From d763448286377b8a0e3f179372e9e292bef3c337 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 11 Mar 2013 09:20:00 +0000 Subject: [PATCH 01/15] Btrfs: update to use fs_state bit Now that we use bit operation to check fs_state, update btrfs_free_fs_root()'s checker, otherwise we get back to memory leak case. Signed-off-by: Liu Bo Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/disk-io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 7d84651e850b..127b23e8323b 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3253,7 +3253,7 @@ void btrfs_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root) if (btrfs_root_refs(&root->root_item) == 0) synchronize_srcu(&fs_info->subvol_srcu); - if (fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) { + if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) { btrfs_free_log(NULL, root); btrfs_free_log_root_tree(NULL, fs_info); } From 835d974fabfa9bff4d173ad03c054ac2f673263f Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 19 Mar 2013 12:13:25 -0400 Subject: [PATCH 02/15] Btrfs: handle a bogus chunk tree nicely If you restore a btrfs-image file system and try to mount that file system we'll panic. That's because btrfs-image restores and just makes one big chunk to envelope the whole disk, since they are really only meant to be messed with by our btrfs-progs. So fix up btrfs_rmap_block and the callers of it for mount so that we no longer panic but instead just return an error and fail to mount. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 35 ++++++++++++++++++++++++++++++----- fs/btrfs/volumes.c | 13 ++++++++++++- 2 files changed, 42 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 350b9b18140c..a8ff25aedca1 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -257,7 +257,8 @@ static int exclude_super_stripes(struct btrfs_root *root, cache->bytes_super += stripe_len; ret = add_excluded_extent(root, cache->key.objectid, stripe_len); - BUG_ON(ret); /* -ENOMEM */ + if (ret) + return ret; } for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) { @@ -265,13 +266,17 @@ static int exclude_super_stripes(struct btrfs_root *root, ret = btrfs_rmap_block(&root->fs_info->mapping_tree, cache->key.objectid, bytenr, 0, &logical, &nr, &stripe_len); - BUG_ON(ret); /* -ENOMEM */ + if (ret) + return ret; while (nr--) { cache->bytes_super += stripe_len; ret = add_excluded_extent(root, logical[nr], stripe_len); - BUG_ON(ret); /* -ENOMEM */ + if (ret) { + kfree(logical); + return ret; + } } kfree(logical); @@ -7964,7 +7969,17 @@ int btrfs_read_block_groups(struct btrfs_root *root) * info has super bytes accounted for, otherwise we'll think * we have more space than we actually do. */ - exclude_super_stripes(root, cache); + ret = exclude_super_stripes(root, cache); + if (ret) { + /* + * We may have excluded something, so call this just in + * case. + */ + free_excluded_extents(root, cache); + kfree(cache->free_space_ctl); + kfree(cache); + goto error; + } /* * check for two cases, either we are full, and therefore @@ -8106,7 +8121,17 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, cache->last_byte_to_unpin = (u64)-1; cache->cached = BTRFS_CACHE_FINISHED; - exclude_super_stripes(root, cache); + ret = exclude_super_stripes(root, cache); + if (ret) { + /* + * We may have excluded something, so call this just in + * case. + */ + free_excluded_extents(root, cache); + kfree(cache->free_space_ctl); + kfree(cache); + return ret; + } add_new_free_space(cache, root->fs_info, chunk_offset, chunk_offset + size); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 5989a92236f7..2854c824ab64 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -4935,7 +4935,18 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree, em = lookup_extent_mapping(em_tree, chunk_start, 1); read_unlock(&em_tree->lock); - BUG_ON(!em || em->start != chunk_start); + if (!em) { + printk(KERN_ERR "btrfs: couldn't find em for chunk %Lu\n", + chunk_start); + return -EIO; + } + + if (em->start != chunk_start) { + printk(KERN_ERR "btrfs: bad chunk start, em=%Lu, wanted=%Lu\n", + em->start, chunk_start); + free_extent_map(em); + return -EIO; + } map = (struct map_lookup *)em->bdev; length = em->len; From 6113077cd319e747875ec71227d2b5cb54e08c76 Mon Sep 17 00:00:00 2001 From: Wang Shilong Date: Tue, 19 Mar 2013 10:57:14 +0000 Subject: [PATCH 03/15] Btrfs: fix missing qgroup reservation before fallocating Steps to reproduce: mkfs.btrfs mount btrfs quota enable btrfs sub create /subv btrfs qgroup limit 10M /subv fallocate --length 20M /subv/data For the above example, fallocating will return successfully which is not expected, we try to fix it by doing qgroup reservation before fallocating. Signed-off-by: Wang Shilong Reviewed-by: Miao Xie Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/file.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 7bdb47faa12e..1be25b92d63c 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2142,6 +2142,7 @@ static long btrfs_fallocate(struct file *file, int mode, { struct inode *inode = file->f_path.dentry->d_inode; struct extent_state *cached_state = NULL; + struct btrfs_root *root = BTRFS_I(inode)->root; u64 cur_offset; u64 last_byte; u64 alloc_start; @@ -2169,6 +2170,11 @@ static long btrfs_fallocate(struct file *file, int mode, ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start); if (ret) return ret; + if (root->fs_info->quota_enabled) { + ret = btrfs_qgroup_reserve(root, alloc_end - alloc_start); + if (ret) + goto out_reserve_fail; + } /* * wait for ordered IO before we have any locks. We'll loop again @@ -2272,6 +2278,9 @@ static long btrfs_fallocate(struct file *file, int mode, &cached_state, GFP_NOFS); out: mutex_unlock(&inode->i_mutex); + if (root->fs_info->quota_enabled) + btrfs_qgroup_free(root, alloc_end - alloc_start); +out_reserve_fail: /* Let go of our reservation. */ btrfs_free_reserved_data_space(inode, alloc_end - alloc_start); return ret; From d9abbf1c3131b679379762700201ae69367f3f62 Mon Sep 17 00:00:00 2001 From: Jan Schmidt Date: Wed, 20 Mar 2013 13:49:48 +0000 Subject: [PATCH 04/15] Btrfs: fix locking on ROOT_REPLACE operations in tree mod log To resolve backrefs, ROOT_REPLACE operations in the tree mod log are required to be tied to at least one KEY_REMOVE_WHILE_FREEING operation. Therefore, those operations must be enclosed by tree_mod_log_write_lock() and tree_mod_log_write_unlock() calls. Those calls are private to the tree_mod_log_* functions, which means that removal of the elements of an old root node must be logged from tree_mod_log_insert_root. This partly reverts and corrects commit ba1bfbd5 (Btrfs: fix a tree mod logging issue for root replacement operations). This fixes the brand-new version of xfstest 276 as of commit cfe73f71. Cc: stable@vger.kernel.org Signed-off-by: Jan Schmidt Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/ctree.c | 30 ++++++++++++++++++++---------- 1 file changed, 20 insertions(+), 10 deletions(-) diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index ecd25a1b4e51..ca9d8f1a3bb6 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -651,6 +651,8 @@ tree_mod_log_insert_root(struct btrfs_fs_info *fs_info, if (tree_mod_dont_log(fs_info, NULL)) return 0; + __tree_mod_log_free_eb(fs_info, old_root); + ret = tree_mod_alloc(fs_info, flags, &tm); if (ret < 0) goto out; @@ -736,7 +738,7 @@ tree_mod_log_search(struct btrfs_fs_info *fs_info, u64 start, u64 min_seq) static noinline void tree_mod_log_eb_copy(struct btrfs_fs_info *fs_info, struct extent_buffer *dst, struct extent_buffer *src, unsigned long dst_offset, - unsigned long src_offset, int nr_items) + unsigned long src_offset, int nr_items, int log_removal) { int ret; int i; @@ -750,10 +752,12 @@ tree_mod_log_eb_copy(struct btrfs_fs_info *fs_info, struct extent_buffer *dst, } for (i = 0; i < nr_items; i++) { - ret = tree_mod_log_insert_key_locked(fs_info, src, - i + src_offset, - MOD_LOG_KEY_REMOVE); - BUG_ON(ret < 0); + if (log_removal) { + ret = tree_mod_log_insert_key_locked(fs_info, src, + i + src_offset, + MOD_LOG_KEY_REMOVE); + BUG_ON(ret < 0); + } ret = tree_mod_log_insert_key_locked(fs_info, dst, i + dst_offset, MOD_LOG_KEY_ADD); @@ -927,7 +931,6 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, ret = btrfs_dec_ref(trans, root, buf, 1, 1); BUG_ON(ret); /* -ENOMEM */ } - tree_mod_log_free_eb(root->fs_info, buf); clean_tree_block(trans, root, buf); *last_ref = 1; } @@ -1046,6 +1049,7 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans, btrfs_set_node_ptr_generation(parent, parent_slot, trans->transid); btrfs_mark_buffer_dirty(parent); + tree_mod_log_free_eb(root->fs_info, buf); btrfs_free_tree_block(trans, root, buf, parent_start, last_ref); } @@ -1750,7 +1754,6 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, goto enospc; } - tree_mod_log_free_eb(root->fs_info, root->node); tree_mod_log_set_root_pointer(root, child); rcu_assign_pointer(root->node, child); @@ -2995,7 +2998,7 @@ static int push_node_left(struct btrfs_trans_handle *trans, push_items = min(src_nritems - 8, push_items); tree_mod_log_eb_copy(root->fs_info, dst, src, dst_nritems, 0, - push_items); + push_items, 1); copy_extent_buffer(dst, src, btrfs_node_key_ptr_offset(dst_nritems), btrfs_node_key_ptr_offset(0), @@ -3066,7 +3069,7 @@ static int balance_node_right(struct btrfs_trans_handle *trans, sizeof(struct btrfs_key_ptr)); tree_mod_log_eb_copy(root->fs_info, dst, src, 0, - src_nritems - push_items, push_items); + src_nritems - push_items, push_items, 1); copy_extent_buffer(dst, src, btrfs_node_key_ptr_offset(0), btrfs_node_key_ptr_offset(src_nritems - push_items), @@ -3218,12 +3221,18 @@ static noinline int split_node(struct btrfs_trans_handle *trans, int mid; int ret; u32 c_nritems; + int tree_mod_log_removal = 1; c = path->nodes[level]; WARN_ON(btrfs_header_generation(c) != trans->transid); if (c == root->node) { /* trying to split the root, lets make a new one */ ret = insert_new_root(trans, root, path, level + 1); + /* + * removal of root nodes has been logged by + * tree_mod_log_set_root_pointer due to locking + */ + tree_mod_log_removal = 0; if (ret) return ret; } else { @@ -3261,7 +3270,8 @@ static noinline int split_node(struct btrfs_trans_handle *trans, (unsigned long)btrfs_header_chunk_tree_uuid(split), BTRFS_UUID_SIZE); - tree_mod_log_eb_copy(root->fs_info, split, c, 0, mid, c_nritems - mid); + tree_mod_log_eb_copy(root->fs_info, split, c, 0, mid, c_nritems - mid, + tree_mod_log_removal); copy_extent_buffer(split, c, btrfs_node_key_ptr_offset(0), btrfs_node_key_ptr_offset(mid), From 1dd05682b3ef6e70409e130bfd83e91770801589 Mon Sep 17 00:00:00 2001 From: Tsutomu Itoh Date: Thu, 21 Mar 2013 04:32:32 +0000 Subject: [PATCH 05/15] Btrfs: fix memory leak in btrfs_create_tree() We should free leaf and root before returning from the error handling code. Signed-off-by: Tsutomu Itoh Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/disk-io.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 127b23e8323b..6d19a0a554aa 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1291,6 +1291,7 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, 0, objectid, NULL, 0, 0, 0); if (IS_ERR(leaf)) { ret = PTR_ERR(leaf); + leaf = NULL; goto fail; } @@ -1334,11 +1335,16 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, btrfs_tree_unlock(leaf); -fail: - if (ret) - return ERR_PTR(ret); - return root; + +fail: + if (leaf) { + btrfs_tree_unlock(leaf); + free_extent_buffer(leaf); + } + kfree(root); + + return ERR_PTR(ret); } static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans, From 4adaa611020fa6ac65b0ac8db78276af4ec04e63 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Tue, 26 Mar 2013 13:07:00 -0400 Subject: [PATCH 06/15] Btrfs: fix race between mmap writes and compression Btrfs uses page_mkwrite to ensure stable pages during crc calculations and mmap workloads. We call clear_page_dirty_for_io before we do any crcs, and this forces any application with the file mapped to wait for the crc to finish before it is allowed to change the file. With compression on, the clear_page_dirty_for_io step is happening after we've compressed the pages. This means the applications might be changing the pages while we are compressing them, and some of those modifications might not hit the disk. This commit adds the clear_page_dirty_for_io before compression starts and makes sure to redirty the page if we have to fallback to uncompressed IO as well. Signed-off-by: Chris Mason Reported-by: Alexandre Oliva cc: stable@vger.kernel.org --- fs/btrfs/extent_io.c | 33 +++++++++++++++++++++++++++++++++ fs/btrfs/extent_io.h | 2 ++ fs/btrfs/inode.c | 14 ++++++++++++++ 3 files changed, 49 insertions(+) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index f173c5af6461..cdee391fc7bf 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -1257,6 +1257,39 @@ int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end) GFP_NOFS); } +int extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end) +{ + unsigned long index = start >> PAGE_CACHE_SHIFT; + unsigned long end_index = end >> PAGE_CACHE_SHIFT; + struct page *page; + + while (index <= end_index) { + page = find_get_page(inode->i_mapping, index); + BUG_ON(!page); /* Pages should be in the extent_io_tree */ + clear_page_dirty_for_io(page); + page_cache_release(page); + index++; + } + return 0; +} + +int extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end) +{ + unsigned long index = start >> PAGE_CACHE_SHIFT; + unsigned long end_index = end >> PAGE_CACHE_SHIFT; + struct page *page; + + while (index <= end_index) { + page = find_get_page(inode->i_mapping, index); + BUG_ON(!page); /* Pages should be in the extent_io_tree */ + account_page_redirty(page); + __set_page_dirty_nobuffers(page); + page_cache_release(page); + index++; + } + return 0; +} + /* * helper function to set both pages and extents in the tree writeback */ diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 6068a1985560..258c92156857 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -325,6 +325,8 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset, unsigned long *map_len); int extent_range_uptodate(struct extent_io_tree *tree, u64 start, u64 end); +int extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end); +int extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end); int extent_clear_unlock_delalloc(struct inode *inode, struct extent_io_tree *tree, u64 start, u64 end, struct page *locked_page, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1f26888825e2..6a6e13c53086 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -353,6 +353,7 @@ static noinline int compress_file_range(struct inode *inode, int i; int will_compress; int compress_type = root->fs_info->compress_type; + int redirty = 0; /* if this is a small write inside eof, kick off a defrag */ if ((end - start + 1) < 16 * 1024 && @@ -415,6 +416,17 @@ again: if (BTRFS_I(inode)->force_compress) compress_type = BTRFS_I(inode)->force_compress; + /* + * we need to call clear_page_dirty_for_io on each + * page in the range. Otherwise applications with the file + * mmap'd can wander in and change the page contents while + * we are compressing them. + * + * If the compression fails for any reason, we set the pages + * dirty again later on. + */ + extent_range_clear_dirty_for_io(inode, start, end); + redirty = 1; ret = btrfs_compress_pages(compress_type, inode->i_mapping, start, total_compressed, pages, @@ -554,6 +566,8 @@ cleanup_and_bail_uncompressed: __set_page_dirty_nobuffers(locked_page); /* unlocked later on in the async handlers */ } + if (redirty) + extent_range_redirty_for_io(inode, start, end); add_async_extent(async_cow, start, end - start + 1, 0, NULL, 0, BTRFS_COMPRESS_NONE); *num_added += 1; From adaa4b8e4d47eeb114513c2f7a172929154b94bd Mon Sep 17 00:00:00 2001 From: Jan Schmidt Date: Thu, 21 Mar 2013 14:30:23 +0000 Subject: [PATCH 07/15] Btrfs: fix EIO from btrfs send in is_extent_unchanged for punched holes When you take a snapshot, punch a hole where there has been data, then take another snapshot and try to send an incremental stream, btrfs send would give you EIO. That is because is_extent_unchanged had no support for holes being punched. With this patch, instead of returning EIO we just return 0 (== the extent is not unchanged) and we're good. Signed-off-by: Jan Schmidt Cc: Alexander Block Signed-off-by: Josef Bacik --- fs/btrfs/send.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 68da757615ae..ed897dc11356 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -3945,12 +3945,10 @@ static int is_extent_unchanged(struct send_ctx *sctx, found_key.type != key.type) { key.offset += right_len; break; - } else { - if (found_key.offset != key.offset + right_len) { - /* Should really not happen */ - ret = -EIO; - goto out; - } + } + if (found_key.offset != key.offset + right_len) { + ret = 0; + goto out; } key = found_key; } From f4881bc7a83eff263789dd524b7c269d138d4af5 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 25 Mar 2013 16:03:35 -0400 Subject: [PATCH 08/15] Btrfs: fix space leak when we fail to reserve metadata space Dave reported a warning when running xfstest 275. We have been leaking delalloc metadata space when our reservations fail. This is because we were improperly calculating how much space to free for our checksum reservations. The problem is we would sometimes free up space that had already been freed in another thread and we would end up with negative usage for the delalloc space. This patch fixes the problem by calculating how much space the other threads would have already freed, and then calculate how much space we need to free had we not done the reservation at all, and then freeing any excess space. This makes xfstests 275 no longer have leaked space. Thanks Cc: stable@vger.kernel.org Reported-by: David Sterba Signed-off-by: Josef Bacik --- fs/btrfs/extent-tree.c | 47 ++++++++++++++++++++++++++++++++++++------ 1 file changed, 41 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index a8ff25aedca1..a22b5cc921ad 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4815,14 +4815,49 @@ out_fail: * If the inodes csum_bytes is the same as the original * csum_bytes then we know we haven't raced with any free()ers * so we can just reduce our inodes csum bytes and carry on. - * Otherwise we have to do the normal free thing to account for - * the case that the free side didn't free up its reserve - * because of this outstanding reservation. */ - if (BTRFS_I(inode)->csum_bytes == csum_bytes) + if (BTRFS_I(inode)->csum_bytes == csum_bytes) { calc_csum_metadata_size(inode, num_bytes, 0); - else - to_free = calc_csum_metadata_size(inode, num_bytes, 0); + } else { + u64 orig_csum_bytes = BTRFS_I(inode)->csum_bytes; + u64 bytes; + + /* + * This is tricky, but first we need to figure out how much we + * free'd from any free-ers that occured during this + * reservation, so we reset ->csum_bytes to the csum_bytes + * before we dropped our lock, and then call the free for the + * number of bytes that were freed while we were trying our + * reservation. + */ + bytes = csum_bytes - BTRFS_I(inode)->csum_bytes; + BTRFS_I(inode)->csum_bytes = csum_bytes; + to_free = calc_csum_metadata_size(inode, bytes, 0); + + + /* + * Now we need to see how much we would have freed had we not + * been making this reservation and our ->csum_bytes were not + * artificially inflated. + */ + BTRFS_I(inode)->csum_bytes = csum_bytes - num_bytes; + bytes = csum_bytes - orig_csum_bytes; + bytes = calc_csum_metadata_size(inode, bytes, 0); + + /* + * Now reset ->csum_bytes to what it should be. If bytes is + * more than to_free then we would have free'd more space had we + * not had an artificially high ->csum_bytes, so we need to free + * the remainder. If bytes is the same or less then we don't + * need to do anything, the other free-ers did the correct + * thing. + */ + BTRFS_I(inode)->csum_bytes = orig_csum_bytes - num_bytes; + if (bytes > to_free) + to_free = bytes - to_free; + else + to_free = 0; + } spin_unlock(&BTRFS_I(inode)->lock); if (dropped) to_free += btrfs_calc_trans_metadata_size(root, dropped); From 6e137ed3f30574f314733d4b7a86ea6523232b14 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 26 Mar 2013 15:26:55 -0400 Subject: [PATCH 09/15] Btrfs: fix space accounting for unlink and rename We are way over-reserving for unlink and rename. Rename is just some random huge number and unlink accounts for tree log operations that don't actually happen during unlink, not to mention the tree log doesn't take from the trans block rsv anyway so it's completely useless. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 6a6e13c53086..8cab424c75f8 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3693,11 +3693,9 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir, * 1 for the dir item * 1 for the dir index * 1 for the inode ref - * 1 for the inode ref in the tree log - * 2 for the dir entries in the log * 1 for the inode */ - trans = btrfs_start_transaction(root, 8); + trans = btrfs_start_transaction(root, 5); if (!IS_ERR(trans) || PTR_ERR(trans) != -ENOSPC) return trans; @@ -8141,7 +8139,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, * inodes. So 5 * 2 is 10, plus 1 for the new link, so 11 total items * should cover the worst case number of items we'll modify. */ - trans = btrfs_start_transaction(root, 20); + trans = btrfs_start_transaction(root, 11); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out_notrans; From db1d607d3ca5cbb283cbb17d648cd7e8dc67cc7b Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 26 Mar 2013 15:29:11 -0400 Subject: [PATCH 10/15] Btrfs: hold the ordered operations mutex when waiting on ordered extents We need to hold the ordered_operations mutex while waiting on ordered extents since we splice and run the ordered extents list. We need to make sure anybody else who wants to wait on ordered extents does actually wait for them to be completed. This will keep us from bailing out of flushing in case somebody is already waiting on ordered extents to complete. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/ordered-data.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index dc08d77b717e..005c45db699e 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -557,6 +557,7 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput) INIT_LIST_HEAD(&splice); INIT_LIST_HEAD(&works); + mutex_lock(&root->fs_info->ordered_operations_mutex); spin_lock(&root->fs_info->ordered_extent_lock); list_splice_init(&root->fs_info->ordered_extents, &splice); while (!list_empty(&splice)) { @@ -600,6 +601,7 @@ void btrfs_wait_ordered_extents(struct btrfs_root *root, int delay_iput) cond_resched(); } + mutex_unlock(&root->fs_info->ordered_operations_mutex); } /* From fdf30d1c1b386e1b73116cc7e0fb14e962b763b0 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 26 Mar 2013 15:31:45 -0400 Subject: [PATCH 11/15] Btrfs: limit the global reserve to 512mb A user reported a problem where he was getting early ENOSPC with hundreds of gigs of free data space and 6 gigs of free metadata space. This is because the global block reserve was taking up the entire free metadata space. This is ridiculous, we have infrastructure in place to throttle if we start using too much of the global reserve, so instead of letting it get this huge just limit it to 512mb so that users can still get work done. This allowed the user to complete his rsync without issues. Thanks Cc: stable@vger.kernel.org Reported-and-tested-by: Stefan Priebe Signed-off-by: Josef Bacik --- fs/btrfs/extent-tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index a22b5cc921ad..0d8478700d78 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4460,7 +4460,7 @@ static void update_global_block_rsv(struct btrfs_fs_info *fs_info) spin_lock(&sinfo->lock); spin_lock(&block_rsv->lock); - block_rsv->size = num_bytes; + block_rsv->size = min_t(u64, num_bytes, 512 * 1024 * 1024); num_bytes = sinfo->bytes_used + sinfo->bytes_pinned + sinfo->bytes_reserved + sinfo->bytes_readonly + From a7975026ff9ddf91ba190ae2b71699dd156395e3 Mon Sep 17 00:00:00 2001 From: Wang Shilong Date: Mon, 25 Mar 2013 11:08:23 +0000 Subject: [PATCH 12/15] Btrfs: fix double free in the btrfs_qgroup_account_ref() The function btrfs_find_all_roots is responsible to allocate memory for 'roots' and free it if errors happen,so the caller should not free it again since the work has been done. Besides,'tmp' is allocated after the function btrfs_find_all_roots, so we can return directly if btrfs_find_all_roots() fails. Signed-off-by: Wang Shilong Reviewed-by: Miao Xie Reviewed-by: Jan Schmidt Signed-off-by: Josef Bacik --- fs/btrfs/qgroup.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 5471e47d6559..b44124dd2370 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1153,7 +1153,7 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, ret = btrfs_find_all_roots(trans, fs_info, node->bytenr, sgn > 0 ? node->seq - 1 : node->seq, &roots); if (ret < 0) - goto out; + return ret; spin_lock(&fs_info->qgroup_lock); quota_root = fs_info->quota_root; @@ -1275,7 +1275,6 @@ int btrfs_qgroup_account_ref(struct btrfs_trans_handle *trans, ret = 0; unlock: spin_unlock(&fs_info->qgroup_lock); -out: ulist_free(roots); ulist_free(tmp); From 39847c4d3d91f487f9ab3d083ee5d0f8419f105c Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Thu, 28 Mar 2013 08:08:20 +0000 Subject: [PATCH 13/15] Btrfs: fix wrong reservation of csums We reserve the space for csums only when we write data into a file, in the other cases, such as tree log, log replay, we don't do reservation, so we can use the reservation of the transaction handle just for the former. And for the latter, we should use the tree's own reservation. But the function - btrfs_csum_file_blocks() didn't differentiate between these two types of the cases, fix it. Signed-off-by: Miao Xie Signed-off-by: Josef Bacik --- fs/btrfs/file-item.c | 2 -- fs/btrfs/inode.c | 2 ++ 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index ec160202be3e..b7e529d2860f 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -728,7 +728,6 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans, return -ENOMEM; sector_sum = sums->sums; - trans->adding_csums = 1; again: next_offset = (u64)-1; found_next = 0; @@ -899,7 +898,6 @@ next_sector: goto again; } out: - trans->adding_csums = 0; btrfs_free_path(path); return ret; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8cab424c75f8..b88381582dab 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1757,8 +1757,10 @@ static noinline int add_pending_csums(struct btrfs_trans_handle *trans, struct btrfs_ordered_sum *sum; list_for_each_entry(sum, list, list) { + trans->adding_csums = 1; btrfs_csum_file_blocks(trans, BTRFS_I(inode)->root->fs_info->csum_root, sum); + trans->adding_csums = 0; } return 0; } From 82d130ff390be67d980d8b6f39e921c0b1d8d8e0 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Thu, 28 Mar 2013 08:12:15 +0000 Subject: [PATCH 14/15] Btrfs: fix wrong return value of btrfs_lookup_csum() If we don't find the expected csum item, but find a csum item which is adjacent to the specified extent, we should return -EFBIG, or we should return -ENOENT. But btrfs_lookup_csum() return -EFBIG even the csum item is not adjacent to the specified extent. Fix it. Signed-off-by: Miao Xie Signed-off-by: Josef Bacik --- fs/btrfs/file-item.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index b7e529d2860f..c4628a201cb3 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -118,9 +118,11 @@ struct btrfs_csum_item *btrfs_lookup_csum(struct btrfs_trans_handle *trans, csums_in_item = btrfs_item_size_nr(leaf, path->slots[0]); csums_in_item /= csum_size; - if (csum_offset >= csums_in_item) { + if (csum_offset == csums_in_item) { ret = -EFBIG; goto fail; + } else if (csum_offset > csums_in_item) { + goto fail; } } item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item); From d8fe29e9dea8d7d61fd140d8779326856478fc62 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 29 Mar 2013 08:09:34 -0600 Subject: [PATCH 15/15] Btrfs: don't drop path when printing out tree errors in scrub A user reported a panic where we were panicing somewhere in tree_backref_for_extent from scrub_print_warning. He only captured the trace but looking at scrub_print_warning we drop the path right before we mess with the extent buffer to print out a bunch of stuff, which isn't right. So fix this by dropping the path after we use the eb if we need to. Thanks, Cc: stable@vger.kernel.org Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/scrub.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 53c3501fa4ca..85e072b956d5 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -542,7 +542,6 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) eb = path->nodes[0]; ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item); item_size = btrfs_item_size_nr(eb, path->slots[0]); - btrfs_release_path(path); if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) { do { @@ -558,7 +557,9 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) ret < 0 ? -1 : ref_level, ret < 0 ? -1 : ref_root); } while (ret != 1); + btrfs_release_path(path); } else { + btrfs_release_path(path); swarn.path = path; swarn.dev = dev; iterate_extent_inodes(fs_info, found_key.objectid,