April 2002 - Wine-Devel - list.winehq.org

Patch: async-file.diff (4 in series)
by Martin Wilck April 12, 2002

April 12, 2002

Patch: async-file.diff Some improvements to the async IO API. - Need a "cleanup" method. - calling the "call_completion" method unconditionally in finish_async (introduced by me) was wrong, because the thread may never be in an alertable wait state -> call it only if user completion function is present. - We need two constant "ops" objects, one with and one without call_completion method (the event field may NOT be used to determine whether the completion must be called or not). - STATUS_CANCELLED is correct for cancelled requests ( corresponds to ERROR_OPERATION_ABORTED ). Modified files: dlls/kernel: comm.c files: file.c include: async.h server: async.c diff -ruNX ignore TMP/wine/dlls/kernel/comm.c MW/wine/dlls/kernel/comm.c --- TMP/wine/dlls/kernel/comm.c Fri Apr 12 11:47:48 2002 +++ MW/wine/dlls/kernel/comm.c Fri Apr 12 12:27:07 2002 @@ -99,14 +99,15 @@ static DWORD commio_get_async_status (const async_private *ovp); static DWORD commio_get_async_count (const async_private *ovp); static void commio_set_async_status (async_private *ovp, const DWORD status); -static void CALLBACK commio_call_completion_func (ULONG_PTR data); +static void commio_async_cleanup (async_private *ovp); static async_ops commio_async_ops = { commio_get_async_status, /* get_status */ commio_set_async_status, /* set_status */ commio_get_async_count, /* get_count */ - commio_call_completion_func /* call_completion */ + NULL, /* call_completion */ + commio_async_cleanup /* cleanup */ }; typedef struct async_commio @@ -131,9 +132,9 @@ return 0; } -static void CALLBACK commio_call_completion_func (ULONG_PTR data) +static void commio_async_cleanup (async_private *ovp) { - HeapFree(GetProcessHeap(), 0, (void*) data); + HeapFree(GetProcessHeap(), 0, ovp ); } /***********************************************************************/ diff -ruNX ignore TMP/wine/files/file.c MW/wine/files/file.c --- TMP/wine/files/file.c Fri Apr 12 11:59:51 2002 +++ MW/wine/files/file.c Fri Apr 12 12:28:22 2002 @@ -88,13 +88,24 @@ static DWORD fileio_get_async_count (const async_private *ovp); static void fileio_set_async_status (async_private *ovp, const DWORD status); static void CALLBACK fileio_call_completion_func (ULONG_PTR data); +static void fileio_async_cleanup (async_private *ovp); static async_ops fileio_async_ops = { fileio_get_async_status, /* get_status */ fileio_set_async_status, /* set_status */ fileio_get_async_count, /* get_count */ - fileio_call_completion_func /* call_completion */ + fileio_call_completion_func, /* call_completion */ + fileio_async_cleanup /* cleanup */ +}; + +static async_ops fileio_nocomp_async_ops = +{ + fileio_get_async_status, /* get_status */ + fileio_set_async_status, /* set_status */ + fileio_get_async_count, /* get_count */ + NULL, /* call_completion */ + fileio_async_cleanup /* cleanup */ }; typedef struct async_fileio @@ -128,12 +139,16 @@ async_fileio *ovp = (async_fileio*) data; TRACE ("data: %p\n", ovp); - if (ovp->completion_func) - ovp->completion_func(ovp->lpOverlapped->Internal, - ovp->lpOverlapped->InternalHigh, - ovp->lpOverlapped); + ovp->completion_func( ovp->lpOverlapped->Internal, + ovp->lpOverlapped->InternalHigh, + ovp->lpOverlapped ); + + fileio_async_cleanup ( &ovp->async ); +} - HeapFree(GetProcessHeap(), 0, ovp); +static void fileio_async_cleanup ( struct async_private *ovp ) +{ + HeapFree ( GetProcessHeap(), 0, ovp ); } /*********************************************************************** @@ -1479,6 +1494,8 @@ { async_fileio *ovp; int fd; + int flags; + enum fd_type type; TRACE("file %d to buf %p num %ld %p func %p\n", hFile, buffer, bytesToRead, overlapped, lpCompletionRoutine); @@ -1490,10 +1507,11 @@ return FALSE; } - fd = FILE_GetUnixHandle( hFile, GENERIC_READ ); - if(fd<0) + fd = FILE_GetUnixHandleType ( hFile, GENERIC_READ, &type, &flags); + if ( fd < 0 ) { - TRACE("Couldn't get FD\n"); + WARN ( "Couldn't get FD\n" ); + SetLastError ( ERROR_INVALID_PARAMETER ); return FALSE; } @@ -1502,11 +1520,10 @@ { TRACE("HeapAlloc Failed\n"); SetLastError(ERROR_NOT_ENOUGH_MEMORY); - close(fd); - return FALSE; + goto error; } - ovp->async.ops = &fileio_async_ops; + ovp->async.ops = ( lpCompletionRoutine ? &fileio_async_ops : &fileio_nocomp_async_ops ); ovp->async.handle = hFile; ovp->async.fd = fd; ovp->async.type = ASYNC_TYPE_READ; @@ -1518,6 +1535,11 @@ ovp->buffer = buffer; return !register_new_async (&ovp->async); + +error: + close (fd); + return FALSE; + } /*********************************************************************** @@ -1691,6 +1713,8 @@ { async_fileio *ovp; int fd; + int flags; + enum fd_type type; TRACE("file %d to buf %p num %ld %p func %p stub\n", hFile, buffer, bytesToWrite, overlapped, lpCompletionRoutine); @@ -1701,7 +1725,7 @@ return FALSE; } - fd = FILE_GetUnixHandle( hFile, GENERIC_WRITE ); + fd = FILE_GetUnixHandleType ( hFile, GENERIC_WRITE, &type, &flags ); if ( fd < 0 ) { TRACE( "Couldn't get FD\n" ); @@ -1713,8 +1737,7 @@ { TRACE("HeapAlloc Failed\n"); SetLastError(ERROR_NOT_ENOUGH_MEMORY); - close (fd); - return FALSE; + goto error; } ovp->async.ops = &fileio_async_ops; @@ -1729,6 +1752,10 @@ ovp->completion_func = lpCompletionRoutine; return !register_new_async (&ovp->async); + +error: + close (fd); + return FALSE; } /*********************************************************************** diff -ruNX ignore TMP/wine/include/async.h MW/wine/include/async.h --- TMP/wine/include/async.h Fri Apr 12 11:47:48 2002 +++ MW/wine/include/async.h Fri Apr 12 12:25:30 2002 @@ -35,6 +35,7 @@ typedef DWORD (*async_get_status)(const struct async_private *ovp); typedef DWORD (*async_get_count)(const struct async_private *ovp); typedef void (*async_set_status)(struct async_private *ovp, const DWORD status); +typedef void (*async_cleanup)(struct async_private *ovp); typedef struct async_ops { @@ -42,6 +43,7 @@ async_set_status set_status; async_get_count get_count; async_call_completion_func call_completion; + async_cleanup cleanup; } async_ops; typedef struct async_private @@ -74,7 +76,10 @@ if( ovp->event != INVALID_HANDLE_VALUE ) NtSetEvent( ovp->event, NULL ); - QueueUserAPC( ovp->ops->call_completion, GetCurrentThread(), (ULONG_PTR)ovp ); + if ( ovp->ops->call_completion ) + QueueUserAPC( ovp->ops->call_completion, GetCurrentThread(), (ULONG_PTR)ovp ); + else + ovp->ops->cleanup ( ovp ); } inline static BOOL __register_async( async_private *ovp, const DWORD status ) @@ -92,7 +97,11 @@ } SERVER_END_REQ; - if ( ret ) ovp->ops->set_status ( ovp, GetLastError() ); + if ( ret ) { + SetLastError( RtlNtStatusToDosError(ret) ); + ovp->ops->set_status ( ovp, ret ); + } + if ( ovp->ops->get_status (ovp) != STATUS_PENDING ) finish_async (ovp); diff -ruNX ignore TMP/wine/server/async.c MW/wine/server/async.c --- TMP/wine/server/async.c Fri Apr 12 11:47:48 2002 +++ MW/wine/server/async.c Fri Apr 12 12:23:56 2002 @@ -71,7 +71,7 @@ { while(q->head) { - async_notify(q->head, STATUS_HANDLES_CLOSED); + async_notify(q->head, STATUS_CANCELLED); destroy_async(q->head); } }

1 0

PATCH: winemaker.diff (3 in series)
by Martin Wilck April 12, 2002

April 12, 2002

PATCH: winemaker.diff (identical to: http://www.winehq.com/hypermail/wine-devel/2002/04/0187.html) Fix winemaker-generated references to dlls that have changed their names in configure.ac. Patch against: Wine CVS 2002-04-12 Modified Files: tools: winemaker diff -ruNX ignore TMP/wine/tools/winemaker MW/wine/tools/winemaker --- TMP/wine/tools/winemaker Tue Apr 9 13:19:57 2002 +++ MW/wine/tools/winemaker Tue Apr 9 13:29:27 2002 @@ -2732,14 +2732,14 @@ WINE_DLL_ROOT="/lib:/lib/dlls:/usr/lib:/usr/lib/dlls:/usr/local/lib:/usr/local/lib/dlls" fi fi -AC_PATH_FILE(WINE_DLL_ROOT,[libntdll.so],[ - AC_MSG_ERROR([Could not find the Wine dlls (libntdll.so)]) +AC_PATH_FILE(WINE_DLL_ROOT,[libntdll.dll.so],[ + AC_MSG_ERROR([Could not find the Wine dlls (libntdll.dll.so)]) ],[$WINE_DLL_ROOT]) -AC_PATH_LIBRARY(WINE_DLL_ROOT,[-lntdll],[$WINE_LIBRARY_PATH -lwine -lwine_unicode],[ - AC_MSG_ERROR([Could not link with the Wine dlls (libntdll.so)]) +AC_PATH_LIBRARY(WINE_DLL_ROOT,[-lntdll.dll],[$WINE_LIBRARY_PATH -lwine -lwine_unicode],[ + AC_MSG_ERROR([Could not link with the Wine dlls (libntdll.dll.so)]) ],[$WINE_DLL_ROOT]) -WINE_DLL_PATH="-L$WINE_DLL_ROOT" +WINE_DLL_PATH="-L$WINE_DLL_ROOT/wine" if test -n "$LD_PATH" then

1 0

PATCH: sock-fd-type.diff (2 in series)
by Martin Wilck April 12, 2002

April 12, 2002

PATCH: sock-fd-type.diff Patch against: Wine CVS 2002-04-12 Test status: Compiles (no errors/warnings). Modified Files: dlls/winsock : socket.c Log message: Martin Wilck <Martin.Wilck(a)Fujitsu-Siemens.com> _get_sock_fd_type : New function, retrieve fd type and flags like FILE_GetUnixHandleType(). WSASendTo() /WSARecvFrom(): Use _get_sock_fd_type(). diff -ruNX ignore TMP/wine/dlls/winsock/socket.c MW/wine/dlls/winsock/socket.c --- TMP/wine/dlls/winsock/socket.c Tue Apr 9 13:27:56 2002 +++ MW/wine/dlls/winsock/socket.c Tue Apr 9 13:28:08 2002 @@ -246,6 +246,13 @@ return fd; } +inline static int _get_sock_fd_type( SOCKET s, enum fd_type *type, int *flags ) +{ + int fd; + if (set_error( wine_server_handle_to_fd( s, GENERIC_READ, &fd, type, flags ) )) return -1; + return fd; +} + static void _enable_event(SOCKET s, unsigned int event, unsigned int sstate, unsigned int cstate) { @@ -1976,15 +1983,16 @@ LPWSAOVERLAPPED lpOverlapped, LPWSAOVERLAPPED_COMPLETION_ROUTINE lpCompletionRoutine ) { - int i, n, fd, err = WSAENOTSOCK; + int i, n, fd, err = WSAENOTSOCK, flags; struct iovec* iovec; struct msghdr msghdr; - + enum fd_type type; + TRACE ("socket %04x, wsabuf %p, nbufs %ld, flags %ld, to %p, tolen %d, ovl %p, func %p\n", s, lpBuffers, dwBufferCount, dwFlags, to, tolen, lpOverlapped, lpCompletionRoutine); - fd = _get_sock_fd(s); + fd = _get_sock_fd_type( s, &type, &flags ); if ( fd == -1 ) goto error; @@ -3269,14 +3277,15 @@ struct iovec* iovec; struct msghdr msghdr; - int fd, i, length, err = WSAENOTSOCK; + int fd, i, length, err = WSAENOTSOCK, flags; + enum fd_type type; TRACE("socket %04x, wsabuf %p, nbufs %ld, flags %ld, from %p, fromlen %ld, ovl %p, func %p\n", s, lpBuffers, dwBufferCount, *lpFlags, lpFrom, (lpFromlen ? *lpFromlen : -1L), lpOverlapped, lpCompletionRoutine); - fd = _get_sock_fd(s); + fd = _get_sock_fd_type( s, &type, &flags ); if (fd == -1) {

1 0

PATCH: async-immediate.diff (1 in series)
by Martin Wilck April 12, 2002

April 12, 2002

PATCH: async-immediate.diff (Resubmission of http://www.winehq.com/hypermail/wine-patches/2002/04/0021.html) This patch uses a more elegant and (IMO) more wine-ish method to check for possible immediate completion of asynchronous IO requests in ReadFile() and WriteFile(). The implementation in current CVS is wrong for 2 reasons: 1. The overlapped event is not set, nor the completion routine called, if immediate completion succeeds. This breaks apps that do not check for immediate success. 2. If there are async requests currently in the queue, immediate completion will bypass the request ordering. This will lead to data corruption on FIFO-type files. The patch fixes this. Patch against: Wine CVS 2002-04-12 Test status: Compiles (no errors/warnings). Tested for regular file IO (ok). The 2002-01-15 patch has been tested for 16-bit serial IO by Lawson Whitney (ok). Modified Files: files : file.c Log message: Martin Wilck <Martin.Wilck(a)Fujitsu-Siemens.com> GetOverlappedResult(): Return ERROR_IO_INCOMPLETE if IO still pending (MSDN docs) ReadFile() / WriteFile(): Use GetOverlappedResult() to check for immediate completion. diff -ruNX ignore TMP/wine/files/file.c MW/wine/files/file.c --- TMP/wine/files/file.c Tue Apr 9 13:20:25 2002 +++ MW/wine/files/file.c Tue Apr 9 13:25:25 2002 @@ -1395,7 +1395,8 @@ if(lpTransferred) *lpTransferred = lpOverlapped->InternalHigh; - SetLastError(lpOverlapped->Internal); + SetLastError ( lpOverlapped->Internal == STATUS_PENDING ? + ERROR_IO_INCOMPLETE : lpOverlapped->Internal ); return (r==WAIT_OBJECT_0); } @@ -1576,41 +1577,21 @@ SetLastError(ERROR_INVALID_PARAMETER); return FALSE; } - - /* see if we can read some data already (this shouldn't block) */ - result = pread( unix_handle, buffer, bytesToRead, OVERLAPPED_OFFSET(overlapped) ); - if ((result < 0) && (errno == ESPIPE)) - result = read( unix_handle, buffer, bytesToRead ); - close(unix_handle); - - if(result<0) - { - if( (errno!=EAGAIN) && (errno!=EINTR) && - ((errno != EFAULT) || IsBadWritePtr( buffer, bytesToRead )) ) - { - FILE_SetDosError(); - return FALSE; - } - else - result = 0; - } - /* if we read enough to keep the app happy, then return now */ - if(result>=bytesToRead) - { - *bytesRead = result; - return TRUE; - } - - /* at last resort, do an overlapped read */ - overlapped->InternalHigh = result; + close(unix_handle); + overlapped->InternalHigh = 0; if(!FILE_ReadFileEx(hFile, buffer, bytesToRead, overlapped, NULL, overlapped->hEvent)) return FALSE; - /* fail on return, with ERROR_IO_PENDING */ - SetLastError(ERROR_IO_PENDING); - return FALSE; + if ( !GetOverlappedResult (hFile, overlapped, bytesRead, FALSE) ) + { + if ( GetLastError() == ERROR_IO_INCOMPLETE ) + SetLastError ( ERROR_IO_PENDING ); + return FALSE; + } + + return TRUE; } if (flags & FD_FLAG_TIMEOUT) { @@ -1790,43 +1771,20 @@ return FALSE; } - /* see if we can write some data already (this shouldn't block) */ - - result = pwrite( unix_handle, buffer, bytesToWrite, OVERLAPPED_OFFSET (overlapped) ); - if ((result < 0) && (errno == ESPIPE)) - result = write( unix_handle, buffer, bytesToWrite ); - close(unix_handle); + overlapped->InternalHigh = 0; - if(result<0) - { - if( (errno!=EAGAIN) && (errno!=EINTR) && - ((errno != EFAULT) || IsBadReadPtr( buffer, bytesToWrite )) ) - { - FILE_SetDosError(); - return FALSE; - } - else - result = 0; - } + if(!FILE_WriteFileEx(hFile, buffer, bytesToWrite, overlapped, NULL, overlapped->hEvent)) + return FALSE; - /* if we wrote enough to keep the app happy, then return now */ - if(result>=bytesToWrite) + if ( !GetOverlappedResult (hFile, overlapped, bytesWritten, FALSE) ) { - *bytesWritten = result; - return TRUE; - } - - /* at last resort, do an overlapped read */ - overlapped->Internal = STATUS_PENDING; - overlapped->InternalHigh = result; - - if(!FILE_WriteFileEx(hFile, buffer, bytesToWrite, overlapped, NULL, overlapped->hEvent)) + if ( GetLastError() == ERROR_IO_INCOMPLETE ) + SetLastError ( ERROR_IO_PENDING ); return FALSE; + } - /* fail on return, with ERROR_IO_PENDING */ - SetLastError(ERROR_IO_PENDING); - return FALSE; + return TRUE; } switch(type)

1 0

Re: Make dumping more robust
by juergen.schmied＠debitel.net April 12, 2002

April 12, 2002

I just rewrote the dumping of pidl's because the dumping itself produces lots of relay/snoop messages. So I made it not call any other functions. Will send it soon. juergen

2 1

Expect Trouble
by Shachar Shemesh April 12, 2002

April 12, 2002

Hi list, I am sorry if this is a baisc question. I have not been able to find anything about it in the docs. I am trying to run a small program I wrote and compiled with VC6 on Windows. I am getting the following error: Invoking /usr/bin/wine.bin HebTest.exe ... Warning: loading builtin kernel32.dll, but native version already present. Expect trouble. Warning: loading builtin user32.dll, but native version already present. Expect trouble. wine: Unhandled exception, starting debugger... err:seh:start_debugger Couldn't start debugger ("debugger/winedbg 134661992 32") (2) Read the Wine Developers Guide on how to set up winedbg or another debugger Wine failed with return code 5 I get the same error whether I am using my own CVS compiled version, or the nightly CVS for debian SID. I get this error regardless of whether I am using fake windows or real windows. I even tried removing the user32.dll file (0 bytes) completely from C:\Windows\System of the fake windows. Attached is the output when running wine with --debugmsg +all, and the binary I am using to run this program (it is a debug build). It's a tiny program of my writing, so unless I unwittedly stole it from somewhere, you hereby have my permission to use the binary as you see fit. Sources are available upon request. Thanks Shachar

1 0

Re: Handling of messages above WM_USER in pack_message
by Alexandre Julliard April 12, 2002

April 12, 2002

Uwe Bonnes <bon(a)elektron.ikp.physik.tu-darmstadt.de> writes: > I don't know if this global approach is valid, but at least it helped me to > get winhlp32 working called with WinHelpA from an application. In what other > way can we cope with user defined messages and their related structures in > inter-process SendMessage? We can't. User messages cannot pass data around, WinHelpA will need to do that differently. The best way to find out how is probably to watch winhlp32 with a message spy under Windows. -- Alexandre Julliard julliard(a)winehq.com

2 1

NT Status vs. Dos Error code in async operations
by Martin Wilck April 12, 2002

April 12, 2002

Hi, If my understanding of the MS docs is correct, the status of overlapped requests is stored as an NT status in lpOverlapped->Internal (As Mike implemented it). The error code returned by GetOverlappedResult() or passed to the completion routine, though, is a DOS error code. Thus in priciple we'd need to convert the status with RtlNtStatusToDosError(), right ? Moreover, if our async read()/write() operations fail, we simply set the status to STATUS_UNSUCCESSFUL which translates to ERROR_GEN_FAILURE - not very meaningful to our users. What we'd need would consequently be a errno->NT status converter to set the overlapped status correctly from the error reported by read or write. I can't seem to find one in the wine code (??). There only seem to be local errno -> DOS error converters, and no DOS error->Nt status converter. What should we do? Martin -- Martin Wilck Phone: +49 5251 8 15113 Fujitsu Siemens Computers Fax: +49 5251 8 20409 Heinz-Nixdorf-Ring 1 mailto:Martin.Wilck(a)Fujitsu-Siemens.com D-33106 Paderborn http://www.fujitsu-siemens.com/primergy

2 1

Async IO on sockets: good news and bad news
by Martin Wilck April 11, 2002

April 11, 2002

Hi, The good news is: I have the basic mechanism for async IO on sockets running. Requests can be scheduled and waited for, just like for regular files. The bad news is: This is not enough for sockets. Problems occur especially when sockets are closed, because the signalling of socket events uses a completely independent mechanism. This leads to all sorts of strange behaviour, mostly to data loss because "FD_CLOSE" is signalled prematurely. Sadly, this means for me: - The whole event signalling mechanism needs to be reworked so that it is consistent with overlapped IO. - Probably the async IO mechanism, too. It seems that it would be good to have a single IO queue where all events are handled, rather than several ones. Unlike now, also things like "peer closed socket" would be scheduled in that queue. Well I need to think more. Currently, I am not even sure how exactly it _should_ work. Martin -- Martin Wilck Phone: +49 5251 8 15113 Fujitsu Siemens Computers Fax: +49 5251 8 20409 Heinz-Nixdorf-Ring 1 mailto:Martin.Wilck(a)Fujitsu-Siemens.com D-33106 Paderborn http://www.fujitsu-siemens.com/primergy

1 0

16-bit apps starting and stopping notification
by Andriy Palamarchuk April 10, 2002

April 10, 2002

I found an issue for a hang of 16-bit application (www.formdocs.com) installer. The app registers with NotifyRegister16 callback function and looks like it hangs while waiting for completion of child process. The last part of relay trace is attached. NotifyRegister16 function of ToolHelp dll is used to register handler for following events: NFY_UNKNOWN NFY_LOADSEG NFY_FREESEG NFY_STARTDLL NFY_STARTTASK NFY_EXITTASK NFY_DELMODULE NFY_RIP NFY_TASKIN NFY_TASKOUT NFY_INCHAR NFY_OUTSTRING NFY_LOGERROR NFY_LOGPARAMERROR Note, this works for 16-bit applications only. These events also used by debuggers to debug 16-bit applications. Can somebody give me a hint how is better to approach implementation of NFY_EXITTASK, NFY_STARTTASK events generation? How important do you think is to implement this functionality, since it is for 16-bit apps only? Andriy Palamarchuk __________________________________________________ Do You Yahoo!? Yahoo! Tax Center - online filing with TurboTax http://taxes.yahoo.com/

1 0