| ||||
| Home Products Download Events Support | ||||
Technical Support
On-Line Manuals
C251 User's Guide
C251 Introduction Compiling Programs Language Extensions Preprocessor 
Advanced Programming Customization Files 
Startup Code Variable Initialization Code Basic I/O Memory Allocation Optimizer General Optimizations 251-Specific Optimizations Options for Code Generation Global Register Coloring Segment Naming Conventions Segment Types Initialization Segments Function Symbols Parameter Passing Symbols Hardware Stack Calculating Stack Size Interfacing C Programs to Assembler Global Variables Function Parameters Passing in Memory Passing in Registers Function Return Values Using the SRC Directive Register Usage Overlaying Segments TINY Model XTINY Model SMALL Model XSMALL Model LARGE Model Data Storage Formats Bit Variables 1-Byte Scalars 2-Byte Scalars 4-Byte Scalars Floating-point Numbers Floating-point Errors Double Numbers Double Errors Absolute Memory Locations Absolute Access Macros Linker Location Controls The _at_ Keyword Debugging Error Messages Library Reference AppendixDouble Numbers
Scalars of type double are stored using eight bytes (64 bits). Type double numbers have a precision of 53 bits, which represents about 13 decimal digits. The format used follows the IEEE-754 standard.
A double number is expressed as the product of two parts: the mantissa and the exponent. The mantissa stores the actual digits of the number. The exponent stores the power to which the mantissa must be raised. For example:
±mantissa × 2exponent
The mantissa represents the actual binary digits of the floating-point number.
The power of two is represented by the exponent. The stored form of the exponent is an 11-bit value from 0 to 2047. The actual value of the exponent is calculated by subtracting 1023 from the stored value (0 to 2047) giving a range of –1023 to +1024.
The mantissa is a 53-bit value (representing about seven decimal digits) whose most significant bit (MSB) is always 1 and is, therefore, not stored. There is also a sign bit that indicates whether the floating-point number is positive or negative.
Tyep double numbers are stored in memory on word boundaries in the following format:
| Address+0 | Address+1 | Address+2 | Address+3 | |
|---|---|---|---|---|
| Contents | SEEE EEEE | EMMM MMMM | MMMM MMMM | MMMM MMMM |
| Address+4 | Address+5 | Address+6 | Address+7 | |
|---|---|---|---|---|
| Contents | MMMM MMMM | MMMM MMMM | MMMM MMMM | MMMM MMMM |
Where
| S | represents the sign bit where 1 is negative and 0 is positive. |
| E | is the exponent with an offset of 1023. |
| M | is the 53-bit normal mantissa. The highest bit is always 1 and, therefore, is not stored. |
| Contact Site Map Press Privacy | Copyright © 2012 ARM Ltd and ARM Germany GmbH. All rights reserved. |
