[MLton-devel] cvs commit: Real32 and Real64 improvements

[Stephen Weeks]

> Bizzare.  I assume that -ffast-math is just using the instructions and I
> thought that they did the `right' thing.

Yeah.  I have no idea how it helps gcc speedwise on x86 to violate
IEEE on arithmetic and comparisons.

> Were we compiling with -ffast-math before? 

No.

> Also does it explain any of the BSD differences since the info pages
> seemed to indicate that there was some difference in the default
> settings for gcc on that OS.

I would guess it doesn't since we didn't use it.



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel