- wine-devel - winehq.org

RE: Handling dialog messages for non-template based dialogs
by Robert Shearman 13 Apr '04

13 Apr '04

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 13 Apr '04

13 Apr '04

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 13 Apr '04

13 Apr '04

"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 13 Apr '04

13 Apr '04

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 13 Apr '04

13 Apr '04

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 12 Apr '04

12 Apr '04

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 12 Apr '04

12 Apr '04

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 12 Apr '04

12 Apr '04

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 12 Apr '04

12 Apr '04

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 12 Apr '04

12 Apr '04

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 12 Apr '04

12 Apr '04

> 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 12 Apr '04

12 Apr '04

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 12 Apr '04

12 Apr '04

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

1 0

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

12 Apr '04

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

1 0

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

12 Apr '04

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

1 0

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

12 Apr '04

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

3 4

None
by None 12 Apr '04

12 Apr '04

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

1 0

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

11 Apr '04

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

1 0

Wine menu woes
by Geoffrey Hausheer 11 Apr '04

11 Apr '04

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

4 3

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

11 Apr '04

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

1 0

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

11 Apr '04

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

2 1

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

11 Apr '04

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

2 1

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

11 Apr '04

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

1 1

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

11 Apr '04

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

2 2

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

11 Apr '04

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

2 2

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

10 Apr '04

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

1 0

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

10 Apr '04

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

2 1

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

10 Apr '04

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

1 0

winehq & 20040408
by Rein Klazes 10 Apr '04

10 Apr '04

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

1 0

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

10 Apr '04

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

6 9

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

09 Apr '04

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

3 2

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

09 Apr '04

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

1 0

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

09 Apr '04

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

1 0

Serial Port Problem
by TN 09 Apr '04

09 Apr '04

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

2 1

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

09 Apr '04

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

4 8

Re: CryptoAPI
by dataphile 09 Apr '04

09 Apr '04

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

1 0

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

09 Apr '04

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

1 0

Re: ShellExecute IE compatibility problems
by Martin Fuchs 09 Apr '04

09 Apr '04

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 09 Apr '04

09 Apr '04

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 08 Apr '04

08 Apr '04

> 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 08 Apr '04

08 Apr '04

> 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 08 Apr '04

08 Apr '04

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 08 Apr '04

08 Apr '04

> 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 08 Apr '04

08 Apr '04

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 08 Apr '04

08 Apr '04

> 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 08 Apr '04

08 Apr '04

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 08 Apr '04

08 Apr '04

"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 08 Apr '04

08 Apr '04

sorry, wrong reply button again.

1 0

DSD_CreateSecondaryBuffer
by Daniel Skorka 07 Apr '04

07 Apr '04

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 07 Apr '04

07 Apr '04

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

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel