I don't think so.

The right fix may be int segments = ceilf(fabsf(thetaArc / ((float)M_PI_2 + 0.001f))) rather than ceilf(fabs(thetaArc / (M_PI_2 + 0.001f))). Because M_PI_2 is a double value and a double value will pass to the function ceilf in your PR.

Thank you!

@rhcad , thanks for your quick reply.

Sorry I didn't think much more then, my fix just takes the xcode's auto-fix suggestion, and the compiler seems happy with passing double value to ceilf, since you just want the integer part, so may be the precision doesn't matter here with ceilf?

Your suggestion is good for type consistency, but I don't think downcast the precision of M_PI_2 is a good choice, after all, precision does matter when it comes to fabs and fabsf, and the compiler did whining about it.

I think you are write. Thank you.