Data Sheet
June 2001
DSP16410B Digital Signal Processor
Agere Systems Inc.
Agere Systems—Proprietary
Use pursuant to Company instructions
241
6 Software Architecture
(continued)
6.2 Registers
(continued)
6.2.3 Register Encodings
(continued)
Table 157. psw0 (Processor Status Word 0) Register
15
LMI
14
13
LLV
12
11
10
9
8—5
4
3—0
LEQ
LMV
SLLV
SLMV
a1V
a1[35:32]
a0V
a0[35:32]
Bit
Field
Value
Description
R/W
Reset
Value
X
§
The ALU or ADDER cannot represent the result in 40 bits or the BMU control operand is out of range.
The ALU/ACS, ADDER, or BMU cannot represent the result in 32 bits. For the BMU, other conditions can also cause mathematical overflow.
§§ The most recent DAU result that was written to that accumulator resulted in mathematical overflow (LMV) with FSAT = 0.
Required for compatibility with DSP16XX family.
In this column, X indicates unknown on powerup reset and unaffected on subsequent reset.
ALU/ACS result or operation if the instruction uses the ALU/ACS; otherwise, ADDER or BMU result, whichever applies.
ALU/ACS result if the DAU operation uses the ALU/ACS; otherwise, ADDER or BMU result, whichever applies.
15
LMI
0
1
0
1
0
1
0
1
0
1
Most recent DAU result
is not negative.
Most recent DAU result
§
is negative (minus).
Most recent DAU result
§
is not zero.
Most recent DAU result
§
is zero (equal).
Most recent DAU operation
§
did not result in logical overflow.
Most recent DAU operation
§
resulted in logical overflow.
Most recent DAU operation did not result in mathematical overflow.
Most recent DAU operation
§
resulted in mathematical overflow.
Previous DAU operation did not result in logical overflow.
Sticky version of LLV that remains active once set by a DAU operation until
explicitly cleared by a write to
psw0
.
Previous DAU operation did not result in mathematical overflow.
Sticky version of LMV that remains active once set by a DAU operation until
explicitly cleared by a write to
psw0
.
The current contents of
a1
is not mathematically overflowed.
The current contents of
a1
is mathematically overflowed.
§§
Reflects the four lower guard bits of
a1
.
The current contents of
a0
is not mathematically overflowed.
The current contents of
a0
is mathematically overflowed.
§§
Reflects the four lower guard bits of
a0
.
R/W
14
LEQ
R/W
X
13
LLV
R/W
X
12
LMV
R/W
X
11
SLLV
R/W
0
10
SLMV
0
1
R/W
0
9
a1V
0
1
R/W
X
8—5
4
a1[35:32]
a0V
—
0
1
—
R/W
R/W
XXXX
X
3—0
a0[35:32]
R/W
XXXX