- Wine-Devel - list.winehq.org

Re: spi.pszProcessName should be UNICODE_STRING
by Paul Vriens June 6, 2005

June 6, 2005

On Mon, 2005-06-06 at 12:15, Felix Nawothnig wrote: > This lets the taskname appear in WinXP's taskmgr.exe. > > ChangeLog: > Make SYSTEM_THREAD_INFORMATION.pszProcessName an UNICODE_STRING > > ______________________________________________________________________ Thanks for the fix. The change is however done to the SYSTEM_PROCESS_INFORMATION struct :-). The process name now shows nicely in Process Explorer as well. I saw it's committed so I'll build on that one. Paul.

1 0

Re: Temp fix for SetWindowTextA()
by Christoph Frick June 6, 2005

June 6, 2005

On Sun, Jun 05, 2005 at 02:26:40PM -0400, Jason Campbell wrote: > A possible better patch may be to see if D2 is a child of the D2 bot > (since most bots can load D2) and allow the title change if it is and > not allow it if it's not a parent/child. there are no stupid questions, so i ask: i doubt that windows has there some "if d2 bot, then remove restrictions"... either this software overwrites this data in the process it spawns (many trainers or actually even mods (Grand Prix Legends) work this way. or windows allows changing the title for the spaning process. so the best thing to do is write a test for this, try it in windows and wine and do the correct thing then? -- cu

1 0

Re: [msvcrt] Implement %I types for printf
by Dmitry Timoshkov June 6, 2005

June 6, 2005

"Jesse Allen" <the3dfxdude(a)gmail.com> wrote: > + if( flags->IntegerLength == 'I' ) > + { > + sprintf(p, "l"); > + p++; > + } > + if( flags->IntegerSize == INTEGERSIZE_LONGLONG ) > + { > + sprintf(p, "l"); > + p++; > + } sprintf is very inefficient here, why not use simple strcpy instead? -- Dmitry.

2 1

Half Life 2 error -- err:mmio:MMIO_ParseExtA
by J. Grant June 6, 2005

June 6, 2005

Hello, I am getting this error when trying to run Half Life2 in err:mmio:MMIO_ParseExtA. Could someone tell me if this is a problem in in WINE, or a problem in my setup? Or not an error at all..? Also if there is already a document on setting up Half Life2 could someone direct me to it please, I could not find one already. $ wcmd Invoking /usr/lib/wine/wine.bin wcmd.exe ... WCMD Version 0.17 Z:\home\now3d>cd C:\opt\games\steam\ C:\opt\games\steam>C:\opt\games\steam\Steam.exe -applaunch 220 C:\opt\games\steam>err:dc:CreateDCW no driver found for L"WINEPS" fixme:winspool:AddPrinterW DocumentPropertiesW on printer 'L"deskjet930c"' fails err:dc:CreateDCW no driver found for L"WINEPS" err:dc:CreateDCW no driver found for L"WINEPS" fixme:winspool:AddPrinterW DocumentPropertiesW on printer 'L"hp930c"' fails err:dc:CreateDCW no driver found for L"WINEPS" err:mmio:MMIO_ParseExtA No . in szFileName: "" err:mmio:MMIO_ParseExtA No . in szFileName: "C:\\opt\\games\\steam\\" You need to install the Mozilla ActiveX control to use Wine's builtin CLSID_WebBrowser from SHDOCVW.DLL fixme:ole:CoCreateInstance no instance created for interface {00000000-0000-0000-c000-000000000046} of class {8856f961-340a-11d0-a96b-00c04fd705a2},hres is 0x80040111 fixme:debug_buffer:RtlCreateQueryDebugBuffer (0, 0): stub fixme:debug_buffer:RtlCreateQueryDebugBuffer (96, 0): returning 0x40473b08 fixme:debug_buffer:RtlQueryProcessDebugInformation (11, 80000001, 0x40473b08): stub fixme:debug_buffer:RtlDestroyQueryDebugBuffer (0x40473b08): stub C:\opt\games\steam> C:\opt\games\steam> C:\opt\games\steam>Shutting down. . . 30client callback thread error C:\opt\games\steam>STEAM.exe "steam://open/registerproduct/220" C:\opt\games\steam>err:mmio:MMIO_ParseExtA No . in szFileName: "" err:mmio:MMIO_ParseExtA No . in szFileName: "C:\\opt\\games\\steam\\" fixme:sync:CreateIoCompletionPort (0xffffffff, (nil), 00000000, 00000000): stub. fixme:crypt:CRYPT_VerifyImage (rsaenh.dll, 0x4046b9a0): not verifying image Shutting down. . . 1completionportmanager.cpp (28) : Assertion Failed: m_hIOCP C:\opt\games\steam>exit -- Homepage: http://jguk.org/ Blog: http://jguk.org/blog.rss Radio: http://jguk.org/#radio

4 7

Re: shell32: BindToObject should fail, if called with empty pidl
by Uwe Bonnes June 6, 2005

June 6, 2005

>>>>> "Michael" == Michael Jung <mjung(a)iss.tu-darmstadt.de> writes: Michael> This fixes the problem in DVD Shrink's SHBrowseForFolder Michael> dialog, reported by Wolfgang Scherer. Can you perhaps provide a testcase? I pays off the long term... Bye -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

2 2

WWN #277 correction.
by Chuck Hall June 5, 2005

June 5, 2005

In the WWN #277 there is mention that I am using Solaris 9 while attempting to get Wine to work. I am actually using Solaris 10 not Solaris 9. What I did was compare what was different between Solaris 9 and Solaris 10 so I could figure out what the problem is/was. -- Have Fun! Chuck Hall

1 0

Re: wine/ include/wine/server.h dlls/ntdll/server. ...
by Robert Shearman June 5, 2005

June 5, 2005

Alexandre Julliard wrote: >Index: wine/include/wine/server.h >diff -u -p wine/include/wine/server.h:1.21 wine/include/wine/server.h:1.22 >--- wine/include/wine/server.h:1.21 Sun Jun 5 20:45:53 2005 >+++ wine/include/wine/server.h Sun Jun 5 20:45:53 2005 >@@ -55,6 +55,7 @@ extern void wine_server_send_fd( int fd > extern int wine_server_fd_to_handle( int fd, unsigned int access, int inherit, obj_handle_t *handle ); > extern int wine_server_handle_to_fd( obj_handle_t handle, unsigned int access, int *unix_fd, int *flags ); > extern void wine_server_release_fd( obj_handle_t handle, int unix_fd ); >+extern void DECLSPEC_NORETURN wine_server_exit_thread( int status ); > > /* do a server call and set the last error code */ > inline static unsigned int wine_server_call_err( void *req_ptr ) > >Index: wine/dlls/ntdll/server.c >diff -u -p wine/dlls/ntdll/server.c:1.21 wine/dlls/ntdll/server.c:1.22 >--- wine/dlls/ntdll/server.c:1.21 Sun Jun 5 20:45:53 2005 >+++ wine/dlls/ntdll/server.c Sun Jun 5 20:45:53 2005 >@@ -119,6 +119,40 @@ static void fatal_perror( const char *er > > > /*********************************************************************** >+ * wine_server_exit_thread (NTDLL.@) >+ */ >+void wine_server_exit_thread( int status ) >+{ >+ struct wine_pthread_thread_info info; >+ ULONG size; >+ >+ RtlAcquirePebLock(); >+ RemoveEntryList( &NtCurrentTeb()->TlsLinks ); >+ RtlReleasePebLock(); >+ >+ info.stack_base = NtCurrentTeb()->DeallocationStack; >+ info.teb_base = NtCurrentTeb(); >+ info.teb_sel = wine_get_fs(); >+ info.exit_status = status; >+ >+ size = 0; >+ NtFreeVirtualMemory( GetCurrentProcess(), &info.stack_base, &size, MEM_RELEASE | MEM_SYSTEM ); >+ info.stack_size = size; >+ >+ size = 0; >+ NtFreeVirtualMemory( GetCurrentProcess(), &info.teb_base, &size, MEM_RELEASE | MEM_SYSTEM ); >+ info.teb_size = size; >+ >+ sigprocmask( SIG_BLOCK, &block_set, NULL ); >+ close( NtCurrentTeb()->wait_fd[0] ); >+ close( NtCurrentTeb()->wait_fd[1] ); >+ close( NtCurrentTeb()->reply_fd ); >+ close( NtCurrentTeb()->request_fd ); >+ wine_pthread_exit_thread( &info ); >+} >+ >+ >+/*********************************************************************** > * server_abort_thread > */ > void server_abort_thread( int status ) > >Index: wine/dlls/ntdll/ntdll.spec >diff -u -p wine/dlls/ntdll/ntdll.spec:1.182 wine/dlls/ntdll/ntdll.spec:1.183 >--- wine/dlls/ntdll/ntdll.spec:1.182 Sun Jun 5 20:45:53 2005 >+++ wine/dlls/ntdll/ntdll.spec Sun Jun 5 20:45:53 2005 >@@ -1027,6 +1027,7 @@ > @ cdecl wine_server_handle_to_fd(long long ptr ptr) > @ cdecl wine_server_release_fd(long long) > @ cdecl wine_server_send_fd(long) >+@ cdecl wine_server_exit_thread(long) > > # Codepages > @ cdecl __wine_init_codepages(ptr ptr ptr) > >Index: wine/dlls/kernel/thread.c >diff -u -p wine/dlls/kernel/thread.c:1.23 wine/dlls/kernel/thread.c:1.24 >--- wine/dlls/kernel/thread.c:1.23 Sun Jun 5 20:45:53 2005 >+++ wine/dlls/kernel/thread.c Sun Jun 5 20:45:53 2005 >@@ -24,7 +24,6 @@ > #include <assert.h> > #include <fcntl.h> > #include <stdarg.h> >-#include <signal.h> > #include <sys/types.h> > #ifdef HAVE_SYS_TIMES_H > #include <sys/times.h> >@@ -238,45 +237,8 @@ void WINAPI ExitThread( DWORD code ) /* > } > else > { >- struct wine_pthread_thread_info info; >- sigset_t block_set; >- ULONG size; >- > LdrShutdownThread(); >- RtlAcquirePebLock(); >- RemoveEntryList( &NtCurrentTeb()->TlsLinks ); >- RtlReleasePebLock(); >- >- info.stack_base = NtCurrentTeb()->DeallocationStack; >- info.teb_base = NtCurrentTeb(); >- info.teb_sel = wine_get_fs(); >- info.exit_status = code; >- >- size = 0; >- NtFreeVirtualMemory( GetCurrentProcess(), &info.stack_base, &size, MEM_RELEASE | MEM_SYSTEM ); >- info.stack_size = size; >- >- size = 0; >- NtFreeVirtualMemory( GetCurrentProcess(), &info.teb_base, &size, MEM_RELEASE | MEM_SYSTEM ); >- info.teb_size = size; >- >- /* block the async signals */ >- sigemptyset( &block_set ); >- sigaddset( &block_set, SIGALRM ); >- sigaddset( &block_set, SIGIO ); >- sigaddset( &block_set, SIGINT ); >- sigaddset( &block_set, SIGHUP ); >- sigaddset( &block_set, SIGUSR1 ); >- sigaddset( &block_set, SIGUSR2 ); >- sigaddset( &block_set, SIGTERM ); >- sigprocmask( SIG_BLOCK, &block_set, NULL ); >- >- close( NtCurrentTeb()->wait_fd[0] ); >- close( NtCurrentTeb()->wait_fd[1] ); >- close( NtCurrentTeb()->reply_fd ); >- close( NtCurrentTeb()->request_fd ); >- >- wine_pthread_exit_thread( &info ); >+ wine_server_exit_thread( code ); > } > } > > > > Couldn't wine_server_exit_thread be merged into LdrShutdownThread to avoid this Wine-specific export? -- Rob Shearman

2 1

Regression : Current CVS gives me a Segmentation Fault
by Paul Vriens June 5, 2005

June 5, 2005

Hi, the current cvs (2005/06/05 14:19:50) gives me a Segmentation Fault on a simple 'wine' command. Regression testing is not ended yet, but the problem wasn't there this morning (with Saturday's CVS). It appears to be http://cvs.winehq.org/patch.py?id=17949 but I'm not a 100% sure on that one yet. Any ideas, or does someone have the same problem ? Cheers, Paul.

3 2

Re: [d3d] Light management API fix to handle an unlimited number of lights
by Christian Costa June 5, 2005

June 5, 2005

Antoine Chavasse wrote: >Direct3D allows to set any number of light, and limits the number of >lights that can be active at the same time. > >The current wine code allows to setup only 16 lights. It made Anarchy >Online crash, because it tried to setup more lights than that and >didn't like to get returned an error when doing so. > >This patch is a rewrite of the d3d light management api and some related code. > >Another patch with test cases for the light api will follow. > >ChangeLog: > - Rewrote the light management API to allow for an unlimited amount >of lights to be set, > and only a subset of them to be enabled. > > > Hi Antoine, May I ask you to wait a little until I finish reorganizing the DDraw directory? This task involves file renaming so if your patch is aplly later, we can save the history of your modifications. :-) Bye, Christian

2 1

Re: [OLE #93] Disconnect IRemUnknown proxy after the others (resend)
by Robert Shearman June 5, 2005

June 5, 2005

Mike Hearn wrote: >On Sun, 05 Jun 2005 11:32:14 -0500, Robert Shearman wrote: > > >>The IRemUnknown proxy shouldn't be disconnected at all. Can you try this >>patch for me? >> >> > >Yes, that works too. This patch is simpler and equivalent though? > > > Yes it is. -- Rob Shearman

1 0