The Master said, “The superior man is frank; the mean man is affectionate.” Analects of Confucius: A review

A hundred blog series. This paper is: the v22. Xx HongMeng kernel source code analysis (assembly) | where the CPU clock in

Hardware architecture:

  • V22. Xx HongMeng kernel source code analysis (assembly) | where the CPU clock in
  • V23. Xx HongMeng kernel source code analysis (assembly parameter) | how to transfer the parameters of the complex
  • V36. Xx HongMeng kernel source code analysis (operating mode) | CPU is inspired, seven wives
  • V38. Xx HongMeng kernel source code analysis (register) | jack Bauer is the busiest storage universe
  • Abnormal v39. Xx HongMeng kernel source code analysis (over) | society is very simple, complex people
  • V40. Xx HongMeng kernel source code analysis (assembly summary) | assembly and lovely as the girl next door
  • V42. Xx HongMeng kernel source code analysis (interrupt switch) | system dynamic due to interrupt
  • V43. Xx HongMeng kernel source code analysis (interrupt concept) | sea father-in-law’s daily work
  • V44. Xx HongMeng kernel source code analysis (interrupt management) | henceforth no longer afraid of interruption

This article through dismantling a very simple assembly code to quickly understand assembly, for reading hongmeng assembly foundation. The rest of this series will examine the hongmeng assembler one by one.

Assembly is simple

  • First: to identify assembly language must be simple, no deep things, nothing more than moving data around, runtime data mainly stay in two places: memory and registers. The register is the internal memory of the CPU, closest to the arithmetic unit, so it is the fastest.

  • The second: The space provided by the user program is called the user stack. The sensitive work of the CPU should be brought back to the company to do. The sensitive work is called the kernel stack. The essence of a system call is that the CPU needs to switch the working mode, i.e. switch the office space.

  • Third: the CPU is pipelined, it only recognizes instructions, and only goes to one place (to the code section of the PC register) to get the instructions to digest. Instruction sets tell the outside world what the CPU can do and provide conversational instructions. Assembly language is a common language that people and CPUS can communicate happily without stumbling. One by one corresponding CPU instructions, and can ensure that the human memory is not good to modular design idea, first look at a section of C compiled into assembly code and then modular.

Square (C -> Assembler)

// compiler: armv7-a clang (trunk)
/ / + + + + + + + + + + + + square (c - > assembly) + + + + + + + + + + + + + + + + + + + + + + + +
int square(intA,int b){
    return a*b;
}
square(int.int):
        sub     sp， sp， #8@ sp minus8To allocate stack space to square, only2STR r0, [sp, #4[sp] @ldr R1, [sp, #4[sp] @ fetch the first parameter to R1 LDR R2, [sp] @ fetch the second parameter to R2 mul r0, R1, r2 @ execute a*b to r0, return the work is always to r0 add sp, sp, #8The @ subroutine returns, equivalent to mov PC, lr, that is, jumps to the call pointCopy the code

Fp (C -> assembler)

/ / + + + + + + + + + + + + fp (c - > assembly) + + + + + + + + + + + + + + + + + + + + + + + +
int fp(int b)
{
    int a = 1;
    return square(a + b, a + b); }fp(int):
        push    {r11， lr}      @r11(fp)/ LR, save main mov r11, sp @r11 used to save sp value, function stack start position sub sp, sp, #8@ sp minus8To allocate stack space to FP, only2STR r0, [sp, #4] @ save the parameter value and put it in SP+4Mov r0, #1         @r0=1STR r0, [sp] @1[SP] @ give the value of SP to r0 LDR r1, [SP, #4] @ SP +4Add R1, r0, R1 @ execute R1 =a+b mov r0, R1 @r0= R1, pass the parameter bl with r0, R1square(int.intMov LR, PC, mov PCsquare(int.intPop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @Copy the code

Main (c -> assembler)

/ / + + + + + + + + + + + + the main (c - > assembly) + + + + + + + + + + + + + + + + + + + + + + + +
int main(a)
{
    int sum = 0;
    for(int a = 0; a <100; a++){
        sum = sum + fp(a);
    }
    return sum;
}
main:
        push    {r11， lr}      @r11(fp)/ LR, mov r11, sp @r11 used to save sp value, function stack start position sub sp, sp, #16@ sp minus16Allocate stack space to main4Stack space to complete the calculation mov R0, #0@ initialize r0 STR r0, [r11, #4 -] @ Execute sum =0}, p, p8] @sum will always occupy SP+8STR r0, [sp, #4@a will always occupy SP+4LBB1_1: @ loop start entry LDR r0, [sp, #4] @ fetch a value to r0 CMP r0, #99@ with99Compare BGT. LBB1_4@ greater than99Mov pc.lbb1_4b.LBB1_2 @ To continue the loop, directly mov PC.lbb1_2.LBB1_2: @ to meet the loop condition8R0, sp+8LDR r0, [sp, #] LDR r0, [sp, #4[@r0 is used to pass the parameter, fetching the value of A to r0 as the parameter bl of fpfp(intMov LR, PC, mov PCfp(int[sp] @ add r0, r0, r1 @ calculate the new SUM, r0 save STR r0, [sp, #8Save the new sum to SP+8Mov pc.lbb1_3.LBB1_3: @ to complete a++ operation entry4]   @SP+4The record is the value of a, and assign r0 add r0, r0, #1@ r0 increase1}, p, p4@ puts the new a value back into SP+4B.LBB1_1 @ jump to compare a <100At.lbb1_4: @ loop end entry LDR r0, [sp, #8Pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @pop {r11, lr} @ Jumping to LR is equivalent to MOV PC, LRCopy the code

The code is a bit long, all annotated, if you can directly understand so congratulations, hongmeng kernel based on the six assembler file also understand. This is the full view of the following C file

File picture

#include <stdio.h>
#include <math.h>

int square(intA,int b){
    return a*b;
}

int fp(int b)
{
    int a = 1;
    return square(a + b, a + b); }int main(a)
{
    int sum = 0;
    for(int a = 0; a <100; a++){
        sum = sum + fp(a);
    }
    return sum;
}
Copy the code

The code is simple and anyone can understand it, it’s typical, it’s representative, it has loops, it has judgments, it has operations, it has multilevel function calls. The structure of the compiled assembly code is not much different from that of C language. The difference is that the realization of the loop uses four modules, which are easy to understand: one is the start block (LBB1_1), a conditional processing block (LBB1_2), a conditional change block (LBB1_3), and the last end block (LBB1_4).

Block by block.

Let’s start with the shortest one

int square(intA,int b){
    return a*b;
}
/ / compiled into
square(int.int):
        sub     sp， sp， #8@ sp minus8To allocate stack space to square, only2STR r0, [sp, #4[sp] @ldr R1, [sp, #4[sp] @ fetch the first parameter to R1 LDR R2, [sp] @ fetch the second parameter to R2 mul r0, R1, r2 @ execute a*b to r0, return the work is always to r0 add sp, sp, #8The @ subroutine returns, equivalent to mov PC, lr, that is, jumps to the call pointCopy the code

Sub sp, sp, #8 = sp-8, sp = sp-8, sp = sp-8, sp = sp-8, sp = sp-8, sp = sp-8, sp = sp-8 The code and the hongmeng kernel stack way, are used to decrease the full stack (FD). What is decreasing full stack? Decrement means that the address at the bottom of the stack is higher than the address at the top of the stack, and the growth direction of the stack is decreasing. Full means that the SP pointer always points to the top of the stack. Each function has its own independent bottom and top of the stack, and the space between them is called the stack frame. Sub sp, sp, #8 = 2 stack space, each stack space is four bytes. Do you want to release it? Add sp sp #8 to free up stack space. [sp, #4] [sp, #4] [sp, #4] [sp, #4] [sp, #4] [sp, #4] [sp, #4] [sp, #4] Be sure to understand decrement full stack, that’s the key! Otherwise you can’t read kernel assembly code.

The participation way

Generally through the register (r0.. R10) passes the parameter to (r0.. r10)

Add r1, r0, r1 @ run r1=a+b mov r0, r@r0 =r1, pass the parameter bl with r0, R1square(int.intMov LR, PC, mov PCsquare(int.int) 
Copy the code

When in square, r0, R1, r0, r1, r0, r1, r0, r1, r0, R1, r0, R1

}, p, p4[sp] @ holds the first parameter STR R1, [sp] @ holds the second parameter LDR r1, [sp, #4[sp] @ : [sp] @ : [sp] @ : [sp] @ : [sp] @ : [sp] @ : [sp] @ : [sp] @ : [sp] @ : [spCopy the code

This is a process problem. The compiler saves parameters first, and it doesn’t matter how you want to use it. In addition, the return value is saved to R0 by default. Square gives the result of (a*b) to R0, and returns r0 to FP, which is the return value of square for FP.

Function calls to main and FP require other functions to be called, so both appear

Push {r11, lr}//....Pop {r11, lr}Copy the code

Fp saves the stack frame range and instruction position of main. Lr saves the position of instruction to which main function is executed. R11 points to the top of the stack. In this way, when FP returns to main after execution, the values of mov PC, LR and PC registers change, indicating that the code executed has changed, and the instructions and stack frames of main are returned to continue the unfinished work.

How to move the memory and register data?

Data stays primarily in two places: memory and registers. Register <-> register, memory <-> register, memory <-> memory handling instructions are different.

STR r1, [sp] @ Register -> memory4] @memory -> registerCopy the code

This is another pair, used between memory <-> registers, known as MOV R0, and R1 for register <-> registers

Ask three questions

First: What if it’s a mutable parameter? How to complete 100 parameters? There are only 12 parameters through the register

Second: Can there be more than one return value?

Third: can data be transported without passing through the CPU?

Intensive reading of the kernel source code

Four code stores synchronous annotation kernel source code, >> view the Gitee repository

Analysis of 100 blogs. Dig deep into the core

Add comments to hongmeng kernel source code process, sort out the following article. Content based on the source code, often in life scene analogy as much as possible into the kernel knowledge of a scene, with a pictorial sense, easy to understand memory. It’s important to speak in a way that others can understand! The 100 blogs are by no means a bunch of ridiculously difficult concepts being put forward by Baidu. That’s not interesting. More hope to make the kernel become lifelike, feel more intimate. It’s hard, it’s hard, but there’s no turning back. 😛 and code bugs need to be constantly debug, there will be many mistakes and omissions in the article and annotation content, please forgive, but will be repeatedly amended, continuous update. Xx represents the number of modifications, refined, concise and comprehensive, and strive to create high-quality content.

Compile build The fundamental tools Loading operation Process management
Compile environment

The build process

Environment script

Build tools

Designed.the gn application

Ninja ninja

 Two-way linked list

Bitmap management

In the stack way

The timer

Atomic operation

Time management

 The ELF format

The ELF parsing

Static link

relocation

Process image

 Process management

Process concept

Fork

Special process

Process recycling

Signal production

Signal consumption

Shell editor

Shell parsing
Process of communication Memory management Ins and outs Task management
spinlocks

The mutex

Process of communication

A semaphore

Incident control

The message queue

 Memory allocation

Memory management

Memory assembly

The memory mapping

Rules of memory

Physical memory

 Total directory

Scheduling the story

Main memory slave

The source code comments

Source structure

Static site

 The clock task

Task scheduling

Task management

The scheduling queue

Scheduling mechanism

Thread concept

Concurrent parallel

The system calls

Task switching
The file system Hardware architecture
File concept

The file system

The index node

Mount the directory

Root file system

Character device

VFS

File handle

Pipeline file

 Compilation basis

Assembly and the cords

Working mode

register

Anomaly over

Assembly summary

Interrupt switch

Interrupt concept

Interrupt management

HongMeng station | into a little bit every day, the original is not easy, welcome to reprint, please indicate the source.