From 034dd340b08be1f2f0477ad16131d609f9dbd53c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 30 Aug 2026 09:22:00 -0700 Subject: Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace Pull tracing fixes from Steven Rostedt: - Fix error output of boot instance creation failure Currently if a boot instance creation fails, instead of printing out the name of the instance that failed, it prints "(null)". That is because it prints "cur_str" that had already been processed by strsep(). Print the saved name instead. While at it, print the error code of the failure. - Fix use-after-free for same named historgrams Histograms can be named so that they can be used in multiple events. But if the named histogram has a variable attached, the second event that uses the named histogram which duplicates it and needs to free the original after duplication leaves the old variable in place and still visible. If another histogram uses than variable, it will use the stale one which will try to reference the freed duplicate histogram and crash the kernel. Free the duplicate variables along with the duplicated histogram data. - Check return value of kthread_run() in event self test The events self tests uses a kthread for testing but does not check if it succeeded in creating a kthread. If the kthread creation were to fail, the code will still try to call kthread_stop() on the error returned. - Fix race between reading trace_pipe and updating subbuffer size If a user is reading the trace_pipe file at the same time they update the ring buffer sub-buffer size, can cause the trace_pipe read to read stale data. Add trace_access_lock() around updating the ring buffer sub-buffer size. - Fix eventfs_inode on failure path in creation of the events directory In the creation of the "events" directory, if after allocating the eventfs_inode a failure is detected, it calls cleanup_ei() which calls free_ei(). The free_ei() will test if eventfs_inode being freed has no children. It is a bug if it does. But on the failure case of the creation of the "events" directory, the children lists have not yet been initialized and the free will trigger a warning because list_empty() on an uninitialized list returns false. Move the initialization into init_ei() where it makes more sense and makes sure that a created eventfs_inode has its lists initialized upon creation. - Check return value of kthread_run() in ftrace direct sample code The sample code that shows how to use the ftrace direct calls does not test the return of kthread_run() to see if it succeeds. Return a failure if the kthread_run() doesn't succeed. - Clear user events state on fork in case of alloc failure On fork, the child gets a pointer to the parent's user events state. It makes a copy of it then updates the child's pointer to it. But if the allocation fails, the duplication function leaves the child with a pointer to its parent's descriptor. When the child cleans up its data, it will free the parent's descriptor while the parent is still using it. In the duplication function, set the child's user_event_mm to NULL before testing if the allocation succeeded, and when it exits it will not free the parent's descriptor. - Fix retry exhaustion in simple ring buffer reader swap simple_ring_buffer_swap_reader_page() starts with retry set to 8 and post-decrements it only after a failed link replacement. On the final attempt, a successful replacement leaves retry at zero, while a failed replacement leaves it at -1. But the check for success expects the retry value to be non-zero and exits with an error on zero. This is the opposite result. Fix it. - Fail nicely when the remote swap_reader_page() returns an error Currently, if the swap_reader_page() of a remote buffer fails, it triggers a WARN_ON_ONCE() and continues normally. Instead, have it exit with an error and a pr_warn() print instead of a full WARNING. * tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: ring-buffer: Stop remote reader update when page swap fails tracing: Fix retry exhaustion in simple ring buffer reader swap tracing/user_events: Clear copied tracing state before fork duplication samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify eventfs: Initialize ei->children and ei->list in init_ei() tracing: Fix use-after-free in trace_pipe read on sub-buffer order change tracing: Fix crash passing ERR_PTR to kthread_stop() tracing: Fix use-after-free with same-name named triggers tracing: Fix logged instance name on creation failure --- fs/nilfs2/bmap.c | 568 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 568 insertions(+) create mode 100644 fs/nilfs2/bmap.c (limited to 'fs/nilfs2/bmap.c') diff --git a/fs/nilfs2/bmap.c b/fs/nilfs2/bmap.c new file mode 100644 index 000000000..83f6ea30c --- /dev/null +++ b/fs/nilfs2/bmap.c @@ -0,0 +1,568 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * NILFS block mapping. + * + * Copyright (C) 2006-2008 Nippon Telegraph and Telephone Corporation. + * + * Written by Koji Sato. + */ + +#include +#include +#include +#include "nilfs.h" +#include "bmap.h" +#include "btree.h" +#include "direct.h" +#include "btnode.h" +#include "mdt.h" +#include "dat.h" +#include "alloc.h" + +struct inode *nilfs_bmap_get_dat(const struct nilfs_bmap *bmap) +{ + struct the_nilfs *nilfs = bmap->b_inode->i_sb->s_fs_info; + + return nilfs->ns_dat; +} + +static int nilfs_bmap_convert_error(struct nilfs_bmap *bmap, + const char *fname, int err) +{ + struct inode *inode = bmap->b_inode; + + if (err == -EINVAL) { + __nilfs_error(inode->i_sb, fname, + "broken bmap (inode number=%llu)", inode->i_ino); + err = -EIO; + } + return err; +} + +/** + * nilfs_bmap_lookup_at_level - find a data block or node block + * @bmap: bmap + * @key: key + * @level: level + * @ptrp: place to store the value associated to @key + * + * Description: nilfs_bmap_lookup_at_level() finds a record whose key + * matches @key in the block at @level of the bmap. The record associated + * with @key is stored in the place pointed to by @ptrp. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOENT - A record associated with @key does not exist. + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_lookup_at_level(struct nilfs_bmap *bmap, __u64 key, int level, + __u64 *ptrp) +{ + sector_t blocknr; + int ret; + + down_read(&bmap->b_sem); + ret = bmap->b_ops->bop_lookup(bmap, key, level, ptrp); + if (ret < 0) + goto out; + + if (NILFS_BMAP_USE_VBN(bmap)) { + ret = nilfs_dat_translate(nilfs_bmap_get_dat(bmap), *ptrp, + &blocknr); + if (!ret) + *ptrp = blocknr; + else if (ret == -ENOENT) { + /* + * If there was no valid entry in DAT for the block + * address obtained by b_ops->bop_lookup, then pass + * internal code -EINVAL to nilfs_bmap_convert_error + * to treat it as metadata corruption. + */ + ret = -EINVAL; + } + } + + out: + up_read(&bmap->b_sem); + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +int nilfs_bmap_lookup_contig(struct nilfs_bmap *bmap, __u64 key, __u64 *ptrp, + unsigned int maxblocks) +{ + int ret; + + down_read(&bmap->b_sem); + ret = bmap->b_ops->bop_lookup_contig(bmap, key, ptrp, maxblocks); + up_read(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +static int nilfs_bmap_do_insert(struct nilfs_bmap *bmap, __u64 key, __u64 ptr) +{ + __u64 keys[NILFS_BMAP_SMALL_HIGH + 1]; + __u64 ptrs[NILFS_BMAP_SMALL_HIGH + 1]; + int ret, n; + + if (bmap->b_ops->bop_check_insert != NULL) { + ret = bmap->b_ops->bop_check_insert(bmap, key); + if (ret > 0) { + n = bmap->b_ops->bop_gather_data( + bmap, keys, ptrs, NILFS_BMAP_SMALL_HIGH + 1); + if (n < 0) + return n; + ret = nilfs_btree_convert_and_insert( + bmap, key, ptr, keys, ptrs, n); + if (ret == 0) + bmap->b_u.u_flags |= NILFS_BMAP_LARGE; + + return ret; + } else if (ret < 0) + return ret; + } + + return bmap->b_ops->bop_insert(bmap, key, ptr); +} + +/** + * nilfs_bmap_insert - insert a new key-record pair into a bmap + * @bmap: bmap + * @key: key + * @rec: record + * + * Description: nilfs_bmap_insert() inserts the new key-record pair specified + * by @key and @rec into @bmap. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EEXIST - A record associated with @key already exists. + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_insert(struct nilfs_bmap *bmap, __u64 key, unsigned long rec) +{ + int ret; + + down_write(&bmap->b_sem); + ret = nilfs_bmap_do_insert(bmap, key, rec); + up_write(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +static int nilfs_bmap_do_delete(struct nilfs_bmap *bmap, __u64 key) +{ + __u64 keys[NILFS_BMAP_LARGE_LOW + 1]; + __u64 ptrs[NILFS_BMAP_LARGE_LOW + 1]; + int ret, n; + + if (bmap->b_ops->bop_check_delete != NULL) { + ret = bmap->b_ops->bop_check_delete(bmap, key); + if (ret > 0) { + n = bmap->b_ops->bop_gather_data( + bmap, keys, ptrs, NILFS_BMAP_LARGE_LOW + 1); + if (n < 0) + return n; + ret = nilfs_direct_delete_and_convert( + bmap, key, keys, ptrs, n); + if (ret == 0) + bmap->b_u.u_flags &= ~NILFS_BMAP_LARGE; + + return ret; + } else if (ret < 0) + return ret; + } + + return bmap->b_ops->bop_delete(bmap, key, false); +} + +/** + * nilfs_bmap_seek_key - seek a valid entry and return its key + * @bmap: bmap struct + * @start: start key number + * @keyp: place to store valid key + * + * Description: nilfs_bmap_seek_key() seeks a valid key on @bmap + * starting from @start, and stores it to @keyp if found. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOENT - No valid entry was found. + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_seek_key(struct nilfs_bmap *bmap, __u64 start, __u64 *keyp) +{ + int ret; + + down_read(&bmap->b_sem); + ret = bmap->b_ops->bop_seek_key(bmap, start, keyp); + up_read(&bmap->b_sem); + + if (ret < 0) + ret = nilfs_bmap_convert_error(bmap, __func__, ret); + return ret; +} + +int nilfs_bmap_last_key(struct nilfs_bmap *bmap, __u64 *keyp) +{ + int ret; + + down_read(&bmap->b_sem); + ret = bmap->b_ops->bop_last_key(bmap, keyp); + up_read(&bmap->b_sem); + + if (ret < 0) + ret = nilfs_bmap_convert_error(bmap, __func__, ret); + return ret; +} + +/** + * nilfs_bmap_delete - delete a key-record pair from a bmap + * @bmap: bmap + * @key: key + * + * Description: nilfs_bmap_delete() deletes the key-record pair specified by + * @key from @bmap. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOENT - A record associated with @key does not exist. + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_delete(struct nilfs_bmap *bmap, __u64 key) +{ + int ret; + + down_write(&bmap->b_sem); + ret = nilfs_bmap_do_delete(bmap, key); + up_write(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +static int nilfs_bmap_do_truncate(struct nilfs_bmap *bmap, __u64 key) +{ + __u64 lastkey; + int ret; + + ret = bmap->b_ops->bop_last_key(bmap, &lastkey); + if (ret < 0) { + if (ret == -ENOENT) + ret = 0; + return ret; + } + + while (key <= lastkey) { + ret = nilfs_bmap_do_delete(bmap, lastkey); + if (ret < 0) + return ret; + ret = bmap->b_ops->bop_last_key(bmap, &lastkey); + if (ret < 0) { + if (ret == -ENOENT) + ret = 0; + return ret; + } + } + return 0; +} + +/** + * nilfs_bmap_truncate - truncate a bmap to a specified key + * @bmap: bmap + * @key: key + * + * Description: nilfs_bmap_truncate() removes key-record pairs whose keys are + * greater than or equal to @key from @bmap. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_truncate(struct nilfs_bmap *bmap, __u64 key) +{ + int ret; + + down_write(&bmap->b_sem); + ret = nilfs_bmap_do_truncate(bmap, key); + up_write(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +/** + * nilfs_bmap_clear - free resources a bmap holds + * @bmap: bmap + * + * Description: nilfs_bmap_clear() frees resources associated with @bmap. + */ +void nilfs_bmap_clear(struct nilfs_bmap *bmap) +{ + down_write(&bmap->b_sem); + if (bmap->b_ops->bop_clear != NULL) + bmap->b_ops->bop_clear(bmap); + up_write(&bmap->b_sem); +} + +/** + * nilfs_bmap_propagate - propagate dirty state + * @bmap: bmap + * @bh: buffer head + * + * Description: nilfs_bmap_propagate() marks the buffers that directly or + * indirectly refer to the block specified by @bh dirty. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_propagate(struct nilfs_bmap *bmap, struct buffer_head *bh) +{ + int ret; + + down_write(&bmap->b_sem); + ret = bmap->b_ops->bop_propagate(bmap, bh); + up_write(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +/** + * nilfs_bmap_lookup_dirty_buffers - collect dirty block buffers + * @bmap: bmap + * @listp: pointer to buffer head list + */ +void nilfs_bmap_lookup_dirty_buffers(struct nilfs_bmap *bmap, + struct list_head *listp) +{ + if (bmap->b_ops->bop_lookup_dirty_buffers != NULL) + bmap->b_ops->bop_lookup_dirty_buffers(bmap, listp); +} + +/** + * nilfs_bmap_assign - assign a new block number to a block + * @bmap: bmap + * @bh: place to store a pointer to the buffer head to which a block + * address is assigned (in/out) + * @blocknr: block number + * @binfo: block information + * + * Description: nilfs_bmap_assign() assigns the block number @blocknr to the + * buffer specified by @bh. The block information is stored in the memory + * pointed to by @binfo, and the buffer head may be replaced as a block + * address is assigned, in which case a pointer to the new buffer head is + * stored in the memory pointed to by @bh. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_assign(struct nilfs_bmap *bmap, + struct buffer_head **bh, + unsigned long blocknr, + union nilfs_binfo *binfo) +{ + int ret; + + down_write(&bmap->b_sem); + ret = bmap->b_ops->bop_assign(bmap, bh, blocknr, binfo); + up_write(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +/** + * nilfs_bmap_mark - mark block dirty + * @bmap: bmap + * @key: key + * @level: level + * + * Description: nilfs_bmap_mark() marks the block specified by @key and @level + * as dirty. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (including metadata corruption). + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_mark(struct nilfs_bmap *bmap, __u64 key, int level) +{ + int ret; + + if (bmap->b_ops->bop_mark == NULL) + return 0; + + down_write(&bmap->b_sem); + ret = bmap->b_ops->bop_mark(bmap, key, level); + up_write(&bmap->b_sem); + + return nilfs_bmap_convert_error(bmap, __func__, ret); +} + +/** + * nilfs_bmap_test_and_clear_dirty - test and clear a bmap dirty state + * @bmap: bmap + * + * Description: nilfs_test_and_clear() is the atomic operation to test and + * clear the dirty state of @bmap. + * + * Return: 1 if @bmap is dirty, or 0 if clear. + */ +int nilfs_bmap_test_and_clear_dirty(struct nilfs_bmap *bmap) +{ + int ret; + + down_write(&bmap->b_sem); + ret = nilfs_bmap_dirty(bmap); + nilfs_bmap_clear_dirty(bmap); + up_write(&bmap->b_sem); + return ret; +} + + +/* + * Internal use only + */ +__u64 nilfs_bmap_data_get_key(const struct nilfs_bmap *bmap, + const struct buffer_head *bh) +{ + loff_t pos = folio_pos(bh->b_folio) + bh_offset(bh); + + return pos >> bmap->b_inode->i_blkbits; +} + +__u64 nilfs_bmap_find_target_seq(const struct nilfs_bmap *bmap, __u64 key) +{ + __s64 diff; + + diff = key - bmap->b_last_allocated_key; + if ((nilfs_bmap_keydiff_abs(diff) < NILFS_INODE_BMAP_SIZE) && + (bmap->b_last_allocated_ptr != NILFS_BMAP_INVALID_PTR) && + (bmap->b_last_allocated_ptr + diff > 0)) + return bmap->b_last_allocated_ptr + diff; + else + return NILFS_BMAP_INVALID_PTR; +} + +#define NILFS_BMAP_GROUP_DIV 8 /* must be power of 2 */ + +__u64 nilfs_bmap_find_target_in_group(const struct nilfs_bmap *bmap) +{ + struct inode *dat = nilfs_bmap_get_dat(bmap); + unsigned long entries_per_group = nilfs_palloc_entries_per_group(dat); + unsigned long group; + u32 index; + + BUILD_BUG_ON_NOT_POWER_OF_2(NILFS_BMAP_GROUP_DIV); + + group = div_u64(bmap->b_inode->i_ino, entries_per_group); + index = bmap->b_inode->i_ino & (NILFS_BMAP_GROUP_DIV - 1); + + return group * entries_per_group + + index * (entries_per_group / NILFS_BMAP_GROUP_DIV); +} + + +static struct lock_class_key nilfs_bmap_dat_lock_key; +static struct lock_class_key nilfs_bmap_mdt_lock_key; + +/** + * nilfs_bmap_read - read a bmap from an inode + * @bmap: bmap + * @raw_inode: on-disk inode + * + * Description: nilfs_bmap_read() initializes the bmap @bmap. + * + * Return: 0 on success, or one of the following negative error codes on + * failure: + * * %-EIO - I/O error (corrupted bmap). + * * %-ENOMEM - Insufficient memory available. + */ +int nilfs_bmap_read(struct nilfs_bmap *bmap, struct nilfs_inode *raw_inode) +{ + if (raw_inode == NULL) + memset(bmap->b_u.u_data, 0, NILFS_BMAP_SIZE); + else + memcpy(bmap->b_u.u_data, raw_inode->i_bmap, NILFS_BMAP_SIZE); + + init_rwsem(&bmap->b_sem); + bmap->b_state = 0; + bmap->b_inode = &NILFS_BMAP_I(bmap)->vfs_inode; + switch (bmap->b_inode->i_ino) { + case NILFS_DAT_INO: + bmap->b_ptr_type = NILFS_BMAP_PTR_P; + bmap->b_last_allocated_key = 0; + bmap->b_last_allocated_ptr = NILFS_BMAP_NEW_PTR_INIT; + lockdep_set_class(&bmap->b_sem, &nilfs_bmap_dat_lock_key); + break; + case NILFS_CPFILE_INO: + case NILFS_SUFILE_INO: + bmap->b_ptr_type = NILFS_BMAP_PTR_VS; + bmap->b_last_allocated_key = 0; + bmap->b_last_allocated_ptr = NILFS_BMAP_INVALID_PTR; + lockdep_set_class(&bmap->b_sem, &nilfs_bmap_mdt_lock_key); + break; + case NILFS_IFILE_INO: + lockdep_set_class(&bmap->b_sem, &nilfs_bmap_mdt_lock_key); + fallthrough; + default: + bmap->b_ptr_type = NILFS_BMAP_PTR_VM; + bmap->b_last_allocated_key = 0; + bmap->b_last_allocated_ptr = NILFS_BMAP_INVALID_PTR; + break; + } + + return (bmap->b_u.u_flags & NILFS_BMAP_LARGE) ? + nilfs_btree_init(bmap) : nilfs_direct_init(bmap); +} + +/** + * nilfs_bmap_write - write back a bmap to an inode + * @bmap: bmap + * @raw_inode: on-disk inode + * + * Description: nilfs_bmap_write() stores @bmap in @raw_inode. + */ +void nilfs_bmap_write(struct nilfs_bmap *bmap, struct nilfs_inode *raw_inode) +{ + memcpy(raw_inode->i_bmap, bmap->b_u.u_data, + NILFS_INODE_BMAP_SIZE * sizeof(__le64)); + if (bmap->b_inode->i_ino == NILFS_DAT_INO) + bmap->b_last_allocated_ptr = NILFS_BMAP_NEW_PTR_INIT; +} + +void nilfs_bmap_init_gc(struct nilfs_bmap *bmap) +{ + memset(&bmap->b_u, 0, NILFS_BMAP_SIZE); + init_rwsem(&bmap->b_sem); + bmap->b_inode = &NILFS_BMAP_I(bmap)->vfs_inode; + bmap->b_ptr_type = NILFS_BMAP_PTR_U; + bmap->b_last_allocated_key = 0; + bmap->b_last_allocated_ptr = NILFS_BMAP_INVALID_PTR; + bmap->b_state = 0; + nilfs_btree_init_gc(bmap); +} + +void nilfs_bmap_save(const struct nilfs_bmap *bmap, + struct nilfs_bmap_store *store) +{ + memcpy(store->data, bmap->b_u.u_data, sizeof(store->data)); + store->last_allocated_key = bmap->b_last_allocated_key; + store->last_allocated_ptr = bmap->b_last_allocated_ptr; + store->state = bmap->b_state; +} + +void nilfs_bmap_restore(struct nilfs_bmap *bmap, + const struct nilfs_bmap_store *store) +{ + memcpy(bmap->b_u.u_data, store->data, sizeof(store->data)); + bmap->b_last_allocated_key = store->last_allocated_key; + bmap->b_last_allocated_ptr = store->last_allocated_ptr; + bmap->b_state = store->state; +} -- cgit v1.3.1