// SPDX-License-Identifier: GPL-2.0-or-later /* * dir.c * * Creates, reads, walks and deletes directory-nodes * * Copyright (C) 2002, 2004 Oracle. All rights reserved. * * Portions of this code from linux/fs/ext3/dir.c * * Copyright (C) 1992, 1993, 1994, 1995 * Remy Card ([email protected]) * Laboratoire MASI - Institut Blaise pascal * Universite Pierre et Marie Curie (Paris VI) * * from * * linux/fs/minix/dir.c * * Copyright (C) 1991, 1992 Linus Torvalds */ #include <linux/fs.h> #include <linux/types.h> #include <linux/slab.h> #include <linux/highmem.h> #include <linux/quotaops.h> #include <linux/sort.h> #include <linux/iversion.h> #include <cluster/masklog.h> #include "ocfs2.h" #include "alloc.h" #include "blockcheck.h" #include "dir.h" #include "dlmglue.h" #include "extent_map.h" #include "file.h" #include "inode.h" #include "journal.h" #include "namei.h" #include "suballoc.h" #include "super.h" #include "sysfile.h" #include "uptodate.h" #include "ocfs2_trace.h" #include "buffer_head_io.h" #define NAMEI_RA_CHUNKS … #define NAMEI_RA_BLOCKS … #define NAMEI_RA_SIZE … static int ocfs2_do_extend_dir(struct super_block *sb, handle_t *handle, struct inode *dir, struct buffer_head *parent_fe_bh, struct ocfs2_alloc_context *data_ac, struct ocfs2_alloc_context *meta_ac, struct buffer_head **new_bh); static int ocfs2_dir_indexed(struct inode *inode); /* * These are distinct checks because future versions of the file system will * want to have a trailing dirent structure independent of indexing. */ static int ocfs2_supports_dir_trailer(struct inode *dir) { … } /* * "new' here refers to the point at which we're creating a new * directory via "mkdir()", but also when we're expanding an inline * directory. In either case, we don't yet have the indexing bit set * on the directory, so the standard checks will fail in when metaecc * is turned off. Only directory-initialization type functions should * use this then. Everything else wants ocfs2_supports_dir_trailer() */ static int ocfs2_new_dir_wants_trailer(struct inode *dir) { … } static inline unsigned int ocfs2_dir_trailer_blk_off(struct super_block *sb) { … } #define ocfs2_trailer_from_bh(_bh, _sb) … /* XXX ocfs2_block_dqtrailer() is similar but not quite - can we make * them more consistent? */ struct ocfs2_dir_block_trailer *ocfs2_dir_trailer_from_size(int blocksize, void *data) { … } /* * XXX: This is executed once on every dirent. We should consider optimizing * it. */ static int ocfs2_skip_dir_trailer(struct inode *dir, struct ocfs2_dir_entry *de, unsigned long offset, unsigned long blklen) { … } static void ocfs2_init_dir_trailer(struct inode *inode, struct buffer_head *bh, u16 rec_len) { … } /* * Link an unindexed block with a dir trailer structure into the index free * list. This function will modify dirdata_bh, but assumes you've already * passed it to the journal. */ static int ocfs2_dx_dir_link_trailer(struct inode *dir, handle_t *handle, struct buffer_head *dx_root_bh, struct buffer_head *dirdata_bh) { … } static int ocfs2_free_list_at_root(struct ocfs2_dir_lookup_result *res) { … } void ocfs2_free_dir_lookup_result(struct ocfs2_dir_lookup_result *res) { … } static int ocfs2_dir_indexed(struct inode *inode) { … } static inline int ocfs2_dx_root_inline(struct ocfs2_dx_root_block *dx_root) { … } /* * Hashing code adapted from ext3 */ #define DELTA … static void TEA_transform(__u32 buf[4], __u32 const in[]) { … } static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) { … } static void ocfs2_dx_dir_name_hash(struct inode *dir, const char *name, int len, struct ocfs2_dx_hinfo *hinfo) { … } /* * bh passed here can be an inode block or a dir data block, depending * on the inode inline data flag. */ static int ocfs2_check_dir_entry(struct inode *dir, struct ocfs2_dir_entry *de, struct buffer_head *bh, char *buf, unsigned int size, unsigned long offset) { … } static inline int ocfs2_match(int len, const char * const name, struct ocfs2_dir_entry *de) { … } /* * Returns 0 if not found, -1 on failure, and 1 on success */ static inline int ocfs2_search_dirblock(struct buffer_head *bh, struct inode *dir, const char *name, int namelen, unsigned long offset, char *first_de, unsigned int bytes, struct ocfs2_dir_entry **res_dir) { … } static struct buffer_head *ocfs2_find_entry_id(const char *name, int namelen, struct inode *dir, struct ocfs2_dir_entry **res_dir) { … } static int ocfs2_validate_dir_block(struct super_block *sb, struct buffer_head *bh) { … } /* * Validate a directory trailer. * * We check the trailer here rather than in ocfs2_validate_dir_block() * because that function doesn't have the inode to test. */ static int ocfs2_check_dir_trailer(struct inode *dir, struct buffer_head *bh) { … } /* * This function forces all errors to -EIO for consistency with its * predecessor, ocfs2_bread(). We haven't audited what returning the * real error codes would do to callers. We log the real codes with * mlog_errno() before we squash them. */ static int ocfs2_read_dir_block(struct inode *inode, u64 v_block, struct buffer_head **bh, int flags) { … } /* * Read the block at 'phys' which belongs to this directory * inode. This function does no virtual->physical block translation - * what's passed in is assumed to be a valid directory block. */ static int ocfs2_read_dir_block_direct(struct inode *dir, u64 phys, struct buffer_head **bh) { … } static int ocfs2_validate_dx_root(struct super_block *sb, struct buffer_head *bh) { … } static int ocfs2_read_dx_root(struct inode *dir, struct ocfs2_dinode *di, struct buffer_head **dx_root_bh) { … } static int ocfs2_validate_dx_leaf(struct super_block *sb, struct buffer_head *bh) { … } static int ocfs2_read_dx_leaf(struct inode *dir, u64 blkno, struct buffer_head **dx_leaf_bh) { … } /* * Read a series of dx_leaf blocks. This expects all buffer_head * pointers to be NULL on function entry. */ static int ocfs2_read_dx_leaves(struct inode *dir, u64 start, int num, struct buffer_head **dx_leaf_bhs) { … } static struct buffer_head *ocfs2_find_entry_el(const char *name, int namelen, struct inode *dir, struct ocfs2_dir_entry **res_dir) { … } static int ocfs2_dx_dir_lookup_rec(struct inode *inode, struct ocfs2_extent_list *el, u32 major_hash, u32 *ret_cpos, u64 *ret_phys_blkno, unsigned int *ret_clen) { … } /* * Returns the block index, from the start of the cluster which this * hash belongs too. */ static inline unsigned int __ocfs2_dx_dir_hash_idx(struct ocfs2_super *osb, u32 minor_hash) { … } static inline unsigned int ocfs2_dx_dir_hash_idx(struct ocfs2_super *osb, struct ocfs2_dx_hinfo *hinfo) { … } static int ocfs2_dx_dir_lookup(struct inode *inode, struct ocfs2_extent_list *el, struct ocfs2_dx_hinfo *hinfo, u32 *ret_cpos, u64 *ret_phys_blkno) { … } static int ocfs2_dx_dir_search(const char *name, int namelen, struct inode *dir, struct ocfs2_dx_root_block *dx_root, struct ocfs2_dir_lookup_result *res) { … } static int ocfs2_find_entry_dx(const char *name, int namelen, struct inode *dir, struct ocfs2_dir_lookup_result *lookup) { … } /* * Try to find an entry of the provided name within 'dir'. * * If nothing was found, -ENOENT is returned. Otherwise, zero is * returned and the struct 'res' will contain information useful to * other directory manipulation functions. * * Caller can NOT assume anything about the contents of the * buffer_heads - they are passed back only so that it can be passed * into any one of the manipulation functions (add entry, delete * entry, etc). As an example, bh in the extent directory case is a * data block, in the inline-data case it actually points to an inode, * in the indexed directory case, multiple buffers are involved. */ int ocfs2_find_entry(const char *name, int namelen, struct inode *dir, struct ocfs2_dir_lookup_result *lookup) { … } /* * Update inode number and type of a previously found directory entry. */ int ocfs2_update_entry(struct inode *dir, handle_t *handle, struct ocfs2_dir_lookup_result *res, struct inode *new_entry_inode) { … } /* * __ocfs2_delete_entry deletes a directory entry by merging it with the * previous entry */ static int __ocfs2_delete_entry(handle_t *handle, struct inode *dir, struct ocfs2_dir_entry *de_del, struct buffer_head *bh, char *first_de, unsigned int bytes) { … } static unsigned int ocfs2_figure_dirent_hole(struct ocfs2_dir_entry *de) { … } static int ocfs2_find_max_rec_len(struct super_block *sb, struct buffer_head *dirblock_bh) { … } static void ocfs2_dx_list_remove_entry(struct ocfs2_dx_entry_list *entry_list, int index) { … } static int ocfs2_delete_entry_dx(handle_t *handle, struct inode *dir, struct ocfs2_dir_lookup_result *lookup) { … } static inline int ocfs2_delete_entry_id(handle_t *handle, struct inode *dir, struct ocfs2_dir_entry *de_del, struct buffer_head *bh) { … } static inline int ocfs2_delete_entry_el(handle_t *handle, struct inode *dir, struct ocfs2_dir_entry *de_del, struct buffer_head *bh) { … } /* * Delete a directory entry. Hide the details of directory * implementation from the caller. */ int ocfs2_delete_entry(handle_t *handle, struct inode *dir, struct ocfs2_dir_lookup_result *res) { … } /* * Check whether 'de' has enough room to hold an entry of * 'new_rec_len' bytes. */ static inline int ocfs2_dirent_would_fit(struct ocfs2_dir_entry *de, unsigned int new_rec_len) { … } static void ocfs2_dx_dir_leaf_insert_tail(struct ocfs2_dx_leaf *dx_leaf, struct ocfs2_dx_entry *dx_new_entry) { … } static void ocfs2_dx_entry_list_insert(struct ocfs2_dx_entry_list *entry_list, struct ocfs2_dx_hinfo *hinfo, u64 dirent_blk) { … } static int __ocfs2_dx_dir_leaf_insert(struct inode *dir, handle_t *handle, struct ocfs2_dx_hinfo *hinfo, u64 dirent_blk, struct buffer_head *dx_leaf_bh) { … } static void ocfs2_dx_inline_root_insert(struct inode *dir, handle_t *handle, struct ocfs2_dx_hinfo *hinfo, u64 dirent_blk, struct ocfs2_dx_root_block *dx_root) { … } static int ocfs2_dx_dir_insert(struct inode *dir, handle_t *handle, struct ocfs2_dir_lookup_result *lookup) { … } static void ocfs2_remove_block_from_free_list(struct inode *dir, handle_t *handle, struct ocfs2_dir_lookup_result *lookup) { … } /* * This expects that a journal write has been reserved on * lookup->dl_prev_leaf_bh or lookup->dl_dx_root_bh */ static void ocfs2_recalc_free_list(struct inode *dir, handle_t *handle, struct ocfs2_dir_lookup_result *lookup) { … } /* we don't always have a dentry for what we want to add, so people * like orphan dir can call this instead. * * The lookup context must have been filled from * ocfs2_prepare_dir_for_insert. */ int __ocfs2_add_entry(handle_t *handle, struct inode *dir, const char *name, int namelen, struct inode *inode, u64 blkno, struct buffer_head *parent_fe_bh, struct ocfs2_dir_lookup_result *lookup) { … } static int ocfs2_dir_foreach_blk_id(struct inode *inode, u64 *f_version, struct dir_context *ctx) { … } /* * NOTE: This function can be called against unindexed directories, * and indexed ones. */ static int ocfs2_dir_foreach_blk_el(struct inode *inode, u64 *f_version, struct dir_context *ctx, bool persist) { … } static int ocfs2_dir_foreach_blk(struct inode *inode, u64 *f_version, struct dir_context *ctx, bool persist) { … } /* * This is intended to be called from inside other kernel functions, * so we fake some arguments. */ int ocfs2_dir_foreach(struct inode *inode, struct dir_context *ctx) { … } /* * ocfs2_readdir() * */ int ocfs2_readdir(struct file *file, struct dir_context *ctx) { … } /* * NOTE: this should always be called with parent dir i_rwsem taken. */ int ocfs2_find_files_on_disk(const char *name, int namelen, u64 *blkno, struct inode *inode, struct ocfs2_dir_lookup_result *lookup) { … } /* * Convenience function for callers which just want the block number * mapped to a name and don't require the full dirent info, etc. */ int ocfs2_lookup_ino_from_name(struct inode *dir, const char *name, int namelen, u64 *blkno) { … } /* Check for a name within a directory. * * Return 0 if the name does not exist * Return -EEXIST if the directory contains the name * * Callers should have i_rwsem + a cluster lock on dir */ int ocfs2_check_dir_for_entry(struct inode *dir, const char *name, int namelen) { … } struct ocfs2_empty_dir_priv { … }; static bool ocfs2_empty_dir_filldir(struct dir_context *ctx, const char *name, int name_len, loff_t pos, u64 ino, unsigned type) { … } static int ocfs2_empty_dir_dx(struct inode *inode, struct ocfs2_empty_dir_priv *priv) { … } /* * routine to check that the specified directory is empty (for rmdir) * * Returns 1 if dir is empty, zero otherwise. * * XXX: This is a performance problem for unindexed directories. */ int ocfs2_empty_dir(struct inode *inode) { … } /* * Fills "." and ".." dirents in a new directory block. Returns dirent for * "..", which might be used during creation of a directory with a trailing * header. It is otherwise safe to ignore the return code. */ static struct ocfs2_dir_entry *ocfs2_fill_initial_dirents(struct inode *inode, struct inode *parent, char *start, unsigned int size) { … } /* * This works together with code in ocfs2_mknod_locked() which sets * the inline-data flag and initializes the inline-data section. */ static int ocfs2_fill_new_dir_id(struct ocfs2_super *osb, handle_t *handle, struct inode *parent, struct inode *inode, struct buffer_head *di_bh) { … } static int ocfs2_fill_new_dir_el(struct ocfs2_super *osb, handle_t *handle, struct inode *parent, struct inode *inode, struct buffer_head *fe_bh, struct ocfs2_alloc_context *data_ac, struct buffer_head **ret_new_bh) { … } static int ocfs2_dx_dir_attach_index(struct ocfs2_super *osb, handle_t *handle, struct inode *dir, struct buffer_head *di_bh, struct buffer_head *dirdata_bh, struct ocfs2_alloc_context *meta_ac, int dx_inline, u32 num_entries, struct buffer_head **ret_dx_root_bh) { … } static int ocfs2_dx_dir_format_cluster(struct ocfs2_super *osb, handle_t *handle, struct inode *dir, struct buffer_head **dx_leaves, int num_dx_leaves, u64 start_blk) { … } /* * Allocates and formats a new cluster for use in an indexed dir * leaf. This version will not do the extent insert, so that it can be * used by operations which need careful ordering. */ static int __ocfs2_dx_dir_new_cluster(struct inode *dir, u32 cpos, handle_t *handle, struct ocfs2_alloc_context *data_ac, struct buffer_head **dx_leaves, int num_dx_leaves, u64 *ret_phys_blkno) { … } static int ocfs2_dx_dir_new_cluster(struct inode *dir, struct ocfs2_extent_tree *et, u32 cpos, handle_t *handle, struct ocfs2_alloc_context *data_ac, struct ocfs2_alloc_context *meta_ac, struct buffer_head **dx_leaves, int num_dx_leaves) { … } static struct buffer_head **ocfs2_dx_dir_kmalloc_leaves(struct super_block *sb, int *ret_num_leaves) { … } static int ocfs2_fill_new_dir_dx(struct ocfs2_super *osb, handle_t *handle, struct inode *parent, struct inode *inode, struct buffer_head *di_bh, struct ocfs2_alloc_context *data_ac, struct ocfs2_alloc_context *meta_ac) { … } int ocfs2_fill_new_dir(struct ocfs2_super *osb, handle_t *handle, struct inode *parent, struct inode *inode, struct buffer_head *fe_bh, struct ocfs2_alloc_context *data_ac, struct ocfs2_alloc_context *meta_ac) { … } static int ocfs2_dx_dir_index_block(struct inode *dir, handle_t *handle, struct buffer_head **dx_leaves, int num_dx_leaves, u32 *num_dx_entries, struct buffer_head *dirent_bh) { … } /* * XXX: This expects dx_root_bh to already be part of the transaction. */ static void ocfs2_dx_dir_index_root_block(struct inode *dir, struct buffer_head *dx_root_bh, struct buffer_head *dirent_bh) { … } /* * Count the number of inline directory entries in di_bh and compare * them against the number of entries we can hold in an inline dx root * block. */ static int ocfs2_new_dx_should_be_inline(struct inode *dir, struct buffer_head *di_bh) { … } /* * Expand rec_len of the rightmost dirent in a directory block so that it * contains the end of our valid space for dirents. We do this during * expansion from an inline directory to one with extents. The first dir block * in that case is taken from the inline data portion of the inode block. * * This will also return the largest amount of contiguous space for a dirent * in the block. That value is *not* necessarily the last dirent, even after * expansion. The directory indexing code wants this value for free space * accounting. We do this here since we're already walking the entire dir * block. * * We add the dir trailer if this filesystem wants it. */ static unsigned int ocfs2_expand_last_dirent(char *start, unsigned int old_size, struct inode *dir) { … } /* * We allocate enough clusters to fulfill "blocks_wanted", but set * i_size to exactly one block. Ocfs2_extend_dir() will handle the * rest automatically for us. * * *first_block_bh is a pointer to the 1st data block allocated to the * directory. */ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, unsigned int blocks_wanted, struct ocfs2_dir_lookup_result *lookup, struct buffer_head **first_block_bh) { … } /* returns a bh of the 1st new block in the allocation. */ static int ocfs2_do_extend_dir(struct super_block *sb, handle_t *handle, struct inode *dir, struct buffer_head *parent_fe_bh, struct ocfs2_alloc_context *data_ac, struct ocfs2_alloc_context *meta_ac, struct buffer_head **new_bh) { … } /* * Assumes you already have a cluster lock on the directory. * * 'blocks_wanted' is only used if we have an inline directory which * is to be turned into an extent based one. The size of the dirent to * insert might be larger than the space gained by growing to just one * block, so we may have to grow the inode by two blocks in that case. * * If the directory is already indexed, dx_root_bh must be provided. */ static int ocfs2_extend_dir(struct ocfs2_super *osb, struct inode *dir, struct buffer_head *parent_fe_bh, unsigned int blocks_wanted, struct ocfs2_dir_lookup_result *lookup, struct buffer_head **new_de_bh) { … } static int ocfs2_find_dir_space_id(struct inode *dir, struct buffer_head *di_bh, const char *name, int namelen, struct buffer_head **ret_de_bh, unsigned int *blocks_wanted) { … } static int ocfs2_find_dir_space_el(struct inode *dir, const char *name, int namelen, struct buffer_head **ret_de_bh) { … } static int dx_leaf_sort_cmp(const void *a, const void *b) { … } static int ocfs2_dx_leaf_same_major(struct ocfs2_dx_leaf *dx_leaf) { … } /* * Find the optimal value to split this leaf on. This expects the leaf * entries to be in sorted order. * * leaf_cpos is the cpos of the leaf we're splitting. insert_hash is * the hash we want to insert. * * This function is only concerned with the major hash - that which * determines which cluster an item belongs to. */ static int ocfs2_dx_dir_find_leaf_split(struct ocfs2_dx_leaf *dx_leaf, u32 leaf_cpos, u32 insert_hash, u32 *split_hash) { … } /* * Transfer all entries in orig_dx_leaves whose major hash is equal to or * larger than split_hash into new_dx_leaves. We use a temporary * buffer (tmp_dx_leaf) to make the changes to the original leaf blocks. * * Since the block offset inside a leaf (cluster) is a constant mask * of minor_hash, we can optimize - an item at block offset X within * the original cluster, will be at offset X within the new cluster. */ static void ocfs2_dx_dir_transfer_leaf(struct inode *dir, u32 split_hash, handle_t *handle, struct ocfs2_dx_leaf *tmp_dx_leaf, struct buffer_head **orig_dx_leaves, struct buffer_head **new_dx_leaves, int num_dx_leaves) { … } static int ocfs2_dx_dir_rebalance_credits(struct ocfs2_super *osb, struct ocfs2_dx_root_block *dx_root) { … } /* * Find the median value in dx_leaf_bh and allocate a new leaf to move * half our entries into. */ static int ocfs2_dx_dir_rebalance(struct ocfs2_super *osb, struct inode *dir, struct buffer_head *dx_root_bh, struct buffer_head *dx_leaf_bh, struct ocfs2_dx_hinfo *hinfo, u32 leaf_cpos, u64 leaf_blkno) { … } static int ocfs2_find_dir_space_dx(struct ocfs2_super *osb, struct inode *dir, struct buffer_head *di_bh, struct buffer_head *dx_root_bh, const char *name, int namelen, struct ocfs2_dir_lookup_result *lookup) { … } static int ocfs2_search_dx_free_list(struct inode *dir, struct buffer_head *dx_root_bh, int namelen, struct ocfs2_dir_lookup_result *lookup) { … } static int ocfs2_expand_inline_dx_root(struct inode *dir, struct buffer_head *dx_root_bh) { … } static int ocfs2_inline_dx_has_space(struct buffer_head *dx_root_bh) { … } static int ocfs2_prepare_dx_dir_for_insert(struct inode *dir, struct buffer_head *di_bh, const char *name, int namelen, struct ocfs2_dir_lookup_result *lookup) { … } /* * Get a directory ready for insert. Any directory allocation required * happens here. Success returns zero, and enough context in the dir * lookup result that ocfs2_add_entry() will be able complete the task * with minimal performance impact. */ int ocfs2_prepare_dir_for_insert(struct ocfs2_super *osb, struct inode *dir, struct buffer_head *parent_fe_bh, const char *name, int namelen, struct ocfs2_dir_lookup_result *lookup) { … } static int ocfs2_dx_dir_remove_index(struct inode *dir, struct buffer_head *di_bh, struct buffer_head *dx_root_bh) { … } int ocfs2_dx_dir_truncate(struct inode *dir, struct buffer_head *di_bh) { … }
Codebase Browser
linux