Data Sheet
June 2001
DSP16410B Digital Signal Processor
Agere Systems Inc.
Agere Systems—Proprietary
Use pursuant to Company instructions
215
6 Software Architecture
(continued)
6.1 Instruction Set Quick Reference
(continued)
Table 130
defines the symbols used in instruction descriptions. Some symbols and characters are part of the
instruction syntax, and must appear as shown within the instruction. Other symbols are representational and are
replaced by other characters. The table groups these two types of symbols separately.
Table 130. Notation Conventions for Instruction Set Descriptions
Symbol
Meaning
Part of
Syntax
*
16-bit x 16-bit multiplication resulting in a 32-bit product.
Exception: When used as a prefix to an address register, denotes register-indirect addressing,
e.g.,
*r3
.
Squaring is a 16-bit x 16-bit multiplication of the operand with itself resulting in a 32-bit product.
40-bit addition
.
40-bit subtraction
.
Register postincrement.
Register postdecrement.
Arithmetic right shift (with sign-extension from bit 39).
Arithmetic left shift (padded with zeros).
Logical right shift (zero guard bits before shift).
Logical left shift (padded with zeros; sign-extended from bit 31).
40-bit bitwise logical AND
.
40-bit bitwise logical OR
.
40-bit bitwise logical exclusive-OR
.
Register shuffle
.
Ones complement (bitwise inverse).
Parentheses enclose multiple operands delimited by commas that are also part of the syntax.
Braces enclose multiple instructions within a cache loop.
The underscore character indicates an accumulator vector (concatenation of the high halves of a
pair of sequential accumulators, e.g.,
a0_1h
).
Lower-case characters appear as shown in the instruction.
Angle brackets enclose items delimited by commas, one of which must be chosen.
Mid braces enclose one or more optional items delimited by commas.
Replaced by either + or –.
Upper-case characters, character strings, and characters plus numerals (e.g., M, CON, and
IM16) are replaced. Replacement tables accompany each instruction group description.
Represents a statement of a DAU function:
F1
MAC.
F1E
Extended MAC.
F2
Special function.
F2E
Extended special function.
F3
ALU.
F3E
Extended ALU.
F4
BMU.
F4E
Extended BMU.
**2
+
–
++
––
>>
<<
>>>
<<<
&
|
^
:
~
( )
{ }
_
The ALU/ACS and ADDER perform 40-bit operations, but the operands can be 16 bits, 32 bits, or 40 bits. In the special case of the split-mode
F1E instruction (
xh=aSPEh±yh, xl=aSPEl±yl, aDE=aSEE+p0+p1, p0=xh**2, p1=xl**2
), the ALU performs two 16-bit addition/subtraction
operations in parallel.
Note that this symbol does not denote compound addressing as it does for the DSP16XX family.
(underscore)
lower-case
±
UPPER-
CASE
F Titles
Not Part
of Syntax
(Replaced)