| Purpose: | To perform arithmetic division |
| Format: | DIVX2: opcode divr.rx, quo.mx
DIVX3: opcode divr.rx, divd.rx, quo.wx |
| Operation: | DIVX2: quo <- quo / divr;
DIVX3: quo <- divd / divr; |
Condition
Codes: |
N <- quo LSS 0;
Z <- quo EQL 0;
V <- overflow Or (divr equal to 0);
C <- 0; |
| Exceptions: | Integer Overflow;
Divide by Zero
Floating Overflow;
Floating Underflow;
Reserved Operand (Float) |
| Opcodes: |
| 86 | DIVB2 | Divide Byte 2 Operand |
| 87 | DIVB3 | Divide Byte 3 Operand |
| A6 | DIVW2 | Divide Word 3 Operand |
| A6 | DIVW3 | Divide Word 2 Operand |
| C6 | DIVL2 | Divide Long 2 Operand |
| C7 | DIVL3 | Divide Long 3 Operand |
| 46 | DIVF2 | Divide Float 2 Operand |
| 47 | DIVF3 | Divide Float 3 Operand |
| 66 | DIVD2 | Divide Double 2 Operand |
| 67 | DIVD3 | Divide Double 3 Operand |
|
| Description: |
In the 2 operand format, the quotient operand is divided by the divisor operand and the quotient operand is replaced with the result. The divisor operand is unaffected.
In the 3 Operand format, the dividend is divided by the divisor operand and the quotient operand is replaced with the result.The dividend and divisor operands are unaffected.
In floating point format, the result is rounded. |
| Notes: |
- 1.
- Integer division is performed such that the remainder (unless it is zero) has the same sign as the dividend; i.e. the result is truncated towards 0.
- 2.
- Integer overflow occurs if and only if the largest negative number is divided by -1. On overflow, the operands are effected as described in 3, below.
- 3.
- If the integer divisor operand is zero, then in 2 operand integer format, the quotient operand is not affected; in 3 operand format the quotient operand is replaced by the dividend operand.
- 4.
- On a floating reserved operand fault, the quotient is unaffected and the condition codes are unpredictable.
- 3.
- On floating underflow, the quotient operand is replaced by 0.
- 4.
- On floating divide by zero or on floating overflow, the quotient operand is replaced by an operand of all 0 bits except for the sign bit of 1 (a reserved operand (-1)). N <- 1; Z <- 0; V <- 1; and C <- 0.
|