- Wine-devel - list.winehq.org

Re: SafeDisc-1 semi-cleaned note...
by Laurent Pinchart 29 May '02

29 May '02

> i hope i am not asking to much by this, but is there > any chance you could send me a copy of the diff's with > some fixme comments so I know exactly which things > need to be cleaned up? I know that you said the > server suspension code needs to be checked, and the > 0x7ffe0000 page needs to be properly implemented, but > im not sure what all would need to be changed to get > this done. Ok, here's a first version of safedisc.safe.diff with comments. I don't think it's usefull to comment safedisc.dmca.diff for now, but if you need some information about it let me know. --- memory/emulate.c Thu Jan 1 01:00:00 1970 +++ memory/emulate.c Thu May 9 17:35:27 2002 @@ -0,0 +1,53 @@ +/* + * Emulation of privileged memory + * + * Copyright 2002 Laurent Pinchart + * + * 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 "windef.h" +#include "wingdi.h" +#include "wine/winuser16.h" +#include "module.h" +#include "wine/debug.h" + +WINE_DEFAULT_DEBUG_CHANNEL(virtual); + +#ifdef __i386__ + +/*********************************************************************** + * MEMORY_SharedUserData + * + * Handles exceptions for the SharedUserData access. + */ +BOOL MEMORY_SharedUserData( LPVOID arg, LPCVOID addr ) +{ + DWORD dwProtection; + BOOL ret; + + TRACE( "MEMORY_SharedUserData\n" ); + + ret = VirtualProtect( (LPVOID)0x7ffe0000, 0x10000, PAGE_READWRITE, &dwProtection ); + if ( ret ) + { + *(LPDWORD)0x7ffe0000 = 0x12345678; + ret = VirtualProtect( (LPVOID)0x7ffe0000, 0x10000, PAGE_READONLY, &dwProtection ); + } + + return ret; +} + +#endif /* __i386__ */ This is the emulation of "SharedUserData". SharedUserData is a page of memory located right before the kernel space (0x80000000), so at address 0x7ffe0000. It contains miscelaneous information, and is read-only in user space and read-write in kernel space. You can find more information about it in ntddk.h (not included in wine). We need 3 things here: 1) Write support for the other fields, not only TickCount (kernel timer). This is optional but quite easy for some fields. 2) Find a way to either remove the read rights after the read access, so that the next read access would trigger an exception again, and the information could be updated (this is IMHO *very* difficult), or periodically update the information contained in that page (mostly TickCount). The later should be easier, but implies a slow down. If would be nice to trigger that periodical update only after the first read of TickCount, so that programs which don't read it wouldn't suffer from that slow down. We could also remove the read rights after some time (instead of right after the read operation). That might be the best way, as it wouldn't imply a slow down (not periodical update). 3) Replace 0x12345678 with a correct value. Don't forget that secdrv.c needs to know that value (or at least a close enough value), so we need a portable and accurate way to set and retrieve it from inside wine. Index: dlls/ntdll/Makefile.in =================================================================== RCS file: /home/wine/wine/dlls/ntdll/Makefile.in,v retrieving revision 1.27 diff -u -r1.27 Makefile.in --- dlls/ntdll/Makefile.in 14 May 2002 20:55:01 -0000 1.27 +++ dlls/ntdll/Makefile.in 23 May 2002 18:54:33 -0000 @@ -32,6 +32,7 @@ $(TOPOBJDIR)/loader/ne/segment.c \ $(TOPOBJDIR)/memory/atom.c \ $(TOPOBJDIR)/memory/codepage.c \ + $(TOPOBJDIR)/memory/emulate.c \ $(TOPOBJDIR)/memory/environ.c \ $(TOPOBJDIR)/memory/global.c \ $(TOPOBJDIR)/memory/heap.c \ When emulate.c will be applied this will be too :-) Index: dlls/ntdll/signal_i386.c =================================================================== RCS file: /home/wine/wine/dlls/ntdll/signal_i386.c,v retrieving revision 1.35 diff -u -r1.35 signal_i386.c --- dlls/ntdll/signal_i386.c 16 May 2002 20:32:16 -0000 1.35 +++ dlls/ntdll/signal_i386.c 23 May 2002 18:54:40 -0000 @@ -679,6 +679,7 @@ { EXCEPTION_RECORD rec; DWORD page_fault_code = EXCEPTION_ACCESS_VIOLATION; + DWORD priv_instr_code = EXCEPTION_PRIV_INSTRUCTION; #ifdef CR2_sig /* we want the page-fault case to be fast */ @@ -708,8 +709,8 @@ case T_SEGNPFLT: /* Segment not present exception */ case T_PROTFLT: /* General protection fault */ case T_UNKNOWN: /* Unknown fault code */ - if (INSTR_EmulateInstruction( context )) return; - rec.ExceptionCode = EXCEPTION_PRIV_INSTRUCTION; + if (INSTR_EmulateInstruction( context, &priv_instr_code )) return; + rec.ExceptionCode = priv_instr_code; break; case T_PAGEFLT: /* Page fault */ #ifdef CR2_sig Index: include/miscemu.h =================================================================== RCS file: /home/wine/wine/include/miscemu.h,v retrieving revision 1.46 diff -u -r1.46 miscemu.h --- include/miscemu.h 10 Mar 2002 00:02:34 -0000 1.46 +++ include/miscemu.h 23 May 2002 18:54:49 -0000 @@ -167,7 +167,7 @@ extern UINT DOSMEM_MapLinearToDos(LPVOID); /* linear Wine to DOS */ /* memory/instr.c */ -extern BOOL INSTR_EmulateInstruction( CONTEXT86 *context ); +extern BOOL INSTR_EmulateInstruction( CONTEXT86 *context, DWORD *exception_code ); /* msdos/interrupts.c */ typedef void (WINAPI *INTPROC)(CONTEXT86*); Index: memory/instr.c =================================================================== RCS file: /home/wine/wine/memory/instr.c,v retrieving revision 1.15 diff -u -r1.15 instr.c --- memory/instr.c 11 May 2002 23:01:12 -0000 1.15 +++ memory/instr.c 23 May 2002 18:54:58 -0000 @@ -396,7 +396,7 @@ * * Emulate a privileged instruction. Returns TRUE if emulation successful. */ -BOOL INSTR_EmulateInstruction( CONTEXT86 *context ) +BOOL INSTR_EmulateInstruction( CONTEXT86 *context, DWORD *exception_code ) { int prefix, segprefix, prefixlen, len, repX, long_op, long_addr; SEGPTR gpHandler; @@ -686,7 +686,9 @@ case 0xcd: /* int <XX> */ if (long_op) { - ERR("int xx from 32-bit code is not supported.\n"); + ERR("int 0x%02x from 32-bit code is not supported.\n", instr[1]); + if (instr[1] == 0x01) + *exception_code = EXCEPTION_ACCESS_VIOLATION; break; /* Unable to emulate it */ } else This probably hasn't been applied because it lacks testing. It seems that some instructions which are emulated by wine (int 0x01 is one of them) should generate an EXCEPTION_PRIV_INSTRUCTION instead of EXCEPTION_ACCESS_VIOLATION. Or maybe my code is not clean enough to commit. I'd like to hear comments from Alexandre about this specific part of the patch. Index: include/wine/server_protocol.h =================================================================== RCS file: /home/wine/wine/include/wine/server_protocol.h,v retrieving revision 1.35 diff -u -r1.35 server_protocol.h --- include/wine/server_protocol.h 25 Apr 2002 22:58:59 -0000 1.35 +++ include/wine/server_protocol.h 23 May 2002 18:54:56 -0000 @@ -296,6 +296,7 @@ struct init_process_done_reply { struct reply_header __header; + int suspended; int debugged; }; @@ -3198,6 +3199,6 @@ struct get_window_properties_reply get_window_properties_reply; }; -#define SERVER_PROTOCOL_VERSION 79 +#define SERVER_PROTOCOL_VERSION 80 #endif /* __WINE_WINE_SERVER_PROTOCOL_H */ Index: scheduler/process.c =================================================================== RCS file: /home/wine/wine/scheduler/process.c,v retrieving revision 1.182 diff -u -r1.182 process.c --- scheduler/process.c 22 May 2002 21:32:50 -0000 1.182 +++ scheduler/process.c 23 May 2002 18:55:04 -0000 @@ -25,6 +25,7 @@ #include <ctype.h> #include <errno.h> #include <fcntl.h> +#include <signal.h> #include <stdlib.h> #include <stdio.h> #include <string.h> @@ -43,6 +44,7 @@ #include "wine/server.h" #include "options.h" #include "wine/debug.h" +#include "global.h" WINE_DEFAULT_DEBUG_CHANNEL(process); WINE_DECLARE_DEBUG_CHANNEL(relay); @@ -126,6 +128,9 @@ /* scheduler/pthread.c */ extern void PTHREAD_init_done(void); +/* memory/emulate.c */ +extern BOOL MEMORY_SharedUserData(LPVOID, LPCVOID); + extern BOOL MAIN_MainInit(void); typedef WORD (WINAPI *pUserSignalProc)( UINT, DWORD, DWORD, HMODULE16 ); @@ -416,6 +421,20 @@ SetStdHandle( STD_ERROR_HANDLE, current_startupinfo.hStdError ); } + /* Reserves a PAGE_NOACCESS at 0x7ffe0000. That page is used as shared + * memory between kernel space and user space (see SharedUserData in + * ntddk.h). It looks like it is at 0xffdf0000 on NT5. That is still to + * be checked. + */ + if ( VirtualAlloc( (LPVOID)0x7ffe0000, 0x10000, MEM_RESERVE|MEM_COMMIT, PAGE_NOACCESS ) == (LPCVOID)0x07ffe0000 ) + { + VIRTUAL_SetFaultHandler( (LPCVOID)0x7ffe0000, MEMORY_SharedUserData, 0 ); + } + else + { + WARN( "Unable to map SharedUserData, SafeDisc protection won't be supported\n" ); + } + /* Now we can use the pthreads routines */ PTHREAD_init_done(); @@ -440,7 +459,7 @@ */ static void start_process(void) { - int debugged, console_app; + int suspended, debugged, console_app; LPTHREAD_START_ROUTINE entry; WINE_MODREF *wm; HFILE main_file = main_exe_file; @@ -481,7 +500,8 @@ req->gui = !console_app; wine_server_add_data( req, main_exe_name, strlen(main_exe_name) ); wine_server_call( req ); - debugged = reply->debugged; + suspended = reply->suspended; + debugged = reply->debugged; } SERVER_END_REQ; @@ -513,6 +533,8 @@ PROCESS_CallUserSignalProc( USIG_PROCESS_LOADED, 0 ); /* Call UserSignalProc ( USIG_PROCESS_RUNNING ... ) only for non-GUI win32 apps */ if (console_app) PROCESS_CallUserSignalProc( USIG_PROCESS_RUNNING, 0 ); + + if (suspended) raise( SIGSTOP ); if (TRACE_ON(relay)) DPRINTF( "%08lx:Starting process %s (entryproc=%p)\n", Index: server/process.c =================================================================== RCS file: /home/wine/wine/server/process.c,v retrieving revision 1.83 diff -u -r1.83 process.c --- server/process.c 26 Apr 2002 19:05:18 -0000 1.83 +++ server/process.c 23 May 2002 18:55:07 -0000 @@ -382,6 +382,14 @@ } +/* signal the parent that a process init is finished */ +void process_init_done( struct process *process ) +{ + set_event( process->init_event ); + release_object( process->init_event ); + process->init_event = NULL; +} + /* get a process from an id (and increment the refcount) */ struct process *get_process_from_id( void *id ) { @@ -896,13 +904,27 @@ if ((process->exe.namelen = get_req_data_size())) process->exe.filename = memdup( get_req_data(), process->exe.namelen ); - generate_startup_debug_events( current->process, req->entry ); - set_event( process->init_event ); - release_object( process->init_event ); - process->init_event = NULL; + generate_startup_debug_events( process, req->entry ); if (req->gui) process->idle_event = create_event( NULL, 0, 1, 0 ); - if (current->suspend + current->process->suspend > 0) stop_thread( current ); - reply->debugged = (current->process->debugger != 0); + + reply->suspended = 0; + reply->debugged = (current->process->debugger != 0); + if (current->suspend + current->process->suspend > 0) + { + /* temporarily clear the init event so that ptrace functions work */ + struct event *init_event = process->init_event; + process->init_event = NULL; + stop_thread( current ); + process->init_event = init_event; + + if (current->attached) + { + reply->suspended = 1; + continue_thread( current ); /* it will suspend itself */ + return; + } + } + process_init_done( process ); } /* open a handle to a process */ Index: server/process.h =================================================================== RCS file: /home/wine/wine/server/process.h,v retrieving revision 1.27 diff -u -r1.27 process.h --- server/process.h 22 Mar 2002 00:21:23 -0000 1.27 +++ server/process.h 23 May 2002 18:55:07 -0000 @@ -90,6 +90,7 @@ extern struct thread *create_process( int fd ); extern struct process *get_process_from_id( void *id ); extern struct process *get_process_from_handle( handle_t handle, unsigned int access ); +extern void process_init_done( struct process *process ); extern int process_set_debugger( struct process *process, struct thread *thread ); extern int debugger_detach( struct process* process, struct thread* debugger ); Index: server/protocol.def =================================================================== RCS file: /home/wine/wine/server/protocol.def,v retrieving revision 1.36 diff -u -r1.36 protocol.def --- server/protocol.def 25 Apr 2002 22:58:59 -0000 1.36 +++ server/protocol.def 23 May 2002 18:55:12 -0000 @@ -276,6 +276,7 @@ int gui; /* is it a GUI process? */ VARARG(filename,string); /* file name of main exe */ @REPLY + int suspended; /* created suspended? */ int debugged; /* being debugged? */ @END Index: server/ptrace.c =================================================================== RCS file: /home/wine/wine/server/ptrace.c,v retrieving revision 1.14 diff -u -r1.14 ptrace.c --- server/ptrace.c 10 Mar 2002 00:18:36 -0000 1.14 +++ server/ptrace.c 23 May 2002 18:55:13 -0000 @@ -79,7 +79,12 @@ switch(sig) { case SIGSTOP: /* continue at once if not suspended */ - if (thread && (thread->process->suspend + thread->suspend)) break; + if (thread && (thread->process->suspend + thread->suspend)) + { + /* check if this was the self-suspend upon process init */ + if (thread->process->init_event) process_init_done( thread->process ); + break; + } /* fall through */ default: /* ignore other signals for now */ if (thread && get_thread_single_step( thread )) Index: server/trace.c =================================================================== RCS file: /home/wine/wine/server/trace.c,v retrieving revision 1.132 diff -u -r1.132 trace.c --- server/trace.c 26 Apr 2002 19:05:18 -0000 1.132 +++ server/trace.c 23 May 2002 18:55:20 -0000 @@ -441,6 +441,7 @@ static void dump_init_process_done_reply( const struct init_process_done_reply *req ) { + fprintf( stderr, " suspended=%d,", req->suspended ); fprintf( stderr, " debugged=%d", req->debugged ); } There are two patches in there. The first one sets a fault handler for the SharedUserData. This looks fine to me, and will probably be applied when memory/emulate.c will be applied. The second one has been written by Alexandre to fix the process suspension at creation time. Once again I'd like to hear from him to know what he doesn't like about the code, as he's the one who wrote it :-) Hope this helps. I'm available if you need more information Laurent Pinchart

3 9

Problem in winpos.c which ends in a infinite loop
by Christian Costa 29 May '02

29 May '02

Hi everyone, I tried to install the game Zork : Grand Inquisitor ( probably a win16 app) and the installer crashed. With the debugger, I Obtained the attached log. As you can see 1) CreateWindowExA is called (9170) 2) X11DRV_CreateWindow send the WM_CREATE message to the window (9168) 3) the window procedure of the app is executed (9162) and call SetForegroundWindow (9161) 4) WINPOS_SetActiveWindows is then called (9160) and send the WM_QUERYNEWPALETTE (9159) 5) When the windows procedure receive the message (9153), it call SetForegroundWindow (9152) 6) return to 4 Commenting the lines (see attached diff) in winpos.c wich send the WM_QUERYNEWPALETTE avoids the infinite loop and makes the installer work. The installer runs perfectly under Windows 98. It seems that the app's windows procedure does not expect a WM_QUERYNEWPALETTE message. If someone that knows this code well have an idea... else I will report this bug to bugzilla. Ciao, Christian 9084 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d428c) 9085 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d42e8) 9086 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d436c) [message.c:1738] in user32.dll.so) (ebp=405d4348) 9087 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d437c) 9088 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d43d4) 9089 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d43f0) 9090 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d4644) 9091 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d4674) 9092 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d469c) 9093 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d46d0) 9094 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d472c) 9095 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d47b0) [message.c:1738] in user32.dll.so) (ebp=405d478c) 9096 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d47c0) 9097 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d4818) 9098 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d4834) 9099 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d4a88) 9100 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d4ab8) 9101 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d4ae0) 9102 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d4b14) 9103 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d4b70) 9104 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d4bf4) [message.c:1738] in user32.dll.so) (ebp=405d4bd0) 9105 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d4c04) 9106 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d4c5c) 9107 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d4c78) 9108 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d4ecc) 9109 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d4efc) 9110 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d4f24) 9111 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d4f58) 9112 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d4fb4) 9113 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d5038) [message.c:1738] in user32.dll.so) (ebp=405d5014) 9114 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d5048) 9115 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d50a0) 9116 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d50bc) 9117 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d5310) 9118 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d5340) 9119 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d5368) 9120 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d539c) 9121 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d53f8) 9122 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d547c) [message.c:1738] in user32.dll.so) (ebp=405d5458) 9123 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d548c) 9124 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d54e4) 9125 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d5500) 9126 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d5754) 9127 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d5784) 9128 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d57ac) 9129 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d57e0) 9130 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d583c) 9131 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d58c0) [message.c:1738] in user32.dll.so) (ebp=405d589c) 9132 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d58d0) 9133 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d5928) 9134 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d5944) 9135 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d5b98) 9136 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d5bc8) 9137 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d5bf0) 9138 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d5c24) 9139 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d5c80) 9140 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d5d04) [message.c:1738] in user32.dll.so) (ebp=405d5ce0) 9141 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d5d14) 9142 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d5d6c) 9143 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d5d88) 9144 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d5fdc) 9145 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d600c) 9146 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d6034) 9147 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d6068) 9148 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d60c4) 9149 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d6148) [message.c:1738] in user32.dll.so) (ebp=405d6124) 9150 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d6158) 9151 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d61b0) 9152 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d61cc) 9153 0x0040535d (splash.exe..text+0x435d in H:\splash\splash.exe) (ebp=405d6420) 9154 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:183] in user32.dll.so) (ebp=405d6450) 9155 0x40698989 (WINPROC_CallProc32WTo32A+0xc9(func=0x404cc0, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2616] in user32.dll.so) (ebp=405d6478) 9156 0x40698fe2 (CallWindowProcW+0xb2(func=0x40ec0412, hwnd=0x10021, msg=0x30f, wParam=0x0, lParam=0x0) [winproc.c:2831] in user32.dll.so) (ebp=405d64ac) 9157 0x406a8e77 (call_window_proc+0xd7(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, unicode=0x1) [message.c:1368] in user32.dll.so) (ebp=405d6508) 9158 0x406a9ab1 (SendMessageTimeoutW+0xe1(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0, flags=0x0, timeout=0xffffffff, res_ptr=0x405d658c) [message.c:1738] in user32.dll.so) (ebp=405d6568) 9159 0x406a9d37 (SendMessageW+0x37(hwnd=0x10021, msg=0x30f, wparam=0x0, lparam=0x0) [message.c:1818] in user32.dll.so) (ebp=405d659c) 9160 0x40691780 (WINPOS_SetActiveWindow+0x230(hWnd=0x10021, fMouse=0x0, fChangeFocus=0x1) [winpos.c:1282] in user32.dll.so) (ebp=405d65f4) 9161 0x406909fa (SetForegroundWindow+0x7a(hwnd=0x10021) [winpos.c:762] in user32.dll.so) (ebp=405d6610) 9162 0x00404b62 (splash.exe..text+0x3b62 in H:\splash\splash.exe) (ebp=405d697c) 9163 0x406924dd (WINPROC_CallWndProc+0x8d(proc=0x404cc0, hwnd=0x10021, msg=0x1, wParam=0x0, lParam=0x405d6d24) [winproc.c:183] in user32.dll.so) (ebp=405d69ac) 9164 0x40698ed2 (CallWindowProcA+0xb2(func=0x404cc0, hwnd=0x10021, msg=0x1, wParam=0x0, lParam=0x405d6d24) [winproc.c:2795] in user32.dll.so) (ebp=405d69e0) 9165 0x406a8ea3 (call_window_proc+0x103(hwnd=0x10021, msg=0x1, wparam=0x0, lparam=0x405d6d24, unicode=0x0) [message.c:1372] in user32.dll.so) (ebp=405d6a3c) 9166 0x406a9c0d (SendMessageTimeoutA+0xed(hwnd=0x10021, msg=0x1, wparam=0x0, lparam=0x405d6d24, flags=0x0, timeout=0xffffffff, res_ptr=0x405d6ac0) [message.c:1786] in user32.dll.so) (ebp=405d6a9c) 9167 0x406a9d87 (SendMessageA+0x37(hwnd=0x10021, msg=0x1, wparam=0x0, lparam=0x405d6d24) [message.c:1829] in user32.dll.so) (ebp=405d6ad0) 9168 0x409b7702 (X11DRV_CreateWindow+0x572(hwnd=0x10021, cs=0x405d6d24, unicode=0x0) [window.c:1005] in x11drv.dll.so) (ebp=405d6b44) 9169 0x4068bd20 (WIN_CreateWindowEx+0x590(cs=0x405d6d24, classAtom=0xc00d, type=0x2) [win.c:1154] in user32.dll.so) (ebp=405d6c0c) 9170 0x4068c17e (CreateWindowExA+0x14e(exStyle=0x0, className=0x411b3c, windowName=0x41c60d50, style=0x80880000, x=0x0, y=0x0, width=0x400, height=0x300, parent=0x0, menu=0x0, instance=0x400000, data=0x0) [win.c:1309] in user32.dll.so) (ebp=405d6d60) 9171 0x00405cd1 (splash.exe..text+0x4cd1 in H:\splash\splash.exe) (ebp=405d6e90) 9172 0x400ca01b (start_process+0x24b [process.c:526] in libntdll.dll.so) (ebp=405d6f38) 9173 0x400ce257 (call_on_thread_stack+0x27(func=0x400c9dd0) [sysdeps.c:105] in libntdll.dll.so) (ebp=405d6ff4) 9174 0x400ce410 (SYSDEPS_CallOnStack+0x14 in libntdll.dll.so) (ebp=00000000) Index: winpos.c =================================================================== RCS file: /home/wine/wine/windows/winpos.c,v retrieving revision 1.133 diff -u -r1.133 winpos.c --- winpos.c 25 May 2002 22:16:13 -0000 1.133 +++ winpos.c 29 May 2002 17:38:49 -0000 @@ -1278,8 +1278,10 @@ PERQDATA_SetActiveWnd( pOldActiveQueue->pQData, 0 ); /* send palette messages */ + /* if (hWnd && SendMessageW( hWnd, WM_QUERYNEWPALETTE, 0, 0L)) SendMessageW( HWND_BROADCAST, WM_PALETTEISCHANGING, (WPARAM)hWnd, 0 ); + */ /* if prev wnd is minimized redraw icon title */ if( IsIconic( hwndPrevActive ) ) WINPOS_RedrawIconTitle(hwndPrevActive);

2 2

RE: Stripping of whitespaces at the end of lines
by Patrik Stridvall 29 May '02

29 May '02

> Enforcing 'no space before the non-blank char on every line' is > orders of magniture more difficult than 'indentation is 4 space' Is it? find . -name \*.c | xargs perl -ne '/^\s*/ && $& =~ / / && do { print }' compared to find . -name \*.c | xargs perl -ne '!/^(?: )*/ && do { print }' Of course both will fail on Wine style function header comments but then you got what you asked for above. Anyway. I fully agree what you (Dimitrie) that the point of indentation is to make it human readable not easy for tools to verify. In fact I think I agree with almost everything you have said on the subject. So for once we agree. :-)

1 0

Quick question on the Cygwin/Mingw port
by Steven Edwards 29 May '02

29 May '02

Hola, I'm on the road traveling till Monday so I sent the question here rather then cygwin support because I figured someone here has to have a little cygwin and mingw experience. I'm using cygwin to host my mingw WINE devel with --target=mingw --host-mingw --build=mingw -mno-cygwin and -fnative-struct and it is still linking to cygwin1.dll. I don't really understand whats going on because it doesn't seem to be using any of the exports from cygwin1.dll but if I try to test the dlls/programs/tools on another box it bitches about missing the dll. If anyone has any insight please let me know or just send this to /dev/null Thanks Steven "Every revolution was once a thought in one man's mind" - Ralph Waldo Emerson

3 3

Use of wine RPC implementation for TAPI 2 client ?
by Yann Dirson 29 May '02

29 May '02

We're investigating the implementation of a TAPI 2 client lib on Linux. Wine looks like one of the projects with an implementation of DCE/MS-RPC. What's the status of this RPC implementation, in terms of specs coverage, and stability ? Do you have an MIDL compiler ? Regards, -- Yann Dirson <Yann.Dirson(a)fr.alcove.com> http://www.alcove.com/ Technical support manager Responsable de l'assistance technique Senior Free-Software Consultant Consultant senior en Logiciels Libres Debian developer (dirson(a)debian.org) Développeur Debian

2 1

More regression errors ...
by Paul Millar 29 May '02

29 May '02

Hi, I've noticed that amongst last night's slew of patches, Mehmet's regression tests were applied without his patches that correct wine's behaviour. Wine now fails regression tests on all platforms. If his patches are delayed for some reason, could/should we use todo_wine to acknowledge that these failures are known? If so, I've attached a trivial patch to do this for the tests that failed for me ... ChangeLog: Mark failing regression tests as todo Cheers, ---- Paul Millar

2 1

Re: [Bug 722] Changed - GDI.EXE
by Andriy Palamarchuk 29 May '02

29 May '02

--- Johan Gill <johane(a)lysator.liu.se> wrote: > On Tue, 28 May 2002 wine-bugs(a)winehq.com wrote: > > > Some rights wouldn't hurt btw. > > + > > + > > + ------- Additional Comments From > Speeddymon(a)yahoo.com 2002-05-28 15:14 ------- > > + Send an email to wine-bugs(a)winehq.com asking > someone to give you additional > > + BugZilla priveleges so that you may work on bug > 722 > > So here's an offical request for privileges. Done. Official congratulations with new privileges :-) Andriy __________________________________________________ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com

1 0

Main window size difference
by Guy L. Albertelli 29 May '02

29 May '02

Has anyone else noticed that the size of main windows under Windows is much different than under Wine. As an example, using the Tab control test program (from the ControlSpy series). Under Windows the overall size appears to be about 819x564. Under Wine it is about 505x268. (I may be off by 1 or 2 pixels, but not 30% to 40%.) And yes my screen resolution under both systems is 1024x768. This obviously makes doing pixel matching, or even item layout matching difficult. Guy Albertelli <<galberte(a)neo.lrun.com>>

2 1

Re: Idea for people with priveleges
by Andriy Palamarchuk 29 May '02

29 May '02

--- Dustin Navea <speeddymon(a)yahoo.com> wrote: > Anyone that can modify the layout of the BugZilla > site > should take this into consideration: > > People with priveleges as you know can pretty much > do > what they need to with a bug, i.e. if it is resolved > and the person working on it forgot to set it as > resolved, anyone with preveleges can do that. But, > what about if we havent heard from the reporter in a > while and we need to knock the bug out, our only > option is to ask the reporter if it has been > resolved, > tell them to try it with a recent wine build and to > try some of the things mentioned in the comments, > and > then mark it as resolved and them after that mark it > as closed, so it is shown as closed and not resolved > on the bug list... > > why not give people with priveleges the ability to > close it without resolving it, so that they only > have > to make 1 post instead of 2? Plus that way we have > bugs that are actually resolved marked so, and ones > that we are unsure of marked as closed but can > always > be reopened... Why do you want to close such bug in the first place? IMHO it is Ok to put the bug in the RESOLVED state and wait some time if the original reporter provides feedback or more information will become available. Keeping the bug in RESOLVED state will insure that somebody will look at the bug one more time. Is this too many steps? >From Bugzilla docs: "QA looks at resolved bugs and to make sure the appropriate action has been taken. If they agree, the bug is marked VERIFIED. Bugs remain in this state until the product ships, at which time the bug is marked CLOSED." Currently we do not have one more "verification layer" for bugs. Andriy __________________________________________________ Do You Yahoo!? Yahoo! - Official partner of 2002 FIFA World Cup http://fifaworldcup.yahoo.com

1 0

Something's wrong with hypermail
by Johan Gill 29 May '02

29 May '02

I've noticed that selecting 'Next message' in the web interface to the mailinglists sometimes skips messages. For example, take a look at my latest messages to wine-patches. Standing at the second 'PATCH: DirectSound', the next message is said to be 'Re: RESUBMIT: PATCH: DirectSound'. Hang on...this is the case for my mmioAdvance patch also! Ok, so my patches seem to confuse the interface :/ /Johan Gill, johane(a)lysator.liu.se

1 0