[MLton-devel] project

[Stephen Weeks]

> I was also thinking of him working on the front-end to have MLton
> generate reasonable (type) error messages on the ast, but I'm not
> sure whether this would be doable in a semester working part-time.

I did a some work a while back to at least get reasonable source
positions.  The work that needs to get done now is a complete front
end rewrite so that type checking occurs before defunctorization and
follows the standard.  That is way too big for 50-60 hours (of my time
evene).

For 50-60 hours, and for undergraduate research, I would think that
any of the following would be fun.  

	Port CML to MLton
	Performance -- use x86 instrumentation to evaluate cache behavior
	SSA optimization -- eliminate redundant derefs

Porting CML might be especially good because he can read Reppy's CML
book, which is a good thing to read anyways.  The drawback is that it
may not involve much with compiler innards.  If he wants that, go for
the SSA optimization, preceeded by the writeVec improvement, which
would give him a good intro to looking at SSA and the optimizer.


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel