Mark Guzdial <guzdial at cc.gatech.edu> wrote...
>I've been spending some time studying the VM lately, and reading 
>papers from John Maloney's VM Workshop last OOPSLA.  The paper from 
>the Dolphin Smalltalk folk suggest that pushing contexts on the stack 
>may be faster than having each context have its own stack, which is 
>what I understand Squeak does.  Has anyone played with this in the 
>Squeak VM?

I'm sure this is true.  Ian wrote a context cache for an earlier Squeak and it sped things up by about 50% as I recall.  What we have now is nicely simple, and we've been holding off on any serious tweaks until we can play around with J3 (How's it going, Ian?).  J3 does even better things with the stack, which allows the basic interpreter to remain simple, yet offers a true high speed solution as well.

	- Dan