From: Charles Oliver Nutter <headius@...>
Date: 2010-12-14T17:58:02+09:00
Subject: [ruby-core:33710] Re: Towards a standardized AST for Ruby code

On Sun, Dec 12, 2010 at 6:34 PM, Rocky Bernstein <rockyb@rubyforge.org> wrote:
> On ��the other hand, I do believe that in the rubinius implementation it is
> possible to have one interpreter/system which suits both competing needs of
> performance and live introspection. I believe this because in a sense
> rubinius *is* already two implementations: a Ruby interpreter plus a JIT for
> it.

JRuby has also been a mixed-mode implementation since early 2008.
Depending on what information you want available for debugging, it's
possible to provide it in interpreted mode; but as we discussed at
RubyKaigi, all optimizing compilers will eventually want to throw away
information about code that isn't necessary for execution, and then
you have to do debugging only with the interpreter (or with a compiler
that doesn't do those optimizations, which may be fine). I don't think
you currently want/need anything we throw away, but we need to be
mindful of that possibility.

- Charlie