understanding linux kernel 4th edition

1, operating system 1.1, the components of the Linux kernel (System):The kernel mainly consists of five parts: process scheduling, memory management, virtual file system, network interface and process communication. (1) Process schedulingProcess scheduling is the CPU's scheduling algorithm for CPU access to multiple processes, making the process macro-parallel ex

= Drv_name "... "; static struct Vortex_chip_info {...} Vortex_info_tbl[] _ _devinitdata = {{"3c590 vortex 10Mbps",... ... ...} static int _ _init vortex_init (void) {...} static void _ _exit Vortex_cleanup (void) {...} Module_init (Vortex_init); Module_exit (Vortex_cleanup);Macro-label-based optimizations:Linux uses a number of macros to set special properties on some functions and data structures. These grand many are defined inInclude/linux/init.h

level:Interrupt Processing Flow:1. Hardware notify pic via IRQ line2. pic puts the interrupt vector into the register3, the processor read registers, such as RAM. By interrupt handling routine processing (ISR)4. Notify picNote: Before the kernel is enabled for interrupts, you must know the correspondence between the IRQ number and the I/O device, otherwise the kernel does not know which device the vector c

[Classic] a thorough understanding of ioremap ing in Linux Kernel) Almost every type of peripherals is carried out by reading and writing registers on the device, usually including control registers, status registers, and data registers. The peripheral registers are usually continuously compiled. AccordingThe CPU architecture is different. The CPU can address th

invalidated before being uninstalled via Umount.2, Linux different, the same file system can be installed multiple times.3, no matter how many times a file system is installed, it has only one super Block object.4, the same installation point can install multiple file systems5, after the installation will be hidden first installed, when top layer top installation is removed, the following system has become availableFile System uninstallation: Umount

System call: The user state process to the kernel, to implement the user-state process calls the hardware Device function or interrupt;Advantages: Making programming easier, freeing users from learning the low-level programming features of hardware devices, improving system-to-security, and enabling the kernel to check for correctness before satisfying requests; improve portability;The difference between a

work on the limit and reduced the driver developer's work.(5) Why do I use the work queue? (The difference between work queue and soft interrupt)The deferred function we described above runs in the interrupt context (a checkpoint for soft interrupts is when DO_IRQ exits), causing some problems: soft interrupts cannot sleep and cannot be blocked. Because the interrupt context is out of the kernel state, there is no process switch, so if the soft inter

Process switch is an important function in preemptible multi-task OS. In essence, the OS Kernel suspends the running process, then, resume the previous suspended process B. Hardware context Each process has its own address space, but all processes share the CPU register physically. Therefore, before a process is resumed, the OS kernel must load the value of the Register when the process is suspended into th

introduced.Each processor of a multi-processor system has a separate hardware high-speed cache, which requires additional hardware circuits to synchronize cached content between different CPUs. As long as a CPU modifies its hardware cache, it must check whether the same data is included in other hardware caches. If yes, it must notify other CPUs to update them with appropriate values. This activity is called high-speed cache listening. These activities are handled by hardware, and the

contiguous and the corresponding data block on disk is also adjacent, then the general block layer can be merged with them, this way to produce a larger memory area is the physical segment.Common block Layer:1. The common block layer is a kernel component that handles requests from all block devices in the systemI\O Scheduler:1. Block device drivers can transmit one sector at a time, but the block i\o layer does not do so because it is time-consuming

The first week of Linux kernel design--from the assembly language to understand how the computer works Song Juning (20135315) I. The experimental process figure 1 writing songchenning5315.c fileFigure 2 Compiling a C file into a 32-bit machine languageFigure 3 Removing the link statement from the. s file to get the final assembly codeSecond, the analysis of stack changesSummarize and clarify the

transmission: The kernel forces the target process to respond to the signal by: A, change the execution status of the target process; b. Start a specific signal processing program; C, both.Suspend signal (pending signal):1, has been produced has not been transmitted to the signal;2. At any time a process has only one pending signal of a given type; the same other signals are not queued, simply discarded3, if the real-time signal, the same type of han

Source: http://blog.21ic.com/user1/4132/archives/2010/67653.html Kernel space and user space are the basis of the operating system theory, that is, the kernel function module runs in the kernel space, and the application program runs in the user space. Modern CPUs all have different operation modes, representing different levels. Different levels have different

transfer2. If the DMA and CPU access the same memory simultaneously, the conflict is resolved by the hardware circuit called the memory arbiter.3, the most used is the disk drive and other need to transfer a large number of bytes of equipment.4, the DMA setting time is longer, a small amount of data use CPU more efficient5. The synchronization DMA is triggered by the process, such as playing movies, music, when playing music, the memory data with DMA to the sound card playback6, asynchronous DM

the current stack top, switches the kernel stack, and saves the EIP (4) address space for the current process, explainingThe switching of the address space does not affect the execution of subsequent switch codes (5) The source location (6) of the change in the process represented by the current macro changes the information about the kernel stack in the Task Status section source location Experiment Summ

Question: What does the field mean in the segment descriptor?Information such as linear address/offset/granularity/privilege limit for the first address of the segment.1, quick access to segment descriptorsWhen the segment selector is loaded, the corresponding segment descriptor is automatically loaded into the corresponding non-programmed CPU register by memory. The GDT or LDT in main memory is no longer accessed after loading.2, Segment selector fieldIndex: Specifies the entry for the correspo

Chapter 1 Introduction.Introduction to operating systems, file systems, and Linux. Single-core and micro-kernel (VxWorks, the micro-kernel usually contains several synchronization primitives, a simple schedulingProgramCommunication Mechanism with processes. Several system processes running on the microkernel implement such OS-level functions as memory allocation,

item cacheRecycle a page box from an index node cacheCycle Recycling:1.Remove the program when the memory is low:1, although Pfra will try to keep a certain free page box, but there is a possibility of memory exhaustion, this time all processes can not be executed, there is no way to release the page box owned by the doorExchange:1. Exchange used to provide backup on disk for non-mapped pages2. Switching is one of the most advanced features of page frame recycling.3. The swap can be used to exp

Written in front: 1. Everything is a document; 2. virtual files in memory; privileged files1. Everything is a file, if we use the command line to view all groupings or open user management in the graphical interfaceYou will find that there are many groups of users, such as sudo groupsThey use a decentralized approach to ensure that the system runs stably while maintaining the root overhead. Complete various tasks2. Our older generation, Longlongage is very dangerous when looking at the

The parent process P->counter = p->priority; for the current process The time slice is a priority value p->signal = 0; Clear the signal bitmap p->alarm = 0; Empty alarm timing P->leader = 0; The leading position of the conversation cannot be the same as the parent process p->utime = P->stime = 0; Some statistical information zero, user state and kernel state time P->cu