[Cfp-interest 3187] Re: Suggested DRAFT Modications to Annex F + G of C26 document

[Damian McGuckin]

I know it is historical reasons, but why is the prototype of frexp()

 	frexp(double x, int *p)

but that of modf() remains

 	modf(double x, double *iptr)

Why not

 	modf(double x, double *p)

Both of these predate IEEE 754 (although probably not before floating 
point standardization was an idea in people's heads in the early 1970s).

Thanks - Damian