next release

[Stephen Weeks]

OK, now that the paper is done, I'd like to get back to getting out the
ever-delayed next release.

As I see it, here's what needs doing:

1. Fix bug in MLton.size.  The useless CPS simplification pass sometimes
   seems to remove the argument, making the size appear to be zero.

2. benchmarking.  I've got a script that sort of works in src/benchmark, but it 
   needs work.

3. fix mlprof to work with native backend

4. documentation

5. integrate new gmp

I'm up for 1, 2, and 4.  And Henry will do 5 any day now (or at least will
emulate it :-).  Matthew, would you be interested in 4?  My guess is that it's
less than a days work.  It just needs some new regular expressions and some
slightly different code for the correlation between the gmon.out data and the
profiling labels.

Also, Suresh, I'd be interested in hearing what you're up to these days.  Are
you still at NEC?  Did anything ever come of the stuff that you and Cristian
were doing with MLton?  Did you make any progress on writing or on the list I
sent?