[MLton] Few problems with porting...

[Matthew Fluet]

> I have encountered a few problems when trying to compile
> a new native version of MLton from the latest CVS
> with and without my own current MinGW changes.

Yes, recent changes tickled a long-standing SSA simplifier bug, which is
the error you are seeing.  Steve described the problem in a little more
detail in a recent commit:
http://www.mlton.org/commitlog

The following seems to work:

cvs update -D "2003-11-15 00:00:00"
## This updates the CVS sources to a point after the bug fix to the
   simplifier, but before the checking that tickles the bug
make all
## Compile this version of MLton; should work with whatever
   working compiler is around
cvs update -A
## This updates the CVS sources to the current point (removing
   all the sticky dates set by the previous update)
make all
## Compile this version of MLton; this requires the previous
   built MLton in order to get around the bug

> Also another issue I have: I'm using SML/NJ 110.0.7 - and as far as I'm
> aware (I could be wrong though?) this is the latest "stable" version.
> So the problem I have is that I can't use bin/check-basis
> Is there a known way to use this with SML/NJ 110.0.7?

We've been using the working versions of SML/NJ for a long time.  The
current CVS MLton sources work with SML/NJ 110.43; the changes to work
with SML/NJ 110.44 are trivial.

Hopefully, we will very soon transition to a point were we don't need
SML/NJ to check the basis.  The new MLton type-checking front-end has
already uncovered a few subtle errors in the basis implementation.

> At least CM.Control does not exist in 110.0.7. I found CM.Control.verbose
> as CM.verbose - but it takes a bool option instead of just a bool.
> I'm not even sure if it's supposed to be the same function.

I assume those are the same.  SML/NJ has moved a lot of the manipulation
of options around between 110.0.7 and 110.44.