/* * Copyright (C) 2007 Oracle. All rights reserved. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public * License v2 as published by the Free Software Foundation. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * General Public License for more details. * * You should have received a copy of the GNU General Public * License along with this program; if not, write to the * Free Software Foundation, Inc., 59 Temple Place - Suite 330, * Boston, MA 021110-1307, USA. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "ctree.h" #include "disk-io.h" #include "transaction.h" #include "btrfs_inode.h" #include "ioctl.h" #include "print-tree.h" #include "tree-log.h" #include "locking.h" #include "compat.h" /* * 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; }; /* 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 inode *inode, struct inode_defrag *defrag) { struct btrfs_root *root = BTRFS_I(inode)->root; struct inode_defrag *entry; struct rb_node **p; struct rb_node *parent = NULL; p = &root->fs_info->defrag_inodes.rb_node; while (*p) { parent = *p; entry = rb_entry(parent, struct inode_defrag, rb_node); if (defrag->ino < entry->ino) p = &parent->rb_left; else if (defrag->ino > entry->ino) 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; goto exists; } } BTRFS_I(inode)->in_defrag = 1; rb_link_node(&defrag->rb_node, parent, p); rb_insert_color(&defrag->rb_node, &root->fs_info->defrag_inodes); return 0; exists: kfree(defrag); return 0; } /* * insert a defrag record for this inode if auto defrag is * enabled */ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans, struct inode *inode) { struct btrfs_root *root = BTRFS_I(inode)->root; struct inode_defrag *defrag; int ret = 0; u64 transid; if (!btrfs_test_opt(root, AUTO_DEFRAG)) return 0; if (root->fs_info->closing) return 0; if (BTRFS_I(inode)->in_defrag) return 0; if (trans) transid = trans->transid; else transid = BTRFS_I(inode)->root->last_trans; defrag = kzalloc(sizeof(*defrag), GFP_NOFS); if (!defrag) return -ENOMEM; defrag->ino = btrfs_ino(inode); defrag->transid = transid; defrag->root = root->root_key.objectid; spin_lock(&root->fs_info->defrag_inodes_lock); if (!BTRFS_I(inode)->in_defrag) ret = __btrfs_add_inode_defrag(inode, defrag); spin_unlock(&root->fs_info->defrag_inodes_lock); return ret; } /* * must be called with the defrag_inodes lock held */ struct inode_defrag *btrfs_find_defrag_inode(struct btrfs_fs_info *info, u64 ino, struct rb_node **next) { struct inode_defrag *entry = NULL; struct rb_node *p; struct rb_node *parent = NULL; p = info->defrag_inodes.rb_node; while (p) { parent = p; entry = rb_entry(parent, struct inode_defrag, rb_node); if (ino < entry->ino) p = parent->rb_left; else if (ino > entry->ino) p = parent->rb_right; else return entry; } if (next) { while (parent && ino > entry->ino) { parent = rb_next(parent); entry = rb_entry(parent, struct inode_defrag, rb_node); } *next = parent; } return NULL; } /* * 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; struct btrfs_root *inode_root; struct inode *inode; struct rb_node *n; struct btrfs_key key; struct btrfs_ioctl_defrag_range_args range; u64 first_ino = 0; int num_defrag; int defrag_batch = 1024; memset(&range, 0, sizeof(range)); range.len = (u64)-1; atomic_inc(&fs_info->defrag_running); spin_lock(&fs_info->defrag_inodes_lock); while(1) { n = NULL; /* find an inode to defrag */ defrag = btrfs_find_defrag_inode(fs_info, first_ino, &n); if (!defrag) { if (n) defrag = rb_entry(n, struct inode_defrag, rb_node); else if (first_ino) { first_ino = 0; continue; } else { break; } } /* remove it from the rbtree */ first_ino = defrag->ino + 1; rb_erase(&defrag->rb_node, &fs_info->defrag_inodes); if (fs_info->closing) goto next_free; spin_unlock(&fs_info->defrag_inodes_lock); /* get the inode */ key.objectid = defrag->root; btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY); key.offset = (u64)-1; inode_root = btrfs_read_fs_root_no_name(fs_info, &key); if (IS_ERR(inode_root)) goto next; key.objectid = defrag->ino; btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY); key.offset = 0; inode = btrfs_iget(fs_info->sb, &key, inode_root, NULL); if (IS_ERR(inode)) goto next; /* do a chunk of defrag */ BTRFS_I(inode)->in_defrag = 0; range.start = defrag->last_offset; num_defrag = btrfs_defrag_file(inode, NULL, &range, defrag->transid, defrag_batch); /* * if we filled the whole defrag batch, there * must be more work to do. Queue this defrag * again */ if (num_defrag == defrag_batch) { defrag->last_offset = range.start; __btrfs_add_inode_defrag(inode, defrag); /* * we don't want to kfree defrag, we added it back to * the rbtree */ defrag = NULL; } 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_add_inode_defrag(inode, defrag); defrag = NULL; } iput(inode); next: spin_lock(&fs_info->defrag_inodes_lock); next_free: kfree(defrag); } spin_unlock(&fs_info->defrag_inodes_lock); 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, int num_pages, 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 = pos & (PAGE_CACHE_SIZE - 1); while (write_bytes > 0) { size_t count = min_t(size_t, PAGE_CACHE_SIZE - offset, write_bytes); struct page *page = prepared_pages[pg]; /* * Copy data from userspace to the current page * * Disable pagefault to avoid recursive lock since * the pages are already locked */ pagefault_disable(); copied = iov_iter_copy_from_user_atomic(page, i, offset, count); pagefault_enable(); /* 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 (!PageUptodate(page) && copied < count) copied = 0; iov_iter_advance(i, copied); write_bytes -= copied; total_copied += copied; /* Return to btrfs_file_aio_write to fault page */ if (unlikely(copied == 0)) break; if (unlikely(copied < PAGE_CACHE_SIZE - offset)) { offset += copied; } else { pg++; offset = 0; } } return total_copied; } /* * unlocks pages after btrfs_file_write is done with them */ void btrfs_drop_pages(struct page **pages, size_t num_pages) { size_t i; 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 */ ClearPageChecked(pages[i]); unlock_page(pages[i]); mark_page_accessed(pages[i]); page_cache_release(pages[i]); } } /* * after copy_from_user, pages need to be dirtied and we need to make * sure holes are created between the current EOF and the start of * any next extents (if required). * * this also makes the decision about creating an inline extent vs * doing real data extents, marking pages dirty and delalloc as required. */ int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode, struct page **pages, size_t num_pages, loff_t pos, size_t write_bytes, struct extent_state **cached) { int err = 0; int i; u64 num_bytes; u64 start_pos; u64 end_of_last_block; u64 end_pos = pos + write_bytes; loff_t isize = i_size_read(inode); start_pos = pos & ~((u64)root->sectorsize - 1); num_bytes = (write_bytes + pos - start_pos + root->sectorsize - 1) & ~((u64)root->sectorsize - 1); end_of_last_block = start_pos + num_bytes - 1; err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block, cached); if (err) return err; for (i = 0; i < num_pages; i++) { struct page *p = pages[i]; SetPageUptodate(p); ClearPageChecked(p); set_page_dirty(p); } /* * we've only changed i_size in ram, and we haven't updated * the disk i_size. There is no need to log the inode * at this time. */ if (end_pos > isize) i_size_write(inode, end_pos); return 0; } /* * this drops all the extents in the cache that intersect the range * [start, end]. Existing extents are split as required. */ int btrfs_drop_extent_cache(struct 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 = &BTRFS_I(inode)->extent_tree; u64 len = end - start + 1; int ret; int testend = 1; unsigned long flags; int compressed = 0; WARN_ON(end < start); if (end == (u64)-1) { len = (u64)-1; testend = 0; } while (1) { if (!split) split = alloc_extent_map(); if (!split2) split2 = alloc_extent_map(); BUG_ON(!split || !split2); write_lock(&em_tree->lock); em = lookup_extent_mapping(em_tree, start, len); if (!em) { write_unlock(&em_tree->lock); break; } flags = em->flags; 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); remove_extent_mapping(em_tree, em); if (em->block_start < EXTENT_MAP_LAST_BYTE && em->start < start) { split->start = em->start; split->len = start - em->start; 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->bdev = em->bdev; split->flags = flags; split->compress_type = em->compress_type; ret = add_extent_mapping(em_tree, split); BUG_ON(ret); free_extent_map(split); split = split2; split2 = NULL; } if (em->block_start < EXTENT_MAP_LAST_BYTE && 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->bdev = em->bdev; split->flags = flags; split->compress_type = em->compress_type; 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 = split->start; } ret = add_extent_mapping(em_tree, split); BUG_ON(ret); free_extent_map(split); split = NULL; } 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); return 0; } /* * 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. * * If an extent intersects the range but is not entirely inside the range * it is either truncated or split. Anything entirely inside the range * is deleted from the tree. */ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode, u64 start, u64 end, u64 *hint_byte, int drop_cache) { struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_buffer *leaf; struct btrfs_file_extent_item *fi; struct btrfs_path *path; struct btrfs_key key; struct btrfs_key new_key; u64 ino = btrfs_ino(inode); u64 search_start = start; u64 disk_bytenr = 0; u64 num_bytes = 0; u64 extent_offset = 0; u64 extent_end = 0; int del_nr = 0; int del_slot = 0; int extent_type; int recow; int ret; if (drop_cache) btrfs_drop_extent_cache(inode, start, end - 1, 0); path = btrfs_alloc_path(); if (!path) return -ENOMEM; while (1) { recow = 0; ret = btrfs_lookup_file_extent(trans, root, path, ino, search_start, -1); if (ret < 0) break; if (ret > 0 && path->slots[0] > 0 && search_start == start) { leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, path->slots[0] - 1); if (key.objectid == ino && key.type == BTRFS_EXTENT_DATA_KEY) path->slots[0]--; } ret = 0; next_slot: leaf = path->nodes[0]; if (path->slots[0] >= btrfs_header_nritems(leaf)) { BUG_ON(del_nr > 0); ret = btrfs_next_leaf(root, path); if (ret < 0) break; if (ret > 0) { ret = 0; break; } leaf = path->nodes[0]; recow = 1; } btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); if (key.objectid > ino || key.type > BTRFS_EXTENT_DATA_KEY || key.offset >= end) break; fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); extent_type = btrfs_file_extent_type(leaf, fi); if (extent_type == BTRFS_FILE_EXTENT_REG || extent_type == BTRFS_FILE_EXTENT_PREALLOC) { disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi); num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi); extent_offset = btrfs_file_extent_offset(leaf, fi); extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { extent_end = key.offset + btrfs_file_extent_inline_len(leaf, fi); } else { WARN_ON(1); extent_end = search_start; } if (extent_end <= search_start) { path->slots[0]++; goto next_slot; } search_start = max(key.offset, start); if (recow) { btrfs_release_path(path); continue; } /* * | - range to drop - | * | -------- extent -------- | */ if (start > key.offset && end < extent_end) { BUG_ON(del_nr > 0); BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE); memcpy(&new_key, &key, sizeof(new_key)); new_key.offset = start; ret = btrfs_duplicate_item(trans, root, path, &new_key); if (ret == -EAGAIN) { btrfs_release_path(path); continue; } if (ret < 0) break; leaf = path->nodes[0]; fi = btrfs_item_ptr(leaf, path->slots[0] - 1, struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, start - key.offset); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); extent_offset += start - key.offset; btrfs_set_file_extent_offset(leaf, fi, extent_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - start); btrfs_mark_buffer_dirty(leaf); if (disk_bytenr > 0) { ret = btrfs_inc_extent_ref(trans, root, disk_bytenr, num_bytes, 0, root->root_key.objectid, new_key.objectid, start - extent_offset); BUG_ON(ret); *hint_byte = disk_bytenr; } key.offset = start; } /* * | ---- range to drop ----- | * | -------- extent -------- | */ if (start <= key.offset && end < extent_end) { BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE); memcpy(&new_key, &key, sizeof(new_key)); new_key.offset = end; btrfs_set_item_key_safe(trans, root, path, &new_key); extent_offset += end - key.offset; btrfs_set_file_extent_offset(leaf, fi, extent_offset); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - end); btrfs_mark_buffer_dirty(leaf); if (disk_bytenr > 0) { inode_sub_bytes(inode, end - key.offset); *hint_byte = disk_bytenr; } break; } search_start = extent_end; /* * | ---- range to drop ----- | * | -------- extent -------- | */ if (start > key.offset && end >= extent_end) { BUG_ON(del_nr > 0); BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE); btrfs_set_file_extent_num_bytes(leaf, fi, start - key.offset); btrfs_mark_buffer_dirty(leaf); if (disk_bytenr > 0) { inode_sub_bytes(inode, extent_end - start); *hint_byte = disk_bytenr; } if (end == extent_end) break; path->slots[0]++; goto next_slot; } /* * | ---- range to drop ----- | * | ------ extent ------ | */ if (start <= key.offset && end >= extent_end) { if (del_nr == 0) { del_slot = path->slots[0]; del_nr = 1; } else { BUG_ON(del_slot + del_nr != path->slots[0]); del_nr++; } if (extent_type == BTRFS_FILE_EXTENT_INLINE) { inode_sub_bytes(inode, extent_end - key.offset); extent_end = ALIGN(extent_end, root->sectorsize); } else if (disk_bytenr > 0) { ret = btrfs_free_extent(trans, root, disk_bytenr, num_bytes, 0, root->root_key.objectid, key.objectid, key.offset - extent_offset); BUG_ON(ret); inode_sub_bytes(inode, extent_end - key.offset); *hint_byte = disk_bytenr; } if (end == extent_end) break; if (path->slots[0] + 1 < btrfs_header_nritems(leaf)) { path->slots[0]++; goto next_slot; } ret = btrfs_del_items(trans, root, path, del_slot, del_nr); BUG_ON(ret); del_nr = 0; del_slot = 0; btrfs_release_path(path); continue; } BUG_ON(1); } if (del_nr > 0) { ret = btrfs_del_items(trans, root, path, del_slot, del_nr); BUG_ON(ret); } btrfs_free_path(path); return ret; } static int 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; btrfs_item_key_to_cpu(leaf, &key, slot); if (key.objectid != objectid || key.type != BTRFS_EXTENT_DATA_KEY) return 0; fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG || btrfs_file_extent_disk_bytenr(leaf, fi) != bytenr || btrfs_file_extent_offset(leaf, fi) != key.offset - orig_offset || btrfs_file_extent_compression(leaf, fi) || btrfs_file_extent_encryption(leaf, fi) || btrfs_file_extent_other_encoding(leaf, fi)) return 0; extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); if ((*start && *start != key.offset) || (*end && *end != extent_end)) return 0; *start = key.offset; *end = extent_end; return 1; } /* * Mark extent in the range start - end as written. * * This changes extent type from 'pre-allocated' to 'regular'. If only * part of extent is marked as written, the extent will be split into * two or three. */ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, struct inode *inode, u64 start, u64 end) { struct btrfs_root *root = BTRFS_I(inode)->root; struct extent_buffer *leaf; struct btrfs_path *path; struct btrfs_file_extent_item *fi; struct btrfs_key key; struct btrfs_key new_key; u64 bytenr; u64 num_bytes; u64 extent_end; u64 orig_offset; u64 other_start; u64 other_end; u64 split; int del_nr = 0; int del_slot = 0; int recow; int ret; u64 ino = btrfs_ino(inode); btrfs_drop_extent_cache(inode, start, end - 1, 0); path = btrfs_alloc_path(); BUG_ON(!path); again: recow = 0; split = start; key.objectid = ino; key.type = BTRFS_EXTENT_DATA_KEY; key.offset = split; ret = btrfs_search_slot(trans, root, &key, path, -1, 1); if (ret < 0) goto out; if (ret > 0 && path->slots[0] > 0) path->slots[0]--; leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); BUG_ON(key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); BUG_ON(btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_PREALLOC); extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); BUG_ON(key.offset > start || extent_end < end); bytenr = btrfs_file_extent_disk_bytenr(leaf, fi); num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi); orig_offset = key.offset - btrfs_file_extent_offset(leaf, fi); memcpy(&new_key, &key, sizeof(new_key)); if (start == key.offset && end < extent_end) { other_start = 0; other_end = start; if (extent_mergeable(leaf, path->slots[0] - 1, ino, bytenr, orig_offset, &other_start, &other_end)) { new_key.offset = end; btrfs_set_item_key_safe(trans, root, path, &new_key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - end); btrfs_set_file_extent_offset(leaf, fi, end - orig_offset); fi = btrfs_item_ptr(leaf, path->slots[0] - 1, struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, end - other_start); btrfs_mark_buffer_dirty(leaf); goto out; } } if (start > key.offset && end == extent_end) { other_start = end; other_end = 0; if (extent_mergeable(leaf, path->slots[0] + 1, ino, bytenr, orig_offset, &other_start, &other_end)) { fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, start - key.offset); path->slots[0]++; new_key.offset = start; btrfs_set_item_key_safe(trans, root, path, &new_key); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, other_end - start); btrfs_set_file_extent_offset(leaf, fi, start - orig_offset); btrfs_mark_buffer_dirty(leaf); goto out; } } while (start > key.offset || end < extent_end) { if (key.offset == start) split = end; new_key.offset = split; ret = btrfs_duplicate_item(trans, root, path, &new_key); if (ret == -EAGAIN) { btrfs_release_path(path); goto again; } BUG_ON(ret < 0); leaf = path->nodes[0]; fi = btrfs_item_ptr(leaf, path->slots[0] - 1, struct btrfs_file_extent_item); btrfs_set_file_extent_num_bytes(leaf, fi, split - key.offset); fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); 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); ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0, root->root_key.objectid, ino, orig_offset); BUG_ON(ret); if (split == start) { key.offset = start; } else { BUG_ON(start != key.offset); path->slots[0]--; extent_end = end; } recow = 1; } other_start = end; other_end = 0; if (extent_mergeable(leaf, path->slots[0] + 1, ino, bytenr, orig_offset, &other_start, &other_end)) { if (recow) { btrfs_release_path(path); goto again; } extent_end = other_end; del_slot = path->slots[0] + 1; del_nr++; ret = btrfs_free_extent(trans, root, bytenr, num_bytes, 0, root->root_key.objectid, ino, orig_offset); BUG_ON(ret); } other_start = 0; other_end = start; if (extent_mergeable(leaf, path->slots[0] - 1, ino, bytenr, orig_offset, &other_start, &other_end)) { if (recow) { btrfs_release_path(path); goto again; } key.offset = other_start; del_slot = path->slots[0]; del_nr++; ret = btrfs_free_extent(trans, root, bytenr, num_bytes, 0, root->root_key.objectid, ino, orig_offset); BUG_ON(ret); } if (del_nr == 0) { fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG); btrfs_mark_buffer_dirty(leaf); } else { fi = btrfs_item_ptr(leaf, del_slot - 1, struct btrfs_file_extent_item); btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG); 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); BUG_ON(ret); } out: btrfs_free_path(path); return 0; } /* * on error we return an unlocked page and the error value * on success we return a locked page and 0 */ static int prepare_uptodate_page(struct page *page, u64 pos) { int ret = 0; if ((pos & (PAGE_CACHE_SIZE - 1)) && !PageUptodate(page)) { ret = btrfs_readpage(NULL, page); if (ret) return ret; lock_page(page); if (!PageUptodate(page)) { unlock_page(page); return -EIO; } } return 0; } /* * this gets pages into the page cache and locks them down, it also properly * waits for data=ordered extents to finish before allowing the pages to be * modified. */ static noinline int prepare_pages(struct btrfs_root *root, struct file *file, struct page **pages, size_t num_pages, loff_t pos, unsigned long first_index, unsigned long last_index, size_t write_bytes) { struct extent_state *cached_state = NULL; int i; unsigned long index = pos >> PAGE_CACHE_SHIFT; struct inode *inode = fdentry(file)->d_inode; int err = 0; int faili = 0; u64 start_pos; u64 last_pos; start_pos = pos & ~((u64)root->sectorsize - 1); last_pos = ((u64)index + num_pages) << PAGE_CACHE_SHIFT; if (start_pos > inode->i_size) { err = btrfs_cont_expand(inode, i_size_read(inode), start_pos); if (err) return err; } again: for (i = 0; i < num_pages; i++) { pages[i] = grab_cache_page(inode->i_mapping, index + i); if (!pages[i]) { faili = i - 1; err = -ENOMEM; goto fail; } if (i == 0) err = prepare_uptodate_page(pages[i], pos); if (i == num_pages - 1) err = prepare_uptodate_page(pages[i], pos + write_bytes); if (err) { page_cache_release(pages[i]); faili = i - 1; goto fail; } wait_on_page_writeback(pages[i]); } err = 0; if (start_pos < inode->i_size) { struct btrfs_ordered_extent *ordered; lock_extent_bits(&BTRFS_I(inode)->io_tree, start_pos, last_pos - 1, 0, &cached_state, GFP_NOFS); ordered = btrfs_lookup_first_ordered_extent(inode, last_pos - 1); if (ordered && ordered->file_offset + ordered->len > start_pos && ordered->file_offset < last_pos) { btrfs_put_ordered_extent(ordered); unlock_extent_cached(&BTRFS_I(inode)->io_tree, start_pos, last_pos - 1, &cached_state, GFP_NOFS); for (i = 0; i < num_pages; i++) { unlock_page(pages[i]); page_cache_release(pages[i]); } btrfs_wait_ordered_range(inode, start_pos, last_pos - start_pos); goto again; } if (ordered) btrfs_put_ordered_extent(ordered); clear_extent_bit(&BTRFS_I(inode)->io_tree, start_pos, last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING, 0, 0, &cached_state, GFP_NOFS); unlock_extent_cached(&BTRFS_I(inode)->io_tree, start_pos, last_pos - 1, &cached_state, GFP_NOFS); } for (i = 0; i < num_pages; i++) { clear_page_dirty_for_io(pages[i]); set_page_extent_mapped(pages[i]); WARN_ON(!PageLocked(pages[i])); } return 0; fail: while (faili >= 0) { unlock_page(pages[faili]); page_cache_release(pages[faili]); faili--; } return err; } static noinline ssize_t __btrfs_buffered_write(struct file *file, struct iov_iter *i, loff_t pos) { struct inode *inode = fdentry(file)->d_inode; struct btrfs_root *root = BTRFS_I(inode)->root; struct page **pages = NULL; unsigned long first_index; unsigned long last_index; size_t num_written = 0; int nrptrs; int ret = 0; nrptrs = min((iov_iter_count(i) + PAGE_CACHE_SIZE - 1) / PAGE_CACHE_SIZE, PAGE_CACHE_SIZE / (sizeof(struct page *))); pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL); if (!pages) return -ENOMEM; first_index = pos >> PAGE_CACHE_SHIFT; last_index = (pos + iov_iter_count(i)) >> PAGE_CACHE_SHIFT; while (iov_iter_count(i) > 0) { size_t offset = pos & (PAGE_CACHE_SIZE - 1); size_t write_bytes = min(iov_iter_count(i), nrptrs * (size_t)PAGE_CACHE_SIZE - offset); size_t num_pages = (write_bytes + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; size_t dirty_pages; size_t copied; WARN_ON(num_pages > nrptrs); /* * Fault pages before locking them in prepare_pages * to avoid recursive lock */ if (unlikely(iov_iter_fault_in_readable(i, write_bytes))) { ret = -EFAULT; break; } ret = btrfs_delalloc_reserve_space(inode, num_pages << PAGE_CACHE_SHIFT); if (ret) break; /* * 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(root, file, pages, num_pages, pos, first_index, last_index, write_bytes); if (ret) { btrfs_delalloc_release_space(inode, num_pages << PAGE_CACHE_SHIFT); break; } copied = btrfs_copy_from_user(pos, num_pages, write_bytes, pages, i); /* * if we have trouble faulting in the pages, fall * back to one page at a time */ if (copied < write_bytes) nrptrs = 1; if (copied == 0) dirty_pages = 0; else dirty_pages = (copied + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; /* * If we had a short copy we need to release the excess delaloc * bytes we reserved. We need to increment outstanding_extents * because btrfs_delalloc_release_space will decrement it, but * we still have an outstanding extent for the chunk we actually * managed to copy. */ if (num_pages > dirty_pages) { if (copied > 0) atomic_inc( &BTRFS_I(inode)->outstanding_extents); btrfs_delalloc_release_space(inode, (num_pages - dirty_pages) << PAGE_CACHE_SHIFT); } if (copied > 0) { ret = btrfs_dirty_pages(root, inode, pages, dirty_pages, pos, copied, NULL); if (ret) { btrfs_delalloc_release_space(inode, dirty_pages << PAGE_CACHE_SHIFT); btrfs_drop_pages(pages, num_pages); break; } } btrfs_drop_pages(pages, num_pages); cond_resched(); balance_dirty_pages_ratelimited_nr(inode->i_mapping, dirty_pages); if (dirty_pages < (root->leafsize >> PAGE_CACHE_SHIFT) + 1) btrfs_btree_balance_dirty(root, 1); btrfs_throttle(root); pos += copied; num_written += copied; } kfree(pages); return num_written ? num_written : ret; } static ssize_t __btrfs_direct_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos, loff_t *ppos, size_t count, size_t ocount) { struct file *file = iocb->ki_filp; struct inode *inode = fdentry(file)->d_inode; struct iov_iter i; ssize_t written; ssize_t written_buffered; loff_t endbyte; int err; written = generic_file_direct_write(iocb, iov, &nr_segs, pos, ppos, count, ocount); /* * the generic O_DIRECT will update in-memory i_size after the * DIOs are done. But our endio handlers that update the on * disk i_size never update past the in memory i_size. So we * need one more update here to catch any additions to the * file */ if (inode->i_size != BTRFS_I(inode)->disk_i_size) { btrfs_ordered_update_i_size(inode, inode->i_size, NULL); mark_inode_dirty(inode); } if (written < 0 || written == count) return written; pos += written; count -= written; iov_iter_init(&i, iov, nr_segs, count, written); written_buffered = __btrfs_buffered_write(file, &i, pos); if (written_buffered < 0) { err = written_buffered; goto out; } endbyte = pos + written_buffered - 1; err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); if (err) goto out; written += written_buffered; *ppos = pos + written_buffered; invalidate_mapping_pages(file->f_mapping, pos >> PAGE_CACHE_SHIFT, endbyte >> PAGE_CACHE_SHIFT); out: return written ? written : err; } static ssize_t btrfs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { struct file *file = iocb->ki_filp; struct inode *inode = fdentry(file)->d_inode; struct btrfs_root *root = BTRFS_I(inode)->root; loff_t *ppos = &iocb->ki_pos; ssize_t num_written = 0; ssize_t err = 0; size_t count, ocount; vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); mutex_lock(&inode->i_mutex); err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); if (err) { mutex_unlock(&inode->i_mutex); goto out; } count = ocount; current->backing_dev_info = inode->i_mapping->backing_dev_info; err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); if (err) { mutex_unlock(&inode->i_mutex); goto out; } if (count == 0) { mutex_unlock(&inode->i_mutex); goto out; } err = file_remove_suid(file); if (err) { mutex_unlock(&inode->i_mutex); goto out; } /* * If BTRFS flips readonly due to some impossible error * (fs_info->fs_state now has BTRFS_SUPER_FLAG_ERROR), * although we have opened a file as writable, we have * to stop this write operation to ensure FS consistency. */ if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) { mutex_unlock(&inode->i_mutex); err = -EROFS; goto out; } file_update_time(file); BTRFS_I(inode)->sequence++; if (unlikely(file->f_flags & O_DIRECT)) { num_written = __btrfs_direct_write(iocb, iov, nr_segs, pos, ppos, count, ocount); } else { struct iov_iter i; iov_iter_init(&i, iov, nr_segs, count, num_written); num_written = __btrfs_buffered_write(file, &i, pos); if (num_written > 0) *ppos = pos + num_written; } mutex_unlock(&inode->i_mutex); /* * we want to make sure fsync finds this change * but we haven't joined a transaction running right now. * * Later on, someone is sure to update the inode and get the * real transid recorded. * * We set last_trans now to the fs_info generation + 1, * this will either be one more than the running transaction * or the generation used for the next transaction if there isn't * one running right now. */ BTRFS_I(inode)->last_trans = root->fs_info->generation + 1; if (num_written > 0 || num_written == -EIOCBQUEUED) { err = generic_write_sync(file, pos, num_written); if (err < 0 && num_written > 0) num_written = err; } out: current->backing_dev_info = NULL; return num_written ? num_written : err; } int btrfs_release_file(struct inode *inode, struct file *filp) { /* * ordered_data_close is set by settattr when we are about to truncate * a file from a non-zero size to a zero size. This tries to * flush down new bytes that may have been written if the * application were using truncate to replace a file in place. */ if (BTRFS_I(inode)->ordered_data_close) { BTRFS_I(inode)->ordered_data_close = 0; btrfs_add_ordered_operation(NULL, BTRFS_I(inode)->root, inode); if (inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT) filemap_flush(inode->i_mapping); } if (filp->private_data) btrfs_ioctl_trans_end(filp); return 0; } /* * fsync call for both files and directories. This logs the inode into * the tree log instead of forcing full commits whenever possible. * * It needs to call filemap_fdatawait so that all ordered extent updates are * in the metadata btree are up to date for copying to the log. * * It drops the inode mutex before doing the tree log commit. This is an * important optimization for directories because holding the mutex prevents * new operations on the dir while we write to disk. */ int btrfs_sync_file(struct file *file, int datasync) { struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; struct btrfs_root *root = BTRFS_I(inode)->root; int ret = 0; struct btrfs_trans_handle *trans; trace_btrfs_sync_file(file, datasync); /* we wait first, since the writeback may change the inode */ root->log_batch++; /* the VFS called filemap_fdatawrite for us */ btrfs_wait_ordered_range(inode, 0, (u64)-1); root->log_batch++; /* * check the transaction that last modified this inode * and see if its already been committed */ if (!BTRFS_I(inode)->last_trans) goto out; /* * if the last transaction that changed this file was before * the current transaction, we can bail out now without any * syncing */ smp_mb(); if (BTRFS_I(inode)->last_trans <= root->fs_info->last_trans_committed) { BTRFS_I(inode)->last_trans = 0; goto out; } /* * ok we haven't committed the transaction yet, lets do a commit */ if (file->private_data) btrfs_ioctl_trans_end(file); trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; } ret = btrfs_log_dentry_safe(trans, root, dentry); if (ret < 0) goto out; /* we've logged all the items and now have a consistent * version of the file in the log. It is possible that * someone will come in and modify the file, but that's * fine because the log is consistent on disk, and we * have references to all of the file's extents * * It is possible that someone will come in and log the * file again, but that will end up using the synchronization * inside btrfs_sync_log to keep things safe. */ mutex_unlock(&dentry->d_inode->i_mutex); if (ret != BTRFS_NO_LOG_SYNC) { if (ret > 0) { ret = btrfs_commit_transaction(trans, root); } else { ret = btrfs_sync_log(trans, root); if (ret == 0) ret = btrfs_end_transaction(trans, root); else ret = btrfs_commit_transaction(trans, root); } } else { ret = btrfs_end_transaction(trans, root); } mutex_lock(&dentry->d_inode->i_mutex); out: return ret > 0 ? -EIO : ret; } static const struct vm_operations_struct btrfs_file_vm_ops = { .fault = filemap_fault, .page_mkwrite = btrfs_page_mkwrite, }; static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma) { struct address_space *mapping = filp->f_mapping; if (!mapping->a_ops->readpage) return -ENOEXEC; file_accessed(filp); vma->vm_ops = &btrfs_file_vm_ops; vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } static long btrfs_fallocate(struct file *file, int mode, loff_t offset, loff_t len) { struct inode *inode = file->f_path.dentry->d_inode; struct extent_state *cached_state = NULL; u64 cur_offset; u64 last_byte; u64 alloc_start; u64 alloc_end; u64 alloc_hint = 0; u64 locked_end; u64 mask = BTRFS_I(inode)->root->sectorsize - 1; struct extent_map *em; int ret; alloc_start = offset & ~mask; alloc_end = (offset + len + mask) & ~mask; /* We only support the FALLOC_FL_KEEP_SIZE mode */ if (mode & ~FALLOC_FL_KEEP_SIZE) return -EOPNOTSUPP; /* * wait for ordered IO before we have any locks. We'll loop again * below with the locks held. */ btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); mutex_lock(&inode->i_mutex); ret = inode_newsize_ok(inode, alloc_end); if (ret) goto out; if (alloc_start > inode->i_size) { ret = btrfs_cont_expand(inode, i_size_read(inode), alloc_start); if (ret) goto out; } ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start); if (ret) goto out; locked_end = alloc_end - 1; while (1) { struct btrfs_ordered_extent *ordered; /* the extent lock is ordered inside the running * transaction */ lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, 0, &cached_state, GFP_NOFS); ordered = btrfs_lookup_first_ordered_extent(inode, alloc_end - 1); if (ordered && ordered->file_offset + ordered->len > 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, GFP_NOFS); /* * we can't wait on the range with the transaction * running or with the extent lock held */ btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); } else { if (ordered) btrfs_put_ordered_extent(ordered); break; } } cur_offset = alloc_start; while (1) { em = btrfs_get_extent(inode, NULL, 0, cur_offset, alloc_end - cur_offset, 0); BUG_ON(IS_ERR_OR_NULL(em)); last_byte = min(extent_map_end(em), alloc_end); last_byte = (last_byte + mask) & ~mask; if (em->block_start == EXTENT_MAP_HOLE || (cur_offset >= inode->i_size && !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { ret = btrfs_prealloc_file_range(inode, mode, cur_offset, last_byte - cur_offset, 1 << inode->i_blkbits, offset + len, &alloc_hint); if (ret < 0) { free_extent_map(em); break; } } free_extent_map(em); cur_offset = last_byte; if (cur_offset >= alloc_end) { ret = 0; break; } } unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, &cached_state, GFP_NOFS); btrfs_free_reserved_data_space(inode, alloc_end - alloc_start); out: mutex_unlock(&inode->i_mutex); return ret; } const struct file_operations btrfs_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, .write = do_sync_write, .aio_read = generic_file_aio_read, .splice_read = generic_file_splice_read, .aio_write = btrfs_file_aio_write, .mmap = btrfs_file_mmap, .open = generic_file_open, .release = btrfs_release_file, .fsync = btrfs_sync_file, .fallocate = btrfs_fallocate, .unlocked_ioctl = btrfs_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = btrfs_ioctl, #endif };