# x cLog2 cExp -> x (when not known whether x may be negative)

T=d f
V=x
R=0, 3, 0.5
F=exp(log2(x))
C=fp_exp(fp_log2(x))
