Shulou(Shulou.com)05/31 Report--

In this article, the editor introduces in detail "what is the underlying architecture of Linux". The content is detailed, the steps are clear, and the details are handled properly. I hope this article "what is the underlying architecture of Linux" can help you solve your doubts.

The working mode of a computer

For a computer, the core is that CPU,CPU is the brain of the computer, and all devices revolve around it.

CPU is connected to other devices through bus (Bus). Among these devices, memory (Memory) is the most important.

CPU alone can not complete the computing task, many complex computing tasks need to save the intermediate results, and then proceed to the next step of calculation based on the intermediate results

CPU and memory are the core components of computing.

CPU itself cannot save so many intermediate results, so it needs to rely on memory

CPU

CPU consists of three parts: operation unit, data unit and control unit.

The computing unit only calculates, but it doesn't know which data to calculate and where to put the results.

If the data calculated by the computing unit has to go through the bus every time and take it directly into the memory, the speed will be very slow, so the data unit appears.

The data unit includes the cache and register group inside CPU, which is small in space but fast in speed.

The control unit is a unified command center that can get the next instruction and then execute it.

This instruction instructs the computing unit to take out some data in the data unit, calculate the result, and put it somewhere in the data unit.

Calculation process

1. Each process has a program on the hard disk, which is binary and stores one-by-one instructions that manipulate some data.

two。 When the process starts to run, there will be separate memory space, isolated but discontiguous-the program will be loaded into the memory space of process An and process B respectively, forming their own code snippets.

3. The data to be manipulated and the calculation results generated by the program will be placed in the data segment (memory).

4. In the control unit of CPU, there is an instruction pointer register, which records the address of the next instruction in memory-the control unit will constantly bring in the instructions of the code segment and put them in the instruction register first.

5. The components of the instruction: what to do + which data to operate-to execute the instruction, you need to give the * part to the operation unit and the second part to the data unit.

6. According to the address of the data, the data unit reads the data from the data segment to the data register, and eventually there are instructions to write the data back to the data segment in memory.

7. There are two registers in CPU, which specifically hold the code segment start address and data segment start address of the current processing process. The current process in the figure is process A.

8. CPU and memory transfer data through the bus. There are two types of data on the bus-address bus (Address Bus): address data, the number of bits determines how wide the address can be accessed-data bus (Data Bus): real data, the number of bits determines how much data can be taken at a time

X86 architecture

Model

The principle of 8086

Universal register

In order to temporarily store data, there are eight 16-bit general-purpose registers inside the 8086 processor, which belong to the data units within the CPU

They are AX, BX, CX, DX, SP, BP, SI and DI

AX, BX, CX and DX can be divided into two 8-bit registers, where H is High,L and Low

In this way, longer data can be temporarily stored, and shorter data can also be temporarily stored.

Control unit

IP register (Instruction Pointer Register) is the instruction pointer register

-points to the location of the next instruction in the code snippet

-CPU will continuously load instructions from the code segment of memory into the instruction queue of CPU according to the IP register, and then give it to the operation unit for execution.

Switching process

-each process is divided into code segments and data segments

-to point to the address space of different processes, there are four 16-bit segment registers, namely CS, DS, SS and ES

CS (Code Segment Register) is a code snippet register through which you can find the location of the code in memory.

DS (Data Segment Register) is a segment register through which you can find the location of the data in memory.

SS (Stack Segment Register) is a stack register, and all operations related to function calls are closely related to the stack.

-A calls BMIT B calls C

-when A calls B, the logic of the B function is executed, so the relevant information about A's operation will be push to the stack.

-when B calls C, similarly, the information related to the operation of B will be push to the stack before running the logic of the C function.

-when C finishes running, the first thing that comes out of pop is BMIT B, and then the instruction after calling C function goes on.

After the B has finished running, then pop comes out and A Magi A continues to run until the end.

Load memory data

If you need to load the data in memory, you can find the data in memory through DS and load it into a general register

For a segment, there is a starting address, and the specific location within the segment is called the offset

Both CS and DS store the starting address of a segment

The offset of the snippet is placed in the IP register

The offset of the data segment is placed in the general register

CS and DS are both 16 bits (starting address), and IP registers and general purpose registers are both 16 bits (offset), but 8086 of the address bus is 20 bits.

Put together 20 bits: starting address

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.