https://www-01.ibm.com/support/knowledgecenter/ssw_aix_61/com.ibm.aix.alangref/idalangref_fsel_fp_instrs.htm
Quote:
The double-precision floating-point operand in floating-point register (FPR) FRA is compared with the value zero. If the value in FRA is greater than or equal to zero, floating point register FRT is set to the contents of floating-point register FRC. If the value in FRA is less than zero or is a NaN, floating point register FRT is set to the contents of floating-point register FRB.The comparison ignores the sign of zero; both +0 and -0 are equal to zero.