Computes the maximums of the four single-precision, floating-point values of a and b (see _mm_max_ps)
Definition at line 214 of file AkSimd.h.