Optimizer is putting result of a log into the wrong reg

[Stephen L Moshier]

On Thu, 23 Sep 1999, Singh, Kirat wrote:
> Oops, sorry about that. I am now passing doubles where doubles are expected
> (I also simplified it a bit). But it *still* puts a NaN into the variable
> DiscRate whereas it clearly should be 0.

Yes, sorry, I tend to think of K&R C rules, where float would promote to
double but not int.