[Zhang San learns C language] - deeply understand data storage

【 Zhang Sanxue C Language 】— Deep understanding of data storage

One , data type

 Different data types represent the size of space opened up by different data in memory  
 eg:int  In memory 4 Bytes 

whole type	Content Be careful ： The contents in brackets are the number of bytes occupied by each type
data type	char（1 byte ） short（2 byte ） int（4 byte ） long（8 byte ）
floating-point	float （8 byte ） double（16 byte ）
Construction type	An array type （ The memory size is decided by yourself ） Structure struct （ Observe memory alignment ） enumeration enum（ Memory elements determine ） Joint type union（ The largest element ）
Pointer types	void Indicates the type , Usually used for the return type of a function , The parameters of the function , Pointer types

C Language has no string type , So use arrays to store strings   eg:char[10]="abcdef"

Two . The storage of integer data in memory

1. Source code , Inverse code , Complement code

 The symbols in the computer have been stored in memory in binary form 
 Data is stored in memory in the form of complement 
 The first is the sign bit , The remaining bits are data bits 
0 Being positive ,1 Negative 

Source code ： Direct binary can be translated into binary in the form of positive and negative numbers

Inverse code ： The sign bits remain the same , The data position is reversed

Complement code ： Inverse code +1

Positive numbers ： Source code = Inverse code = Complement code

negative ： Store... As a complement

eg:
 Positive numbers 
date:10
bin:00000000 00000000 00000000 00001010
 Source code = Inverse code = Complement code =00000000 00000000 00000000 00001010

 negative 
date:-10
bin: 10000000 00000000 00000000 00001010
 Inverse code ：11111111 11111111 11111111 11110101
 Complement code ：11111111 11111111 11111111 11110110

2. The reason why data is stored in the form of complement in memory

a. The use of complement can unify the symbol bit and data field

b.CPU Only adders , Using complement can make addition and subtraction unified

3, data storage

Byte order ：

Concept ：cpu The order in which data in memory is accessed in bytes

Because there are high and low addresses in memory , There are high and low binary data

The host byte order mainly depends on cpu framework : X86- The small end ; MIPS- Big end
Byte order is mainly used in network communication , Communication between different hosts , If the byte order is different, it will cause data ambiguity

classification ：int=0x11223344

Big end storage ： The low address stores the high address Storage in memory 11 22 33 44

Small end storage ： The low address stores the low address Storage in memory 44 33 22 11

notes ： The size end is a high-frequency examination site in interviews and examinations , To learn more, click the link below to learn .

http://t.csdn.cn/q270n

3、 ... and . The storage of floating-point data in memory

1,IEEE754 The standard document stipulates , The representation of any floating-point number in memory is divided into three parts

Symbol	Memory occupied by each bit of single precision floating-point numbers	Memory occupied by each bit of double precision floating-point numbers
s— Sign bit	1 A bit	1 A bit
E— Exponential position	8 A bit	11 A bit
M— mantissa	23 A bit	52 A bit

Floating point numbers are expressed in exponential form in the computer , Decompose into ： Numeral character , mantissa , Exponential sign , Index Four parts .

The number symbol accounts for 1 Bit binary , The positive and negative of a number .0 Express positive ,1 Negative

Exponential operators account for 1 Bit binary , The positive and negative of the index .

The mantissa represents the significant number of floating-point numbers ,0.xxxxxxx, But don't save the beginning 0 Sum point .

The effective number of the index .

How many places does the index occupy , How many bits does the mantissa occupy , It's up to the computer system to decide .

Two special cases :

E For all 0: At this time, the real index value is 1-127=-126,M Represent true value , Usually used to indicate positive and negative 0

E For all 1: At this time, the real index value is 255-127=128,M For all 0 Hour means positive and negative infinity

2. Value range

type	Number of bits	Significant figures	Range of values
float	32	6-7	-3.410(-38)～3.410(38)
double	64	15-16	-1.710(-308)～1.710(308)
long double	128	18-19	-1.210(-4932)～1.210(4932)

If storage is more important than accuracy , Consider using float type . contrary , If accuracy is the most important condition , Then use double type .

Floating point variables can be promoted to larger Radix types （ from float Type to double type ）. When you perform arithmetic on floating-point variables , Usually there will be an improvement . This arithmetic is always performed with the same precision as the variable with the highest precision .

float f_short;
double f_long;
long double f_longer;
gf_short = f_short * f_long

 In the previous example , Variable f_short Promote to type  double  And with f_long Multiply ; then , Result rounded to type  float, Then give to f_short.

3. Application example analysis

stay C and C++ in , The following assignment statement ：

float  a=0.1; 

Compiler error ：warning C4305: ‘initializing’ : truncation from 'const double ’ to 'float ’

reason ： stay C/C++ in , To the right of the equal sign 0.1, Think it's a float, But the compiler thinks it's a double（ Because decimals default to double）, So I have to report this warning, It's usually changed to 0.1f It's all right. .

Welcome to exchange , If it works , Remember to comment + Collection + Pay attention to ！ meow ~~

Today's rubbish ： I didn't understand until I grew up , Nothing is difficult in the world , I'm just afraid that those who have a mind .