Linux Kernel Internals Linux 内核底层

Syscalls, VFS, memory, io_uring, eBPF, tracing — notes on how the kernel actually works under the hood. Drawn from mainline kernel sources and public documentation.

系统调用、VFS、内存管理、io_uring、eBPF、tracing——内核底层到底是怎么跑的。所有内容均来自主线内核源码和公开资料。

Posts in this topic · 本话题下的文章

  • io-uring Linux Next-Generation Asynchronous I/O InterfaceEN

    io_uring is an asynchronous I/O interface introduced in Linux 5.1 (May 2019), designed by Jens Axboe to replace the older aio and POSIX AIO interfaces. This post covers what problems it was built t...

  • Linux内存管理深度解析(一):从文件系统缓存到现代内存优化技术中文

    在Linux操作系统中,内存管理是系统性能和稳定性的基石。随着硬件技术的发展和应用场景的日益复杂,Linux内核的内存管理机制也在不断演进。本文将基于对Linux内存管理核心概念的深入理解,结合最新的技术进展和互联网信息,为您呈现一份全面且更新的Linux内存管理指南。我们将从文件系统缓存和匿名页的交换机制入手,逐步探讨页面回收、脏页回写等关键环节,并引入zRAM、Zswap等现代内存优化技...

  • fspy (1): Trace Filesystem Activity and I/O Patterns in Your Linux ApplicationsEN

    Linux has good tools for tracing syscalls (strace) and profiling CPU, but watching filesystem activity at the file level — which files an application opens, how often, how much data — is surprising...

  • Challenges of Deploying eBPF-Based Tracing in Embedded Systems, and Alternatives in Embedded Platforms (Libtracefs/libtraceevent)EN

    Challenges of Deploying eBPF-Based Tracing in Embedded Systems, and Alternatives in Embedded Platforms (Libtracefs/libtraceevent)Presentation OverviewThis page summarizes the key insights from my p...

  • Printing Stack Traces in the Linux KernelEN

    When something goes wrong in the kernel, the first thing I usually want to know is how the code got there. A stack trace tells you that. This post is my notes on the few ways to print one, and what...

  • Shared Memory in Linux Inter-Process CommunicationEN

    Shared Memory in Linux Inter-Process Communication (IPC)In Linux, shared memory is an efficient method for inter-process communication (IPC). Through shared memory, multiple processes can access th...

  • Linux System Calls: An In-Depth LookEN

    What are System Calls?System calls (syscalls) are the fundamental interface through which user-space applications interact with the Linux kernel. They are the gateway for processes to request privi...

  • Linux VFS架构、数据结构与文件系统注册中文

    Linux VFS深度解析:架构、数据结构与文件系统注册1. VFS概述Virtual File System(虚拟文件系统,VFS)是Linux内核中的一个软件层,它为用户空间程序提供了统一的文件系统接口。VFS的核心作用是抽象底层文件系统的差异,使得应用程序可以使用相同的系统调用(如open、read、write)来操作不同类型的文件系统(ext4、XFS、Btrfs等)。1.1 VFS...

  • Linux 设备管理全解析:从 devfs 到 udev/sysfs 的演进中文

    Linux 设备管理全解析:从 devfs 到 udev/sysfs 的演进引言在 Linux 系统中,设备管理是操作系统的核心功能之一。遵循”一切皆文件”的 Unix 哲学,Linux 将硬件设备抽象为文件,存放在 /dev 目录下供用户进程操作。然而,设备管理机制经历了从静态到动态、从内核空间到用户空间的重大演进。本文将深入探讨 Linux 设备管理的发展历程,重点解析 devfs、tm...

  • Heap and Stack, one Minimalist heap AllocatorEN

    Introduction to the HeapWhat is the Heap?The heap is a region of memory used for dynamic memory allocation — memory that is allocated and freed at runtime rather than at compile time.In C, you acce...