1af96c1e3STobin C. Harding.. SPDX-License-Identifier: GPL-2.0 2af96c1e3STobin C. Harding 3af96c1e3STobin C. Harding========================================= 4af96c1e3STobin C. HardingOverview of the Linux Virtual File System 5af96c1e3STobin C. Harding========================================= 6af96c1e3STobin C. Harding 7af96c1e3STobin C. HardingOriginal author: Richard Gooch <rgooch@atnf.csiro.au> 8af96c1e3STobin C. Harding 9af96c1e3STobin C. Harding- Copyright (C) 1999 Richard Gooch 10af96c1e3STobin C. Harding- Copyright (C) 2005 Pekka Enberg 11af96c1e3STobin C. Harding 12af96c1e3STobin C. Harding 13af96c1e3STobin C. HardingIntroduction 14af96c1e3STobin C. Harding============ 15af96c1e3STobin C. Harding 16af96c1e3STobin C. HardingThe Virtual File System (also known as the Virtual Filesystem Switch) is 17af96c1e3STobin C. Hardingthe software layer in the kernel that provides the filesystem interface 18af96c1e3STobin C. Hardingto userspace programs. It also provides an abstraction within the 19af96c1e3STobin C. Hardingkernel which allows different filesystem implementations to coexist. 20af96c1e3STobin C. Harding 21af96c1e3STobin C. HardingVFS system calls open(2), stat(2), read(2), write(2), chmod(2) and so on 22af96c1e3STobin C. Hardingare called from a process context. Filesystem locking is described in 23ec23eb54SMauro Carvalho Chehabthe document Documentation/filesystems/locking.rst. 24af96c1e3STobin C. Harding 25af96c1e3STobin C. Harding 26af96c1e3STobin C. HardingDirectory Entry Cache (dcache) 27af96c1e3STobin C. Harding------------------------------ 28af96c1e3STobin C. Harding 29af96c1e3STobin C. HardingThe VFS implements the open(2), stat(2), chmod(2), and similar system 30af96c1e3STobin C. Hardingcalls. The pathname argument that is passed to them is used by the VFS 31af96c1e3STobin C. Hardingto search through the directory entry cache (also known as the dentry 32af96c1e3STobin C. Hardingcache or dcache). This provides a very fast look-up mechanism to 33af96c1e3STobin C. Hardingtranslate a pathname (filename) into a specific dentry. Dentries live 34af96c1e3STobin C. Hardingin RAM and are never saved to disc: they exist only for performance. 35af96c1e3STobin C. Harding 36af96c1e3STobin C. HardingThe dentry cache is meant to be a view into your entire filespace. As 37af96c1e3STobin C. Hardingmost computers cannot fit all dentries in the RAM at the same time, some 38af96c1e3STobin C. Hardingbits of the cache are missing. In order to resolve your pathname into a 39af96c1e3STobin C. Hardingdentry, the VFS may have to resort to creating dentries along the way, 40af96c1e3STobin C. Hardingand then loading the inode. This is done by looking up the inode. 41af96c1e3STobin C. Harding 42af96c1e3STobin C. Harding 43af96c1e3STobin C. HardingThe Inode Object 44af96c1e3STobin C. Harding---------------- 45af96c1e3STobin C. Harding 46af96c1e3STobin C. HardingAn individual dentry usually has a pointer to an inode. Inodes are 47af96c1e3STobin C. Hardingfilesystem objects such as regular files, directories, FIFOs and other 48af96c1e3STobin C. Hardingbeasts. They live either on the disc (for block device filesystems) or 49af96c1e3STobin C. Hardingin the memory (for pseudo filesystems). Inodes that live on the disc 50af96c1e3STobin C. Hardingare copied into the memory when required and changes to the inode are 51af96c1e3STobin C. Hardingwritten back to disc. A single inode can be pointed to by multiple 52af96c1e3STobin C. Hardingdentries (hard links, for example, do this). 53af96c1e3STobin C. Harding 54af96c1e3STobin C. HardingTo look up an inode requires that the VFS calls the lookup() method of 55af96c1e3STobin C. Hardingthe parent directory inode. This method is installed by the specific 56af96c1e3STobin C. Hardingfilesystem implementation that the inode lives in. Once the VFS has the 57af96c1e3STobin C. Hardingrequired dentry (and hence the inode), we can do all those boring things 58af96c1e3STobin C. Hardinglike open(2) the file, or stat(2) it to peek at the inode data. The 59af96c1e3STobin C. Hardingstat(2) operation is fairly simple: once the VFS has the dentry, it 60af96c1e3STobin C. Hardingpeeks at the inode data and passes some of it back to userspace. 61af96c1e3STobin C. Harding 62af96c1e3STobin C. Harding 63af96c1e3STobin C. HardingThe File Object 64af96c1e3STobin C. Harding--------------- 65af96c1e3STobin C. Harding 66af96c1e3STobin C. HardingOpening a file requires another operation: allocation of a file 67af96c1e3STobin C. Hardingstructure (this is the kernel-side implementation of file descriptors). 68af96c1e3STobin C. HardingThe freshly allocated file structure is initialized with a pointer to 69af96c1e3STobin C. Hardingthe dentry and a set of file operation member functions. These are 70af96c1e3STobin C. Hardingtaken from the inode data. The open() file method is then called so the 71af96c1e3STobin C. Hardingspecific filesystem implementation can do its work. You can see that 72af96c1e3STobin C. Hardingthis is another switch performed by the VFS. The file structure is 73af96c1e3STobin C. Hardingplaced into the file descriptor table for the process. 74af96c1e3STobin C. Harding 75af96c1e3STobin C. HardingReading, writing and closing files (and other assorted VFS operations) 76af96c1e3STobin C. Hardingis done by using the userspace file descriptor to grab the appropriate 77af96c1e3STobin C. Hardingfile structure, and then calling the required file structure method to 78af96c1e3STobin C. Hardingdo whatever is required. For as long as the file is open, it keeps the 79af96c1e3STobin C. Hardingdentry in use, which in turn means that the VFS inode is still in use. 80af96c1e3STobin C. Harding 81af96c1e3STobin C. Harding 82af96c1e3STobin C. HardingRegistering and Mounting a Filesystem 83af96c1e3STobin C. Harding===================================== 84af96c1e3STobin C. Harding 85af96c1e3STobin C. HardingTo register and unregister a filesystem, use the following API 86af96c1e3STobin C. Hardingfunctions: 87af96c1e3STobin C. Harding 88af96c1e3STobin C. Harding.. code-block:: c 89af96c1e3STobin C. Harding 90af96c1e3STobin C. Harding #include <linux/fs.h> 91af96c1e3STobin C. Harding 92af96c1e3STobin C. Harding extern int register_filesystem(struct file_system_type *); 93af96c1e3STobin C. Harding extern int unregister_filesystem(struct file_system_type *); 94af96c1e3STobin C. Harding 95af96c1e3STobin C. HardingThe passed struct file_system_type describes your filesystem. When a 96af96c1e3STobin C. Hardingrequest is made to mount a filesystem onto a directory in your 97af96c1e3STobin C. Hardingnamespace, the VFS will call the appropriate mount() method for the 98af96c1e3STobin C. Hardingspecific filesystem. New vfsmount referring to the tree returned by 99af96c1e3STobin C. Harding->mount() will be attached to the mountpoint, so that when pathname 100af96c1e3STobin C. Hardingresolution reaches the mountpoint it will jump into the root of that 101af96c1e3STobin C. Hardingvfsmount. 102af96c1e3STobin C. Harding 103af96c1e3STobin C. HardingYou can see all filesystems that are registered to the kernel in the 104af96c1e3STobin C. Hardingfile /proc/filesystems. 105af96c1e3STobin C. Harding 106af96c1e3STobin C. Harding 107af96c1e3STobin C. Hardingstruct file_system_type 108af96c1e3STobin C. Harding----------------------- 109af96c1e3STobin C. Harding 110af96c1e3STobin C. HardingThis describes the filesystem. As of kernel 2.6.39, the following 111af96c1e3STobin C. Hardingmembers are defined: 112af96c1e3STobin C. Harding 113af96c1e3STobin C. Harding.. code-block:: c 114af96c1e3STobin C. Harding 1156a2195a1SLiao Pingfang struct file_system_type { 116af96c1e3STobin C. Harding const char *name; 117af96c1e3STobin C. Harding int fs_flags; 118af96c1e3STobin C. Harding struct dentry *(*mount) (struct file_system_type *, int, 119af96c1e3STobin C. Harding const char *, void *); 120af96c1e3STobin C. Harding void (*kill_sb) (struct super_block *); 121af96c1e3STobin C. Harding struct module *owner; 122af96c1e3STobin C. Harding struct file_system_type * next; 123af96c1e3STobin C. Harding struct list_head fs_supers; 124af96c1e3STobin C. Harding struct lock_class_key s_lock_key; 125af96c1e3STobin C. Harding struct lock_class_key s_umount_key; 126af96c1e3STobin C. Harding }; 127af96c1e3STobin C. Harding 128ee5dc049STobin C. Harding``name`` 129ee5dc049STobin C. Harding the name of the filesystem type, such as "ext2", "iso9660", 130af96c1e3STobin C. Harding "msdos" and so on 131af96c1e3STobin C. Harding 132ee5dc049STobin C. Harding``fs_flags`` 133ee5dc049STobin C. Harding various flags (i.e. FS_REQUIRES_DEV, FS_NO_DCACHE, etc.) 134af96c1e3STobin C. Harding 135ee5dc049STobin C. Harding``mount`` 136ee5dc049STobin C. Harding the method to call when a new instance of this filesystem should 137af96c1e3STobin C. Harding be mounted 138af96c1e3STobin C. Harding 139ee5dc049STobin C. Harding``kill_sb`` 140ee5dc049STobin C. Harding the method to call when an instance of this filesystem should be 141ee5dc049STobin C. Harding shut down 142af96c1e3STobin C. Harding 143af96c1e3STobin C. Harding 144ee5dc049STobin C. Harding``owner`` 145ee5dc049STobin C. Harding for internal VFS use: you should initialize this to THIS_MODULE 146ee5dc049STobin C. Harding in most cases. 147ee5dc049STobin C. Harding 148ee5dc049STobin C. Harding``next`` 149ee5dc049STobin C. Harding for internal VFS use: you should initialize this to NULL 150af96c1e3STobin C. Harding 151af96c1e3STobin C. Harding s_lock_key, s_umount_key: lockdep-specific 152af96c1e3STobin C. Harding 153af96c1e3STobin C. HardingThe mount() method has the following arguments: 154af96c1e3STobin C. Harding 155ee5dc049STobin C. Harding``struct file_system_type *fs_type`` 156ee5dc049STobin C. Harding describes the filesystem, partly initialized by the specific 157ee5dc049STobin C. Harding filesystem code 158af96c1e3STobin C. Harding 159ee5dc049STobin C. Harding``int flags`` 160ee5dc049STobin C. Harding mount flags 161af96c1e3STobin C. Harding 162ee5dc049STobin C. Harding``const char *dev_name`` 163ee5dc049STobin C. Harding the device name we are mounting. 164af96c1e3STobin C. Harding 165ee5dc049STobin C. Harding``void *data`` 166ee5dc049STobin C. Harding arbitrary mount options, usually comes as an ASCII string (see 167ee5dc049STobin C. Harding "Mount Options" section) 168af96c1e3STobin C. Harding 169af96c1e3STobin C. HardingThe mount() method must return the root dentry of the tree requested by 170af96c1e3STobin C. Hardingcaller. An active reference to its superblock must be grabbed and the 171af96c1e3STobin C. Hardingsuperblock must be locked. On failure it should return ERR_PTR(error). 172af96c1e3STobin C. Harding 173af96c1e3STobin C. HardingThe arguments match those of mount(2) and their interpretation depends 174af96c1e3STobin C. Hardingon filesystem type. E.g. for block filesystems, dev_name is interpreted 175af96c1e3STobin C. Hardingas block device name, that device is opened and if it contains a 176af96c1e3STobin C. Hardingsuitable filesystem image the method creates and initializes struct 177af96c1e3STobin C. Hardingsuper_block accordingly, returning its root dentry to caller. 178af96c1e3STobin C. Harding 179af96c1e3STobin C. Harding->mount() may choose to return a subtree of existing filesystem - it 180af96c1e3STobin C. Hardingdoesn't have to create a new one. The main result from the caller's 181af96c1e3STobin C. Hardingpoint of view is a reference to dentry at the root of (sub)tree to be 182af96c1e3STobin C. Hardingattached; creation of new superblock is a common side effect. 183af96c1e3STobin C. Harding 184af96c1e3STobin C. HardingThe most interesting member of the superblock structure that the mount() 185af96c1e3STobin C. Hardingmethod fills in is the "s_op" field. This is a pointer to a "struct 186af96c1e3STobin C. Hardingsuper_operations" which describes the next level of the filesystem 187af96c1e3STobin C. Hardingimplementation. 188af96c1e3STobin C. Harding 189af96c1e3STobin C. HardingUsually, a filesystem uses one of the generic mount() implementations 190af96c1e3STobin C. Hardingand provides a fill_super() callback instead. The generic variants are: 191af96c1e3STobin C. Harding 192ee5dc049STobin C. Harding``mount_bdev`` 193ee5dc049STobin C. Harding mount a filesystem residing on a block device 194af96c1e3STobin C. Harding 195ee5dc049STobin C. Harding``mount_nodev`` 196ee5dc049STobin C. Harding mount a filesystem that is not backed by a device 197af96c1e3STobin C. Harding 198ee5dc049STobin C. Harding``mount_single`` 199ee5dc049STobin C. Harding mount a filesystem which shares the instance between all mounts 200af96c1e3STobin C. Harding 201af96c1e3STobin C. HardingA fill_super() callback implementation has the following arguments: 202af96c1e3STobin C. Harding 203ee5dc049STobin C. Harding``struct super_block *sb`` 204ee5dc049STobin C. Harding the superblock structure. The callback must initialize this 205ee5dc049STobin C. Harding properly. 206af96c1e3STobin C. Harding 207ee5dc049STobin C. Harding``void *data`` 208ee5dc049STobin C. Harding arbitrary mount options, usually comes as an ASCII string (see 209ee5dc049STobin C. Harding "Mount Options" section) 210af96c1e3STobin C. Harding 211ee5dc049STobin C. Harding``int silent`` 212ee5dc049STobin C. Harding whether or not to be silent on error 213af96c1e3STobin C. Harding 214af96c1e3STobin C. Harding 215af96c1e3STobin C. HardingThe Superblock Object 216af96c1e3STobin C. Harding===================== 217af96c1e3STobin C. Harding 218af96c1e3STobin C. HardingA superblock object represents a mounted filesystem. 219af96c1e3STobin C. Harding 220af96c1e3STobin C. Harding 221af96c1e3STobin C. Hardingstruct super_operations 222af96c1e3STobin C. Harding----------------------- 223af96c1e3STobin C. Harding 224af96c1e3STobin C. HardingThis describes how the VFS can manipulate the superblock of your 225af96c1e3STobin C. Hardingfilesystem. As of kernel 2.6.22, the following members are defined: 226af96c1e3STobin C. Harding 227af96c1e3STobin C. Harding.. code-block:: c 228af96c1e3STobin C. Harding 229af96c1e3STobin C. Harding struct super_operations { 230af96c1e3STobin C. Harding struct inode *(*alloc_inode)(struct super_block *sb); 231af96c1e3STobin C. Harding void (*destroy_inode)(struct inode *); 232af96c1e3STobin C. Harding 233af96c1e3STobin C. Harding void (*dirty_inode) (struct inode *, int flags); 234af96c1e3STobin C. Harding int (*write_inode) (struct inode *, int); 235af96c1e3STobin C. Harding void (*drop_inode) (struct inode *); 236af96c1e3STobin C. Harding void (*delete_inode) (struct inode *); 237af96c1e3STobin C. Harding void (*put_super) (struct super_block *); 238af96c1e3STobin C. Harding int (*sync_fs)(struct super_block *sb, int wait); 239af96c1e3STobin C. Harding int (*freeze_fs) (struct super_block *); 240af96c1e3STobin C. Harding int (*unfreeze_fs) (struct super_block *); 241af96c1e3STobin C. Harding int (*statfs) (struct dentry *, struct kstatfs *); 242af96c1e3STobin C. Harding int (*remount_fs) (struct super_block *, int *, char *); 243af96c1e3STobin C. Harding void (*clear_inode) (struct inode *); 244af96c1e3STobin C. Harding void (*umount_begin) (struct super_block *); 245af96c1e3STobin C. Harding 246af96c1e3STobin C. Harding int (*show_options)(struct seq_file *, struct dentry *); 247af96c1e3STobin C. Harding 248af96c1e3STobin C. Harding ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t); 249af96c1e3STobin C. Harding ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); 250af96c1e3STobin C. Harding int (*nr_cached_objects)(struct super_block *); 251af96c1e3STobin C. Harding void (*free_cached_objects)(struct super_block *, int); 252af96c1e3STobin C. Harding }; 253af96c1e3STobin C. Harding 254af96c1e3STobin C. HardingAll methods are called without any locks being held, unless otherwise 255af96c1e3STobin C. Hardingnoted. This means that most methods can block safely. All methods are 256af96c1e3STobin C. Hardingonly called from a process context (i.e. not from an interrupt handler 257af96c1e3STobin C. Hardingor bottom half). 258af96c1e3STobin C. Harding 259ee5dc049STobin C. Harding``alloc_inode`` 260ee5dc049STobin C. Harding this method is called by alloc_inode() to allocate memory for 261ee5dc049STobin C. Harding struct inode and initialize it. If this function is not 262af96c1e3STobin C. Harding defined, a simple 'struct inode' is allocated. Normally 263af96c1e3STobin C. Harding alloc_inode will be used to allocate a larger structure which 264af96c1e3STobin C. Harding contains a 'struct inode' embedded within it. 265af96c1e3STobin C. Harding 266ee5dc049STobin C. Harding``destroy_inode`` 267ee5dc049STobin C. Harding this method is called by destroy_inode() to release resources 268ee5dc049STobin C. Harding allocated for struct inode. It is only required if 269af96c1e3STobin C. Harding ->alloc_inode was defined and simply undoes anything done by 270af96c1e3STobin C. Harding ->alloc_inode. 271af96c1e3STobin C. Harding 272ee5dc049STobin C. Harding``dirty_inode`` 273a38ed483SEric Biggers this method is called by the VFS when an inode is marked dirty. 274a38ed483SEric Biggers This is specifically for the inode itself being marked dirty, 275a38ed483SEric Biggers not its data. If the update needs to be persisted by fdatasync(), 276a38ed483SEric Biggers then I_DIRTY_DATASYNC will be set in the flags argument. 277af96c1e3STobin C. Harding 278ee5dc049STobin C. Harding``write_inode`` 279ee5dc049STobin C. Harding this method is called when the VFS needs to write an inode to 280ee5dc049STobin C. Harding disc. The second parameter indicates whether the write should 281ee5dc049STobin C. Harding be synchronous or not, not all filesystems check this flag. 282af96c1e3STobin C. Harding 283ee5dc049STobin C. Harding``drop_inode`` 284ee5dc049STobin C. Harding called when the last access to the inode is dropped, with the 285ee5dc049STobin C. Harding inode->i_lock spinlock held. 286af96c1e3STobin C. Harding 287af96c1e3STobin C. Harding This method should be either NULL (normal UNIX filesystem 288ee5dc049STobin C. Harding semantics) or "generic_delete_inode" (for filesystems that do 289ee5dc049STobin C. Harding not want to cache inodes - causing "delete_inode" to always be 290af96c1e3STobin C. Harding called regardless of the value of i_nlink) 291af96c1e3STobin C. Harding 292ee5dc049STobin C. Harding The "generic_delete_inode()" behavior is equivalent to the old 293ee5dc049STobin C. Harding practice of using "force_delete" in the put_inode() case, but 294ee5dc049STobin C. Harding does not have the races that the "force_delete()" approach had. 295af96c1e3STobin C. Harding 296ee5dc049STobin C. Harding``delete_inode`` 297ee5dc049STobin C. Harding called when the VFS wants to delete an inode 298af96c1e3STobin C. Harding 299ee5dc049STobin C. Harding``put_super`` 300ee5dc049STobin C. Harding called when the VFS wishes to free the superblock 301af96c1e3STobin C. Harding (i.e. unmount). This is called with the superblock lock held 302af96c1e3STobin C. Harding 303ee5dc049STobin C. Harding``sync_fs`` 304ee5dc049STobin C. Harding called when VFS is writing out all dirty data associated with a 305ee5dc049STobin C. Harding superblock. The second parameter indicates whether the method 306af96c1e3STobin C. Harding should wait until the write out has been completed. Optional. 307af96c1e3STobin C. Harding 308ee5dc049STobin C. Harding``freeze_fs`` 309ee5dc049STobin C. Harding called when VFS is locking a filesystem and forcing it into a 310ee5dc049STobin C. Harding consistent state. This method is currently used by the Logical 311ee5dc049STobin C. Harding Volume Manager (LVM). 312af96c1e3STobin C. Harding 313ee5dc049STobin C. Harding``unfreeze_fs`` 314ee5dc049STobin C. Harding called when VFS is unlocking a filesystem and making it writable 315af96c1e3STobin C. Harding again. 316af96c1e3STobin C. Harding 317ee5dc049STobin C. Harding``statfs`` 318ee5dc049STobin C. Harding called when the VFS needs to get filesystem statistics. 319af96c1e3STobin C. Harding 320ee5dc049STobin C. Harding``remount_fs`` 321ee5dc049STobin C. Harding called when the filesystem is remounted. This is called with 322ee5dc049STobin C. Harding the kernel lock held 323af96c1e3STobin C. Harding 324ee5dc049STobin C. Harding``clear_inode`` 325ee5dc049STobin C. Harding called then the VFS clears the inode. Optional 326af96c1e3STobin C. Harding 327ee5dc049STobin C. Harding``umount_begin`` 328ee5dc049STobin C. Harding called when the VFS is unmounting a filesystem. 329af96c1e3STobin C. Harding 330ee5dc049STobin C. Harding``show_options`` 331ee5dc049STobin C. Harding called by the VFS to show mount options for /proc/<pid>/mounts. 332ee5dc049STobin C. Harding (see "Mount Options" section) 333af96c1e3STobin C. Harding 334ee5dc049STobin C. Harding``quota_read`` 335ee5dc049STobin C. Harding called by the VFS to read from filesystem quota file. 336af96c1e3STobin C. Harding 337ee5dc049STobin C. Harding``quota_write`` 338ee5dc049STobin C. Harding called by the VFS to write to filesystem quota file. 339af96c1e3STobin C. Harding 340ee5dc049STobin C. Harding``nr_cached_objects`` 341ee5dc049STobin C. Harding called by the sb cache shrinking function for the filesystem to 342ee5dc049STobin C. Harding return the number of freeable cached objects it contains. 343af96c1e3STobin C. Harding Optional. 344af96c1e3STobin C. Harding 345ee5dc049STobin C. Harding``free_cache_objects`` 346ee5dc049STobin C. Harding called by the sb cache shrinking function for the filesystem to 347ee5dc049STobin C. Harding scan the number of objects indicated to try to free them. 348ee5dc049STobin C. Harding Optional, but any filesystem implementing this method needs to 349ee5dc049STobin C. Harding also implement ->nr_cached_objects for it to be called 350ee5dc049STobin C. Harding correctly. 351af96c1e3STobin C. Harding 352af96c1e3STobin C. Harding We can't do anything with any errors that the filesystem might 353ee5dc049STobin C. Harding encountered, hence the void return type. This will never be 354ee5dc049STobin C. Harding called if the VM is trying to reclaim under GFP_NOFS conditions, 355ee5dc049STobin C. Harding hence this method does not need to handle that situation itself. 356af96c1e3STobin C. Harding 357ee5dc049STobin C. Harding Implementations must include conditional reschedule calls inside 358ee5dc049STobin C. Harding any scanning loop that is done. This allows the VFS to 359ee5dc049STobin C. Harding determine appropriate scan batch sizes without having to worry 360ee5dc049STobin C. Harding about whether implementations will cause holdoff problems due to 361ee5dc049STobin C. Harding large scan batch sizes. 362af96c1e3STobin C. Harding 363af96c1e3STobin C. HardingWhoever sets up the inode is responsible for filling in the "i_op" 364af96c1e3STobin C. Hardingfield. This is a pointer to a "struct inode_operations" which describes 365af96c1e3STobin C. Hardingthe methods that can be performed on individual inodes. 366af96c1e3STobin C. Harding 367af96c1e3STobin C. Harding 368af96c1e3STobin C. Hardingstruct xattr_handlers 369af96c1e3STobin C. Harding--------------------- 370af96c1e3STobin C. Harding 371af96c1e3STobin C. HardingOn filesystems that support extended attributes (xattrs), the s_xattr 372af96c1e3STobin C. Hardingsuperblock field points to a NULL-terminated array of xattr handlers. 373af96c1e3STobin C. HardingExtended attributes are name:value pairs. 374af96c1e3STobin C. Harding 375ee5dc049STobin C. Harding``name`` 376ee5dc049STobin C. Harding Indicates that the handler matches attributes with the specified 377ee5dc049STobin C. Harding name (such as "system.posix_acl_access"); the prefix field must 378ee5dc049STobin C. Harding be NULL. 379af96c1e3STobin C. Harding 380ee5dc049STobin C. Harding``prefix`` 381ee5dc049STobin C. Harding Indicates that the handler matches all attributes with the 382ee5dc049STobin C. Harding specified name prefix (such as "user."); the name field must be 383ee5dc049STobin C. Harding NULL. 384af96c1e3STobin C. Harding 385ee5dc049STobin C. Harding``list`` 386ee5dc049STobin C. Harding Determine if attributes matching this xattr handler should be 387ee5dc049STobin C. Harding listed for a particular dentry. Used by some listxattr 388ee5dc049STobin C. Harding implementations like generic_listxattr. 389af96c1e3STobin C. Harding 390ee5dc049STobin C. Harding``get`` 391ee5dc049STobin C. Harding Called by the VFS to get the value of a particular extended 392ee5dc049STobin C. Harding attribute. This method is called by the getxattr(2) system 393ee5dc049STobin C. Harding call. 394af96c1e3STobin C. Harding 395ee5dc049STobin C. Harding``set`` 396ee5dc049STobin C. Harding Called by the VFS to set the value of a particular extended 397ee5dc049STobin C. Harding attribute. When the new value is NULL, called to remove a 3988286de7cSRandy Dunlap particular extended attribute. This method is called by the 399ee5dc049STobin C. Harding setxattr(2) and removexattr(2) system calls. 400af96c1e3STobin C. Harding 401af96c1e3STobin C. HardingWhen none of the xattr handlers of a filesystem match the specified 402af96c1e3STobin C. Hardingattribute name or when a filesystem doesn't support extended attributes, 403af96c1e3STobin C. Hardingthe various ``*xattr(2)`` system calls return -EOPNOTSUPP. 404af96c1e3STobin C. Harding 405af96c1e3STobin C. Harding 406af96c1e3STobin C. HardingThe Inode Object 407af96c1e3STobin C. Harding================ 408af96c1e3STobin C. Harding 409af96c1e3STobin C. HardingAn inode object represents an object within the filesystem. 410af96c1e3STobin C. Harding 411af96c1e3STobin C. Harding 412af96c1e3STobin C. Hardingstruct inode_operations 413af96c1e3STobin C. Harding----------------------- 414af96c1e3STobin C. Harding 415af96c1e3STobin C. HardingThis describes how the VFS can manipulate an inode in your filesystem. 416af96c1e3STobin C. HardingAs of kernel 2.6.22, the following members are defined: 417af96c1e3STobin C. Harding 418af96c1e3STobin C. Harding.. code-block:: c 419af96c1e3STobin C. Harding 420af96c1e3STobin C. Harding struct inode_operations { 421549c7297SChristian Brauner int (*create) (struct user_namespace *, struct inode *,struct dentry *, umode_t, bool); 422af96c1e3STobin C. Harding struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int); 423af96c1e3STobin C. Harding int (*link) (struct dentry *,struct inode *,struct dentry *); 424af96c1e3STobin C. Harding int (*unlink) (struct inode *,struct dentry *); 425549c7297SChristian Brauner int (*symlink) (struct user_namespace *, struct inode *,struct dentry *,const char *); 426549c7297SChristian Brauner int (*mkdir) (struct user_namespace *, struct inode *,struct dentry *,umode_t); 427af96c1e3STobin C. Harding int (*rmdir) (struct inode *,struct dentry *); 428549c7297SChristian Brauner int (*mknod) (struct user_namespace *, struct inode *,struct dentry *,umode_t,dev_t); 429549c7297SChristian Brauner int (*rename) (struct user_namespace *, struct inode *, struct dentry *, 430af96c1e3STobin C. Harding struct inode *, struct dentry *, unsigned int); 431af96c1e3STobin C. Harding int (*readlink) (struct dentry *, char __user *,int); 432af96c1e3STobin C. Harding const char *(*get_link) (struct dentry *, struct inode *, 433af96c1e3STobin C. Harding struct delayed_call *); 434549c7297SChristian Brauner int (*permission) (struct user_namespace *, struct inode *, int); 435af96c1e3STobin C. Harding int (*get_acl)(struct inode *, int); 436549c7297SChristian Brauner int (*setattr) (struct user_namespace *, struct dentry *, struct iattr *); 437549c7297SChristian Brauner int (*getattr) (struct user_namespace *, const struct path *, struct kstat *, u32, unsigned int); 438af96c1e3STobin C. Harding ssize_t (*listxattr) (struct dentry *, char *, size_t); 439af96c1e3STobin C. Harding void (*update_time)(struct inode *, struct timespec *, int); 440af96c1e3STobin C. Harding int (*atomic_open)(struct inode *, struct dentry *, struct file *, 441af96c1e3STobin C. Harding unsigned open_flag, umode_t create_mode); 442549c7297SChristian Brauner int (*tmpfile) (struct user_namespace *, struct inode *, struct dentry *, umode_t); 443549c7297SChristian Brauner int (*set_acl)(struct user_namespace *, struct inode *, struct posix_acl *, int); 444*4c5b4799SMiklos Szeredi int (*fileattr_set)(struct user_namespace *mnt_userns, 445*4c5b4799SMiklos Szeredi struct dentry *dentry, struct fileattr *fa); 446*4c5b4799SMiklos Szeredi int (*fileattr_get)(struct dentry *dentry, struct fileattr *fa); 447af96c1e3STobin C. Harding }; 448af96c1e3STobin C. Harding 449af96c1e3STobin C. HardingAgain, all methods are called without any locks being held, unless 450af96c1e3STobin C. Hardingotherwise noted. 451af96c1e3STobin C. Harding 452ee5dc049STobin C. Harding``create`` 453ee5dc049STobin C. Harding called by the open(2) and creat(2) system calls. Only required 454ee5dc049STobin C. Harding if you want to support regular files. The dentry you get should 455ee5dc049STobin C. Harding not have an inode (i.e. it should be a negative dentry). Here 456ee5dc049STobin C. Harding you will probably call d_instantiate() with the dentry and the 457ee5dc049STobin C. Harding newly created inode 458af96c1e3STobin C. Harding 459ee5dc049STobin C. Harding``lookup`` 460ee5dc049STobin C. Harding called when the VFS needs to look up an inode in a parent 461af96c1e3STobin C. Harding directory. The name to look for is found in the dentry. This 462af96c1e3STobin C. Harding method must call d_add() to insert the found inode into the 463af96c1e3STobin C. Harding dentry. The "i_count" field in the inode structure should be 464af96c1e3STobin C. Harding incremented. If the named inode does not exist a NULL inode 465af96c1e3STobin C. Harding should be inserted into the dentry (this is called a negative 466ee5dc049STobin C. Harding dentry). Returning an error code from this routine must only be 467ee5dc049STobin C. Harding done on a real error, otherwise creating inodes with system 468af96c1e3STobin C. Harding calls like create(2), mknod(2), mkdir(2) and so on will fail. 469af96c1e3STobin C. Harding If you wish to overload the dentry methods then you should 470ee5dc049STobin C. Harding initialise the "d_dop" field in the dentry; this is a pointer to 471ee5dc049STobin C. Harding a struct "dentry_operations". This method is called with the 472ee5dc049STobin C. Harding directory inode semaphore held 473af96c1e3STobin C. Harding 474ee5dc049STobin C. Harding``link`` 475ee5dc049STobin C. Harding called by the link(2) system call. Only required if you want to 476ee5dc049STobin C. Harding support hard links. You will probably need to call 477af96c1e3STobin C. Harding d_instantiate() just as you would in the create() method 478af96c1e3STobin C. Harding 479ee5dc049STobin C. Harding``unlink`` 480ee5dc049STobin C. Harding called by the unlink(2) system call. Only required if you want 481ee5dc049STobin C. Harding to support deleting inodes 482af96c1e3STobin C. Harding 483ee5dc049STobin C. Harding``symlink`` 484ee5dc049STobin C. Harding called by the symlink(2) system call. Only required if you want 485ee5dc049STobin C. Harding to support symlinks. You will probably need to call 486af96c1e3STobin C. Harding d_instantiate() just as you would in the create() method 487af96c1e3STobin C. Harding 488ee5dc049STobin C. Harding``mkdir`` 489ee5dc049STobin C. Harding called by the mkdir(2) system call. Only required if you want 490af96c1e3STobin C. Harding to support creating subdirectories. You will probably need to 491af96c1e3STobin C. Harding call d_instantiate() just as you would in the create() method 492af96c1e3STobin C. Harding 493ee5dc049STobin C. Harding``rmdir`` 494ee5dc049STobin C. Harding called by the rmdir(2) system call. Only required if you want 495af96c1e3STobin C. Harding to support deleting subdirectories 496af96c1e3STobin C. Harding 497ee5dc049STobin C. Harding``mknod`` 498ee5dc049STobin C. Harding called by the mknod(2) system call to create a device (char, 499ee5dc049STobin C. Harding block) inode or a named pipe (FIFO) or socket. Only required if 500ee5dc049STobin C. Harding you want to support creating these types of inodes. You will 501ee5dc049STobin C. Harding probably need to call d_instantiate() just as you would in the 502ee5dc049STobin C. Harding create() method 503af96c1e3STobin C. Harding 504ee5dc049STobin C. Harding``rename`` 505ee5dc049STobin C. Harding called by the rename(2) system call to rename the object to have 506ee5dc049STobin C. Harding the parent and name given by the second inode and dentry. 507af96c1e3STobin C. Harding 508af96c1e3STobin C. Harding The filesystem must return -EINVAL for any unsupported or 509af96c1e3STobin C. Harding unknown flags. Currently the following flags are implemented: 510ee5dc049STobin C. Harding (1) RENAME_NOREPLACE: this flag indicates that if the target of 511ee5dc049STobin C. Harding the rename exists the rename should fail with -EEXIST instead of 512ee5dc049STobin C. Harding replacing the target. The VFS already checks for existence, so 513ee5dc049STobin C. Harding for local filesystems the RENAME_NOREPLACE implementation is 514ee5dc049STobin C. Harding equivalent to plain rename. 515af96c1e3STobin C. Harding (2) RENAME_EXCHANGE: exchange source and target. Both must 516ee5dc049STobin C. Harding exist; this is checked by the VFS. Unlike plain rename, source 517ee5dc049STobin C. Harding and target may be of different type. 518af96c1e3STobin C. Harding 519ee5dc049STobin C. Harding``get_link`` 520ee5dc049STobin C. Harding called by the VFS to follow a symbolic link to the inode it 521ee5dc049STobin C. Harding points to. Only required if you want to support symbolic links. 522ee5dc049STobin C. Harding This method returns the symlink body to traverse (and possibly 523ee5dc049STobin C. Harding resets the current position with nd_jump_link()). If the body 524ee5dc049STobin C. Harding won't go away until the inode is gone, nothing else is needed; 525ee5dc049STobin C. Harding if it needs to be otherwise pinned, arrange for its release by 526ee5dc049STobin C. Harding having get_link(..., ..., done) do set_delayed_call(done, 527ee5dc049STobin C. Harding destructor, argument). In that case destructor(argument) will 528ee5dc049STobin C. Harding be called once VFS is done with the body you've returned. May 529ee5dc049STobin C. Harding be called in RCU mode; that is indicated by NULL dentry 530af96c1e3STobin C. Harding argument. If request can't be handled without leaving RCU mode, 531af96c1e3STobin C. Harding have it return ERR_PTR(-ECHILD). 532af96c1e3STobin C. Harding 533af96c1e3STobin C. Harding If the filesystem stores the symlink target in ->i_link, the 534af96c1e3STobin C. Harding VFS may use it directly without calling ->get_link(); however, 535af96c1e3STobin C. Harding ->get_link() must still be provided. ->i_link must not be 536af96c1e3STobin C. Harding freed until after an RCU grace period. Writing to ->i_link 537af96c1e3STobin C. Harding post-iget() time requires a 'release' memory barrier. 538af96c1e3STobin C. Harding 539ee5dc049STobin C. Harding``readlink`` 540ee5dc049STobin C. Harding this is now just an override for use by readlink(2) for the 541af96c1e3STobin C. Harding cases when ->get_link uses nd_jump_link() or object is not in 542af96c1e3STobin C. Harding fact a symlink. Normally filesystems should only implement 543af96c1e3STobin C. Harding ->get_link for symlinks and readlink(2) will automatically use 544af96c1e3STobin C. Harding that. 545af96c1e3STobin C. Harding 546ee5dc049STobin C. Harding``permission`` 547ee5dc049STobin C. Harding called by the VFS to check for access rights on a POSIX-like 548af96c1e3STobin C. Harding filesystem. 549af96c1e3STobin C. Harding 550ee5dc049STobin C. Harding May be called in rcu-walk mode (mask & MAY_NOT_BLOCK). If in 551ee5dc049STobin C. Harding rcu-walk mode, the filesystem must check the permission without 552ee5dc049STobin C. Harding blocking or storing to the inode. 553af96c1e3STobin C. Harding 554ee5dc049STobin C. Harding If a situation is encountered that rcu-walk cannot handle, 555ee5dc049STobin C. Harding return 556af96c1e3STobin C. Harding -ECHILD and it will be called again in ref-walk mode. 557af96c1e3STobin C. Harding 558ee5dc049STobin C. Harding``setattr`` 559ee5dc049STobin C. Harding called by the VFS to set attributes for a file. This method is 560ee5dc049STobin C. Harding called by chmod(2) and related system calls. 561af96c1e3STobin C. Harding 562ee5dc049STobin C. Harding``getattr`` 563ee5dc049STobin C. Harding called by the VFS to get attributes of a file. This method is 564ee5dc049STobin C. Harding called by stat(2) and related system calls. 565af96c1e3STobin C. Harding 566ee5dc049STobin C. Harding``listxattr`` 567ee5dc049STobin C. Harding called by the VFS to list all extended attributes for a given 568ee5dc049STobin C. Harding file. This method is called by the listxattr(2) system call. 569af96c1e3STobin C. Harding 570ee5dc049STobin C. Harding``update_time`` 571ee5dc049STobin C. Harding called by the VFS to update a specific time or the i_version of 572ee5dc049STobin C. Harding an inode. If this is not defined the VFS will update the inode 573ee5dc049STobin C. Harding itself and call mark_inode_dirty_sync. 574af96c1e3STobin C. Harding 575ee5dc049STobin C. Harding``atomic_open`` 576ee5dc049STobin C. Harding called on the last component of an open. Using this optional 577ee5dc049STobin C. Harding method the filesystem can look up, possibly create and open the 578ee5dc049STobin C. Harding file in one atomic operation. If it wants to leave actual 579ee5dc049STobin C. Harding opening to the caller (e.g. if the file turned out to be a 580ee5dc049STobin C. Harding symlink, device, or just something filesystem won't do atomic 581ee5dc049STobin C. Harding open for), it may signal this by returning finish_no_open(file, 582ee5dc049STobin C. Harding dentry). This method is only called if the last component is 583ee5dc049STobin C. Harding negative or needs lookup. Cached positive dentries are still 584ee5dc049STobin C. Harding handled by f_op->open(). If the file was created, FMODE_CREATED 585ee5dc049STobin C. Harding flag should be set in file->f_mode. In case of O_EXCL the 586ee5dc049STobin C. Harding method must only succeed if the file didn't exist and hence 587ee5dc049STobin C. Harding FMODE_CREATED shall always be set on success. 588af96c1e3STobin C. Harding 589ee5dc049STobin C. Harding``tmpfile`` 590ee5dc049STobin C. Harding called in the end of O_TMPFILE open(). Optional, equivalent to 591ee5dc049STobin C. Harding atomically creating, opening and unlinking a file in given 592ee5dc049STobin C. Harding directory. 593af96c1e3STobin C. Harding 594*4c5b4799SMiklos Szeredi``fileattr_get`` 595*4c5b4799SMiklos Szeredi called on ioctl(FS_IOC_GETFLAGS) and ioctl(FS_IOC_FSGETXATTR) to 596*4c5b4799SMiklos Szeredi retrieve miscellaneous file flags and attributes. Also called 597*4c5b4799SMiklos Szeredi before the relevant SET operation to check what is being changed 598*4c5b4799SMiklos Szeredi (in this case with i_rwsem locked exclusive). If unset, then 599*4c5b4799SMiklos Szeredi fall back to f_op->ioctl(). 600*4c5b4799SMiklos Szeredi 601*4c5b4799SMiklos Szeredi``fileattr_set`` 602*4c5b4799SMiklos Szeredi called on ioctl(FS_IOC_SETFLAGS) and ioctl(FS_IOC_FSSETXATTR) to 603*4c5b4799SMiklos Szeredi change miscellaneous file flags and attributes. Callers hold 604*4c5b4799SMiklos Szeredi i_rwsem exclusive. If unset, then fall back to f_op->ioctl(). 605*4c5b4799SMiklos Szeredi 606af96c1e3STobin C. Harding 607af96c1e3STobin C. HardingThe Address Space Object 608af96c1e3STobin C. Harding======================== 609af96c1e3STobin C. Harding 610af96c1e3STobin C. HardingThe address space object is used to group and manage pages in the page 611af96c1e3STobin C. Hardingcache. It can be used to keep track of the pages in a file (or anything 612af96c1e3STobin C. Hardingelse) and also track the mapping of sections of the file into process 613af96c1e3STobin C. Hardingaddress spaces. 614af96c1e3STobin C. Harding 615af96c1e3STobin C. HardingThere are a number of distinct yet related services that an 616af96c1e3STobin C. Hardingaddress-space can provide. These include communicating memory pressure, 617af96c1e3STobin C. Hardingpage lookup by address, and keeping track of pages tagged as Dirty or 618af96c1e3STobin C. HardingWriteback. 619af96c1e3STobin C. Harding 620af96c1e3STobin C. HardingThe first can be used independently to the others. The VM can try to 621af96c1e3STobin C. Hardingeither write dirty pages in order to clean them, or release clean pages 622af96c1e3STobin C. Hardingin order to reuse them. To do this it can call the ->writepage method 623af96c1e3STobin C. Hardingon dirty pages, and ->releasepage on clean pages with PagePrivate set. 624af96c1e3STobin C. HardingClean pages without PagePrivate and with no external references will be 625af96c1e3STobin C. Hardingreleased without notice being given to the address_space. 626af96c1e3STobin C. Harding 627af96c1e3STobin C. HardingTo achieve this functionality, pages need to be placed on an LRU with 628af96c1e3STobin C. Hardinglru_cache_add and mark_page_active needs to be called whenever the page 629af96c1e3STobin C. Hardingis used. 630af96c1e3STobin C. Harding 631af96c1e3STobin C. HardingPages are normally kept in a radix tree index by ->index. This tree 632af96c1e3STobin C. Hardingmaintains information about the PG_Dirty and PG_Writeback status of each 633af96c1e3STobin C. Hardingpage, so that pages with either of these flags can be found quickly. 634af96c1e3STobin C. Harding 635af96c1e3STobin C. HardingThe Dirty tag is primarily used by mpage_writepages - the default 636af96c1e3STobin C. Harding->writepages method. It uses the tag to find dirty pages to call 637af96c1e3STobin C. Harding->writepage on. If mpage_writepages is not used (i.e. the address 638af96c1e3STobin C. Hardingprovides its own ->writepages) , the PAGECACHE_TAG_DIRTY tag is almost 639af96c1e3STobin C. Hardingunused. write_inode_now and sync_inode do use it (through 640af96c1e3STobin C. Harding__sync_single_inode) to check if ->writepages has been successful in 641af96c1e3STobin C. Hardingwriting out the whole address_space. 642af96c1e3STobin C. Harding 643af96c1e3STobin C. HardingThe Writeback tag is used by filemap*wait* and sync_page* functions, via 644af96c1e3STobin C. Hardingfilemap_fdatawait_range, to wait for all writeback to complete. 645af96c1e3STobin C. Harding 646af96c1e3STobin C. HardingAn address_space handler may attach extra information to a page, 647af96c1e3STobin C. Hardingtypically using the 'private' field in the 'struct page'. If such 648af96c1e3STobin C. Hardinginformation is attached, the PG_Private flag should be set. This will 649af96c1e3STobin C. Hardingcause various VM routines to make extra calls into the address_space 650af96c1e3STobin C. Hardinghandler to deal with that data. 651af96c1e3STobin C. Harding 652af96c1e3STobin C. HardingAn address space acts as an intermediate between storage and 653af96c1e3STobin C. Hardingapplication. Data is read into the address space a whole page at a 654af96c1e3STobin C. Hardingtime, and provided to the application either by copying of the page, or 655af96c1e3STobin C. Hardingby memory-mapping the page. Data is written into the address space by 656af96c1e3STobin C. Hardingthe application, and then written-back to storage typically in whole 657af96c1e3STobin C. Hardingpages, however the address_space has finer control of write sizes. 658af96c1e3STobin C. Harding 659af96c1e3STobin C. HardingThe read process essentially only requires 'readpage'. The write 660af96c1e3STobin C. Hardingprocess is more complicated and uses write_begin/write_end or 661af96c1e3STobin C. Hardingset_page_dirty to write data into the address_space, and writepage and 662af96c1e3STobin C. Hardingwritepages to writeback data to storage. 663af96c1e3STobin C. Harding 664af96c1e3STobin C. HardingAdding and removing pages to/from an address_space is protected by the 665af96c1e3STobin C. Hardinginode's i_mutex. 666af96c1e3STobin C. Harding 667af96c1e3STobin C. HardingWhen data is written to a page, the PG_Dirty flag should be set. It 668af96c1e3STobin C. Hardingtypically remains set until writepage asks for it to be written. This 669af96c1e3STobin C. Hardingshould clear PG_Dirty and set PG_Writeback. It can be actually written 670af96c1e3STobin C. Hardingat any point after PG_Dirty is clear. Once it is known to be safe, 671af96c1e3STobin C. HardingPG_Writeback is cleared. 672af96c1e3STobin C. Harding 673af96c1e3STobin C. HardingWriteback makes use of a writeback_control structure to direct the 6748286de7cSRandy Dunlapoperations. This gives the writepage and writepages operations some 675af96c1e3STobin C. Hardinginformation about the nature of and reason for the writeback request, 676af96c1e3STobin C. Hardingand the constraints under which it is being done. It is also used to 677af96c1e3STobin C. Hardingreturn information back to the caller about the result of a writepage or 678af96c1e3STobin C. Hardingwritepages request. 679af96c1e3STobin C. Harding 680af96c1e3STobin C. Harding 681af96c1e3STobin C. HardingHandling errors during writeback 682af96c1e3STobin C. Harding-------------------------------- 683af96c1e3STobin C. Harding 684af96c1e3STobin C. HardingMost applications that do buffered I/O will periodically call a file 685af96c1e3STobin C. Hardingsynchronization call (fsync, fdatasync, msync or sync_file_range) to 686af96c1e3STobin C. Hardingensure that data written has made it to the backing store. When there 687af96c1e3STobin C. Hardingis an error during writeback, they expect that error to be reported when 688af96c1e3STobin C. Hardinga file sync request is made. After an error has been reported on one 689af96c1e3STobin C. Hardingrequest, subsequent requests on the same file descriptor should return 690af96c1e3STobin C. Harding0, unless further writeback errors have occurred since the previous file 691af96c1e3STobin C. Hardingsyncronization. 692af96c1e3STobin C. Harding 693af96c1e3STobin C. HardingIdeally, the kernel would report errors only on file descriptions on 694af96c1e3STobin C. Hardingwhich writes were done that subsequently failed to be written back. The 695af96c1e3STobin C. Hardinggeneric pagecache infrastructure does not track the file descriptions 696af96c1e3STobin C. Hardingthat have dirtied each individual page however, so determining which 697af96c1e3STobin C. Hardingfile descriptors should get back an error is not possible. 698af96c1e3STobin C. Harding 699af96c1e3STobin C. HardingInstead, the generic writeback error tracking infrastructure in the 700af96c1e3STobin C. Hardingkernel settles for reporting errors to fsync on all file descriptions 701af96c1e3STobin C. Hardingthat were open at the time that the error occurred. In a situation with 702af96c1e3STobin C. Hardingmultiple writers, all of them will get back an error on a subsequent 703af96c1e3STobin C. Hardingfsync, even if all of the writes done through that particular file 704af96c1e3STobin C. Hardingdescriptor succeeded (or even if there were no writes on that file 705af96c1e3STobin C. Hardingdescriptor at all). 706af96c1e3STobin C. Harding 707af96c1e3STobin C. HardingFilesystems that wish to use this infrastructure should call 708af96c1e3STobin C. Hardingmapping_set_error to record the error in the address_space when it 709af96c1e3STobin C. Hardingoccurs. Then, after writing back data from the pagecache in their 710af96c1e3STobin C. Hardingfile->fsync operation, they should call file_check_and_advance_wb_err to 711af96c1e3STobin C. Hardingensure that the struct file's error cursor has advanced to the correct 712af96c1e3STobin C. Hardingpoint in the stream of errors emitted by the backing device(s). 713af96c1e3STobin C. Harding 714af96c1e3STobin C. Harding 715af96c1e3STobin C. Hardingstruct address_space_operations 716af96c1e3STobin C. Harding------------------------------- 717af96c1e3STobin C. Harding 718af96c1e3STobin C. HardingThis describes how the VFS can manipulate mapping of a file to page 719af96c1e3STobin C. Hardingcache in your filesystem. The following members are defined: 720af96c1e3STobin C. Harding 721af96c1e3STobin C. Harding.. code-block:: c 722af96c1e3STobin C. Harding 723af96c1e3STobin C. Harding struct address_space_operations { 724af96c1e3STobin C. Harding int (*writepage)(struct page *page, struct writeback_control *wbc); 725af96c1e3STobin C. Harding int (*readpage)(struct file *, struct page *); 726af96c1e3STobin C. Harding int (*writepages)(struct address_space *, struct writeback_control *); 727af96c1e3STobin C. Harding int (*set_page_dirty)(struct page *page); 7288151b4c8SMatthew Wilcox (Oracle) void (*readahead)(struct readahead_control *); 729af96c1e3STobin C. Harding int (*readpages)(struct file *filp, struct address_space *mapping, 730af96c1e3STobin C. Harding struct list_head *pages, unsigned nr_pages); 731af96c1e3STobin C. Harding int (*write_begin)(struct file *, struct address_space *mapping, 732af96c1e3STobin C. Harding loff_t pos, unsigned len, unsigned flags, 733af96c1e3STobin C. Harding struct page **pagep, void **fsdata); 734af96c1e3STobin C. Harding int (*write_end)(struct file *, struct address_space *mapping, 735af96c1e3STobin C. Harding loff_t pos, unsigned len, unsigned copied, 736af96c1e3STobin C. Harding struct page *page, void *fsdata); 737af96c1e3STobin C. Harding sector_t (*bmap)(struct address_space *, sector_t); 738af96c1e3STobin C. Harding void (*invalidatepage) (struct page *, unsigned int, unsigned int); 739af96c1e3STobin C. Harding int (*releasepage) (struct page *, int); 740af96c1e3STobin C. Harding void (*freepage)(struct page *); 741af96c1e3STobin C. Harding ssize_t (*direct_IO)(struct kiocb *, struct iov_iter *iter); 742af96c1e3STobin C. Harding /* isolate a page for migration */ 743af96c1e3STobin C. Harding bool (*isolate_page) (struct page *, isolate_mode_t); 744af96c1e3STobin C. Harding /* migrate the contents of a page to the specified target */ 745af96c1e3STobin C. Harding int (*migratepage) (struct page *, struct page *); 746af96c1e3STobin C. Harding /* put migration-failed page back to right list */ 747af96c1e3STobin C. Harding void (*putback_page) (struct page *); 748af96c1e3STobin C. Harding int (*launder_page) (struct page *); 749af96c1e3STobin C. Harding 750af96c1e3STobin C. Harding int (*is_partially_uptodate) (struct page *, unsigned long, 751af96c1e3STobin C. Harding unsigned long); 752af96c1e3STobin C. Harding void (*is_dirty_writeback) (struct page *, bool *, bool *); 753af96c1e3STobin C. Harding int (*error_remove_page) (struct mapping *mapping, struct page *page); 754af96c1e3STobin C. Harding int (*swap_activate)(struct file *); 755af96c1e3STobin C. Harding int (*swap_deactivate)(struct file *); 756af96c1e3STobin C. Harding }; 757af96c1e3STobin C. Harding 758ee5dc049STobin C. Harding``writepage`` 759ee5dc049STobin C. Harding called by the VM to write a dirty page to backing store. This 760ee5dc049STobin C. Harding may happen for data integrity reasons (i.e. 'sync'), or to free 761ee5dc049STobin C. Harding up memory (flush). The difference can be seen in 762ee5dc049STobin C. Harding wbc->sync_mode. The PG_Dirty flag has been cleared and 763ee5dc049STobin C. Harding PageLocked is true. writepage should start writeout, should set 764ee5dc049STobin C. Harding PG_Writeback, and should make sure the page is unlocked, either 765ee5dc049STobin C. Harding synchronously or asynchronously when the write operation 766ee5dc049STobin C. Harding completes. 767af96c1e3STobin C. Harding 768af96c1e3STobin C. Harding If wbc->sync_mode is WB_SYNC_NONE, ->writepage doesn't have to 769af96c1e3STobin C. Harding try too hard if there are problems, and may choose to write out 770af96c1e3STobin C. Harding other pages from the mapping if that is easier (e.g. due to 771af96c1e3STobin C. Harding internal dependencies). If it chooses not to start writeout, it 772ee5dc049STobin C. Harding should return AOP_WRITEPAGE_ACTIVATE so that the VM will not 773ee5dc049STobin C. Harding keep calling ->writepage on that page. 774af96c1e3STobin C. Harding 775af96c1e3STobin C. Harding See the file "Locking" for more details. 776af96c1e3STobin C. Harding 777ee5dc049STobin C. Harding``readpage`` 778ee5dc049STobin C. Harding called by the VM to read a page from backing store. The page 779ee5dc049STobin C. Harding will be Locked when readpage is called, and should be unlocked 780ee5dc049STobin C. Harding and marked uptodate once the read completes. If ->readpage 781ee5dc049STobin C. Harding discovers that it needs to unlock the page for some reason, it 782ee5dc049STobin C. Harding can do so, and then return AOP_TRUNCATED_PAGE. In this case, 783ee5dc049STobin C. Harding the page will be relocated, relocked and if that all succeeds, 784ee5dc049STobin C. Harding ->readpage will be called again. 785af96c1e3STobin C. Harding 786ee5dc049STobin C. Harding``writepages`` 787ee5dc049STobin C. Harding called by the VM to write out pages associated with the 788e9b2f15bSJulia Lawall address_space object. If wbc->sync_mode is WB_SYNC_ALL, then 789af96c1e3STobin C. Harding the writeback_control will specify a range of pages that must be 790e9b2f15bSJulia Lawall written out. If it is WB_SYNC_NONE, then a nr_to_write is 791ee5dc049STobin C. Harding given and that many pages should be written if possible. If no 792ee5dc049STobin C. Harding ->writepages is given, then mpage_writepages is used instead. 793ee5dc049STobin C. Harding This will choose pages from the address space that are tagged as 794ee5dc049STobin C. Harding DIRTY and will pass them to ->writepage. 795af96c1e3STobin C. Harding 796ee5dc049STobin C. Harding``set_page_dirty`` 797ee5dc049STobin C. Harding called by the VM to set a page dirty. This is particularly 798ee5dc049STobin C. Harding needed if an address space attaches private data to a page, and 799ee5dc049STobin C. Harding that data needs to be updated when a page is dirtied. This is 800ee5dc049STobin C. Harding called, for example, when a memory mapped page gets modified. 801af96c1e3STobin C. Harding If defined, it should set the PageDirty flag, and the 802af96c1e3STobin C. Harding PAGECACHE_TAG_DIRTY tag in the radix tree. 803af96c1e3STobin C. Harding 8048151b4c8SMatthew Wilcox (Oracle)``readahead`` 8058151b4c8SMatthew Wilcox (Oracle) Called by the VM to read pages associated with the address_space 8068151b4c8SMatthew Wilcox (Oracle) object. The pages are consecutive in the page cache and are 8078151b4c8SMatthew Wilcox (Oracle) locked. The implementation should decrement the page refcount 8088151b4c8SMatthew Wilcox (Oracle) after starting I/O on each page. Usually the page will be 8098151b4c8SMatthew Wilcox (Oracle) unlocked by the I/O completion handler. If the filesystem decides 8108151b4c8SMatthew Wilcox (Oracle) to stop attempting I/O before reaching the end of the readahead 8118151b4c8SMatthew Wilcox (Oracle) window, it can simply return. The caller will decrement the page 8128151b4c8SMatthew Wilcox (Oracle) refcount and unlock the remaining pages for you. Set PageUptodate 8138151b4c8SMatthew Wilcox (Oracle) if the I/O completes successfully. Setting PageError on any page 8148151b4c8SMatthew Wilcox (Oracle) will be ignored; simply unlock the page if an I/O error occurs. 8158151b4c8SMatthew Wilcox (Oracle) 816ee5dc049STobin C. Harding``readpages`` 817ee5dc049STobin C. Harding called by the VM to read pages associated with the address_space 818ee5dc049STobin C. Harding object. This is essentially just a vector version of readpage. 819ee5dc049STobin C. Harding Instead of just one page, several pages are requested. 820af96c1e3STobin C. Harding readpages is only used for read-ahead, so read errors are 821af96c1e3STobin C. Harding ignored. If anything goes wrong, feel free to give up. 8228151b4c8SMatthew Wilcox (Oracle) This interface is deprecated and will be removed by the end of 8238151b4c8SMatthew Wilcox (Oracle) 2020; implement readahead instead. 824af96c1e3STobin C. Harding 825ee5dc049STobin C. Harding``write_begin`` 826ee5dc049STobin C. Harding Called by the generic buffered write code to ask the filesystem 827ee5dc049STobin C. Harding to prepare to write len bytes at the given offset in the file. 828ee5dc049STobin C. Harding The address_space should check that the write will be able to 829ee5dc049STobin C. Harding complete, by allocating space if necessary and doing any other 830ee5dc049STobin C. Harding internal housekeeping. If the write will update parts of any 831ee5dc049STobin C. Harding basic-blocks on storage, then those blocks should be pre-read 832ee5dc049STobin C. Harding (if they haven't been read already) so that the updated blocks 833ee5dc049STobin C. Harding can be written out properly. 834af96c1e3STobin C. Harding 835ee5dc049STobin C. Harding The filesystem must return the locked pagecache page for the 836ee5dc049STobin C. Harding specified offset, in ``*pagep``, for the caller to write into. 837af96c1e3STobin C. Harding 838ee5dc049STobin C. Harding It must be able to cope with short writes (where the length 839ee5dc049STobin C. Harding passed to write_begin is greater than the number of bytes copied 840ee5dc049STobin C. Harding into the page). 841af96c1e3STobin C. Harding 842af96c1e3STobin C. Harding flags is a field for AOP_FLAG_xxx flags, described in 843af96c1e3STobin C. Harding include/linux/fs.h. 844af96c1e3STobin C. Harding 845af96c1e3STobin C. Harding A void * may be returned in fsdata, which then gets passed into 846af96c1e3STobin C. Harding write_end. 847af96c1e3STobin C. Harding 848ee5dc049STobin C. Harding Returns 0 on success; < 0 on failure (which is the error code), 849ee5dc049STobin C. Harding in which case write_end is not called. 850af96c1e3STobin C. Harding 851ee5dc049STobin C. Harding``write_end`` 852ee5dc049STobin C. Harding After a successful write_begin, and data copy, write_end must be 853ee5dc049STobin C. Harding called. len is the original len passed to write_begin, and 854ee5dc049STobin C. Harding copied is the amount that was able to be copied. 855af96c1e3STobin C. Harding 856ee5dc049STobin C. Harding The filesystem must take care of unlocking the page and 857ee5dc049STobin C. Harding releasing it refcount, and updating i_size. 858af96c1e3STobin C. Harding 859ee5dc049STobin C. Harding Returns < 0 on failure, otherwise the number of bytes (<= 860ee5dc049STobin C. Harding 'copied') that were able to be copied into pagecache. 861af96c1e3STobin C. Harding 862ee5dc049STobin C. Harding``bmap`` 863ee5dc049STobin C. Harding called by the VFS to map a logical block offset within object to 864ee5dc049STobin C. Harding physical block number. This method is used by the FIBMAP ioctl 865ee5dc049STobin C. Harding and for working with swap-files. To be able to swap to a file, 866ee5dc049STobin C. Harding the file must have a stable mapping to a block device. The swap 867ee5dc049STobin C. Harding system does not go through the filesystem but instead uses bmap 868ee5dc049STobin C. Harding to find out where the blocks in the file are and uses those 869ee5dc049STobin C. Harding addresses directly. 870af96c1e3STobin C. Harding 871ee5dc049STobin C. Harding``invalidatepage`` 872ee5dc049STobin C. Harding If a page has PagePrivate set, then invalidatepage will be 873ee5dc049STobin C. Harding called when part or all of the page is to be removed from the 874ee5dc049STobin C. Harding address space. This generally corresponds to either a 875af96c1e3STobin C. Harding truncation, punch hole or a complete invalidation of the address 876af96c1e3STobin C. Harding space (in the latter case 'offset' will always be 0 and 'length' 877af96c1e3STobin C. Harding will be PAGE_SIZE). Any private data associated with the page 878ee5dc049STobin C. Harding should be updated to reflect this truncation. If offset is 0 879ee5dc049STobin C. Harding and length is PAGE_SIZE, then the private data should be 880ee5dc049STobin C. Harding released, because the page must be able to be completely 881ee5dc049STobin C. Harding discarded. This may be done by calling the ->releasepage 882ee5dc049STobin C. Harding function, but in this case the release MUST succeed. 883af96c1e3STobin C. Harding 884ee5dc049STobin C. Harding``releasepage`` 885ee5dc049STobin C. Harding releasepage is called on PagePrivate pages to indicate that the 886ee5dc049STobin C. Harding page should be freed if possible. ->releasepage should remove 887ee5dc049STobin C. Harding any private data from the page and clear the PagePrivate flag. 888ee5dc049STobin C. Harding If releasepage() fails for some reason, it must indicate failure 889ee5dc049STobin C. Harding with a 0 return value. releasepage() is used in two distinct 890ee5dc049STobin C. Harding though related cases. The first is when the VM finds a clean 891ee5dc049STobin C. Harding page with no active users and wants to make it a free page. If 892ee5dc049STobin C. Harding ->releasepage succeeds, the page will be removed from the 893ee5dc049STobin C. Harding address_space and become free. 894af96c1e3STobin C. Harding 895af96c1e3STobin C. Harding The second case is when a request has been made to invalidate 896ee5dc049STobin C. Harding some or all pages in an address_space. This can happen through 897ee5dc049STobin C. Harding the fadvise(POSIX_FADV_DONTNEED) system call or by the 898ee5dc049STobin C. Harding filesystem explicitly requesting it as nfs and 9fs do (when they 899ee5dc049STobin C. Harding believe the cache may be out of date with storage) by calling 900ee5dc049STobin C. Harding invalidate_inode_pages2(). If the filesystem makes such a call, 901ee5dc049STobin C. Harding and needs to be certain that all pages are invalidated, then its 902ee5dc049STobin C. Harding releasepage will need to ensure this. Possibly it can clear the 903ee5dc049STobin C. Harding PageUptodate bit if it cannot free private data yet. 904af96c1e3STobin C. Harding 905ee5dc049STobin C. Harding``freepage`` 906ee5dc049STobin C. Harding freepage is called once the page is no longer visible in the 907ee5dc049STobin C. Harding page cache in order to allow the cleanup of any private data. 908ee5dc049STobin C. Harding Since it may be called by the memory reclaimer, it should not 909ee5dc049STobin C. Harding assume that the original address_space mapping still exists, and 910ee5dc049STobin C. Harding it should not block. 911af96c1e3STobin C. Harding 912ee5dc049STobin C. Harding``direct_IO`` 913ee5dc049STobin C. Harding called by the generic read/write routines to perform direct_IO - 914ee5dc049STobin C. Harding that is IO requests which bypass the page cache and transfer 915ee5dc049STobin C. Harding data directly between the storage and the application's address 916ee5dc049STobin C. Harding space. 917af96c1e3STobin C. Harding 918ee5dc049STobin C. Harding``isolate_page`` 919ee5dc049STobin C. Harding Called by the VM when isolating a movable non-lru page. If page 920ee5dc049STobin C. Harding is successfully isolated, VM marks the page as PG_isolated via 921ee5dc049STobin C. Harding __SetPageIsolated. 922af96c1e3STobin C. Harding 923ee5dc049STobin C. Harding``migrate_page`` 924ee5dc049STobin C. Harding This is used to compact the physical memory usage. If the VM 925ee5dc049STobin C. Harding wants to relocate a page (maybe off a memory card that is 926ee5dc049STobin C. Harding signalling imminent failure) it will pass a new page and an old 927ee5dc049STobin C. Harding page to this function. migrate_page should transfer any private 928ee5dc049STobin C. Harding data across and update any references that it has to the page. 929af96c1e3STobin C. Harding 930ee5dc049STobin C. Harding``putback_page`` 931ee5dc049STobin C. Harding Called by the VM when isolated page's migration fails. 932af96c1e3STobin C. Harding 933ee5dc049STobin C. Harding``launder_page`` 934ee5dc049STobin C. Harding Called before freeing a page - it writes back the dirty page. 935ee5dc049STobin C. Harding To prevent redirtying the page, it is kept locked during the 936ee5dc049STobin C. Harding whole operation. 937af96c1e3STobin C. Harding 938ee5dc049STobin C. Harding``is_partially_uptodate`` 939ee5dc049STobin C. Harding Called by the VM when reading a file through the pagecache when 940ee5dc049STobin C. Harding the underlying blocksize != pagesize. If the required block is 941ee5dc049STobin C. Harding up to date then the read can complete without needing the IO to 942ee5dc049STobin C. Harding bring the whole page up to date. 943af96c1e3STobin C. Harding 944ee5dc049STobin C. Harding``is_dirty_writeback`` 945ee5dc049STobin C. Harding Called by the VM when attempting to reclaim a page. The VM uses 946ee5dc049STobin C. Harding dirty and writeback information to determine if it needs to 947ee5dc049STobin C. Harding stall to allow flushers a chance to complete some IO. 948ee5dc049STobin C. Harding Ordinarily it can use PageDirty and PageWriteback but some 949ee5dc049STobin C. Harding filesystems have more complex state (unstable pages in NFS 950ee5dc049STobin C. Harding prevent reclaim) or do not set those flags due to locking 951ee5dc049STobin C. Harding problems. This callback allows a filesystem to indicate to the 952ee5dc049STobin C. Harding VM if a page should be treated as dirty or writeback for the 953ee5dc049STobin C. Harding purposes of stalling. 954af96c1e3STobin C. Harding 955ee5dc049STobin C. Harding``error_remove_page`` 956ee5dc049STobin C. Harding normally set to generic_error_remove_page if truncation is ok 957ee5dc049STobin C. Harding for this address space. Used for memory failure handling. 958af96c1e3STobin C. Harding Setting this implies you deal with pages going away under you, 959af96c1e3STobin C. Harding unless you have them locked or reference counts increased. 960af96c1e3STobin C. Harding 961ee5dc049STobin C. Harding``swap_activate`` 962ee5dc049STobin C. Harding Called when swapon is used on a file to allocate space if 963ee5dc049STobin C. Harding necessary and pin the block lookup information in memory. A 964ee5dc049STobin C. Harding return value of zero indicates success, in which case this file 965ee5dc049STobin C. Harding can be used to back swapspace. 966af96c1e3STobin C. Harding 967ee5dc049STobin C. Harding``swap_deactivate`` 968ee5dc049STobin C. Harding Called during swapoff on files where swap_activate was 969ee5dc049STobin C. Harding successful. 970af96c1e3STobin C. Harding 971af96c1e3STobin C. Harding 972af96c1e3STobin C. HardingThe File Object 973af96c1e3STobin C. Harding=============== 974af96c1e3STobin C. Harding 975af96c1e3STobin C. HardingA file object represents a file opened by a process. This is also known 976af96c1e3STobin C. Hardingas an "open file description" in POSIX parlance. 977af96c1e3STobin C. Harding 978af96c1e3STobin C. Harding 979af96c1e3STobin C. Hardingstruct file_operations 980af96c1e3STobin C. Harding---------------------- 981af96c1e3STobin C. Harding 982af96c1e3STobin C. HardingThis describes how the VFS can manipulate an open file. As of kernel 983af96c1e3STobin C. Harding4.18, the following members are defined: 984af96c1e3STobin C. Harding 985af96c1e3STobin C. Harding.. code-block:: c 986af96c1e3STobin C. Harding 987af96c1e3STobin C. Harding struct file_operations { 988af96c1e3STobin C. Harding struct module *owner; 989af96c1e3STobin C. Harding loff_t (*llseek) (struct file *, loff_t, int); 990af96c1e3STobin C. Harding ssize_t (*read) (struct file *, char __user *, size_t, loff_t *); 991af96c1e3STobin C. Harding ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *); 992af96c1e3STobin C. Harding ssize_t (*read_iter) (struct kiocb *, struct iov_iter *); 993af96c1e3STobin C. Harding ssize_t (*write_iter) (struct kiocb *, struct iov_iter *); 994af96c1e3STobin C. Harding int (*iopoll)(struct kiocb *kiocb, bool spin); 995af96c1e3STobin C. Harding int (*iterate) (struct file *, struct dir_context *); 996af96c1e3STobin C. Harding int (*iterate_shared) (struct file *, struct dir_context *); 997af96c1e3STobin C. Harding __poll_t (*poll) (struct file *, struct poll_table_struct *); 998af96c1e3STobin C. Harding long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long); 999af96c1e3STobin C. Harding long (*compat_ioctl) (struct file *, unsigned int, unsigned long); 1000af96c1e3STobin C. Harding int (*mmap) (struct file *, struct vm_area_struct *); 1001af96c1e3STobin C. Harding int (*open) (struct inode *, struct file *); 1002af96c1e3STobin C. Harding int (*flush) (struct file *, fl_owner_t id); 1003af96c1e3STobin C. Harding int (*release) (struct inode *, struct file *); 1004af96c1e3STobin C. Harding int (*fsync) (struct file *, loff_t, loff_t, int datasync); 1005af96c1e3STobin C. Harding int (*fasync) (int, struct file *, int); 1006af96c1e3STobin C. Harding int (*lock) (struct file *, int, struct file_lock *); 1007af96c1e3STobin C. Harding ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int); 1008af96c1e3STobin C. Harding unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); 1009af96c1e3STobin C. Harding int (*check_flags)(int); 1010af96c1e3STobin C. Harding int (*flock) (struct file *, int, struct file_lock *); 1011af96c1e3STobin C. Harding ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); 1012af96c1e3STobin C. Harding ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); 1013af96c1e3STobin C. Harding int (*setlease)(struct file *, long, struct file_lock **, void **); 1014af96c1e3STobin C. Harding long (*fallocate)(struct file *file, int mode, loff_t offset, 1015af96c1e3STobin C. Harding loff_t len); 1016af96c1e3STobin C. Harding void (*show_fdinfo)(struct seq_file *m, struct file *f); 1017af96c1e3STobin C. Harding #ifndef CONFIG_MMU 1018af96c1e3STobin C. Harding unsigned (*mmap_capabilities)(struct file *); 1019af96c1e3STobin C. Harding #endif 1020af96c1e3STobin C. Harding ssize_t (*copy_file_range)(struct file *, loff_t, struct file *, loff_t, size_t, unsigned int); 1021af96c1e3STobin C. Harding loff_t (*remap_file_range)(struct file *file_in, loff_t pos_in, 1022af96c1e3STobin C. Harding struct file *file_out, loff_t pos_out, 1023af96c1e3STobin C. Harding loff_t len, unsigned int remap_flags); 1024af96c1e3STobin C. Harding int (*fadvise)(struct file *, loff_t, loff_t, int); 1025af96c1e3STobin C. Harding }; 1026af96c1e3STobin C. Harding 1027af96c1e3STobin C. HardingAgain, all methods are called without any locks being held, unless 1028af96c1e3STobin C. Hardingotherwise noted. 1029af96c1e3STobin C. Harding 1030ee5dc049STobin C. Harding``llseek`` 1031ee5dc049STobin C. Harding called when the VFS needs to move the file position index 1032af96c1e3STobin C. Harding 1033ee5dc049STobin C. Harding``read`` 1034ee5dc049STobin C. Harding called by read(2) and related system calls 1035af96c1e3STobin C. Harding 1036ee5dc049STobin C. Harding``read_iter`` 1037ee5dc049STobin C. Harding possibly asynchronous read with iov_iter as destination 1038af96c1e3STobin C. Harding 1039ee5dc049STobin C. Harding``write`` 1040ee5dc049STobin C. Harding called by write(2) and related system calls 1041af96c1e3STobin C. Harding 1042ee5dc049STobin C. Harding``write_iter`` 1043ee5dc049STobin C. Harding possibly asynchronous write with iov_iter as source 1044af96c1e3STobin C. Harding 1045ee5dc049STobin C. Harding``iopoll`` 1046ee5dc049STobin C. Harding called when aio wants to poll for completions on HIPRI iocbs 1047af96c1e3STobin C. Harding 1048ee5dc049STobin C. Harding``iterate`` 1049ee5dc049STobin C. Harding called when the VFS needs to read the directory contents 1050af96c1e3STobin C. Harding 1051ee5dc049STobin C. Harding``iterate_shared`` 1052ee5dc049STobin C. Harding called when the VFS needs to read the directory contents when 1053ee5dc049STobin C. Harding filesystem supports concurrent dir iterators 1054af96c1e3STobin C. Harding 1055ee5dc049STobin C. Harding``poll`` 1056ee5dc049STobin C. Harding called by the VFS when a process wants to check if there is 1057af96c1e3STobin C. Harding activity on this file and (optionally) go to sleep until there 1058af96c1e3STobin C. Harding is activity. Called by the select(2) and poll(2) system calls 1059af96c1e3STobin C. Harding 1060ee5dc049STobin C. Harding``unlocked_ioctl`` 1061ee5dc049STobin C. Harding called by the ioctl(2) system call. 1062af96c1e3STobin C. Harding 1063ee5dc049STobin C. Harding``compat_ioctl`` 1064ee5dc049STobin C. Harding called by the ioctl(2) system call when 32 bit system calls are 1065ee5dc049STobin C. Harding used on 64 bit kernels. 1066af96c1e3STobin C. Harding 1067ee5dc049STobin C. Harding``mmap`` 1068ee5dc049STobin C. Harding called by the mmap(2) system call 1069af96c1e3STobin C. Harding 1070ee5dc049STobin C. Harding``open`` 1071ee5dc049STobin C. Harding called by the VFS when an inode should be opened. When the VFS 1072af96c1e3STobin C. Harding opens a file, it creates a new "struct file". It then calls the 1073af96c1e3STobin C. Harding open method for the newly allocated file structure. You might 1074ee5dc049STobin C. Harding think that the open method really belongs in "struct 1075ee5dc049STobin C. Harding inode_operations", and you may be right. I think it's done the 1076ee5dc049STobin C. Harding way it is because it makes filesystems simpler to implement. 1077ee5dc049STobin C. Harding The open() method is a good place to initialize the 1078af96c1e3STobin C. Harding "private_data" member in the file structure if you want to point 1079af96c1e3STobin C. Harding to a device structure 1080af96c1e3STobin C. Harding 1081ee5dc049STobin C. Harding``flush`` 1082ee5dc049STobin C. Harding called by the close(2) system call to flush a file 1083af96c1e3STobin C. Harding 1084ee5dc049STobin C. Harding``release`` 1085ee5dc049STobin C. Harding called when the last reference to an open file is closed 1086af96c1e3STobin C. Harding 1087ee5dc049STobin C. Harding``fsync`` 1088ee5dc049STobin C. Harding called by the fsync(2) system call. Also see the section above 1089af96c1e3STobin C. Harding entitled "Handling errors during writeback". 1090af96c1e3STobin C. Harding 1091ee5dc049STobin C. Harding``fasync`` 1092ee5dc049STobin C. Harding called by the fcntl(2) system call when asynchronous 1093af96c1e3STobin C. Harding (non-blocking) mode is enabled for a file 1094af96c1e3STobin C. Harding 1095ee5dc049STobin C. Harding``lock`` 1096ee5dc049STobin C. Harding called by the fcntl(2) system call for F_GETLK, F_SETLK, and 1097ee5dc049STobin C. Harding F_SETLKW commands 1098af96c1e3STobin C. Harding 1099ee5dc049STobin C. Harding``get_unmapped_area`` 1100ee5dc049STobin C. Harding called by the mmap(2) system call 1101af96c1e3STobin C. Harding 1102ee5dc049STobin C. Harding``check_flags`` 1103ee5dc049STobin C. Harding called by the fcntl(2) system call for F_SETFL command 1104af96c1e3STobin C. Harding 1105ee5dc049STobin C. Harding``flock`` 1106ee5dc049STobin C. Harding called by the flock(2) system call 1107af96c1e3STobin C. Harding 1108ee5dc049STobin C. Harding``splice_write`` 1109ee5dc049STobin C. Harding called by the VFS to splice data from a pipe to a file. This 1110af96c1e3STobin C. Harding method is used by the splice(2) system call 1111af96c1e3STobin C. Harding 1112ee5dc049STobin C. Harding``splice_read`` 1113ee5dc049STobin C. Harding called by the VFS to splice data from file to a pipe. This 1114af96c1e3STobin C. Harding method is used by the splice(2) system call 1115af96c1e3STobin C. Harding 1116ee5dc049STobin C. Harding``setlease`` 1117ee5dc049STobin C. Harding called by the VFS to set or release a file lock lease. setlease 1118af96c1e3STobin C. Harding implementations should call generic_setlease to record or remove 1119af96c1e3STobin C. Harding the lease in the inode after setting it. 1120af96c1e3STobin C. Harding 1121ee5dc049STobin C. Harding``fallocate`` 1122ee5dc049STobin C. Harding called by the VFS to preallocate blocks or punch a hole. 1123af96c1e3STobin C. Harding 1124ee5dc049STobin C. Harding``copy_file_range`` 1125ee5dc049STobin C. Harding called by the copy_file_range(2) system call. 1126af96c1e3STobin C. Harding 1127ee5dc049STobin C. Harding``remap_file_range`` 1128ee5dc049STobin C. Harding called by the ioctl(2) system call for FICLONERANGE and FICLONE 1129ee5dc049STobin C. Harding and FIDEDUPERANGE commands to remap file ranges. An 1130ee5dc049STobin C. Harding implementation should remap len bytes at pos_in of the source 1131ee5dc049STobin C. Harding file into the dest file at pos_out. Implementations must handle 1132ee5dc049STobin C. Harding callers passing in len == 0; this means "remap to the end of the 1133ee5dc049STobin C. Harding source file". The return value should the number of bytes 1134ee5dc049STobin C. Harding remapped, or the usual negative error code if errors occurred 1135ee5dc049STobin C. Harding before any bytes were remapped. The remap_flags parameter 1136ee5dc049STobin C. Harding accepts REMAP_FILE_* flags. If REMAP_FILE_DEDUP is set then the 1137ee5dc049STobin C. Harding implementation must only remap if the requested file ranges have 1138cb56ecaeSJulia Lawall identical contents. If REMAP_FILE_CAN_SHORTEN is set, the caller is 1139ee5dc049STobin C. Harding ok with the implementation shortening the request length to 1140ee5dc049STobin C. Harding satisfy alignment or EOF requirements (or any other reason). 1141af96c1e3STobin C. Harding 1142ee5dc049STobin C. Harding``fadvise`` 1143ee5dc049STobin C. Harding possibly called by the fadvise64() system call. 1144af96c1e3STobin C. Harding 1145af96c1e3STobin C. HardingNote that the file operations are implemented by the specific 1146af96c1e3STobin C. Hardingfilesystem in which the inode resides. When opening a device node 1147af96c1e3STobin C. Harding(character or block special) most filesystems will call special 1148af96c1e3STobin C. Hardingsupport routines in the VFS which will locate the required device 1149af96c1e3STobin C. Hardingdriver information. These support routines replace the filesystem file 1150af96c1e3STobin C. Hardingoperations with those for the device driver, and then proceed to call 1151af96c1e3STobin C. Hardingthe new open() method for the file. This is how opening a device file 1152af96c1e3STobin C. Hardingin the filesystem eventually ends up calling the device driver open() 1153af96c1e3STobin C. Hardingmethod. 1154af96c1e3STobin C. Harding 1155af96c1e3STobin C. Harding 1156af96c1e3STobin C. HardingDirectory Entry Cache (dcache) 1157af96c1e3STobin C. Harding============================== 1158af96c1e3STobin C. Harding 1159af96c1e3STobin C. Harding 1160af96c1e3STobin C. Hardingstruct dentry_operations 1161af96c1e3STobin C. Harding------------------------ 1162af96c1e3STobin C. Harding 1163af96c1e3STobin C. HardingThis describes how a filesystem can overload the standard dentry 1164af96c1e3STobin C. Hardingoperations. Dentries and the dcache are the domain of the VFS and the 1165af96c1e3STobin C. Hardingindividual filesystem implementations. Device drivers have no business 1166af96c1e3STobin C. Hardinghere. These methods may be set to NULL, as they are either optional or 1167af96c1e3STobin C. Hardingthe VFS uses a default. As of kernel 2.6.22, the following members are 1168af96c1e3STobin C. Hardingdefined: 1169af96c1e3STobin C. Harding 1170af96c1e3STobin C. Harding.. code-block:: c 1171af96c1e3STobin C. Harding 1172af96c1e3STobin C. Harding struct dentry_operations { 1173af96c1e3STobin C. Harding int (*d_revalidate)(struct dentry *, unsigned int); 1174af96c1e3STobin C. Harding int (*d_weak_revalidate)(struct dentry *, unsigned int); 1175af96c1e3STobin C. Harding int (*d_hash)(const struct dentry *, struct qstr *); 1176af96c1e3STobin C. Harding int (*d_compare)(const struct dentry *, 1177af96c1e3STobin C. Harding unsigned int, const char *, const struct qstr *); 1178af96c1e3STobin C. Harding int (*d_delete)(const struct dentry *); 1179af96c1e3STobin C. Harding int (*d_init)(struct dentry *); 1180af96c1e3STobin C. Harding void (*d_release)(struct dentry *); 1181af96c1e3STobin C. Harding void (*d_iput)(struct dentry *, struct inode *); 1182af96c1e3STobin C. Harding char *(*d_dname)(struct dentry *, char *, int); 1183af96c1e3STobin C. Harding struct vfsmount *(*d_automount)(struct path *); 1184af96c1e3STobin C. Harding int (*d_manage)(const struct path *, bool); 1185af96c1e3STobin C. Harding struct dentry *(*d_real)(struct dentry *, const struct inode *); 1186af96c1e3STobin C. Harding }; 1187af96c1e3STobin C. Harding 1188ee5dc049STobin C. Harding``d_revalidate`` 1189ee5dc049STobin C. Harding called when the VFS needs to revalidate a dentry. This is 1190ee5dc049STobin C. Harding called whenever a name look-up finds a dentry in the dcache. 1191ee5dc049STobin C. Harding Most local filesystems leave this as NULL, because all their 1192ee5dc049STobin C. Harding dentries in the dcache are valid. Network filesystems are 1193ee5dc049STobin C. Harding different since things can change on the server without the 1194ee5dc049STobin C. Harding client necessarily being aware of it. 1195af96c1e3STobin C. Harding 1196ee5dc049STobin C. Harding This function should return a positive value if the dentry is 1197ee5dc049STobin C. Harding still valid, and zero or a negative error code if it isn't. 1198af96c1e3STobin C. Harding 1199ee5dc049STobin C. Harding d_revalidate may be called in rcu-walk mode (flags & 1200ee5dc049STobin C. Harding LOOKUP_RCU). If in rcu-walk mode, the filesystem must 1201ee5dc049STobin C. Harding revalidate the dentry without blocking or storing to the dentry, 1202ee5dc049STobin C. Harding d_parent and d_inode should not be used without care (because 1203ee5dc049STobin C. Harding they can change and, in d_inode case, even become NULL under 1204ee5dc049STobin C. Harding us). 1205af96c1e3STobin C. Harding 1206ee5dc049STobin C. Harding If a situation is encountered that rcu-walk cannot handle, 1207ee5dc049STobin C. Harding return 1208af96c1e3STobin C. Harding -ECHILD and it will be called again in ref-walk mode. 1209af96c1e3STobin C. Harding 1210ee5dc049STobin C. Harding``_weak_revalidate`` 1211ee5dc049STobin C. Harding called when the VFS needs to revalidate a "jumped" dentry. This 1212ee5dc049STobin C. Harding is called when a path-walk ends at dentry that was not acquired 1213ee5dc049STobin C. Harding by doing a lookup in the parent directory. This includes "/", 1214ee5dc049STobin C. Harding "." and "..", as well as procfs-style symlinks and mountpoint 1215ee5dc049STobin C. Harding traversal. 1216af96c1e3STobin C. Harding 1217ee5dc049STobin C. Harding In this case, we are less concerned with whether the dentry is 1218ee5dc049STobin C. Harding still fully correct, but rather that the inode is still valid. 1219ee5dc049STobin C. Harding As with d_revalidate, most local filesystems will set this to 1220ee5dc049STobin C. Harding NULL since their dcache entries are always valid. 1221af96c1e3STobin C. Harding 1222ee5dc049STobin C. Harding This function has the same return code semantics as 1223ee5dc049STobin C. Harding d_revalidate. 1224af96c1e3STobin C. Harding 1225af96c1e3STobin C. Harding d_weak_revalidate is only called after leaving rcu-walk mode. 1226af96c1e3STobin C. Harding 1227ee5dc049STobin C. Harding``d_hash`` 1228ee5dc049STobin C. Harding called when the VFS adds a dentry to the hash table. The first 1229af96c1e3STobin C. Harding dentry passed to d_hash is the parent directory that the name is 1230af96c1e3STobin C. Harding to be hashed into. 1231af96c1e3STobin C. Harding 1232af96c1e3STobin C. Harding Same locking and synchronisation rules as d_compare regarding 1233af96c1e3STobin C. Harding what is safe to dereference etc. 1234af96c1e3STobin C. Harding 1235ee5dc049STobin C. Harding``d_compare`` 1236ee5dc049STobin C. Harding called to compare a dentry name with a given name. The first 1237af96c1e3STobin C. Harding dentry is the parent of the dentry to be compared, the second is 1238ee5dc049STobin C. Harding the child dentry. len and name string are properties of the 1239ee5dc049STobin C. Harding dentry to be compared. qstr is the name to compare it with. 1240af96c1e3STobin C. Harding 1241af96c1e3STobin C. Harding Must be constant and idempotent, and should not take locks if 1242ee5dc049STobin C. Harding possible, and should not or store into the dentry. Should not 1243ee5dc049STobin C. Harding dereference pointers outside the dentry without lots of care 1244ee5dc049STobin C. Harding (eg. d_parent, d_inode, d_name should not be used). 1245af96c1e3STobin C. Harding 1246ee5dc049STobin C. Harding However, our vfsmount is pinned, and RCU held, so the dentries 1247ee5dc049STobin C. Harding and inodes won't disappear, neither will our sb or filesystem 1248ee5dc049STobin C. Harding module. ->d_sb may be used. 1249af96c1e3STobin C. Harding 1250ee5dc049STobin C. Harding It is a tricky calling convention because it needs to be called 1251ee5dc049STobin C. Harding under "rcu-walk", ie. without any locks or references on things. 1252af96c1e3STobin C. Harding 1253ee5dc049STobin C. Harding``d_delete`` 1254ee5dc049STobin C. Harding called when the last reference to a dentry is dropped and the 1255ee5dc049STobin C. Harding dcache is deciding whether or not to cache it. Return 1 to 1256ee5dc049STobin C. Harding delete immediately, or 0 to cache the dentry. Default is NULL 1257ee5dc049STobin C. Harding which means to always cache a reachable dentry. d_delete must 1258ee5dc049STobin C. Harding be constant and idempotent. 1259af96c1e3STobin C. Harding 1260ee5dc049STobin C. Harding``d_init`` 1261ee5dc049STobin C. Harding called when a dentry is allocated 1262af96c1e3STobin C. Harding 1263ee5dc049STobin C. Harding``d_release`` 1264ee5dc049STobin C. Harding called when a dentry is really deallocated 1265af96c1e3STobin C. Harding 1266ee5dc049STobin C. Harding``d_iput`` 1267ee5dc049STobin C. Harding called when a dentry loses its inode (just prior to its being 1268ee5dc049STobin C. Harding deallocated). The default when this is NULL is that the VFS 1269ee5dc049STobin C. Harding calls iput(). If you define this method, you must call iput() 1270ee5dc049STobin C. Harding yourself 1271af96c1e3STobin C. Harding 1272ee5dc049STobin C. Harding``d_dname`` 1273ee5dc049STobin C. Harding called when the pathname of a dentry should be generated. 1274ee5dc049STobin C. Harding Useful for some pseudo filesystems (sockfs, pipefs, ...) to 1275ee5dc049STobin C. Harding delay pathname generation. (Instead of doing it when dentry is 1276ee5dc049STobin C. Harding created, it's done only when the path is needed.). Real 1277ee5dc049STobin C. Harding filesystems probably dont want to use it, because their dentries 1278ee5dc049STobin C. Harding are present in global dcache hash, so their hash should be an 1279ee5dc049STobin C. Harding invariant. As no lock is held, d_dname() should not try to 1280ee5dc049STobin C. Harding modify the dentry itself, unless appropriate SMP safety is used. 1281ee5dc049STobin C. Harding CAUTION : d_path() logic is quite tricky. The correct way to 1282ee5dc049STobin C. Harding return for example "Hello" is to put it at the end of the 1283ee5dc049STobin C. Harding buffer, and returns a pointer to the first char. 1284ee5dc049STobin C. Harding dynamic_dname() helper function is provided to take care of 1285ee5dc049STobin C. Harding this. 1286af96c1e3STobin C. Harding 1287af96c1e3STobin C. Harding Example : 1288af96c1e3STobin C. Harding 1289af96c1e3STobin C. Harding.. code-block:: c 1290af96c1e3STobin C. Harding 1291af96c1e3STobin C. Harding static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen) 1292af96c1e3STobin C. Harding { 1293af96c1e3STobin C. Harding return dynamic_dname(dentry, buffer, buflen, "pipe:[%lu]", 1294af96c1e3STobin C. Harding dentry->d_inode->i_ino); 1295af96c1e3STobin C. Harding } 1296af96c1e3STobin C. Harding 1297ee5dc049STobin C. Harding``d_automount`` 1298ee5dc049STobin C. Harding called when an automount dentry is to be traversed (optional). 1299ee5dc049STobin C. Harding This should create a new VFS mount record and return the record 1300ee5dc049STobin C. Harding to the caller. The caller is supplied with a path parameter 1301ee5dc049STobin C. Harding giving the automount directory to describe the automount target 1302ee5dc049STobin C. Harding and the parent VFS mount record to provide inheritable mount 1303ee5dc049STobin C. Harding parameters. NULL should be returned if someone else managed to 1304ee5dc049STobin C. Harding make the automount first. If the vfsmount creation failed, then 1305ee5dc049STobin C. Harding an error code should be returned. If -EISDIR is returned, then 1306ee5dc049STobin C. Harding the directory will be treated as an ordinary directory and 1307ee5dc049STobin C. Harding returned to pathwalk to continue walking. 1308af96c1e3STobin C. Harding 1309ee5dc049STobin C. Harding If a vfsmount is returned, the caller will attempt to mount it 1310ee5dc049STobin C. Harding on the mountpoint and will remove the vfsmount from its 1311ee5dc049STobin C. Harding expiration list in the case of failure. The vfsmount should be 1312ee5dc049STobin C. Harding returned with 2 refs on it to prevent automatic expiration - the 1313ee5dc049STobin C. Harding caller will clean up the additional ref. 1314af96c1e3STobin C. Harding 1315ee5dc049STobin C. Harding This function is only used if DCACHE_NEED_AUTOMOUNT is set on 1316ee5dc049STobin C. Harding the dentry. This is set by __d_instantiate() if S_AUTOMOUNT is 1317ee5dc049STobin C. Harding set on the inode being added. 1318af96c1e3STobin C. Harding 1319ee5dc049STobin C. Harding``d_manage`` 1320ee5dc049STobin C. Harding called to allow the filesystem to manage the transition from a 1321ee5dc049STobin C. Harding dentry (optional). This allows autofs, for example, to hold up 1322ee5dc049STobin C. Harding clients waiting to explore behind a 'mountpoint' while letting 1323ee5dc049STobin C. Harding the daemon go past and construct the subtree there. 0 should be 1324ee5dc049STobin C. Harding returned to let the calling process continue. -EISDIR can be 1325ee5dc049STobin C. Harding returned to tell pathwalk to use this directory as an ordinary 1326ee5dc049STobin C. Harding directory and to ignore anything mounted on it and not to check 1327ee5dc049STobin C. Harding the automount flag. Any other error code will abort pathwalk 1328ee5dc049STobin C. Harding completely. 1329af96c1e3STobin C. Harding 1330af96c1e3STobin C. Harding If the 'rcu_walk' parameter is true, then the caller is doing a 1331ee5dc049STobin C. Harding pathwalk in RCU-walk mode. Sleeping is not permitted in this 1332ee5dc049STobin C. Harding mode, and the caller can be asked to leave it and call again by 1333ee5dc049STobin C. Harding returning -ECHILD. -EISDIR may also be returned to tell 1334ee5dc049STobin C. Harding pathwalk to ignore d_automount or any mounts. 1335af96c1e3STobin C. Harding 1336ee5dc049STobin C. Harding This function is only used if DCACHE_MANAGE_TRANSIT is set on 1337ee5dc049STobin C. Harding the dentry being transited from. 1338af96c1e3STobin C. Harding 1339ee5dc049STobin C. Harding``d_real`` 1340ee5dc049STobin C. Harding overlay/union type filesystems implement this method to return 1341ee5dc049STobin C. Harding one of the underlying dentries hidden by the overlay. It is 1342ee5dc049STobin C. Harding used in two different modes: 1343af96c1e3STobin C. Harding 1344ee5dc049STobin C. Harding Called from file_dentry() it returns the real dentry matching 1345ee5dc049STobin C. Harding the inode argument. The real dentry may be from a lower layer 1346ee5dc049STobin C. Harding already copied up, but still referenced from the file. This 1347ee5dc049STobin C. Harding mode is selected with a non-NULL inode argument. 1348af96c1e3STobin C. Harding 1349af96c1e3STobin C. Harding With NULL inode the topmost real underlying dentry is returned. 1350af96c1e3STobin C. Harding 1351af96c1e3STobin C. HardingEach dentry has a pointer to its parent dentry, as well as a hash list 1352af96c1e3STobin C. Hardingof child dentries. Child dentries are basically like files in a 1353af96c1e3STobin C. Hardingdirectory. 1354af96c1e3STobin C. Harding 1355af96c1e3STobin C. Harding 1356af96c1e3STobin C. HardingDirectory Entry Cache API 1357af96c1e3STobin C. Harding-------------------------- 1358af96c1e3STobin C. Harding 1359af96c1e3STobin C. HardingThere are a number of functions defined which permit a filesystem to 1360af96c1e3STobin C. Hardingmanipulate dentries: 1361af96c1e3STobin C. Harding 1362ee5dc049STobin C. Harding``dget`` 1363ee5dc049STobin C. Harding open a new handle for an existing dentry (this just increments 1364af96c1e3STobin C. Harding the usage count) 1365af96c1e3STobin C. Harding 1366ee5dc049STobin C. Harding``dput`` 1367ee5dc049STobin C. Harding close a handle for a dentry (decrements the usage count). If 1368af96c1e3STobin C. Harding the usage count drops to 0, and the dentry is still in its 1369af96c1e3STobin C. Harding parent's hash, the "d_delete" method is called to check whether 1370ee5dc049STobin C. Harding it should be cached. If it should not be cached, or if the 1371ee5dc049STobin C. Harding dentry is not hashed, it is deleted. Otherwise cached dentries 1372ee5dc049STobin C. Harding are put into an LRU list to be reclaimed on memory shortage. 1373af96c1e3STobin C. Harding 1374ee5dc049STobin C. Harding``d_drop`` 1375ee5dc049STobin C. Harding this unhashes a dentry from its parents hash list. A subsequent 1376ee5dc049STobin C. Harding call to dput() will deallocate the dentry if its usage count 1377ee5dc049STobin C. Harding drops to 0 1378af96c1e3STobin C. Harding 1379ee5dc049STobin C. Harding``d_delete`` 1380ee5dc049STobin C. Harding delete a dentry. If there are no other open references to the 1381ee5dc049STobin C. Harding dentry then the dentry is turned into a negative dentry (the 1382ee5dc049STobin C. Harding d_iput() method is called). If there are other references, then 1383ee5dc049STobin C. Harding d_drop() is called instead 1384af96c1e3STobin C. Harding 1385ee5dc049STobin C. Harding``d_add`` 1386ee5dc049STobin C. Harding add a dentry to its parents hash list and then calls 1387af96c1e3STobin C. Harding d_instantiate() 1388af96c1e3STobin C. Harding 1389ee5dc049STobin C. Harding``d_instantiate`` 1390ee5dc049STobin C. Harding add a dentry to the alias hash list for the inode and updates 1391ee5dc049STobin C. Harding the "d_inode" member. The "i_count" member in the inode 1392ee5dc049STobin C. Harding structure should be set/incremented. If the inode pointer is 1393ee5dc049STobin C. Harding NULL, the dentry is called a "negative dentry". This function 1394ee5dc049STobin C. Harding is commonly called when an inode is created for an existing 1395ee5dc049STobin C. Harding negative dentry 1396af96c1e3STobin C. Harding 1397ee5dc049STobin C. Harding``d_lookup`` 1398ee5dc049STobin C. Harding look up a dentry given its parent and path name component It 1399ee5dc049STobin C. Harding looks up the child of that given name from the dcache hash 1400ee5dc049STobin C. Harding table. If it is found, the reference count is incremented and 1401ee5dc049STobin C. Harding the dentry is returned. The caller must use dput() to free the 1402ee5dc049STobin C. Harding dentry when it finishes using it. 1403af96c1e3STobin C. Harding 1404af96c1e3STobin C. Harding 1405af96c1e3STobin C. HardingMount Options 1406af96c1e3STobin C. Harding============= 1407af96c1e3STobin C. Harding 1408af96c1e3STobin C. Harding 1409af96c1e3STobin C. HardingParsing options 1410af96c1e3STobin C. Harding--------------- 1411af96c1e3STobin C. Harding 1412af96c1e3STobin C. HardingOn mount and remount the filesystem is passed a string containing a 1413af96c1e3STobin C. Hardingcomma separated list of mount options. The options can have either of 1414af96c1e3STobin C. Hardingthese forms: 1415af96c1e3STobin C. Harding 1416af96c1e3STobin C. Harding option 1417af96c1e3STobin C. Harding option=value 1418af96c1e3STobin C. Harding 1419af96c1e3STobin C. HardingThe <linux/parser.h> header defines an API that helps parse these 1420af96c1e3STobin C. Hardingoptions. There are plenty of examples on how to use it in existing 1421af96c1e3STobin C. Hardingfilesystems. 1422af96c1e3STobin C. Harding 1423af96c1e3STobin C. Harding 1424af96c1e3STobin C. HardingShowing options 1425af96c1e3STobin C. Harding--------------- 1426af96c1e3STobin C. Harding 1427af96c1e3STobin C. HardingIf a filesystem accepts mount options, it must define show_options() to 1428af96c1e3STobin C. Hardingshow all the currently active options. The rules are: 1429af96c1e3STobin C. Harding 1430af96c1e3STobin C. Harding - options MUST be shown which are not default or their values differ 1431af96c1e3STobin C. Harding from the default 1432af96c1e3STobin C. Harding 1433af96c1e3STobin C. Harding - options MAY be shown which are enabled by default or have their 1434af96c1e3STobin C. Harding default value 1435af96c1e3STobin C. Harding 1436af96c1e3STobin C. HardingOptions used only internally between a mount helper and the kernel (such 1437af96c1e3STobin C. Hardingas file descriptors), or which only have an effect during the mounting 1438af96c1e3STobin C. Harding(such as ones controlling the creation of a journal) are exempt from the 1439af96c1e3STobin C. Hardingabove rules. 1440af96c1e3STobin C. Harding 1441af96c1e3STobin C. HardingThe underlying reason for the above rules is to make sure, that a mount 1442af96c1e3STobin C. Hardingcan be accurately replicated (e.g. umounting and mounting again) based 1443af96c1e3STobin C. Hardingon the information found in /proc/mounts. 1444af96c1e3STobin C. Harding 1445af96c1e3STobin C. Harding 1446af96c1e3STobin C. HardingResources 1447af96c1e3STobin C. Harding========= 1448af96c1e3STobin C. Harding 1449af96c1e3STobin C. Harding(Note some of these resources are not up-to-date with the latest kernel 1450af96c1e3STobin C. Harding version.) 1451af96c1e3STobin C. Harding 1452af96c1e3STobin C. HardingCreating Linux virtual filesystems. 2002 1453c69f22f2SAlexander A. Klimov <https://lwn.net/Articles/13325/> 1454af96c1e3STobin C. Harding 1455af96c1e3STobin C. HardingThe Linux Virtual File-system Layer by Neil Brown. 1999 1456af96c1e3STobin C. Harding <http://www.cse.unsw.edu.au/~neilb/oss/linux-commentary/vfs.html> 1457af96c1e3STobin C. Harding 1458af96c1e3STobin C. HardingA tour of the Linux VFS by Michael K. Johnson. 1996 1459c69f22f2SAlexander A. Klimov <https://www.tldp.org/LDP/khg/HyperNews/get/fs/vfstour.html> 1460af96c1e3STobin C. Harding 1461af96c1e3STobin C. HardingA small trail through the Linux kernel by Andries Brouwer. 2001 1462c69f22f2SAlexander A. Klimov <https://www.win.tue.nl/~aeb/linux/vfs/trail.html> 1463