that said, please be *conservative* on how  you optimize floats, i'd rather not have accidental unexpected numerical instability.  Its much easier to validate correctness of optimizations for Words and Ints rather than Floats and Doubles, the latter are complicated critters


On Fri, Jul 19, 2013 at 1:25 PM, Bryan O'Sullivan <bos@serpentine.com> wrote:

On Fri, Jul 19, 2013 at 9:58 AM, Jan Stolarek <jan.stolarek@p.lodz.pl> wrote:
It seems that currently there are no built-in constant folding rules for Integer div and mod. I plan on adding those rules, but before I do that I wanted to ask whether there is a good reason that these rules don't exist?

I would guess that Integer arithmetic doesn't affect people's performance much, because a lot of people steer away from Integer. I'd be focusing my attention on fixed-width numeric types and floating point, as likely to have a much larger real-world impact.

_______________________________________________
ghc-devs mailing list
ghc-devs@haskell.org
http://www.haskell.org/mailman/listinfo/ghc-devs