A Divendres, 3 de maig de 2013, George Pontis va escriure:
> > -----Original Message-----
> > From: Leopold Palomo-Avellaneda [mailto:[email protected]]
> > Sent: Thursday, May 02, 2013 1:37 AM
> > To: [email protected]
> > Cc: George Pontis
> > Subject: Re: [Xenomai] Qt - Xenomai compatibility problem
> > 
> > A Dimarts, 30 d'abril de 2013, George Pontis va escriure:
> > > I took a snapshot from git of the current Xenomai code and put it in the
> > hands of the software
> > > developers. They reported that Qt would not build after patching with
> > Xenomai. The problem was that
> > > Qt uses "signals" as a keyword, and that conflicts with the function
> > prototype for rt_task_notify
> > > under xenomai/native:
> > >
> > >
> > > int rt_task_notify(RT_TASK *task, rt_sigset_t signals);
> > >
> > >
> > > Although this is not a Xenomai problem, it might save some users the
> > annoyance if the name "signals"
> > > could change.
> > 
> > I'm not a C++ expert, but to solve this you must encapsulate all the 
Xenomai
> > calls of your software in a namespace. After that, I think that you could 
use
> > the signal word without problem of collision with another signal word.
> > 
> > Regards,
> > 
> > Leopold
> 
> That that won't work because Qt defines the keyword "signals" using #define,
> which is handled by the preprocessor.  The preprocessor ignores namespaces
> so namespaces wouldn't help in this case.
> 
oops,

thanks for the info. I didn't know that. I could only say that it could not be 
a good idea to make a Qt program with realtime, IMHO. Quoting Jan Kiszka from 
another mail in the list, talking about a program written in Java:

>Writing some non-RT configuration and monitoring front-end in Java is
>fine, but the actual control task is better done in C (or C++ if you
>like). The RT process containing the control task could then have a
>another thread, low priority, that does the non-RT communication with
>the front-end via TCP and transfers any parameter updates to the RT task
>(lock-free or synchronized with a compatible locking mechanism). That's
>the standard design pattern for real-time control applications.

in my opinion this could be applied to a Qt program.

Otherwise, good info about the namespaces and preprocessors.

Regards,

Leopold
-- 
--
Linux User 152692
Catalonia

_______________________________________________
Xenomai mailing list
[email protected]
http://www.xenomai.org/mailman/listinfo/xenomai

Reply via email to