Multiplier Instruction Options - Analog Devices ADSP-BF53x Blackfin Reference

Multiply Accumulators (Multipliers)

Multiplier Instruction Options

The following descriptions of multiplier instruction options provide an
overview. Not all options are available for all instructions. For informa-
tion about how to use these options with their respective instructions, see
Chapter 15, "Arithmetic Operations."
default
(IS)
(FU)
(IU)
(T)
(TFU)
(ISS2)
2-40 ADSP-BF53x/BF56x Blackfin Processor Programming Reference
No option; input data is signed fraction.
Input data operands are signed integer. No shift
correction is made.
Input data operands are unsigned fraction. No shift
correction is made.
Input data operands are unsigned integer. No shift
correction is made.
Input data operands are signed fraction. When
copying to the destination half register, truncates
the lower 16 bits of the Accumulator contents.
Input data operands are unsigned fraction. When
copying to the destination half register, truncates
the lower 16 bits of the Accumulator contents.
If multiplying and accumulating to a register:
Input data operands are signed integer. When copy-
ing to the destination register, Accumulator
contents are scaled (multiplied x2 by a one-place
shift-left). If scaling produces a signed value larger
than 32 bits, the number is saturated to its maxi-
mum positive or negative value.