Re: wine/ server/trace.c server/thread.h server/th ...

2003-03-15 Thread Rein Klazes 
On 14 Mar 2003 12:33:30 -0800, you wrote:

> > | err:seh:EXC_DefaultHandling Exception frame is not in stack limits => unable to 
> > dispatch exception.
> > 
> > One thread stays alive, but the half-constructed main window is frozen.
> 
> Does this help?
> 
> Index: scheduler/sysdeps.c

Yes thats it. Thanks,

Rein.
-- 
Rein Klazes
[EMAIL PROTECTED]

Re: wine/ server/trace.c server/thread.h server/th ...

2003-03-14 Thread Alexandre Julliard 
Rein Klazes <[EMAIL PROTECTED]> writes:

> | trace:seh:EXC_RtlRaiseException code=c005 flags=0 addr=0x41f32ea1
> | trace:seh:EXC_RtlRaiseException  info[0]=0001
> | trace:seh:EXC_RtlRaiseException  info[1]=
> | trace:seh:EXC_CallHandler calling handler at 0x43fcf5 code=c005 flags=0
> | trace:seh:EXC_RtlUnwind code=c005 flags=2
> | trace:seh:EXC_RtlRaiseException code=c028 flags=1 addr=(nil)
> | err:seh:EXC_DefaultHandling Exception frame is not in stack limits => unable to 
> dispatch exception.
> 
> One thread stays alive, but the half-constructed main window is frozen.

Does this help?

Index: scheduler/sysdeps.c
===
RCS file: /opt/cvs-commit/wine/scheduler/sysdeps.c,v
retrieving revision 1.54
diff -u -r1.54 sysdeps.c
--- scheduler/sysdeps.c 27 Feb 2003 21:11:13 -  1.54
+++ scheduler/sysdeps.c 14 Mar 2003 20:30:57 -
@@ -162,8 +162,8 @@
 static void SYSDEPS_StartThread( TEB *teb )
 {
 SYSDEPS_SetCurThread( teb );
-CLIENT_InitThread();
 SIGNAL_Init();
+CLIENT_InitThread();
 __TRY
 {
 teb->startup();

-- 
Alexandre Julliard
[EMAIL PROTECTED]

Re: wine/ server/trace.c server/thread.h server/th ...

2003-03-14 Thread Rein Klazes 
On Tue, 4 Mar 2003 16:18:44 -0600, you wrote:

> ChangeSet ID: 7374
> CVSROOT:  /opt/cvs-commit
> Module name:  wine
> Changes by:   [EMAIL PROTECTED]   2003/03/04 16:18:44
> 
> Modified files:
>   server : trace.c thread.h thread.c ptrace.c process.c 
>context_sparc.c context_powerpc.c 
>context_i386.c 
>   scheduler  : process.c client.c 
>   include/wine   : server_protocol.h 
>   dlls/ntdll : signal_sparc.c signal_powerpc.c signal_i386.c 
> 
> Log message:
>   Use SIGUSR1 instead of SIGSTOP to suspend threads.

This patch causes a regression running thebat! mail client.

Running with --debugmsg +seh gives:

| trace:seh:EXC_RtlRaiseException code=c005 flags=0 addr=0x41f32ea1
| trace:seh:EXC_RtlRaiseException  info[0]=0001
| trace:seh:EXC_RtlRaiseException  info[1]=
| trace:seh:EXC_CallHandler calling handler at 0x43fcf5 code=c005 flags=0
| trace:seh:EXC_RtlUnwind code=c005 flags=2
| trace:seh:EXC_RtlRaiseException code=c028 flags=1 addr=(nil)
| err:seh:EXC_DefaultHandling Exception frame is not in stack limits => unable to 
dispatch exception.

One thread stays alive, but the half-constructed main window is frozen.

If I run with --debugmsg +relay there is no crash and the program
manages to display the main windows after 20 miutes.

Starting form winedbg doesn't seem to work, a first chance exception
occurs soon after startup (long before the one above) and:

| Wine-dbg>wine client error:b: read: Bad address
| wine client error:b: sendmsg: Bad file descriptor

I don't know how to proceed from here.

Rein.
-- 
Rein Klazes
[EMAIL PROTECTED]