Virtual Filesystem (VFS)

The kernel's filesystem abstraction layer

What is VFS?

The Virtual Filesystem Switch (VFS) is an abstraction layer that allows the kernel to support many different filesystems (ext4, btrfs, tmpfs, procfs, NFS...) through a common interface. When your program calls read(), it goes through VFS, which dispatches to the appropriate filesystem implementation.

VFS makes the following possible: - A single open(), read(), write() API works for ext4, NFS, /proc, /sys, pipes, and sockets - Files can be accessed across mounted filesystems transparently - Filesystems can be developed as kernel modules

The four VFS objects

VFS defines four core objects:

Object	Represents	Kernel struct
Superblock	A mounted filesystem instance	struct super_block
Inode	A file or directory	struct inode
Dentry	A path component (name → inode mapping)	struct dentry
File	An open file (open() creates one)	struct file

Each has an associated operations table (vtable) that the filesystem must implement:

Object	Operations struct
Superblock	struct super_operations
Inode	struct inode_operations
Dentry	struct dentry_operations
File	struct file_operations

Contents

Fundamentals

• VFS Objects: superblock, inode, dentry, file — The four core objects in detail
• Path Resolution — How /usr/bin/bash becomes an inode
• File Operations — The file_operations vtable and open/read/write
• Filesystem Registration and Mounting — How filesystems plug into VFS

Lifecycle

• Life of a write() syscall — From write() to disk
• Dentry and Inode Caches — Caching for performance