[MLton-devel] project

[Matthew Fluet]

Here are some more that I've had on my mind:

SSA optimization -- partial redundancy elimination; or at least some form
                    of loop invariant motion.

[PRE isn't as easy as I'd hoped.  There is a discrepency between our
notion of SSA and other notions; in particular, our notion that every
variable is unique makes it more difficult to find common expressions than
it is with versioned variables.  Of course, I don't really understand the
utility of an optimization that fails miserably if I alpha-vary variables,
but that's getting a little OT.]

SSA optimization -- in the removeUnused pass, a datatype with a single
                    variant can be represented by a tuple; don't know if
                    this happens much in practice, but I seem to recall
                    seeing before

XML optimization -- lambda lifting before polyvariance

There is probably a lot of room to play with the closure conversion
representations.  The specific problem I have is with the polyvariance
pass that results in alpha-equivalent functions in benchmarks/logic.sml.





-------------------------------------------------------
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