// SPDX-License-Identifier: GPL-2.0-only /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. */ #include <linux/spinlock.h> #include <linux/completion.h> #include <linux/buffer_head.h> #include <linux/blkdev.h> #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> #include <linux/iomap.h> #include <linux/ktime.h> #include "gfs2.h" #include "incore.h" #include "bmap.h" #include "glock.h" #include "inode.h" #include "meta_io.h" #include "quota.h" #include "rgrp.h" #include "log.h" #include "super.h" #include "trans.h" #include "dir.h" #include "util.h" #include "aops.h" #include "trace_gfs2.h" /* This doesn't need to be that large as max 64 bit pointers in a 4k * block is 512, so __u16 is fine for that. It saves stack space to * keep it small. */ struct metapath { … }; static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length); /** * gfs2_unstuffer_folio - unstuff a stuffed inode into a block cached by a folio * @ip: the inode * @dibh: the dinode buffer * @block: the block number that was allocated * @folio: The folio. * * Returns: errno */ static int gfs2_unstuffer_folio(struct gfs2_inode *ip, struct buffer_head *dibh, u64 block, struct folio *folio) { … } static int __gfs2_unstuff_inode(struct gfs2_inode *ip, struct folio *folio) { … } /** * gfs2_unstuff_dinode - Unstuff a dinode when the data has grown too big * @ip: The GFS2 inode to unstuff * * This routine unstuffs a dinode and returns it to a "normal" state such * that the height can be grown in the traditional way. * * Returns: errno */ int gfs2_unstuff_dinode(struct gfs2_inode *ip) { … } /** * find_metapath - Find path through the metadata tree * @sdp: The superblock * @block: The disk block to look up * @mp: The metapath to return the result in * @height: The pre-calculated height of the metadata tree * * This routine returns a struct metapath structure that defines a path * through the metadata of inode "ip" to get to block "block". * * Example: * Given: "ip" is a height 3 file, "offset" is 101342453, and this is a * filesystem with a blocksize of 4096. * * find_metapath() would return a struct metapath structure set to: * mp_fheight = 3, mp_list[0] = 0, mp_list[1] = 48, and mp_list[2] = 165. * * That means that in order to get to the block containing the byte at * offset 101342453, we would load the indirect block pointed to by pointer * 0 in the dinode. We would then load the indirect block pointed to by * pointer 48 in that indirect block. We would then load the data block * pointed to by pointer 165 in that indirect block. * * ---------------------------------------- * | Dinode | | * | | 4| * | |0 1 2 3 4 5 9| * | | 6| * ---------------------------------------- * | * | * V * ---------------------------------------- * | Indirect Block | * | 5| * | 4 4 4 4 4 5 5 1| * |0 5 6 7 8 9 0 1 2| * ---------------------------------------- * | * | * V * ---------------------------------------- * | Indirect Block | * | 1 1 1 1 1 5| * | 6 6 6 6 6 1| * |0 3 4 5 6 7 2| * ---------------------------------------- * | * | * V * ---------------------------------------- * | Data block containing offset | * | 101342453 | * | | * | | * ---------------------------------------- * */ static void find_metapath(const struct gfs2_sbd *sdp, u64 block, struct metapath *mp, unsigned int height) { … } static inline unsigned int metapath_branch_start(const struct metapath *mp) { … } /** * metaptr1 - Return the first possible metadata pointer in a metapath buffer * @height: The metadata height (0 = dinode) * @mp: The metapath */ static inline __be64 *metaptr1(unsigned int height, const struct metapath *mp) { … } /** * metapointer - Return pointer to start of metadata in a buffer * @height: The metadata height (0 = dinode) * @mp: The metapath * * Return a pointer to the block number of the next height of the metadata * tree given a buffer containing the pointer to the current height of the * metadata tree. */ static inline __be64 *metapointer(unsigned int height, const struct metapath *mp) { … } static inline const __be64 *metaend(unsigned int height, const struct metapath *mp) { … } static void clone_metapath(struct metapath *clone, struct metapath *mp) { … } static void gfs2_metapath_ra(struct gfs2_glock *gl, __be64 *start, __be64 *end) { … } static inline struct buffer_head * metapath_dibh(struct metapath *mp) { … } static int __fillup_metapath(struct gfs2_inode *ip, struct metapath *mp, unsigned int x, unsigned int h) { … } /** * lookup_metapath - Walk the metadata tree to a specific point * @ip: The inode * @mp: The metapath * * Assumes that the inode's buffer has already been looked up and * hooked onto mp->mp_bh[0] and that the metapath has been initialised * by find_metapath(). * * If this function encounters part of the tree which has not been * allocated, it returns the current height of the tree at the point * at which it found the unallocated block. Blocks which are found are * added to the mp->mp_bh[] list. * * Returns: error */ static int lookup_metapath(struct gfs2_inode *ip, struct metapath *mp) { … } /** * fillup_metapath - fill up buffers for the metadata path to a specific height * @ip: The inode * @mp: The metapath * @h: The height to which it should be mapped * * Similar to lookup_metapath, but does lookups for a range of heights * * Returns: error or the number of buffers filled */ static int fillup_metapath(struct gfs2_inode *ip, struct metapath *mp, int h) { … } static sector_t metapath_to_block(struct gfs2_sbd *sdp, struct metapath *mp) { … } static void release_metapath(struct metapath *mp) { … } /** * gfs2_extent_length - Returns length of an extent of blocks * @bh: The metadata block * @ptr: Current position in @bh * @eob: Set to 1 if we hit "end of block" * * Returns: The length of the extent (minimum of one block) */ static inline unsigned int gfs2_extent_length(struct buffer_head *bh, __be64 *ptr, int *eob) { … } enum walker_status { … }; /* * gfs2_metadata_walker - walk an indirect block * @mp: Metapath to indirect block * @ptrs: Number of pointers to look at * * When returning WALK_FOLLOW, the walker must update @mp to point at the right * indirect block to follow. */ gfs2_metadata_walker; /* * gfs2_walk_metadata - walk a tree of indirect blocks * @inode: The inode * @mp: Starting point of walk * @max_len: Maximum number of blocks to walk * @walker: Called during the walk * * Returns 1 if the walk was stopped by @walker, 0 if we went past @max_len or * past the end of metadata, and a negative error code otherwise. */ static int gfs2_walk_metadata(struct inode *inode, struct metapath *mp, u64 max_len, gfs2_metadata_walker walker) { … } static enum walker_status gfs2_hole_walker(struct metapath *mp, unsigned int ptrs) { … } /** * gfs2_hole_size - figure out the size of a hole * @inode: The inode * @lblock: The logical starting block number * @len: How far to look (in blocks) * @mp: The metapath at lblock * @iomap: The iomap to store the hole size in * * This function modifies @mp. * * Returns: errno on error */ static int gfs2_hole_size(struct inode *inode, sector_t lblock, u64 len, struct metapath *mp, struct iomap *iomap) { … } static inline void gfs2_indirect_init(struct metapath *mp, struct gfs2_glock *gl, unsigned int i, unsigned offset, u64 bn) { … } enum alloc_state { … }; /** * __gfs2_iomap_alloc - Build a metadata tree of the requested height * @inode: The GFS2 inode * @iomap: The iomap structure * @mp: The metapath, with proper height information calculated * * In this routine we may have to alloc: * i) Indirect blocks to grow the metadata tree height * ii) Indirect blocks to fill in lower part of the metadata tree * iii) Data blocks * * This function is called after __gfs2_iomap_get, which works out the * total number of blocks which we need via gfs2_alloc_size. * * We then do the actual allocation asking for an extent at a time (if * enough contiguous free blocks are available, there will only be one * allocation request per call) and uses the state machine to initialise * the blocks in order. * * Right now, this function will allocate at most one indirect block * worth of data -- with a default block size of 4K, that's slightly * less than 2M. If this limitation is ever removed to allow huge * allocations, we would probably still want to limit the iomap size we * return to avoid stalling other tasks during huge writes; the next * iomap iteration would then find the blocks already allocated. * * Returns: errno on error */ static int __gfs2_iomap_alloc(struct inode *inode, struct iomap *iomap, struct metapath *mp) { … } #define IOMAP_F_GFS2_BOUNDARY … /** * gfs2_alloc_size - Compute the maximum allocation size * @inode: The inode * @mp: The metapath * @size: Requested size in blocks * * Compute the maximum size of the next allocation at @mp. * * Returns: size in blocks */ static u64 gfs2_alloc_size(struct inode *inode, struct metapath *mp, u64 size) { … } /** * __gfs2_iomap_get - Map blocks from an inode to disk blocks * @inode: The inode * @pos: Starting position in bytes * @length: Length to map, in bytes * @flags: iomap flags * @iomap: The iomap structure * @mp: The metapath * * Returns: errno */ static int __gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length, unsigned flags, struct iomap *iomap, struct metapath *mp) { … } static struct folio * gfs2_iomap_get_folio(struct iomap_iter *iter, loff_t pos, unsigned len) { … } static void gfs2_iomap_put_folio(struct inode *inode, loff_t pos, unsigned copied, struct folio *folio) { … } static const struct iomap_folio_ops gfs2_iomap_folio_ops = …; static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos, loff_t length, unsigned flags, struct iomap *iomap, struct metapath *mp) { … } static int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length, unsigned flags, struct iomap *iomap, struct iomap *srcmap) { … } static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length, ssize_t written, unsigned flags, struct iomap *iomap) { … } const struct iomap_ops gfs2_iomap_ops = …; /** * gfs2_block_map - Map one or more blocks of an inode to a disk block * @inode: The inode * @lblock: The logical block number * @bh_map: The bh to be mapped * @create: True if its ok to alloc blocks to satify the request * * The size of the requested mapping is defined in bh_map->b_size. * * Clears buffer_mapped(bh_map) and leaves bh_map->b_size unchanged * when @lblock is not mapped. Sets buffer_mapped(bh_map) and * bh_map->b_size to indicate the size of the mapping when @lblock and * successive blocks are mapped, up to the requested size. * * Sets buffer_boundary() if a read of metadata will be required * before the next block can be mapped. Sets buffer_new() if new * blocks were allocated. * * Returns: errno */ int gfs2_block_map(struct inode *inode, sector_t lblock, struct buffer_head *bh_map, int create) { … } int gfs2_get_extent(struct inode *inode, u64 lblock, u64 *dblock, unsigned int *extlen) { … } int gfs2_alloc_extent(struct inode *inode, u64 lblock, u64 *dblock, unsigned int *extlen, bool *new) { … } /* * NOTE: Never call gfs2_block_zero_range with an open transaction because it * uses iomap write to perform its actions, which begin their own transactions * (iomap_begin, get_folio, etc.) */ static int gfs2_block_zero_range(struct inode *inode, loff_t from, unsigned int length) { … } #define GFS2_JTRUNC_REVOKES … /** * gfs2_journaled_truncate - Wrapper for truncate_pagecache for jdata files * @inode: The inode being truncated * @oldsize: The original (larger) size * @newsize: The new smaller size * * With jdata files, we have to journal a revoke for each block which is * truncated. As a result, we need to split this into separate transactions * if the number of pages being truncated gets too large. */ static int gfs2_journaled_truncate(struct inode *inode, u64 oldsize, u64 newsize) { … } static int trunc_start(struct inode *inode, u64 newsize) { … } int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length, struct iomap *iomap) { … } int gfs2_iomap_alloc(struct inode *inode, loff_t pos, loff_t length, struct iomap *iomap) { … } /** * sweep_bh_for_rgrps - find an rgrp in a meta buffer and free blocks therein * @ip: inode * @rd_gh: holder of resource group glock * @bh: buffer head to sweep * @start: starting point in bh * @end: end point in bh * @meta: true if bh points to metadata (rather than data) * @btotal: place to keep count of total blocks freed * * We sweep a metadata buffer (provided by the metapath) for blocks we need to * free, and free them all. However, we do it one rgrp at a time. If this * block has references to multiple rgrps, we break it into individual * transactions. This allows other processes to use the rgrps while we're * focused on a single one, for better concurrency / performance. * At every transaction boundary, we rewrite the inode into the journal. * That way the bitmaps are kept consistent with the inode and we can recover * if we're interrupted by power-outages. * * Returns: 0, or return code if an error occurred. * *btotal has the total number of blocks freed */ static int sweep_bh_for_rgrps(struct gfs2_inode *ip, struct gfs2_holder *rd_gh, struct buffer_head *bh, __be64 *start, __be64 *end, bool meta, u32 *btotal) { … } static bool mp_eq_to_hgt(struct metapath *mp, __u16 *list, unsigned int h) { … } /** * find_nonnull_ptr - find a non-null pointer given a metapath and height * @sdp: The superblock * @mp: starting metapath * @h: desired height to search * @end_list: See punch_hole(). * @end_aligned: See punch_hole(). * * Assumes the metapath is valid (with buffers) out to height h. * Returns: true if a non-null pointer was found in the metapath buffer * false if all remaining pointers are NULL in the buffer */ static bool find_nonnull_ptr(struct gfs2_sbd *sdp, struct metapath *mp, unsigned int h, __u16 *end_list, unsigned int end_aligned) { … } enum dealloc_states { … }; static inline void metapointer_range(struct metapath *mp, int height, __u16 *start_list, unsigned int start_aligned, __u16 *end_list, unsigned int end_aligned, __be64 **start, __be64 **end) { … } static inline bool walk_done(struct gfs2_sbd *sdp, struct metapath *mp, int height, __u16 *end_list, unsigned int end_aligned) { … } /** * punch_hole - deallocate blocks in a file * @ip: inode to truncate * @offset: the start of the hole * @length: the size of the hole (or 0 for truncate) * * Punch a hole into a file or truncate a file at a given position. This * function operates in whole blocks (@offset and @length are rounded * accordingly); partially filled blocks must be cleared otherwise. * * This function works from the bottom up, and from the right to the left. In * other words, it strips off the highest layer (data) before stripping any of * the metadata. Doing it this way is best in case the operation is interrupted * by power failure, etc. The dinode is rewritten in every transaction to * guarantee integrity. */ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length) { … } static int trunc_end(struct gfs2_inode *ip) { … } /** * do_shrink - make a file smaller * @inode: the inode * @newsize: the size to make the file * * Called with an exclusive lock on @inode. The @size must * be equal to or smaller than the current inode size. * * Returns: errno */ static int do_shrink(struct inode *inode, u64 newsize) { … } /** * do_grow - Touch and update inode size * @inode: The inode * @size: The new size * * This function updates the timestamps on the inode and * may also increase the size of the inode. This function * must not be called with @size any smaller than the current * inode size. * * Although it is not strictly required to unstuff files here, * earlier versions of GFS2 have a bug in the stuffed file reading * code which will result in a buffer overrun if the size is larger * than the max stuffed file size. In order to prevent this from * occurring, such files are unstuffed, but in other cases we can * just update the inode size directly. * * Returns: 0 on success, or -ve on error */ static int do_grow(struct inode *inode, u64 size) { … } /** * gfs2_setattr_size - make a file a given size * @inode: the inode * @newsize: the size to make the file * * The file size can grow, shrink, or stay the same size. This * is called holding i_rwsem and an exclusive glock on the inode * in question. * * Returns: errno */ int gfs2_setattr_size(struct inode *inode, u64 newsize) { … } int gfs2_truncatei_resume(struct gfs2_inode *ip) { … } int gfs2_file_dealloc(struct gfs2_inode *ip) { … } /** * gfs2_free_journal_extents - Free cached journal bmap info * @jd: The journal * */ void gfs2_free_journal_extents(struct gfs2_jdesc *jd) { … } /** * gfs2_add_jextent - Add or merge a new extent to extent cache * @jd: The journal descriptor * @lblock: The logical block at start of new extent * @dblock: The physical block at start of new extent * @blocks: Size of extent in fs blocks * * Returns: 0 on success or -ENOMEM */ static int gfs2_add_jextent(struct gfs2_jdesc *jd, u64 lblock, u64 dblock, u64 blocks) { … } /** * gfs2_map_journal_extents - Cache journal bmap info * @sdp: The super block * @jd: The journal to map * * Create a reusable "extent" mapping from all logical * blocks to all physical blocks for the given journal. This will save * us time when writing journal blocks. Most journals will have only one * extent that maps all their logical blocks. That's because gfs2.mkfs * arranges the journal blocks sequentially to maximize performance. * So the extent would map the first block for the entire file length. * However, gfs2_jadd can happen while file activity is happening, so * those journals may not be sequential. Less likely is the case where * the users created their own journals by mounting the metafs and * laying it out. But it's still possible. These journals might have * several extents. * * Returns: 0 on success, or error on failure */ int gfs2_map_journal_extents(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd) { … } /** * gfs2_write_alloc_required - figure out if a write will require an allocation * @ip: the file being written to * @offset: the offset to write to * @len: the number of bytes being written * * Returns: 1 if an alloc is required, 0 otherwise */ int gfs2_write_alloc_required(struct gfs2_inode *ip, u64 offset, unsigned int len) { … } static int stuffed_zero_range(struct inode *inode, loff_t offset, loff_t length) { … } static int gfs2_journaled_truncate_range(struct inode *inode, loff_t offset, loff_t length) { … } int __gfs2_punch_hole(struct file *file, loff_t offset, loff_t length) { … } static int gfs2_map_blocks(struct iomap_writepage_ctx *wpc, struct inode *inode, loff_t offset, unsigned int len) { … } const struct iomap_writeback_ops gfs2_writeback_ops = …;
Codebase Browser
linux