NS32FV16 Datasheet, PDF (52/104 Pages) Texas Instruments – Advanced Imaging/Communication Signal Processors

3 0 Functional Description (Continued)
Operation
Operation
aligned integer X Y
integer Z
for (n40 n ^k LENG n00)
Z 2n 4 (integer) (Z 2n 0 X n low *
Y n low)
Z 2n01 4 (integer) (Z 2n01 0 X n high
* Y n high)
real X Y Z
for (n40 n ^k LENG n00)
Z n 4 (real) (Z n 0 X n * Y n )
Note When PARAM CLR is set to ‘‘1’’ only multiplication is performed
without addition When PARAM SUB is set to ‘‘1’’ the ‘‘a’’ sign is
replaced by a ‘‘b’’ sign
VARMAD Vector Aligned Real Multiply and Add
Note When PARAM CLR is set to ‘‘1’’ only multiplication is done without
addition When PARAM SUB is set to ‘‘1’’ the ‘‘a’’ sign is replaced
by a ‘‘b’’ sign
The VARMAD instruction multiplies corresponding elements
of the X and Y real vectors and adds or subtracts the result
to the real vector Z This result is placed in the Z output
VAIMADS Vector Aligned Integer Multiply and Add
Saturated
The VAIMADS instruction multiplies corresponding ele-
ments of the X and Y integer vectors and adds or subtracts
the result as an integer value to the integer vector Z This
result is placed in the Z output vector The saturation logic
provides clamping of the accumulator results before writing
the result back to the Z vector whenever the result cannot
te ^{be represented correctly within the limits of the integer data}
type
Syntax
EXEC VAIMADS
15
10000
le ^Operation
11 10
0
101 0101 1100
aligned integer X Y
integer Z
for (n40 n ^k LENG n00)
o ^{Z 2n 4 (integer) (Z 2n 0 X n low *}
Y n low)
Z 2n01 4 (integer) (Z 2n01 0 X n high *
Y n high)
s _{VRMAD Vector Real Multiply and Add}
The VRMAD instruction multiplies corresponding elements
of the X and Y real vectors and adds or subtracts the result
b ^{to the real vector Z This result is placed in the Z output}
vector
Syntax
EXEC VRMAD
O ¹⁵
11 10
0
vector
Syntax
EXEC VARMAD
15
10000
Operation
11 10
0
100 0000 1110
aligned real X Y Z
for (n40 n ^k LENG n00)
Z n low 4 (real) (Z n low 0 X n low *
Y n low)
Z n high 4 (real) (Z n high 0 X n high
* Y n high)
Note When PARAM CLR is set to ‘‘1’’ only multiplication is performed
without addition When PARAM SUB is set to ‘‘1’’ the ‘‘a’’ sign is
replaced by a ‘‘b’’ sign
VEMAD Vector Extended Multiply and Add
The VEMAD instruction multiplies corresponding elements
of the X and Y real vectors and adds or subtracts the result
as an extended-precision value to the extended-precision
vector Z This result is placed in the Z output vector
Syntax
EXEC VEMAD
15
11 10
0
10000
101 0001 0010
10000
100 0011 0011
51