Parrot on RTEMS

Stefan O'Rear stefanor at cox.net
Fri Apr 16 22:34:43 UTC 2010


On Fri, Apr 16, 2010 at 03:29:40PM -0700, Jonathan Leto wrote:
> Howdy,
> 
> > I'm not sure I follow here.  The Parrot VM is not designed for real time
> > operation, and it cannot provide timing guarantees for several reasons
> > (most importantly, our stop-the-world garbage collector).  Without
> > multitasking, Parrot cannot be used to control other systems that do
> > depend on real time.  So, what role is left for Parrot to fill?
> 
> Firstly, if the process is short-lived, disabling the garbage
> collector via -G gets us around the issue of not having a real-time
> GC. Also, many people have shown interest in getting a real-time GC in
> Parrot, and having Parrot on RTEMS gives us a place to develop and
> test it.
> 
> There is also the fact that everything running on real-time hardware
> doesn't necessarily have to be "real-time" at the same time scale.

How can a "single process operating system" have short-lived processes?

-Stefan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
URL: <http://lists.parrot.org/pipermail/parrot-dev/attachments/20100416/7884fe58/attachment.pgp>


More information about the parrot-dev mailing list