14.06.2015 Views

Power ISA™ Version 2.03 - Power.org

Power ISA™ Version 2.03 - Power.org

Power ISA™ Version 2.03 - Power.org

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

<strong>Version</strong> <strong>2.03</strong><br />

Chapter 4. Floating-Point Processor [Category:<br />

Floating-Point]<br />

4.1 Floating-Point Processor Overview 89<br />

4.2 Floating-Point Processor Registers 90<br />

4.2.1 Floating-Point Registers . . . . . . . 90<br />

4.2.2 Floating-Point Status and Control<br />

Register. . . . . . . . . . . . . . . . . . . . . . . . . 91<br />

4.3 Floating-Point Data . . . . . . . . . . . . . 93<br />

4.3.1 Data Format. . . . . . . . . . . . . . . . . 93<br />

4.3.2 Value Representation . . . . . . . . . 94<br />

4.3.3 Sign of Result . . . . . . . . . . . . . . . 95<br />

4.3.4 Normalization and<br />

Denormalization . . . . . . . . . . . . . . . . . . 95<br />

4.3.5 Data Handling and Precision . . . . 96<br />

4.3.5.1 Single-Precision Operands. . . . 96<br />

4.3.5.2 Integer-Valued Operands . . . . . 97<br />

4.3.6 Rounding . . . . . . . . . . . . . . . . . . . 97<br />

4.4 Floating-Point Exceptions. . . . . . . . 98<br />

4.4.1 Invalid Operation Exception. . . . 100<br />

4.4.1.1 Definition. . . . . . . . . . . . . . . . . 100<br />

4.4.1.2 Action . . . . . . . . . . . . . . . . . . . 100<br />

4.4.2 Zero Divide Exception . . . . . . . . 101<br />

4.4.2.1 Definition. . . . . . . . . . . . . . . . . 101<br />

4.4.2.2 Action . . . . . . . . . . . . . . . . . . . 101<br />

4.4.3 Overflow Exception . . . . . . . . . . 101<br />

4.4.3.1 Definition. . . . . . . . . . . . . . . . . 101<br />

4.4.3.2 Action . . . . . . . . . . . . . . . . . . . 101<br />

4.4.4 Underflow Exception . . . . . . . . . 102<br />

4.4.4.1 Definition. . . . . . . . . . . . . . . . . 102<br />

4.4.4.2 Action . . . . . . . . . . . . . . . . . . . 102<br />

4.4.5 Inexact Exception . . . . . . . . . . . 103<br />

4.4.5.1 Definition. . . . . . . . . . . . . . . . . 103<br />

4.4.5.2 Action . . . . . . . . . . . . . . . . . . . 103<br />

4.5 Floating-Point Execution Models . 103<br />

4.5.1 Execution Model for IEEE Operations<br />

. . . . . . . . . . . . . . . . . . . . . . . . . . . 103<br />

4.5.2 Execution Model for<br />

Multiply-Add Type Instructions. . . . . . . 105<br />

4.6 Floating-Point Processor Instructions .<br />

106<br />

4.6.1 Floating-Point Storage Access<br />

Instructions . . . . . . . . . . . . . . . . . . . . . 106<br />

4.6.1.1 Storage Access Exceptions. . . 106<br />

4.6.2 Floating-Point Load Instructions. 106<br />

4.6.3 Floating-Point Store Instructions 109<br />

4.6.4 Floating-Point Move Instructions 113<br />

4.6.5 Floating-Point Arithmetic Instructions<br />

114<br />

4.6.5.1 Floating-Point Elementary Arithmetic<br />

Instructions. . . . . . . . . . . . . . . . . 114<br />

4.6.5.2 Floating-Point Multiply-Add Instructions<br />

. . . . . . . . . . . . . . . . . . . . . . . . . . . 118<br />

4.6.6 Floating-Point Rounding and Conversion<br />

Instructions . . . . . . . . . . . . . . . 120<br />

4.6.6.1 Floating-Point Rounding Instruction.<br />

. . . . . . . . . . . . . . . . . . . . . . . . . . . 120<br />

4.6.6.2 Floating-Point Convert To/From<br />

Integer Instructions . . . . . . . . . . . . . . . 120<br />

4.6.6.3 Floating Round to Integer Instructions<br />

[Category: Floating-Point.Phased-In]<br />

122<br />

4.6.7 Floating-Point Compare Instructions<br />

124<br />

4.6.8 Floating-Point Select Instruction 125<br />

4.6.9 Floating-Point Status and Control<br />

Register Instructions . . . . . . . . . . . . . . 125<br />

4.1 Floating-Point Processor<br />

Overview<br />

This chapter describes the registers and instructions<br />

that make up the Floating-Point Processor facility.<br />

The processor (augmented by appropriate software<br />

support, where required) implements a floating-point<br />

system compliant with the ANSI/IEEE Standard<br />

754-1985, "IEEE Standard for Binary Floating-Point<br />

Arithmetic" (hereafter referred to as "the IEEE standard").<br />

That standard defines certain required "operations"<br />

(addition, subtraction, etc.). Herein, the term<br />

"floating-point operation" is used to refer to one of<br />

these required operations and to additional operations<br />

defined (e.g., those performed by Multiply-Add or<br />

Chapter 4. Floating-Point Processor [Category: Floating-Point]<br />

89

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!