Re: still problem with semaphore

Hi Diego,

The only way I can contribute to this, is if I can get hold of Ubuntu or
Debian packages for the kernel you're using and the RTAI libraries, such
that I can just compile and run Orocos in an RTAI virtual machine, and then
confirm your issue. Just looking at the RTAI 3.9 API won't reveal a thing...

For clarity, the bug is not in Orocos/Thread.cpp itself but in how we
translate rtos_* functions to the RTAI API, and possibly,how RTAI was
configured. The most safe way is to turn OS_AGNOSTIC off, such that we use
the RTAI structs directly, and not our own copies of it...

Peter


On Wed, May 8, 2013 at 6:41 PM, <[email protected]> wrote:

> Dear Peter,
> sorry for bothering you again, but I am still fighting against orocos
> (lxrt).
> And, up to now, it wins. This is my last shot beofre giving up.
> I have installed it on several machines, but I keep having the same
> problem.
> By doing a more careful debug, I can see that (task-test) for example stops
> on rtos_sem_wait(&(task->sem)) of void *thread_function(void* t)
> in rtt/os/Thread.cpp (line 116 of orocos-toolchain-2.6). I am trying to
> understand the code to figure out who should raise the semaphore. But it is
> very complex. Please, can you help me to figure out wath's wrong.
>
> Thanks in advance. Best,
>
>
> Diego
> --
> Orocos-Dev mailing list
> [email protected]
> http://lists.mech.kuleuven.be/mailman/listinfo/orocos-dev
>

-- 
Orocos-Dev mailing list
[email protected]
http://lists.mech.kuleuven.be/mailman/listinfo/orocos-dev