Merge branch 'for-linus-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs fixes from Chris Mason:
 "The most user visible change here is a fix for our recent superblock
  validation checks that were causing problems on non-4k pagesized
  systems"

* 'for-linus-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  Btrfs: btrfs_check_super_valid: Allow 4096 as stripesize
  btrfs: remove build fixup for qgroup_account_snapshot
  btrfs: use new error message helper in qgroup_account_snapshot
  btrfs: avoid blocking open_ctree from cleaner_kthread
  Btrfs: don't BUG_ON() in btrfs_orphan_add
  btrfs: account for non-CoW'd blocks in btrfs_abort_transaction
  Btrfs: check if extent buffer is aligned to sectorsize
  btrfs: Use correct format specifier

11 files changed, 61 insertions, 43 deletions

diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index b677a6ea6001..7706c8dc5fa6 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -2645,7 +2645,7 @@ static void btrfsic_dump_tree_sub(const struct btrfsic_state *state,
          * This algorithm is recursive because the amount of used stack space
          * is very small and the max recursion depth is limited.
          */
-        indent_add = sprintf(buf, "%c-%llu(%s/%llu/%d)",
+        indent_add = sprintf(buf, "%c-%llu(%s/%llu/%u)",
                              btrfsic_get_block_type(state, block),
                              block->logical_bytenr, block->dev_state->name,
                              block->dev_bytenr, block->mirror_num);
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 46025688f1d0..6276add8538a 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -1554,6 +1554,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,
                        trans->transid, root->fs_info->generation);
 
         if (!should_cow_block(trans, root, buf)) {
+                trans->dirty = true;
                 *cow_ret = buf;
                 return 0;
         }
@@ -2512,6 +2513,8 @@ read_block_for_search(struct btrfs_trans_handle *trans,
                 if (!btrfs_buffer_uptodate(tmp, 0, 0))
                         ret = -EIO;
                 free_extent_buffer(tmp);
+        } else {
+                ret = PTR_ERR(tmp);
         }
         return ret;
 }
@@ -2775,8 +2778,10 @@ again:
                          * then we don't want to set the path blocking,
                          * so we test it here
                          */
-                        if (!should_cow_block(trans, root, b))
+                        if (!should_cow_block(trans, root, b)) {
+                                trans->dirty = true;
                                 goto cow_done;
+                        }
 
                         /*
                          * must have write locks on this node and the
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 1142127f6e5e..54cca7a1572b 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1098,7 +1098,7 @@ void readahead_tree_block(struct btrfs_root *root, u64 bytenr)
         struct inode *btree_inode = root->fs_info->btree_inode;
 
         buf = btrfs_find_create_tree_block(root, bytenr);
-        if (!buf)
+        if (IS_ERR(buf))
                 return;
         read_extent_buffer_pages(&BTRFS_I(btree_inode)->io_tree,
                                  buf, 0, WAIT_NONE, btree_get_extent, 0);
@@ -1114,7 +1114,7 @@ int reada_tree_block_flagged(struct btrfs_root *root, u64 bytenr,
         int ret;
 
         buf = btrfs_find_create_tree_block(root, bytenr);
-        if (!buf)
+        if (IS_ERR(buf))
                 return 0;
 
         set_bit(EXTENT_BUFFER_READAHEAD, &buf->bflags);
@@ -1172,8 +1172,8 @@ struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr,
         int ret;
 
         buf = btrfs_find_create_tree_block(root, bytenr);
-        if (!buf)
-                return ERR_PTR(-ENOMEM);
+        if (IS_ERR(buf))
+                return buf;
 
         ret = btree_read_extent_buffer_pages(root, buf, 0, parent_transid);
         if (ret) {
@@ -1806,6 +1806,13 @@ static int cleaner_kthread(void *arg)
                 if (btrfs_need_cleaner_sleep(root))
                         goto sleep;
 
+                /*
+                 * Do not do anything if we might cause open_ctree() to block
+                 * before we have finished mounting the filesystem.
+                 */
+                if (!root->fs_info->open)
+                        goto sleep;
+
                 if (!mutex_trylock(&root->fs_info->cleaner_mutex))
                         goto sleep;
 
@@ -2520,7 +2527,6 @@ int open_ctree(struct super_block *sb,
         int num_backups_tried = 0;
         int backup_index = 0;
         int max_active;
-        bool cleaner_mutex_locked = false;
 
         tree_root = fs_info->tree_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
         chunk_root = fs_info->chunk_root = btrfs_alloc_root(fs_info, GFP_KERNEL);
@@ -2999,13 +3005,6 @@ retry_root_backup:
                 goto fail_sysfs;
         }
 
-        /*
-         * Hold the cleaner_mutex thread here so that we don't block
-         * for a long time on btrfs_recover_relocation.  cleaner_kthread
-         * will wait for us to finish mounting the filesystem.
-         */
-        mutex_lock(&fs_info->cleaner_mutex);
-        cleaner_mutex_locked = true;
         fs_info->cleaner_kthread = kthread_run(cleaner_kthread, tree_root,
                                                "btrfs-cleaner");
         if (IS_ERR(fs_info->cleaner_kthread))
@@ -3065,8 +3064,10 @@ retry_root_backup:
                 ret = btrfs_cleanup_fs_roots(fs_info);
                 if (ret)
                         goto fail_qgroup;
-                /* We locked cleaner_mutex before creating cleaner_kthread. */
+
+                mutex_lock(&fs_info->cleaner_mutex);
                 ret = btrfs_recover_relocation(tree_root);
+                mutex_unlock(&fs_info->cleaner_mutex);
                 if (ret < 0) {
                         btrfs_warn(fs_info, "failed to recover relocation: %d",
                                         ret);
@@ -3074,8 +3075,6 @@ retry_root_backup:
                         goto fail_qgroup;
                 }
         }
-        mutex_unlock(&fs_info->cleaner_mutex);
-        cleaner_mutex_locked = false;
 
         location.objectid = BTRFS_FS_TREE_OBJECTID;
         location.type = BTRFS_ROOT_ITEM_KEY;
@@ -3189,10 +3188,6 @@ fail_cleaner:
         filemap_write_and_wait(fs_info->btree_inode->i_mapping);
 
 fail_sysfs:
-        if (cleaner_mutex_locked) {
-                mutex_unlock(&fs_info->cleaner_mutex);
-                cleaner_mutex_locked = false;
-        }
         btrfs_sysfs_remove_mounted(fs_info);
 
 fail_fsdev_sysfs:
@@ -4139,7 +4134,8 @@ static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
                 ret = -EINVAL;
         }
         if (!is_power_of_2(btrfs_super_stripesize(sb)) ||
-            btrfs_super_stripesize(sb) != sectorsize) {
+                ((btrfs_super_stripesize(sb) != sectorsize) &&
+                        (btrfs_super_stripesize(sb) != 4096))) {
                 btrfs_err(fs_info, "invalid stripesize %u",
                        btrfs_super_stripesize(sb));
                 ret = -EINVAL;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 689d25ac6a68..29e5d000bbee 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -8016,8 +8016,9 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
         struct extent_buffer *buf;
 
         buf = btrfs_find_create_tree_block(root, bytenr);
-        if (!buf)
-                return ERR_PTR(-ENOMEM);
+        if (IS_ERR(buf))
+                return buf;
+
         btrfs_set_header_generation(buf, trans->transid);
         btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level);
         btrfs_tree_lock(buf);
@@ -8044,7 +8045,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root,
                 set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
                          buf->start + buf->len - 1, GFP_NOFS);
         }
-        trans->blocks_used++;
+        trans->dirty = true;
         /* this returns a buffer locked for blocking */
         return buf;
 }
@@ -8659,8 +8660,9 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
         next = btrfs_find_tree_block(root->fs_info, bytenr);
         if (!next) {
                 next = btrfs_find_create_tree_block(root, bytenr);
-                if (!next)
-                        return -ENOMEM;
+                if (IS_ERR(next))
+                        return PTR_ERR(next);
+
                 btrfs_set_buffer_lockdep_class(root->root_key.objectid, next,
                                                level - 1);
                 reada = 1;
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index a3412d68ad37..aaee3ef55ed8 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -4892,18 +4892,25 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
         int uptodate = 1;
         int ret;
 
+        if (!IS_ALIGNED(start, fs_info->tree_root->sectorsize)) {
+                btrfs_err(fs_info, "bad tree block start %llu", start);
+                return ERR_PTR(-EINVAL);
+        }
+
         eb = find_extent_buffer(fs_info, start);
         if (eb)
                 return eb;
 
         eb = __alloc_extent_buffer(fs_info, start, len);
         if (!eb)
-                return NULL;
+                return ERR_PTR(-ENOMEM);
 
         for (i = 0; i < num_pages; i++, index++) {
                 p = find_or_create_page(mapping, index, GFP_NOFS|__GFP_NOFAIL);
-                if (!p)
+                if (!p) {
+                        exists = ERR_PTR(-ENOMEM);
                         goto free_eb;
+                }
 
                 spin_lock(&mapping->private_lock);
                 if (PagePrivate(p)) {
@@ -4948,8 +4955,10 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
                 set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 again:
         ret = radix_tree_preload(GFP_NOFS);
-        if (ret)
+        if (ret) {
+                exists = ERR_PTR(ret);
                 goto free_eb;
+        }
 
         spin_lock(&fs_info->buffer_lock);
         ret = radix_tree_insert(&fs_info->buffer_radix,
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 8b1212e8f7a8..d2be95cfb6d1 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3271,7 +3271,16 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode)
         /* grab metadata reservation from transaction handle */
         if (reserve) {
                 ret = btrfs_orphan_reserve_metadata(trans, inode);
-                BUG_ON(ret); /* -ENOSPC in reservation; Logic error? JDM */
+                ASSERT(!ret);
+                if (ret) {
+                        atomic_dec(&root->orphan_inodes);
+                        clear_bit(BTRFS_INODE_ORPHAN_META_RESERVED,
+                                  &BTRFS_I(inode)->runtime_flags);
+                        if (insert)
+                                clear_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
+                                          &BTRFS_I(inode)->runtime_flags);
+                        return ret;
+                }
         }
 
         /* insert an orphan item to track this unlinked/truncated file */
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 4339b6613f19..60e7179ed4b7 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -235,7 +235,7 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
         trans->aborted = errno;
         /* Nothing used. The other threads that have joined this
          * transaction may be able to continue. */
-        if (!trans->blocks_used && list_empty(&trans->new_bgs)) {
+        if (!trans->dirty && list_empty(&trans->new_bgs)) {
                 const char *errstr;
 
                 errstr = btrfs_decode_error(errno);
@@ -1807,6 +1807,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
                         }
                 }
                 sb->s_flags &= ~MS_RDONLY;
+
+                fs_info->open = 1;
         }
 out:
         wake_up_process(fs_info->transaction_kthread);
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index f6e24cb423ae..765845742fde 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -1311,11 +1311,6 @@ int btrfs_defrag_root(struct btrfs_root *root)
         return ret;
 }
 
-/* Bisesctability fixup, remove in 4.8 */
-#ifndef btrfs_std_error
-#define btrfs_std_error btrfs_handle_fs_error
-#endif
-
 /*
  * Do all special snapshot related qgroup dirty hack.
  *
@@ -1385,7 +1380,7 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
         switch_commit_roots(trans->transaction, fs_info);
         ret = btrfs_write_and_wait_transaction(trans, src);
         if (ret)
-                btrfs_std_error(fs_info, ret,
+                btrfs_handle_fs_error(fs_info, ret,
                         "Error while writing out transaction for qgroup");
 
 out:
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 9fe0ec2bf0fe..c5abee4f01ad 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -110,7 +110,6 @@ struct btrfs_trans_handle {
         u64 chunk_bytes_reserved;
         unsigned long use_count;
         unsigned long blocks_reserved;
-        unsigned long blocks_used;
         unsigned long delayed_ref_updates;
         struct btrfs_transaction *transaction;
         struct btrfs_block_rsv *block_rsv;
@@ -121,6 +120,7 @@ struct btrfs_trans_handle {
         bool can_flush_pending_bgs;
         bool reloc_reserved;
         bool sync;
+        bool dirty;
         unsigned int type;
         /*
          * this root is only needed to validate that the root passed to
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index b7665af471d8..c05f69a8ec42 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -2422,8 +2422,8 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
                 root_owner = btrfs_header_owner(parent);
 
                 next = btrfs_find_create_tree_block(root, bytenr);
-                if (!next)
-                        return -ENOMEM;
+                if (IS_ERR(next))
+                        return PTR_ERR(next);
 
                 if (*level == 1) {
                         ret = wc->process_func(root, next, wc, ptr_gen);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 548faaa9e169..2f631b58ae00 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -6607,8 +6607,8 @@ int btrfs_read_sys_array(struct btrfs_root *root)
          * overallocate but we can keep it as-is, only the first page is used.
          */
         sb = btrfs_find_create_tree_block(root, BTRFS_SUPER_INFO_OFFSET);
-        if (!sb)
-                return -ENOMEM;
+        if (IS_ERR(sb))
+                return PTR_ERR(sb);
         set_extent_buffer_uptodate(sb);
         btrfs_set_buffer_lockdep_class(root->root_key.objectid, sb, 0);
         /*