Re: [avr-libc-dev] Bug in math library

[Colin O'Flynn]

> Sorry for the diversion, just meant to field a thought that may both
> simplify float routine implementations and use, although "non-standard".

Perhaps this is something that we could just use #define's to work with? It 
should be trivial to have a +inf put in place of a NaN, then there is the 
option of compiling with either NaN generated or +inf.

Of course that is just more options, and it would be likely that it would be 
never changed from the default value...

The only other problem: avr-libc doesn't support -0 I think (going from source 
code comments), which would mean there could be no -inf option. The result of 
+/- inf for a zero would be a logical option in some cases, but the result of 
always +inf seems fairly useless to me.

Regards,

 -Colin