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

Re: [DMUSIC] IDirectMusicSegment8 loading of Wave Segments
by Rok Mandeljc May 26, 2004

May 26, 2004

> PS: > Rok, i have you have implemented IDirectMusicSegment8 interface on > IDirectMusicWaveImpl. > Why ? > What it your "sample" who permit to check this behavior ? > For testing my patch you only have, for example, to open a .wav file using > 3DAudio.exe sdk sample Hey, I'm sorry but I cannot really remember (have experienced computer crash and it will take some time for me to organise everything again). I'm not even sure if IDirectMusicSegment should be implemented for it, but i think my test was to create native IDirectMusicWave via CoCreateInstance and then try to QueryInterface it with all possible IIDs; I can't remember result, but I think it was supported (unless I made a mistake ;=) - Rok

2 1

Assume MAX_PATH length for pszDisplayName
by Uwe Bonnes May 26, 2004

May 26, 2004

Changelog: dlls/shell32/brsfolders.c: SHBrowseForFolderA LPBROWSEINFOA->pszDisplayName is assumed to have MAX_PATH legth Fixes a crash in Xilinx ISE when vreating a new project with a long path -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- Index: wine/dlls/shell32/brsfolder.c =================================================================== RCS file: /home/wine/wine/dlls/shell32/brsfolder.c,v retrieving revision 1.51 diff -u -w -5 -r1.51 brsfolder.c --- wine/dlls/shell32/brsfolder.c 19 Apr 2004 20:12:14 -0000 1.51 +++ wine/dlls/shell32/brsfolder.c 26 May 2004 13:56:29 -0000 @@ -460,11 +460,12 @@ bi.hwndOwner = lpbi->hwndOwner; bi.pidlRoot = lpbi->pidlRoot; if (lpbi->pszDisplayName) { len = MultiByteToWideChar(CP_ACP, 0, lpbi->pszDisplayName, -1, NULL, 0); - bi.pszDisplayName = HeapAlloc(GetProcessHeap(), 0, len * sizeof(WCHAR)); + /*lpbi->pszDisplayName is assumed to be MAX_PATH (MSDN) */ + bi.pszDisplayName = HeapAlloc(GetProcessHeap(), 0, MAX_PATH * sizeof(WCHAR)); MultiByteToWideChar(CP_ACP, 0, lpbi->pszDisplayName, -1, bi.pszDisplayName, len); } else bi.pszDisplayName = NULL;

1 0

Type in msvcrt
by Uwe Bonnes May 26, 2004

May 26, 2004

Changelog: dlls/msvcrt/file.c: _setmode() Reset only the affected flag Make Xilinx ise not hang at startup Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- Index: wine/dlls/msvcrt/file.c =================================================================== RCS file: /home/wine/wine/dlls/msvcrt/file.c,v retrieving revision 1.67 diff -u -w -5 -r1.67 file.c --- wine/dlls/msvcrt/file.c 13 May 2004 00:23:25 -0000 1.67 +++ wine/dlls/msvcrt/file.c 26 May 2004 12:17:06 -0000 @@ -1437,11 +1448,11 @@ if (mode & (~(_O_TEXT|_O_BINARY))) FIXME("fd (%d) mode (0x%08x) unknown\n",fd,mode); if ((mode & _O_TEXT) == _O_TEXT) MSVCRT_fdesc[fd].xflag |= WX_TEXT; else - MSVCRT_fdesc[fd].xflag &= WX_TEXT; + MSVCRT_fdesc[fd].xflag &= ~WX_TEXT; return ret; } /********************************************************************* * _stati64 (MSVCRT.@)

1 0

Re: Re: [OLE] Better handling of References
by fenix＠club-internet.fr May 26, 2004

May 26, 2004

>On Wed, 26 May 2004 03:55:25 +0200, Raphael wrote: >> I'm trying to install Splinter Cell Demo (installshield). >> And i have seen: >> - one typelib was released while typeinfos (childs of typelib) are in use >> - the install "crash" (catched and returns an E_UNEXPECTED error) when it >> wants to call ITypeInfo_fnGetContainingTypeLib (who want to AddRef the >> freshly released TypeLib) > >Hi Raphael, Hi Mike >This has been a known problem for a long time now, with several attempts >to fix it, all of them have been incorrect. It seems your approach is a >new one and interesting, but I'm not sure this is how Windows does it. Well Soft References are used in many places in COM (for example D3D textures/surfaces use it on windows) >Does anybody know exactly what Windows does in this situation? It might be >worth writing a test case to find out. > >Objects in a semi-freed state seem like a slightly odd way to do things. > >> Now i have another crash later, but it seems a problem with an invoke method >> who failed :( >> If any ole/rpc expert can look at this problem i'll be very happy :) > >Well, if you could provide more detail (backtrace etc) on the crash it >would be good. I can't have a backtrace as i can't get winedbg working correctly on installshield programs :( But i'll send you the traces later (when i'll be back home). For what i have see it's a problem of ITypeInfo::Invoke when it should returns a Pointer to an IUnknown Variant. InstallShield complains about Objects References not set !! >A few comments on the patch: > >- --(This->ref); >+ /** escape decrement for soft_references */ >+ if (This->ref > 0) { >+ --(This->ref); >+ } else { >+ ERR("trying to Release again an ITypeLib2 in freing state waiting for ITypeInfo (%p) childs Releases\n", This); >+ } > >I don't remember whether this object is supposed to be thread safe or not, >but should this not be an InterlockedDecrement? Yes, but as the code of ITypeLib2/ITypeInfo don't seems thread safe at all ... (anyway i'll fix it) :) >I think the error could be better phrased as: > >ERR("Reference count on an ITypeLib2 object (%p) already zero: waiting for >ITypeInfo children to be released as well"); :) >or words to that effect. > >+ if (This->soft_ref > 0) { >+ ERR("trying to Release an ITypeLib2 while some ITypeInfo (%p) childs are actually with Refs > 1. So Dont free it now \n", This); >+ /** maybe scaning ITypeInfo list to dump ITypeInfos with Refs > 1 */ >+ return 1; >+ } > >Is it actually an error to do that? If you free a typelib while refs are >held on child typeinfos, is this a bug in the app or a legitimate thing to >do? If it's not a bug in the app we probably shouldn't use ERR here. Weel, we had problems like that on d3d8 (a long time ago) about ref counting. For me, seems more a wine bug (we have to found where missing AddRefs as to be done) than an app bug (new traces can help us to find what happened) >--- variant.c 30 Apr 2004 18:32:58 -0000 1.97 >+++ variant.c 26 May 2004 01:46:44 -0000 >@@ -605,6 +605,8 @@ > > hres = VARIANT_ValidateType(V_VT(pVarg)); > >+ if (V_VT(pVarg) == VT_EMPTY) return hres; >+ > if (SUCCEEDED(hres)) > { > if (!V_ISBYREF(pVarg)) > >This change looks OK though I'm not sure why it's necessary. Does the >Splinter Cell installer actually need this? No, i have only to no have a lot of debug traces (i have added on my tree) on Variant* Operations when clearing already empty Variants >thanks -mike > Regards, Raphael

2 1

Re: [OLE] Better handling of References
by Mike Hearn May 26, 2004

May 26, 2004

Oops, sending to wine-devel too (this is what I get for using a newsreader...) On Wed, 26 May 2004 03:55:25 +0200, Raphael wrote: > I'm trying to install Splinter Cell Demo (installshield). > And i have seen: > - one typelib was released while typeinfos (childs of typelib) are in use > - the install "crash" (catched and returns an E_UNEXPECTED error) when it > wants to call ITypeInfo_fnGetContainingTypeLib (who want to AddRef the > freshly released TypeLib) Hi Raphael, This has been a known problem for a long time now, with several attempts to fix it, all of them have been incorrect. It seems your approach is a new one and interesting, but I'm not sure this is how Windows does it. Does anybody know exactly what Windows does in this situation? It might be worth writing a test case to find out. Objects in a semi-freed state seem like a slightly odd way to do things. > Now i have another crash later, but it seems a problem with an invoke method > who failed :( > If any ole/rpc expert can look at this problem i'll be very happy :) Well, if you could provide more detail (backtrace etc) on the crash it would be good. A few comments on the patch: - --(This->ref); + /** escape decrement for soft_references */ + if (This->ref > 0) { + --(This->ref); + } else { + ERR("trying to Release again an ITypeLib2 in freing state waiting for ITypeInfo (%p) childs Releases\n", This); + } I don't remember whether this object is supposed to be thread safe or not, but should this not be an InterlockedDecrement? I think the error could be better phrased as: ERR("Reference count on an ITypeLib2 object (%p) already zero: waiting for ITypeInfo children to be released as well"); or words to that effect. + if (This->soft_ref > 0) { + ERR("trying to Release an ITypeLib2 while some ITypeInfo (%p) childs are actually with Refs > 1. So Dont free it now \n", This); + /** maybe scaning ITypeInfo list to dump ITypeInfos with Refs > 1 */ + return 1; + } Is it actually an error to do that? If you free a typelib while refs are held on child typeinfos, is this a bug in the app or a legitimate thing to do? If it's not a bug in the app we probably shouldn't use ERR here. ++(This->ref); + if (This->ref > 1) { + if (NULL != This->pTypeLib) ++(This->pTypeLib->soft_ref); + } + Ditto on thread safety. Please forgive me if this object is not free threaded. --- variant.c 30 Apr 2004 18:32:58 -0000 1.97 +++ variant.c 26 May 2004 01:46:44 -0000 @@ -605,6 +605,8 @@ hres = VARIANT_ValidateType(V_VT(pVarg)); + if (V_VT(pVarg) == VT_EMPTY) return hres; + if (SUCCEEDED(hres)) { if (!V_ISBYREF(pVarg)) This change looks OK though I'm not sure why it's necessary. Does the Splinter Cell installer actually need this? thanks -mike

1 0

CoRegisterMessageFilter
by Simon Harvey May 26, 2004

May 26, 2004

hello I was wondering if anybody was doing work on or was in the process of implementing CoRegisterMessageFilter in dlls/ole32/ole2.c ? I would like to take a stab at it but i wont if somebody else is going to do it. thanks simon Find local movie times and trailers on Yahoo! Movies. http://au.movies.yahoo.com

2 1

Bad wine today.
by Rein Klazes May 26, 2004

May 26, 2004

Hi, I wonder if I am the only one that sees this with any windows prog that I try: |$ wine sol |x11drv: Can't open display: |Please ensure that your X server is running and that $DISPLAY is set correctly. |$ (other X programs run fine). I did the usual things when in trouble: a make clean, fresh cvs update. That did not help. Debugmessages "+all", does not give any messages. Version win98 or win2k makes no difference. Reverting to the yesterdays cvs, makes wine work again. Any idea? Rein. -- Rein Klazes rklazes(a)xs4all.nl

2 2

Crash and corrupted registry
by Christian Costa May 25, 2004

May 25, 2004

Hi, This is the second time I had to reboot when by box is stuck while running a windows apps. In both case the registry had been corrupted. This is very annoying. Could we have the wineserver do a backup of registry files at startup so we can retreive the previous version if a corruption occurs? That would be very usefull. Thanks, Christian

2 2

Re: Bad wine today.
by Ivan May 25, 2004

May 25, 2004

Current CVS works for me. Ivan.

1 0

Re: additional download for winehq download page
by Ivan May 25, 2004

May 25, 2004

< And if a prog doesn't provide the dll it should have at least a readme to tell people where to get it. I can tell you out of experience that this isn't always true. What about the attached patches? Ivan.

1 0

Re: Strip comments from end of line in INI parser
by Alexandre Julliard May 25, 2004

May 25, 2004

Mike Hearn <mike(a)navi.cx> writes: > Mike Hearn <mike(a)navi.cx> > Strip comments and whitespace from end of line in PROFILE_CopyString That will make it impossible to have an ini entry containing a ';'. I really doubt that Windows does it this way; could you please provide a test case? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: winegcc: Use $(CC) for building generated .spec.c
by Dimitrie O. Paun May 25, 2004

May 25, 2004

On Tue, May 25, 2004 at 12:56:27AM -0700, Jon Griffiths wrote: > Building and linking a C++ dll to a Winelib app currently fails using > winegcc as it will use $(CXX) to compile the .spec.c file. This means > it won't link correctly to libwine.so due to the name mangling. I > tried 2 solutions: changing winebuild to generate extern "C" {} > declarations where needed, and changing winegcc to use $(CC) instead. > Of the two, I think the latter is neater. The following patch > implements it. I thought that gcc will figure the langauge based on the extension. So g++ file.c would actually compile as a C file... But this may not work for other compilers. We could do it ourselves in compile(), choose $CC or $CXX based on the extension. However, I'm not sure it's better than your solution. In fact, it's a lot more risky. -- Dimi.

2 1

Proprietary Hardware and related stuff - Get them work with Linux
by Gianfranco Luceri May 25, 2004

May 25, 2004

Hello, I own a Matrox Mistyque with Rainbow Runner Studio and Rainbow Runner TV (not G-Series) add-on cards and there's no way of getting them all work with Linux due to the lack of documentation about a proprietary chip (MGA-something-I-cant-remember-any-more). The only supported OSes are Win95 (fully operational, not taking into account the Os flaws) and WinNT 4.0 (some "minor" features aren't available). Maybe WINE is the only way to get it (and similar hardware) work with Linux. Infact the driver and related software could be installed and executed with WINE, assuming that WINE could succesfully access the I/O port used by the hardware (and maybe linux is not even aware of that hardware!) Maybe I'll try that by myself and post the (maybe bad) results I'll get (at least a divorce from my wife). Any comments, hint, post wuold be appreciated. There's a lot o people owning a Rainbow Runner (not G-Series) or hardware with the same problem out there. So please help. Thank you. DonalDuck

1 0

mingw crossbuild broke
by Kevin Koltzau May 25, 2004

May 25, 2004

the mingw crossbuild broke last night with mmap.c: In function `wine_mmap_add_reserved_area': mmap.c:291: error: `PROT_NONE' undeclared (first use in this function) mmap.c:291: error: (Each undeclared identifier is reported only once mmap.c:291: error: for each function it appears in.) mmap.c:291: error: `MAP_FIXED' undeclared (first use in this function) should those defines be included in one of our headers?

1 0

failure report: picasa-1.6
by Dan Kegel May 25, 2004

May 25, 2004

I tried running the demo of Picasa from http://www.lifescapeinc.com/picasa/ under Wine-20040213 and Wine-20040505, but the installer just put up "Out of Memory" boxes :-( This progam uses some unusual installer app (it announces which one early on when you run it under Windows). Anyone seen this before? -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

4 8

Re: [DMUSIC] Make Unreal2 playable
by Dmitry Timoshkov May 25, 2004

May 25, 2004

"Raphael" <fenix(a)club-internet.fr> wrote: > @@ -131,6 +149,7 @@ > } > > static BOOL PostMessageToProcessMsgThread(IDirectMusicPerformance8Impl* This, UINT iMsg) { > + return FALSE; > if (FALSE == This->procThreadTicStarted && PROCESSMSG_EXIT != iMsg) { > This->procThread = CreateThread(NULL, 0, ProcessMsgThread, This, 0, &This->procThreadId); > if (NULL == This->procThread) return FALSE; Are you sure you want this change to be included? -- Dmitry.

1 0

Re: General serial communications issues
by David Purucker May 25, 2004

May 25, 2004

>>I have tried several terminal programs (Hyperterm and TeraTerm) under > > several builds of > > Wine including Wine-20040505 and they seem to have the same generic >problem. > >We are using serial ports under Wine for a current project, and they >work OK (apart from a few warnings). The machine I'm using doesn't >have X, so I haven't tried HyperTerminal or TeraTerm, but our >console-mode applications operate pretty much as expected. > >I'm using two patches applied to a recent CVS. The first is a patch >Eric Pouech recently posted: > > http://www.winehq.org/hypermail/wine-devel/2004/05/0297.html > >The second is a patch of my own to supress warnings when using a >pseudo-tty as a serial port: > > http://www.winehq.org/hypermail/wine-devel/2004/05/0278.html > >But our apps worked OK before that; they just generated some warnings. >Still, these patches might help, or at least cut down on the noise and >help you find the real error. > >TeraTerm has source available, so it should be one of the easier >applications to debug. > >----ScottG. Thanks Scott, for the prompt reply. I tried the patch posted by Eric Pouech and still no luck, the program still appears to receive nothing back. I can type directly into the terminal a ATDT command to dial a number and I get a connect, but never see a response on the screen. I have tested it on several different machines with different wine builds, but could someone please test this, just to make sure that I am not doing something trivially wrong. Thanks again _________________________________________________________________ Is your PC infected? Get a FREE online computer virus scan from McAfee® Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

Re: additional download for winehq download page
by Ivan May 25, 2004

May 25, 2004

> This may be a fine thing to do for another project, but this is > not a Wine thing, sorry. OK, but at least the vb runtime and the win installer should be there. Nobody is implementing, or, AFAIK, planning to implement these parts of windows at the moment, and without them many programs simply have no hope of working with wine. We may also want to list dcom, as we even had it up on the sf page at one time. Ivan.

2 1

exec-shield patch 7
by Mike McCormack May 25, 2004

May 25, 2004

Description: This patch allows platforms using exec-shield patch against the Linux kernel (eg. Fedora Core) to run Wine without disabling exec-shield. License: LGPL Changes in version 7: * fix bug that caused crash when CreateProcess was called * update to work with the new functions for reserving memory Changes in version 6: * include ulimit workaround * resolve conflict with recent patch to WineHQ CVS tip Changes in version 5: * reserve the shared heap area as well as the PE load area * allow multiple reserve areas in libwine Changes in version 4: * set the WINEPEEXEPATH environment variable correctly Changes in version 3: * 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 Changes in verison 2: * replace the shared object loader instead of using a bootstrap exe Version 1: * create a bootstrap exe that reserves memory before loading wine-pthread or wine-kthread Index: dlls/kernel/process.c =================================================================== RCS file: /home/wine/wine/dlls/kernel/process.c,v retrieving revision 1.63 diff -u -r1.63 process.c --- dlls/kernel/process.c 25 May 2004 01:29:24 -0000 1.63 +++ dlls/kernel/process.c 25 May 2004 04:59:54 -0000 @@ -1257,7 +1257,8 @@ * * Build the environment of a new child process. */ -static char **build_envp( const WCHAR *envW ) +static char **build_envp( const WCHAR *envW, + const char *workdir, const char *unixpath ) { const WCHAR *p; char **envp; @@ -1270,7 +1271,7 @@ 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) ))) { @@ -1283,12 +1284,19 @@ 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 (workdir) *envptr++ = alloc_env_string( "WINEPEEXEPATH=", unixpath ); + /* now put the Windows environment strings */ for (p = env; *p; p += strlen(p) + 1) { 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; @@ -1319,7 +1327,7 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmdline, 0 ); - char **envp = build_envp( env ); + char **envp = build_envp( env, NULL, NULL ); close( fd[0] ); /* Reset signals that we previously set to SIG_IGN */ @@ -1354,14 +1362,8 @@ RTL_USER_PROCESS_PARAMETERS *params; UNICODE_STRING image_str, cmdline_str, curdir_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 (cur_dir) RtlInitUnicodeString( &curdir_str, cur_dir ); if (startup->lpDesktop) RtlInitUnicodeString( &desktop, startup->lpDesktop ); @@ -1415,10 +1417,16 @@ pid_t pid; int err; char dummy = 0; + WCHAR buffer[MAX_PATH]; if (!env) env = GetEnvironmentStringsW(); - if (!(params = create_user_params( filename, cmd_line, cur_dir, 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, cur_dir, startup ))) return FALSE; /* create the synchronization pipes */ @@ -1444,7 +1452,20 @@ if (!(pid = fork())) /* child */ { char **argv = build_argv( cmd_line, 1 ); - char **envp = build_envp( env ); + char **envp = build_envp( env, NULL, NULL ); + char *unixpath; + + unixpath = wine_get_unix_file_name( buffer ); + + /* 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, unixdir, unixpath ); + HeapFree( GetProcessHeap(), 0, unixpath ); close( startfd[1] ); close( execfd[0] ); @@ -1456,8 +1477,6 @@ /* 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 1.4 diff -u -r1.4 config.c --- libs/wine/config.c 11 May 2004 04:29:18 -0000 1.4 +++ libs/wine/config.c 25 May 2004 04:59:54 -0000 @@ -271,6 +271,23 @@ 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 ) @@ -285,7 +302,7 @@ 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 */ @@ -294,7 +311,7 @@ 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] ); } @@ -311,7 +328,7 @@ 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 25 May 2004 04:59:54 -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 25 May 2004 04:59:54 -0000 @@ -29,6 +29,7 @@ #ifdef HAVE_PTHREAD_H # include <pthread.h> #endif +#include <sys/resource.h> #include "wine/library.h" @@ -75,6 +76,23 @@ { const char *loader = getenv( "WINELOADER" ); const char *threads = get_threading(); + char cwd[MAX_PATH], *cwd_var; + + /* hack to work-around broken address space layout in RHEL3 */ + struct rlimit limit; + getrlimit( RLIMIT_STACK, &limit ); + limit.rlim_cur = limit.rlim_max; + setrlimit( RLIMIT_STACK, &limit ); + + 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 +104,9 @@ 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 1.91 diff -u -r1.91 main.c --- loader/main.c 25 May 2004 01:29:24 -0000 1.91 +++ loader/main.c 25 May 2004 04:59:54 -0000 @@ -20,14 +20,52 @@ #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 ) + { +#if 0 + printf("vars(*%p) = %p -> (%p, %08x) (%p, %08x) (%p, %08x)\n", + &wine_preload_vars, wine_preload_vars, + wine_preload_vars->pe_address, + wine_preload_vars->pe_size, + wine_preload_vars->heap_address, + wine_preload_vars->heap_size, + wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); +#endif + + /* let wine know which area was mapped for the PE EXE */ + wine_mmap_add_reserved_area( wine_preload_vars->pe_address, + wine_preload_vars->pe_size); + + /* let wine know which area was mapped for the system heap */ + wine_mmap_add_reserved_area( wine_preload_vars->heap_address, + wine_preload_vars->heap_size); + + /* unmap the preloader itself */ + munmap( wine_preload_vars->preload_base, + wine_preload_vars->preload_size ); + wine_preload_vars = NULL; + } + + 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 2004-04-15 17:29:17.000000000 +0900 @@ -0,0 +1,10 @@ + +struct wine_preload_info +{ + void *pe_address; + size_t pe_size; + void *heap_address; + size_t heap_size; + void *preload_base; + size_t preload_size; +}; --- /dev/null 1994-07-18 08:46:18.000000000 +0900 +++ loader/preload.h 2004-04-15 17:29:17.000000000 +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 2004-05-25 15:04:52.000000000 +0900 @@ -0,0 +1,750 @@ +/* + 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 +#undef DUMP_PERESERVE + +/* older systems may not define these */ +#ifndef PT_TLS +#define PT_TLS 7 +#endif +#ifndef PT_GNU_EH_FRAME +#define PT_GNU_EH_FRAME 0x6474e550 +#endif +#ifndef PT_GNU_STACK +#define PT_GNU_STACK 0x6474e551 +#endif +#ifndef PT_GNU_RELRO +#define PT_GNU_RELRO 0x6474e552 +#endif + +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[1] != 'E' ) || + ( header->e_ident[2] != 'L' ) || + ( header->e_ident[3] != '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 PT_GNU_STACK: + case PT_GNU_RELRO: + 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( "file %s\n", pe_file ); + 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); + + /* reserve the memory that wine needs for the global heap */ + vars.heap_size = 0x400000; + vars.heap_address = (void*)0x65430000; + mmap( vars.heap_address, vars.heap_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" +); +

1 0

Re: Resend - Unitialised EDITSTATE
by Alexandre Julliard May 24, 2004

May 24, 2004

jeffl <jeffl(a)defcen.gov.au> writes: > People, when trying to run DebugApiSpy.exe in encounter a divide by > zero. It seems there is a circumstance where the EDITSTATE block can be > unitialised. I have not found this yet and I am not sure how to track > it down. This patch fixes the instance where the problem occurs for me. That's not a fix, that's just hiding the problem. You need to figure out why the value was not initialized at this point, it's probably a message order thing. -- Alexandre Julliard julliard(a)winehq.org

1 0

Calendars
by William Lahti May 24, 2004

May 24, 2004

I am trying to implement EnumCalendarInfo and would like to know if any locales have multiple calendars, where the settings for the extra calendars would be in the nls file, and if any calendars have their own seperate file.

3 3

additional download for winehq download page
by Ivan May 24, 2004

May 24, 2004

This patch adds lots of download to the winehq download page. Sending only to wine-devel for now, as I'm not sure they are all needed, but they could all be useful in some situation or another. Comment welcome. Ivan. p.s. Maybe we should have a download page just for m$ components?

3 2

Re: additional download for winehq download page
by Ivan May 24, 2004

May 24, 2004

> The link to the images is broken, it is winedows.gif when it should be > windows.gif. Yes, I've just noticed that. Ivan.

1 0

Re: Check that at least one drive is defined on startup
by Alexandre Julliard May 24, 2004

May 24, 2004

Mike Hearn <mh(a)codeweavers.com> writes: > Mike Hearn <mh(a)codeweavers.com> > Add a check to process_init() to ensure at least one drive is defined, > as things break if that isn't the case. There is no need to define the Drives registry key, everything should work fine without it. If something doesn't, that's a bug. -- Alexandre Julliard julliard(a)winehq.org

2 3

Re: Split the dsound test
by Robert Reif May 24, 2004

May 24, 2004

Francois Gouget wrote: > > You will notice the following problems: > * the test takes awfully long (>6 minutes here). It loops through the > known formats and for each of them it plays, in turn, the regular > test, a test through a duplicated buffer, with a listener, etc. > * as a result of the above, a user listening to the test does not > know what to expect, which makes it hard to know whether the sound was > played correctly. The corresponding Microsoft test takes hours to complete. If time is an issue, then the test should be split up into a quick test and an exhaustive test. Also, volume and pan tests on primary and secondary buffers is not the same thing as 3D. They just happen to be similar now because we don't support hardware acceleration yet.

2 2