[BC] Block Closures, Version 2
Tim Rowledge
tim at sumeru.stanford.edu
Thu Feb 14 19:12:20 UTC 2002
Scott A Crosby <crosby at qwes.math.cmu.edu> is claimed by the authorities to have written:
> > How does one make a "cheaper" stack?
>
> Cheaper bytecode main loop?
Separate issue, surely. It's hard to see much direct intersection
between the bytecode loop and the stack.
>
> Maybe a shared stack, for both locals and temps, rather than several
> stacks/vectors for locals, args, and temporaries. This would likely help
> with the CPU caches. (I'm not sure what squeak does, nor how your version
> changes this. I'm just elaborating what I meant.)
locals _are_ temps in contexts. So are arguments. They all go together
at the top of the context's stack. Take a look at the method activation
code for precise details, or peek at the diagrams etc in my chapter in
the nuBlu book. I guess I'll have to update that a bit when we've
finished the changes for BCs.
> Its good news for me, but bad news for squeak work.
congratulations; I don't know what rice is like, but UIUC does have
longtime Smalltalk connections.
tim
--
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Strange OpCodes: IPL: Invent Program Lines
More information about the Squeak-dev
mailing list
|