Merge branch 'for-linus-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs fixes from Chris Mason:
"These are bug fixes, including a really old fsync bug, and a few trace
points to help us track down problems in the quota code"

* 'for-linus-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix file/data loss caused by fsync after rename and new inode
btrfs: Reset IO error counters before start of device replacing
btrfs: Add qgroup tracing
Btrfs: don't use src fd for printk
btrfs: fallback to vmalloc in btrfs_compare_tree
btrfs: handle non-fatal errors in btrfs_qgroup_inherit()
btrfs: Output more info for enospc_debug mount option
Btrfs: fix invalid reference in replace_path
Btrfs: Improve FL_KEEP_SIZE handling in fallocate

fs/btrfs/ctree.c

···
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/rbtree.h>
+#include <linux/vmalloc.h>
 #include "ctree.h"
 #include "disk-io.h"
 #include "transaction.h"
···
 		goto out;
 	}
 
-	tmp_buf = kmalloc(left_root->nodesize, GFP_KERNEL);
+	tmp_buf = kmalloc(left_root->nodesize, GFP_KERNEL | __GFP_NOWARN);
 	if (!tmp_buf) {
-		ret = -ENOMEM;
-		goto out;
+		tmp_buf = vmalloc(left_root->nodesize);
+		if (!tmp_buf) {
+			ret = -ENOMEM;
+			goto out;
+		}
 	}
 
 	left_path->search_commit_root = 1;
···
 out:
 	btrfs_free_path(left_path);
 	btrfs_free_path(right_path);
-	kfree(tmp_buf);
+	kvfree(tmp_buf);
 	return ret;
 }
 

fs/btrfs/dev-replace.c

···
 	dev_replace->cursor_right = 0;
 	dev_replace->is_valid = 1;
 	dev_replace->item_needs_writeback = 1;
+	atomic64_set(&dev_replace->num_write_errors, 0);
+	atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
 	args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
 	btrfs_dev_replace_unlock(dev_replace, 1);
 

fs/btrfs/extent-tree.c

···
 	u64 dev_min = 1;
 	u64 dev_nr = 0;
 	u64 target;
+	int debug;
 	int index;
 	int full = 0;
 	int ret = 0;
 
+	debug = btrfs_test_opt(root, ENOSPC_DEBUG);
+
 	block_group = btrfs_lookup_block_group(root->fs_info, bytenr);
 
 	/* odd, couldn't find the block group, leave it alone */
-	if (!block_group)
+	if (!block_group) {
+		if (debug)
+			btrfs_warn(root->fs_info,
+				   "can't find block group for bytenr %llu",
+				   bytenr);
 		return -1;
+	}
 
 	min_free = btrfs_block_group_used(&block_group->item);
 
···
 		 * this is just a balance, so if we were marked as full
 		 * we know there is no space for a new chunk
 		 */
-		if (full)
+		if (full) {
+			if (debug)
+				btrfs_warn(root->fs_info,
+					"no space to alloc new chunk for block group %llu",
+					block_group->key.objectid);
 			goto out;
+		}
 
 		index = get_block_group_index(block_group);
 	}
···
 			ret = -1;
 		}
 	}
+	if (debug && ret == -1)
+		btrfs_warn(root->fs_info,
+			"no space to allocate a new chunk for block group %llu",
+			block_group->key.objectid);
 	mutex_unlock(&root->fs_info->chunk_mutex);
 	btrfs_end_transaction(trans, root);
 out:

fs/btrfs/file.c

···
 		return ret;
 
 	inode_lock(inode);
-	ret = inode_newsize_ok(inode, alloc_end);
-	if (ret)
-		goto out;
+
+	if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size) {
+		ret = inode_newsize_ok(inode, offset + len);
+		if (ret)
+			goto out;
+	}
 
 	/*
 	 * TODO: Move these two operations after we have checked

fs/btrfs/ioctl.c

···
 
 		src_inode = file_inode(src.file);
 		if (src_inode->i_sb != file_inode(file)->i_sb) {
-			btrfs_info(BTRFS_I(src_inode)->root->fs_info,
+			btrfs_info(BTRFS_I(file_inode(file))->root->fs_info,
 				   "Snapshot src from another FS");
 			ret = -EXDEV;
 		} else if (!inode_owner_or_capable(src_inode)) {

fs/btrfs/qgroup.c

···
 	u64 bytenr = record->bytenr;
 
 	assert_spin_locked(&delayed_refs->lock);
+	trace_btrfs_qgroup_insert_dirty_extent(record);
 
 	while (*p) {
 		parent_node = *p;
···
 		cur_old_count = btrfs_qgroup_get_old_refcnt(qg, seq);
 		cur_new_count = btrfs_qgroup_get_new_refcnt(qg, seq);
 
+		trace_qgroup_update_counters(qg->qgroupid, cur_old_count,
+					     cur_new_count);
+
 		/* Rfer update part */
 		if (cur_old_count == 0 && cur_new_count > 0) {
 			qg->rfer += num_bytes;
···
 		goto out_free;
 	BUG_ON(!fs_info->quota_root);
 
+	trace_btrfs_qgroup_account_extent(bytenr, num_bytes, nr_old_roots,
+					  nr_new_roots);
+
 	qgroups = ulist_alloc(GFP_NOFS);
 	if (!qgroups) {
 		ret = -ENOMEM;
···
 	while ((node = rb_first(&delayed_refs->dirty_extent_root))) {
 		record = rb_entry(node, struct btrfs_qgroup_extent_record,
 				  node);
+
+		trace_btrfs_qgroup_account_extents(record);
 
 		if (!ret) {
 			/*
···
 }
 
 /*
- * copy the acounting information between qgroups. This is necessary when a
- * snapshot or a subvolume is created
+ * Copy the acounting information between qgroups. This is necessary
+ * when a snapshot or a subvolume is created. Throwing an error will
+ * cause a transaction abort so we take extra care here to only error
+ * when a readonly fs is a reasonable outcome.
  */
 int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans,
 			 struct btrfs_fs_info *fs_info, u64 srcid, u64 objectid,
···
 		       2 * inherit->num_excl_copies;
 		for (i = 0; i < nums; ++i) {
 			srcgroup = find_qgroup_rb(fs_info, *i_qgroups);
-			if (!srcgroup) {
-				ret = -EINVAL;
-				goto out;
-			}
 
-			if ((srcgroup->qgroupid >> 48) <= (objectid >> 48)) {
-				ret = -EINVAL;
-				goto out;
-			}
+			/*
+			 * Zero out invalid groups so we can ignore
+			 * them later.
+			 */
+			if (!srcgroup ||
+			    ((srcgroup->qgroupid >> 48) <= (objectid >> 48)))
+				*i_qgroups = 0ULL;
+
 			++i_qgroups;
 		}
 	}
···
 	 */
 	if (inherit) {
 		i_qgroups = (u64 *)(inherit + 1);
-		for (i = 0; i < inherit->num_qgroups; ++i) {
+		for (i = 0; i < inherit->num_qgroups; ++i, ++i_qgroups) {
+			if (*i_qgroups == 0)
+				continue;
 			ret = add_qgroup_relation_item(trans, quota_root,
 						       objectid, *i_qgroups);
-			if (ret)
+			if (ret && ret != -EEXIST)
 				goto out;
 			ret = add_qgroup_relation_item(trans, quota_root,
 						       *i_qgroups, objectid);
-			if (ret)
+			if (ret && ret != -EEXIST)
 				goto out;
-			++i_qgroups;
 		}
+		ret = 0;
 	}
 
 
···
 
 	i_qgroups = (u64 *)(inherit + 1);
 	for (i = 0; i < inherit->num_qgroups; ++i) {
-		ret = add_relation_rb(quota_root->fs_info, objectid,
-				      *i_qgroups);
-		if (ret)
-			goto unlock;
+		if (*i_qgroups) {
+			ret = add_relation_rb(quota_root->fs_info, objectid,
+					      *i_qgroups);
+			if (ret)
+				goto unlock;
+		}
 		++i_qgroups;
 	}
 
-	for (i = 0; i <  inherit->num_ref_copies; ++i) {
+	for (i = 0; i <  inherit->num_ref_copies; ++i, i_qgroups += 2) {
 		struct btrfs_qgroup *src;
 		struct btrfs_qgroup *dst;
+
+		if (!i_qgroups[0] || !i_qgroups[1])
+			continue;
 
 		src = find_qgroup_rb(fs_info, i_qgroups[0]);
 		dst = find_qgroup_rb(fs_info, i_qgroups[1]);
···
 
 		dst->rfer = src->rfer - level_size;
 		dst->rfer_cmpr = src->rfer_cmpr - level_size;
-		i_qgroups += 2;
 	}
-	for (i = 0; i <  inherit->num_excl_copies; ++i) {
+	for (i = 0; i <  inherit->num_excl_copies; ++i, i_qgroups += 2) {
 		struct btrfs_qgroup *src;
 		struct btrfs_qgroup *dst;
+
+		if (!i_qgroups[0] || !i_qgroups[1])
+			continue;
 
 		src = find_qgroup_rb(fs_info, i_qgroups[0]);
 		dst = find_qgroup_rb(fs_info, i_qgroups[1]);
···
 
 		dst->excl = src->excl + level_size;
 		dst->excl_cmpr = src->excl_cmpr + level_size;
-		i_qgroups += 2;
 	}
 
 unlock:

fs/btrfs/relocation.c

···
 			eb = read_tree_block(dest, old_bytenr, old_ptr_gen);
 			if (IS_ERR(eb)) {
 				ret = PTR_ERR(eb);
+				break;
 			} else if (!extent_buffer_uptodate(eb)) {
 				ret = -EIO;
 				free_extent_buffer(eb);

fs/btrfs/tree-log.c

···
 	return ret;
 }
 
+/*
+ * When we are logging a new inode X, check if it doesn't have a reference that
+ * matches the reference from some other inode Y created in a past transaction
+ * and that was renamed in the current transaction. If we don't do this, then at
+ * log replay time we can lose inode Y (and all its files if it's a directory):
+ *
+ * mkdir /mnt/x
+ * echo "hello world" > /mnt/x/foobar
+ * sync
+ * mv /mnt/x /mnt/y
+ * mkdir /mnt/x                 # or touch /mnt/x
+ * xfs_io -c fsync /mnt/x
+ * <power fail>
+ * mount fs, trigger log replay
+ *
+ * After the log replay procedure, we would lose the first directory and all its
+ * files (file foobar).
+ * For the case where inode Y is not a directory we simply end up losing it:
+ *
+ * echo "123" > /mnt/foo
+ * sync
+ * mv /mnt/foo /mnt/bar
+ * echo "abc" > /mnt/foo
+ * xfs_io -c fsync /mnt/foo
+ * <power fail>
+ *
+ * We also need this for cases where a snapshot entry is replaced by some other
+ * entry (file or directory) otherwise we end up with an unreplayable log due to
+ * attempts to delete the snapshot entry (entry of type BTRFS_ROOT_ITEM_KEY) as
+ * if it were a regular entry:
+ *
+ * mkdir /mnt/x
+ * btrfs subvolume snapshot /mnt /mnt/x/snap
+ * btrfs subvolume delete /mnt/x/snap
+ * rmdir /mnt/x
+ * mkdir /mnt/x
+ * fsync /mnt/x or fsync some new file inside it
+ * <power fail>
+ *
+ * The snapshot delete, rmdir of x, mkdir of a new x and the fsync all happen in
+ * the same transaction.
+ */
+static int btrfs_check_ref_name_override(struct extent_buffer *eb,
+					 const int slot,
+					 const struct btrfs_key *key,
+					 struct inode *inode)
+{
+	int ret;
+	struct btrfs_path *search_path;
+	char *name = NULL;
+	u32 name_len = 0;
+	u32 item_size = btrfs_item_size_nr(eb, slot);
+	u32 cur_offset = 0;
+	unsigned long ptr = btrfs_item_ptr_offset(eb, slot);
+
+	search_path = btrfs_alloc_path();
+	if (!search_path)
+		return -ENOMEM;
+	search_path->search_commit_root = 1;
+	search_path->skip_locking = 1;
+
+	while (cur_offset < item_size) {
+		u64 parent;
+		u32 this_name_len;
+		u32 this_len;
+		unsigned long name_ptr;
+		struct btrfs_dir_item *di;
+
+		if (key->type == BTRFS_INODE_REF_KEY) {
+			struct btrfs_inode_ref *iref;
+
+			iref = (struct btrfs_inode_ref *)(ptr + cur_offset);
+			parent = key->offset;
+			this_name_len = btrfs_inode_ref_name_len(eb, iref);
+			name_ptr = (unsigned long)(iref + 1);
+			this_len = sizeof(*iref) + this_name_len;
+		} else {
+			struct btrfs_inode_extref *extref;
+
+			extref = (struct btrfs_inode_extref *)(ptr +
+							       cur_offset);
+			parent = btrfs_inode_extref_parent(eb, extref);
+			this_name_len = btrfs_inode_extref_name_len(eb, extref);
+			name_ptr = (unsigned long)&extref->name;
+			this_len = sizeof(*extref) + this_name_len;
+		}
+
+		if (this_name_len > name_len) {
+			char *new_name;
+
+			new_name = krealloc(name, this_name_len, GFP_NOFS);
+			if (!new_name) {
+				ret = -ENOMEM;
+				goto out;
+			}
+			name_len = this_name_len;
+			name = new_name;
+		}
+
+		read_extent_buffer(eb, name, name_ptr, this_name_len);
+		di = btrfs_lookup_dir_item(NULL, BTRFS_I(inode)->root,
+					   search_path, parent,
+					   name, this_name_len, 0);
+		if (di && !IS_ERR(di)) {
+			ret = 1;
+			goto out;
+		} else if (IS_ERR(di)) {
+			ret = PTR_ERR(di);
+			goto out;
+		}
+		btrfs_release_path(search_path);
+
+		cur_offset += this_len;
+	}
+	ret = 0;
+out:
+	btrfs_free_path(search_path);
+	kfree(name);
+	return ret;
+}
+
 /* log a single inode in the tree log.
  * At least one parent directory for this inode must exist in the tree
  * or be logged already.
···
 
 		if (min_key.type == BTRFS_INODE_ITEM_KEY)
 			need_log_inode_item = false;
+
+		if ((min_key.type == BTRFS_INODE_REF_KEY ||
+		     min_key.type == BTRFS_INODE_EXTREF_KEY) &&
+		    BTRFS_I(inode)->generation == trans->transid) {
+			ret = btrfs_check_ref_name_override(path->nodes[0],
+							    path->slots[0],
+							    &min_key, inode);
+			if (ret < 0) {
+				err = ret;
+				goto out_unlock;
+			} else if (ret > 0) {
+				err = 1;
+				btrfs_set_log_full_commit(root->fs_info, trans);
+				goto out_unlock;
+			}
+		}
 
 		/* Skip xattrs, we log them later with btrfs_log_all_xattrs() */
 		if (min_key.type == BTRFS_XATTR_ITEM_KEY) {

include/trace/events/btrfs.h

···
 struct extent_buffer;
 struct btrfs_work;
 struct __btrfs_workqueue;
-struct btrfs_qgroup_operation;
+struct btrfs_qgroup_extent_record;
 
 #define show_ref_type(type)						\
 	__print_symbolic(type,						\
···
 
 	TP_ARGS(ref_root, reserved)
 );
+
+DECLARE_EVENT_CLASS(btrfs_qgroup_extent,
+	TP_PROTO(struct btrfs_qgroup_extent_record *rec),
+
+	TP_ARGS(rec),
+
+	TP_STRUCT__entry(
+		__field(	u64,  bytenr		)
+		__field(	u64,  num_bytes		)
+	),
+
+	TP_fast_assign(
+		__entry->bytenr		= rec->bytenr,
+		__entry->num_bytes	= rec->num_bytes;
+	),
+
+	TP_printk("bytenr = %llu, num_bytes = %llu",
+		  (unsigned long long)__entry->bytenr,
+		  (unsigned long long)__entry->num_bytes)
+);
+
+DEFINE_EVENT(btrfs_qgroup_extent, btrfs_qgroup_account_extents,
+
+	TP_PROTO(struct btrfs_qgroup_extent_record *rec),
+
+	TP_ARGS(rec)
+);
+
+DEFINE_EVENT(btrfs_qgroup_extent, btrfs_qgroup_insert_dirty_extent,
+
+	TP_PROTO(struct btrfs_qgroup_extent_record *rec),
+
+	TP_ARGS(rec)
+);
+
+TRACE_EVENT(btrfs_qgroup_account_extent,
+
+	TP_PROTO(u64 bytenr, u64 num_bytes, u64 nr_old_roots, u64 nr_new_roots),
+
+	TP_ARGS(bytenr, num_bytes, nr_old_roots, nr_new_roots),
+
+	TP_STRUCT__entry(
+		__field(	u64,  bytenr			)
+		__field(	u64,  num_bytes			)
+		__field(	u64,  nr_old_roots		)
+		__field(	u64,  nr_new_roots		)
+	),
+
+	TP_fast_assign(
+		__entry->bytenr		= bytenr;
+		__entry->num_bytes	= num_bytes;
+		__entry->nr_old_roots	= nr_old_roots;
+		__entry->nr_new_roots	= nr_new_roots;
+	),
+
+	TP_printk("bytenr = %llu, num_bytes = %llu, nr_old_roots = %llu, "
+		  "nr_new_roots = %llu",
+		  __entry->bytenr,
+		  __entry->num_bytes,
+		  __entry->nr_old_roots,
+		  __entry->nr_new_roots)
+);
+
+TRACE_EVENT(qgroup_update_counters,
+
+	TP_PROTO(u64 qgid, u64 cur_old_count, u64 cur_new_count),
+
+	TP_ARGS(qgid, cur_old_count, cur_new_count),
+
+	TP_STRUCT__entry(
+		__field(	u64,  qgid			)
+		__field(	u64,  cur_old_count		)
+		__field(	u64,  cur_new_count		)
+	),
+
+	TP_fast_assign(
+		__entry->qgid		= qgid;
+		__entry->cur_old_count	= cur_old_count;
+		__entry->cur_new_count	= cur_new_count;
+	),
+
+	TP_printk("qgid = %llu, cur_old_count = %llu, cur_new_count = %llu",
+		  __entry->qgid,
+		  __entry->cur_old_count,
+		  __entry->cur_new_count)
+);
+
 #endif /* _TRACE_BTRFS_H */
 
 /* This part must be outside protection */