[MLton-devel] icfp

[Suresh Jagannathan]

Right, just in the compiler.   Shao had a paper in ICFP'98 where
he discusses the cost of maintaining types in Flint, and the various
tricks employed to minimize type representation overheads.  By
running the profiler on a self-compile, I was hoping we'd be able
make a similar argument that allocation and time overhead in translating
to a simply-typed IL is relatively low.



On Monday 24 February 2003 06:00 pm, Henry Cejtin wrote:
> I'm confused.  When you talk about `manipulating types', you are just
> talking about in the compiler right?  No other ML  programs  manipulate 
> types  since programs only manipulate values.
>
> What I was talking about was the cost of regular ML programs.
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> MLton-devel mailing list
> MLton-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mlton-devel


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel