SH7261 datasheet(145/1348 Pages) RENESAS

145 / 1348 page

Section 5 Exception Handling

Rev. 2.00 Sep. 07, 2007 Page 113 of 1312

REJ09B0320-0200

5.7.5

Integer Division Instructions

When an integer division instruction performs division by zero or the result of integer division

overflows, integer division instruction exception handling starts. The instructions that may become

the source of division-by-zero exception are DIVU and DIVS. The only source instruction of

overflow exception is DIVS, and overflow exception occurs only when the negative maximum

value is divided by

−1. The CPU operates as follows:

1. The exception service routine start address which corresponds to the integer division

instruction exception that occurred is fetched from the exception handling vector table.

2. The status register (SR) is saved to the stack.

3. The program counter (PC) is saved to the stack. The PC value saved is the start address of the

integer division instruction at which the exception occurred.

4. After jumping to the address fetched from the exception handling vector table, program

execution starts. The jump that occurs is not a delayed branch.

5.7.6

Floating-point Operation Instruction

When the bits V, Z, O, U, or I in the enabled field of the floating point status/control register

(FPSCR) are set, a FPU exception is generated. This means that instructions that cause disabled

operation exception defined by IEEE754 standard, division exception by zero, overflow (possible

instruction), underflow (possible instruction), or imprecise exception (possible instruction) are

yielded. Floating-point operation instructions that can be exception sources are FADD, FSUB,

FMUL, FDIV, FMAC, FCMP/EQ, FCMP/GT, FLOAT, FTRC, FCNVDS, FCNVSD, and

FSQRT.

FPU exceptions occur only when the said enabled bits are set. When the FPU detects exception

sources, the FPU operation stops and exception occurrence is notified to the CPU. The CPU starts

the exception handling as follows:

1. The exception service routine start address which corresponds to the FPU exception that

occurred is fetched from the exception handling vector table.

2. The status register (SR) is saved to the stack.

3. The program counter (PC) is saved to the stack. The PC value saved is the start address of the

instruction to be executed after the last executed instruction.

4. After jumping to the address fetched from the exception handling vector table, program

execution starts. The jump that occurs is not a delayed branch.

SH7261 Datasheet(PDF) 145 Page - Renesas Technology Corp