diff options
Diffstat (limited to 'fs/btrfs/file.c')
| -rw-r--r-- | fs/btrfs/file.c | 3438 |
1 files changed, 1738 insertions, 1700 deletions
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 11204dbbe053..7a501e73d880 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -18,466 +18,64 @@ #include <linux/iversion.h> #include <linux/fsverity.h> #include "ctree.h" +#include "direct-io.h" #include "disk-io.h" #include "transaction.h" #include "btrfs_inode.h" -#include "print-tree.h" #include "tree-log.h" #include "locking.h" -#include "volumes.h" #include "qgroup.h" #include "compression.h" #include "delalloc-space.h" #include "reflink.h" #include "subpage.h" - -static struct kmem_cache *btrfs_inode_defrag_cachep; -/* - * when auto defrag is enabled we - * queue up these defrag structs to remember which - * inodes need defragging passes - */ -struct inode_defrag { - struct rb_node rb_node; - /* objectid */ - u64 ino; - /* - * transid where the defrag was added, we search for - * extents newer than this - */ - u64 transid; - - /* root objectid */ - u64 root; - - /* last offset we were able to defrag */ - u64 last_offset; - - /* if we've wrapped around back to zero once already */ - int cycled; -}; - -static int __compare_inode_defrag(struct inode_defrag *defrag1, - struct inode_defrag *defrag2) -{ - if (defrag1->root > defrag2->root) - return 1; - else if (defrag1->root < defrag2->root) - return -1; - else if (defrag1->ino > defrag2->ino) - return 1; - else if (defrag1->ino < defrag2->ino) - return -1; - else - return 0; -} - -/* pop a record for an inode into the defrag tree. The lock - * must be held already - * - * If you're inserting a record for an older transid than an - * existing record, the transid already in the tree is lowered - * - * If an existing record is found the defrag item you - * pass in is freed - */ -static int __btrfs_add_inode_defrag(struct btrfs_inode *inode, - struct inode_defrag *defrag) -{ - struct btrfs_fs_info *fs_info = inode->root->fs_info; - struct inode_defrag *entry; - struct rb_node **p; - struct rb_node *parent = NULL; - int ret; - - p = &fs_info->defrag_inodes.rb_node; - while (*p) { - parent = *p; - entry = rb_entry(parent, struct inode_defrag, rb_node); - - ret = __compare_inode_defrag(defrag, entry); - if (ret < 0) - p = &parent->rb_left; - else if (ret > 0) - p = &parent->rb_right; - else { - /* if we're reinserting an entry for - * an old defrag run, make sure to - * lower the transid of our existing record - */ - if (defrag->transid < entry->transid) - entry->transid = defrag->transid; - if (defrag->last_offset > entry->last_offset) - entry->last_offset = defrag->last_offset; - return -EEXIST; - } - } - set_bit(BTRFS_INODE_IN_DEFRAG, &inode->runtime_flags); - rb_link_node(&defrag->rb_node, parent, p); - rb_insert_color(&defrag->rb_node, &fs_info->defrag_inodes); - return 0; -} - -static inline int __need_auto_defrag(struct btrfs_fs_info *fs_info) -{ - if (!btrfs_test_opt(fs_info, AUTO_DEFRAG)) - return 0; - - if (btrfs_fs_closing(fs_info)) - return 0; - - return 1; -} - -/* - * insert a defrag record for this inode if auto defrag is - * enabled - */ -int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode) -{ - struct btrfs_root *root = inode->root; - struct btrfs_fs_info *fs_info = root->fs_info; - struct inode_defrag *defrag; - u64 transid; - int ret; - - if (!__need_auto_defrag(fs_info)) - return 0; - - if (test_bit(BTRFS_INODE_IN_DEFRAG, &inode->runtime_flags)) - return 0; - - if (trans) - transid = trans->transid; - else - transid = inode->root->last_trans; - - defrag = kmem_cache_zalloc(btrfs_inode_defrag_cachep, GFP_NOFS); - if (!defrag) - return -ENOMEM; - - defrag->ino = btrfs_ino(inode); - defrag->transid = transid; - defrag->root = root->root_key.objectid; - - spin_lock(&fs_info->defrag_inodes_lock); - if (!test_bit(BTRFS_INODE_IN_DEFRAG, &inode->runtime_flags)) { - /* - * If we set IN_DEFRAG flag and evict the inode from memory, - * and then re-read this inode, this new inode doesn't have - * IN_DEFRAG flag. At the case, we may find the existed defrag. - */ - ret = __btrfs_add_inode_defrag(inode, defrag); - if (ret) - kmem_cache_free(btrfs_inode_defrag_cachep, defrag); - } else { - kmem_cache_free(btrfs_inode_defrag_cachep, defrag); - } - spin_unlock(&fs_info->defrag_inodes_lock); - return 0; -} - -/* - * Requeue the defrag object. If there is a defrag object that points to - * the same inode in the tree, we will merge them together (by - * __btrfs_add_inode_defrag()) and free the one that we want to requeue. - */ -static void btrfs_requeue_inode_defrag(struct btrfs_inode *inode, - struct inode_defrag *defrag) -{ - struct btrfs_fs_info *fs_info = inode->root->fs_info; - int ret; - - if (!__need_auto_defrag(fs_info)) - goto out; - - /* - * Here we don't check the IN_DEFRAG flag, because we need merge - * them together. - */ - spin_lock(&fs_info->defrag_inodes_lock); - ret = __btrfs_add_inode_defrag(inode, defrag); - spin_unlock(&fs_info->defrag_inodes_lock); - if (ret) - goto out; - return; -out: - kmem_cache_free(btrfs_inode_defrag_cachep, defrag); -} - -/* - * pick the defragable inode that we want, if it doesn't exist, we will get - * the next one. - */ -static struct inode_defrag * -btrfs_pick_defrag_inode(struct btrfs_fs_info *fs_info, u64 root, u64 ino) -{ - struct inode_defrag *entry = NULL; - struct inode_defrag tmp; - struct rb_node *p; - struct rb_node *parent = NULL; - int ret; - - tmp.ino = ino; - tmp.root = root; - - spin_lock(&fs_info->defrag_inodes_lock); - p = fs_info->defrag_inodes.rb_node; - while (p) { - parent = p; - entry = rb_entry(parent, struct inode_defrag, rb_node); - - ret = __compare_inode_defrag(&tmp, entry); - if (ret < 0) - p = parent->rb_left; - else if (ret > 0) - p = parent->rb_right; - else - goto out; - } - - if (parent && __compare_inode_defrag(&tmp, entry) > 0) { - parent = rb_next(parent); - if (parent) - entry = rb_entry(parent, struct inode_defrag, rb_node); - else - entry = NULL; - } -out: - if (entry) - rb_erase(parent, &fs_info->defrag_inodes); - spin_unlock(&fs_info->defrag_inodes_lock); - return entry; -} - -void btrfs_cleanup_defrag_inodes(struct btrfs_fs_info *fs_info) -{ - struct inode_defrag *defrag; - struct rb_node *node; - - spin_lock(&fs_info->defrag_inodes_lock); - node = rb_first(&fs_info->defrag_inodes); - while (node) { - rb_erase(node, &fs_info->defrag_inodes); - defrag = rb_entry(node, struct inode_defrag, rb_node); - kmem_cache_free(btrfs_inode_defrag_cachep, defrag); - - cond_resched_lock(&fs_info->defrag_inodes_lock); - - node = rb_first(&fs_info->defrag_inodes); - } - spin_unlock(&fs_info->defrag_inodes_lock); -} - -#define BTRFS_DEFRAG_BATCH 1024 - -static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info, - struct inode_defrag *defrag) -{ - struct btrfs_root *inode_root; - struct inode *inode; - struct btrfs_ioctl_defrag_range_args range; - int num_defrag; - int ret; - - /* get the inode */ - inode_root = btrfs_get_fs_root(fs_info, defrag->root, true); - if (IS_ERR(inode_root)) { - ret = PTR_ERR(inode_root); - goto cleanup; - } - - inode = btrfs_iget(fs_info->sb, defrag->ino, inode_root); - btrfs_put_root(inode_root); - if (IS_ERR(inode)) { - ret = PTR_ERR(inode); - goto cleanup; - } - - /* do a chunk of defrag */ - clear_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags); - memset(&range, 0, sizeof(range)); - range.len = (u64)-1; - range.start = defrag->last_offset; - - sb_start_write(fs_info->sb); - num_defrag = btrfs_defrag_file(inode, NULL, &range, defrag->transid, - BTRFS_DEFRAG_BATCH); - sb_end_write(fs_info->sb); - /* - * if we filled the whole defrag batch, there - * must be more work to do. Queue this defrag - * again - */ - if (num_defrag == BTRFS_DEFRAG_BATCH) { - defrag->last_offset = range.start; - btrfs_requeue_inode_defrag(BTRFS_I(inode), defrag); - } else if (defrag->last_offset && !defrag->cycled) { - /* - * we didn't fill our defrag batch, but - * we didn't start at zero. Make sure we loop - * around to the start of the file. - */ - defrag->last_offset = 0; - defrag->cycled = 1; - btrfs_requeue_inode_defrag(BTRFS_I(inode), defrag); - } else { - kmem_cache_free(btrfs_inode_defrag_cachep, defrag); - } - - iput(inode); - return 0; -cleanup: - kmem_cache_free(btrfs_inode_defrag_cachep, defrag); - return ret; -} - -/* - * run through the list of inodes in the FS that need - * defragging - */ -int btrfs_run_defrag_inodes(struct btrfs_fs_info *fs_info) -{ - struct inode_defrag *defrag; - u64 first_ino = 0; - u64 root_objectid = 0; - - atomic_inc(&fs_info->defrag_running); - while (1) { - /* Pause the auto defragger. */ - if (test_bit(BTRFS_FS_STATE_REMOUNTING, - &fs_info->fs_state)) - break; - - if (!__need_auto_defrag(fs_info)) - break; - - /* find an inode to defrag */ - defrag = btrfs_pick_defrag_inode(fs_info, root_objectid, - first_ino); - if (!defrag) { - if (root_objectid || first_ino) { - root_objectid = 0; - first_ino = 0; - continue; - } else { - break; - } - } - - first_ino = defrag->ino + 1; - root_objectid = defrag->root; - - __btrfs_run_defrag_inode(fs_info, defrag); - } - atomic_dec(&fs_info->defrag_running); - - /* - * during unmount, we use the transaction_wait queue to - * wait for the defragger to stop - */ - wake_up(&fs_info->transaction_wait); - return 0; -} - -/* simple helper to fault in pages and copy. This should go away - * and be replaced with calls into generic code. - */ -static noinline int btrfs_copy_from_user(loff_t pos, size_t write_bytes, - struct page **prepared_pages, - struct iov_iter *i) -{ - size_t copied = 0; - size_t total_copied = 0; - int pg = 0; - int offset = offset_in_page(pos); - - while (write_bytes > 0) { - size_t count = min_t(size_t, - PAGE_SIZE - offset, write_bytes); - struct page *page = prepared_pages[pg]; - /* - * Copy data from userspace to the current page - */ - copied = copy_page_from_iter_atomic(page, offset, count, i); - - /* Flush processor's dcache for this page */ - flush_dcache_page(page); - - /* - * if we get a partial write, we can end up with - * partially up to date pages. These add - * a lot of complexity, so make sure they don't - * happen by forcing this copy to be retried. - * - * The rest of the btrfs_file_write code will fall - * back to page at a time copies after we return 0. - */ - if (unlikely(copied < count)) { - if (!PageUptodate(page)) { - iov_iter_revert(i, copied); - copied = 0; - } - if (!copied) - break; - } - - write_bytes -= copied; - total_copied += copied; - offset += copied; - if (offset == PAGE_SIZE) { - pg++; - offset = 0; - } - } - return total_copied; -} +#include "fs.h" +#include "accessors.h" +#include "extent-tree.h" +#include "file-item.h" +#include "ioctl.h" +#include "file.h" +#include "super.h" +#include "print-tree.h" /* - * unlocks pages after btrfs_file_write is done with them + * Unlock folio after btrfs_file_write() is done with it. */ -static void btrfs_drop_pages(struct btrfs_fs_info *fs_info, - struct page **pages, size_t num_pages, +static void btrfs_drop_folio(struct btrfs_fs_info *fs_info, struct folio *folio, u64 pos, u64 copied) { - size_t i; u64 block_start = round_down(pos, fs_info->sectorsize); u64 block_len = round_up(pos + copied, fs_info->sectorsize) - block_start; ASSERT(block_len <= U32_MAX); - for (i = 0; i < num_pages; i++) { - /* page checked is some magic around finding pages that - * have been modified without going through btrfs_set_page_dirty - * clear it here. There should be no need to mark the pages - * accessed as prepare_pages should have marked them accessed - * in prepare_pages via find_or_create_page() - */ - btrfs_page_clamp_clear_checked(fs_info, pages[i], block_start, - block_len); - unlock_page(pages[i]); - put_page(pages[i]); - } + /* + * Folio checked is some magic around finding folios that have been + * modified without going through btrfs_dirty_folio(). Clear it here. + * There should be no need to mark the pages accessed as + * prepare_one_folio() should have marked them accessed in + * prepare_one_folio() via find_or_create_page() + */ + btrfs_folio_clamp_clear_checked(fs_info, folio, block_start, block_len); + folio_unlock(folio); + folio_put(folio); } /* - * After btrfs_copy_from_user(), update the following things for delalloc: - * - Mark newly dirtied pages as DELALLOC in the io tree. + * After copy_folio_from_iter_atomic(), update the following things for delalloc: + * - Mark newly dirtied folio as DELALLOC in the io tree. * Used to advise which range is to be written back. - * - Mark modified pages as Uptodate/Dirty and not needing COW fixup + * - Mark modified folio as Uptodate/Dirty and not needing COW fixup * - Update inode size for past EOF write */ -int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, - size_t num_pages, loff_t pos, size_t write_bytes, - struct extent_state **cached, bool noreserve) +int btrfs_dirty_folio(struct btrfs_inode *inode, struct folio *folio, loff_t pos, + size_t write_bytes, struct extent_state **cached, bool noreserve) { struct btrfs_fs_info *fs_info = inode->root->fs_info; - int err = 0; - int i; + int ret = 0; u64 num_bytes; u64 start_pos; u64 end_of_last_block; - u64 end_pos = pos + write_bytes; + const u64 end_pos = pos + write_bytes; loff_t isize = i_size_read(&inode->vfs_inode); unsigned int extra_bits = 0; @@ -488,9 +86,9 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, extra_bits |= EXTENT_NORESERVE; start_pos = round_down(pos, fs_info->sectorsize); - num_bytes = round_up(write_bytes + pos - start_pos, - fs_info->sectorsize); + num_bytes = round_up(end_pos - start_pos, fs_info->sectorsize); ASSERT(num_bytes <= U32_MAX); + ASSERT(folio_pos(folio) <= pos && folio_next_pos(folio) >= end_pos); end_of_last_block = start_pos + num_bytes - 1; @@ -498,22 +96,18 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, * The pages may have already been dirty, clear out old accounting so * we can set things up properly */ - clear_extent_bit(&inode->io_tree, start_pos, end_of_last_block, - EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, - 0, 0, cached); + btrfs_clear_extent_bit(&inode->io_tree, start_pos, end_of_last_block, + EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, + cached); - err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block, + ret = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block, extra_bits, cached); - if (err) - return err; - - for (i = 0; i < num_pages; i++) { - struct page *p = pages[i]; + if (ret) + return ret; - btrfs_page_clamp_set_uptodate(fs_info, p, start_pos, num_bytes); - btrfs_page_clamp_clear_checked(fs_info, p, start_pos, num_bytes); - btrfs_page_clamp_set_dirty(fs_info, p, start_pos, num_bytes); - } + btrfs_folio_clamp_set_uptodate(fs_info, folio, start_pos, num_bytes); + btrfs_folio_clamp_clear_checked(fs_info, folio, start_pos, num_bytes); + btrfs_folio_clamp_set_dirty(fs_info, folio, start_pos, num_bytes); /* * we've only changed i_size in ram, and we haven't updated @@ -526,159 +120,6 @@ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, } /* - * this drops all the extents in the cache that intersect the range - * [start, end]. Existing extents are split as required. - */ -void btrfs_drop_extent_cache(struct btrfs_inode *inode, u64 start, u64 end, - int skip_pinned) -{ - struct extent_map *em; - struct extent_map *split = NULL; - struct extent_map *split2 = NULL; - struct extent_map_tree *em_tree = &inode->extent_tree; - u64 len = end - start + 1; - u64 gen; - int ret; - int testend = 1; - unsigned long flags; - int compressed = 0; - bool modified; - - WARN_ON(end < start); - if (end == (u64)-1) { - len = (u64)-1; - testend = 0; - } - while (1) { - int no_splits = 0; - - modified = false; - if (!split) - split = alloc_extent_map(); - if (!split2) - split2 = alloc_extent_map(); - if (!split || !split2) - no_splits = 1; - - write_lock(&em_tree->lock); - em = lookup_extent_mapping(em_tree, start, len); - if (!em) { - write_unlock(&em_tree->lock); - break; - } - flags = em->flags; - gen = em->generation; - if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) { - if (testend && em->start + em->len >= start + len) { - free_extent_map(em); - write_unlock(&em_tree->lock); - break; - } - start = em->start + em->len; - if (testend) - len = start + len - (em->start + em->len); - free_extent_map(em); - write_unlock(&em_tree->lock); - continue; - } - compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags); - clear_bit(EXTENT_FLAG_PINNED, &em->flags); - clear_bit(EXTENT_FLAG_LOGGING, &flags); - modified = !list_empty(&em->list); - if (no_splits) - goto next; - - if (em->start < start) { - split->start = em->start; - split->len = start - em->start; - - if (em->block_start < EXTENT_MAP_LAST_BYTE) { - split->orig_start = em->orig_start; - split->block_start = em->block_start; - - if (compressed) - split->block_len = em->block_len; - else - split->block_len = split->len; - split->orig_block_len = max(split->block_len, - em->orig_block_len); - split->ram_bytes = em->ram_bytes; - } else { - split->orig_start = split->start; - split->block_len = 0; - split->block_start = em->block_start; - split->orig_block_len = 0; - split->ram_bytes = split->len; - } - - split->generation = gen; - split->flags = flags; - split->compress_type = em->compress_type; - replace_extent_mapping(em_tree, em, split, modified); - free_extent_map(split); - split = split2; - split2 = NULL; - } - if (testend && em->start + em->len > start + len) { - u64 diff = start + len - em->start; - - split->start = start + len; - split->len = em->start + em->len - (start + len); - split->flags = flags; - split->compress_type = em->compress_type; - split->generation = gen; - - if (em->block_start < EXTENT_MAP_LAST_BYTE) { - split->orig_block_len = max(em->block_len, - em->orig_block_len); - - split->ram_bytes = em->ram_bytes; - if (compressed) { - split->block_len = em->block_len; - split->block_start = em->block_start; - split->orig_start = em->orig_start; - } else { - split->block_len = split->len; - split->block_start = em->block_start - + diff; - split->orig_start = em->orig_start; - } - } else { - split->ram_bytes = split->len; - split->orig_start = split->start; - split->block_len = 0; - split->block_start = em->block_start; - split->orig_block_len = 0; - } - - if (extent_map_in_tree(em)) { - replace_extent_mapping(em_tree, em, split, - modified); - } else { - ret = add_extent_mapping(em_tree, split, - modified); - ASSERT(ret == 0); /* Logic error */ - } - free_extent_map(split); - split = NULL; - } -next: - if (extent_map_in_tree(em)) - remove_extent_mapping(em_tree, em); - write_unlock(&em_tree->lock); - - /* once for us */ - free_extent_map(em); - /* once for the tree*/ - free_extent_map(em); - } - if (split) - free_extent_map(split); - if (split2) - free_extent_map(split2); -} - -/* * this is very complex, but the basic idea is to drop all extents * in the range start - end. hint_block is filled in with a block number * that would be a good hint to the block allocator for this file. @@ -700,7 +141,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info = root->fs_info; struct extent_buffer *leaf; struct btrfs_file_extent_item *fi; - struct btrfs_ref ref = { 0 }; struct btrfs_key key; struct btrfs_key new_key; u64 ino = btrfs_ino(inode); @@ -718,7 +158,6 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, int modify_tree = -1; int update_refs; int found = 0; - int leafs_visited = 0; struct btrfs_path *path = args->path; args->bytes_found = 0; @@ -736,12 +175,12 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, } if (args->drop_cache) - btrfs_drop_extent_cache(inode, args->start, args->end - 1, 0); + btrfs_drop_extent_map_range(inode, args->start, args->end - 1, false); - if (args->start >= inode->disk_i_size && !args->replace_extent) + if (data_race(args->start >= inode->disk_i_size) && !args->replace_extent) modify_tree = 0; - update_refs = (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID); + update_refs = (btrfs_root_id(root) != BTRFS_TREE_LOG_OBJECTID); while (1) { recow = 0; ret = btrfs_lookup_file_extent(trans, root, path, ino, @@ -756,11 +195,14 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, path->slots[0]--; } ret = 0; - leafs_visited++; next_slot: leaf = path->nodes[0]; if (path->slots[0] >= btrfs_header_nritems(leaf)) { - BUG_ON(del_nr > 0); + if (WARN_ON(del_nr > 0)) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } ret = btrfs_next_leaf(root, path); if (ret < 0) break; @@ -768,7 +210,6 @@ next_slot: ret = 0; break; } - leafs_visited++; leaf = path->nodes[0]; recow = 1; } @@ -837,7 +278,11 @@ next_slot: * | -------- extent -------- | */ if (args->start > key.offset && args->end < extent_end) { - BUG_ON(del_nr > 0); + if (WARN_ON(del_nr > 0)) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } if (extent_type == BTRFS_FILE_EXTENT_INLINE) { ret = -EOPNOTSUPP; break; @@ -867,19 +312,24 @@ next_slot: btrfs_set_file_extent_offset(leaf, fi, extent_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - args->start); - btrfs_mark_buffer_dirty(leaf); if (update_refs && disk_bytenr > 0) { - btrfs_init_generic_ref(&ref, - BTRFS_ADD_DELAYED_REF, - disk_bytenr, num_bytes, 0); - btrfs_init_data_ref(&ref, - root->root_key.objectid, - new_key.objectid, - args->start - extent_offset, - 0, false); + struct btrfs_ref ref = { + .action = BTRFS_ADD_DELAYED_REF, + .bytenr = disk_bytenr, + .num_bytes = num_bytes, + .parent = 0, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), + }; + btrfs_init_data_ref(&ref, new_key.objectid, + args->start - extent_offset, + 0, false); ret = btrfs_inc_extent_ref(trans, &ref); - BUG_ON(ret); /* -ENOMEM */ + if (unlikely(ret)) { + btrfs_abort_transaction(trans, ret); + break; + } } key.offset = args->start; } @@ -901,13 +351,12 @@ next_slot: memcpy(&new_key, &key, sizeof(new_key)); new_key.offset = args->end; - btrfs_set_item_key_safe(fs_info, path, &new_key); + btrfs_set_item_key_safe(trans, path, &new_key); extent_offset += args->end - key.offset; btrfs_set_file_extent_offset(leaf, fi, extent_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - args->end); - btrfs_mark_buffer_dirty(leaf); if (update_refs && disk_bytenr > 0) args->bytes_found += args->end - key.offset; break; @@ -919,7 +368,11 @@ next_slot: * | -------- extent -------- | */ if (args->start > key.offset && args->end >= extent_end) { - BUG_ON(del_nr > 0); + if (WARN_ON(del_nr > 0)) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } if (extent_type == BTRFS_FILE_EXTENT_INLINE) { ret = -EOPNOTSUPP; break; @@ -927,7 +380,6 @@ next_slot: btrfs_set_file_extent_num_bytes(leaf, fi, args->start - key.offset); - btrfs_mark_buffer_dirty(leaf); if (update_refs && disk_bytenr > 0) args->bytes_found += extent_end - args->start; if (args->end == extent_end) @@ -947,7 +399,11 @@ delete_extent_item: del_slot = path->slots[0]; del_nr = 1; } else { - BUG_ON(del_slot + del_nr != path->slots[0]); + if (WARN_ON(del_slot + del_nr != path->slots[0])) { + btrfs_print_leaf(leaf); + ret = -EINVAL; + break; + } del_nr++; } @@ -957,16 +413,22 @@ delete_extent_item: extent_end = ALIGN(extent_end, fs_info->sectorsize); } else if (update_refs && disk_bytenr > 0) { - btrfs_init_generic_ref(&ref, - BTRFS_DROP_DELAYED_REF, - disk_bytenr, num_bytes, 0); - btrfs_init_data_ref(&ref, - root->root_key.objectid, - key.objectid, - key.offset - extent_offset, 0, - false); + struct btrfs_ref ref = { + .action = BTRFS_DROP_DELAYED_REF, + .bytenr = disk_bytenr, + .num_bytes = num_bytes, + .parent = 0, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), + }; + btrfs_init_data_ref(&ref, key.objectid, + key.offset - extent_offset, + 0, false); ret = btrfs_free_extent(trans, &ref); - BUG_ON(ret); /* -ENOMEM */ + if (unlikely(ret)) { + btrfs_abort_transaction(trans, ret); + break; + } args->bytes_found += extent_end - key.offset; } @@ -980,7 +442,7 @@ delete_extent_item: ret = btrfs_del_items(trans, root, path, del_slot, del_nr); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); break; } @@ -1014,7 +476,7 @@ delete_extent_item: * which case it unlocked our path, so check path->locks[0] matches a * write lock. */ - if (!ret && args->replace_extent && leafs_visited == 1 && + if (!ret && args->replace_extent && path->locks[0] == BTRFS_WRITE_LOCK && btrfs_leaf_free_space(leaf) >= sizeof(struct btrfs_item) + args->extent_item_size) { @@ -1029,7 +491,8 @@ delete_extent_item: if (btrfs_comp_cpu_keys(&key, &slot_key) > 0) path->slots[0]++; } - btrfs_setup_item_for_insert(root, path, &key, args->extent_item_size); + btrfs_setup_item_for_insert(trans, root, path, &key, + args->extent_item_size); args->extent_inserted = true; } @@ -1043,20 +506,19 @@ out: return ret; } -static int extent_mergeable(struct extent_buffer *leaf, int slot, - u64 objectid, u64 bytenr, u64 orig_offset, - u64 *start, u64 *end) +static bool extent_mergeable(struct extent_buffer *leaf, int slot, u64 objectid, + u64 bytenr, u64 orig_offset, u64 *start, u64 *end) { struct btrfs_file_extent_item *fi; struct btrfs_key key; u64 extent_end; if (slot < 0 || slot >= btrfs_header_nritems(leaf)) - return 0; + return false; btrfs_item_key_to_cpu(leaf, &key, slot); if (key.objectid != objectid || key.type != BTRFS_EXTENT_DATA_KEY) - return 0; + return false; fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG || @@ -1065,15 +527,15 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot, btrfs_file_extent_compression(leaf, fi) || btrfs_file_extent_encryption(leaf, fi) || btrfs_file_extent_other_encoding(leaf, fi)) - return 0; + return false; extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); if ((*start && *start != key.offset) || (*end && *end != extent_end)) - return 0; + return false; *start = key.offset; *end = extent_end; - return 1; + return true; } /* @@ -1086,10 +548,9 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot, int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, struct btrfs_inode *inode, u64 start, u64 end) { - struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_root *root = inode->root; struct extent_buffer *leaf; - struct btrfs_path *path; + BTRFS_PATH_AUTO_FREE(path); struct btrfs_file_extent_item *fi; struct btrfs_ref ref = { 0 }; struct btrfs_key key; @@ -1125,21 +586,20 @@ again: leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); - if (key.objectid != ino || - key.type != BTRFS_EXTENT_DATA_KEY) { + if (unlikely(key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY)) { ret = -EINVAL; btrfs_abort_transaction(trans, ret); goto out; } fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); - if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_PREALLOC) { + if (unlikely(btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_PREALLOC)) { ret = -EINVAL; btrfs_abort_transaction(trans, ret); goto out; } extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); - if (key.offset > start || extent_end < end) { + if (unlikely(key.offset > start || extent_end < end)) { ret = -EINVAL; btrfs_abort_transaction(trans, ret); goto out; @@ -1157,7 +617,7 @@ again: ino, bytenr, orig_offset, &other_start, &other_end)) { new_key.offset = end; - btrfs_set_item_key_safe(fs_info, path, &new_key); + btrfs_set_item_key_safe(trans, path, &new_key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); btrfs_set_file_extent_generation(leaf, fi, @@ -1172,7 +632,6 @@ again: trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, end - other_start); - btrfs_mark_buffer_dirty(leaf); goto out; } } @@ -1191,7 +650,7 @@ again: trans->transid); path->slots[0]++; new_key.offset = start; - btrfs_set_item_key_safe(fs_info, path, &new_key); + btrfs_set_item_key_safe(trans, path, &new_key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); @@ -1201,7 +660,6 @@ again: other_end - start); btrfs_set_file_extent_offset(leaf, fi, start - orig_offset); - btrfs_mark_buffer_dirty(leaf); goto out; } } @@ -1216,7 +674,7 @@ again: btrfs_release_path(path); goto again; } - if (ret < 0) { + if (unlikely(ret < 0)) { btrfs_abort_transaction(trans, ret); goto out; } @@ -1235,14 +693,16 @@ again: btrfs_set_file_extent_offset(leaf, fi, split - orig_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - split); - btrfs_mark_buffer_dirty(leaf); - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, bytenr, - num_bytes, 0); - btrfs_init_data_ref(&ref, root->root_key.objectid, ino, - orig_offset, 0, false); + ref.action = BTRFS_ADD_DELAYED_REF; + ref.bytenr = bytenr; + ref.num_bytes = num_bytes; + ref.parent = 0; + ref.owning_root = btrfs_root_id(root); + ref.ref_root = btrfs_root_id(root); + btrfs_init_data_ref(&ref, ino, orig_offset, 0, false); ret = btrfs_inc_extent_ref(trans, &ref); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); goto out; } @@ -1250,7 +710,7 @@ again: if (split == start) { key.offset = start; } else { - if (start != key.offset) { + if (unlikely(start != key.offset)) { ret = -EINVAL; btrfs_abort_transaction(trans, ret); goto out; @@ -1263,10 +723,14 @@ again: other_start = end; other_end = 0; - btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, bytenr, - num_bytes, 0); - btrfs_init_data_ref(&ref, root->root_key.objectid, ino, orig_offset, - 0, false); + + ref.action = BTRFS_DROP_DELAYED_REF; + ref.bytenr = bytenr; + ref.num_bytes = num_bytes; + ref.parent = 0; + ref.owning_root = btrfs_root_id(root); + ref.ref_root = btrfs_root_id(root); + btrfs_init_data_ref(&ref, ino, orig_offset, 0, false); if (extent_mergeable(leaf, path->slots[0] + 1, ino, bytenr, orig_offset, &other_start, &other_end)) { @@ -1278,7 +742,7 @@ again: del_slot = path->slots[0] + 1; del_nr++; ret = btrfs_free_extent(trans, &ref); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); goto out; } @@ -1296,7 +760,7 @@ again: del_slot = path->slots[0]; del_nr++; ret = btrfs_free_extent(trans, &ref); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); goto out; } @@ -1307,7 +771,6 @@ again: btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG); btrfs_set_file_extent_generation(leaf, fi, trans->transid); - btrfs_mark_buffer_dirty(leaf); } else { fi = btrfs_item_ptr(leaf, del_slot - 1, struct btrfs_file_extent_item); @@ -1316,137 +779,131 @@ again: btrfs_set_file_extent_generation(leaf, fi, trans->transid); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - key.offset); - btrfs_mark_buffer_dirty(leaf); ret = btrfs_del_items(trans, root, path, del_slot, del_nr); - if (ret < 0) { + if (unlikely(ret < 0)) { btrfs_abort_transaction(trans, ret); goto out; } } out: - btrfs_free_path(path); return ret; } /* - * on error we return an unlocked page and the error value - * on success we return a locked page and 0 + * On error return an unlocked folio and the error value + * On success return a locked folio and 0 */ -static int prepare_uptodate_page(struct inode *inode, - struct page *page, u64 pos, - bool force_uptodate) +static int prepare_uptodate_folio(struct inode *inode, struct folio *folio, u64 pos, + u64 len) { + u64 clamp_start = max_t(u64, pos, folio_pos(folio)); + u64 clamp_end = min_t(u64, pos + len, folio_next_pos(folio)); + const u32 blocksize = inode_to_fs_info(inode)->sectorsize; int ret = 0; - if (((pos & (PAGE_SIZE - 1)) || force_uptodate) && - !PageUptodate(page)) { - ret = btrfs_readpage(NULL, page); - if (ret) - return ret; - lock_page(page); - if (!PageUptodate(page)) { - unlock_page(page); - return -EIO; - } + if (folio_test_uptodate(folio)) + return 0; - /* - * Since btrfs_readpage() will unlock the page before it - * returns, there is a window where btrfs_releasepage() can be - * called to release the page. Here we check both inode - * mapping and PagePrivate() to make sure the page was not - * released. - * - * The private flag check is essential for subpage as we need - * to store extra bitmap using page->private. - */ - if (page->mapping != inode->i_mapping || !PagePrivate(page)) { - unlock_page(page); - return -EAGAIN; - } + if (IS_ALIGNED(clamp_start, blocksize) && + IS_ALIGNED(clamp_end, blocksize)) + return 0; + + ret = btrfs_read_folio(NULL, folio); + if (ret) + return ret; + folio_lock(folio); + if (unlikely(!folio_test_uptodate(folio))) { + folio_unlock(folio); + return -EIO; + } + + /* + * Since btrfs_read_folio() will unlock the folio before it returns, + * there is a window where btrfs_release_folio() can be called to + * release the page. Here we check both inode mapping and page + * private to make sure the page was not released. + * + * The private flag check is essential for subpage as we need to store + * extra bitmap using folio private. + */ + if (folio->mapping != inode->i_mapping || !folio_test_private(folio)) { + folio_unlock(folio); + return -EAGAIN; } return 0; } +static gfp_t get_prepare_gfp_flags(struct inode *inode, bool nowait) +{ + gfp_t gfp; + + gfp = btrfs_alloc_write_mask(inode->i_mapping); + if (nowait) { + gfp &= ~__GFP_DIRECT_RECLAIM; + gfp |= GFP_NOWAIT; + } + + return gfp; +} + /* - * this just gets pages into the page cache and locks them down. + * Get folio into the page cache and lock it. */ -static noinline int prepare_pages(struct inode *inode, struct page **pages, - size_t num_pages, loff_t pos, - size_t write_bytes, bool force_uptodate) +static noinline int prepare_one_folio(struct inode *inode, struct folio **folio_ret, + loff_t pos, size_t write_bytes, + bool nowait) { - int i; - unsigned long index = pos >> PAGE_SHIFT; - gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); - int err = 0; - int faili; + const pgoff_t index = pos >> PAGE_SHIFT; + gfp_t mask = get_prepare_gfp_flags(inode, nowait); + fgf_t fgp_flags = (nowait ? FGP_WRITEBEGIN | FGP_NOWAIT : FGP_WRITEBEGIN) | + fgf_set_order(write_bytes); + struct folio *folio; + int ret = 0; - for (i = 0; i < num_pages; i++) { again: - pages[i] = find_or_create_page(inode->i_mapping, index + i, - mask | __GFP_WRITE); - if (!pages[i]) { - faili = i - 1; - err = -ENOMEM; - goto fail; - } - - err = set_page_extent_mapped(pages[i]); - if (err < 0) { - faili = i; - goto fail; - } - - if (i == 0) - err = prepare_uptodate_page(inode, pages[i], pos, - force_uptodate); - if (!err && i == num_pages - 1) - err = prepare_uptodate_page(inode, pages[i], - pos + write_bytes, false); - if (err) { - put_page(pages[i]); - if (err == -EAGAIN) { - err = 0; - goto again; - } - faili = i - 1; - goto fail; + folio = __filemap_get_folio(inode->i_mapping, index, fgp_flags, mask); + if (IS_ERR(folio)) + return PTR_ERR(folio); + + ret = set_folio_extent_mapped(folio); + if (ret < 0) { + folio_unlock(folio); + folio_put(folio); + return ret; + } + ret = prepare_uptodate_folio(inode, folio, pos, write_bytes); + if (ret) { + /* The folio is already unlocked. */ + folio_put(folio); + if (!nowait && ret == -EAGAIN) { + ret = 0; + goto again; } - wait_on_page_writeback(pages[i]); + return ret; } - + *folio_ret = folio; return 0; -fail: - while (faili >= 0) { - unlock_page(pages[faili]); - put_page(pages[faili]); - faili--; - } - return err; - } /* - * This function locks the extent and properly waits for data=ordered extents - * to finish before allowing the pages to be modified if need. + * Locks the extent and properly waits for data=ordered extents to finish + * before allowing the folios to be modified if need. * - * The return value: + * Return: * 1 - the extent is locked * 0 - the extent is not locked, and everything is OK - * -EAGAIN - need re-prepare the pages - * the other < 0 number - Something wrong happens + * -EAGAIN - need to prepare the folios again */ static noinline int -lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages, - size_t num_pages, loff_t pos, - size_t write_bytes, - u64 *lockstart, u64 *lockend, +lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct folio *folio, + loff_t pos, size_t write_bytes, + u64 *lockstart, u64 *lockend, bool nowait, struct extent_state **cached_state) { struct btrfs_fs_info *fs_info = inode->root->fs_info; u64 start_pos; u64 last_pos; - int i; int ret = 0; start_pos = round_down(pos, fs_info->sectorsize); @@ -1455,20 +912,28 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages, if (start_pos < inode->vfs_inode.i_size) { struct btrfs_ordered_extent *ordered; - lock_extent_bits(&inode->io_tree, start_pos, last_pos, - cached_state); + if (nowait) { + if (!btrfs_try_lock_extent(&inode->io_tree, start_pos, + last_pos, cached_state)) { + folio_unlock(folio); + folio_put(folio); + return -EAGAIN; + } + } else { + btrfs_lock_extent(&inode->io_tree, start_pos, last_pos, + cached_state); + } + ordered = btrfs_lookup_ordered_range(inode, start_pos, last_pos - start_pos + 1); if (ordered && ordered->file_offset + ordered->num_bytes > start_pos && ordered->file_offset <= last_pos) { - unlock_extent_cached(&inode->io_tree, start_pos, - last_pos, cached_state); - for (i = 0; i < num_pages; i++) { - unlock_page(pages[i]); - put_page(pages[i]); - } - btrfs_start_ordered_extent(ordered, 1); + btrfs_unlock_extent(&inode->io_tree, start_pos, last_pos, + cached_state); + folio_unlock(folio); + folio_put(folio); + btrfs_start_ordered_extent(ordered); btrfs_put_ordered_extent(ordered); return -EAGAIN; } @@ -1481,98 +946,96 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages, } /* - * We should be called after prepare_pages() which should have locked + * We should be called after prepare_one_folio() which should have locked * all pages in the range. */ - for (i = 0; i < num_pages; i++) - WARN_ON(!PageLocked(pages[i])); + WARN_ON(!folio_test_locked(folio)); return ret; } -static int check_can_nocow(struct btrfs_inode *inode, loff_t pos, +/* + * Check if we can do nocow write into the range [@pos, @pos + @write_bytes) + * + * @pos: File offset. + * @write_bytes: The length to write, will be updated to the nocow writeable + * range. + * @nowait: Indicate if we can block or not (non-blocking IO context). + * + * This function will flush ordered extents in the range to ensure proper + * nocow checks. + * + * Return: + * > 0 If we can nocow, and updates @write_bytes. + * 0 If we can't do a nocow write. + * -EAGAIN If we can't do a nocow write because snapshotting of the inode's + * root is in progress or because we are in a non-blocking IO + * context and need to block (@nowait is true). + * < 0 If an error happened. + * + * NOTE: Callers need to call btrfs_check_nocow_unlock() if we return > 0. + */ +int btrfs_check_nocow_lock(struct btrfs_inode *inode, loff_t pos, size_t *write_bytes, bool nowait) { struct btrfs_fs_info *fs_info = inode->root->fs_info; struct btrfs_root *root = inode->root; + struct extent_state *cached_state = NULL; u64 lockstart, lockend; - u64 num_bytes; - int ret; + u64 cur_offset; + int ret = 0; if (!(inode->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC))) return 0; - if (!nowait && !btrfs_drew_try_write_lock(&root->snapshot_lock)) + if (!btrfs_drew_try_write_lock(&root->snapshot_lock)) return -EAGAIN; lockstart = round_down(pos, fs_info->sectorsize); lockend = round_up(pos + *write_bytes, fs_info->sectorsize) - 1; - num_bytes = lockend - lockstart + 1; if (nowait) { - struct btrfs_ordered_extent *ordered; - - if (!try_lock_extent(&inode->io_tree, lockstart, lockend)) + if (!btrfs_try_lock_ordered_range(inode, lockstart, lockend, + &cached_state)) { + btrfs_drew_write_unlock(&root->snapshot_lock); return -EAGAIN; - - ordered = btrfs_lookup_ordered_range(inode, lockstart, - num_bytes); - if (ordered) { - btrfs_put_ordered_extent(ordered); - ret = -EAGAIN; - goto out_unlock; } } else { - btrfs_lock_and_flush_ordered_range(inode, lockstart, - lockend, NULL); + btrfs_lock_and_flush_ordered_range(inode, lockstart, lockend, + &cached_state); } - ret = can_nocow_extent(&inode->vfs_inode, lockstart, &num_bytes, - NULL, NULL, NULL, false); - if (ret <= 0) { - ret = 0; - if (!nowait) - btrfs_drew_write_unlock(&root->snapshot_lock); - } else { - *write_bytes = min_t(size_t, *write_bytes , - num_bytes - pos + lockstart); + cur_offset = lockstart; + while (cur_offset < lockend) { + u64 num_bytes = lockend - cur_offset + 1; + + ret = can_nocow_extent(inode, cur_offset, &num_bytes, NULL, nowait); + if (ret <= 0) { + /* + * If cur_offset == lockstart it means we haven't found + * any extent against which we can NOCOW, so unlock the + * snapshot lock. + */ + if (cur_offset == lockstart) + btrfs_drew_write_unlock(&root->snapshot_lock); + break; + } + cur_offset += num_bytes; } -out_unlock: - unlock_extent(&inode->io_tree, lockstart, lockend); - return ret; -} + btrfs_unlock_extent(&inode->io_tree, lockstart, lockend, &cached_state); -static int check_nocow_nolock(struct btrfs_inode *inode, loff_t pos, - size_t *write_bytes) -{ - return check_can_nocow(inode, pos, write_bytes, true); -} + /* + * cur_offset > lockstart means there's at least a partial range we can + * NOCOW, and that range can cover one or more extents. + */ + if (cur_offset > lockstart) { + *write_bytes = min_t(size_t, *write_bytes, cur_offset - pos); + return 1; + } -/* - * Check if we can do nocow write into the range [@pos, @pos + @write_bytes) - * - * @pos: File offset - * @write_bytes: The length to write, will be updated to the nocow writeable - * range - * - * This function will flush ordered extents in the range to ensure proper - * nocow checks. - * - * Return: - * >0 and update @write_bytes if we can do nocow write - * 0 if we can't do nocow write - * -EAGAIN if we can't get the needed lock or there are ordered extents - * for * (nowait == true) case - * <0 if other error happened - * - * NOTE: Callers need to release the lock by btrfs_check_nocow_unlock(). - */ -int btrfs_check_nocow_lock(struct btrfs_inode *inode, loff_t pos, - size_t *write_bytes) -{ - return check_can_nocow(inode, pos, write_bytes, false); + return ret; } void btrfs_check_nocow_unlock(struct btrfs_inode *inode) @@ -1580,51 +1043,25 @@ void btrfs_check_nocow_unlock(struct btrfs_inode *inode) btrfs_drew_write_unlock(&inode->root->snapshot_lock); } -static void update_time_for_write(struct inode *inode) -{ - struct timespec64 now; - - if (IS_NOCMTIME(inode)) - return; - - now = current_time(inode); - if (!timespec64_equal(&inode->i_mtime, &now)) - inode->i_mtime = now; - - if (!timespec64_equal(&inode->i_ctime, &now)) - inode->i_ctime = now; - - if (IS_I_VERSION(inode)) - inode_inc_iversion(inode); -} - -static int btrfs_write_check(struct kiocb *iocb, struct iov_iter *from, - size_t count) +int btrfs_write_check(struct kiocb *iocb, size_t count) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); loff_t pos = iocb->ki_pos; int ret; loff_t oldsize; - loff_t start_pos; - - if (iocb->ki_flags & IOCB_NOWAIT) { - size_t nocow_bytes = count; - /* We will allocate space in case nodatacow is not set, so bail */ - if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes) <= 0) - return -EAGAIN; - /* - * There are holes in the range or parts of the range that must - * be COWed (shared extents, RO block groups, etc), so just bail - * out. - */ - if (nocow_bytes < count) - return -EAGAIN; - } + /* + * Quickly bail out on NOWAIT writes if we don't have the nodatacow or + * prealloc flags, as without those flags we always have to COW. We will + * later check if we can really COW into the target range (using + * can_nocow_extent() at btrfs_get_blocks_direct_write()). + */ + if ((iocb->ki_flags & IOCB_NOWAIT) && + !(BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC))) + return -EAGAIN; - current->backing_dev_info = inode_to_bdi(inode); ret = file_remove_privs(file); if (ret) return ret; @@ -1635,445 +1072,376 @@ static int btrfs_write_check(struct kiocb *iocb, struct iov_iter *from, * need to start yet another transaction to update the inode as we will * update the inode when we finish writing whatever data we write. */ - update_time_for_write(inode); + if (!IS_NOCMTIME(inode)) { + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); + inode_inc_iversion(inode); + } - start_pos = round_down(pos, fs_info->sectorsize); oldsize = i_size_read(inode); - if (start_pos > oldsize) { + if (pos > oldsize) { /* Expand hole size to cover write data, preventing empty gap */ loff_t end_pos = round_up(pos + count, fs_info->sectorsize); ret = btrfs_cont_expand(BTRFS_I(inode), oldsize, end_pos); - if (ret) { - current->backing_dev_info = NULL; + if (ret) return ret; - } } return 0; } -static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb, - struct iov_iter *i) +static void release_space(struct btrfs_inode *inode, struct extent_changeset *data_reserved, + u64 start, u64 len, bool only_release_metadata) { - struct file *file = iocb->ki_filp; - loff_t pos; - struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct page **pages = NULL; - struct extent_changeset *data_reserved = NULL; - u64 release_bytes = 0; - u64 lockstart; - u64 lockend; - size_t num_written = 0; - int nrptrs; - ssize_t ret; - bool only_release_metadata = false; - bool force_page_uptodate = false; - loff_t old_isize = i_size_read(inode); - unsigned int ilock_flags = 0; - - if (iocb->ki_flags & IOCB_NOWAIT) - ilock_flags |= BTRFS_ILOCK_TRY; + if (len == 0) + return; - ret = btrfs_inode_lock(inode, ilock_flags); - if (ret < 0) - return ret; + if (only_release_metadata) { + btrfs_check_nocow_unlock(inode); + btrfs_delalloc_release_metadata(inode, len, true); + } else { + const struct btrfs_fs_info *fs_info = inode->root->fs_info; - ret = generic_write_checks(iocb, i); - if (ret <= 0) - goto out; + btrfs_delalloc_release_space(inode, data_reserved, + round_down(start, fs_info->sectorsize), + len, true); + } +} - ret = btrfs_write_check(iocb, i, ret); - if (ret < 0) - goto out; +/* + * Reserve data and metadata space for this buffered write range. + * + * Return >0 for the number of bytes reserved, which is always block aligned. + * Return <0 for error. + */ +static ssize_t reserve_space(struct btrfs_inode *inode, + struct extent_changeset **data_reserved, + u64 start, size_t *len, bool nowait, + bool *only_release_metadata) +{ + const struct btrfs_fs_info *fs_info = inode->root->fs_info; + const unsigned int block_offset = (start & (fs_info->sectorsize - 1)); + size_t reserve_bytes; + int ret; - pos = iocb->ki_pos; - nrptrs = min(DIV_ROUND_UP(iov_iter_count(i), PAGE_SIZE), - PAGE_SIZE / (sizeof(struct page *))); - nrptrs = min(nrptrs, current->nr_dirtied_pause - current->nr_dirtied); - nrptrs = max(nrptrs, 8); - pages = kmalloc_array(nrptrs, sizeof(struct page *), GFP_KERNEL); - if (!pages) { - ret = -ENOMEM; - goto out; - } + ret = btrfs_check_data_free_space(inode, data_reserved, start, *len, nowait); + if (ret < 0) { + int can_nocow; - while (iov_iter_count(i) > 0) { - struct extent_state *cached_state = NULL; - size_t offset = offset_in_page(pos); - size_t sector_offset; - size_t write_bytes = min(iov_iter_count(i), - nrptrs * (size_t)PAGE_SIZE - - offset); - size_t num_pages; - size_t reserve_bytes; - size_t dirty_pages; - size_t copied; - size_t dirty_sectors; - size_t num_sectors; - int extents_locked; + if (nowait && (ret == -ENOSPC || ret == -EAGAIN)) + return -EAGAIN; /* - * Fault pages before locking them in prepare_pages - * to avoid recursive lock + * If we don't have to COW at the offset, reserve metadata only. + * write_bytes may get smaller than requested here. */ - if (unlikely(fault_in_iov_iter_readable(i, write_bytes))) { - ret = -EFAULT; - break; - } + can_nocow = btrfs_check_nocow_lock(inode, start, len, nowait); + if (can_nocow < 0) + ret = can_nocow; + if (can_nocow > 0) + ret = 0; + if (ret) + return ret; + *only_release_metadata = true; + } - only_release_metadata = false; - sector_offset = pos & (fs_info->sectorsize - 1); + reserve_bytes = round_up(*len + block_offset, fs_info->sectorsize); + WARN_ON(reserve_bytes == 0); + ret = btrfs_delalloc_reserve_metadata(inode, reserve_bytes, + reserve_bytes, nowait); + if (ret) { + if (!*only_release_metadata) + btrfs_free_reserved_data_space(inode, *data_reserved, + start, *len); + else + btrfs_check_nocow_unlock(inode); - extent_changeset_release(data_reserved); - ret = btrfs_check_data_free_space(BTRFS_I(inode), - &data_reserved, pos, - write_bytes); - if (ret < 0) { - /* - * If we don't have to COW at the offset, reserve - * metadata only. write_bytes may get smaller than - * requested here. - */ - if (btrfs_check_nocow_lock(BTRFS_I(inode), pos, - &write_bytes) > 0) - only_release_metadata = true; - else - break; - } + if (nowait && ret == -ENOSPC) + ret = -EAGAIN; + return ret; + } + return reserve_bytes; +} - num_pages = DIV_ROUND_UP(write_bytes + offset, PAGE_SIZE); - WARN_ON(num_pages > nrptrs); - reserve_bytes = round_up(write_bytes + sector_offset, - fs_info->sectorsize); - WARN_ON(reserve_bytes == 0); - ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), - reserve_bytes); - if (ret) { - if (!only_release_metadata) - btrfs_free_reserved_data_space(BTRFS_I(inode), - data_reserved, pos, - write_bytes); - else - btrfs_check_nocow_unlock(BTRFS_I(inode)); - break; - } +/* Shrink the reserved data and metadata space from @reserved_len to @new_len. */ +static void shrink_reserved_space(struct btrfs_inode *inode, + struct extent_changeset *data_reserved, + u64 reserved_start, u64 reserved_len, + u64 new_len, bool only_release_metadata) +{ + const u64 diff = reserved_len - new_len; - release_bytes = reserve_bytes; -again: - /* - * This is going to setup the pages array with the number of - * pages we want, so we don't really need to worry about the - * contents of pages from loop to loop - */ - ret = prepare_pages(inode, pages, num_pages, - pos, write_bytes, - force_page_uptodate); - if (ret) { - btrfs_delalloc_release_extents(BTRFS_I(inode), - reserve_bytes); - break; - } + ASSERT(new_len <= reserved_len); + btrfs_delalloc_shrink_extents(inode, reserved_len, new_len); + if (only_release_metadata) + btrfs_delalloc_release_metadata(inode, diff, true); + else + btrfs_delalloc_release_space(inode, data_reserved, + reserved_start + new_len, diff, true); +} - extents_locked = lock_and_cleanup_extent_if_need( - BTRFS_I(inode), pages, - num_pages, pos, write_bytes, &lockstart, - &lockend, &cached_state); - if (extents_locked < 0) { - if (extents_locked == -EAGAIN) - goto again; - btrfs_delalloc_release_extents(BTRFS_I(inode), - reserve_bytes); - ret = extents_locked; - break; - } +/* Calculate the maximum amount of bytes we can write into one folio. */ +static size_t calc_write_bytes(const struct btrfs_inode *inode, + const struct iov_iter *iter, u64 start) +{ + const size_t max_folio_size = mapping_max_folio_size(inode->vfs_inode.i_mapping); + + return min(max_folio_size - (start & (max_folio_size - 1)), + iov_iter_count(iter)); +} + +/* + * Do the heavy-lifting work to copy one range into one folio of the page cache. + * + * Return > 0 in case we copied all bytes or just some of them. + * Return 0 if no bytes were copied, in which case the caller should retry. + * Return <0 on error. + */ +static int copy_one_range(struct btrfs_inode *inode, struct iov_iter *iter, + struct extent_changeset **data_reserved, u64 start, + bool nowait) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct extent_state *cached_state = NULL; + size_t write_bytes = calc_write_bytes(inode, iter, start); + size_t copied; + const u64 reserved_start = round_down(start, fs_info->sectorsize); + u64 reserved_len; + struct folio *folio = NULL; + int extents_locked; + u64 lockstart; + u64 lockend; + bool only_release_metadata = false; + const unsigned int bdp_flags = (nowait ? BDP_ASYNC : 0); + int ret; - copied = btrfs_copy_from_user(pos, write_bytes, pages, i); + /* + * Fault all pages before locking them in prepare_one_folio() to avoid + * recursive lock. + */ + if (unlikely(fault_in_iov_iter_readable(iter, write_bytes))) + return -EFAULT; + extent_changeset_release(*data_reserved); + ret = reserve_space(inode, data_reserved, start, &write_bytes, nowait, + &only_release_metadata); + if (ret < 0) + return ret; + reserved_len = ret; + /* Write range must be inside the reserved range. */ + ASSERT(reserved_start <= start); + ASSERT(start + write_bytes <= reserved_start + reserved_len); - num_sectors = BTRFS_BYTES_TO_BLKS(fs_info, reserve_bytes); - dirty_sectors = round_up(copied + sector_offset, - fs_info->sectorsize); - dirty_sectors = BTRFS_BYTES_TO_BLKS(fs_info, dirty_sectors); +again: + ret = balance_dirty_pages_ratelimited_flags(inode->vfs_inode.i_mapping, + bdp_flags); + if (ret) { + btrfs_delalloc_release_extents(inode, reserved_len); + release_space(inode, *data_reserved, reserved_start, reserved_len, + only_release_metadata); + return ret; + } - /* - * if we have trouble faulting in the pages, fall - * back to one page at a time - */ - if (copied < write_bytes) - nrptrs = 1; + ret = prepare_one_folio(&inode->vfs_inode, &folio, start, write_bytes, false); + if (ret) { + btrfs_delalloc_release_extents(inode, reserved_len); + release_space(inode, *data_reserved, reserved_start, reserved_len, + only_release_metadata); + return ret; + } - if (copied == 0) { - force_page_uptodate = true; - dirty_sectors = 0; - dirty_pages = 0; - } else { - force_page_uptodate = false; - dirty_pages = DIV_ROUND_UP(copied + offset, - PAGE_SIZE); - } + /* + * The reserved range goes beyond the current folio, shrink the reserved + * space to the folio boundary. + */ + if (reserved_start + reserved_len > folio_next_pos(folio)) { + const u64 last_block = folio_next_pos(folio); + + shrink_reserved_space(inode, *data_reserved, reserved_start, + reserved_len, last_block - reserved_start, + only_release_metadata); + write_bytes = last_block - start; + reserved_len = last_block - reserved_start; + } + + extents_locked = lock_and_cleanup_extent_if_need(inode, folio, start, + write_bytes, &lockstart, + &lockend, nowait, + &cached_state); + if (extents_locked < 0) { + if (!nowait && extents_locked == -EAGAIN) + goto again; - if (num_sectors > dirty_sectors) { - /* release everything except the sectors we dirtied */ - release_bytes -= dirty_sectors << fs_info->sectorsize_bits; - if (only_release_metadata) { - btrfs_delalloc_release_metadata(BTRFS_I(inode), - release_bytes, true); - } else { - u64 __pos; - - __pos = round_down(pos, - fs_info->sectorsize) + - (dirty_pages << PAGE_SHIFT); - btrfs_delalloc_release_space(BTRFS_I(inode), - data_reserved, __pos, - release_bytes, true); - } - } + btrfs_delalloc_release_extents(inode, reserved_len); + release_space(inode, *data_reserved, reserved_start, reserved_len, + only_release_metadata); + ret = extents_locked; + return ret; + } - release_bytes = round_up(copied + sector_offset, - fs_info->sectorsize); + copied = copy_folio_from_iter_atomic(folio, offset_in_folio(folio, start), + write_bytes, iter); + flush_dcache_folio(folio); - ret = btrfs_dirty_pages(BTRFS_I(inode), pages, - dirty_pages, pos, copied, - &cached_state, only_release_metadata); + if (unlikely(copied < write_bytes)) { + u64 last_block; /* - * If we have not locked the extent range, because the range's - * start offset is >= i_size, we might still have a non-NULL - * cached extent state, acquired while marking the extent range - * as delalloc through btrfs_dirty_pages(). Therefore free any - * possible cached extent state to avoid a memory leak. + * The original write range doesn't need an uptodate folio as + * the range is block aligned. But now a short copy happened. + * We cannot handle it without an uptodate folio. + * + * So just revert the range and we will retry. */ - if (extents_locked) - unlock_extent_cached(&BTRFS_I(inode)->io_tree, - lockstart, lockend, &cached_state); - else - free_extent_state(cached_state); - - btrfs_delalloc_release_extents(BTRFS_I(inode), reserve_bytes); - if (ret) { - btrfs_drop_pages(fs_info, pages, num_pages, pos, copied); - break; + if (!folio_test_uptodate(folio)) { + iov_iter_revert(iter, copied); + copied = 0; } - release_bytes = 0; - if (only_release_metadata) - btrfs_check_nocow_unlock(BTRFS_I(inode)); - - btrfs_drop_pages(fs_info, pages, num_pages, pos, copied); - - cond_resched(); + /* No copied bytes, unlock, release reserved space and exit. */ + if (copied == 0) { + if (extents_locked) + btrfs_unlock_extent(&inode->io_tree, lockstart, lockend, + &cached_state); + else + btrfs_free_extent_state(cached_state); + btrfs_delalloc_release_extents(inode, reserved_len); + release_space(inode, *data_reserved, reserved_start, reserved_len, + only_release_metadata); + btrfs_drop_folio(fs_info, folio, start, copied); + return 0; + } - balance_dirty_pages_ratelimited(inode->i_mapping); + /* Release the reserved space beyond the last block. */ + last_block = round_up(start + copied, fs_info->sectorsize); - pos += copied; - num_written += copied; + shrink_reserved_space(inode, *data_reserved, reserved_start, + reserved_len, last_block - reserved_start, + only_release_metadata); + reserved_len = last_block - reserved_start; } - kfree(pages); - - if (release_bytes) { - if (only_release_metadata) { - btrfs_check_nocow_unlock(BTRFS_I(inode)); - btrfs_delalloc_release_metadata(BTRFS_I(inode), - release_bytes, true); - } else { - btrfs_delalloc_release_space(BTRFS_I(inode), - data_reserved, - round_down(pos, fs_info->sectorsize), - release_bytes, true); - } - } + ret = btrfs_dirty_folio(inode, folio, start, copied, &cached_state, + only_release_metadata); + /* + * If we have not locked the extent range, because the range's start + * offset is >= i_size, we might still have a non-NULL cached extent + * state, acquired while marking the extent range as delalloc through + * btrfs_dirty_page(). Therefore free any possible cached extent state + * to avoid a memory leak. + */ + if (extents_locked) + btrfs_unlock_extent(&inode->io_tree, lockstart, lockend, &cached_state); + else + btrfs_free_extent_state(cached_state); - extent_changeset_free(data_reserved); - if (num_written > 0) { - pagecache_isize_extended(inode, old_isize, iocb->ki_pos); - iocb->ki_pos += num_written; + btrfs_delalloc_release_extents(inode, reserved_len); + if (ret) { + btrfs_drop_folio(fs_info, folio, start, copied); + release_space(inode, *data_reserved, reserved_start, reserved_len, + only_release_metadata); + return ret; } -out: - btrfs_inode_unlock(inode, ilock_flags); - return num_written ? num_written : ret; -} - -static ssize_t check_direct_IO(struct btrfs_fs_info *fs_info, - const struct iov_iter *iter, loff_t offset) -{ - const u32 blocksize_mask = fs_info->sectorsize - 1; - - if (offset & blocksize_mask) - return -EINVAL; - - if (iov_iter_alignment(iter) & blocksize_mask) - return -EINVAL; + if (only_release_metadata) + btrfs_check_nocow_unlock(inode); - return 0; + btrfs_drop_folio(fs_info, folio, start, copied); + return copied; } -static ssize_t btrfs_direct_write(struct kiocb *iocb, struct iov_iter *from) +ssize_t btrfs_buffered_write(struct kiocb *iocb, struct iov_iter *iter) { - const bool is_sync_write = (iocb->ki_flags & IOCB_DSYNC); struct file *file = iocb->ki_filp; - struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); loff_t pos; - ssize_t written = 0; - ssize_t written_buffered; - size_t prev_left = 0; - loff_t endbyte; - ssize_t err; + struct inode *inode = file_inode(file); + struct extent_changeset *data_reserved = NULL; + size_t num_written = 0; + ssize_t ret; + loff_t old_isize; unsigned int ilock_flags = 0; + const bool nowait = (iocb->ki_flags & IOCB_NOWAIT); - if (iocb->ki_flags & IOCB_NOWAIT) + if (nowait) ilock_flags |= BTRFS_ILOCK_TRY; - /* If the write DIO is within EOF, use a shared lock */ - if (iocb->ki_pos + iov_iter_count(from) <= i_size_read(inode)) - ilock_flags |= BTRFS_ILOCK_SHARED; + ret = btrfs_inode_lock(BTRFS_I(inode), ilock_flags); + if (ret < 0) + return ret; -relock: - err = btrfs_inode_lock(inode, ilock_flags); - if (err < 0) - return err; + /* + * We can only trust the isize with inode lock held, or it can race with + * other buffered writes and cause incorrect call of + * pagecache_isize_extended() to overwrite existing data. + */ + old_isize = i_size_read(inode); - err = generic_write_checks(iocb, from); - if (err <= 0) { - btrfs_inode_unlock(inode, ilock_flags); - return err; - } + ret = generic_write_checks(iocb, iter); + if (ret <= 0) + goto out; - err = btrfs_write_check(iocb, from, err); - if (err < 0) { - btrfs_inode_unlock(inode, ilock_flags); + ret = btrfs_write_check(iocb, ret); + if (ret < 0) goto out; - } pos = iocb->ki_pos; - /* - * Re-check since file size may have changed just before taking the - * lock or pos may have changed because of O_APPEND in generic_write_check() - */ - if ((ilock_flags & BTRFS_ILOCK_SHARED) && - pos + iov_iter_count(from) > i_size_read(inode)) { - btrfs_inode_unlock(inode, ilock_flags); - ilock_flags &= ~BTRFS_ILOCK_SHARED; - goto relock; + while (iov_iter_count(iter) > 0) { + ret = copy_one_range(BTRFS_I(inode), iter, &data_reserved, pos, nowait); + if (ret < 0) + break; + pos += ret; + num_written += ret; + cond_resched(); } - if (check_direct_IO(fs_info, from, pos)) { - btrfs_inode_unlock(inode, ilock_flags); - goto buffered; + extent_changeset_free(data_reserved); + if (num_written > 0) { + pagecache_isize_extended(inode, old_isize, iocb->ki_pos); + iocb->ki_pos += num_written; } +out: + btrfs_inode_unlock(BTRFS_I(inode), ilock_flags); + return num_written ? num_written : ret; +} - /* - * We remove IOCB_DSYNC so that we don't deadlock when iomap_dio_rw() - * calls generic_write_sync() (through iomap_dio_complete()), because - * that results in calling fsync (btrfs_sync_file()) which will try to - * lock the inode in exclusive/write mode. - */ - if (is_sync_write) - iocb->ki_flags &= ~IOCB_DSYNC; - - /* - * The iov_iter can be mapped to the same file range we are writing to. - * If that's the case, then we will deadlock in the iomap code, because - * it first calls our callback btrfs_dio_iomap_begin(), which will create - * an ordered extent, and after that it will fault in the pages that the - * iov_iter refers to. During the fault in we end up in the readahead - * pages code (starting at btrfs_readahead()), which will lock the range, - * find that ordered extent and then wait for it to complete (at - * btrfs_lock_and_flush_ordered_range()), resulting in a deadlock since - * obviously the ordered extent can never complete as we didn't submit - * yet the respective bio(s). This always happens when the buffer is - * memory mapped to the same file range, since the iomap DIO code always - * invalidates pages in the target file range (after starting and waiting - * for any writeback). - * - * So here we disable page faults in the iov_iter and then retry if we - * got -EFAULT, faulting in the pages before the retry. - */ -again: - from->nofault = true; - err = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops, &btrfs_dio_ops, - IOMAP_DIO_PARTIAL, written); - from->nofault = false; - - /* No increment (+=) because iomap returns a cumulative value. */ - if (err > 0) - written = err; +static ssize_t btrfs_encoded_write(struct kiocb *iocb, struct iov_iter *from, + const struct btrfs_ioctl_encoded_io_args *encoded) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + loff_t count; + ssize_t ret; - if (iov_iter_count(from) > 0 && (err == -EFAULT || err > 0)) { - const size_t left = iov_iter_count(from); + btrfs_inode_lock(BTRFS_I(inode), 0); + count = encoded->len; + ret = generic_write_checks_count(iocb, &count); + if (ret == 0 && count != encoded->len) { /* - * We have more data left to write. Try to fault in as many as - * possible of the remainder pages and retry. We do this without - * releasing and locking again the inode, to prevent races with - * truncate. - * - * Also, in case the iov refers to pages in the file range of the - * file we want to write to (due to a mmap), we could enter an - * infinite loop if we retry after faulting the pages in, since - * iomap will invalidate any pages in the range early on, before - * it tries to fault in the pages of the iov. So we keep track of - * how much was left of iov in the previous EFAULT and fallback - * to buffered IO in case we haven't made any progress. + * The write got truncated by generic_write_checks_count(). We + * can't do a partial encoded write. */ - if (left == prev_left) { - err = -ENOTBLK; - } else { - fault_in_iov_iter_readable(from, left); - prev_left = left; - goto again; - } + ret = -EFBIG; } - - btrfs_inode_unlock(inode, ilock_flags); - - /* - * Add back IOCB_DSYNC. Our caller, btrfs_file_write_iter(), will do - * the fsync (call generic_write_sync()). - */ - if (is_sync_write) - iocb->ki_flags |= IOCB_DSYNC; - - /* If 'err' is -ENOTBLK then it means we must fallback to buffered IO. */ - if ((err < 0 && err != -ENOTBLK) || !iov_iter_count(from)) + if (ret || encoded->len == 0) goto out; -buffered: - pos = iocb->ki_pos; - written_buffered = btrfs_buffered_write(iocb, from); - if (written_buffered < 0) { - err = written_buffered; - goto out; - } - /* - * Ensure all data is persisted. We want the next direct IO read to be - * able to read what was just written. - */ - endbyte = pos + written_buffered - 1; - err = btrfs_fdatawrite_range(inode, pos, endbyte); - if (err) - goto out; - err = filemap_fdatawait_range(inode->i_mapping, pos, endbyte); - if (err) + ret = btrfs_write_check(iocb, encoded->len); + if (ret < 0) goto out; - written += written_buffered; - iocb->ki_pos = pos + written_buffered; - invalidate_mapping_pages(file->f_mapping, pos >> PAGE_SHIFT, - endbyte >> PAGE_SHIFT); + + ret = btrfs_do_encoded_write(iocb, from, encoded); out: - return err < 0 ? err : written; + btrfs_inode_unlock(BTRFS_I(inode), 0); + return ret; } -static ssize_t btrfs_file_write_iter(struct kiocb *iocb, - struct iov_iter *from) +ssize_t btrfs_do_write_iter(struct kiocb *iocb, struct iov_iter *from, + const struct btrfs_ioctl_encoded_io_args *encoded) { struct file *file = iocb->ki_filp; struct btrfs_inode *inode = BTRFS_I(file_inode(file)); - ssize_t num_written = 0; - const bool sync = iocb->ki_flags & IOCB_DSYNC; + ssize_t num_written, num_sync; + if (unlikely(btrfs_is_shutdown(inode->root->fs_info))) + return -EIO; /* * If the fs flips readonly due to some impossible error, although we * have opened a file as writable, we have to stop this write operation @@ -2082,38 +1450,46 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, if (BTRFS_FS_ERROR(inode->root->fs_info)) return -EROFS; - if (!(iocb->ki_flags & IOCB_DIRECT) && - (iocb->ki_flags & IOCB_NOWAIT)) + if (encoded && (iocb->ki_flags & IOCB_NOWAIT)) return -EOPNOTSUPP; - if (sync) - atomic_inc(&inode->sync_writers); - - if (iocb->ki_flags & IOCB_DIRECT) + if (encoded) { + num_written = btrfs_encoded_write(iocb, from, encoded); + num_sync = encoded->len; + } else if (iocb->ki_flags & IOCB_DIRECT) { num_written = btrfs_direct_write(iocb, from); - else + num_sync = num_written; + } else { num_written = btrfs_buffered_write(iocb, from); + num_sync = num_written; + } btrfs_set_inode_last_sub_trans(inode); - if (num_written > 0) - num_written = generic_write_sync(iocb, num_written); - - if (sync) - atomic_dec(&inode->sync_writers); + if (num_sync > 0) { + num_sync = generic_write_sync(iocb, num_sync); + if (num_sync < 0) + num_written = num_sync; + } - current->backing_dev_info = NULL; return num_written; } +static ssize_t btrfs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) +{ + return btrfs_do_write_iter(iocb, from, NULL); +} + int btrfs_release_file(struct inode *inode, struct file *filp) { struct btrfs_file_private *private = filp->private_data; - if (private && private->filldir_buf) + if (private) { kfree(private->filldir_buf); - kfree(private); - filp->private_data = NULL; + btrfs_free_extent_state(private->llseek_cached_state); + kfree(private); + filp->private_data = NULL; + } /* * Set by setattr when we are about to truncate a file from a non-zero @@ -2127,7 +1503,7 @@ int btrfs_release_file(struct inode *inode, struct file *filp) return 0; } -static int start_ordered_ops(struct inode *inode, loff_t start, loff_t end) +static int start_ordered_ops(struct btrfs_inode *inode, loff_t start, loff_t end) { int ret; struct blk_plug plug; @@ -2139,9 +1515,7 @@ static int start_ordered_ops(struct inode *inode, loff_t start, loff_t end) * several segments of stripe length (currently 64K). */ blk_start_plug(&plug); - atomic_inc(&BTRFS_I(inode)->sync_writers); ret = btrfs_fdatawrite_range(inode, start, end); - atomic_dec(&BTRFS_I(inode)->sync_writers); blk_finish_plug(&plug); return ret; @@ -2149,10 +1523,10 @@ static int start_ordered_ops(struct inode *inode, loff_t start, loff_t end) static inline bool skip_inode_logging(const struct btrfs_log_ctx *ctx) { - struct btrfs_inode *inode = BTRFS_I(ctx->inode); + struct btrfs_inode *inode = ctx->inode; struct btrfs_fs_info *fs_info = inode->root->fs_info; - if (btrfs_inode_in_log(inode, fs_info->generation) && + if (btrfs_inode_in_log(inode, btrfs_get_fs_generation(fs_info)) && list_empty(&ctx->ordered_extents)) return true; @@ -2163,7 +1537,7 @@ static inline bool skip_inode_logging(const struct btrfs_log_ctx *ctx) * and for a fast fsync we don't wait for that, we only wait for the * writeback to complete. */ - if (inode->last_trans <= fs_info->last_trans_committed && + if (inode->last_trans <= btrfs_get_last_trans_committed(fs_info) && (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags) || list_empty(&ctx->ordered_extents))) return true; @@ -2185,14 +1559,21 @@ static inline bool skip_inode_logging(const struct btrfs_log_ctx *ctx) int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) { struct dentry *dentry = file_dentry(file); - struct inode *inode = d_inode(dentry); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_inode *inode = BTRFS_I(d_inode(dentry)); + struct btrfs_root *root = inode->root; + struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_trans_handle *trans; struct btrfs_log_ctx ctx; int ret = 0, err; u64 len; bool full_sync; + bool skip_ilock = false; + + if (current->journal_info == BTRFS_TRANS_DIO_WRITE_STUB) { + skip_ilock = true; + current->journal_info = NULL; + btrfs_assert_inode_locked(inode); + } trace_btrfs_sync_file(file, datasync); @@ -2220,19 +1601,14 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) if (ret) goto out; - btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP); + if (skip_ilock) + down_write(&inode->i_mmap_lock); + else + btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP); atomic_inc(&root->log_batch); /* - * Always check for the full sync flag while holding the inode's lock, - * to avoid races with other tasks. The flag must be either set all the - * time during logging or always off all the time while logging. - */ - full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, - &BTRFS_I(inode)->runtime_flags); - - /* * Before we acquired the inode's lock and the mmap lock, someone may * have dirtied more pages in the target range. We need to make sure * that writeback for any such pages does not start while we are logging @@ -2252,11 +1628,24 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) */ ret = start_ordered_ops(inode, start, end); if (ret) { - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + if (skip_ilock) + up_write(&inode->i_mmap_lock); + else + btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); goto out; } /* + * Always check for the full sync flag while holding the inode's lock, + * to avoid races with other tasks. The flag must be either set all the + * time during logging or always off all the time while logging. + * We check the flag here after starting delalloc above, because when + * running delalloc the full sync flag may be set if we need to drop + * extra extent map ranges due to temporary memory allocation failures. + */ + full_sync = test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags); + + /* * We have to do this here to avoid the priority inversion of waiting on * IO of a lower priority task while holding a transaction open. * @@ -2273,15 +1662,29 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) */ if (full_sync || btrfs_is_zoned(fs_info)) { ret = btrfs_wait_ordered_range(inode, start, len); + clear_bit(BTRFS_INODE_COW_WRITE_ERROR, &inode->runtime_flags); } else { /* * Get our ordered extents as soon as possible to avoid doing * checksum lookups in the csum tree, and use instead the * checksums attached to the ordered extents. */ - btrfs_get_ordered_extents_for_logging(BTRFS_I(inode), - &ctx.ordered_extents); - ret = filemap_fdatawait_range(inode->i_mapping, start, end); + btrfs_get_ordered_extents_for_logging(inode, &ctx.ordered_extents); + ret = filemap_fdatawait_range(inode->vfs_inode.i_mapping, start, end); + if (ret) + goto out_release_extents; + + /* + * Check and clear the BTRFS_INODE_COW_WRITE_ERROR now after + * starting and waiting for writeback, because for buffered IO + * it may have been set during the end IO callback + * (end_bbio_data_write() -> btrfs_finish_ordered_extent()) in + * case an error happened and we need to wait for ordered + * extents to complete so that any extent maps that point to + * unwritten locations are dropped and we don't log them. + */ + if (test_and_clear_bit(BTRFS_INODE_COW_WRITE_ERROR, &inode->runtime_flags)) + ret = btrfs_wait_ordered_range(inode, start, len); } if (ret) @@ -2289,15 +1692,13 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) atomic_inc(&root->log_batch); - smp_mb(); if (skip_inode_logging(&ctx)) { /* * We've had everything committed since the last time we were * modified so clear this flag in case it was set for whatever * reason, it's no longer relevant. */ - clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, - &BTRFS_I(inode)->runtime_flags); + clear_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags); /* * An ordered extent might have started before and completed * already with io errors, in which case the inode was not @@ -2305,10 +1706,12 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) * for any errors that might have happened since we last * checked called fsync. */ - ret = filemap_check_wb_err(inode->i_mapping, file->f_wb_err); + ret = filemap_check_wb_err(inode->vfs_inode.i_mapping, file->f_wb_err); goto out_release_extents; } + btrfs_init_log_ctx_scratch_eb(&ctx); + /* * We use start here because we will need to wait on the IO to complete * in btrfs_sync_log, which could require joining a transaction (for @@ -2328,10 +1731,19 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) trans->in_fsync = true; ret = btrfs_log_dentry_safe(trans, dentry, &ctx); + /* + * Scratch eb no longer needed, release before syncing log or commit + * transaction, to avoid holding unnecessary memory during such long + * operations. + */ + if (ctx.scratch_eb) { + free_extent_buffer(ctx.scratch_eb); + ctx.scratch_eb = NULL; + } btrfs_release_log_ctx_extents(&ctx); if (ret < 0) { /* Fallthrough and commit/free transaction. */ - ret = 1; + ret = BTRFS_LOG_FORCE_COMMIT; } /* we've logged all the items and now have a consistent @@ -2344,29 +1756,71 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) * file again, but that will end up using the synchronization * inside btrfs_sync_log to keep things safe. */ - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + if (skip_ilock) + up_write(&inode->i_mmap_lock); + else + btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + + if (ret == BTRFS_NO_LOG_SYNC) { + ret = btrfs_end_transaction(trans); + goto out; + } - if (ret != BTRFS_NO_LOG_SYNC) { + /* We successfully logged the inode, attempt to sync the log. */ + if (!ret) { + ret = btrfs_sync_log(trans, root, &ctx); if (!ret) { - ret = btrfs_sync_log(trans, root, &ctx); - if (!ret) { - ret = btrfs_end_transaction(trans); - goto out; - } - } - if (!full_sync) { - ret = btrfs_wait_ordered_range(inode, start, len); - if (ret) { - btrfs_end_transaction(trans); - goto out; - } + ret = btrfs_end_transaction(trans); + goto out; } - ret = btrfs_commit_transaction(trans); - } else { + } + + /* + * At this point we need to commit the transaction because we had + * btrfs_need_log_full_commit() or some other error. + * + * If we didn't do a full sync we have to stop the trans handle, wait on + * the ordered extents, start it again and commit the transaction. If + * we attempt to wait on the ordered extents here we could deadlock with + * something like fallocate() that is holding the extent lock trying to + * start a transaction while some other thread is trying to commit the + * transaction while we (fsync) are currently holding the transaction + * open. + */ + if (!full_sync) { ret = btrfs_end_transaction(trans); + if (ret) + goto out; + ret = btrfs_wait_ordered_range(inode, start, len); + if (ret) + goto out; + + /* + * This is safe to use here because we're only interested in + * making sure the transaction that had the ordered extents is + * committed. We aren't waiting on anything past this point, + * we're purely getting the transaction and committing it. + */ + trans = btrfs_attach_transaction_barrier(root); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + + /* + * We committed the transaction and there's no currently + * running transaction, this means everything we care + * about made it to disk and we are done. + */ + if (ret == -ENOENT) + ret = 0; + goto out; + } } + + ret = btrfs_commit_transaction(trans); out: + free_extent_buffer(ctx.scratch_eb); ASSERT(list_empty(&ctx.list)); + ASSERT(list_empty(&ctx.conflict_inodes)); err = file_check_and_advance_wb_err(file); if (!ret) ret = err; @@ -2374,56 +1828,260 @@ out: out_release_extents: btrfs_release_log_ctx_extents(&ctx); - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + if (skip_ilock) + up_write(&inode->i_mmap_lock); + else + btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); goto out; } +/* + * btrfs_page_mkwrite() is not allowed to change the file size as it gets + * called from a page fault handler when a page is first dirtied. Hence we must + * be careful to check for EOF conditions here. We set the page up correctly + * for a written page which means we get ENOSPC checking when writing into + * holes and correct delalloc and unwritten extent mapping on filesystems that + * support these features. + * + * We are not allowed to take the i_mutex here so we have to play games to + * protect against truncate races as the page could now be beyond EOF. Because + * truncate_setsize() writes the inode size before removing pages, once we have + * the page lock we can determine safely if the page is beyond EOF. If it is not + * beyond EOF, then the page is guaranteed safe against truncation until we + * unlock the page. + */ +static vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf) +{ + struct page *page = vmf->page; + struct folio *folio = page_folio(page); + struct btrfs_inode *inode = BTRFS_I(file_inode(vmf->vma->vm_file)); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct extent_io_tree *io_tree = &inode->io_tree; + struct btrfs_ordered_extent *ordered; + struct extent_state *cached_state = NULL; + struct extent_changeset *data_reserved = NULL; + unsigned long zero_start; + loff_t size; + size_t fsize = folio_size(folio); + int ret; + bool only_release_metadata = false; + u64 reserved_space; + u64 page_start; + u64 page_end; + u64 end; + + reserved_space = fsize; + + sb_start_pagefault(inode->vfs_inode.i_sb); + page_start = folio_pos(folio); + page_end = page_start + folio_size(folio) - 1; + end = page_end; + + /* + * Reserving delalloc space after obtaining the page lock can lead to + * deadlock. For example, if a dirty page is locked by this function + * and the call to btrfs_delalloc_reserve_space() ends up triggering + * dirty page write out, then the btrfs_writepages() function could + * end up waiting indefinitely to get a lock on the page currently + * being processed by btrfs_page_mkwrite() function. + */ + ret = btrfs_check_data_free_space(inode, &data_reserved, page_start, + reserved_space, false); + if (ret < 0) { + size_t write_bytes = reserved_space; + + if (btrfs_check_nocow_lock(inode, page_start, &write_bytes, false) <= 0) + goto out_noreserve; + + only_release_metadata = true; + + /* + * Can't write the whole range, there may be shared extents or + * holes in the range, bail out with @only_release_metadata set + * to true so that we unlock the nocow lock before returning the + * error. + */ + if (write_bytes < reserved_space) + goto out_noreserve; + } + ret = btrfs_delalloc_reserve_metadata(inode, reserved_space, + reserved_space, false); + if (ret < 0) { + if (!only_release_metadata) + btrfs_free_reserved_data_space(inode, data_reserved, + page_start, reserved_space); + goto out_noreserve; + } + + ret = file_update_time(vmf->vma->vm_file); + if (ret < 0) + goto out; +again: + down_read(&inode->i_mmap_lock); + folio_lock(folio); + size = i_size_read(&inode->vfs_inode); + + if ((folio->mapping != inode->vfs_inode.i_mapping) || + (page_start >= size)) { + /* Page got truncated out from underneath us. */ + goto out_unlock; + } + folio_wait_writeback(folio); + + btrfs_lock_extent(io_tree, page_start, page_end, &cached_state); + ret = set_folio_extent_mapped(folio); + if (ret < 0) { + btrfs_unlock_extent(io_tree, page_start, page_end, &cached_state); + goto out_unlock; + } + + /* + * We can't set the delalloc bits if there are pending ordered + * extents. Drop our locks and wait for them to finish. + */ + ordered = btrfs_lookup_ordered_range(inode, page_start, fsize); + if (ordered) { + btrfs_unlock_extent(io_tree, page_start, page_end, &cached_state); + folio_unlock(folio); + up_read(&inode->i_mmap_lock); + btrfs_start_ordered_extent(ordered); + btrfs_put_ordered_extent(ordered); + goto again; + } + + if (folio_contains(folio, (size - 1) >> PAGE_SHIFT)) { + reserved_space = round_up(size - page_start, fs_info->sectorsize); + if (reserved_space < fsize) { + const u64 to_free = fsize - reserved_space; + + end = page_start + reserved_space - 1; + if (only_release_metadata) + btrfs_delalloc_release_metadata(inode, to_free, true); + else + btrfs_delalloc_release_space(inode, data_reserved, + end + 1, to_free, true); + } + } + + /* + * page_mkwrite gets called when the page is firstly dirtied after it's + * faulted in, but write(2) could also dirty a page and set delalloc + * bits, thus in this case for space account reason, we still need to + * clear any delalloc bits within this page range since we have to + * reserve data&meta space before lock_page() (see above comments). + */ + btrfs_clear_extent_bit(io_tree, page_start, end, + EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | + EXTENT_DEFRAG, &cached_state); + + ret = btrfs_set_extent_delalloc(inode, page_start, end, 0, &cached_state); + if (ret < 0) { + btrfs_unlock_extent(io_tree, page_start, page_end, &cached_state); + goto out_unlock; + } + + /* Page is wholly or partially inside EOF. */ + if (page_start + folio_size(folio) > size) + zero_start = offset_in_folio(folio, size); + else + zero_start = fsize; + + if (zero_start != fsize) + folio_zero_range(folio, zero_start, folio_size(folio) - zero_start); + + btrfs_folio_clear_checked(fs_info, folio, page_start, fsize); + btrfs_folio_set_dirty(fs_info, folio, page_start, end + 1 - page_start); + btrfs_folio_set_uptodate(fs_info, folio, page_start, end + 1 - page_start); + + btrfs_set_inode_last_sub_trans(inode); + + if (only_release_metadata) + btrfs_set_extent_bit(io_tree, page_start, end, EXTENT_NORESERVE, + &cached_state); + + btrfs_unlock_extent(io_tree, page_start, page_end, &cached_state); + up_read(&inode->i_mmap_lock); + + btrfs_delalloc_release_extents(inode, fsize); + if (only_release_metadata) + btrfs_check_nocow_unlock(inode); + sb_end_pagefault(inode->vfs_inode.i_sb); + extent_changeset_free(data_reserved); + return VM_FAULT_LOCKED; + +out_unlock: + folio_unlock(folio); + up_read(&inode->i_mmap_lock); +out: + btrfs_delalloc_release_extents(inode, fsize); + if (only_release_metadata) + btrfs_delalloc_release_metadata(inode, reserved_space, true); + else + btrfs_delalloc_release_space(inode, data_reserved, page_start, + reserved_space, true); + extent_changeset_free(data_reserved); +out_noreserve: + if (only_release_metadata) + btrfs_check_nocow_unlock(inode); + + sb_end_pagefault(inode->vfs_inode.i_sb); + + if (ret < 0) + return vmf_error(ret); + + /* Make the VM retry the fault. */ + return VM_FAULT_NOPAGE; +} + static const struct vm_operations_struct btrfs_file_vm_ops = { .fault = filemap_fault, .map_pages = filemap_map_pages, .page_mkwrite = btrfs_page_mkwrite, }; -static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma) +static int btrfs_file_mmap_prepare(struct vm_area_desc *desc) { + struct file *filp = desc->file; struct address_space *mapping = filp->f_mapping; - if (!mapping->a_ops->readpage) + if (unlikely(btrfs_is_shutdown(inode_to_fs_info(file_inode(filp))))) + return -EIO; + if (!mapping->a_ops->read_folio) return -ENOEXEC; file_accessed(filp); - vma->vm_ops = &btrfs_file_vm_ops; + desc->vm_ops = &btrfs_file_vm_ops; return 0; } -static int hole_mergeable(struct btrfs_inode *inode, struct extent_buffer *leaf, - int slot, u64 start, u64 end) +static bool hole_mergeable(struct btrfs_inode *inode, struct extent_buffer *leaf, + int slot, u64 start, u64 end) { struct btrfs_file_extent_item *fi; struct btrfs_key key; if (slot < 0 || slot >= btrfs_header_nritems(leaf)) - return 0; + return false; btrfs_item_key_to_cpu(leaf, &key, slot); if (key.objectid != btrfs_ino(inode) || key.type != BTRFS_EXTENT_DATA_KEY) - return 0; + return false; fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG) - return 0; + return false; if (btrfs_file_extent_disk_bytenr(leaf, fi)) - return 0; + return false; if (key.offset == end) - return 1; + return true; if (key.offset + btrfs_file_extent_num_bytes(leaf, fi) == start) - return 1; - return 0; + return true; + return false; } static int fill_holes(struct btrfs_trans_handle *trans, @@ -2435,7 +2093,6 @@ static int fill_holes(struct btrfs_trans_handle *trans, struct extent_buffer *leaf; struct btrfs_file_extent_item *fi; struct extent_map *hole_em; - struct extent_map_tree *em_tree = &inode->extent_tree; struct btrfs_key key; int ret; @@ -2469,7 +2126,7 @@ static int fill_holes(struct btrfs_trans_handle *trans, btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes); btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes); btrfs_set_file_extent_offset(leaf, fi, 0); - btrfs_mark_buffer_dirty(leaf); + btrfs_set_file_extent_generation(leaf, fi, trans->transid); goto out; } @@ -2477,7 +2134,7 @@ static int fill_holes(struct btrfs_trans_handle *trans, u64 num_bytes; key.offset = offset; - btrfs_set_item_key_safe(fs_info, path, &key); + btrfs_set_item_key_safe(trans, path, &key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); num_bytes = btrfs_file_extent_num_bytes(leaf, fi) + end - @@ -2485,45 +2142,36 @@ static int fill_holes(struct btrfs_trans_handle *trans, btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes); btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes); btrfs_set_file_extent_offset(leaf, fi, 0); - btrfs_mark_buffer_dirty(leaf); + btrfs_set_file_extent_generation(leaf, fi, trans->transid); goto out; } btrfs_release_path(path); - ret = btrfs_insert_file_extent(trans, root, btrfs_ino(inode), - offset, 0, 0, end - offset, 0, end - offset, 0, 0, 0); + ret = btrfs_insert_hole_extent(trans, root, btrfs_ino(inode), offset, + end - offset); if (ret) return ret; out: btrfs_release_path(path); - hole_em = alloc_extent_map(); + hole_em = btrfs_alloc_extent_map(); if (!hole_em) { - btrfs_drop_extent_cache(inode, offset, end - 1, 0); - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags); + btrfs_drop_extent_map_range(inode, offset, end - 1, false); + btrfs_set_inode_full_sync(inode); } else { hole_em->start = offset; hole_em->len = end - offset; hole_em->ram_bytes = hole_em->len; - hole_em->orig_start = offset; - hole_em->block_start = EXTENT_MAP_HOLE; - hole_em->block_len = 0; - hole_em->orig_block_len = 0; - hole_em->compress_type = BTRFS_COMPRESS_NONE; + hole_em->disk_bytenr = EXTENT_MAP_HOLE; + hole_em->disk_num_bytes = 0; hole_em->generation = trans->transid; - do { - btrfs_drop_extent_cache(inode, offset, end - 1, 0); - write_lock(&em_tree->lock); - ret = add_extent_mapping(em_tree, hole_em, 1); - write_unlock(&em_tree->lock); - } while (ret == -EEXIST); - free_extent_map(hole_em); + ret = btrfs_replace_extent_map_range(inode, hole_em, true); + btrfs_free_extent_map(hole_em); if (ret) - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, - &inode->runtime_flags); + btrfs_set_inode_full_sync(inode); } return 0; @@ -2541,28 +2189,46 @@ static int find_first_non_hole(struct btrfs_inode *inode, u64 *start, u64 *len) struct extent_map *em; int ret = 0; - em = btrfs_get_extent(inode, NULL, 0, + em = btrfs_get_extent(inode, NULL, round_down(*start, fs_info->sectorsize), round_up(*len, fs_info->sectorsize)); if (IS_ERR(em)) return PTR_ERR(em); /* Hole or vacuum extent(only exists in no-hole mode) */ - if (em->block_start == EXTENT_MAP_HOLE) { + if (em->disk_bytenr == EXTENT_MAP_HOLE) { ret = 1; *len = em->start + em->len > *start + *len ? 0 : *start + *len - em->start - em->len; *start = em->start + em->len; } - free_extent_map(em); + btrfs_free_extent_map(em); return ret; } -static int btrfs_punch_hole_lock_range(struct inode *inode, - const u64 lockstart, - const u64 lockend, - struct extent_state **cached_state) +/* + * Check if there is no folio in the range. + * + * We cannot utilize filemap_range_has_page() in a filemap with large folios + * as we can hit the following false positive: + * + * start end + * | | + * |//|//|//|//| | | | | | | | |//|//| + * \ / \ / + * Folio A Folio B + * + * That large folio A and B cover the start and end indexes. + * In that case filemap_range_has_page() will always return true, but the above + * case is fine for btrfs_punch_hole_lock_range() usage. + * + * So here we only ensure that no other folios is in the range, excluding the + * head/tail large folio. + */ +static bool check_range_has_page(struct inode *inode, u64 start, u64 end) { + struct folio_batch fbatch; + bool ret = false; /* * For subpage case, if the range is not at page boundary, we could * have pages at the leading/tailing part of the range. @@ -2570,45 +2236,66 @@ static int btrfs_punch_hole_lock_range(struct inode *inode, * will always return true. * So here we need to do extra page alignment for * filemap_range_has_page(). + * + * And do not decrease page_lockend right now, as it can be 0. */ - const u64 page_lockstart = round_up(lockstart, PAGE_SIZE); - const u64 page_lockend = round_down(lockend + 1, PAGE_SIZE) - 1; + const u64 page_lockstart = round_up(start, PAGE_SIZE); + const u64 page_lockend = round_down(end + 1, PAGE_SIZE); + const pgoff_t start_index = page_lockstart >> PAGE_SHIFT; + const pgoff_t end_index = (page_lockend - 1) >> PAGE_SHIFT; + pgoff_t tmp = start_index; + int found_folios; + + /* The same page or adjacent pages. */ + if (page_lockend <= page_lockstart) + return false; + + folio_batch_init(&fbatch); + found_folios = filemap_get_folios(inode->i_mapping, &tmp, end_index, &fbatch); + for (int i = 0; i < found_folios; i++) { + struct folio *folio = fbatch.folios[i]; + + /* A large folio begins before the start. Not a target. */ + if (folio->index < start_index) + continue; + /* A large folio extends beyond the end. Not a target. */ + if (folio_next_index(folio) > end_index) + continue; + /* A folio doesn't cover the head/tail index. Found a target. */ + ret = true; + break; + } + folio_batch_release(&fbatch); + return ret; +} +static void btrfs_punch_hole_lock_range(struct inode *inode, + const u64 lockstart, const u64 lockend, + struct extent_state **cached_state) +{ while (1) { - struct btrfs_ordered_extent *ordered; - int ret; - truncate_pagecache_range(inode, lockstart, lockend); - lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend, - cached_state); - ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode), - lockend); - + btrfs_lock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend, + cached_state); /* - * We need to make sure we have no ordered extents in this range - * and nobody raced in and read a page in this range, if we did - * we need to try again. + * We can't have ordered extents in the range, nor dirty/writeback + * pages, because we have locked the inode's VFS lock in exclusive + * mode, we have locked the inode's i_mmap_lock in exclusive mode, + * we have flushed all delalloc in the range and we have waited + * for any ordered extents in the range to complete. + * We can race with anyone reading pages from this range, so after + * locking the range check if we have pages in the range, and if + * we do, unlock the range and retry. */ - if ((!ordered || - (ordered->file_offset + ordered->num_bytes <= lockstart || - ordered->file_offset > lockend)) && - !filemap_range_has_page(inode->i_mapping, - page_lockstart, page_lockend)) { - if (ordered) - btrfs_put_ordered_extent(ordered); + if (!check_range_has_page(inode, lockstart, lockend)) break; - } - if (ordered) - btrfs_put_ordered_extent(ordered); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, - lockend, cached_state); - ret = btrfs_wait_ordered_range(inode, lockstart, - lockend - lockstart + 1); - if (ret) - return ret; + + btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend, + cached_state); } - return 0; + + btrfs_assert_inode_range_clean(BTRFS_I(inode), lockstart, lockend); } static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, @@ -2624,7 +2311,6 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, struct extent_buffer *leaf; struct btrfs_key key; int slot; - struct btrfs_ref ref = { 0 }; int ret; if (replace_len == 0) @@ -2654,7 +2340,6 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, btrfs_set_file_extent_num_bytes(leaf, extent, replace_len); if (extent_info->is_new_extent) btrfs_set_file_extent_generation(leaf, extent, trans->transid); - btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); ret = btrfs_inode_set_file_extent_range(inode, extent_info->file_offset, @@ -2680,14 +2365,17 @@ static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans, extent_info->qgroup_reserved, &key); } else { + struct btrfs_ref ref = { + .action = BTRFS_ADD_DELAYED_REF, + .bytenr = extent_info->disk_offset, + .num_bytes = extent_info->disk_len, + .owning_root = btrfs_root_id(root), + .ref_root = btrfs_root_id(root), + }; u64 ref_offset; - btrfs_init_generic_ref(&ref, BTRFS_ADD_DELAYED_REF, - extent_info->disk_offset, - extent_info->disk_len, 0); ref_offset = extent_info->file_offset - extent_info->data_offset; - btrfs_init_data_ref(&ref, root->root_key.objectid, - btrfs_ino(inode), ref_offset, 0, false); + btrfs_init_data_ref(&ref, btrfs_ino(inode), ref_offset, 0, false); ret = btrfs_inc_extent_ref(trans, &ref); } @@ -2717,7 +2405,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, u64 min_size = btrfs_calc_insert_metadata_size(fs_info, 1); u64 ino_size = round_up(inode->vfs_inode.i_size, fs_info->sectorsize); struct btrfs_trans_handle *trans = NULL; - struct btrfs_block_rsv *rsv; + struct btrfs_block_rsv rsv; unsigned int rsv_count; u64 cur_offset; u64 len = end - start; @@ -2726,13 +2414,9 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, if (end <= start) return -EINVAL; - rsv = btrfs_alloc_block_rsv(fs_info, BTRFS_BLOCK_RSV_TEMP); - if (!rsv) { - ret = -ENOMEM; - goto out; - } - rsv->size = btrfs_calc_insert_metadata_size(fs_info, 1); - rsv->failfast = 1; + btrfs_init_metadata_block_rsv(fs_info, &rsv, BTRFS_BLOCK_RSV_TEMP); + rsv.size = btrfs_calc_insert_metadata_size(fs_info, 1); + rsv.failfast = true; /* * 1 - update the inode @@ -2749,13 +2433,14 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, if (IS_ERR(trans)) { ret = PTR_ERR(trans); trans = NULL; - goto out_free; + goto out_release; } - ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, + ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, &rsv, min_size, false); - BUG_ON(ret); - trans->block_rsv = rsv; + if (WARN_ON(ret)) + goto out_trans; + trans->block_rsv = &rsv; cur_offset = start; drop_args.path = path; @@ -2777,9 +2462,9 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, * got EOPNOTSUPP via prealloc then we messed up and * need to abort. */ - if (ret && - (ret != -EOPNOTSUPP || - (extent_info && extent_info->is_new_extent))) + if (unlikely(ret && + (ret != -EOPNOTSUPP || + (extent_info && extent_info->is_new_extent)))) btrfs_abort_transaction(trans, ret); break; } @@ -2790,7 +2475,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, cur_offset < ino_size) { ret = fill_holes(trans, inode, path, cur_offset, drop_args.drop_end); - if (ret) { + if (unlikely(ret)) { /* * If we failed then we didn't insert our hole * entries for the area we dropped, so now the @@ -2810,7 +2495,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, ret = btrfs_inode_clear_file_extent_range(inode, cur_offset, drop_args.drop_end - cur_offset); - if (ret) { + if (unlikely(ret)) { /* * We couldn't clear our area, so we could * presumably adjust up and corrupt the fs, so @@ -2829,7 +2514,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, ret = btrfs_insert_replace_extent(trans, inode, path, extent_info, replace_len, drop_args.bytes_found); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); break; } @@ -2838,7 +2523,25 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, extent_info->file_offset += replace_len; } - ret = btrfs_update_inode(trans, root, inode); + /* + * We are releasing our handle on the transaction, balance the + * dirty pages of the btree inode and flush delayed items, and + * then get a new transaction handle, which may now point to a + * new transaction in case someone else may have committed the + * transaction we used to replace/drop file extent items. So + * bump the inode's iversion and update mtime and ctime except + * if we are called from a dedupe context. This is because a + * power failure/crash may happen after the transaction is + * committed and before we finish replacing/dropping all the + * file extent items we need. + */ + inode_inc_iversion(&inode->vfs_inode); + + if (!extent_info || extent_info->update_times) + inode_set_mtime_to_ts(&inode->vfs_inode, + inode_set_ctime_current(&inode->vfs_inode)); + + ret = btrfs_update_inode(trans, inode); if (ret) break; @@ -2853,9 +2556,10 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, } ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, - rsv, min_size, false); - BUG_ON(ret); /* shouldn't happen */ - trans->block_rsv = rsv; + &rsv, min_size, false); + if (WARN_ON(ret)) + break; + trans->block_rsv = &rsv; cur_offset = drop_args.drop_end; len = end - cur_offset; @@ -2877,7 +2581,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, * maps for the replacement extents (or holes). */ if (extent_info && !extent_info->is_new_extent) - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &inode->runtime_flags); + btrfs_set_inode_full_sync(inode); if (ret) goto out_trans; @@ -2905,7 +2609,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, cur_offset < drop_args.drop_end) { ret = fill_holes(trans, inode, path, cur_offset, drop_args.drop_end); - if (ret) { + if (unlikely(ret)) { /* Same comment as above. */ btrfs_abort_transaction(trans, ret); goto out_trans; @@ -2914,7 +2618,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, /* See the comment in the loop above for the reasoning here. */ ret = btrfs_inode_clear_file_extent_range(inode, cur_offset, drop_args.drop_end - cur_offset); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); goto out_trans; } @@ -2924,7 +2628,7 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, ret = btrfs_insert_replace_extent(trans, inode, path, extent_info, extent_info->data_len, drop_args.bytes_found); - if (ret) { + if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); goto out_trans; } @@ -2932,22 +2636,22 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, out_trans: if (!trans) - goto out_free; + goto out_release; trans->block_rsv = &fs_info->trans_block_rsv; if (ret) btrfs_end_transaction(trans); else *trans_out = trans; -out_free: - btrfs_free_block_rsv(fs_info, rsv); -out: +out_release: + btrfs_block_rsv_release(fs_info, &rsv, (u64)-1, NULL); return ret; } -static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) +static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len) { - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct inode *inode = file_inode(file); + struct btrfs_fs_info *fs_info = inode_to_fs_info(inode); struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_state *cached_state = NULL; struct btrfs_path *path; @@ -2956,18 +2660,20 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) u64 lockend; u64 tail_start; u64 tail_len; - u64 orig_start = offset; + const u64 orig_start = offset; + const u64 orig_end = offset + len - 1; int ret = 0; bool same_block; u64 ino_size; bool truncated_block = false; bool updated_inode = false; - ret = btrfs_wait_ordered_range(inode, offset, len); + btrfs_inode_lock(BTRFS_I(inode), BTRFS_ILOCK_MMAP); + + ret = btrfs_wait_ordered_range(BTRFS_I(inode), offset, len); if (ret) - return ret; + goto out_only_mutex; - btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP); ino_size = round_up(inode->i_size, fs_info->sectorsize); ret = find_first_non_hole(BTRFS_I(inode), &offset, &len); if (ret < 0) @@ -2978,24 +2684,23 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) goto out_only_mutex; } - lockstart = round_up(offset, btrfs_inode_sectorsize(BTRFS_I(inode))); - lockend = round_down(offset + len, - btrfs_inode_sectorsize(BTRFS_I(inode))) - 1; + ret = file_modified(file); + if (ret) + goto out_only_mutex; + + lockstart = round_up(offset, fs_info->sectorsize); + lockend = round_down(offset + len, fs_info->sectorsize) - 1; same_block = (BTRFS_BYTES_TO_BLKS(fs_info, offset)) == (BTRFS_BYTES_TO_BLKS(fs_info, offset + len - 1)); /* - * We needn't truncate any block which is beyond the end of the file - * because we are sure there is no data there. - */ - /* * Only do this if we are in the same block and we aren't doing the * entire block. */ if (same_block && len < fs_info->sectorsize) { if (offset < ino_size) { truncated_block = true; - ret = btrfs_truncate_block(BTRFS_I(inode), offset, len, - 0); + ret = btrfs_truncate_block(BTRFS_I(inode), offset + len - 1, + orig_start, orig_end); } else { ret = 0; } @@ -3005,9 +2710,9 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) /* zero back part of the first block */ if (offset < ino_size) { truncated_block = true; - ret = btrfs_truncate_block(BTRFS_I(inode), offset, 0, 0); + ret = btrfs_truncate_block(BTRFS_I(inode), offset, orig_start, orig_end); if (ret) { - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + btrfs_inode_unlock(BTRFS_I(inode), BTRFS_ILOCK_MMAP); return ret; } } @@ -3042,8 +2747,8 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) if (tail_start + tail_len < ino_size) { truncated_block = true; ret = btrfs_truncate_block(BTRFS_I(inode), - tail_start + tail_len, - 0, 1); + tail_start + tail_len - 1, + orig_start, orig_end); if (ret) goto out_only_mutex; } @@ -3055,10 +2760,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) goto out_only_mutex; } - ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend, - &cached_state); - if (ret) - goto out_only_mutex; + btrfs_punch_hole_lock_range(inode, lockstart, lockend, &cached_state); path = btrfs_alloc_path(); if (!path) { @@ -3074,14 +2776,14 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) ASSERT(trans != NULL); inode_inc_iversion(inode); - inode->i_mtime = inode->i_ctime = current_time(inode); - ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); + inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); + ret = btrfs_update_inode(trans, BTRFS_I(inode)); updated_inode = true; btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); out: - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend, - &cached_state); + btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend, + &cached_state); out_only_mutex: if (!updated_inode && truncated_block && !ret) { /* @@ -3091,24 +2793,23 @@ out_only_mutex: * for detecting, at fsync time, if the inode isn't yet in the * log tree or it's there but not up to date. */ - struct timespec64 now = current_time(inode); + struct timespec64 now = inode_set_ctime_current(inode); inode_inc_iversion(inode); - inode->i_mtime = now; - inode->i_ctime = now; + inode_set_mtime_to_ts(inode, now); trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { ret = PTR_ERR(trans); } else { int ret2; - ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); + ret = btrfs_update_inode(trans, BTRFS_I(inode)); ret2 = btrfs_end_transaction(trans); if (!ret) ret = ret2; } } - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + btrfs_inode_unlock(BTRFS_I(inode), BTRFS_ILOCK_MMAP); return ret; } @@ -3156,20 +2857,30 @@ static int btrfs_fallocate_update_isize(struct inode *inode, { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; + u64 range_start; + u64 range_end; int ret; int ret2; if (mode & FALLOC_FL_KEEP_SIZE || end <= i_size_read(inode)) return 0; + range_start = round_down(i_size_read(inode), root->fs_info->sectorsize); + range_end = round_up(end, root->fs_info->sectorsize); + + ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), range_start, + range_end - range_start); + if (ret) + return ret; + trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) return PTR_ERR(trans); - inode->i_ctime = current_time(inode); + inode_set_ctime_current(inode); i_size_write(inode, end); btrfs_inode_safe_disk_i_size_write(BTRFS_I(inode), 0); - ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); + ret = btrfs_update_inode(trans, BTRFS_I(inode)); ret2 = btrfs_end_transaction(trans); return ret ? ret : ret2; @@ -3184,23 +2895,23 @@ enum { static int btrfs_zero_range_check_range_boundary(struct btrfs_inode *inode, u64 offset) { - const u64 sectorsize = btrfs_inode_sectorsize(inode); + const u64 sectorsize = inode->root->fs_info->sectorsize; struct extent_map *em; int ret; offset = round_down(offset, sectorsize); - em = btrfs_get_extent(inode, NULL, 0, offset, sectorsize); + em = btrfs_get_extent(inode, NULL, offset, sectorsize); if (IS_ERR(em)) return PTR_ERR(em); - if (em->block_start == EXTENT_MAP_HOLE) + if (em->disk_bytenr == EXTENT_MAP_HOLE) ret = RANGE_BOUNDARY_HOLE; - else if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) + else if (em->flags & EXTENT_FLAG_PREALLOC) ret = RANGE_BOUNDARY_PREALLOC_EXTENT; else ret = RANGE_BOUNDARY_WRITTEN_EXTENT; - free_extent_map(em); + btrfs_free_extent_map(em); return ret; } @@ -3214,15 +2925,15 @@ static int btrfs_zero_range(struct inode *inode, struct extent_changeset *data_reserved = NULL; int ret; u64 alloc_hint = 0; - const u64 sectorsize = btrfs_inode_sectorsize(BTRFS_I(inode)); + const u64 sectorsize = fs_info->sectorsize; + const u64 orig_start = offset; + const u64 orig_end = offset + len - 1; u64 alloc_start = round_down(offset, sectorsize); u64 alloc_end = round_up(offset + len, sectorsize); u64 bytes_to_reserve = 0; bool space_reserved = false; - inode_dio_wait(inode); - - em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, alloc_start, + em = btrfs_get_extent(BTRFS_I(inode), NULL, alloc_start, alloc_end - alloc_start); if (IS_ERR(em)) { ret = PTR_ERR(em); @@ -3237,8 +2948,7 @@ static int btrfs_zero_range(struct inode *inode, * extents and holes, we drop all the existing extents and allocate a * new prealloc extent, so that we get a larger contiguous disk extent. */ - if (em->start <= alloc_start && - test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { + if (em->start <= alloc_start && (em->flags & EXTENT_FLAG_PREALLOC)) { const u64 em_end = em->start + em->len; if (em_end >= offset + len) { @@ -3247,7 +2957,7 @@ static int btrfs_zero_range(struct inode *inode, * do nothing except updating the inode's i_size if * needed. */ - free_extent_map(em); + btrfs_free_extent_map(em); ret = btrfs_fallocate_update_isize(inode, offset + len, mode); goto out; @@ -3260,36 +2970,35 @@ static int btrfs_zero_range(struct inode *inode, ASSERT(IS_ALIGNED(alloc_start, sectorsize)); len = offset + len - alloc_start; offset = alloc_start; - alloc_hint = em->block_start + em->len; + alloc_hint = btrfs_extent_map_block_start(em) + em->len; } - free_extent_map(em); + btrfs_free_extent_map(em); if (BTRFS_BYTES_TO_BLKS(fs_info, offset) == BTRFS_BYTES_TO_BLKS(fs_info, offset + len - 1)) { - em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, alloc_start, - sectorsize); + em = btrfs_get_extent(BTRFS_I(inode), NULL, alloc_start, sectorsize); if (IS_ERR(em)) { ret = PTR_ERR(em); goto out; } - if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { - free_extent_map(em); + if (em->flags & EXTENT_FLAG_PREALLOC) { + btrfs_free_extent_map(em); ret = btrfs_fallocate_update_isize(inode, offset + len, mode); goto out; } - if (len < sectorsize && em->block_start != EXTENT_MAP_HOLE) { - free_extent_map(em); - ret = btrfs_truncate_block(BTRFS_I(inode), offset, len, - 0); + if (len < sectorsize && em->disk_bytenr != EXTENT_MAP_HOLE) { + btrfs_free_extent_map(em); + ret = btrfs_truncate_block(BTRFS_I(inode), offset + len - 1, + orig_start, orig_end); if (!ret) ret = btrfs_fallocate_update_isize(inode, offset + len, mode); return ret; } - free_extent_map(em); + btrfs_free_extent_map(em); alloc_start = round_down(offset, sectorsize); alloc_end = alloc_start + sectorsize; goto reserve_space; @@ -3313,7 +3022,8 @@ static int btrfs_zero_range(struct inode *inode, alloc_start = round_down(offset, sectorsize); ret = 0; } else if (ret == RANGE_BOUNDARY_WRITTEN_EXTENT) { - ret = btrfs_truncate_block(BTRFS_I(inode), offset, 0, 0); + ret = btrfs_truncate_block(BTRFS_I(inode), offset, + orig_start, orig_end); if (ret) goto out; } else { @@ -3330,8 +3040,8 @@ static int btrfs_zero_range(struct inode *inode, alloc_end = round_up(offset + len, sectorsize); ret = 0; } else if (ret == RANGE_BOUNDARY_WRITTEN_EXTENT) { - ret = btrfs_truncate_block(BTRFS_I(inode), offset + len, - 0, 1); + ret = btrfs_truncate_block(BTRFS_I(inode), offset + len - 1, + orig_start, orig_end); if (ret) goto out; } else { @@ -3351,23 +3061,21 @@ reserve_space: if (ret < 0) goto out; space_reserved = true; - ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend, - &cached_state); - if (ret) - goto out; + btrfs_punch_hole_lock_range(inode, lockstart, lockend, + &cached_state); ret = btrfs_qgroup_reserve_data(BTRFS_I(inode), &data_reserved, alloc_start, bytes_to_reserve); if (ret) { - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, - lockend, &cached_state); + btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, + lockend, &cached_state); goto out; } ret = btrfs_prealloc_file_range(inode, mode, alloc_start, alloc_end - alloc_start, - i_blocksize(inode), + fs_info->sectorsize, offset + len, &alloc_hint); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, - lockend, &cached_state); + btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend, + &cached_state); /* btrfs_prealloc_file_range releases reserved space on error */ if (ret) { space_reserved = false; @@ -3392,7 +3100,7 @@ static long btrfs_fallocate(struct file *file, int mode, struct extent_changeset *data_reserved = NULL; struct falloc_range *range; struct falloc_range *tmp; - struct list_head reserve_list; + LIST_HEAD(reserve_list); u64 cur_offset; u64 last_byte; u64 alloc_start; @@ -3400,12 +3108,18 @@ static long btrfs_fallocate(struct file *file, int mode, u64 alloc_hint = 0; u64 locked_end; u64 actual_end = 0; + u64 data_space_needed = 0; + u64 data_space_reserved = 0; + u64 qgroup_reserved = 0; struct extent_map *em; - int blocksize = btrfs_inode_sectorsize(BTRFS_I(inode)); + int blocksize = BTRFS_I(inode)->root->fs_info->sectorsize; int ret; + if (unlikely(btrfs_is_shutdown(inode_to_fs_info(inode)))) + return -EIO; + /* Do not allow fallocate in ZONED mode */ - if (btrfs_is_zoned(btrfs_sb(inode->i_sb))) + if (btrfs_is_zoned(inode_to_fs_info(inode))) return -EOPNOTSUPP; alloc_start = round_down(offset, blocksize); @@ -3418,21 +3132,9 @@ static long btrfs_fallocate(struct file *file, int mode, return -EOPNOTSUPP; if (mode & FALLOC_FL_PUNCH_HOLE) - return btrfs_punch_hole(inode, offset, len); - - /* - * Only trigger disk allocation, don't trigger qgroup reserve - * - * For qgroup space, it will be checked later. - */ - if (!(mode & FALLOC_FL_ZERO_RANGE)) { - ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode), - alloc_end - alloc_start); - if (ret < 0) - return ret; - } + return btrfs_punch_hole(file, offset, len); - btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP); + btrfs_inode_lock(BTRFS_I(inode), BTRFS_ILOCK_MMAP); if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size) { ret = inode_newsize_ok(inode, offset + len); @@ -3440,6 +3142,10 @@ static long btrfs_fallocate(struct file *file, int mode, goto out; } + ret = file_modified(file); + if (ret) + goto out; + /* * TODO: Move these two operations after we have checked * accurate reserved space, or fallocate can still fail but @@ -3458,117 +3164,109 @@ static long btrfs_fallocate(struct file *file, int mode, * need to zero out the end of the block if i_size lands in the * middle of a block. */ - ret = btrfs_truncate_block(BTRFS_I(inode), inode->i_size, 0, 0); + ret = btrfs_truncate_block(BTRFS_I(inode), inode->i_size, + inode->i_size, (u64)-1); if (ret) goto out; } /* - * wait for ordered IO before we have any locks. We'll loop again - * below with the locks held. + * We have locked the inode at the VFS level (in exclusive mode) and we + * have locked the i_mmap_lock lock (in exclusive mode). Now before + * locking the file range, flush all dealloc in the range and wait for + * all ordered extents in the range to complete. After this we can lock + * the file range and, due to the previous locking we did, we know there + * can't be more delalloc or ordered extents in the range. */ - ret = btrfs_wait_ordered_range(inode, alloc_start, + ret = btrfs_wait_ordered_range(BTRFS_I(inode), alloc_start, alloc_end - alloc_start); if (ret) goto out; if (mode & FALLOC_FL_ZERO_RANGE) { ret = btrfs_zero_range(inode, offset, len, mode); - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); + btrfs_inode_unlock(BTRFS_I(inode), BTRFS_ILOCK_MMAP); return ret; } locked_end = alloc_end - 1; - while (1) { - struct btrfs_ordered_extent *ordered; + btrfs_lock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, + &cached_state); - /* the extent lock is ordered inside the running - * transaction - */ - lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start, - locked_end, &cached_state); - ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode), - locked_end); - - if (ordered && - ordered->file_offset + ordered->num_bytes > alloc_start && - ordered->file_offset < alloc_end) { - btrfs_put_ordered_extent(ordered); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, - alloc_start, locked_end, - &cached_state); - /* - * we can't wait on the range with the transaction - * running or with the extent lock held - */ - ret = btrfs_wait_ordered_range(inode, alloc_start, - alloc_end - alloc_start); - if (ret) - goto out; - } else { - if (ordered) - btrfs_put_ordered_extent(ordered); - break; - } - } + btrfs_assert_inode_range_clean(BTRFS_I(inode), alloc_start, locked_end); /* First, check if we exceed the qgroup limit */ - INIT_LIST_HEAD(&reserve_list); while (cur_offset < alloc_end) { - em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, cur_offset, + em = btrfs_get_extent(BTRFS_I(inode), NULL, cur_offset, alloc_end - cur_offset); if (IS_ERR(em)) { ret = PTR_ERR(em); break; } - last_byte = min(extent_map_end(em), alloc_end); - actual_end = min_t(u64, extent_map_end(em), offset + len); + last_byte = min(btrfs_extent_map_end(em), alloc_end); + actual_end = min_t(u64, btrfs_extent_map_end(em), offset + len); last_byte = ALIGN(last_byte, blocksize); - if (em->block_start == EXTENT_MAP_HOLE || + if (em->disk_bytenr == EXTENT_MAP_HOLE || (cur_offset >= inode->i_size && - !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { - ret = add_falloc_range(&reserve_list, cur_offset, - last_byte - cur_offset); + !(em->flags & EXTENT_FLAG_PREALLOC))) { + const u64 range_len = last_byte - cur_offset; + + ret = add_falloc_range(&reserve_list, cur_offset, range_len); if (ret < 0) { - free_extent_map(em); + btrfs_free_extent_map(em); break; } ret = btrfs_qgroup_reserve_data(BTRFS_I(inode), - &data_reserved, cur_offset, - last_byte - cur_offset); + &data_reserved, cur_offset, range_len); if (ret < 0) { - cur_offset = last_byte; - free_extent_map(em); + btrfs_free_extent_map(em); break; } - } else { - /* - * Do not need to reserve unwritten extent for this - * range, free reserved data space first, otherwise - * it'll result in false ENOSPC error. - */ - btrfs_free_reserved_data_space(BTRFS_I(inode), - data_reserved, cur_offset, - last_byte - cur_offset); + qgroup_reserved += range_len; + data_space_needed += range_len; } - free_extent_map(em); + btrfs_free_extent_map(em); cur_offset = last_byte; } + if (!ret && data_space_needed > 0) { + /* + * We are safe to reserve space here as we can't have delalloc + * in the range, see above. + */ + ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode), + data_space_needed); + if (!ret) + data_space_reserved = data_space_needed; + } + /* * If ret is still 0, means we're OK to fallocate. * Or just cleanup the list and exit. */ list_for_each_entry_safe(range, tmp, &reserve_list, list) { - if (!ret) + if (!ret) { ret = btrfs_prealloc_file_range(inode, mode, range->start, - range->len, i_blocksize(inode), + range->len, blocksize, offset + len, &alloc_hint); - else + /* + * btrfs_prealloc_file_range() releases space even + * if it returns an error. + */ + data_space_reserved -= range->len; + qgroup_reserved -= range->len; + } else if (data_space_reserved > 0) { btrfs_free_reserved_data_space(BTRFS_I(inode), - data_reserved, range->start, - range->len); + data_reserved, range->start, + range->len); + data_space_reserved -= range->len; + qgroup_reserved -= range->len; + } else if (qgroup_reserved > 0) { + btrfs_qgroup_free_data(BTRFS_I(inode), data_reserved, + range->start, range->len, NULL); + qgroup_reserved -= range->len; + } list_del(&range->list); kfree(range); } @@ -3581,35 +3279,331 @@ static long btrfs_fallocate(struct file *file, int mode, */ ret = btrfs_fallocate_update_isize(inode, actual_end, mode); out_unlock: - unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, - &cached_state); + btrfs_unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, + &cached_state); out: - btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); - /* Let go of our reservation. */ - if (ret != 0 && !(mode & FALLOC_FL_ZERO_RANGE)) - btrfs_free_reserved_data_space(BTRFS_I(inode), data_reserved, - cur_offset, alloc_end - cur_offset); + btrfs_inode_unlock(BTRFS_I(inode), BTRFS_ILOCK_MMAP); extent_changeset_free(data_reserved); return ret; } -static loff_t find_desired_extent(struct btrfs_inode *inode, loff_t offset, - int whence) +/* + * Helper for btrfs_find_delalloc_in_range(). Find a subrange in a given range + * that has unflushed and/or flushing delalloc. There might be other adjacent + * subranges after the one it found, so btrfs_find_delalloc_in_range() keeps + * looping while it gets adjacent subranges, and merging them together. + */ +static bool find_delalloc_subrange(struct btrfs_inode *inode, u64 start, u64 end, + struct extent_state **cached_state, + bool *search_io_tree, + u64 *delalloc_start_ret, u64 *delalloc_end_ret) { + u64 len = end + 1 - start; + u64 delalloc_len = 0; + struct btrfs_ordered_extent *oe; + u64 oe_start; + u64 oe_end; + + /* + * Search the io tree first for EXTENT_DELALLOC. If we find any, it + * means we have delalloc (dirty pages) for which writeback has not + * started yet. + */ + if (*search_io_tree) { + spin_lock(&inode->lock); + if (inode->delalloc_bytes > 0) { + spin_unlock(&inode->lock); + *delalloc_start_ret = start; + delalloc_len = btrfs_count_range_bits(&inode->io_tree, + delalloc_start_ret, end, + len, EXTENT_DELALLOC, 1, + cached_state); + } else { + spin_unlock(&inode->lock); + } + } + + if (delalloc_len > 0) { + /* + * If delalloc was found then *delalloc_start_ret has a sector size + * aligned value (rounded down). + */ + *delalloc_end_ret = *delalloc_start_ret + delalloc_len - 1; + + if (*delalloc_start_ret == start) { + /* Delalloc for the whole range, nothing more to do. */ + if (*delalloc_end_ret == end) + return true; + /* Else trim our search range for ordered extents. */ + start = *delalloc_end_ret + 1; + len = end + 1 - start; + } + } else { + /* No delalloc, future calls don't need to search again. */ + *search_io_tree = false; + } + + /* + * Now also check if there's any ordered extent in the range. + * We do this because: + * + * 1) When delalloc is flushed, the file range is locked, we clear the + * EXTENT_DELALLOC bit from the io tree and create an extent map and + * an ordered extent for the write. So we might just have been called + * after delalloc is flushed and before the ordered extent completes + * and inserts the new file extent item in the subvolume's btree; + * + * 2) We may have an ordered extent created by flushing delalloc for a + * subrange that starts before the subrange we found marked with + * EXTENT_DELALLOC in the io tree. + * + * We could also use the extent map tree to find such delalloc that is + * being flushed, but using the ordered extents tree is more efficient + * because it's usually much smaller as ordered extents are removed from + * the tree once they complete. With the extent maps, we may have them + * in the extent map tree for a very long time, and they were either + * created by previous writes or loaded by read operations. + */ + oe = btrfs_lookup_first_ordered_range(inode, start, len); + if (!oe) + return (delalloc_len > 0); + + /* The ordered extent may span beyond our search range. */ + oe_start = max(oe->file_offset, start); + oe_end = min(oe->file_offset + oe->num_bytes - 1, end); + + btrfs_put_ordered_extent(oe); + + /* Don't have unflushed delalloc, return the ordered extent range. */ + if (delalloc_len == 0) { + *delalloc_start_ret = oe_start; + *delalloc_end_ret = oe_end; + return true; + } + + /* + * We have both unflushed delalloc (io_tree) and an ordered extent. + * If the ranges are adjacent returned a combined range, otherwise + * return the leftmost range. + */ + if (oe_start < *delalloc_start_ret) { + if (oe_end < *delalloc_start_ret) + *delalloc_end_ret = oe_end; + *delalloc_start_ret = oe_start; + } else if (*delalloc_end_ret + 1 == oe_start) { + *delalloc_end_ret = oe_end; + } + + return true; +} + +/* + * Check if there's delalloc in a given range. + * + * @inode: The inode. + * @start: The start offset of the range. It does not need to be + * sector size aligned. + * @end: The end offset (inclusive value) of the search range. + * It does not need to be sector size aligned. + * @cached_state: Extent state record used for speeding up delalloc + * searches in the inode's io_tree. Can be NULL. + * @delalloc_start_ret: Output argument, set to the start offset of the + * subrange found with delalloc (may not be sector size + * aligned). + * @delalloc_end_ret: Output argument, set to he end offset (inclusive value) + * of the subrange found with delalloc. + * + * Returns true if a subrange with delalloc is found within the given range, and + * if so it sets @delalloc_start_ret and @delalloc_end_ret with the start and + * end offsets of the subrange. + */ +bool btrfs_find_delalloc_in_range(struct btrfs_inode *inode, u64 start, u64 end, + struct extent_state **cached_state, + u64 *delalloc_start_ret, u64 *delalloc_end_ret) +{ + u64 cur_offset = round_down(start, inode->root->fs_info->sectorsize); + u64 prev_delalloc_end = 0; + bool search_io_tree = true; + bool ret = false; + + while (cur_offset <= end) { + u64 delalloc_start; + u64 delalloc_end; + bool delalloc; + + delalloc = find_delalloc_subrange(inode, cur_offset, end, + cached_state, &search_io_tree, + &delalloc_start, + &delalloc_end); + if (!delalloc) + break; + + if (prev_delalloc_end == 0) { + /* First subrange found. */ + *delalloc_start_ret = max(delalloc_start, start); + *delalloc_end_ret = delalloc_end; + ret = true; + } else if (delalloc_start == prev_delalloc_end + 1) { + /* Subrange adjacent to the previous one, merge them. */ + *delalloc_end_ret = delalloc_end; + } else { + /* Subrange not adjacent to the previous one, exit. */ + break; + } + + prev_delalloc_end = delalloc_end; + cur_offset = delalloc_end + 1; + cond_resched(); + } + + return ret; +} + +/* + * Check if there's a hole or delalloc range in a range representing a hole (or + * prealloc extent) found in the inode's subvolume btree. + * + * @inode: The inode. + * @whence: Seek mode (SEEK_DATA or SEEK_HOLE). + * @start: Start offset of the hole region. It does not need to be sector + * size aligned. + * @end: End offset (inclusive value) of the hole region. It does not + * need to be sector size aligned. + * @start_ret: Return parameter, used to set the start of the subrange in the + * hole that matches the search criteria (seek mode), if such + * subrange is found (return value of the function is true). + * The value returned here may not be sector size aligned. + * + * Returns true if a subrange matching the given seek mode is found, and if one + * is found, it updates @start_ret with the start of the subrange. + */ +static bool find_desired_extent_in_hole(struct btrfs_inode *inode, int whence, + struct extent_state **cached_state, + u64 start, u64 end, u64 *start_ret) +{ + u64 delalloc_start; + u64 delalloc_end; + bool delalloc; + + delalloc = btrfs_find_delalloc_in_range(inode, start, end, cached_state, + &delalloc_start, &delalloc_end); + if (delalloc && whence == SEEK_DATA) { + *start_ret = delalloc_start; + return true; + } + + if (delalloc && whence == SEEK_HOLE) { + /* + * We found delalloc but it starts after out start offset. So we + * have a hole between our start offset and the delalloc start. + */ + if (start < delalloc_start) { + *start_ret = start; + return true; + } + /* + * Delalloc range starts at our start offset. + * If the delalloc range's length is smaller than our range, + * then it means we have a hole that starts where the delalloc + * subrange ends. + */ + if (delalloc_end < end) { + *start_ret = delalloc_end + 1; + return true; + } + + /* There's delalloc for the whole range. */ + return false; + } + + if (!delalloc && whence == SEEK_HOLE) { + *start_ret = start; + return true; + } + + /* + * No delalloc in the range and we are seeking for data. The caller has + * to iterate to the next extent item in the subvolume btree. + */ + return false; +} + +static loff_t find_desired_extent(struct file *file, loff_t offset, int whence) +{ + struct btrfs_inode *inode = BTRFS_I(file->f_mapping->host); + struct btrfs_file_private *private; struct btrfs_fs_info *fs_info = inode->root->fs_info; - struct extent_map *em = NULL; struct extent_state *cached_state = NULL; - loff_t i_size = inode->vfs_inode.i_size; + struct extent_state **delalloc_cached_state; + const loff_t i_size = i_size_read(&inode->vfs_inode); + const u64 ino = btrfs_ino(inode); + struct btrfs_root *root = inode->root; + struct btrfs_path *path; + struct btrfs_key key; + u64 last_extent_end; u64 lockstart; u64 lockend; u64 start; - u64 len; - int ret = 0; + int ret; + bool found = false; if (i_size == 0 || offset >= i_size) return -ENXIO; /* + * Quick path. If the inode has no prealloc extents and its number of + * bytes used matches its i_size, then it can not have holes. + */ + if (whence == SEEK_HOLE && + !(inode->flags & BTRFS_INODE_PREALLOC) && + inode_get_bytes(&inode->vfs_inode) == i_size) + return i_size; + + spin_lock(&inode->lock); + private = file->private_data; + spin_unlock(&inode->lock); + + if (private && private->owner_task != current) { + /* + * Not allocated by us, don't use it as its cached state is used + * by the task that allocated it and we don't want neither to + * mess with it nor get incorrect results because it reflects an + * invalid state for the current task. + */ + private = NULL; + } else if (!private) { + private = kzalloc(sizeof(*private), GFP_KERNEL); + /* + * No worries if memory allocation failed. + * The private structure is used only for speeding up multiple + * lseek SEEK_HOLE/DATA calls to a file when there's delalloc, + * so everything will still be correct. + */ + if (private) { + bool free = false; + + private->owner_task = current; + + spin_lock(&inode->lock); + if (file->private_data) + free = true; + else + file->private_data = private; + spin_unlock(&inode->lock); + + if (free) { + kfree(private); + private = NULL; + } + } + } + + if (private) + delalloc_cached_state = &private->llseek_cached_state; + else + delalloc_cached_state = NULL; + + /* * offset can be negative, in this case we start finding DATA/HOLE from * the very start of the file. */ @@ -3620,45 +3614,174 @@ static loff_t find_desired_extent(struct btrfs_inode *inode, loff_t offset, if (lockend <= lockstart) lockend = lockstart + fs_info->sectorsize; lockend--; - len = lockend - lockstart + 1; - lock_extent_bits(&inode->io_tree, lockstart, lockend, &cached_state); + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + path->reada = READA_FORWARD; + + key.objectid = ino; + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = start; + + last_extent_end = lockstart; + + btrfs_lock_extent(&inode->io_tree, lockstart, lockend, &cached_state); + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + goto out; + } else if (ret > 0 && path->slots[0] > 0) { + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0] - 1); + if (key.objectid == ino && key.type == BTRFS_EXTENT_DATA_KEY) + path->slots[0]--; + } while (start < i_size) { - em = btrfs_get_extent_fiemap(inode, start, len); - if (IS_ERR(em)) { - ret = PTR_ERR(em); - em = NULL; - break; + struct extent_buffer *leaf = path->nodes[0]; + struct btrfs_file_extent_item *extent; + u64 extent_end; + u8 type; + + if (path->slots[0] >= btrfs_header_nritems(leaf)) { + ret = btrfs_next_leaf(root, path); + if (ret < 0) + goto out; + else if (ret > 0) + break; + + leaf = path->nodes[0]; } - if (whence == SEEK_HOLE && - (em->block_start == EXTENT_MAP_HOLE || - test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) - break; - else if (whence == SEEK_DATA && - (em->block_start != EXTENT_MAP_HOLE && - !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) + btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); + if (key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY) break; - start = em->start + em->len; - free_extent_map(em); - em = NULL; + extent_end = btrfs_file_extent_end(path); + + /* + * In the first iteration we may have a slot that points to an + * extent that ends before our start offset, so skip it. + */ + if (extent_end <= start) { + path->slots[0]++; + continue; + } + + /* We have an implicit hole, NO_HOLES feature is likely set. */ + if (last_extent_end < key.offset) { + u64 search_start = last_extent_end; + u64 found_start; + + /* + * First iteration, @start matches @offset and it's + * within the hole. + */ + if (start == offset) + search_start = offset; + + found = find_desired_extent_in_hole(inode, whence, + delalloc_cached_state, + search_start, + key.offset - 1, + &found_start); + if (found) { + start = found_start; + break; + } + /* + * Didn't find data or a hole (due to delalloc) in the + * implicit hole range, so need to analyze the extent. + */ + } + + extent = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_file_extent_item); + type = btrfs_file_extent_type(leaf, extent); + + /* + * Can't access the extent's disk_bytenr field if this is an + * inline extent, since at that offset, it's where the extent + * data starts. + */ + if (type == BTRFS_FILE_EXTENT_PREALLOC || + (type == BTRFS_FILE_EXTENT_REG && + btrfs_file_extent_disk_bytenr(leaf, extent) == 0)) { + /* + * Explicit hole or prealloc extent, search for delalloc. + * A prealloc extent is treated like a hole. + */ + u64 search_start = key.offset; + u64 found_start; + + /* + * First iteration, @start matches @offset and it's + * within the hole. + */ + if (start == offset) + search_start = offset; + + found = find_desired_extent_in_hole(inode, whence, + delalloc_cached_state, + search_start, + extent_end - 1, + &found_start); + if (found) { + start = found_start; + break; + } + /* + * Didn't find data or a hole (due to delalloc) in the + * implicit hole range, so need to analyze the next + * extent item. + */ + } else { + /* + * Found a regular or inline extent. + * If we are seeking for data, adjust the start offset + * and stop, we're done. + */ + if (whence == SEEK_DATA) { + start = max_t(u64, key.offset, offset); + found = true; + break; + } + /* + * Else, we are seeking for a hole, check the next file + * extent item. + */ + } + + start = extent_end; + last_extent_end = extent_end; + path->slots[0]++; + if (fatal_signal_pending(current)) { + ret = -EINTR; + goto out; + } cond_resched(); } - free_extent_map(em); - unlock_extent_cached(&inode->io_tree, lockstart, lockend, - &cached_state); - if (ret) { - offset = ret; - } else { - if (whence == SEEK_DATA && start >= i_size) - offset = -ENXIO; - else - offset = min_t(loff_t, start, i_size); + + /* We have an implicit hole from the last extent found up to i_size. */ + if (!found && start < i_size) { + found = find_desired_extent_in_hole(inode, whence, + delalloc_cached_state, start, + i_size - 1, &start); + if (!found) + start = i_size; } - return offset; +out: + btrfs_unlock_extent(&inode->io_tree, lockstart, lockend, &cached_state); + btrfs_free_path(path); + + if (ret < 0) + return ret; + + if (whence == SEEK_DATA && start >= i_size) + return -ENXIO; + + return min_t(loff_t, start, i_size); } static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence) @@ -3670,9 +3793,9 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence) return generic_file_llseek(file, offset, whence); case SEEK_DATA: case SEEK_HOLE: - btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED); - offset = find_desired_extent(BTRFS_I(inode), offset, whence); - btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); + btrfs_inode_lock(BTRFS_I(inode), BTRFS_ILOCK_SHARED); + offset = find_desired_extent(file, offset, whence); + btrfs_inode_unlock(BTRFS_I(inode), BTRFS_ILOCK_SHARED); break; } @@ -3686,7 +3809,10 @@ static int btrfs_file_open(struct inode *inode, struct file *filp) { int ret; - filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC; + if (unlikely(btrfs_is_shutdown(inode_to_fs_info(inode)))) + return -EIO; + + filp->f_mode |= FMODE_NOWAIT | FMODE_CAN_ODIRECT; ret = fsverity_file_open(inode, filp); if (ret) @@ -3694,97 +3820,13 @@ static int btrfs_file_open(struct inode *inode, struct file *filp) return generic_file_open(inode, filp); } -static int check_direct_read(struct btrfs_fs_info *fs_info, - const struct iov_iter *iter, loff_t offset) -{ - int ret; - int i, seg; - - ret = check_direct_IO(fs_info, iter, offset); - if (ret < 0) - return ret; - - if (!iter_is_iovec(iter)) - return 0; - - for (seg = 0; seg < iter->nr_segs; seg++) - for (i = seg + 1; i < iter->nr_segs; i++) - if (iter->iov[seg].iov_base == iter->iov[i].iov_base) - return -EINVAL; - return 0; -} - -static ssize_t btrfs_direct_read(struct kiocb *iocb, struct iov_iter *to) -{ - struct inode *inode = file_inode(iocb->ki_filp); - size_t prev_left = 0; - ssize_t read = 0; - ssize_t ret; - - if (fsverity_active(inode)) - return 0; - - if (check_direct_read(btrfs_sb(inode->i_sb), to, iocb->ki_pos)) - return 0; - - btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED); -again: - /* - * This is similar to what we do for direct IO writes, see the comment - * at btrfs_direct_write(), but we also disable page faults in addition - * to disabling them only at the iov_iter level. This is because when - * reading from a hole or prealloc extent, iomap calls iov_iter_zero(), - * which can still trigger page fault ins despite having set ->nofault - * to true of our 'to' iov_iter. - * - * The difference to direct IO writes is that we deadlock when trying - * to lock the extent range in the inode's tree during he page reads - * triggered by the fault in (while for writes it is due to waiting for - * our own ordered extent). This is because for direct IO reads, - * btrfs_dio_iomap_begin() returns with the extent range locked, which - * is only unlocked in the endio callback (end_bio_extent_readpage()). - */ - pagefault_disable(); - to->nofault = true; - ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops, - IOMAP_DIO_PARTIAL, read); - to->nofault = false; - pagefault_enable(); - - /* No increment (+=) because iomap returns a cumulative value. */ - if (ret > 0) - read = ret; - - if (iov_iter_count(to) > 0 && (ret == -EFAULT || ret > 0)) { - const size_t left = iov_iter_count(to); - - if (left == prev_left) { - /* - * We didn't make any progress since the last attempt, - * fallback to a buffered read for the remainder of the - * range. This is just to avoid any possibility of looping - * for too long. - */ - ret = read; - } else { - /* - * We made some progress since the last retry or this is - * the first time we are retrying. Fault in as many pages - * as possible and retry. - */ - fault_in_iov_iter_writeable(to, left); - prev_left = left; - goto again; - } - } - btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); - return ret < 0 ? ret : read; -} - static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) { ssize_t ret = 0; + if (unlikely(btrfs_is_shutdown(inode_to_fs_info(file_inode(iocb->ki_filp))))) + return -EIO; + if (iocb->ki_flags & IOCB_DIRECT) { ret = btrfs_direct_read(iocb, to); if (ret < 0 || !iov_iter_count(to) || @@ -3795,15 +3837,26 @@ static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) return filemap_read(iocb, to, ret); } +static ssize_t btrfs_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + if (unlikely(btrfs_is_shutdown(inode_to_fs_info(file_inode(in))))) + return -EIO; + + return filemap_splice_read(in, ppos, pipe, len, flags); +} + const struct file_operations btrfs_file_operations = { .llseek = btrfs_file_llseek, .read_iter = btrfs_file_read_iter, - .splice_read = generic_file_splice_read, + .splice_read = btrfs_file_splice_read, .write_iter = btrfs_file_write_iter, .splice_write = iter_file_splice_write, - .mmap = btrfs_file_mmap, + .mmap_prepare = btrfs_file_mmap_prepare, .open = btrfs_file_open, .release = btrfs_release_file, + .get_unmapped_area = thp_get_unmapped_area, .fsync = btrfs_sync_file, .fallocate = btrfs_fallocate, .unlocked_ioctl = btrfs_ioctl, @@ -3811,27 +3864,13 @@ const struct file_operations btrfs_file_operations = { .compat_ioctl = btrfs_compat_ioctl, #endif .remap_file_range = btrfs_remap_file_range, + .uring_cmd = btrfs_uring_cmd, + .fop_flags = FOP_BUFFER_RASYNC | FOP_BUFFER_WASYNC, }; -void __cold btrfs_auto_defrag_exit(void) -{ - kmem_cache_destroy(btrfs_inode_defrag_cachep); -} - -int __init btrfs_auto_defrag_init(void) -{ - btrfs_inode_defrag_cachep = kmem_cache_create("btrfs_inode_defrag", - sizeof(struct inode_defrag), 0, - SLAB_MEM_SPREAD, - NULL); - if (!btrfs_inode_defrag_cachep) - return -ENOMEM; - - return 0; -} - -int btrfs_fdatawrite_range(struct inode *inode, loff_t start, loff_t end) +int btrfs_fdatawrite_range(struct btrfs_inode *inode, loff_t start, loff_t end) { + struct address_space *mapping = inode->vfs_inode.i_mapping; int ret; /* @@ -3848,10 +3887,9 @@ int btrfs_fdatawrite_range(struct inode *inode, loff_t start, loff_t end) * know better and pull this out at some point in the future, it is * right and you are wrong. */ - ret = filemap_fdatawrite_range(inode->i_mapping, start, end); - if (!ret && test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, - &BTRFS_I(inode)->runtime_flags)) - ret = filemap_fdatawrite_range(inode->i_mapping, start, end); + ret = filemap_fdatawrite_range(mapping, start, end); + if (!ret && test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, &inode->runtime_flags)) + ret = filemap_fdatawrite_range(mapping, start, end); return ret; } |