Analog Devices ADSP-BF53x Blackfin Reference page 558

Instruction Overview
Table 15-2. Multiply 16-Bit Operands Options
Option Description for
Register Half Destination
Default Signed fraction. Multiply 1.15 * 1.15 to
produce 1.31 results after left-shift cor-
rection. Round 1.31 format value at bit
16. (RND_MOD bit in the ASTAT
register controls the rounding.) Saturate
the result to 1.15 precision in destina-
tion register half. Result is between
minimum -1 and maximum 1-2
expressed in hex, between minimum
0x8000 and maximum 0x7FFF).
(FU) Unsigned fraction. Multiply 0.16 *
0.16 to produce 0.32 results. No shift
correction. Round 0.32 format value at
bit 16. (RND_MOD bit in the ASTAT
register controls the rounding.) Satu-
rate the result to 0.16 precision in desti-
nation register half. Result is between
minimum 0 and maximum 1-2
expressed in hex, between minimum
0x0000 and maximum 0xFFFF).
(IS) Signed integer. Multiply 16.0 * 16.0 to
produce 32.0 results. No shift correc-
tion. Extract the lower 16 bits. Saturate
for 16.0 precision in destination register
half. Result is between minimum -2
and maximum 2
hex, between minimum 0x8000 and
maximum 0x7FFF).
(IU) Unsigned integer. Multiply 16.0 * 16.0
to produce 32.0 results. No shift correc-
tion. Extract the lower 16 bits. Saturate
for 16.0 precision in destination register
half. Result is between minimum 0 and
maximum 2
between minimum 0x0000 and maxi-
mum 0xFFFF).
15-46 ADSP-BF53x/BF56x Blackfin Processor Programming Reference
-15
(or,
-16
(or,
15
-1 (or, expressed in
16
-1 (or, expressed in hex,
Description for
32-Bit Register Destination
Signed fraction. Multiply 1.15 * 1.15 to
produce 1.31 results after left-shift cor-
rection. Saturate results between mini-
mum -1 and maximum 1-2
The resulting hexadecimal range is mini-
mum 0x8000 0000 through maximum
0x7FFF FFFF.
Unsigned fraction. Multiply 0.16 * 0.16
to produce 0.32 results. No shift correc-
tion. Saturate results between minimum
0 and maximum 1-2
Unsigned integer. Multiply 16.0 * 16.0 to
produce 32.0 results. No shift correction.
Saturate results between minimum 0 and
32
maximum 2 -1.
In either case, the resulting hexadecimal
range is minimum 0x0000 0000 through
maximum 0xFFFF FFFF.
Signed integer. Multiply 16.0 * 16.0 to
produce 32.0 results. No shift correction.
Saturate integer results between mini-
31
mum -2 and maximum 2
15
Not applicable. Use (IS).
-31
.
-32
.
31
-1.