[Bsd-sharp-list] pthread problems

[Robert Wojciechowski]

Two quick comments:

 1) Have you tried running the application a few dozens times to see if it occurs after repeated executions?

 2) Is SCHED_ULE stable in 6.2? I can switch to SCHED_ULE on a development box here, but I'm hesitant to switch to it on my production servers after hearing it was not very stable from messages such as:

  http://lists.freebsd.org/pipermail/freebsd-questions/2006-November/136433.html

-- Robert

> -----Original Message-----
> From: Phillip Neumann [mailto:pneumann at gmail.com]
> Sent: Saturday, January 20, 2007 7:58 PM
> To: Robert Wojciechowski
> Cc: bsd-sharp-list at forge.novell.com
> Subject: RE: [Bsd-sharp-list] pthread problems
> 
> Hello robert.
> 
> I just test mono 1.2.2.1 with my AMD64 X2 mashine running 6.2-STABLE
> (RELENG_6)
> 
> as you see it works ok:
> 
> [killfill at worm ~]$ cat prueba.cs
>  using System;
> 
>   class EntryPoint {
>     static void Main(string[] args) {
>       Console.WriteLine("hello world");
>     }
>   }
> [killfill at worm ~]$ mcs prueba.cs && mono prueba.exe
> hello world
> [killfill at worm ~]$ sysctl hw.ncpu
> hw.ncpu: 2
> 
> 
> This is a GENERIC Kernel with SCHED_ULE and SMP enabled.
> make.conf file contains this:
> CFLAGS=    -O2 -pipe
> 
> 
> 
> Cannot reproduce your problem.
> Maybe you could try SCHED_ULE to see if its that.
> 
> good luck!
> 
> 
> 
> El vie, 29-12-2006 a las 13:44 -0500, Robert Wojciechowski escribió:
> > I can exhibit the problem by writing a simple program such as:
> >
> >   using System;
> >
> >   class EntryPoint {
> >     static void Main(string[] args) {
> >       Console.WriteLine("hello world");
> >     }
> >   }
> >
> > and compiling with mcs/gmcs like so to get the error:
> >
> > [dev2:~]> mcs test.cs
> > Thread 81e7e00 has exited with leftover thread-specific data after 4
> > destructor iterations
> >
> > Of course I get the error when I execute it, too:
> >
> > [dev2:~]> mono test.exe
> > hello world
> > Thread 81e7e00 has exited with leftover thread-specific data after 4
> > destructor iterations
> >