Motorola DSP96002 - Manual (Page 374)

A - 186 DSP96002 USER’S MANUAL MOTOROLA
INTRZ Floating-Point INTRZ
to Integer Conversion with Round to Zero
Operation:
Integer(D)
→
D.L (parallel data bus move)
Assembler Syntax:
INTRZ D
(move syntax - see the Move instruction de-
scription.)
Description:
Convert the specified floating-point operand to 32-bit, 2’s complement integer rounding towards zero. The
result is stored in the low portion of D. The high and middle portions of D remain unchanged. Since this
operation is frequently required (e. g., truncation assignment), this instruction has been implemented to
eliminate the need to change the rounding mode associated with INT.
The following table lists the results for some special cases:
Source operand Result
Greater than 2
31
- 1 $7FFFFFFF
Less than -2
31
$80000000
+infinity $7FFFFFFF
-infinity $80000000
NaN $FFFFFFFF
Input Operand(s) Precision: SEP Floating-Point.
Output Operand Precision: 32-bit integer.
CCR Condition Codes:
C - Not affected.
V - Set if source operand is a NaN, infinity, or its magnitude is too big to be represent-
able in the integer number range. Cleared otherwise.
Z - Set if result is zero. Cleared otherwise.
N - Set if source operand is negative. Cleared otherwise.
I - Set if source operand is infinity. Cleared otherwise.
LR - Not affected.
–
R - Not affected.
A - Not affected.