[Vm-dev] Cog and Exupery

[bryce at kampjes.demon.co.uk]

Eliot Miranda writes:
 >  On Wed, Mar 4, 2009 at 4:29 PM, Igor Stasenko <siguctua at gmail.com> wrote:
 > 
 > I don't think it's unrealistic at all.  It's my aim too!  There are other
 > Smaltalk VMs out there that are faster than VW.  I just think Bryce is
 > making some strange decisions (not eliminating contexts early on) and I want
 > to try and understand why he is going a different way from me.  I might be
 > very wrong in my approach.  On the other hand, Bryce might be going the
 > wrong way, and my questions might help.

What are you asking here?
 1) Why I haven't implemented any serious context optimisation yet?
 2) Why I haven't implemented a context cache/stack?

The answer to 1) is debugging a compiler and adaptive inlining at
the same time seemed needlessly difficult so I've been working
on getting the compiler ready and reliable first. Up until now
the goal was to get everything needed to benefit from adaptive
inlining working except adaptive inlining.

The answer to 2) is adaptive inlining is the main send optimisation
planned. It's needed to expose the entire loops to allow further
optimisation. Once adaptive optimisation is in place and working then
other major send optimisations may be considered based on real
measurements

Bryce