Re: [AD] float math function inconsistencies

[Nick Trout <chinbillybilbo@xxxxxxxxxx>]

Any thoughts on this?

On 26 January 2013 20:11, Nick Trout <chinbillybilbo@xxxxxxxxxx> wrote:

I noticed whilst going through the code that float math functions are sometimes used inconsistently, e.g. fabs instead of fabsf where <math.h> declares:

double fabs(double);

float fabsf(float);

The compiler should promote the argument to a double. This might not be inefficient on some platforms, but it is inconsistent, and could inefficient on some lesser platforms.

Cheers, Nick