Lines Matching full:bias
19 argument is larger. The "bias" listed for the floating point operations
20 indicates how NaN comparisons are treated: "Gt bias" instructions return 1 for
21 NaN comparisons, and "lt bias" instructions return -1.
44 2d: cmpl-float <i>(lt bias)</i><br/>
45 2e: cmpg-float <i>(gt bias)</i><br/>
46 2f: cmpl-double <i>(lt bias)</i><br/>
47 30: cmpg-double <i>(gt bias)</i><br/>
95 For the -float and -double variants, an addition "bias" specifies what
99 A "lt bias" results in vA'=-1.
102 A "gt bias" results in vA'=1.
