May 2002 - Wine-devel - list.winehq.org

Regression in CreateDIBSection
by Mehmet YASAR 10 May '02

10 May '02

Hi, the following patch incorporated in CVS may 3rd has broken BabyGammon (you can get it at http://www.directfichiers.com/babygammon/BAFR100DOGT90.EXE ) With it I have black background instead of the normal backgammon game. If somebody could look at this ... Mehmet YASAR Extract from logs : .... 08073078:Call user32.LoadImageA(00400000,0000006a,00000000,00000000,00000000,00002000) ret=00407f17 08073078:Call x11drv.CreateDIBSection(403844f4,403bb334,00000000,00000000,00000000,00000000,00000000) ret=407b4ac2 trace:bitmap:X11DRV_DIB_CreateDIBSection format (114,19), planes 1, bpp 8, size 2204, colors 256 (RGB) 08073078:Call x11drv.GetDeviceCaps(403844f4,0000000c) ret=407b173b 08073078:Ret x11drv.GetDeviceCaps() retval=00000010 ret=407b173b 08073078:Call x11drv.GetDeviceCaps(403844f4,0000000e) ret=407b173b 08073078:Ret x11drv.GetDeviceCaps() retval=00000001 ret=407b173b trace:bitmap:CreateBitmap 114x19, 65536 colors returning 00000870 trace:bitmap:X11DRV_DIB_DoProtectDIBSection Changed protection from 4 to 4 08073078:Ret x11drv.CreateDIBSection() retval=00000870 ret=407b4ac2 .... Index: wine/graphics/x11drv/dib.c diff -u wine/graphics/x11drv/dib.c:1.91 wine/graphics/x11drv/dib.c:1.92 --- wine/graphics/x11drv/dib.c:1.91 Wed Apr 24 16:32:11 2002 +++ wine/graphics/x11drv/dib.c Sat May 4 13:32:48 2002 @@ -5793,16 +5793,8 @@ InitializeCriticalSection(&(dib->lock)); if (VIRTUAL_SetFaultHandler(bm.bmBits, X11DRV_DIB_FaultHandler, (LPVOID)res)) { - if (section || offset) - { - X11DRV_DIB_DoProtectDIBSection( bmp, PAGE_READWRITE ); - if (dib) dib->status = DIB_Status_AppMod; - } - else - { - X11DRV_DIB_DoProtectDIBSection( bmp, PAGE_READONLY ); - if (dib) dib->status = DIB_Status_InSync; - } + X11DRV_DIB_DoProtectDIBSection( bmp, PAGE_READWRITE ); + if (dib) dib->status = DIB_Status_AppMod; } }

2 2

Re: wine/ ./configure ./configure.ac include/confi ...
by Joerg Mayer 10 May '02

10 May '02

On Thu, May 09, 2002 at 08:33:41PM -0500, Alexandre Julliard wrote: > Log message: > Steven Edwards <Steven_Ed4153(a)yahoo.com> > Detect snprintf && _snprintf, use _snprintf on stupid platforms > (windows). I'm afraid that this is the wrong solution to this problem: People using sNprintf do this sometimes for security reasons. Just ignoring the length is very likely to cause some problems with applications that would have been otherwise secure (I'm thinking of chatclients etc). snprintf should either be implemented or the application simply should not work, but making a secure app insecure is not the way to go. ciao Jörg -- Joerg Mayer <jmayer(a)loplof.de> I found out that "pro" means "instead of" (as in proconsul). Now I know what proactive means.

2 2

Request For Comments: emulation of Windows file attributes on POSIX
by Jakob Eriksson 10 May '02

10 May '02

I filed a bug that comments Wine can't create a HIDDEN file, so that Wine does not later find it with FindFirstFile(). http://bugs.winehq.com/show_bug.cgi?id=660 there was a little discussion there and I suggested a) a mode where Wine reads the special info from VFAT and NTFS filesystems. (linux kernel enhancement probably necessary) b) for POSIX filesystems, an emulation layer where Wine emulates the attributes by storing them in a special file in each directory. Andriy Palamarchuk commented with this: There are plans on integration with Samba and this functionality can be implemented there for communication with Windows machines. However, I can't decide whether it is worth to implement these features for Posix file system. I suggest you to discuss the matter on wine-devel. Is it worth it at all? Is it worth it if I implement it as a Wine learning exercise and all you Wine demigods have to do is either say "ok" or "bad, to slow, let's ditch it" or whatever. I don't understand that about Samba integration. Can they read the special attributes on VFAT and NTFS? Are they planning to emulate it? -- regards, Jakob Eriksson The wages of sin is debugging. -- Ron Jeffries

2 1

A little configure change help needed for the mingw/msys port.
by Steven Edwards 09 May '02

09 May '02

Preface: If you have the time to help thanks, if not I will keep fighting with it and should be able To get it sooner or latter and thanks anyway. After the recent patch for cygwin I found the 2 changes I need to make to configure/configure.ac for a mingw host/target but I'm having trouble with the syntax of configure to get it to do what I need it to do. My current porting work has been based on some dirty hacks to configure that didn't Really fix the problem without breaking the checks for other platforms. Now I just need some help making These 2 changes. Change #1. at line 5552 its checking for dlopen which I think is in dlfcn.h. Mingw doesn't have this and I don't think the mingw port needs this but configure returns. else LIBEXT="a" fi This of course causes configure to bomb because we need to build a dll/so. I cant get my syntax right To let it know not to set LIBEXT=a here if it's a mingw host. Change #2 On line 9465 there is the $LIBEXT = so/$target_os *cygwin* check. If I make Change #1 via my dirty breaking of configure and I change the case $target_os to *mingw* then it generates My makefiles right. My problem is really stupid because if I copy the cygwin check and make a mingw check It bitches about unexepected EOF in the configure script where I'm just missing a if/fi or something. Like I said, its really minor. Once this is done then you should at least be able to ./configure on msys just like cygwin if they fix the dlltool/wrap on msys. Thanks Steven "Every revolution was once a thought in one man's mind" - Ralph Waldo Emerson

2 2

Help debugging wine with CAM app
by Kyle 09 May '02

09 May '02

I'd like to get involved in wine development. I'm starting by trying to get wine to run my favorite CAM application. Anyone who could give me a little coaching on debugging would be much appreciated. I'm using a recent CVS download of wine. To replicate the problem, do: Download the zip file at ftp://ftp.actarg.com/pub/misc/mtc.tgz. Unzip it under /c/Program Files. Then CD into MTC/TurboNest/Program and run: winedbg TurboNst.exe --debugmsg trace+string,+relay 2>/tmp/debug.out Search in the debug file for the string: TURBONST.ent Right around there is about where it starts to go into an endless loop. I have been single stepping all around in there and I can't figure out what is going on. One wierd thing: In the debugger, do break INT_Int31Handler cont then single step into CallRMInt on line: if (!Dosvm.CallRMInt && !DPMI_LoadDosSystem()), it doesn't seem to call DPMI_LoadDosSystem() at the correct address it calls some other address which looks like a jump table of some kind I'm not sure if that's the problem, but in any case, there's a point soon thereafter where it gets stuck doing: trace:string:lstrcmpiA strcmpi "" and "TURBONST.ent" over and over again. One other interesting thing: I found out by accident that if I patch into wine an illegal operation as follows: Put into the last line of procedure: BOOL WINAPI SetCurrentDirectoryA( LPCSTR dir ) a command that access memory illegally such as: FIXME("TRUE:%s\n",drive); //drive is an int--not a string Then run the program. Wine reports the illegal memory access, but then goes on to run the program successfully (other than that the fonts look funny and that function doesn't work right). Anyone that has experience debugging wine, I'm interested in learning more about how to do it. If I can get some experience under my belt, I'd love to help developing wine and finding more bugs. Thanks in advance! Kyle Bateman

2 1

Possible fix for DLLSeparation, Bug #543
by Uwe Bonnes 09 May '02

09 May '02

Hallo, appended patch exports FILE_DupUnixHandle() as wine_dup_unix_handle() from kernel32. As FILE_DupUnixHandle uses it's own server-call, I thinks we need it as wine-exported function. Please comment if it is the right approach and apply if appropriate Thanks -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Free Software: If you contribute nothing, expect nothing -- Index: wine/include/winbase.h =================================================================== RCS file: /home/wine/wine/include/winbase.h,v retrieving revision 1.142 diff -u -r1.142 winbase.h --- wine/include/winbase.h 25 Apr 2002 21:34:24 -0000 1.142 +++ wine/include/winbase.h 9 May 2002 21:24:45 -0000 @@ -1747,7 +1747,7 @@ /* Wine internal functions */ BOOL WINAPI wine_get_unix_file_name( LPCSTR dos, LPSTR buffer, DWORD len ); - +HANDLE WINAPI wine_dup_unix_handle( int fd, DWORD access, BOOL inherit ); /* a few optimizations for i386/gcc */ Index: wine/include/file.h =================================================================== RCS file: /home/wine/wine/include/file.h,v retrieving revision 1.41 diff -u -r1.41 file.h --- wine/include/file.h 5 Apr 2002 22:53:57 -0000 1.41 +++ wine/include/file.h 9 May 2002 21:24:56 -0000 @@ -69,7 +69,6 @@ extern int FILE_strcasecmp( const char *str1, const char *str2 ); extern int FILE_strncasecmp( const char *str1, const char *str2, int len ); extern void FILE_SetDosError(void); -extern HANDLE FILE_DupUnixHandle( int fd, DWORD access, BOOL inherit ); extern int FILE_GetUnixHandle( HANDLE handle, DWORD access ); extern BOOL FILE_Stat( LPCSTR unixName, BY_HANDLE_FILE_INFORMATION *info ); extern HFILE16 FILE_Dup2( HFILE16 hFile1, HFILE16 hFile2 ); Index: wine/dlls/kernel/kernel32.spec =================================================================== RCS file: /home/wine/wine/dlls/kernel/kernel32.spec,v retrieving revision 1.50 diff -u -r1.50 kernel32.spec --- wine/dlls/kernel/kernel32.spec 9 May 2002 00:05:50 -0000 1.50 +++ wine/dlls/kernel/kernel32.spec 9 May 2002 21:25:50 -0000 @@ -1011,3 +1011,6 @@ # Unix files @ stdcall wine_get_unix_file_name(str ptr long) wine_get_unix_file_name + +#Unix handles +@ stdcall wine_dup_unix_handle(long long long) wine_dup_unix_handle Index: wine/files/file.c =================================================================== RCS file: /home/wine/wine/files/file.c,v retrieving revision 1.145 diff -u -r1.145 file.c --- wine/files/file.c 26 Apr 2002 18:31:19 -0000 1.145 +++ wine/files/file.c 9 May 2002 21:27:29 -0000 @@ -307,12 +307,12 @@ /*********************************************************************** - * FILE_DupUnixHandle + * wine_dup_unix_handle * * Duplicate a Unix handle into a task handle. * Returns 0 on failure. */ -HANDLE FILE_DupUnixHandle( int fd, DWORD access, BOOL inherit ) +HANDLE WINAPI wine_dup_unix_handle( int fd, DWORD access, BOOL inherit ) { HANDLE ret; Index: wine/dlls/x11drv/x11drv_main.c =================================================================== RCS file: /home/wine/wine/dlls/x11drv/x11drv_main.c,v retrieving revision 1.49 diff -u -r1.49 x11drv_main.c --- wine/dlls/x11drv/x11drv_main.c 24 Apr 2002 21:32:11 -0000 1.49 +++ wine/dlls/x11drv/x11drv_main.c 9 May 2002 21:29:06 -0000 @@ -457,7 +457,7 @@ fcntl( ConnectionNumber(data->display), F_SETFD, 1 ); /* set close on exec flag */ if (synchronous) XSynchronize( data->display, True ); wine_tsx11_unlock(); - data->display_fd = FILE_DupUnixHandle( ConnectionNumber(data->display), + data->display_fd = wine_dup_unix_handle( ConnectionNumber(data->display), GENERIC_READ | SYNCHRONIZE, FALSE ); data->process_event_count = 0; NtCurrentTeb()->driver_data = data; Index: wine/scheduler/process.c =================================================================== RCS file: /home/wine/wine/scheduler/process.c,v retrieving revision 1.179 diff -u -r1.179 process.c --- wine/scheduler/process.c 29 Apr 2002 19:34:06 -0000 1.179 +++ wine/scheduler/process.c 9 May 2002 21:29:30 -0000 @@ -299,9 +299,9 @@ /* no parent, and no new console requested, create a simple console with bare handles to * unix stdio input & output streams (aka simple console) */ - SetStdHandle( STD_INPUT_HANDLE, FILE_DupUnixHandle( 0, GENERIC_READ|SYNCHRONIZE, TRUE )); - SetStdHandle( STD_OUTPUT_HANDLE, FILE_DupUnixHandle( 1, GENERIC_WRITE|SYNCHRONIZE, TRUE )); - SetStdHandle( STD_ERROR_HANDLE, FILE_DupUnixHandle( 1, GENERIC_WRITE|SYNCHRONIZE, TRUE )); + SetStdHandle( STD_INPUT_HANDLE, wine_dup_unix_handle ( 0, GENERIC_READ|SYNCHRONIZE, TRUE )); + SetStdHandle( STD_OUTPUT_HANDLE, wine_dup_unix_handle ( 1, GENERIC_WRITE|SYNCHRONIZE, TRUE )); + SetStdHandle( STD_ERROR_HANDLE, wine_dup_unix_handle ( 1, GENERIC_WRITE|SYNCHRONIZE, TRUE )); } else if (!(main_create_flags & (DETACHED_PROCESS|CREATE_NEW_CONSOLE))) {

1 0

Re: FindFirstFile
by Kari Hurtta 09 May '02

09 May '02

http://www.winehq.com/hypermail/wine-devel/2002/05/0253.html | Vincent | | --- files/file.c.orig Thu May 9 10:55:51 2002 | +++ files/file.c Sun Apr 28 20:01:55 2002 | @@ -659,7 +659,7 @@ | if (!(st->st_mode & S_IWUSR)) | info->dwFileAttributes |= FILE_ATTRIBUTE_READONLY; | | - RtlSecondsSince1970ToTime( st->st_mtime, &info->ftCreationTime ); | + RtlSecondsSince1970ToTime( st->st_ctime, &info->ftCreationTime ); | RtlSecondsSince1970ToTime( st->st_mtime, &info->ftLastWriteTime ); | RtlSecondsSince1970ToTime( st->st_atime, &info->ftLastAccessTime ); Isn't creationtime usually set to MIN(st->st_mtime,st->st_ctime)

1 0

Testframework primer needed
by Uwe Bonnes 09 May '02

09 May '02

Hallo, I can't find a primer in the wine distribution. I think it should be included. Possible subjects: Writing good test, running test on wine and winxx, ... Also is there a directive for a failing test to immediatly abort or skip some part of the program flow? The reason I ask: Somwhow ../dlls/kernel/testfile.xxx was craeted readonly before and some test didn't erase it so it was left behind. On the next run dlls/kernel/tests/file.c test__hread went into an infinite loop about that, cluttering up the screen, making it hard to find the reason. Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

3 3

Account for possible missing backslah in GetTempPath
by Uwe Bonnes 09 May '02

09 May '02

Changelog: files/directory.c: GetTempPath[A|W] Acount for a possible missing backslash for size calculation -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Free Software: If you contribute nothing, expect nothing -- Index: wine/files/directory.c =================================================================== RCS file: /home/wine/wine/files/directory.c,v retrieving revision 1.45 diff -u -r1.45 directory.c --- wine/files/directory.c 9 Mar 2002 23:50:37 -0000 1.45 +++ wine/files/directory.c 9 May 2002 16:31:31 -0000 @@ -186,6 +186,21 @@ if (!(ret = GetEnvironmentVariableA( "TEMP", path, count ))) if (!(ret = GetCurrentDirectoryA( count, path ))) return 0; + if ((ret) && (ret +1 > count)) + /* We need to return the required length including the + trailing backslash. So we have to acually get the string + to test if the backslash is there */ + { + char *buffer; + UINT ret1; + if ((buffer = HeapAlloc( GetProcessHeap(), 0, ret ))) + { + ret1 = GetTempPathA(ret,buffer); + if (buffer[ret1-1] != '\\') + ret++; + HeapFree( GetProcessHeap(), 0, buffer ); + } + } if (count && (ret < count - 1) && (path[ret-1] != '\\')) { path[ret++] = '\\'; @@ -207,6 +222,21 @@ if (!(ret = GetEnvironmentVariableW( temp, path, count ))) if (!(ret = GetCurrentDirectoryW( count, path ))) return 0; + if ((ret) && (ret +1 > count)) + /* We need to return the required length including the + trailing backslash. So we have to acually get the string + to test if the backslash is there */ + { + WCHAR *buffer; + UINT ret1; + if ((buffer = HeapAlloc( GetProcessHeap(), 0, ret ))) + { + ret1 = GetTempPathW(ret,buffer); + if (buffer[ret1-1] != '\\') + ret++; + HeapFree( GetProcessHeap(), 0, buffer ); + } + } if (count && (ret < count - 1) && (path[ret-1] != '\\')) { path[ret++] = '\\';

1 0

test handling of trailing backslash in dlls/kernel/tests/path.c
by Uwe Bonnes 09 May '02

09 May '02

Changelog: dlls/kernel/tests/path.c: test_InitPathA Test what happens when TMP variable has no trailing backslash A patch to correct that hbehaviour follows. -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Free Software: If you contribute nothing, expect nothing -- Index: wine/dlls/kernel/tests/path.c =================================================================== RCS file: /home/wine/wine/dlls/kernel/tests/path.c,v retrieving revision 1.3 diff -u -r1.3 path.c --- wine/dlls/kernel/tests/path.c 6 May 2002 20:13:16 -0000 1.3 +++ wine/dlls/kernel/tests/path.c 9 May 2002 16:23:44 -0000 @@ -19,6 +19,7 @@ */ #include <stdio.h> +#include <assert.h> #include "wine/test.h" #include "winbase.h" #include "winuser.h" @@ -330,20 +331,22 @@ { CHAR tmppath[MAX_PATH], /*path to TEMP */ tmpstr[MAX_PATH], - tmpstr1[MAX_PATH]; + tmpstr1[MAX_PATH], + envstr[MAX_PATH]; DWORD len,len1; INT id; HANDLE hndl; /* Test GetTempPathA */ - len=GetTempPathA(MAX_PATH,tmppath); + len=GetTempPathA(0,tmppath); ok(len!=0 && len < MAX_PATH,"GetTempPathA failed"); + len1=GetTempPathA(len,tmppath); ok(HAS_TRAIL_SLASH_A(tmppath), "GetTempPathA returned a path that did not end in '\\'"); lstrcpyA(tmpstr,"aaaaaaaa"); len1=GetTempPathA(len,tmpstr); - ok(len1==len+1, - "GetTempPathA should return string length %d instead of %d",len+1,len1); + ok(len1+1==len, + "GetTempPathA should return string %s length %d instead of %d",tmpstr, len,len1+1); if(WIN2K_PLUS(version)) { /* in Win2k, the path won't be modified, but in win98, wine it is */ todo_wine { @@ -351,6 +354,20 @@ "GetTempPathA should not have modified the buffer"); } } + + /* Now set the TMP variable to GetTempPath without the trailing backslash + to test for completion of backslash */ + len1 = GetEnvironmentVariableA("TMP",envstr,MAX_PATH); + ok(len1< MAX_PATH, "TMP Envirinment Variable too long"); + tmppath[len-2] = '\0'; + ok(SetEnvironmentVariableA("TMP",tmppath),"SetEnvironmentVariable(\"TMP\") failed"); + len1=GetTempPathA(0,tmppath); + len1=GetTempPathA(len1,tmppath); + ok(HAS_TRAIL_SLASH_A(tmppath), + "GetTempPathA didn't correct a TMP setting without trailing '\\'"); + ok(SetEnvironmentVariableA("TMP",envstr),"Can't reset TMP environment variable"); + len1=GetTempPathA(len,tmppath); + /* Test GetTmpFileNameA The only test we do here is whether GetTempFileNameA passes or not. We do not thoroughly test this function yet (specifically, whether

1 0