[MLton-devel] Minimizing space usage in ML (fwd)

[Stephen Weeks]

> If Matthew is right, then it would seem his conservative approach wouldn't
> be bad.  Hm, I guess you would need to count on the fact that things in this
> semi-mutable mode would be relatively directly accessible from the stack, and
> to not consider things like that immutable regardless of the type. 

I'm not following.  There is no semi-mutable mode.  Arrays are
mutable, vectors aren't.  We have a magic cast in the basis library
that lets us promise to the compiler/runtime that we won't modify an
array any more.  The runtime never has make a conservative guess.  Why
does being accessible by the stack matter?

> Maybe having fromArray set the bit isn't too bad given that a write
> happened recently so it is probably in the L1 cache.

I want to point out that with the new header word approach, we're not
actually setting a bit, we're changing the header word from one index
to another.


-------------------------------------------------------
This sf.net emial is sponsored by: Influence the future 
of Java(TM) technology. Join the Java Community 
Process(SM) (JCP(SM)) program now. 
http://ad.doubleclick.net/clk;4699841;7576301;v?http://www.sun.com/javavote
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel