + CHAR unixpath[MAX_PATH]; + + unixpath[0]=0; + wine_get_unix_file_name( buffer, unixpath, MAX_PATH ); + + /* 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, unixpath ); close( startfd[1] ); close( execfd[0] ); @@ -1279,8 +1300,6 @@ /* 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 14 Apr 2004 08:45:09 -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 14 Apr 2004 08:45:09 -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 main.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 14 Apr 2004 08:45:09 -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 14 Apr 2004 08:45:09 -0000 @@ -20,19 +20,44 @@ #include <stdio.h> #include <stdlib.h> +#include <unistd.h> +#include <sys/mman.h> #include "wine/library.h" +#include "main.h" + +/* this variable is set by the wine-specific shared object loader */ +struct wine_preload_info *wine_preload_vars; + /********************************************************************** * main */ int main( int argc, char *argv[] ) { - char error[1024]; + char error[1024], *dir; + + if( wine_preload_vars ) + { + /* printf("vars(*%p) = %p -> (%p, %08x) (%p, %08x)\n", + &wine_preload_vars, wine_preload_vars, + wine_preload_vars->pe_address, + wine_preload_vars->pe_size, + wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); */ + + /* let wine know which area was mapped */ + wine_set_pe_load_area(wine_preload_vars->pe_address, + wine_preload_vars->pe_size); + + /* unmap the preloader itself */ + munmap( wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); + wine_preload_vars = NULL; + } -#if 0 - static char pe_load[256*1024*1024] __attribute__((aligned(4096))); - wine_set_pe_load_area( pe_load, sizeof(pe_load) ); -#endif + 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/main.h 2004-04-11 16:58:25.000000000 +0900 @@ -0,0 +1,8 @@ + +struct wine_preload_info +{ + void *pe_address; + size_t pe_size; + void *preload_base; + size_t preload_size; +}; --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.h 2004-04-11 16:58:25.000000000 +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); + --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.c 2004-04-11 16:58:25.000000000 +0900 @@ -0,0 +1,727 @@ +/* + 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 + * + * This program will be invoked by linking it as the shared object loader + * using the link time flags --shared-loader=ld-winepreload.so . Since we + * are not using an absolute path, an executable using this shared object + * loader must be invoked from the directory that contains this loader. + * + * 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 + * + * If this program is used as the shared object loader. the only difference + * that the loaded programs should see is that this loader will be mapped + * into memory when it starts. + * + * Limitations: + * * using a non-absolution path for the shared object loader is not standard + */ + +/* + * 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 "main.h" +#include <elf.h> + +/* the following headers are needed to get the definition for IMAGE_NT_HEADERS */ +#include <stdarg.h> +#include "windef.h" +#include "winbase.h" + +/* debugging */ +#undef DUMP_SEGMENTS +#undef DUMP_ARG +#undef DUMP_AUX_INFO +#undef DUMP_SYMS + +unsigned int GL(dl_pagesize) = 0x0000; +extern const unsigned char _end[]; + +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; +} + +#ifdef DUMP_AUX_INFO +/* + * Dump interesting bits of the ELF auxv_t structure that is passed + * as the 4th parameter to the _start function + */ +static void wld_dump_auxilary( ElfW(auxv_t) *av ) +{ + for ( ; av->a_type != AT_NULL; av++) + switch (av->a_type) + { + case AT_PAGESZ: + 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_get_auxilary + * + * Get a field of the auxillary structure + */ +int wld_get_auxilary( ElfW(auxv_t) *av, int type, int *val ) +{ + for ( ; av->a_type != AT_NULL; av++) + if( av->a_type == type ) + { + *val = av->a_un.a_val; + return 1; + } + return 0; +} + +/* + * 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, 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; +} + +#define ROUND_SIZE(addr,size) \ + (((int)(size) + ((int)(addr) & page_mask) + page_mask) & ~page_mask) + +/* + * 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( char *pe_file, void **base, size_t *size ) +{ + IMAGE_DOS_HEADER dos; + IMAGE_NT_HEADERS nt; + int unix_fd, page_mask = GL(dl_pagesize) - 1; + + if( !pe_file ) + return; + + unix_fd = open( pe_file, O_RDONLY, 0); + if( unix_fd < 0 ) + return; + + /* + * The following code is borrowed from + * get_image_params() in wine/server/mapping.c + */ + if (read( unix_fd, &dos, sizeof(dos) ) != sizeof(dos)) goto error; + if (dos.e_magic != IMAGE_DOS_SIGNATURE) goto error; + if (lseek( unix_fd, dos.e_lfanew, SEEK_SET ) == -1) goto error; + + if (read( unix_fd, &nt.Signature, sizeof(nt.Signature) ) != sizeof(nt.Signature)) goto error; + if (nt.Signature != IMAGE_NT_SIGNATURE) goto error; + if (read( unix_fd, &nt.FileHeader, sizeof(nt.FileHeader) ) != sizeof(nt.FileHeader)) goto error; + /* zero out Optional header in the case it's not present or partial */ + memset(&nt.OptionalHeader, 0, sizeof(nt.OptionalHeader)); + if (read( unix_fd, &nt.OptionalHeader, nt.FileHeader.SizeOfOptionalHeader) != nt.FileHeader.SizeOfOptionalHeader) goto error; + + *size = ROUND_SIZE( 0, nt.OptionalHeader.SizeOfImage ); + *base = (void *)nt.OptionalHeader.ImageBase; + +#ifdef DUMP_PERESERVE + wld_printf( "parsed PE image... base %x size %x\n", *base, *size ); +#endif + +error: + close( unix_fd ); +} + +/* + * Find a symbol in the symbol table of the executable loaded + */ +void *wld_find_symbol( const ElfW(Phdr) *phdr, int num, char *var ) +{ + const ElfW(Dyn) *dyn = NULL; + void *ptr = NULL; + const ElfW(Phdr) *ph; + const ElfW(Sym) *symtab = NULL; + const char *strings = NULL; + + /* check the values */ +#ifdef DUMP_SYMS + wld_printf("%x %x\n", phdr, num ); +#endif + if( ( phdr == NULL ) || ( num == 0 ) ) + { + wld_printf("could not find PT_DYNAMIC header entry\n"); + return NULL; + } + + /* parse the (already loaded) ELF executable's header */ + for (ph = phdr; ph < &phdr[num]; ++ph) + { + if( PT_DYNAMIC == ph->p_type ) + { + dyn = (void *) ph->p_vaddr; + num = ph->p_memsz / sizeof (Elf32_Dyn); + break; + } + } + if( !dyn ) + { + wld_printf("No PT_DYNAMIC section\n"); + return ptr; + } + + while( dyn->d_tag ) + { + if( dyn->d_tag == DT_STRTAB ) + strings = (const char*) dyn->d_un.d_ptr; + if( dyn->d_tag == DT_SYMTAB ) + symtab = (const ElfW(Sym) *)dyn->d_un.d_ptr; +#ifdef DUMP_SYMS + wld_printf("%x %x\n", dyn->d_tag, dyn->d_un.d_ptr ); +#endif + dyn++; + } + + if( (!symtab) || (!strings) ) + { + wld_printf("No symbol table found\n"); + return ptr; + } + + /* FIXME: this crashes if it doesn't find the symbol we want */ + do + { + if( ( ELF32_ST_BIND(symtab->st_info) == STT_OBJECT ) && + ( 0 == strcmp( strings+symtab->st_name, var ) ) ) + { + ptr = (void*)symtab->st_value; + break; + } + symtab++; + } + while( symtab->st_name ); + +#ifdef DUMP_SYMS + wld_printf("Found %s -> %x\n", strings+symtab->st_name, symtab->st_value ); +#endif + + return ptr; +} + +/* + * 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, phnum = 0; + unsigned char buf[0x800]; + void *start; + char **argv = (char**) &argc, **p, *pe_file = NULL; + struct link_map map; + ElfW(auxv_t)* av; + ElfW(Phdr) *elf_phdr = NULL; + struct wine_preload_info **wine_preload_vars; + static struct wine_preload_info vars; + + 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, find the name of the PE file */ + while( *p ) + { + const char pefile[] = "WINEPEEXEPATH="; + if( 0 == memcmp( pefile, *p, sizeof pefile - 1 ) ) + pe_file = (*p) + sizeof pefile - 1 ; + p++; + } + + av = (ElfW(auxv_t)*) (p+1); + wld_get_auxilary( av, AT_PAGESZ, &dl_pagesize ); +#ifdef DUMP_AUX_INFO + wld_dump_auxilary( av ); +#endif + + /* reserve memory that Wine needs */ + vars.pe_size = 0x01000000; + vars.pe_address = (void*)0x00400000; + wld_reserve_wine_memory( pe_file, &vars.pe_address, &vars.pe_size ); + mmap( vars.pe_address, vars.pe_size, PROT_NONE, + MAP_FIXED|MAP_PRIVATE|MAP_ANON, -1, 0); + + /* 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 ); + start = wld_map_so( fd, buf, sizeof buf, &map ); + if( !start ) + exit(0); + close( fd ); + + /* + * Pass information to the wine loader in a static variable + * so it doesn't get destroyed after we leave this stack frame + */ + wld_get_auxilary( av, AT_PHDR, (int*)&elf_phdr ); + wld_get_auxilary( av, AT_PHNUM, (int*)&phnum ); + wine_preload_vars = wld_find_symbol( elf_phdr, phnum, "wine_preload_vars" ); + if( wine_preload_vars ) + { + unsigned char *base = (void*)0x7ff81000; /* FIXME - define properly */ + vars.preload_base = (void*) base; + vars.preload_size = (size_t) (_end - base); + *wine_preload_vars = &vars; + } + + 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" +); +

2 2

Re: Remove "Path" lines from sample config
by Saulius Krasuckas April 14, 2004

April 14, 2004

> Changelog: > Remove "Path" lines from sample config. > Setup the links for the default DOS drives in wineinstall. hey, it is plain and simple change but it makes me disappointed right after second reinstall of Wine. GNU ln when encounters existing target (link to directory, in fact) creates a sym-link with the name of original target already in mentioned symlinked dir. that we are getting is like a: ln: `/home/s2/.wine/dosdevices/c:/c': File exists and this: lrwxrwxrwx 1 s2 s2 10 Apr 14 15:36 /home/s2/.wine/dosdevices/c: -> /home/s2/c lrwxrwxrwx 1 s2 s2 10 Apr 14 15:58 /home/s2/.wine/dosdevices/c:/c -> /home/s2/c as that does not belong to Wine binary (it is more to config of Wine-user), i'd not like to change existing dosdevices/ symlinks during "make uninstall" or even "tools/wineinstall". OK, maybe target should only be removed if it is not a symlink. so, what about some additional small bash function? Index: wine/tools/wineinstall =================================================================== RCS file: /home/wine/wine/tools/wineinstall,v retrieving revision 1.67 diff -u -r1.67 wineinstall --- wine/tools/wineinstall 12 Apr 2004 22:05:00 -0000 1.67 +++ wine/tools/wineinstall 14 Apr 2004 14:54:05 -0000 @@ -140,6 +140,16 @@ link_app winebrowser "$CROOT/windows/winebrowser.exe" } +function link_file_safely { + # makes link to target if link does not exist already + # parameters: + # $1 - path to target, + # $2 - path to link name + if [ -L $2 ]; then return 1; fi + if [ -e $2 ]; then return 2; fi + ln -s $1 $2 +} + # startup... echo "WINE Installer v0.74" @@ -600,12 +610,12 @@ echo "Preparing to install default Wine registry entries..." # Make sure we are on a Windows drive - ln -s /mnt/fd0 ~/.wine/dosdevices/a: - ln -s $CROOT ~/.wine/dosdevices/c: - ln -s /cdrom ~/.wine/dosdevices/d: - ln -s /tmp ~/.wine/dosdevices/e: - ln -s ~ ~/.wine/dosdevices/f: - ln -s / ~/.wine/dosdevices/z: + link_file_safely /mnt/fd0 ~/.wine/dosdevices/a: + link_file_safely $CROOT ~/.wine/dosdevices/c: + link_file_safely /cdrom ~/.wine/dosdevices/d: + link_file_safely /tmp ~/.wine/dosdevices/e: + link_file_safely ~ ~/.wine/dosdevices/f: + link_file_safely / ~/.wine/dosdevices/z: echo "Installing default Wine registry entries..." echo PS and yes, that double check in "link_file_safely" function (one for existing target/link with a given name and second about target being some file of other kind) may be overhead. PPS IMHO it could be nice to eliminate overhead of multiple passing string "~/.wine/dosdevices/" by assigning its value to some script variable and by putting out that block into separate function, say some "mk_dosdevices" or like. ughm?

1 0

Re: exec-shield workaround take 4
by Mike Hearn April 14, 2004

April 14, 2004

On Wed, 14 Apr 2004 18:44:45 +0900, Mike McCormack wrote: > Anybody have a FreeBSD machine to test this on? If it still uses syscalls surely that won't work on anything other than Linux?

2 1

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 14, 2004

April 14, 2004

Thanks for the feedback. Excellent catches! Hope my next mail to wine-patches gets through! Regards Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Tuesday, April 13, 2004 7:38 PM > To: Santosh Siddheshwar; Robert Shearman; wine-devel(a)winehq.org > Cc: santosh.s(a)sonata-software.com > Subject: Re: Handling dialog messages for non-template based dialogs > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > I didnt get the second point that you mention > > Another one might be that WM_CREATE must be always > > passed to DEFDLG_Proc regardless of what result an application > > returned to DefDlgProc16/A/W. > > Here is a snippet of DefDlgProc16 (DefDlgProcA/W are the same): > > LRESULT WINAPI DefDlgProc16( HWND16 hwnd, UINT16 msg, WPARAM16 wParam, > LPARAM lParam ) > { > WNDPROC16 dlgproc; > HWND hwnd32 = WIN_Handle32( hwnd ); > BOOL result = FALSE; > > ===== > Perhaps WM_CREATE handling code should be moved here: > if (msg == WM_CREATE) > { > wndPtr->flags |= WIN_ISDIALOG; > blah-blah-blah > } > ===== > > SetWindowLongW( hwnd32, DWL_MSGRESULT, 0 ); > > if ((dlgproc = (WNDPROC16)DEFDLG_GetDlgProc( hwnd32 ))) > { > /* Call dialog procedure */ > result = CallWindowProc16( dlgproc, hwnd, msg, wParam, lParam ); > /* 16 bit dlg procs only return BOOL16 */ > if( WINPROC_GetProcType( (WNDPROC)dlgproc ) == WIN_PROC_16 ) > result = LOWORD(result); > } > > if (!result && IsWindow(hwnd32)) <== here we check whether an app > handled the message > { > /* callback didn't process this message */ > > switch(msg) > { > case WM_ERASEBKGND: > > Probably WM_CREATE handling should be performed even before control > is passed to an application. > > -- > 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. *********************************************************************

2 1

Game support
by James Perry April 14, 2004

April 14, 2004

Hi everyone, I've been following Wine development for a while but now I think it's time to contribute something. I don't use Windows anymore (except for PowerPoint at work), but I decided it would be good to be able to play some games on Linux. So I grabbed Wine from CVS, compiled it up and tested some games, with the intention of finding where the problems were and hopefully fixing them. I have put my results on a web page at: http://www.epcc.ed.ac.uk/~jamesp/wine-games.html to avoid making this posting even longer. Note that I haven't looked in detail at most of these yet, I just tried them out once to see what would happen. I'm gradually working through the list, debugging them. I've already submitted one patch (to solve a D3D blend mode problem that affected SonicR). I have traced a number of the other problems to specific aspects of Wine, but I'm not sure how to go about fixing them. Any advice on these points from experienced developers would be welcome: 1. MusicVR exits with this error: X Error of failed request: GLXBadDrawable Major opcode of failed request: 145 (GLX) Minor opcode of failed request: 11 (X_GLXSwapBuffers) Serial number of failed request: 10522 Current serial number in output stream: 10523 I traced it to the line: glXMakeCurrent(default_display, None, NULL) in dlls/opengl32/wgl.c. My OpenGL (nVidia) doesn't seem to like the drawable being set to None, although according to the GLX docs, this should be OK. Simply commenting out the line makes the game work fine, but I don't think this is the right fix. Any ideas? 2. Settlers III crashes almost straight away, in the code of the PE-Crypt protection system. This is caused by the processor's direction flag (which controls whether pointers are incremented or decremented by string instructions) getting reset somewhere, probably in Wine code. Can anyone comment on treatment of this flag in Wine? Is it likely to be clobbered in lots of places? Basically, is there any hope of getting this working? I'm not sure where to start with this one. 3. South Park Rally creates its DirectInput object using the generic OLE functions, which doesn't work in Wine right now. I have some initial code to support this, but not sure if it does everything it needs to. I have: - copied regsvr.c from the DirectSound DLL and edited it to register the DirectInput classes, and implement the DirectInput Dll[Un]RegisterServer methods - removed the stubs of those functions from dinput_main.c - added regsvr.c to the list of source files in dinput's Makefile.in, and added ole32 and advapi32 to the list of imports It would also be good if the default registry created by winebuild contained the necessary entries to make this work (I had to add them by hand), but I couldn't find where the initial registry entries come from. If this all sounds OK, I can submit it as a patch. Although even with this support, South Park Rally fails with another DirectX problem. One step at a time though... 4. The CD audio in Sonic 3D doesn't play in Wine yet. I found there were two separate problems causing this. First, it's trying to use CD device 2. No idea where this number comes from, as I have only one CD-ROM in the system, and the game doesn't seem to be specifying device 2 anywhere. Anyone know where mcicda gets the device ID from, save me spending ages looking for it? Second, it's using TMSF (track,minutes,seconds,frames) format to specify locations on the CD. These appear as colon-separated values in the MCI commands (e.g. "02:00:00") but the argument parser chokes on them, expecting a single integer. I got around this by hacking MCI_GetDWord to recognise the colons and parse these arguments correctly, but maybe that's not the way to do it... any suggestions? Sorry this post is so long. Finally, I'd just like to say great job everyone! I was impressed that things like DirectX, InstallShield and even SecuROM copy protection are working so well - they're flawless on some of the games I tried. It's really come on since I first tried it back in 2000. Cheers, James

6 12

Re: shell32 - file not found
by Uwe Bonnes April 14, 2004

April 14, 2004

>>>>> "Martin" == Martin Fuchs <martin-fuchs(a)gmx.net> writes: Martin> Changelog: return "file not found" error values instead of Martin> E_INVALIDARG if _ILCreateFromPathA() failed to find files ... Martin> 0x80070002L; /* file not found */ } else { pidlTemp = ... Don't we have a symbolic error for that? -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

2 1

(no subject)
by j.cooper1＠comcast.net April 13, 2004

April 13, 2004

I am trying to compile Wine (the latest CVS) on Mac os x. If I emailed the wrong list sorry I noticed what I think is a bug in file dll/kernel/console.c around line 1003 exists function below when compiling it gives me an error about x being undefined or something similar COORD WINAPI GetLargestConsoleWindowSize(HANDLE hConsoleOutput) { COORD c; c.X = 80; c.Y = 24; TRACE("(%p), returning %dx%d\n", hConsoleOutput, c.X, x.Y); return c; } #endif /* defined(__i386__) */ Changing the x.Y to c.X seems to fix the problem COORD WINAPI GetLargestConsoleWindowSize(HANDLE hConsoleOutput) { COORD c; c.X = 80; c.Y = 24; TRACE("(%p), returning %dx%d\n", hConsoleOutput, c.X, x.Y); return c; } I am not a programmer nor do I know how to create a patch or if I am correct in my assumptions. Could some one please help me out? Jeff ps.. I now get an error compiling dbghelp. I am not on the computer I am compiling on so I can't copy and paste the error. If there are typos in the error that is because I am copying the code by typing it up elf_module.c:627: error: previous declaration of 'elf_read_wine_loader_dbg_info' dbghelp_private.h:275: error: previous declaration of 'elf_read_wine_loader_dbg_info' elf module.c:71: warning: '__wine_dbch___default' defined but not used make[2]: *** [elf_module.o] Error 1 make[1]: *** [dbghelp] Error 2 make: *** [dlls] Error 2

2 1

RE: CryptoAPI
by Juan Lang April 13, 2004

April 13, 2004

Hi Robert, Robert wrote: > It depends on which part you are trying to implement. > For the Certificate Store functions I would recommend > implementing the "OID support functions" first, I was taking a look at implementing the certificate store functions, and I was thinking (hoping?) I could avoid the OID stuff. Maybe I'm missing something, but a cert will be stored in the registry in some well-known format, and as long as Wine handles the registry I/O bit, OpenSSL ought to be able to read the certs, right? So CertOpenStore/CertEnumStore et al wouldn't need to do any asn.1 decoding, at least not from the get-go. What did I miss? --Juan __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/

2 3

exec-shield workaround take 3
by Mike McCormack April 13, 2004

April 13, 2004

Hi, Here's a new patch for exec-shield. Changes against the last patch: * reserve only the area required by the PE EXE if launched from CreateProcess (else just a default slab) * pass the reserved area to wine, so wine knows about it * unmap the preloader after it is no longer used Again, comments, improvements, flames and test reports are all welcome! Mike ----------------------------------------------------------------------- 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 11 Apr 2004 07:14:20 -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 char *unixpath ) { 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 += 6; if ((envp = malloc( count * sizeof(*envp) ))) { @@ -1093,6 +1094,11 @@ 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 ); + if (unixpath) *envptr++ = alloc_env_string( "WINEPEEXEPATH=", unixpath ); + /* now put the Windows environment strings */ for (p = env; *p; p += strlen(p) + 1) { @@ -1131,7 +1137,7 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmdline, 0 ); - char **envp = build_envp( env, NULL ); + char **envp = build_envp( env, NULL, NULL, NULL ); close( fd[0] ); /* Reset signals that we previously set to SIG_IGN */ @@ -1166,14 +1172,8 @@ RTL_USER_PROCESS_PARAMETERS *params; UNICODE_STRING image_str, cmdline_str, desktop, title; NTSTATUS status; - WCHAR buffer[MAX_PATH]; - - if(!GetLongPathNameW( filename, buffer, MAX_PATH )) - lstrcpynW( buffer, filename, MAX_PATH ); - if(!GetFullPathNameW( buffer, MAX_PATH, buffer, NULL )) - lstrcpynW( buffer, filename, MAX_PATH ); - RtlInitUnicodeString( &image_str, buffer ); + RtlInitUnicodeString( &image_str, filename ); RtlInitUnicodeString( &cmdline_str, cmdline ); if (startup->lpDesktop) RtlInitUnicodeString( &desktop, startup->lpDesktop ); if (startup->lpTitle) RtlInitUnicodeString( &title, startup->lpTitle ); @@ -1225,6 +1225,7 @@ pid_t pid; int err; char dummy = 0; + WCHAR buffer[MAX_PATH]; if (!env) { @@ -1232,7 +1233,12 @@ extra_env = DRIVE_BuildEnv(); } - if (!(params = create_user_params( filename, cmd_line, startup ))) + if(!GetLongPathNameW( filename, buffer, MAX_PATH )) + lstrcpynW( buffer, filename, MAX_PATH ); + if(!GetFullPathNameW( buffer, MAX_PATH, buffer, NULL )) + lstrcpynW( buffer, filename, MAX_PATH ); + + if (!(params = create_user_params( buffer, cmd_line, startup ))) { if (extra_env) HeapFree( GetProcessHeap(), 0, extra_env ); return FALSE; @@ -1263,7 +1269,20 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmd_line, 1 ); - char **envp = build_envp( env, extra_env ); + char **envp; + CHAR unixpath[MAX_PATH]; + + unixpath[0]=0; + wine_get_unix_file_name( buffer, unixpath, MAX_PATH ); + + /* 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, unixpath ); close( startfd[1] ); close( execfd[0] ); @@ -1275,8 +1294,6 @@ /* 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 11 Apr 2004 07:14:21 -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 11 Apr 2004 07:14:21 -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 main.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 11 Apr 2004 07:14:21 -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 11 Apr 2004 07:14:21 -0000 @@ -20,19 +20,44 @@ #include <stdio.h> #include <stdlib.h> +#include <unistd.h> +#include <sys/mman.h> #include "wine/library.h" +#include "main.h" + +/* this variable is set by the wine-specific shared object loader */ +struct wine_preload_info *wine_preload_vars; + /********************************************************************** * main */ int main( int argc, char *argv[] ) { - char error[1024]; + char error[1024], *dir; + + if( wine_preload_vars ) + { + /* printf("vars(*%p) = %p -> (%p, %08x) (%p, %08x)\n", + &wine_preload_vars, wine_preload_vars, + wine_preload_vars->pe_address, + wine_preload_vars->pe_size, + wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); */ + + /* let wine know which area was mapped */ + wine_set_pe_load_area(wine_preload_vars->pe_address, + wine_preload_vars->pe_size); + + /* unmap the preloader itself */ + munmap( wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); + wine_preload_vars = NULL; + } -#if 0 - static char pe_load[256*1024*1024] __attribute__((aligned(4096))); - wine_set_pe_load_area( pe_load, sizeof(pe_load) ); -#endif + 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/main.h 2004-04-11 16:58:25.000000000 +0900 @@ -0,0 +1,8 @@ + +struct wine_preload_info +{ + void *pe_address; + size_t pe_size; + void *preload_base; + size_t preload_size; +}; --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.h 2004-04-11 16:58:25.000000000 +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); + --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.c 2004-04-11 16:58:25.000000000 +0900 @@ -0,0 +1,727 @@ +/* + 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 + * + * This program will be invoked by linking it as the shared object loader + * using the link time flags --shared-loader=ld-winepreload.so . Since we + * are not using an absolute path, an executable using this shared object + * loader must be invoked from the directory that contains this loader. + * + * 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 + * + * If this program is used as the shared object loader. the only difference + * that the loaded programs should see is that this loader will be mapped + * into memory when it starts. + * + * Limitations: + * * using a non-absolution path for the shared object loader is not standard + */ + +/* + * 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 "main.h" +#include <elf.h> + +/* the following headers are needed to get the definition for IMAGE_NT_HEADERS */ +#include <stdarg.h> +#include "windef.h" +#include "winbase.h" + +/* debugging */ +#undef DUMP_SEGMENTS +#undef DUMP_ARG +#undef DUMP_AUX_INFO +#undef DUMP_SYMS + +unsigned int GL(dl_pagesize) = 0x0000; +extern const unsigned char _end[]; + +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; +} + +#ifdef DUMP_AUX_INFO +/* + * Dump interesting bits of the ELF auxv_t structure that is passed + * as the 4th parameter to the _start function + */ +static void wld_dump_auxilary( ElfW(auxv_t) *av ) +{ + for ( ; av->a_type != AT_NULL; av++) + switch (av->a_type) + { + case AT_PAGESZ: + 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_get_auxilary + * + * Get a field of the auxillary structure + */ +int wld_get_auxilary( ElfW(auxv_t) *av, int type, int *val ) +{ + for ( ; av->a_type != AT_NULL; av++) + if( av->a_type == type ) + { + *val = av->a_un.a_val; + return 1; + } + return 0; +} + +/* + * 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, 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; +} + +#define ROUND_SIZE(addr,size) \ + (((int)(size) + ((int)(addr) & page_mask) + page_mask) & ~page_mask) + +/* + * 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( char *pe_file, void **base, size_t *size ) +{ + IMAGE_DOS_HEADER dos; + IMAGE_NT_HEADERS nt; + int unix_fd, page_mask = GL(dl_pagesize) - 1; + + if( !pe_file ) + return; + + unix_fd = open( pe_file, O_RDONLY, 0); + if( unix_fd < 0 ) + return; + + /* + * The following code is borrowed from + * get_image_params() in wine/server/mapping.c + */ + if (read( unix_fd, &dos, sizeof(dos) ) != sizeof(dos)) goto error; + if (dos.e_magic != IMAGE_DOS_SIGNATURE) goto error; + if (lseek( unix_fd, dos.e_lfanew, SEEK_SET ) == -1) goto error; + + if (read( unix_fd, &nt.Signature, sizeof(nt.Signature) ) != sizeof(nt.Signature)) goto error; + if (nt.Signature != IMAGE_NT_SIGNATURE) goto error; + if (read( unix_fd, &nt.FileHeader, sizeof(nt.FileHeader) ) != sizeof(nt.FileHeader)) goto error; + /* zero out Optional header in the case it's not present or partial */ + memset(&nt.OptionalHeader, 0, sizeof(nt.OptionalHeader)); + if (read( unix_fd, &nt.OptionalHeader, nt.FileHeader.SizeOfOptionalHeader) != nt.FileHeader.SizeOfOptionalHeader) goto error; + + *size = ROUND_SIZE( 0, nt.OptionalHeader.SizeOfImage ); + *base = (void *)nt.OptionalHeader.ImageBase; + +#ifdef DUMP_PERESERVE + wld_printf( "parsed PE image... base %x size %x\n", *base, *size ); +#endif + +error: + close( unix_fd ); +} + +/* + * Find a symbol in the symbol table of the executable loaded + */ +void *wld_find_symbol( const ElfW(Phdr) *phdr, int num, char *var ) +{ + const ElfW(Dyn) *dyn = NULL; + void *ptr = NULL; + const ElfW(Phdr) *ph; + const ElfW(Sym) *symtab = NULL; + const char *strings = NULL; + + /* check the values */ +#ifdef DUMP_SYMS + wld_printf("%x %x\n", phdr, num ); +#endif + if( ( phdr == NULL ) || ( num == 0 ) ) + { + wld_printf("could not find PT_DYNAMIC header entry\n"); + return NULL; + } + + /* parse the (already loaded) ELF executable's header */ + for (ph = phdr; ph < &phdr[num]; ++ph) + { + if( PT_DYNAMIC == ph->p_type ) + { + dyn = (void *) ph->p_vaddr; + num = ph->p_memsz / sizeof (Elf32_Dyn); + break; + } + } + if( !dyn ) + { + wld_printf("No PT_DYNAMIC section\n"); + return ptr; + } + + while( dyn->d_tag ) + { + if( dyn->d_tag == DT_STRTAB ) + strings = (const char*) dyn->d_un.d_ptr; + if( dyn->d_tag == DT_SYMTAB ) + symtab = (const ElfW(Sym) *)dyn->d_un.d_ptr; +#ifdef DUMP_SYMS + wld_printf("%x %x\n", dyn->d_tag, dyn->d_un.d_ptr ); +#endif + dyn++; + } + + if( (!symtab) || (!strings) ) + { + wld_printf("No symbol table found\n"); + return ptr; + } + + /* FIXME: this crashes if it doesn't find the symbol we want */ + do + { + if( ( ELF32_ST_BIND(symtab->st_info) == STT_OBJECT ) && + ( 0 == strcmp( strings+symtab->st_name, var ) ) ) + { + ptr = (void*)symtab->st_value; + break; + } + symtab++; + } + while( symtab->st_name ); + +#ifdef DUMP_SYMS + wld_printf("Found %s -> %x\n", strings+symtab->st_name, symtab->st_value ); +#endif + + return ptr; +} + +/* + * 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, phnum = 0; + unsigned char buf[0x800]; + void *start; + char **argv = (char**) &argc, **p, *pe_file = NULL; + struct link_map map; + ElfW(auxv_t)* av; + ElfW(Phdr) *elf_phdr = NULL; + struct wine_preload_info **wine_preload_vars; + static struct wine_preload_info vars; + + 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, find the name of the PE file */ + while( *p ) + { + const char pefile[] = "WINEPEEXEPATH="; + if( 0 == memcmp( pefile, *p, sizeof pefile - 1 ) ) + pe_file = (*p) + sizeof pefile - 1 ; + p++; + } + + av = (ElfW(auxv_t)*) (p+1); + wld_get_auxilary( av, AT_PAGESZ, &dl_pagesize ); +#ifdef DUMP_AUX_INFO + wld_dump_auxilary( av ); +#endif + + /* reserve memory that Wine needs */ + vars.pe_size = 0x01000000; + vars.pe_address = (void*)0x00400000; + wld_reserve_wine_memory( pe_file, &vars.pe_address, &vars.pe_size ); + mmap( vars.pe_address, vars.pe_size, PROT_NONE, + MAP_FIXED|MAP_PRIVATE|MAP_ANON, -1, 0); + + /* 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 ); + start = wld_map_so( fd, buf, sizeof buf, &map ); + if( !start ) + exit(0); + close( fd ); + + /* + * Pass information to the wine loader in a static variable + * so it doesn't get destroyed after we leave this stack frame + */ + wld_get_auxilary( av, AT_PHDR, (int*)&elf_phdr ); + wld_get_auxilary( av, AT_PHNUM, (int*)&phnum ); + wine_preload_vars = wld_find_symbol( elf_phdr, phnum, "wine_preload_vars" ); + if( wine_preload_vars ) + { + unsigned char *base = (void*)0x7ff81000; /* FIXME - define properly */ + vars.preload_base = (void*) base; + vars.preload_size = (size_t) (_end - base); + *wine_preload_vars = &vars; + } + + 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" +); +

2 6

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 13, 2004

April 13, 2004

Hi, I didnt get the second point that you mention Another one might be that WM_CREATE must be always passed to DEFDLG_Proc regardless of what result an application returned to DefDlgProc16/A/W. Regards Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Tuesday, April 13, 2004 5:46 PM > To: Robert Shearman; wine-devel(a)winehq.org > Cc: santosh.s(a)sonata-software.com > Subject: Re: Handling dialog messages for non-template based dialogs > > "Robert Shearman" <R.J.Shearman(a)warwick.ac.uk> wrote: > > > > + if (!(dlgInfo = HeapAlloc( GetProcessHeap(), 0, > > sizeof(*dlgInfo) ))) return 0; > > > > Minor point, but you probably want to return -1 here to say that window > > creation failed. > > Good catch. Another one might be that WM_CREATE must be always > passed to DEFDLG_Proc regardless of what result an application > returned to DefDlgProc16/A/W. > > -- > 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. *********************************************************************

2 1

RE: Handling dialog messages for non-template based dialogs
by Robert Shearman April 13, 2004

April 13, 2004

Santosh Siddheshwar 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. > <<patch.diff>> > > diff -urN wine-20040408/windows/defdlg.c wine-bin/windows/defdlg.c > --- wine-20040408/windows/defdlg.c 2003-09-06 04:45:40.000000000 +0530 > +++ wine-bin/windows/defdlg.c 2004-04-13 11:53:02.000000000 +0530 > @@ -168,6 +168,27 @@ > { > switch(msg) > { > + case WM_CREATE: > + { > + /* Allocate memory for DIALOGINFO structure and store in DWL_DIALOGINFO > + structure. Also flag the window as a dialog type */ > + WND* wndPtr; > + DIALOGINFO* dlgInfo; > + if (!(dlgInfo = HeapAlloc( GetProcessHeap(), 0, sizeof(*dlgInfo) ))) return 0; Minor point, but you probably want to return -1 here to say that window creation failed. > + dlgInfo->hwndFocus = 0; > + dlgInfo->hUserFont = 0; > + dlgInfo->hMenu = 0; > + dlgInfo->xBaseUnit = 0; > + dlgInfo->yBaseUnit = 0; Rob

2 1

[PATCH] SHLWapi fixes - Need cleanup
by Flameeyes April 13, 2004

April 13, 2004

Hi, Some time ago I fixed some stubs on SHLWapi lib, but the patch needs some cleanup. I haven't the time to do them (because they are not "natural" for me, and when I tried to post the cleaned up version I missed a lot of cleans. If someone can take a look at it, clean it up and possibly merge it, the patch can be downloaded from http://flameeyes.web.ctonet.it/patches/wine-shlwapi_20031219.diff . Should apply well also in the current version, because it never created conflicts with code in CVS. HTH -- Flameeyes <dgp85(a)users.sourceforge.net> http://flameeyes.web.ctonet.it/

1 0

Re: Trace flags passed to SetWindowPos for easier debugging.
by Dmitry Timoshkov April 13, 2004

April 13, 2004

"Mike Hearn" <mh(a)codeweavers.com> wrote: > Mike Hearn <mike(a)navi.cx> > Trace flags passed to SetWindowPos for easier debugging. Did you try to add +win? Is that what you are looiking for? windows/winpos.c,1176: BOOL WINAPI SetWindowPos( HWND hwnd, HWND hwndInsertAfter, INT x, INT y, INT cx, INT cy, UINT flags ) { WINDOWPOS winpos; TRACE("hwnd %p, after %p, %d,%d (%dx%d), flags %08x\n", hwnd, hwndInsertAfter, x, y, cx, cy, flags); if(TRACE_ON(win)) dump_winpos_flags(flags); ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ -- Dmitry.

2 1

Wine-20040408 Results (so far) for MFC App
by greensh＠knology.net April 13, 2004

April 13, 2004

Now it works very nicely... I spent Easter debugging most of the crashes I was having due to the GUI menus (even in winders), from variables not freed, etc and now I cannot get it to crash...yet. I'm using MSVC6 for all the development under VMware 4.5.1. The GUI code is completely MFC with 16 color bitmaps for resources and all underlying solver code, etc is C++/C. This code puts a P4 laptop and memory under moderate load when run... so it's a good test. Below are the ONLY messages I get from Wine-20040408 when I run it from the command-line; wine sokoban.exe (then it opens the GUI) and then do many things with it; data file I/O, parsing many text files, GUI menus, clicks, etc. As you can see below the help file opens fine, but does not display text properly. I used Shalom Help (free) for winders (it works fine on winxp as I've tested). Any ideas why these messages appear and how to prevent them? I can send app screenshots and provide code snippets. Below these is my entire wine config. fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! fixme:winhelp:HLPFILE_UncompressRLE Null length 2, next is 0 fixme:winhelp:HLPFILE_UncompressRLE Null length 2, next is 140 fixme:winhelp:HLPFILE_UncompressRLE Buffer X-flow: d(108628) instead of d(108640) fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! fixme:dc:LockWindowUpdate (0x10029), partial stub! fixme:dc:LockWindowUpdate ((nil)), partial stub! ~/.wine/config: [Drive C] "Path" = "/home/greensh/.wine/c" "Type" = "hd" [Drive D] "Path" = "/mnt/cdrw" "Type" = "cdrom" ; make sure that device is correct and has proper permissions ! "Device" = "/dev/scd0" [Drive E] "Path" = "/tmp" "Type" = "hd" [Drive F] "Path" = "/home/greensh" "Type" = "hd" [Drive Z] "Path" = "/" "Type" = "hd" [wine] "Windows" = "c:\\windows" "System" = "c:\\windows\\system" "Temp" = "e:\\" "Path" = "c:\\windows;c:\\windows\\system;e:\\;e:\\test;f:\\" "Profile" = "c:\\windows\\Profiles\\Administrator" "GraphicsDriver" = "x11drv" ;"ShowDirSymlinks" = "1" ;"ShowDotFiles" = "1" "ShellLinker" = "wineshelllink" # [wineconf] [Version] ; Windows version to imitate (win95,win98,winme,nt351,nt40,win2k,winxp, win2k3,win20,win30,win31) ;"Windows" = "win98" ; DOS version to imitate ;"DOS" = "6.22" ; Be careful here, wrong DllOverrides settings have the potential ; to pretty much kill your setup. [DllOverrides] ; some dlls you may want to change "oleaut32" = "builtin, native" "ole32" = "builtin, native" "comdlg32" = "builtin, native" "shell32" = "builtin, native" "shfolder" = "builtin, native" "shlwapi" = "builtin, native" "shdocvw" = "builtin, native" "advapi32" = "builtin, native" "msvcrt" = "native, builtin" "mciavi.drv" = "native, builtin" "mcianim.drv" = "native, builtin" "msi" = "native, builtin" ; you can specify applications too ; this one will apply for all notepad.exe ;"*notepad.exe" = "native, builtin" ; this one will apply only for a particular file ;"C:\\windows\\regedit.exe" = "native, builtin" ; default for all other dlls "*" = "builtin, native" [x11drv] ; Number of colors to allocate from the system palette "AllocSystemColors" = "100" ; Use a private color map "PrivateColorMap" = "N" ; Favor correctness over speed in some graphics operations "PerfectGraphics" = "N" ; Color depth to use on multi-depth screens ;;"ScreenDepth" = "16" ; Allow the window manager to manage created windows "Managed" = "Y" ; Use a desktop window of 640x480 for Wine ;"Desktop" = "640x480" ; Use XFree86 DGA extension if present ; (make sure /dev/mem is accessible by you !) "UseDGA" = "Y" ; Use XVidMode extension if present "UseXVidMode" = "Y" ; Use XRandR extension if present "UseXRandR" = "Y" ; Use the take focus protocol "UseTakeFocus" = "Y" ; Enable DirectX mouse grab "DXGrab" = "N" ; Create the desktop window with a double-buffered visual ; (useful to play OpenGL games) "DesktopDoubleBuffered" = "N" ; Run in synchronous mode (useful for debugging X11 problems) ;;"Synchronous" = "Y" ; ; Use the Render extension to render client side fonts (default "Y") ;;"ClientSideWithRender" = "Y" ; Fallback on X core requests to render client side fonts (default "Y") ;;"ClientSideWithCore" = "Y" ; Set both of the previous two to "N" in order to force X11 server side fonts ; ; Anti-alias fonts if using the Render extension (default "Y") ;;"ClientSideAntiAliasWithRender" = "Y" ; Anti-alias fonts if using core requests fallback (default "Y") ;;"ClientSideAntiAliasWithCore" = "Y" ; [fonts] ;Read the Fonts topic in the Wine User Guide before adding aliases ;See a couple of examples for russian users below "Resolution" = "96" "Default" = "-adobe-helvetica-" "DefaultFixed" = "fixed" "DefaultSerif" = "-adobe-times-" "DefaultSansSerif" = "-adobe-helvetica-" ;; default TrueType fonts with russian koi8-r encoding ;"Default" = "-monotype-arial-*-*-*--*-*-*-*-*-*-koi8-r" ;"DefaultFixed" = "-monotype-courier new-*-*-*--*-*-*-*-*-*-koi8-r" ;"DefaultSerif" = "-monotype-times new roman-*-*-*--*-*-*-*-*-*-koi8-r" ;"DefaultSansSerif" = "-monotype-arial-*-*-*--*-*-*-*-*-*-koi8-r" ;; default cyrillic bitmap X fonts ;"Default" = "-cronyx-helvetica-" ;"DefaultFixed" = "fixed" ;"DefaultSerif" = "-cronyx-times-" ;"DefaultSansSerif" = "-cronyx-helvetica-" ; the TrueType font dirs you want to make accessible to wine [FontDirs] ;"dir1" = "/usr/X11R6/lib/X11/fonts/TrueType" ;"dir2" = "/usr/share/fonts/truetype" ;"dir3" = "/usr/X11R6/lib/X11/fonts/TT" ;"dir4" = "/usr/share/fonts/TT" [ppdev] ;; key: io-base of the emulated port ;; value : parport-device{,timeout} ;; timeout for auto closing an open device ( not yet implemented) ;"378" = "/dev/parport0" ;"278" = "/dev/parport1" ;"3bc" = "/dev/parport2" [spooler] "FILE:" = "tmp.ps" "LPT1:" = "|lpr" "LPT2:" = "|gs -sDEVICE=bj200 -sOutputFile=/tmp/fred -q -" "LPT3:" = "/dev/lp3" [ports] ;"read" = "0x779,0x379,0x280-0x2a0" ;"write" = "0x779,0x379,0x280-0x2a0" [Debug] ;"RelayExclude" = "RtlEnterCriticalSection;RtlLeaveCriticalSection" ;"RelayInclude" = "user32.CreateWindowA" ;"RelayFromExclude" = "user32;x11drv" ;"RelayFromInclude" = "sol.exe" ;"SnoopExclude" = "RtlEnterCriticalSection;RtlLeaveCriticalSection" ;"SpyExclude" = "WM_SIZE;WM_TIMER;" [registry] ;These are all booleans. Y/y/T/t/1 are true, N/n/F/f/0 are false. ;Defaults are read all, write to Home ; Where to find the global registries ;"GlobalRegistryDir" = "/etc"; ; Global registries (stored in /etc) "LoadGlobalRegistryFiles" = "Y" ; Home registries (stored in ~user/.wine/) "LoadHomeRegistryFiles" = "Y" ; Load Windows registries from the Windows directory "LoadWindowsRegistryFiles" = "Y" ; TRY to write all changes to home registries "WritetoHomeRegistryFiles" = "Y" ; Registry periodic save timeout in seconds ; "PeriodicSave" = "600" ; Save only modified keys "SaveOnlyUpdatedKeys" = "Y" [Clipboard] "ClearAllSelections" = "0" "PersistentSelection" = "1" "UsePrimary" = "0" ; List of all directories directly contain .AFM files [afmdirs] "1" = "/usr/share/ghostscript/fonts" "2" = "/usr/share/a2ps/afm" "3" = "/usr/share/enscript" "4" = "/usr/X11R6/lib/X11/fonts/Type1" [WinMM] ; Uncomment the "Drivers" line matching your sound setting. "Drivers" = "wineoss.drv" ; default for most common configurations ;"Drivers" = "winearts.drv" ; for KDE ;"Drivers" = "winealsa.drv" ; for ALSA users ;"Drivers" = "winejack.drv" ; for Jack sound server ;"Drivers" = "winenas.drv" ; for NAS sound system ;"Drivers" = "wineaudioio.drv" ; for Solaris machines ;"Drivers" = "" ; to disable sound "WaveMapper" = "msacm.drv" "MidiMapper" = "midimap.drv" [dsound] ;; HEL only: Number of waveOut fragments ahead to mix in new buffers. ;"HELmargin" = "5" ;; HEL only: Number of waveOut fragments ahead to queue to driver. ;"HELqueue" = "5" ;; Max number of fragments to prebuffer ;"SndQueueMax" = "28" ;; Min number of fragments to prebuffer ;"SndQueueMin" = "12" ;; Forces emulation mode (using wave api) ;"HardwareAcceleration" = "Emulation" ;; Sets default playback device (0 - number of devices - 1) ;"DefaultPlayback" = "0" ; use first device (/dev/dsp) ;"DefaultPlayback" = "1" ; use second device (/dev/dsp1) ;"DefaultPlayback" = "2" ; use third device (/dev/dsp2) ;; Sets default capture device (0 - number of devices - 1) ;"DefaultCapture" = "0" ; use first device (/dev/dsp) ;"DefaultCapture" = "1" ; use second device (/dev/dsp1) ;"DefaultCapture" = "2" ; use third device (/dev/dsp2) [Network] ;; Use the DNS (Unix) host name always as NetBIOS "ComputerName" (boolean, default "Y"). ;; Set to N if you need a persistent NetBIOS ComputerName that possibly differs ;; from the Unix host name. You'll need to set ComputerName in ;; HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\ComputerName\ComputerNam e, too. ;"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\\_INS0432._MP\\x11drv] "Desktop" = "640x480" [AppDefaults\\_INS0466._MP\\x11drv] "Desktop" = "640x480" [AppDefaults\\_INS0576._MP\\x11drv] "Desktop" = "640x480" [AppDefaults\\_INS5176._MP\\x11drv] "Desktop" = "640x480" [AppDefaults\\_INS5576._MP\\x11drv] "Desktop" = "640x480" ;[AppDefaults\\iexplore.exe\\DllOverrides] ;"shlwapi" = "native" ;"rpcrt4" = "native" ;"ole32" = "native" ;"shdocvw" = "native" ;"wininet" = "native" ;"shfolder" = "native" ;"shell32" = "native" ;"shell" = "native" ;"comctl32" = "native" ; ;[AppDefaults\\setup.exe\\x11drv] ;"Desktop" = "800x600" ; ;[AppDefaults\\sol.exe\\Version] ;"Windows" = "nt40" ; ;; Some games (Quake 2, UT) refuse to accept emulated dsound devices. ;; You can add an AppDefault entry like this for such cases. ;[AppDefaults\\pickygame.exe\\dsound] ;"EmulDriver" = "N" # [/wineconf]

3 2

Windows installer XML (WiX) released on sourceforce by Microsoft
by David.Goodenough＠DGA.co.uk April 13, 2004

April 13, 2004

Given the problems that Wine has had with Installers, had anyone noticed that WiX has now been released by Microsoft (under the CPL license) on Sourceforge. http://seattlepi.nwsource.com/business/168652_msftnotebook12.html It seems to be accessible as http://wix.sf.net, but that redirects to a blog site on http://blogs.msdn.com/robmen/category/4625.aspx. David

1 0

Re: smb: separate files
by Alexandre Julliard April 12, 2004

April 12, 2004

Juan Lang <juan_lang(a)yahoo.com> writes: > Alexandre, this is somewhat of a test patch in that it > doesn't "do" anything besides split files up. What > I'm hoping to do is share smb.c, and to a lesser > extent netbios.c, between netapi32 and wherever else > it's needed (currently only kernel32). smbfile.c > isn't needed by netapi32. The file stuff is going to be removed anyway, so I'm not sure there's much point in splitting the file. File I/O has to be done by the Linux kernel, we can't do it right in userspace. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: Remove "Path" lines from sample config
by Vincent Béron April 12, 2004

April 12, 2004

Le dim 11/04/2004 à 20:56, Sylvain Petreolle a écrit : > Vincent, > >Changelog: > >Remove "Path" lines from sample config. > >Setup the links for the default DOS drives in wineinstall. > >+ ln -s /mnt/fd0 ~/.wine/dosdevices/a: > >+ ln -s $CROOT ~/.wine/dosdevices/c: > >+ ln -s /cdrom ~/.wine/dosdevices/d: > The cdrom is usually more often /mnt/cdrom rather than /cdrom, isnt't it ? Having only used RedHat, I can't say for sure what it's like for other distros. I just copied what was in the sample config file. Vincent

2 1

Can't load/print with a 16 bit app, wine 20040408
by ceztko＠libero.it April 12, 2004

April 12, 2004

I can't load a 16 bit app...The problem is correlated to the printing system: i get an "run-time Error 482: Printer Error". These are the logs: Log WINEDEBUG="warn+all" ..... warn:dosfs:DOSFS_FindUnixName L"test\\WINEPS16.dll" not found in '/home/franzo/. wine/dosdevices/e:' warn:file:CreateFileW Unable to get full filename from L"e:\\test\\WINEPS16.dll" (GLE 2) warn:dosfs:DOSFS_FindUnixName L"WINEPS16.dll" not found in '/home/franzo/.wine/d osdevices/f:' warn:file:CreateFileW Unable to get full filename from L"f:\\WINEPS16.dll" (GLE 2) warn:module:load_dll Failed to load module L"WINEPS16"; status=c0000135 err:dc:CreateDCW no driver found for L"WINEPS16" Log WINEDEBUG="psdrv" ..... trace:psdrv:PSDRV_FindPrinterInfo No 'Paper Size' for printer 'HP' trace:psdrv:PSDRV_ExtDeviceMode (Driver=(null), hwnd=(nil), devOut=(nil), Device ='HP', Port='LPT1:', devIn=(nil), Profile='(null)', Mode=0000) trace:psdrv:PSDRV_FindPrinterInfo 'HP' trace:psdrv:PSDRV_ExtDeviceMode (Driver=(null), hwnd=(nil), devOut=0x41b6ce42, D evice='HP', Port='LPT1:', devIn=(nil), Profile='(null)', Mode=0002) err:dc:CreateDCW no driver found for L"WINEPS16" In /usr/local/lib/wine/ i created this situation: -rwxr-xr-x 1 root root 1851820 Apr 12 15:47 wineps.dll.so lrwxrwxrwx 1 root root 13 Apr 12 19:52 wineps.drv.so -> wineps.dll.so lrwxrwxrwx 1 root root 13 Apr 12 17:46 wineps16.dll.so -> wineps.dll.so lrwxrwxrwx 1 root root 13 Apr 12 15:47 wineps16.drv.so -> wineps.dll.so My configuration: In the registry: [System\\CurrentControlSet\\Control\\Print\\Environments\\Windows 4.0\\Drivers\\PS Driver] 1081792881 "Configuration File"="wineps16" "Data File"="<datafile?>" "Datatype"="RAW" "Dependent Files"=str(7):"" "Driver"="wineps16" "Help File"="<helpfile?>" "Monitor"="<monitor name?>" "Version"=dword:00000400 [System\\CurrentControlSet\\Control\\Print\\Printers\\HP] 1081784615 "Attributes"=dword:00000000 "Default Devmode"=hex:48,50,00,65,20,50,6f,73,74,53,63,72,69,70,74,20,44,72,69,\ 76,65,72,00,00,00,00,00,00,00,00,00,00,0a,03,01,00,9c,00,08,00,13,6b,00,00,\ 01,00,01,00,ea,0a,6f,08,64,00,01,00,07,00,00,00,02,00,01,00,00,00,03,00,00,\ 00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,\ 00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,\ 00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,\ 00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00,00 "Name"="HP" "Port"="LPT1:" "Print Processor"="WinPrint" "Printer Driver"="PS Driver" "Priority"=dword:00000000 "Start Time"=dword:00000000 "Status"=dword:00000000 "Until Time"=dword:00000000 [System\\CurrentControlSet\\Control\\Print\\Printers\\HP\\PrinterDriverData] 1081784392 "PPD File"="/usr/local/share/wine/generic.ppd" [System\\CurrentControlSet\\Control\\Print\\Printers\\HP\\PrinterDriverData\\FontSubTable] 1081784392 "Arial"="Helvetica" "Courier New"="Courier" "Helv"="Helvetica" "Times New Roman"="Times" In win.ini: [windows] device=HP,WINEPS16,LPT1: [devices] HP=WINEPS16,LPT1: [PrinterPorts] HP=WINEPS16,LPT1:,15,45 In config [spooler] "FILE:" = "tmp.ps" "LPT1:" = "|lpr" "LPT2:" = "|gs -sDEVICE=bj200 -sOutputFile=/tmp/fred -q -" "LPT3:" = "/dev/lp3" Wine version is the last: 20040408. I can provide more information if needed. Thanks! Francesco Pretto P.S.: a last time WINEDEBUG="+all" log (using gentoo ebuild wine, differently than the ones before, taked using wine source unmodified), this identify the last activity before the loading of the error pop-up: ........ trace:psdrv:PSDRV_ParsePPD INPUTSLOTS 'Tray3' Name 'Tray 3' (259) Invocation '<</ManualFeed false /MediaPosition 1>>setpagedevice' trace:psdrv:PSDRV_ParsePPD INPUTSLOTS 'Manual' Name 'Manual Feed' (260) Invocation '<</ManualFeed true /Media Position 3>>setpagedevice' trace:reg:RegQueryValueExA (0x14c,"Paper Size",(nil),(nil),0x41584b84,0x41584b9c=4) trace:reg:RegQueryInfoKeyW (0x14c,(nil),0,(nil),(nil),(nil),0x41584adc,0x41584ae0,0x41584ae4,(nil),(nil)) trace:reg:RegEnumValueW (0x14c,0,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,1,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,2,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,3,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegQueryInfoKeyW (0x14c,(nil),0,(nil),(nil),(nil),0x41584adc,0x41584ae0,0x41584ae4,(nil),(nil)) trace:reg:RegEnumValueW (0x14c,0,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,1,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,2,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,3,0x415b7ad8,0x41584ad4,(nil),(nil),0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,0,0x415b7ad8,0x41584ad4,(nil),0x41584ad0,0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,1,0x415b7ad8,0x41584ad4,(nil),0x41584ad0,0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,2,0x415b7ad8,0x41584ad4,(nil),0x41584ad0,0x415b7b00,0x41584ad8) trace:reg:RegEnumValueW (0x14c,3,0x415b7ad8,0x41584ad4,(nil),0x41584ad0,0x415b7b00,0x41584ad8) err:dc:CreateDCW no driver found for L"WINEPS16" trace:int:WIN87_fpmath (cs:eip=125f:17dc es=1207 bx=0001 ax=1207 dx=1207) trace:resource:LoadString16 inst=19e6 id=01e2 buff=41a992fe len=256 trace:resource:LoadString16 strlen = 19 trace:resource:LoadString16 'Errore di stampante' loaded ! trace:resource:LoadString16 inst=19e6 id=277b buff=41a86810 len=500 trace:resource:LoadString16 strlen = 22 trace:resource:LoadString16 'Errore di run-time '|'' loaded !

1 0

current version of "wine_adopt_thread"
by Paul Davis April 12, 2004

April 12, 2004

I offer this for Alexandre and other's consideration and comments. This is already in use in libfst, which is in turn used by Ardour, a native linux digital audio workstation, and gAlan, a native linux modular synthesis system. There is still no proper cleanup on failure, and nothing is done if the adopted thread calls pthread_exit(). Both are simple to fix. --p ---------------------------------------------------------------------- /* * winelib code - Initializes wine as shared library * * Copyright 2004 Novell, Inc. (http://www.novell.com/) * * This 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. * * This 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 this library; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ #include <stdio.h> #include <windows.h> #include <windows/ntstatus.h> #include <setjmp.h> #include <signal.h> #include <fcntl.h> #include <wine/library.h> #include <sys/mman.h> #include <dlfcn.h> #include <limits.h> #include <errno.h> #include <string.h> #include "thread.h" #include "pthread.h" #include "server.h" #include "libwinelib.h" #include "fst.h" #include <winternl.h> typedef NTSTATUS WINAPI (*NtProtectVirtualMemory_ptr)(HANDLE, PVOID*, ULONG*, ULONG, ULONG*); typedef NTSTATUS WINAPI (*NtFreeVirtualMemory_ptr)(HANDLE, PVOID, ULONG*, ULONG); typedef NTSTATUS WINAPI (*NtAllocateVirtualMemory_ptr)(HANDLE, PVOID*, PVOID, ULONG*, ULONG, ULONG); typedef VOID WINAPI (*RtlAcquirePebLock_ptr) (VOID); typedef VOID WINAPI (*RtlReleasePebLock_ptr) (VOID); static NtProtectVirtualMemory_ptr ntprotectvm; static NtAllocateVirtualMemory_ptr ntallocatevm; static NtFreeVirtualMemory_ptr ntfreevm; static RtlAcquirePebLock_ptr acquirepeblock; static RtlReleasePebLock_ptr releasepeblock; static LIST_ENTRY tls_link_head; static PEB* peb; static HTASK16 htask16; static unsigned int (*server_call)(void*); static void (*send_fd)(int); static void (*init_thread)(int,int,void*); static jmp_buf jump; static void (*sharedwine_signal_handler)(int signal, siginfo_t*, void*) = NULL; void* initial_teb; static void setup_signal (int sig) { struct sigaction sigact; struct sigaction oldact; sigact.sa_handler = (void (*)(int)) sharedwine_signal_handler; sigact.sa_flags = SA_RESTART|SA_SIGINFO; sigact.sa_sigaction = (void (*)(int signal, siginfo_t* info, void *ptr)) sharedwine_signal_handler; sigemptyset (&sigact.sa_mask); sigaddset (&sigact.sa_mask, SIGINT); sigaddset (&sigact.sa_mask, SIGUSR2); if (sigaction (sig, &sigact, NULL)) { fst_error ("could not install signal handler for signal %d", sig); } } void redirect_signals () { setup_signal (SIGINT); setup_signal (SIGFPE); setup_signal (SIGILL); setup_signal (SIGSEGV); setup_signal (SIGABRT); setup_signal (SIGTERM); setup_signal (SIGBUS); setup_signal (SIGTRAP); } /* The per-thread signal stack size */ #ifdef __i386__ #define SIGNAL_STACK_SIZE 4096 #else #define SIGNAL_STACK_SIZE 0 /* we don't need a signal stack on non-i386 */ #endif /* from winnt.h, but hard to make it appear */ #define MEM_SYSTEM 0x80000000 /*********************************************************************** * alloc_teb (no difference from libwine, but its static there) */ static TEB *shared_alloc_teb( ULONG *size ) { TEB *teb; *size = SIGNAL_STACK_SIZE + sizeof(TEB); teb = wine_anon_mmap( NULL, *size, PROT_READ | PROT_WRITE | PROT_EXEC, 0 ); if (teb == (TEB *)-1) return NULL; if (!(teb->teb_sel = wine_ldt_alloc_fs())) { munmap( teb, *size ); return NULL; } teb->Tib.ExceptionList = (void *)~0UL; teb->Tib.StackBase = (void *)~0UL; teb->Tib.Self = &teb->Tib; teb->Peb = peb; teb->StaticUnicodeString.Buffer = teb->StaticUnicodeBuffer; teb->StaticUnicodeString.MaximumLength = sizeof(teb->StaticUnicodeBuffer); return teb; } /*********************************************************************** * free_teb (no change from libwine, but its static there) */ static inline void shared_free_teb( TEB *teb ) { ULONG size = 0; void *addr = teb; ntfreevm ( GetCurrentProcess(), &addr, &size, MEM_RELEASE ); wine_ldt_free_fs( teb->teb_sel ); munmap( teb, SIGNAL_STACK_SIZE + sizeof(TEB) ); } static int proxy_pripe_request[2]; static int proxy_pripe_status[2]; static pthread_mutex_t proxy_thread_lock; static DWORD WINAPI new_thread_proxy (LPVOID param) { int request_pipe[2]; NTSTATUS status; struct __server_request_info sri; struct new_thread_request *req = &sri.u.req.new_thread_request; int fd = -1; TEB *teb; struct wine_pthread_thread_info thread_info; ULONG size; if (pipe (proxy_pripe_request)) { fst_error ("cannot set up proxy request pipe (%s)", strerror (errno)); return -1; } if (pipe (proxy_pripe_status)) { fst_error ("cannot set up proxy reply pipe (%s)", strerror (errno)); return -1; } while (1) { /* wait for a request size to be written */ if (read (proxy_pripe_request[0], &sri, sizeof (sri)) != sizeof (sri)) { fst_error ("cannot read new thread request from proxy request pipe (%s)", strerror (errno)); return -1; } if (pipe (request_pipe) >= 0) { /* set up a new request pipe */ fcntl (request_pipe[1], F_SETFD, 1); /* set close on exec flag */ /* server_*/ send_fd (request_pipe[0]); req->request_fd = request_pipe[0]; if ((status = server_call (req)) == 0) { fd = request_pipe[1]; } else { close (request_pipe[0]); close (request_pipe[1]); } } /* create a new TEB. the "thread_info" structure seems to be the most convenient way of passing back the result. */ teb = shared_alloc_teb (&size); thread_info.stack_base = NULL; thread_info.stack_size = 0; thread_info.teb_base = teb; thread_info.teb_size = size; thread_info.teb_sel = teb->teb_sel; /* send back the result, request fd and the newly allocated TEB info */ if (write (proxy_pripe_status[1], &sri, sizeof (sri)) != sizeof (sri)) { fst_error ("cannot write request/reply back to proxy reply pipe (%s)", strerror (errno)); /* XXX free teb */ return -1; } if (write (proxy_pripe_status[1], &fd, sizeof (fd)) != sizeof (status)) { fst_error ("cannot write status back to proxy reply pipe (%s)", strerror (errno)); /* XXX free teb */ return -1; } if (write (proxy_pripe_status[1], &thread_info, sizeof (thread_info)) != sizeof (thread_info)) { fst_error ("cannot write thread info to proxy reply pipe (%s)", strerror (errno)); /* XXX free teb */ return -1; } close (request_pipe[0]); } return; } extern int wine_shared_premain (); int WINAPI WinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPSTR lpszCmdLine, int nCmdShow) { int jmpstat = 1; if (CreateThread (NULL, 0, new_thread_proxy, NULL, 0, NULL) == NULL) { fst_error ("could not create new thread proxy"); } if (wine_shared_premain ()) { jmpstat = 2; } // return to caller. longjmp (jump, jmpstat); } /* WineLoadLibrary is used by System.Windows.Forms to import the Wine dlls */ void * WineLoadLibrary(unsigned char *dll) { return(LoadLibraryA(dll)); } void * WineGetProcAddress(void *handle, unsigned char *function) { return(GetProcAddress(handle, function)); } int SharedWineInit (void (*sighandler)(int,siginfo_t*,void*)) { unsigned char Error[1024]=""; char *WineArguments[] = {"sharedapp", LIBPATH "/libfst.so", NULL}; void* stackbase; size_t stacksize; void *ntdll; void *ntso; char ntdllpath[PATH_MAX+1]; char* dlerr; sharedwine_signal_handler = sighandler; if (setjmp (jump) == 0) { wine_init (2, WineArguments, Error, sizeof (Error)); if (Error[0]!='\0') { fst_error ("Wine initialization error:%s\n", Error); return -1; } } redirect_signals (); initial_teb = NtCurrentTeb(); /* set the stack limits so that any exception handling based on tracking stack boundaries has a chance of working. */ NtCurrentTeb()->Tib.StackBase = (void*) ~0UL; NtCurrentTeb()->Tib.StackLimit = 0; putenv ("_WINE_SHAREDLIB_PATH=" DLLPATH); /* Loading this will pull in many other DLLs that would otherwise be loaded+unloaded over and over again during plugin discovery. */ if (WineLoadLibrary ("user32.dll") == NULL) { fst_error ("cannot load Windows DLL user32"); return 1; } if (WineLoadLibrary ("comdlg32.dll") == NULL) { fst_error ("cannot load Windows DLL comdlg32"); return 1; } if (WineLoadLibrary ("oleaut32.dll") == NULL) { fst_error ("cannot load Windows DLL oleaut32"); return 1; } /* set up the functions and data we need to be able to adopt threads. */ #define GET_SYMBOL(res,handle,name) if ((res = WineGetProcAddress (handle,name)) == NULL) { fst_error ("cannot find Windows function %s", name); return -1; } if ((ntdll = WineLoadLibrary( "ntdll.dll")) == NULL) { fst_error ("cannot load Windows DLL ntdll"); return 1; } GET_SYMBOL (ntallocatevm, ntdll, "NtAllocateVirtualMemory"); GET_SYMBOL (ntfreevm, ntdll, "NtFreeVirtualMemory"); GET_SYMBOL (ntprotectvm, ntdll, "NtProtectVirtualMemory"); GET_SYMBOL (acquirepeblock, ntdll, "RtlAcquirePebLock"); GET_SYMBOL (releasepeblock, ntdll, "RtlReleasePebLock"); tls_link_head = NtCurrentTeb()->TlsLinks; peb = NtCurrentTeb()->Peb; htask16 = NtCurrentTeb()->htask16; sprintf (ntdllpath, "%s/ntdll.dll.so", DLLPATH); if ((ntso = dlopen (ntdllpath, RTLD_NOW|RTLD_GLOBAL)) == NULL) { fst_error ("cannot open NT DLL (%s)", ntdllpath); return -1; } server_call = dlsym (ntso, "wine_server_call"); if ((dlerr = dlerror ()) != NULL) { fst_error ("cannot find wine_server_call (%s)", dlerr); return -1; } send_fd = dlsym (ntso, "wine_server_send_fd"); if ((dlerr = dlerror ()) != NULL) { fst_error ("cannot find wine_server_send_fd (%s)", dlerr); return -1; } init_thread = dlsym (ntso, "server_init_thread"); if ((dlerr = dlerror ()) != NULL) { fst_error ("cannot find server_init_thread (%s)", dlerr); return -1; } return 0; } /*********************************************************************** * adopt_thread - take an existing linux thread (of some kind), * make the wineserver aware of it, and set up the TEB so that * we can execute win32 functions in it. */ int wine_adopt_thread (void) { struct wine_pthread_thread_info thread_info; struct debug_info* debug_info; TEB* teb; int request_fd; int tid; stack_t altstack; void *addr; ULONG size; SERVER_START_REQ( new_thread ) { req->suspend = FALSE; req->inherit = 0; /* FIXME */ pthread_mutex_lock (&proxy_thread_lock); if (write (proxy_pripe_request[1], &__req, sizeof (__req)) != sizeof (__req) || read (proxy_pripe_status[0], &__req, sizeof (__req)) != sizeof (__req) || read (proxy_pripe_status[0], &request_fd, sizeof (request_fd)) != sizeof (request_fd) || read (proxy_pripe_status[0], &thread_info, sizeof (thread_info)) != sizeof (thread_info)) { fst_error ("cannot read/write proxy thread request (%s)", strerror (errno)); } pthread_mutex_unlock (&proxy_thread_lock); if (request_fd >= 0) { tid = reply->tid; } } SERVER_END_REQ; if (request_fd < 0) goto error; debug_info = (struct debug_info *) malloc (sizeof (struct debug_info)); debug_info->str_pos = debug_info->strings; debug_info->out_pos = debug_info->output; teb = (TEB*) thread_info.teb_base; size = thread_info.teb_size; teb->tibflags = TEBF_WIN32; teb->exit_code = STILL_ACTIVE; teb->request_fd = -1; teb->request_fd = -1; teb->reply_fd = -1; teb->wait_fd[0] = -1; teb->wait_fd[1] = -1; teb->debug_info = debug_info; teb->htask16 = htask16; wine_pthread_init_current_teb (&thread_info); teb->ClientId.UniqueProcess = (HANDLE)GetCurrentProcessId(); teb->ClientId.UniqueThread = (HANDLE)tid; teb->request_fd = request_fd; /* don't bother with signal initialization since we redirect all signals back to linux anyway. */ /* SIGNAL_Init(); */ /* server_*/ init_thread( thread_info.pid, thread_info.tid, NULL ); wine_pthread_init_thread( &thread_info ); /* create a memory view for the TEB */ ntallocatevm ( GetCurrentProcess(), &addr, teb, &size, MEM_SYSTEM, PAGE_EXECUTE_READWRITE ); /* allocate a memory view for the stack */ size = thread_info.stack_size; ntallocatevm ( GetCurrentProcess(), &teb->DeallocationStack, thread_info.stack_base, &size, MEM_SYSTEM, PAGE_EXECUTE_READWRITE ); /* limit is lower than base since the stack grows down */ teb->Tib.StackBase = (char *)thread_info.stack_base + thread_info.stack_size; teb->Tib.StackLimit = thread_info.stack_base; /* setup the guard page */ size = 1; ntprotectvm ( GetCurrentProcess(), &teb->DeallocationStack, &size, PAGE_EXECUTE_READWRITE | PAGE_GUARD, NULL ); acquirepeblock (); InsertHeadList( &tls_link_head, &teb->TlsLinks ); releasepeblock (); return 0; error: shared_free_teb (teb); /* XXX deallocate VM */ /* close thread handle */ close (request_fd); return -1; }

1 0

Winelib rookie...
by antonio serra April 12, 2004

April 12, 2004

Hi! I'm trying to port a windows code to linux via winelib. After running winemaker I get the project files and run ./configure. At a certain point the following message appears on the console: checking for windef.h... configure: error: Could not find the Wine headers (windef.h) I have the windef.h file in a directory named /usr/include/wine/windows/. Do I need to setup any environment variables or some kind of soft links? TIA, AS __________________________________ Do you Yahoo!? Yahoo! Tax Center - File online by April 15th http://taxes.yahoo.com/filing.html

2 1

Re: Remove "Path" lines from sample config
by Ivan Leo Murray-Smith April 12, 2004

April 12, 2004

> The cdrom is usually more often /mnt/cdrom rather than /cdrom, isnt't it ? Yes, it usually is under /mnt/, also, the floppy is /mnt/floppy under mandrake, and probably other distros. Ivan.

1 0

Re: Need to disable DTR on serial port
by Geoffrey Hausheer April 12, 2004

April 12, 2004

On Mon, 12 Apr 2004 09:26:30 +0200, "Rein Klazes" said: > > 1) it might be better to call COMM_WhackModem() after tcsetattr() so > > previous flow control settings don't interfere with setting the RTS/DTS > > lines. > > > > 2) #ifdef guards around TIOCM_DTR and TIOCM_RTS? > > > > Agreed, I will fix this in a moment. > Well, I tried Rein's latest patch out, and it still doesn't let me communicate with the device (using either my test program or TzMax). I am not sure if the patch isn't working (I can see the relevant code being called in winedbg), if there is something else required that wine isn't doing, or if it is a timing issue. In case it is interesting, here is my test app. This appears to be the minimum set of commands needed to communicate with the device. Note the tight write/query loop. Adding even a 'Sleep(1)' within that loop prevents the communication from working in Windows. I'm open to any other suggestions. Thanks, .Geoff

1 0

RE: Handling dialog messages for non-template based dialogs
by Santosh Siddheshwar April 12, 2004

April 12, 2004

OK. I will resend the fix after incorporating your feedback. Thanks Santosh Siddheshwar > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)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 April 12, 2004

April 12, 2004

"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 April 12, 2004

April 12, 2004

"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 April 12, 2004

April 12, 2004

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

Re: win32 build error log
by Ivan Leo Murray-Smith April 11, 2004

April 11, 2004

> 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 April 11, 2004

April 11, 2004

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 April 11, 2004

April 11, 2004

>> 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 April 11, 2004

April 11, 2004

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

2 1

Re: exec-shield workaround take 2
by Dan Kegel April 11, 2004

April 11, 2004

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 April 11, 2004

April 11, 2004

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

abnt2 keyboard fix
by Rafael Ávila de Espíndola April 11, 2004

April 11, 2004

-----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

Abi Word ported to Winelib!
by Dimitrie O. Paun April 10, 2004

April 10, 2004

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 April 10, 2004

April 10, 2004

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 April 10, 2004

April 10, 2004

> 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 April 10, 2004

April 10, 2004

"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 April 10, 2004

April 10, 2004

> 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 April 10, 2004

April 10, 2004

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 April 10, 2004

April 10, 2004

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 April 9, 2004

April 9, 2004

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 April 9, 2004

April 9, 2004

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

Serial Port Problem
by TN April 9, 2004

April 9, 2004

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 April 9, 2004

April 9, 2004

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 April 9, 2004

April 9, 2004

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 April 9, 2004

April 9, 2004

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

Re: ShellExecute IE compatibility problems
by Martin Fuchs April 9, 2004

April 9, 2004

On 09.04.2004 02:03:37 Francois Gouget wrote: > > This patch fixes two IE compatibility problems. > > The first one is a mundane issue. IE calls ShellExecuteEx with a NULL > verb. Our implementation fails when it should use some sophisticated > algorithm described in the MSDN to choose a default verb. Well most of > the time we just want 'open' anyway so that's what this patch does. I've > left it as an exercise to the reader to head over to the MSDN and > implement the series of fallbacks we're supposed to implement. This is what SHELL_FindExecutable() tries to do when enumerating the registry keys. What is wrong with that - doesn't this work for you? I don't think we should go back and use "open" again if the verb parameter is NULL. Please correct the current algorithm, if it's not fully compliant. Regards, Martin

1 0

Re: Workaround a SoundBlaster 16 OSS bug
by Andreas Mohr April 9, 2004

April 9, 2004

Hi, On Fri, Apr 09, 2004 at 02:27:19AM +0200, Francois Gouget wrote: > Note: AFAIK Andreas' kernel patch has never been applied :-( In fact it has been, fortunately, just about two weeks ago... By Andreas Kies in 2.4.26-pre3. He'd already patched 2.6.x a bit earlier. I've been trying to get the slightest bit of attention about that DoS at that time, but after two very detailed fruitless mails to LKML, I just gave up: http://www.uwsg.iu.edu/hypermail/linux/kernel/0206.0/0087.html http://www.uwsg.iu.edu/hypermail/linux/kernel/0206.2/0033.html Andreas Kies told me that it seemed that they now cared a slight wee bit more about small fringe problems such as DoSes in 2.4.x... </sarcasm> > (now you know how to mount a DOS on these systems!) Ow, watch your words! This can be VERY easily misinterpreted as mounting a DOS partition somewhere!? (in fact I did fall for it the first couple of seconds) > Patch the fragment size if it does not match the current sound > format. This saves us from trigerring a kernel freeze on some OSS 2.4 + > SounBlaster 16 systems. Crap, I thought I'd sent a Wine patch at that time; it would have been the easiest thing to do. Anyway, glad to see it fixed now. Greetings, Andi (been finishing my wireless driver any single day for about 5 months now :))

2 1

Re: Regression
by Ivan Leo Murray-Smith April 8, 2004

April 8, 2004

> Tried to install on an updated (like 45 min old) CVS version to test it, > seems to get in an endless loop here. Alexandre committed a fix http://www.winehq.com/hypermail/wine-cvs/2004/04/0095.html Everything seems OK again, and hopefully will stay so. Ivan.

2 1

Re: Regression
by Ivan Leo Murray-Smith April 8, 2004

April 8, 2004

> Tried to install on an updated (like 45 min old) CVS version to test it, > seems to get in an endless loop here. You're wright, it loops :( Ivan.

1 0

Re: Re: inheriting exec-shield avoidance
by chmorgan＠charter.net April 8, 2004

April 8, 2004

Any idea when the kernel and wine changes will be going in? Then we can start asking people with Fedora to disable exec-shield and prelinking OR upgrade to kernel 2.6.X and the latest wine. There are people everyday that come onto #winehq with problems due to prelinking/exec-shield. Thanks, Chris > > From: "Dimitrie O. Paun" <dimi(a)intelliware.ca> > Date: 2004/04/08 Thu PM 03:10:35 EDT > To: Mike McCormack <mike(a)codeweavers.com> > CC: Peter Riocreux <par+wine_devel(a)silistix.com>, <wine-devel(a)winehq.org> > Subject: Re: inheriting exec-shield avoidance > > On Thu, 8 Apr 2004, Mike McCormack wrote: > > > I've attached the source code. The idea is to run it like this: > > > > wld /home/mike/wine/loader/wine-pthread my.exe > > Cool. One small nit: for consistency, can we name it wineld instead? > Or is it winepreld more apropriate? I'm thinking that wineld will > seem to be the counterpart of winegcc... > > -- > Dimi. > > > >

1 0

RE: Status of TAPI
by Robert van Herk April 8, 2004

April 8, 2004

> This (dutch..) site provides some TAPI patches for Wine that are not > in CVS: > http://home.wanadoo.nl/wijn/girotel/ > > Since I'm dutch, I'd thought I'd explain a bit on what this site says. Wijn is dutch for wine by the way . The site explains how to patch wine in order to be able to use girotel software. Girotel is a telebanking system using dail up by modem that was quite popular a few years ago. Anyway, now-a-days most people use Internet for telebanking, not dail up modem. Although the site for that specific bank is not mozilla compatible , it can be used from Konqueror. The patches seems to be specially suited in order to use this Girotel system. Greetings, Robert

1 0

Status of TAPI
by Andreas Rosenberg April 8, 2004

April 8, 2004

Hi guys! I'd like to run a TAPI based application in Wine (updated from CVS today). I extracted the vendor specific TAPI driver and the registry keys and included both into a Wine setup. Then I tried to do setup the TAPI driver using the telephon.cpl (taken from Win98): wine rundll32.exe shell32.dll,Control_RunDLL telephon.cpl first I had a problem with the Wine implementation of the TAPI32.DLL. err:module:import_dll No implementation for TAPI32.dll.internalConfig imported from L"C:\\windows\\system\\telephon.cpl", setting to 0xdeadbeef ... No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\TELEPHON.CPL' (0x75a70000) Unhandled exception: page fault on read access to 0xdeadbeef in 32-bit code (0x404c5b7e). who had dug in that DEAD BEEF? ;-) Then I took the TAPI32.DLL from a Win98 installation. This gave me another error: fixme:ole:RpcServerRegisterAuthInfoA ((null),10,(nil),(nil)): stub fixme:seh:EXC_RtlRaiseException call to unimplemented function rpcrt4.dll.NdrClientCall (last message repeated several times). At last I took the rpcrt4.dll from a Windows98 installation and tried it again. Then it took quite a time (~15 sec) until the Wine debugger came up. wine rundll32.exe shell32.dll,Control_RunDLL telephon.cpl wine: Unhandled exception (thread 0009), starting debugger... WineDbg starting on pid 8 Can't find file for ELF 'wine-kthread' ((nil)) Loaded debug information from ELF 'wine-pthread' ((nil)) Breakpoint 1 at 0x4000dd60 (_end+0x3fefb8d0) Loaded debug information from ELF '/usr/local/lib/libwine.so.1' (0x40028000) No debug information in ELF '/lib/i686/libpthread.so.0' (0x40040000) No debug information in ELF '/lib/i686/libc.so.6' (0x40091000) No debug information in ELF '/lib/libdl.so.2' (0x401c4000) No debug information in ELF '/lib/ld-linux.so.2' (0x40000000) Loaded debug information from ELF '/usr/local/lib/wine/ntdll.dll.so' (0x401c8000) Loaded debug information from ELF '/usr/local/lib/libwine_unicode.so.1' (0x40238000) No debug information in ELF '/lib/i686/libm.so.6' (0x4032b000) Loaded debug information from ELF '/usr/local/lib/wine/kernel32.dll.so' (0x40460000) Loaded debug information from ELF '/usr/local/lib/wine/rundll32.exe.so' (0x405a2000) Loaded debug information from ELF '/usr/local/lib/wine/user32.dll.so' (0x406c0000) Loaded debug information from ELF '/usr/local/lib/wine/gdi32.dll.so' (0x407f8000) Loaded debug information from ELF '/usr/local/lib/wine/advapi32.dll.so' (0x40878000) No debug information in ELF '/usr/lib/libfreetype.so.6' (0x408b7000) No debug information in ELF '/usr/lib/libfontconfig.so.1' (0x4090d000) No debug information in ELF '/usr/lib/libexpat.so.0' (0x40935000) Loaded debug information from ELF '/usr/local/lib/wine/x11drv.dll.so' (0x40955000) No debug information in ELF '/usr/X11R6/lib/libSM.so.6' (0x4001c000) No debug information in ELF '/usr/X11R6/lib/libICE.so.6' (0x409d3000) No debug information in ELF '/usr/X11R6/lib/libXrandr.so.2' (0x405b6000) No debug information in ELF '/usr/X11R6/lib/libXrender.so.1' (0x409ea000) No debug information in ELF '/usr/X11R6/lib/libXext.so.6' (0x409f2000) No debug information in ELF '/usr/X11R6/lib/libX11.so.6' (0x40a00000) No debug information in ELF '/usr/lib/tls/libGL.so' (0x40afc000) No debug information in ELF '/usr/lib/tls/libGLcore.so.1' (0x40b67000) No debug information in ELF '/usr/X11R6/lib/libXcursor.so.1' (0x410e4000) Unsupported debug information in ELF '/usr/X11R6/lib/X11/locale/lib/common/xlcDef.so.2' (0x40025000) Unsupported debug information in ELF '/usr/X11R6/lib/X11/locale/lib/common/ximcp.so.2' (0x41131000) No debug information in ELF '/usr/lib/gconv/ISO8859-1.so' (0x408a9000) Loaded debug information from ELF '/usr/local/lib/wine/imm32.dll.so' (0x41150000) Loaded debug information from ELF '/usr/local/lib/wine/shell32.dll.so' (0x4116e000) Loaded debug information from ELF '/usr/local/lib/wine/ole32.dll.so' (0x411f6000) Loaded debug information from ELF '/usr/local/lib/wine/shlwapi.dll.so' (0x4126a000) Loaded debug information from ELF '/usr/local/lib/wine/comctl32.dll.so' (0x412c1000) No debug information in 32bit DLL 'C:\windows\system\rundll32.exe' (0x405b0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\NTDLL.DLL' (0x401e0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\KERNEL32.DLL' (0x40490000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\ADVAPI32.DLL' (0x40890000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\GDI32.DLL' (0x40810000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\USER32.DLL' (0x406e0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\X11DRV.DLL' (0x40970000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\IMM32.DLL' (0x41160000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\RPCRT4.DLL' (0x7fb90000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\OLE32.DLL' (0x41210000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\SHLWAPI.DLL' (0x41280000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\COMCTL32.DLL' (0x412d0000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\SHELL32.DLL' (0x41190000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\MSVCRT.DLL' (0x78000000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\TAPI32.DLL' (0x7f960000) No debug information in 32bit DLL 'C:\WINDOWS\SYSTEM\TELEPHON.CPL' (0x75a70000) Unhandled exception: page fault on read access to 0xda62061d in 32-bit code (0x401f999a). In 32-bit mode. 0x401f999a (RtlpWaitForCriticalSection+0x6a [critsection.c:204] in NTDLL.DLL): movl 0x1c(%eax),%eax 204 if ((!crit->DebugInfo) || (!crit->DebugInfo->Spare[1])) continue; The wait time could be an indication that a timeout happened. Anybody has a clue what could be the problem? Regards Andy

4 5

Re: MSYS's sh.exe has trouble reading from stdin on Wine?
by Pouech Eric DMI AEI CAEN April 8, 2004

April 8, 2004

> wine --debugmsg +all -- sh.exe - does wineconsole -- sh.exe - work ? I fear it's the same old pb of not mapping win32 console semantics on standard unix files without winconsole (anyway, check for the CreateFile which returns 0xffffffff, likely on $CONIN) A+

1 0

MSYS's sh.exe has trouble reading from stdin on Wine?
by Dan Kegel April 8, 2004

April 8, 2004

Someone's having trouble building my crosstool on msys, so I tried installing msys on my Wine (a home-built 20040213). (This is slightly less crazy than installing Cygwin, I think, but it's still a pretty hard test, I bet.) The installer http://prdownloads.sf.net/mingw/MSYS-1.0.10.exe?download ran just fine until it tried to run the postinstall shell script. Turns out running msys's sh.exe hangs. Running it with wine --debugmsg +all -- sh.exe - and waiting until the output settles into an infinite repeat, I see the repeating section is 0009:Call kernel32.WaitForSingleObject(00000084,0000000a) ret=7105f927 0009: select( flags=12, cookie=0x4060eea8, timeout=1081401202.858741 (+0.009956), handles={0x84} ) 0009: select() = PENDING 0009: *wakeup* signaled=258 cookie=0x4060eea8 0009:Ret kernel32.WaitForSingleObject() retval=00000102 ret=7105f927 0009:Call ntdll.RtlEnterCriticalSection(710a8b0c) ret=71070c98 0009:Ret ntdll.RtlEnterCriticalSection() retval=00000000 ret=71070c98 0009:Call ntdll.RtlLeaveCriticalSection(710a8b0c) ret=71070cb0 0009:Ret ntdll.RtlLeaveCriticalSection() retval=00000000 ret=71070cb0 0009:Call kernel32.PeekConsoleInputA(ffffffff,4060f150,00000001,4060f12c) ret=7105687a 0009: read_console_input( handle=0xffffffff, flush=0 ) 0009: read_console_input() = OBJECT_TYPE_MISMATCH { read=-1, rec={} } 0009:Ret kernel32.PeekConsoleInputA() retval=00000000 ret=7105687a 0009:Call ntdll.RtlEnterCriticalSection(710a8b0c) ret=71070c98 0009:Ret ntdll.RtlEnterCriticalSection() retval=00000000 ret=71070c98 0009:Call ntdll.RtlLeaveCriticalSection(710a8b0c) ret=71070cb0 0009:Ret ntdll.RtlLeaveCriticalSection() retval=00000000 ret=71070cb0 0009:Call kernel32.WaitForSingleObject(00000084,0000000a) ret=7105f927 So it kind of looks like it's having trouble reading from stdin? Can someone have a look? It'd be so awesome to be able to run msys on Wine... - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

1 0

Re: abnt2 keyboard
by Dmitry Timoshkov April 7, 2004

April 7, 2004

"Rafael аvila de EspМndola" <rafael.espindola(a)ic.unicamp.br> wrote: > The attached patch fixes the problem that wine has with the abnt2 (Brazilian) > keyboard. The problem was recently discussed in wine-devel. If you still can't provide more arguments as on wine-devel I don't see a reason why the patch can be accepted. Especially taking into account that the patch contains magic numbers without a little bit of explanation what they are. > While reading dlls/x11drv/keyboard.c I notice that wine does a linear search > to convert a virtKey to a keycode. Could this be optimized/simplified by > pre-computing a vkey2keyc[(e->state & NumLockMask)!=0] table? Feel free to send a patch. -- Dmitry.

1 0

[Fwd: DSD_CreateSecondaryBuffer]
by Robert Reif April 7, 2004

April 7, 2004

sorry, wrong reply button again.

1 0

DSD_CreateSecondaryBuffer
by Daniel Skorka April 7, 2004

April 7, 2004

Hi, said function is a stub in Wine-20040309. Is anyone working on implementing it? If not, any pointers on doing so, especially on how this fuction differs from DSD_CreatePrimaryBuffer? Thanks, Daniel Skorka

1 0

Regression
by Ivan Leo Murray-Smith April 7, 2004

April 7, 2004

This patch http://www.winehq.com/hypermail/wine-cvs/2004/03/0434.html breaks the decompression system used by lots of microsoft software, a screenshot is attached. This breaks the installer of the word 97 viewer, and probably of all office viewers, and it may affect dcom and the installer of other Windows components or Microsoft software. This should be fixed before the next release. The CVS regression testing was done with the word viewer available at http://www.microsoft.com/downloads/details.aspx?FamilyID=9bbb9e60-e4f3-436d… Ivan.

2 1

Re: Documentation: Clarified need for conformance tests.
by Francois Gouget April 7, 2004

April 7, 2004

On Wed, 7 Apr 2004, Michael Jacobsen wrote: > Hi there. This is my first go at submitting a patch. Let's see if it works. It needs a small s/avalible/available/ and it would be nice to follow the 2-space indentation used in the rest of the doc. Besides that it looks good. Thanks for your contribution. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ Any sufficiently advanced Operating System is indistinguishable from Linux

1 0

article
by Pouech Eric DMI AEI CAEN April 7, 2004

April 7, 2004

read on from lwn.net http://www-106.ibm.com/developerworks/linux/library/l-wine/?ca=dgr-lnxw04Fi…

1 0

Virus on wine-patches@winehq.org
by Rolf Kalbermatter April 7, 2004

April 7, 2004

Could someone please remove the following postings from the archive as they contain viruses? http://www.winehq.org/hypermail/wine-patches/2004/04/0082.html http://www.winehq.org/hypermail/wine-patches/2004/04/0084.html Rolf Kalbermatter

3 2

Re: development speed
by Ivan Leo Murray-Smith April 7, 2004

April 7, 2004

> The website stats are interesting too: Are these available somewhere? Ivan.

2 1

arch tip #1
by Mike Hearn April 7, 2004

April 7, 2004

At Dimis request, I've added a new script to the root of the gateway tree. The genpatch program takes a series of patch ids as parameters and can generate and send an email to wine-patches with the combined diff of all the given commits. For example: let's say you are working on a feature. You do 3 commits, to keep the changes physically separate, however you decide the commits are too fine grained to submit all at once, so you want to submit them together. Genpatch makes that easy. Say you commit patch-4, patch-5 and patch-6: ./genpatch -m 4 5 6 will compute the combined differential of all those commits, take the commit log and summary of the first and generate an email from them. The patch will include and new files that were added. Have fun! thanks -mike

1 0

Re: wine/ include/config.h.in dlls/x11drv/xrandr.c ...
by Mike Hearn April 7, 2004

April 7, 2004

On Tue, 06 Apr 2004 22:57:36 -0500, Alexandre Julliard wrote: > Log message: > Mike McCormack <mike(a)codeweavers.com> > Load the XRandR extension dynamically (with help from Alex Pasadyn). Are we sure this is a good idea? Dynamic version of the X extension libraries are not available on every distro, as traditionally XFree has shipped them statically. I'm not even sure they have stable interfaces. It seems wiser to just statically link libXrandr, seeing that it's only 7.7k

1 0

Window management thoughts
by Mike Hearn April 7, 2004

April 7, 2004

So, me and Mike have been discussing some ideas for window management in Wine. Currently it hasn't turned into code, but I thought I'd write up what our thoughts were so others could comment and maybe be inspired to write patches. The problem: Currently Wine decides whether to make a window managed or not based on a series of heuristics, in is_window_toplevel in dlls/x11drv/window.c Unfortunately these heuristics are frequently wrong, as they have only limited information to work with (window styles, mostly). In CrossOver, is_window_toplevel is easily the most hacked function in the entire codebase. This manifests itself in a variety of ways: * Windows are marked unmanaged when they shouldn't be. For instance, see WinAmp, Trillian, the Counter-Strike CD Key screen etc: there are tons of examples of this problem. This at best means the window is "nailed" to the screen, at worst it makes the program unusable. * The classic "my game starts but I can't type" problem where keyboard input goes to the terminal instead of the window, due to the lack of focus management in unmanaged windows. * Some programs display splash screens and then pop up errors or question dialogs underneath them. Unmanaged windows are always-on-top -> you cannot get to the underlying managed window and the app appears to hang. * Fullscreen windows are really hard to do properly because they must be managed to work correctly in KDE/GNOME. We currently make certain windows unmanaged typically because we need to control their position or rendering with a high degree of accuracy. Unfortunately some X window managers ignore positioning hints, even when we ask them not to. There is no way to make things like menus and tooltips work correctly with these WMs, so we take the window out of the usual window management mechanism entirely. Unfortunately due to the design of the win32 API there is no simple way to tell the difference between a large menu and the Trillian main window. We could mark our own menus specially to make them unmanaged and have everything else be managed, but we then run into problems with programs that draw their own menus and tooltips, of which there are a truly shocking number. So what is a possible solution? Well, one way forward is to implement another mode, in which Wine makes all windows managed and uses a variety of WM hints to get the desired behaviour. For instance, the PPosition flag asks the WM to place the window where the application requests it to. Some WMs respect this, others do not, in yet others it's a toggleable option. For people with WMs that meet the requirements of Wine all windows could be made managed, and for people that don't use such WMs the old way can still be used. IIRC it's possible to ask the WM for identification which means this setting could be largely autodetected by having blacklists of WMs to use the old mode for (or vice-versa). Unfortunately there's no way to know how well this approach would work, or even if it would work at all, short of trying it. thanks -mike

5 6

Re: Re: wineserver: correctly free the debug ctx
by Pouech Eric DMI AEI CAEN April 7, 2004

April 7, 2004

> It seems to me this will cause the context to never be freed. I don't > understand the problem with multiple processes, the context is bound > to the debugger thread so it shouldn't make any difference how many > processes are being debugged. Could you please give more details on > the problem? well, problem (sigsev in wineserver) happened when quitting windbg after debugging several processes. I'll try to reproduce it over this weekend. A+

1 0

Re: Game support
by James Perry April 7, 2004

April 7, 2004

Sorry, meant to send this to the list first time around... >> But I narrowed it down to 3 Wine calls in the >> critical loop: SetEvent, WaitForSingleObject and ResetEvent. >> I tried wrapping each of these functions with >> __asm__("pushfl\n"); at the start and __asm__("popfl\n"); at >> the end to preserve the flags - with this, PECrypt works >> fine and the game gets as far as trying to initialise >> DirectX. Not exactly a neat fix though. > > > Did you have to do all of them, or is it just one function that needs this > treatment? SetEvent definitely needs it. The others possibly don't. But I think it would be more robust to preserve the flag through all of them anyway, in case someone makes a change in future that causes one of the others to clear the direction flag too, which seems quite possible. > Are you sure it's not that we need to set the flag before > jumping to the entrypoint? The flag is normally cleared. PECrypt sets it explicitly, then expects it to remain set across those three API calls. > PECrypt sounds like it should run very early on > in the programs lifetime: how much stuff does it do before hitting this > bug? AFAIK, PECrypt just decrypts and decompresses the executable at startup, then it goes away and the program runs as normal. This bug does happen very early on. > If there is really widespread software that depends on > this behaviour though there is precedent for including a hack for it - see > the DllMain wrapper which preserves %ebx for an example. Certainly anything compressed with PECrypt will depend on the flag being preserved. In general I'd imagine it's quite rare for programs to run with the direction flag set for long periods though. Cheers, James

1 0

ole32.dll_CoRegisterPSClsid
by Paul Ortyl April 6, 2004

April 6, 2004

Hi, If only ole32.dll.CoRegisterPSClsid() is implemented (empty stub returning '1' is enough) then the application "GeoGeogrid® for Windows©" starts and allows at least some basic operations. (tested with "Baden-Würtemberg AMTLICHE TOPOGRAPHISHE KARTEN" Top50, Version 4) For more info on GeoGrid you may look at: http://swpat.ffii.org/patents/effects/geogrid/index.en.html http://www.eads.net/frame/lang/en/xml/content/OF00000000400004/4/20/534204.… Basically all official topographical maps in Germany are available only through this proprietary application. It would be really useful if ole32.dll.CoRegisterPSClsid() could be implemented in (next) release... TIA Paul Ortyl -- Don't take life too seriously; you will never get out of it alive. -- Elbert Hubbard

1 0

AbiWord on ReactOS
by Steven Edwards April 6, 2004

April 6, 2004

Note there are still a few minor bugs but for the most part it seems we have a real application working on ReactOS. Thanks Steven __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/

1 0

Re: WININET: move to Windows sockets
by Juan Lang April 6, 2004

April 6, 2004

> Geoff wrote: > > Or we write our own BIO that wraps up whatever > > glue is desirable on the wine-side. Worst-case > > (and it's not that bad an alternative) is that > > you use a memory-base BIO to encapsulate all I/O > > and worry about moving data to and from "sockets", > > whatever they look like, afterwards. > > That's what I've always done. <snip> > So it's not a bad way to go at all. Yep, in fact the "real" way to do this as far as Wine is concerned is to implement SSL in secur32.dll, since this is where Windows does it. It looks about like this, although as Geoff pointed out to me the API is a bit broken. Doing a BIO inside wininet.dll might be more expedient for Hans, who I think is trying to get as much of Wine as possible ported to MinGW. You guys that know SSL better than I do: patches welcome :) --Juan __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/

2 1

Re: Game support
by Ivan Leo Murray-Smith April 6, 2004

April 6, 2004

> btw, are there already any automated tools for parsing Wine logs? There is the examine-relay perl script in the tools directory, but it's only for relay traces. Ivan.

1 0

Re: development speed
by Brian Vincent April 6, 2004

April 6, 2004

> Check that out - 5mb of patches for the last 3 months running. A meg of > patches just 6 days into the new month already! > > It seems Wine is moving faster than ever before. The website stats are interesting too: Daily Avg | Monthly Totals Hits Pages Visits| Sites KBytes Visits Pages Files Hits Mar 2004 267733 75495 16582 |268931 157346721 514060 2340347 7064537 8299737 Feb 2004 278180 79217 17199 |279399 203400907 498774 2297305 6994822 8067232 Jan 2004 135477 44930 8341 |154109 90054340 258586 1392848 3591136 4199791 Dec 2003 213993 71246 14408 |243882 138401056 446653 2208643 5725553 6633804 Nov 2003 244427 85277 15766 |254363 153748583 472983 2558310 6268477 7332826 Oct 2003 275663 111913 17352 |260400 161803798 537913 3469326 7430846 8545582 Sep 2003 156860 50508 10393 |173365 93441224 311790 1515269 4048260 4705814 Aug 2003 209530 69084 12812 |218040 124270932 397177 2141613 5612004 6495449 Jul 2003 243646 105136 12253 |197296 123038700 379853 3259216 6704428 7553047 Jun 2003 253998 109167 12250 |160468 91985851 306270 2729185 5642823 6349961 May 2003 262218 111683 13764 |207954 105320483 399181 3238809 6336486 7604349 Skipping to the KBytes and Hits columns, things are up. Even more interesting is in Feburary we got Slashdotted and in March we didn't. So it appears people are interested. Generally things begin to taper off as we approach summer. Last year we got some good PR by doing an interview series, so I'm planning on doing that again. Anyone know what happened on October 5th and 6th that generated so many hits? -Brian

1 0

Compiling Wine with BiDi support - instructions
by Shachar Shemesh April 6, 2004

April 6, 2004

Hi list, Here are the formal instructions for compiling Wine with bidi support. When doing so, it is recommended that you use a fairly recent version of ICU (2.6 and up), or else there is going to be a runtime soft dependancy on some ICU files in the resulting Wine. No big deal - if these files are not there, Wine will work just fine, only without BiDi support. Still, this is not necessary. Instructions: 1. If you use a precompiled package (relatvely rare for ICU, unfortunatly), make sure it has the static libraries as well. ICU does not have these by default. If they are there, you may skip to step 8. Please note, however, that this will add about 7MB to the size of the resulting gdi.dll.so. 2. Grab ICU from the web http://oss.software.ibm.com/icu. 3. Extract the sources file. 4. CD into the ICU directory, and then into the "source" subdirectory. 5. Delete all files with the "mk" extension under icu/source/data. e.g. run "find data -name \*.mk -exec rm \{\} \;". Please trust me on this one, it's ok to do it. This step saves the aforementioned 7MB. 6. ./runConfigureICU LinuxRedHat --enable-static. If you are only building this for use with Wine, you can also add "--enable-shared=no --enable-64bit-libs=no --prefix=~/icuinstall". If not, you may wish to skip the previous step as well. 7. run "make" and "make install". Grab a lunch. Just one will do. 8. If your ICU library files are installed in /usr/lib and /usr/include, compile wine as usual. If they are installed anywhere else, you will need to set the "ICU_LIB_DIR" environment variable to wherever it was that they are installed. You will also have to add the include path. For example, if you compiled and installed icu according to my example above, you will need to set ICU_LIB_DIR to "~/icuinstall/lib", and also CPPFLAGS to "-I~/icuinstall/include". 9. Make sure that including "ubidi.h" and the following linking step passes in "configure". Also check whether include/config.h has "HAVE_ICU" set. 10. To be really sure that your wine supports BiDi, run the program given in this email. A few things to note: * The above assumes that your Wine already has http://www.winehq.org/hypermail/wine-patches/2004/04/0088.html integrated. * If you are compiling Wine on Debian, and want to use the Debian supplied ICU library for whatever reason (for example - because the packaging system pretty much requires you to), the above procedure needs to be slightly altered. ICU changed their static library naming scheme over some recent version. The Debian packaged ICU (2.1) still uses the old scheme. Either use Wine without the aforementioned patch, or set ICUUC_LIB and ICUDATA_LIB to the full path to libicuuc.a and libicudata.a respectively (they are called "libsicuuc.a" and "libsicudata.a" in newer version of ICU). Please note that this means there is a runtime dependancy on some ICU files for BiDi to work, as explained above. Instead of reposting the program for testing whether Win32 supports reordering, the link to the file in the list's archive is at http://www.winehq.org/hypermail/wine-devel/2003/08/att-0175/01-biditest.c. Enjoy -- Shachar Shemesh Lingnu OpenSource Consulting http://www.lingnu.com/

1 0

Re: Dos INT15
by Dmitry Timoshkov April 6, 2004

April 6, 2004

"Florian Goth" <Captainsifff(a)gmx.de> wrote: > Added some more stubs after examining the dosbox source code Could you please try to stick to the formatting style used in the file in your changes? -- Dmitry.

1 0

development speed
by Mike Hearn April 6, 2004

April 6, 2004

Hey guys, http://www.winehq.org/hypermail/wine-patches/ Check that out - 5mb of patches for the last 3 months running. A meg of patches just 6 days into the new month already! It seems Wine is moving faster than ever before. To put things in perspective, the Wine source tree is 10mb - that means in only 2 months as many lines of patch were generated as exists in Wine itself! thanks -mike

2 2

Re: WININET: move to Windows sockets
by Dan Kegel April 6, 2004

April 6, 2004

Geoff wrote: > Or we write our own BIO that wraps up whatever glue is desirable on the > wine-side. Worst-case (and it's not that bad an alternative) is that you > use a memory-base BIO to encapsulate all I/O and worry about moving data > to and from "sockets", whatever they look like, afterwards. That's what I've always done. And it seems I've managed (via the rant at http://kegel.com/ssl/api.html) to convince Sun to make that the standard way to do ssl with NIO under the next JDK, kinda. So it's not a bad way to go at all. - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

1 0

DDraw issues, etc.
by Tim Hentenaar April 6, 2004

April 6, 2004

Hi all, I've implemented SetClipList and GetClipList in the DirectDraw Clipper, and I'm wondering how I should go about implementing clipping in DirectDraw's Blt() (if it isn't already done somewhere). M$'s documentation isn't too great, so I figured I should ask around. There's a game one of my relatives wants to run under Linux (it's a children's program), and the major problem is that it dosen't seem to handle clipping properly when blitting. (e.g. every time you move the mouse, it calls SetClipList). FYI, the binkw32.dll that came with the program was originally causing it to hang, but I patched the DLL and it works fine. If anyone else has had such problems, I could provide a patch. Also, I'm going to take a look at some of the DirectSound stuff too. The sound isn't very good quality for some reason, but under the WineX 3.3 CVS, sound works perfectly. Tim

3 11

Re: WININET: move to Windows sockets
by Juan Lang April 5, 2004

April 5, 2004

Hans wrote: > On Monday 05 April 2004 19:24, Alexandre Julliard > wrote: > > There's no real reason that we cannot use Unix > > sockets on Unix and Windows socket on Reactos with > > the same code, it just needs a bit of header magic; > > and this will be useful for Winelib apps too. > > Hmm, I'd like to agree with that. I'd be rather nervous about the mixing of OpenSSL with the use of Winsock, at least as is. OpenSSL is itself using socket read/write calls, and it expects them to be the correct ones: in particular, the UNIX ones. I believe the only way to do this correctly is to write a new OpenSSL BIO that uses the same read/write/open/close calls as the rest of netconnection.c, and this can be combined with the aforementioned header magic to get what you're after. Copying and pasting an existing OpenSSL BIO would probably suffice. --Juan __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/

2 1

Re: wineserver: correctly free the debug ctx
by Alexandre Julliard April 5, 2004

April 5, 2004

Eric Pouech <pouech-eric(a)wanadoo.fr> writes: > when a debugging thread is attached to several processes, the debug > context (shared across all processes) is destroyed when the first > process terminates. This patch destroys it when the last process > attached terminates. It seems to me this will cause the context to never be freed. I don't understand the problem with multiple processes, the context is bound to the debugger thread so it shouldn't make any difference how many processes are being debugged. Could you please give more details on the problem? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: WININET: move to Windows sockets
by Steven Edwards April 5, 2004

April 5, 2004

Hello Hans, --- Hans Leidekker <hans(a)it.vu.nl> wrote: > > This makes wininet build on MinGW. > > Changelog: > Move to Windows sockets. > I asked Alexandre about a patch to move Wininet to Winsock and he said it took to much of a performance hit to merge. Is there a way we can bench mark it to show how much of a hit Winsock vs Unix sockets is going to give us? He suggested that we find a way to make Wininet use Unix sockets on Unix and Winsock on Windows. Thanks Steven __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/

6 7

CryptoAPI
by unknown＠example.com April 5, 2004

April 5, 2004

Has anybody started implementing CryptoApi in WINE?? If no mayby someone can give me some clues how to start... I am not bad i C++ I think, but I have never make thing like that. I don't promise I will do that, but I can try... :)

3 2

Re: CryptoAPI
by Juan Lang April 5, 2004

April 5, 2004

Jacek wrote: > Has anybody started implementing CryptoApi in WINE?? > If no mayby someone can give me some clues how to > start... I am not bad i C++ I think, but I > have never make thing like that. I don't promise I > will do that, but I can try... :) Hi Jacek, take a look at Wine's rsabase.dll, that DLL doesn't do very much yet. All its functions (CPAcquireContext, etc) are described here: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/security/s… If you have access to a Windows box it's easier: you can load the DLL itself and call all the functions and try to reproduce what it's doing. Good luck, --Juan __________________________________ Do you Yahoo!? Yahoo! Small Business $15K Web Design Giveaway http://promotions.yahoo.com/design_giveaway/

1 0

Re: Fwd: Re: appdb - improvement in serching out bugs
by tony_lambregts＠telusplanet.net April 5, 2004

April 5, 2004

Quoting hatky <hatky(a)hatky.myz.info>: >> No, no, no... read the patch again. > Ok, sorry wasn't what I tought you are trying to do, this patch is ok > p.s. I didn't think someone actually changed the urls in bugzilla. sweet :). I have been going through them 100 at a time. I've got 1-400 done at this time. I figure I will be done sometime before Christmas :P. If anyone wants to help with this I would be glad to have the help... Check out http://bugs.winehq.org/show_bug.cgi?id=560 for more more on this. -- Tony Lambregts

1 0

NtPulseEvent
by Robert Lunnon April 5, 2004

April 5, 2004

What is the status of NtPulseEvent it emits a diagnostic messages fixme:ntdll:NtPulseEvent (7c,0) yet it looks to me that NtPulseEvent is implemented but the fixme is bogus Anyone know about this ? BTW Has anyone experience with the Skype application under linux, does it work there ? Bob b

2 1

Should winemaker be able to handle this
by Bill Medland April 5, 2004

April 5, 2004

I haven't been following the winemaker stuff so I don't know whether it is supposed to be able to handle this (and is broken) or not. I am having to play with the "build a builtin dll to wrap a linux so" concept again after a year or so since the last time, as discussed in the Winlib manual, chapter 5. I see that winemaker has changed since then. In particular the bit that is not working for my case is the -D, -I, -L and -l options. What I want is to run winemaker to generate the build environment and then do the build. What I used toi have was: #! /bin/bash # Run this script in order to pass the correct arguments to winemaker so that # it can set up the build environment winemaker --nosource-fix --nogenerated-specs --dll --single-target mytarget --nomfc -DMY_SETTING -I$MY_INC_DIR -L$MY_LIB_DIR -lmylinuxso . I see that we don't bother with generating spec files any longer. However I see that the -D, -I, -L and -l options don't find their way into the Makefile these days. Is that by design or a bug? (i.e. do I have to fix winemaker or do I need to look at a different way of doing it) -- Bill Medland mailto:billmedland(a)mercuryspeed.com http://webhome.idirect.com/~kbmed

4 5

RE: Handling dialogs created without using templates
by Santosh Siddheshwar April 5, 2004

April 5, 2004

Hi, Thanks. Would you be able to share details of the changes you did for handling MDI windows? Regards Santosh > -----Original Message----- > From: Dmitry Timoshkov [SMTP:dmitry(a)baikal.ru] > Sent: Tuesday, March 30, 2004 7:40 PM > To: Santosh Siddheshwar > Cc: wine-devel(a)winehq.com > Subject: Re: Handling dialogs created without using templates > > "Santosh Siddheshwar" <santosh.s(a)sonata-software.com> wrote: > > > Yes attached is a sample which is pretty similar to what I described > > below. > > Thanks. Now I see where the problem is. We have to hook dialog creation > code into common window creation code, something similar to what I've done > recently for MDI. Alexandre, are my thoughts correct? > > -- > 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. *********************************************************************

2 1

Re: Re: dbghelp
by Pouech Eric DMI AEI CAEN April 5, 2004

April 5, 2004

> UnDecorateName is still a stub, while our winedump has some a good > framework for UnDecorateName, also translating into a GCC way and not in a > VC++ way. How should these things go together? See the comment in dbghelp.c !! My first thouhgt would be to make a .a lib out of winedump code, and use it in winedump and msvcrt and / or dbghelp. I haven't studied yet the best way to share the code between msvcrt and dbghelp A+

1 0

Re: appdb - improvement in serching out bugs
by tony_lambregts＠telusplanet.net April 4, 2004

April 4, 2004

Quoting hatky <hatky(a)hatky.myz.info>: > > On Sunday 04 April 2004 06:21, tony_lambregts(a)telusplanet.net wrote: > > The current code can get bugs that have nothing to do with the AppDB. > This > > should fix it. > > > > -- > > > > Tony Lambregts > > I think this is wrong, you are saying the url in bugzilla should be exactly > "appdb #" where it should realy be > "http://bugs.winehq,org/show_bug.cgi?id=#" > > No, no, no... read the patch again. The current code is correct as far as it goes, but it grabs the substring. So when the appdb# is 68 it grabs bugs that for appid 680, 681... 689. It will also grab bugs that just happen to have 68 in the url. My testing with this has been limited somewhat since I have not been able to set up the appdb on my machine but I know enough to know that the code as I have submitted will do what it needs to do. -- Tony Lambregts

2 1

Re: appdb - improvement in serching out bugs
by tony_lambregts＠telusplanet.net April 4, 2004

April 4, 2004

Quoting hatky <hatky(a)hatky.myz.info>: > > On Sunday 04 April 2004 06:21, tony_lambregts(a)telusplanet.net wrote: > > The current code can get bugs that have nothing to do with the AppDB. > This > > should fix it. > > > > -- > > > > Tony Lambregts > > I think this is wrong, you are saying the url in bugzilla should be exactly > "appdb #" where it should realy be > "http://bugs.winehq,org/show_bug.cgi?id=#" > > No, no, no... read the patch again. The current code is correct as far as it goes, but it grabs the substring. So when the appdb# is 68 it grabs bugs that for appid 680, 681... 689. It will also grab bugs that just happen to have 68 in the url. My testing with this has been limited somewhat since I have not been able to set up the appdb on my machine but I know enough to know that the code as I have submitted will do what it needs to do. -- Tony Lambregts

1 0

Re: dbghelp
by Uwe Bonnes April 4, 2004

April 4, 2004

>>>>> "Eric" == Eric Pouech <pouech-eric(a)wanadoo.fr> writes: Eric> Eric Pouech a écrit : >> This is a first (and incomplete) shot at implementing the dbghelp >> DLL. Eric> forgot to say that: - this patch must be applied before the ones Eric> on imagehlp & taskmgr - code is highly derivated from winedbg Eric> current code Eric> I'm also working on a winedbg version based on this DLL, but it Eric> isn't ready yet (as dbghelp seems to get a bit more mature, I Eric> thought some people might be interested in testing it) UnDecorateName is still a stub, while our winedump has some a good framework for UnDecorateName, also translating into a GCC way and not in a VC++ way. How should these things go together? Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

1 0

Personal tree with interesting patches
by Mike Hearn April 4, 2004

April 4, 2004

Hi, As I'm pretty busy right now and won't have time to work on the various misc projects I've been working on up until recently for the foreseeable future, I'm checking them into my personal arch archive for people to play with and maybe work on themselves. My archive co-ordinates: http://navi.cx/~mike/archive/2004 mike(a)navi.cx--2004 so, you'd use: $ tla register-archive http://navi.cx/~mike/archive/2004 There are currently only two interesting branches: wine--mainline--0.9: misc patches that were never applied for whatever reason, debug delay, lockwindowupdate (disabled), that sort of thing as i find more patches around my system i'll check them in for safekeeping wine--loadarea-reserve--0.9 execshield/prelink work, see the log message in patch-1 for details on what remains to be done Have fun! thanks -mike

2 2

installation program
by Brian Vincent (C) April 4, 2004

April 4, 2004

On the To Do List (c.4) is an item for having a new graphical Wine install program. Anyone ever looked at Frank Hendrickson's WineTools program? http://franksworld.net/winetools.html It used to be written in some funky language, now it's all shell script. Anyway, if you look at the script (wt210) you'll notice he uses xdialog ( http://thgodef.nerim.net/xdialog/ ) for the GUI. So here's what I'm wondering.. How feasible would it be for Wine to create a new Winelib program (say, wdialog) and use a similar concept as Frank? wdialog would simply replicate the functionality of xdialog. We could potentially fail gracefully into shell mode if Wine can't start. It might be a cheap and simple way to get a GUI front-end. Of course, all this requires sane defaults so Wine can actually run wdialog. Any thoughts? --------------- Brian Vincent Copper Mountain Telecom vincentb(a)coppercolorado.com

1 0

Re: appdb - improvement in serching out bugs
by hatky April 4, 2004

April 4, 2004

On Sunday 04 April 2004 06:21, tony_lambregts(a)telusplanet.net wrote: > The current code can get bugs that have nothing to do with the AppDB. This > should fix it. > > -- > > Tony Lambregts I think this is wrong, you are saying the url in bugzilla should be exactly "appdb #" where it should realy be "http://bugs.winehq,org/show_bug.cgi?id=#" -- Hatky, Worshiper of wine (http://www.winehq.org/) Impossible is only an opinion.

1 0

Microsoft SDK samples
by Jonathan Wilson April 3, 2004

April 3, 2004

Has anyone given thought to using the Microsoft SDK samples (e.g. DirectX samples) to improve WINE? i.e. take a particular sample and work away at it untill what you get in WINE matches what you get in Windows. Although I suspect that there are things that the MS samples dont cover :)

2 1

Re: Documentation for GDI Region Functions
by Alexandre Julliard April 3, 2004

April 3, 2004

Robert Shearman <R.J.Shearman(a)warwick.ac.uk> writes: > and renamed the variables to Microsoft conventions. Please don't do that. The variable names don't matter at all for compatibility, so we don't have to follow the brain-damaged Microsoft conventions for them. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: Developer's path to Windows/*NIX multi-platform?
by Dimitrie O. Paun April 3, 2004

April 3, 2004

You ask: > I am willing to re-write my best Windows software from scratch, but I > insist that the result is native or almost on Linux so there are not any > weird GUI glitches. Do you think Wine is the way to go for me, or am I > better off writing individual versions and keeping the Windows software > native and then producing QT or GTK versions for *NIX? I'm at a cross-roads > here, since I'll be dedicating years of work in whatever direction I take. IMO Wine is the perfect tool for you. We know, it's not perfect. But the best use of your time (and the most efficient) would be to: 1. Fix any problems in your apps (like reliance on strange Windows corner cases, use of very obscure features) 2. Help out fix Wine so that your apps run properly. This will not only help us a lot, but it will help you make good use of your time, and it will allow you to do cross platform development (even for Windows) from your Linux box. Please direct questions to wine-devel(a)winehq.org, where people can help you out. -- Dimi.

7 9

W95INF32 implementation, incase it's useful to anybody
by Mike McCormack April 2, 2004

April 2, 2004

Hi All, In the process of understanding another bug, I did a little work to implement w95inf32, which appears to be a small IE/win9x only DLL and which is probably of not much use to Wine. It appears to be a think wrapper 32bit around 4 functions in SETUPX. So this is just for documentation purposes, and incase anybody is curious or finds a real need for this DLL. Mike --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ dlls/w95inf32/Makefile.in 2004-03-30 12:03:27.000000000 +0900 @@ -0,0 +1,20 @@ +#MKDLL_SKIP +TOPSRCDIR = @top_srcdir@ +TOPOBJDIR = ../.. +SRCDIR = @srcdir@ +VPATH = @srcdir@ +MODULE = w95inf32.dll +IMPORTS = setupapi kernel32 +ALTNAMES = w95inf16.dll + +SPEC_SRCS16 = $(ALTNAMES:.dll=.spec) + +C_SRCS = \ + w95inf32.c + +(a)MAKE_DLL_RULES@ + +w95inf16.spec.c: w95inf16.spec + $(WINEBUILD) $(DEFS) $(DLLFLAGS) -o $@ --main-module $(MODULE) --spec $(SRCDIR)/w95inf16.spec + +### Dependencies: --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ dlls/w95inf32/w95inf32.c 2004-03-29 17:07:20.000000000 +0900 @@ -0,0 +1,91 @@ +/* + * W95INF32 - This is a small 32bit interface to SETUPX + * + * Copyright 2004 Mike McCormack for CodeWeavers + * + * This 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. + * + * This 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 this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <stdarg.h> +#include "windef.h" +#include "winbase.h" +#include "winuser.h" + +#include "setupapi.h" + +#include "wine/debug.h" + +WINE_DEFAULT_DEBUG_CHANNEL(w95inf32); + +BOOL WINAPI CtlSetLddPath32( LPSTR x, LPSTR y ) +{ + FIXME("%s %s\n", debugstr_a(x), debugstr_a(y) ); + return FALSE; +} + +BOOL WINAPI GenFormStrWithoutPlaceHolders32( LPSTR dst, LPCSTR src, LPCSTR filename ) +{ + HINF hinf; + DWORD ret = 0; + + FIXME("%p %s %s\n", dst, debugstr_a(src), debugstr_a(filename) ); + + hinf = SetupOpenInfFileA( filename, NULL, INF_STYLE_WIN4, NULL ); + if( !hinf ) + return GetLastError(); + + /* FIXME: call LookupStringInTable? */ + strncpy(dst, src, MAX_INF_STRING_LENGTH ); + + SetupCloseInfFile( hinf ); + + return ret; +} + +DWORD WINAPI GenInstall32(LPCSTR filename, LPCSTR section, LPCSTR directory, DWORD flags) +{ + HINF hinf; + void *context; + DWORD ret = 0; + + ERR("%s, %s, %s, %08lx\n", filename, section, directory, flags ); + + hinf = SetupOpenInfFileA( filename, NULL, INF_STYLE_WIN4, NULL ); + if( !hinf ) + return GetLastError(); + + context = SetupInitDefaultQueueCallback( 0 ); + if (!SetupInstallFromInfSectionA( 0, hinf, section, flags, 0, NULL, + SP_COPY_NEWER_OR_SAME, SetupDefaultQueueCallbackA, + context, 0, 0 )) + { + ret = GetLastError(); + ERR("Failed %08lx\n",ret); + } + + SetupTermDefaultQueueCallback( context ); + + SetupCloseInfFile( hinf ); + + return ret; +} + +BOOL WINAPI GetSETUPXErrorText32(long a, long b, long c) +{ + FIXME("\n"); + return FALSE; +} + --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ dlls/w95inf32/w95inf32.spec 2004-03-29 17:04:15.000000000 +0900 @@ -0,0 +1,9 @@ +# w95inf32 functions have @nn in their names... +# winebuild removes the last @xx in a function name. + +1 stdcall CtlSetLddPath32(a)8@8(str str) CtlSetLddPath32 +2 stdcall GenFormStrWithoutPlaceHolders32(a)12@12(ptr str str) GenFormStrWithoutPlaceHolders32 +3 stdcall GenInstall32(a)20@20(str str str long) GenInstall32 +4 stdcall GetSETUPXErrorText32(a)12@12(long long long) GetSETUPXErrorText32 +5 stub w95thk_ThunkData32 +

1 0