Memory ==c language 1

One 、 Why do programs need memory to run

1、 The purpose of a computer program is

• Results and processes ： Use functions as analogies , The formal parameters of a function are the data to be processed , Function ontology is code , The return value of the function is the result , The execution process of a function is a process .
• Code example ：

  int add(int a,int b){
        // This function is executed to get the result 
  	return a+b;
  }
  void add(int a,int b{
        // The execution of this function focuses on the process 
  	int c ;
  	c=a+b;
  	printf("c=%d\n",c);
  	return c;
  }
  int add(int a,int b){
        // The execution of this function focuses on the process , Focus on results 
  	int c;
  	c=a+b;
  	printf("c=%d\n",c);
  	return c;
  }
  

2、 The running process of a computer program

• The running process of a computer program is the process in which multiple functions in the program run successively , The process of function operation is the process of data processing .

3、 Von Neumann structure and Harvard structure

• Von Neumann structure ： Data and code together , Together .（x86）
  Harvard structure ： Data and code are placed separately , Independent addressing .（ARM）
• Code ： It's a function
  data ： Global variables 、 local variable
• Example ： stay S5PV210 Running in Linux On the system , When running the application , All the application code and data are placed in DRAM in , This is the von Neumann structure ; In a single chip computer , We burn the program to Flash in . Then the program is Flash Running in the Central Plains , The data involved in the program （ Global variables 、 local variable ） Can't put Flash in , Must be on RAM（SRAM） in , This is the Harvard structure .

4、 Dynamic memory DRAM And static memory SRAM

• (1) DRAM The storage principle of is the way that capacitors store charge ,SRAM The storage principle of is the way of trigger .
  
• （2）SDRAM： Synchronous dynamic random access memory .

5、 Why do programs need memory to run ？

• Memory is used to store data （ And code ）, Global variables 、 Local variables, etc , And data is the component and processing object of the program , Therefore, memory is the essential requirement of program operation . The simpler the program, the smaller the memory it needs to run , The more complex the program, the more memory it needs to run .
• Memory management is a very important knowledge point of programming , We learn many key points of programming for memory management , For example, data structures and algorithms .

6、 Think deeply ： How to manage memory

• From the perspective of operating system ：
  • The operating system holds all the hardware resources , Because there's a lot of memory , So the operating system divides the memory into pages （ That is, the block , It's usually 4KB）, Then manage it in pages . The page is managed in bytes . In fact, the principle of operating system memory management is very complex , But for programmers , Just make good use of the memory management provided by the operating system API that will do . For example, in C Used in language malloc、free To apply for and release memory
  • Without operating system , The program needs to operate memory directly , Programmers need to calculate and arrange memory by themselves .
• From the perspective of programming language ： Different languages provide different memory management interfaces
  • ① assembly ： There is no memory management , Memory management depends on the programmer himself , Directly use memory address when operating memory , Very trouble
  • ②C Language ： The compiler helps us directly manage memory addresses , Access memory through the variable name provided by the compiler , If you need a large block of memory under the operating system, you can use API（malloc、free） To access system memory , And bare metal programs need large blocks of memory, and need to define their own data to solve .
  • ③c++： The use of memory is further encapsulated , We can use new To create objects （ It's just about allocating memory for objects ）, Use after use delete To delete objects （ That is, freeing up memory ）. therefore c++ Better memory management than c The language is more advanced , But memory management still needs to be implemented by programmers , Not releasing after use will cause memory leakage .
  • ④java/c# etc. : These languages do not manipulate memory , Instead, the memory is manipulated through a virtual machine . Virtual machine as our programmer's agent , Help us deal with memory release .
• summary ：java/c# It seems better than c/c++ More advanced , But the memory recycling of virtual machine needs to pay a certain performance price , When we care about the speed of developing programs , Will use java/c# Other languages ; When we are very concerned about the performance of the program , Will use c/c++.

Two 、 position 、 byte 、 Half word 、 Word concept and memory bit width

1、 What is memory ？

• ① Hardware angle ： Memory is actually an accessory of a computer （ Memory module ）. According to different implementation principles , Memory can be divided into SRAM and DRAM（SDRAM\DDR1SDRAM\DDR2SDRAM、LPDDR5SDRAM）
• ② Logical angle ： Can read and write at random （ Given an address, you can access ）; Memory is naturally used to store variables in programming ,c A variable of language corresponds to a small segment in memory .

2、 Logical abstraction diagram of memory

• Logically, memory is composed of infinite small lattices , Each small grid corresponds to an address
• In reality, the memory size is limited , for example 32 Bit operating system （32 Bit system means 32 Bit data line , But the general address line is also 32 position , That is, the memory address is only 32 Bit binary number , So the logical size is 2^32） The memory limit is 4GB. actually 32 The memory in the bit system is less than or equal to 4GB Of .

3、 Bits and bytes 、 Words and half words

• The unit of memory unit size is 4 individual ： position （1bit） byte （8bit） Half word （ It's usually 16bit） word （ It's usually 32bit）

4、 Memory bit width

• In terms of hardware ： The hardware implementation of memory itself has width , Some memory modules are 8 Bit , Some are 16 Bit . It should be emphasized that memory chips can be connected in parallel , Through parallel connection 8 Bit memory chip can realize 16 Bits or 32 Bit of memory .（ Digital circuit has mentioned this knowledge ）
• Logically speaking ： The memory bit width is logically arbitrary , But the actual hardware bit width is not arbitrary , So we need to program according to the hardware characteristics of different bit widths
• Memory bit width is the number of bits of data that can be transferred in a clock cycle , The larger the number of bits, the larger the amount of data that can be transmitted instantaneously , This is one of the important parameters of memory

3、 ... and 、 Memory addressing 、 Memory alignment

1、 Memory addressing method

• The memory unit and address are uniquely 、 Bind forever .
• Address and space are two aspects of a memory unit .

2、 The key ： Memory addressing is in bytes

3、 The relationship between memory and data type

• c The data types in the language are char、short、 int 、long、 float 、double.
• Data types are used to define variables , Variables need to be stored in memory , So the data type must match the memory to get the best performance .
• stay 32 The best way to define variables in bit system is to use int, Because it's efficient . The reason is that the system itself is 32 Bit structure , A processing 32 Bit data . stay 32 Defined in the bit system bool Type variables are also used inside the system int Realized .
• In actual programming, we should choose between memory and efficiency according to the actual situation .

4、 Memory alignment

• Alignment access works well with hardware , So it's very efficient ;
• Non aligned access is not compatible with hardware , So the efficiency is not high

Four 、C How language operates memory

1、C Language encapsulates memory address

• Use variable names to access memory
• What data types mean ： Represents the length and parsing method of a memory lattice
• A function is an encapsulation of a piece of code , The essence of function name is the first address of this code , So the essence of function name is also a memory address .

2、 Using pointers to access memory indirectly

• About type , Whether ordinary variable type or pointer variable type , Just remember that ： Type is only for the following numbers or symbols （ Represents the memory address ） The length and parsing method of the memory represented

3、 Using arrays to manage memory

• There is no essential difference between array management memory and variables , It's just that symbols are analyzed in different ways .
• The first element in the array is called the first element , The address of the first byte of the first element is the first address of the first element .

5、 ... and 、 The structure of memory management

1、 The significance of data structure

• Data structure is to study how data is organized in memory 、 Processing knowledge .

2、 Simplest data structure ： Array

• Organize variables of the same type , Easy to manage .
• advantage ： Store in memory in order , Subscripts are available for random access .
  Inferiority ： All elements must be of the same type , When defining, the size must be given and cannot be modified .

3、 The structure is on the stage

• Structure was invented to solve the first defect of array ： All elements in an array must be of the same type .
• Code example ：

struct people{
    
	int age;
	char name[10];
	int height;
};

4、 Digression ： Embedded pointer in the structure to realize object-oriented ( node )

• Process and object oriented ： Process oriented refers to the step-by-step splitting of problem solving steps , As required , Finally solve the problem ; Object-oriented is to decompose the problem transaction into various objects , Classification to solve problems .
• C Languages are process-oriented , but C Written in language Linux The system is object-oriented .
• In a non object-oriented language , It's not necessarily impossible to implement object-oriented code , It's just easier to use object-oriented language to realize object-oriented , Because language itself does a lot of work .
• Code example ：

struct s{
    
	int age;// Common variables 
	int(*pfunc)(void);// A function pointer , Point to int xxx（void） Such functions 
}
// Using such a structure can realize object-oriented , In this way, the structure containing function pointers is similar to that in object-oriented class

6、 ... and 、 Memory management stack （stack）

1、 What is stack? ？

• A stack is a data structure ,C Stack is used in language memory management to save function parameters and local variables .

2、 The characteristics of stack managed memory （ Small memory 、 automation ）

• First in, then out （FILO）, And the queue is first in, first out （FIFO）.
• The characteristic of stack is that the entrance is the exit , There's only one mouth , The other one is blocked , So it's first in, then out . The characteristic of the queue is that there are both entrances and exits , You have to go in through the entrance , Come out of the exit , So it's first in, first out .

3、 Examples of stack applications ： local variable

• c Local variables in language are implemented by stack , Defining local variables , The corresponding operation is stack （ automation );
  When the function exits the local variable logout , The corresponding operation is stack （ automation ）.
• The advantages of stacks ： convenient 、 Neither allocation nor recycling need to be operated by programmers ,c Language is done automatically
• Define local variables but not initialized , Values are random , Why? ？
  Because the trestle is used repeatedly , And it is not cleared after each use

4、 Advantages and disadvantages of stack

• advantage ： automation , There is no need for programmers to manually operate memory
• Inferiority ： First , The trestle has sizes , So the stack memory size is not easy to set . So it's too small to overflow , Too big to waste memory （ Similar to arrays ）. secondly , Stack overflow is very harmful , Be sure to avoid . So we are C You can't define too many or too large local variables in a language （ For example, we cannot define int a[10000]; When using recursion to solve problems, we must pay attention to recursive convergence ）.

7、 ... and 、 Heap of memory management （heap）

1、 What is a heap

• Heap is a way of memory management , Its characteristic is freedom （ Apply at any time 、 Release 、 The size of the block is arbitrary ）
• Heap memory is the partition of the operating system to the heap manager （ A piece of code for the operating system , Belongs to the operating system memory management unit ） To manage , And then to the user （ User process ） Provide API（malloc and free） Heap to use .
• When do I need to use heap memory ？
  When the memory capacity is large 、 When it needs to be used and released repeatedly , A lot of data structures （ Like linked lists ） The use of requires the use of heap memory .

2、 Characteristics of heap memory management

• Unlimited capacity （ Regular needs can be met ）.
• Both application and release need to be done manually , If it is not released after use, it will cause memory leakage , The accumulation of memory leaks will lead to memory overflow and even system crash .
• （ Memory leak ： apply (new/malloc) I got a piece of memory , But there is no manual release (delete/free) Memory , As a result, the pointer has disappeared , And what the pointer points to is still , Can't control this memory ）

3、C Language operation heap memory interface

• （1） There are three functions with similar functions when applying for heap memory ：malloc、calloc、realloc.
  • void *malloc(size_t size); // Request a specified amount of memory
  • void *calloc(size_t nmemb, size_t size); // apply nmemb Memory units , Every size byte
  • void *realloc(void *ptr, size_t size); // Change the size of the space already requested
• （2） Heap memory release is the simplest , Call directly free Function
  void free(void *ptr);
• (3) Code example ：

//  For example, to apply for 10 individual int Element memory 
malloc(40);
calloc(10,4);

• (4)C The number of array elements must be given when the language defines an array , And can no longer be modified . And in the Java In advanced languages , You can modify the size of the array , The principle is to recreate a new array , Then release the original array .realloc The implementation principle of is similar .

4、 Advantages and disadvantages of heaps

• advantage ： Flexible use
• Inferiority ： Programmers need to deal with details such as release , It's easy to make mistakes , Rely heavily on programmer level .

8、 ... and 、 Complex data structures

1、 Linked list 、 Hashtable 、 Binary tree 、 Map, etc.

• Linked list ： Linked list is the most important ,Linux Linked lists are often used in the kernel , The writing of drivers and applications often requires linked lists , So you must master the linked list ： Definition of structure , Creation of linked list , Addition of linked list 、 Delete 、 check , The reverse order of the linked list
• Hashtable ： Not very often , Generally, you don't need to realize it yourself , There are many hash tables directly implemented by others . We need to understand the principle of hash table 、 characteristic 、 Applicable scenario .
• Binary tree 、 Map, etc. ： Used to deal with specific problems , Rarely used , Don't get too tangled up .

2、 Why more complex data structures are needed

• Because there are many practical problems in reality , Specific problems may require specific data structures and algorithms to solve .

3、 The relationship between data structure and algorithm

• The invention of data structure is to cooperate with a certain algorithm , The algorithm is to deal with specific problems , Its implementation depends on specific data structures

4、 How to learn data structure and algorithm

• The data structure is consistent with the algorithm , We should study together .
• Data structure and algorithm are not all the key points for embedded , Don't study blindly .

5、 give an example

• Linux When the kernel manages character device drivers , Hash table used （hash table, Hash table ）, Therefore, many characteristics of character device driver are related to the characteristics of hash table .