June 2003 - Wine-devel - list.winehq.org

The Matrix DVD installer hangs
by Gerhard W. Gruber 14 Jun '03

14 Jun '03

Some days ago there was a patch for fixing installer hangs. I just tried to install from The Matrix DVD and the installer starts, but hangs after it a short while running. -- Gerhard Gruber Für jedes menschliche Problem gibt es immer eine einfache Lösung: Klar, einleuchtend und falsch. (Henry Louis Mencken)

1 0

CoSetState
by Mike Hearn 14 Jun '03

14 Jun '03

Can anybody enlighten me as to what this is supposed to do? It appears to be undocumented, or at least if it's on MSDN it's not where it should be and Google/MSDN Search can't find it. In particular, if you run RhymBox with native oleaut32,rpcrt4 but builtin ole32, shortly after calling this function it appears to crash. If I remove the Release() call, it no longer crashes but seems to freeze the chat window. Looking at the source, all it seems to do is store the passed IUnknown in a global var, with a note that it should be thread local. That doesn't seem to be a very useful function to have, so it must do more than that. Anybody? I think I'm going to try and make it thread local.... thanks -mike

1 0

Re: DEVENUM.DLL Implementation
by Robert Shearman 14 Jun '03

14 Jun '03

> Date: Mon, 9 Jun 2003 23:59:01 +0200 > From: Lionel Ulmer <lionel.ulmer(a)free.fr> > To: wine-devel(a)winehq.com > Subject: Re: DEVENUM.DLL Implementation > > Alexandre, Robert, > > Do you have any idea why this patch was never comitted to Wine ? It's probably best to ask Alexandre about this, but it either slipped through the cracks (and I didn't want to submit it again until I was completely happy with it) or it was due to depending on other unimplemented behaviour (IFilterMapper interface). > Maybe this > will prevent me to do 'quartz,devenum=n' and only 'quartz=n' to get movies > in some applications :-) It should work alright if you can get the patch to apply properly, but IIRC a few changes conflict with it. I still have a few things I'd like to improve (like using DirectSound functions to enumerate the DirectSound devices available) and I will submit another patch when this is done in a couple of weeks. As for the future, work on Quartz shouldn't be too hard to do. A minimal working system would be a File Source 'codec', video and audio renderers and the graph builder, along with my device enumerator implementation. Rob

5 6

Re: Process termination & exit code
by Alexandre Julliard 14 Jun '03

14 Jun '03

Eric Pouech <pouech-eric(a)wanadoo.fr> writes: > as reported by Igor Sysoev on WD, exit process code wasn't properly > implemented on *BSD > it appeared (thanks Igor for all the testings and traces) that after > terminate_process request was issued, the client/server pipe got > closed because the running thread was actually killed, erasing the > exit_code that terminate_process had set. > This patch actually fixes it. I don't think we need a new thread state just for that; how about something like this? Index: server/process.c =================================================================== RCS file: /home/winehq/opt/cvs-commit/wine/server/process.c,v retrieving revision 1.104 diff -u -p -r1.104 process.c --- server/process.c 18 Mar 2003 05:04:33 -0000 1.104 +++ server/process.c 13 Jun 2003 23:18:03 -0000 @@ -688,7 +688,7 @@ void kill_process( struct process *proce while (thread) { struct thread *next = thread->proc_next; - thread->exit_code = exit_code; + if (exit_code) thread->exit_code = exit_code; if (thread != skip) kill_thread( thread, 1 ); thread = next; } Index: server/request.c =================================================================== RCS file: /home/winehq/opt/cvs-commit/wine/server/request.c,v retrieving revision 1.79 diff -u -p -r1.79 request.c --- server/request.c 28 Mar 2003 00:36:12 -0000 1.79 +++ server/request.c 13 Jun 2003 23:18:04 -0000 @@ -382,7 +382,7 @@ int receive_fd( struct process *process if (ret > 0) fprintf( stderr, "Protocol error: process %p: partial recvmsg %d for fd\n", process, ret ); - kill_process( process, NULL, 1 ); + kill_process( process, NULL, (ret > 0) ); } else { -- Alexandre Julliard julliard(a)winehq.com

3 3

NPTL questions again
by Mike Hearn 14 Jun '03

14 Jun '03

Hi, If I want to use a standard kernel from kernel.org, ie not the RH9 stock one, on a RH9 installation, is there anything I need to be aware of? Everything else seems fine, except that I get wierd threading lockups when running Wine. Going back to the original kernel fixes them. Is there any way to rid my RH9 install of NPTL completely? The LD_ASSUME_KERNEL trick doesn't work, and I'd like to stick with this kernel I have now but I need to be able to run Wine. Alternatively if anybody knows of a patch to add NPTL to a standard 2.4.20/21 kernel, then please let me know. thanks -mike

1 0

Re: DSound buffer underrun problems with X-Plane
by Christian Neumair 14 Jun '03

14 Jun '03

Am Sam, 2003-06-14 um 07.54 schrieb Jerry Jenkins: > Which sound driver did your specify in your wine config file -- > "~/.wine/config" or something similiar? > If it's winearts, maybe > http://www.winehq.com/hypermail/wine-patches/2003/03/0316.html > is what you need. When I made the backtrace (see below) I used wineoss. Now that you made me remember that there are other sound drivers, too, I tried winealsa - and voila - it works. At least these buffer uderruns occur way more sparse. It's a first success, thanks! > Otherwise could you do a trace for me? I don't think my 56k modem is > capable of downloading the program your mentioned :-(. Run your program as: > wine --debugmsg +wave,+dsound YourProgram.exe >& trace > where YourProgram.exe the file name of X-Plane and email the file directly > to me. The trace way done one minute or so, it just logged the messages during the startup splash screen where you hear engine sound - so it should be an ideal test candidate. I uploaded the trace under [1]. But be warned, it's pretty huge - 350 KB bzip2'ped, ~17 MB bunzip2'ed. regs, Chris [1] http://gnome-de.org/~manny/wine.xplane.trace.bz2

1 0

couple of questions about pthreads (scheduler/pthread.c) in wine
by Juraj Hercek 14 Jun '03

14 Jun '03

Hi, I've ported an application which uses a library - let's call it aceutilities.so - which in turn uses ACE library (http://www.cs.wustl.edu/~schmidt/ACE.html) ACE uses libpthread.so library. So the dependency is like this: my_application -> aceutilities.so -indirectly via ace library-> libpthread.so Wine's scheduler/pthread.c (compiled code resides in ntdll.dll.so) defines couple of symbols, which are defined (if we don't take symbol versions into consideration) also in libpthread.so. Now you probably feel what's my problem: After loading application into memory, dynamic linker loads wine stuff first, then when loading aceutilities.so into memory, pthread symbols are defined in module ntdll.dll.so - already loaded into memory - and so it doesn't try to load libpthread.so, because pthread symbols are already resolved. My questions are: 1) Why Winelib uses it's own implementation of pthreads (at least it looks like this is intention)? 2) Is it possible to force linker to use symbols from libpthread.so instead of symbols from ntdll.dll.so? (I don't know how to tell linker to use symbol, e.g. pthread_cond_wait, from specified library, i.e. /lib/libpthread.so.0). Thanks for help, Juraj

4 5

Re: wineconsole init
by PETREOLLE Sylvain 14 Jun '03

14 Jun '03

Could we have settings that avoid the error message about ncurses ? To have working non-buffered keyboard input in dos programs, I must disable curses support.

2 1

new compile problem - curses.c:790: conflicting types for `WCCURSES_InitBackend'
by hatky 14 Jun '03

14 Jun '03

I was updating the cvs after the last commit rain from alexander and it does not compile to be sure I removed the tree and checked it out again and I still get the error: gcc -c -I. -I. -I../../include -I../../include -D_REENTRANT -fPIC -DUNICODE -Wall -mpreferred-stack-boundary=2 -gstabs+ -Wpointer-arith -g -O2 -o curses.o curses.c curses.c:790: conflicting types for `WCCURSES_InitBackend' winecon_private.h:96: previous declaration of `WCCURSES_InitBackend' make[2]: *** [curses.o] Error 1 make[2]: Leaving directory `/home/hatky/wine-source/wine/programs/wineconsole' make[1]: *** [wineconsole] Error 2 make[1]: Leaving directory `/home/hatky/wine-source/wine/programs' make: *** [programs] Error 2 I did not notice anyone else asking about this so I wonder it this is some version of a package I have.... I have mandrake9.1 with the latest updates... It is a completly clean tree (just checked out fully), Hatky. __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com

1 0

Re: advapi32: added Ace/Acl functions
by PETREOLLE Sylvain 13 Jun '03

13 Jun '03

Could you resubmit your patch using cvs diff -u ? This would be great.

1 0