added patch for ACE/TAO 5.1.14 beta release

Joel Sherrill joel.sherrill at OARcorp.com
Thu Mar 29 20:11:31 UTC 2001



Rosimildo daSilva wrote:
> 
> From: Joel Sherrill <joel.sherrill at OARcorp.com>
> To: Rosimildo da Silva <rdasilva at connecttel.com>
> Cc: rtems-snapshots at OARcorp.com <rtems-snapshots at OARcorp.com>;
> rtems-users at OARcorp.com <rtems-users at OARcorp.com>
> Date: Thursday, March 29, 2001 9:38 AM
> Subject: Re: added patch for ACE/TAO 5.1.14 beta release
> 
> >It compiles cleanly.  I did a loopback network configuration
> >to make everything happy (and target independent) so the
> >tests would link.
> >
> >I think that now if someone just tried to run the ACE tests,
> >that would move things along.
> >
> >TAO built but there is no point in trying it until we pass
> >the ACE tests.
> >
> >> For C++ people that want to build their applications
> >> on top of a wonderful framework, ACE is one of the
> >> best ones that I ever seen. ACE has nothing to do
> >> with CORBA. It is a multi-platform framwork for
> >> threads/semaphores/sockets/timers, and a zillion
> >> design patterns implemented in a nice OO way.
> >
> >There are a few free things built on top of ACE.
> >TAO is one, JAWS (adaptive web server) is another.
> >
> >> TAO is one of the best designed CORBA ORBs.
> >
> >I have been reading the paper on its pluggage protocols.
> >That part looks real nice.  They are working on the size
> >issue.
> >
> >> Joel,
> >>
> >> I presume you build them as separated libraries,
> >> so people that want use ACE only, would be able
> >> to do so.
> >
> >I built it exactly as it is delivered with minimal
> >RTEMS changes to the Makefiles.  Basically we
> >use a special RTEMS Makefile to ask the name of
> >the C and C++ compilers along with arguments.
> >This makes the RTEMS configurery in ACE
> >BSP independent.  This does not appear to
> >be the case for other embedded RTOS ports.
> >look at "ace/*psos*" and "include/makeinclude/*psos*"
> >for an example.  The *vxw* files are not as
> >numerous but have CPU dependent conditional in them.
> 
> Excellent. So, I presume that the requirements to test this
> *enough memory*, a network aware BSP, and the POSIX api. ?

That sounds correct.  I **suspect** that many of the ACE
tests can be run using the loopback network device.  So
technically you can probably run them on a simulator
although I did not try.  I probably should though.
 
> Rosimildo.

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel at OARcorp.com                 On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985



More information about the users mailing list