On 06/12/2013 12:51 PM, Leopold Palomo-Avellaneda wrote:

> Hi,
> 
> I have reviewed the code, the compiler options and I have arrived to the 
> conclusion that I was wrong and probably it was nothing about compiler 
> options.
> 
> However, more deeply doubts have come and I would to ask to the clever people 
> here if someone could light me. It's curious because there's another person 
> in 
> the orocos list that has similar problems, and probably our problems came 
> from 
> the same place.
> 
> The soem library, use some pthread to not block functions to send and receive 
> packets. I think that I'm having problems here. I can setup a socket, but I 
> cannot send any packet message in a simple program.
> 
> I got the idea of putting a:
>    struct sched_param param = { .sched_priority = 1 };
>    pthread_setschedparam(pthread_self(), SCHED_FIFO, &param);
> 
> in the send function, before rt_dev_send.
> 
> The raw-ethernet example use it, but if I put the same parameters I got a 
> segmentation fault, The pthread library didn't like the 
> pthread_setschedparam. 
> I have reviewed and compiled by hand linking against pthread_rt , but the 
> same.
> 
> Someone have some idea to solve this?


Are you doing this from the main thread? If yes, what is the current
stack size limit for this process? If the process is started directly
from shell, ulimit -s should give the current stack size limit.


-- 
                                                                Gilles.

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
RTnet-users mailing list
RTnet-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rtnet-users

Reply via email to