[Mono-list] Mono contributions

[Peter Drayton]

+1. I'm with Nelson on this - interpreters have all sorts of nice
characteristics IMO, in that they are easier to write, easier to read,
and easier to port. I think these are important characteristics for a
bootstrap like this, where you want the core system to materialize and
stabilize fast enough to retain community interest, and to also be
approachable.

IMO a well-designed VES should be able to combine JITted and interpreted
code, too, and change the mix over time (ie. reJIT code if access
patterns change, or pitch JITted functions and go back to interpreting
if memory is tight). 

Are these currently goals for Mono? Is there a list somewhere of the
design goals for the Mono VES?

--Peter
http://staff.develop.com/peterd

> -----Original Message-----
> From: mono-list-admin@ximian.com [mailto:mono-list-admin@ximian.com]
On
> Behalf Of Nelson Minar
> Sent: Monday, July 09, 2001 4:52 PM
> To: mono-list@ximian.com
> Subject: Re: [Mono-list] Mono contributions
> 
> Mono is the most exciting thing I've seen all year! Thanks, Ximian!
> 
> >	* We are trying to build a few things before we do the JIT
> >		* An interpreter
> 
> I'd love to have an interpreter - JITs are often superfluous, at least
> for the JVM. If folks are interested in this topic, this article is
> interesting:
> 
>
http://www2.fit.qut.edu.au/CompSci/PLAS//ComponentPascal/VirtualMachines
.p
> s
> 
> It makes the case that the details of typing in the CLR make an
> interpreter much more difficult than in the JVM. Also has some
> suggestions about how to work around it.
> 
>                                                      nelson@monkey.org
> .       .      .     .    .   .  . . http://www.media.mit.edu/~nelson/
> 
> _______________________________________________
> Mono-list maillist  -  Mono-list@ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-list