Computer Hardware User's Guide

5-1
Data Formats and Floating-Point Operation
In the ’C3x architecture, data is organized into three fundamental types: integer,
unsigned integer, and floating-point. The terms integer and signed integer are
equivalent. The ’C3x supports short and single-precision formats for signed and
unsigned integers. It also supports short, single-precision, and extended-
precision formats for floating-point data.
Floating-point operations make fast, trouble-free, accurate, and precise com-
putations. Specifically, the ’C3x implementation of floating-point arithmetic facili-
tates floating-point operations at integer speeds, while preventing problems with
overflow, operand alignment, and other burdensome tasks that are common in
integer operations.
This chapter discusses data formats and floating-point operations supported
in the ’C3x.
Topic Page
5.1 Integer Formats 5-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2 Unsigned-Integer Formats 5-3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3 Floating-Point Formats 5-4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.4 Floating-Point Conversion (IEEE Std. 754) 5-14. . . . . . . . . . . . . . . . . . . . .
5.5 Floating-Point Multiplication 5-26. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.6 Floating-Point Addition and Subtraction 5-32. . . . . . . . . . . . . . . . . . . . . . .
5.7 Normalization Using the NORM Instruction 5-37. . . . . . . . . . . . . . . . . . . . .
5.8 Rounding (RND Instruction) 5-39. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.9 Floating-Point to Integer Conversion (FIX Instruction) 5-41. . . . . . . . . . .
5.10 Integer to Floating-Point Conversion (FLOAT Instruction) 5-43. . . . . . .
5.11 Fast Logarithms on a Floating-Point Device 5-44. . . . . . . . . . . . . . . . . . . .
Chapter 5