class Floating Point Classificator

Library:

Floating Point Arithmetic

Behavior

This component classifies the floating-point value on its west input. For a fully defined input, exactly one of the Zero, Subnormal, Normal, Infinite, and NaN outputs is 1. The Negative output reports the encoded sign bit independently, so it can be 1 at the same time as any class output, including for negative zero or a NaN with its sign bit set.

Zero recognizes both positive and negative zero. Subnormal recognizes an exponent of zero with a nonzero fraction. Normal recognizes a nonzero, nonmaximum exponent. The NaN output covers all NaN encodings; the component does not distinguish quiet from signaling NaNs. An input containing unknown or error bits is treated as NaN by the current value conversion.

This component has no ERR output.

Pins

West edge (input, bit width matches Float size)

The floating-point value to classify.

East edge, top to bottom (six outputs, bit width 1 each)

Negative, Zero, Subnormal, Normal, Infinite, and NaN, in that order.

Attributes

Poke Tool Behavior

None.

Text Tool Behavior

None.

Back to Floating Point Arithmetic library