Analog Devices ADSP-BF53x Blackfin, Reference

Page: 96 / 1042

Multiply Accumulators (Multipliers)
2-36 ADSP-BF53x/BF56x Blackfin Processor Programming Reference
Inputs are treated as fractional or integer, unsigned or two’s-complement.
Multiplier instructions include:
• Multiplication
• Multiply and accumulate with addition, rounding optional
• Multiply and accumulate with subtraction, rounding optional
• Dual versions of the above
Multiplier Operation
Each multiplier has two 32-bit inputs from which it derives the two 16-bit
operands. For single multiply and accumulate instructions, these operands
can be any Data registers in the Data Register File. Each multiplier can
accumulate results in its Accumulator register,
A1
or
A0
. The accumulator
results can be saturated to 32 or 40 bits. The multiplier result can also be
written directly to a 16- or 32-bit destination register with optional
rounding.
Each multiplier instruction determines whether the inputs are either both
in integer format or both in fractional format. The format of the result
matches the format of the inputs. In MAC0, both inputs are treated as
signed or unsigned. In MAC1, there is a mixed-mode option.
If both inputs are fractional and signed, the multiplier automatically shifts
the result left one bit to remove the redundant sign bit. Unsigned frac-
tional, integer, and mixed modes do not perform a shift for sign bit
correction. Multiplier instruction options specify the data format of the
inputs. See
“Multiplier Instruction Options” on page 2-40 for more
information.