Floating-point encoding on IBM i

Valid values for floating-point encoding

ENFUND

Undefined floating-point encoding.

Floating-point numbers are represented using an encoding that is undefined.

ENFNOR

Normal IEEE (The Institute of Electrical and Electronics Engineers) float encoding. Floating-point numbers are represented using the standard IEEE floating-point format, with the bytes arranged as follows:

• The least significant byte in the mantissa has the highest address of any of the bytes in the number; the byte containing the exponent has the lowest address
• The least significant bit in each byte is next to the byte with the next higher address; the most significant bit in each byte is next to the byte with the next lower address

Details of the IEEE float encoding might be found in IEEE Standard 754.

ENFREV

Reversed IEEE float encoding.

Floating-point numbers are represented in the same way as ENFNOR, but with the bytes arranged in reverse order. The bits within each byte are arranged in the same way as ENFNOR.

ENF390

System/390 architecture float encoding.

Floating-point numbers are represented using the standard System/390 floating-point format; this is also used by System/370.

Parent topic: Machine encodings on IBM i