intraprocedural flattening

Stephen Weeks MLton@sourcelight.com
Fri, 20 Jul 2001 15:43:52 -0700


> Excellent.  The changes are larger than I would have expected, but I'm sure
> that that is just because of some random hot loop that used to be allocating.

I'll say.  On psdes-random, it cuts the allocation from 1,500,000,752 to 776.