SH7721 datasheet(116/1524 Pages) RENESAS

116 / 1524 page

Section 2

CPU

Rev. 3.00 Jan. 18, 2008 Page 54 of 1458

REJ09B0033-0300

2.5

Features of CPU Core Instructions

2.5.1

Instruction Execution Method

(1)

Instruction Length

All instructions have a fixed length of 16 bits and are executed in the sequential pipeline. In the

sequential pipeline, almost all instructions can be executed in one cycle. All data items are handles

in longword (32 bits). Memory can be accessed in byte, word, or longword. In this case, Memory

byte or word data is sign-extended and operated on as longword data. Immediate data is sign-

extended to longword size for arithmetic operations (MOV, ADD, and CMP/EQ instructions) or

zero-extended to longword size for logical operations (TST, AND, OR, and XOR instructions).

(2)

Load/Store Architecture

Basic operations are executed between registers. In operations involving memory, data is first

loaded into a register (load/store architecture). However, bit manipulation instructions such as

AND are executed directly on memory.

(3)

Delayed Branching

Unconditional branch instructions are executed as delayed branches. With a delayed branch

instruction, the branch is made after execution of the instruction (called the slot instruction)

immediately following the delayed branch instruction. This minimizes disruption of the pipeline

when a branch is made.

This LSI supports two types of conditional branch instructions: delayed branch instruction or

normal branch instruction.

Example:

BRA

TARGET

ADD

R1, R0

; ADD is executed before branching to

the

TARGET

SH7721 Datasheet(PDF) 116 Page - Renesas Technology Corp