> There's a more fundamental issue though. After spending time
> looking at the benefits of a JIT and thinking about the "yet
> another switch/goto implementation" conversation, I'm starting
> to think that a JIT will be almost useless for Parrot.
>
> JITs help when the VM is focused on lots of small instructions
> with well-known, static semantics. Perl's use of Parrot is going
> to be focused almost completely on PMC vtable ops. A JIT has
> no advantage over a threaded interpreter.
Sorry but I don't think this is true, great work has been done with
jits on highly dynamic language, this is actually where they shine
the most. You may be surprise to learn that the hotspot vim/jit from
sun comes from a smalltalk vm and before that from the self vm which
was extremely dynamic. You should do a web search for animorphic smalltalk
and self to find some of the publications (a lot were done on self).
Here is a pointer from one of the original member of the self group
(now working on java since the group was bought by sun)
http://groups.google.com/groups?q=animorphic&hl=en&rnum=1&selm=Pine.GSO.4.21
..0110211003390.13224-100000%40explorer.ce.chalmers.se
this newsgroup article has interesting links.
>
> About the only place where a JIT might really win big is in
> regexps.
>
> Have other people come to the same conclusion?
>
> Is there any interest in a less dynamic dialect of Perl that can
> take advantage of a JIT? Should we feed a request to p6-language
> to think about this?
>
> - Ken
