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

Pull btrfs updates from Chris Mason:
 "This has our merge window series of cleanups and fixes.  These target
  a wide range of issues, but do include some important fixes for
  qgroups, O_DIRECT, and fsync handling.  Jeff Mahoney moved around a
  few definitions to make them easier for userland to consume.

  Also whiteout support is included now that issues with overlayfs have
  been cleared up.

  I have one more fix pending for page faults during btrfs_copy_from_user,
  but I wanted to get this bulk out the door first"

* 'for-linus-4.7' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (90 commits)
  btrfs: fix memory leak during RAID 5/6 device replacement
  Btrfs: add semaphore to synchronize direct IO writes with fsync
  Btrfs: fix race between block group relocation and nocow writes
  Btrfs: fix race between fsync and direct IO writes for prealloc extents
  Btrfs: fix number of transaction units for renames with whiteout
  Btrfs: pin logs earlier when doing a rename exchange operation
  Btrfs: unpin logs if rename exchange operation fails
  Btrfs: fix inode leak on failure to setup whiteout inode in rename
  btrfs: add support for RENAME_EXCHANGE and RENAME_WHITEOUT
  Btrfs: pin log earlier when renaming
  Btrfs: unpin log if rename operation fails
  Btrfs: don't do unnecessary delalloc flushes when relocating
  Btrfs: don't wait for unrelated IO to finish before relocation
  Btrfs: fix empty symlink after creating symlink and fsync parent dir
  Btrfs: fix for incorrect directory entries after fsync log replay
  btrfs: build fixup for qgroup_account_snapshot
  btrfs: qgroup: Fix qgroup accounting when creating snapshot
  Btrfs: fix fspath error deallocation
  btrfs: make find_workspace warn if there are no workspaces
  btrfs: make find_workspace always succeed
  ...

fs/btrfs/backref.c

@@ -1991,7 +1991,7 @@ struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root,
 
 	ifp = kmalloc(sizeof(*ifp), GFP_NOFS);
 	if (!ifp) {
-		kfree(fspath);
+		vfree(fspath);
 		return ERR_PTR(-ENOMEM);
 	}
 

fs/btrfs/btrfs_inode.h

@@ -196,6 +196,16 @@ struct btrfs_inode {
 	struct list_head delayed_iput;
 	long delayed_iput_count;
 
+	/*
+	 * To avoid races between lockless (i_mutex not held) direct IO writes
+	 * and concurrent fsync requests. Direct IO writes must acquire read
+	 * access on this semaphore for creating an extent map and its
+	 * corresponding ordered extent. The fast fsync path must acquire write
+	 * access on this semaphore before it collects ordered extents and
+	 * extent maps.
+	 */
+	struct rw_semaphore dio_sem;
+
 	struct inode vfs_inode;
 };
 

fs/btrfs/compression.c

@@ -743,8 +743,11 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 static struct {
 	struct list_head idle_ws;
 	spinlock_t ws_lock;
-	int num_ws;
-	atomic_t alloc_ws;
+	/* Number of free workspaces */
+	int free_ws;
+	/* Total number of allocated workspaces */
+	atomic_t total_ws;
+	/* Waiters for a free workspace */
 	wait_queue_head_t ws_wait;
 } btrfs_comp_ws[BTRFS_COMPRESS_TYPES];
 
@@ -758,16 +761,34 @@ void __init btrfs_init_compress(void)
 	int i;
 
 	for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
+		struct list_head *workspace;
+
 		INIT_LIST_HEAD(&btrfs_comp_ws[i].idle_ws);
 		spin_lock_init(&btrfs_comp_ws[i].ws_lock);
-		atomic_set(&btrfs_comp_ws[i].alloc_ws, 0);
+		atomic_set(&btrfs_comp_ws[i].total_ws, 0);
 		init_waitqueue_head(&btrfs_comp_ws[i].ws_wait);
+
+		/*
+		 * Preallocate one workspace for each compression type so
+		 * we can guarantee forward progress in the worst case
+		 */
+		workspace = btrfs_compress_op[i]->alloc_workspace();
+		if (IS_ERR(workspace)) {
+			printk(KERN_WARNING
+	"BTRFS: cannot preallocate compression workspace, will try later");
+		} else {
+			atomic_set(&btrfs_comp_ws[i].total_ws, 1);
+			btrfs_comp_ws[i].free_ws = 1;
+			list_add(workspace, &btrfs_comp_ws[i].idle_ws);
+		}
 	}
 }
 
 /*
- * this finds an available workspace or allocates a new one
- * ERR_PTR is returned if things go bad.
+ * This finds an available workspace or allocates a new one.
+ * If it's not possible to allocate a new one, waits until there's one.
+ * Preallocation makes a forward progress guarantees and we do not return
+ * errors.
  */
 static struct list_head *find_workspace(int type)
 {
@@ -777,36 +798,58 @@ static struct list_head *find_workspace(int type)
 
 	struct list_head *idle_ws	= &btrfs_comp_ws[idx].idle_ws;
 	spinlock_t *ws_lock		= &btrfs_comp_ws[idx].ws_lock;
-	atomic_t *alloc_ws		= &btrfs_comp_ws[idx].alloc_ws;
+	atomic_t *total_ws		= &btrfs_comp_ws[idx].total_ws;
 	wait_queue_head_t *ws_wait	= &btrfs_comp_ws[idx].ws_wait;
-	int *num_ws			= &btrfs_comp_ws[idx].num_ws;
+	int *free_ws			= &btrfs_comp_ws[idx].free_ws;
 again:
 	spin_lock(ws_lock);
 	if (!list_empty(idle_ws)) {
 		workspace = idle_ws->next;
 		list_del(workspace);
-		(*num_ws)--;
+		(*free_ws)--;
 		spin_unlock(ws_lock);
 		return workspace;
 
 	}
-	if (atomic_read(alloc_ws) > cpus) {
+	if (atomic_read(total_ws) > cpus) {
 		DEFINE_WAIT(wait);
 
 		spin_unlock(ws_lock);
 		prepare_to_wait(ws_wait, &wait, TASK_UNINTERRUPTIBLE);
-		if (atomic_read(alloc_ws) > cpus && !*num_ws)
+		if (atomic_read(total_ws) > cpus && !*free_ws)
 			schedule();
 		finish_wait(ws_wait, &wait);
 		goto again;
 	}
-	atomic_inc(alloc_ws);
+	atomic_inc(total_ws);
 	spin_unlock(ws_lock);
 
 	workspace = btrfs_compress_op[idx]->alloc_workspace();
 	if (IS_ERR(workspace)) {
-		atomic_dec(alloc_ws);
+		atomic_dec(total_ws);
 		wake_up(ws_wait);
+
+		/*
+		 * Do not return the error but go back to waiting. There's a
+		 * workspace preallocated for each type and the compression
+		 * time is bounded so we get to a workspace eventually. This
+		 * makes our caller's life easier.
+		 *
+		 * To prevent silent and low-probability deadlocks (when the
+		 * initial preallocation fails), check if there are any
+		 * workspaces at all.
+		 */
+		if (atomic_read(total_ws) == 0) {
+			static DEFINE_RATELIMIT_STATE(_rs,
+					/* once per minute */ 60 * HZ,
+					/* no burst */ 1);
+
+			if (__ratelimit(&_rs)) {
+				printk(KERN_WARNING
+			    "no compression workspaces, low memory, retrying");
+			}
+		}
+		goto again;
 	}
 	return workspace;
 }
@@ -820,21 +863,21 @@ static void free_workspace(int type, struct list_head *workspace)
 	int idx = type - 1;
 	struct list_head *idle_ws	= &btrfs_comp_ws[idx].idle_ws;
 	spinlock_t *ws_lock		= &btrfs_comp_ws[idx].ws_lock;
-	atomic_t *alloc_ws		= &btrfs_comp_ws[idx].alloc_ws;
+	atomic_t *total_ws		= &btrfs_comp_ws[idx].total_ws;
 	wait_queue_head_t *ws_wait	= &btrfs_comp_ws[idx].ws_wait;
-	int *num_ws			= &btrfs_comp_ws[idx].num_ws;
+	int *free_ws			= &btrfs_comp_ws[idx].free_ws;
 
 	spin_lock(ws_lock);
-	if (*num_ws < num_online_cpus()) {
+	if (*free_ws < num_online_cpus()) {
 		list_add(workspace, idle_ws);
-		(*num_ws)++;
+		(*free_ws)++;
 		spin_unlock(ws_lock);
 		goto wake;
 	}
 	spin_unlock(ws_lock);
 
 	btrfs_compress_op[idx]->free_workspace(workspace);
-	atomic_dec(alloc_ws);
+	atomic_dec(total_ws);
 wake:
 	/*
 	 * Make sure counter is updated before we wake up waiters.
@@ -857,7 +900,7 @@ static void free_workspaces(void)
 			workspace = btrfs_comp_ws[i].idle_ws.next;
 			list_del(workspace);
 			btrfs_compress_op[i]->free_workspace(workspace);
-			atomic_dec(&btrfs_comp_ws[i].alloc_ws);
+			atomic_dec(&btrfs_comp_ws[i].total_ws);
 		}
 	}
 }
@@ -894,8 +937,6 @@ int btrfs_compress_pages(int type, struct address_space *mapping,
 	int ret;
 
 	workspace = find_workspace(type);
-	if (IS_ERR(workspace))
-		return PTR_ERR(workspace);
 
 	ret = btrfs_compress_op[type-1]->compress_pages(workspace, mapping,
 						      start, len, pages,
@@ -930,8 +971,6 @@ static int btrfs_decompress_biovec(int type, struct page **pages_in,
 	int ret;
 
 	workspace = find_workspace(type);
-	if (IS_ERR(workspace))
-		return PTR_ERR(workspace);
 
 	ret = btrfs_compress_op[type-1]->decompress_biovec(workspace, pages_in,
 							 disk_start,
@@ -952,8 +991,6 @@ int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
 	int ret;
 
 	workspace = find_workspace(type);
-	if (IS_ERR(workspace))
-		return PTR_ERR(workspace);
 
 	ret = btrfs_compress_op[type-1]->decompress(workspace, data_in,
 						  dest_page, start_byte,

fs/btrfs/ctree.c

@@ -1011,7 +1011,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 			return ret;
 		if (refs == 0) {
 			ret = -EROFS;
-			btrfs_std_error(root->fs_info, ret, NULL);
+			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			return ret;
 		}
 	} else {
@@ -1928,7 +1928,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		child = read_node_slot(root, mid, 0);
 		if (!child) {
 			ret = -EROFS;
-			btrfs_std_error(root->fs_info, ret, NULL);
+			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			goto enospc;
 		}
 
@@ -2031,7 +2031,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		 */
 		if (!left) {
 			ret = -EROFS;
-			btrfs_std_error(root->fs_info, ret, NULL);
+			btrfs_handle_fs_error(root->fs_info, ret, NULL);
 			goto enospc;
 		}
 		wret = balance_node_right(trans, root, mid, left);

fs/btrfs/delayed-inode.c

@@ -134,7 +134,7 @@ static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node(
 	/* cached in the btrfs inode and can be accessed */
 	atomic_add(2, &node->refs);
 
-	ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
+	ret = radix_tree_preload(GFP_NOFS);
 	if (ret) {
 		kmem_cache_free(delayed_node_cache, node);
 		return ERR_PTR(ret);

fs/btrfs/dev-replace.c

@@ -44,9 +44,6 @@ static void btrfs_dev_replace_update_device_in_mapping_tree(
 						struct btrfs_fs_info *fs_info,
 						struct btrfs_device *srcdev,
 						struct btrfs_device *tgtdev);
-static int btrfs_dev_replace_find_srcdev(struct btrfs_root *root, u64 srcdevid,
-					 char *srcdev_name,
-					 struct btrfs_device **device);
 static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info);
 static int btrfs_dev_replace_kthread(void *data);
 static int btrfs_dev_replace_continue_on_mount(struct btrfs_fs_info *fs_info);
@@ -305,8 +302,8 @@ void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info)
 		dev_replace->cursor_left_last_write_of_item;
 }
 
-int btrfs_dev_replace_start(struct btrfs_root *root,
-			    struct btrfs_ioctl_dev_replace_args *args)
+int btrfs_dev_replace_start(struct btrfs_root *root, char *tgtdev_name,
+				u64 srcdevid, char *srcdev_name, int read_src)
 {
 	struct btrfs_trans_handle *trans;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -315,29 +312,16 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 	struct btrfs_device *tgt_device = NULL;
 	struct btrfs_device *src_device = NULL;
 
-	switch (args->start.cont_reading_from_srcdev_mode) {
-	case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS:
-	case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID:
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	if ((args->start.srcdevid == 0 && args->start.srcdev_name[0] == '\0') ||
-	    args->start.tgtdev_name[0] == '\0')
-		return -EINVAL;
-
 	/* the disk copy procedure reuses the scrub code */
 	mutex_lock(&fs_info->volume_mutex);
-	ret = btrfs_dev_replace_find_srcdev(root, args->start.srcdevid,
-					    args->start.srcdev_name,
-					    &src_device);
+	ret = btrfs_find_device_by_devspec(root, srcdevid,
+					    srcdev_name, &src_device);
 	if (ret) {
 		mutex_unlock(&fs_info->volume_mutex);
 		return ret;
 	}
 
-	ret = btrfs_init_dev_replace_tgtdev(root, args->start.tgtdev_name,
+	ret = btrfs_init_dev_replace_tgtdev(root, tgtdev_name,
 					    src_device, &tgt_device);
 	mutex_unlock(&fs_info->volume_mutex);
 	if (ret)
@@ -364,18 +348,17 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 		break;
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
-		args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
+		ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
 		goto leave;
 	}
 
-	dev_replace->cont_reading_from_srcdev_mode =
-		args->start.cont_reading_from_srcdev_mode;
+	dev_replace->cont_reading_from_srcdev_mode = read_src;
 	WARN_ON(!src_device);
 	dev_replace->srcdev = src_device;
 	WARN_ON(!tgt_device);
 	dev_replace->tgtdev = tgt_device;
 
-	btrfs_info_in_rcu(root->fs_info,
+	btrfs_info_in_rcu(fs_info,
 		      "dev_replace from %s (devid %llu) to %s started",
 		      src_device->missing ? "<missing disk>" :
 		        rcu_str_deref(src_device->name),
@@ -396,14 +379,13 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 	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);
 
 	ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
 	if (ret)
-		btrfs_err(root->fs_info, "kobj add dev failed %d\n", ret);
+		btrfs_err(fs_info, "kobj add dev failed %d\n", ret);
 
-	btrfs_wait_ordered_roots(root->fs_info, -1);
+	btrfs_wait_ordered_roots(root->fs_info, -1, 0, (u64)-1);
 
 	/* force writing the updated state information to disk */
 	trans = btrfs_start_transaction(root, 0);
@@ -421,11 +403,9 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 			      btrfs_device_get_total_bytes(src_device),
 			      &dev_replace->scrub_progress, 0, 1);
 
-	ret = btrfs_dev_replace_finishing(root->fs_info, ret);
-	/* don't warn if EINPROGRESS, someone else might be running scrub */
+	ret = btrfs_dev_replace_finishing(fs_info, ret);
 	if (ret == -EINPROGRESS) {
-		args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
-		ret = 0;
+		ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
 	} else {
 		WARN_ON(ret);
 	}
@@ -440,6 +420,35 @@ int btrfs_dev_replace_start(struct btrfs_root *root,
 	return ret;
 }
 
+int btrfs_dev_replace_by_ioctl(struct btrfs_root *root,
+			    struct btrfs_ioctl_dev_replace_args *args)
+{
+	int ret;
+
+	switch (args->start.cont_reading_from_srcdev_mode) {
+	case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS:
+	case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if ((args->start.srcdevid == 0 && args->start.srcdev_name[0] == '\0') ||
+	    args->start.tgtdev_name[0] == '\0')
+		return -EINVAL;
+
+	ret = btrfs_dev_replace_start(root, args->start.tgtdev_name,
+					args->start.srcdevid,
+					args->start.srcdev_name,
+					args->start.cont_reading_from_srcdev_mode);
+	args->result = ret;
+	/* don't warn if EINPROGRESS, someone else might be running scrub */
+	if (ret == BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS)
+		ret = 0;
+
+	return ret;
+}
+
 /*
  * blocked until all flighting bios are finished.
  */
@@ -495,7 +504,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 		mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
 		return ret;
 	}
-	btrfs_wait_ordered_roots(root->fs_info, -1);
+	btrfs_wait_ordered_roots(root->fs_info, -1, 0, (u64)-1);
 
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
@@ -560,10 +569,9 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 	ASSERT(list_empty(&src_device->resized_list));
 	tgt_device->commit_total_bytes = src_device->commit_total_bytes;
 	tgt_device->commit_bytes_used = src_device->bytes_used;
-	if (fs_info->sb->s_bdev == src_device->bdev)
-		fs_info->sb->s_bdev = tgt_device->bdev;
-	if (fs_info->fs_devices->latest_bdev == src_device->bdev)
-		fs_info->fs_devices->latest_bdev = tgt_device->bdev;
+
+	btrfs_assign_next_active_device(fs_info, src_device, tgt_device);
+
 	list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
 	fs_info->fs_devices->rw_devices++;
 
@@ -626,25 +634,6 @@ static void btrfs_dev_replace_update_device_in_mapping_tree(
 	write_unlock(&em_tree->lock);
 }
 
-static int btrfs_dev_replace_find_srcdev(struct btrfs_root *root, u64 srcdevid,
-					 char *srcdev_name,
-					 struct btrfs_device **device)
-{
-	int ret;
-
-	if (srcdevid) {
-		ret = 0;
-		*device = btrfs_find_device(root->fs_info, srcdevid, NULL,
-					    NULL);
-		if (!*device)
-			ret = -ENOENT;
-	} else {
-		ret = btrfs_find_device_missing_or_by_path(root, srcdev_name,
-							   device);
-	}
-	return ret;
-}
-
 void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
 			      struct btrfs_ioctl_dev_replace_args *args)
 {

fs/btrfs/dev-replace.h

@@ -25,8 +25,10 @@ int btrfs_init_dev_replace(struct btrfs_fs_info *fs_info);
 int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
 			  struct btrfs_fs_info *fs_info);
 void btrfs_after_dev_replace_commit(struct btrfs_fs_info *fs_info);
-int btrfs_dev_replace_start(struct btrfs_root *root,
+int btrfs_dev_replace_by_ioctl(struct btrfs_root *root,
 			    struct btrfs_ioctl_dev_replace_args *args);
+int btrfs_dev_replace_start(struct btrfs_root *root, char *tgtdev_name,
+				u64 srcdevid, char *srcdev_name, int read_src);
 void btrfs_dev_replace_status(struct btrfs_fs_info *fs_info,
 			      struct btrfs_ioctl_dev_replace_args *args);
 int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info,

fs/btrfs/disk-io.c

@@ -1640,7 +1640,7 @@ int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 {
 	int ret;
 
-	ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
+	ret = radix_tree_preload(GFP_NOFS);
 	if (ret)
 		return ret;
 
@@ -2417,7 +2417,7 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info,
 	/* returns with log_tree_root freed on success */
 	ret = btrfs_recover_log_trees(log_tree_root);
 	if (ret) {
-		btrfs_std_error(tree_root->fs_info, ret,
+		btrfs_handle_fs_error(tree_root->fs_info, ret,
 			    "Failed to recover log tree");
 		free_extent_buffer(log_tree_root->node);
 		kfree(log_tree_root);
@@ -2517,6 +2517,7 @@ 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);
@@ -2713,7 +2714,7 @@ int open_ctree(struct super_block *sb,
 	 * Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k).
 	 */
 	if (btrfs_check_super_csum(bh->b_data)) {
-		printk(KERN_ERR "BTRFS: superblock checksum mismatch\n");
+		btrfs_err(fs_info, "superblock checksum mismatch");
 		err = -EINVAL;
 		brelse(bh);
 		goto fail_alloc;
@@ -2733,7 +2734,7 @@ int open_ctree(struct super_block *sb,
 
 	ret = btrfs_check_super_valid(fs_info, sb->s_flags & MS_RDONLY);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: superblock contains fatal errors\n");
+		btrfs_err(fs_info, "superblock contains fatal errors");
 		err = -EINVAL;
 		goto fail_alloc;
 	}
@@ -2768,9 +2769,9 @@ int open_ctree(struct super_block *sb,
 	features = btrfs_super_incompat_flags(disk_super) &
 		~BTRFS_FEATURE_INCOMPAT_SUPP;
 	if (features) {
-		printk(KERN_ERR "BTRFS: couldn't mount because of "
-		       "unsupported optional features (%Lx).\n",
-		       features);
+		btrfs_err(fs_info,
+		    "cannot mount because of unsupported optional features (%llx)",
+		    features);
 		err = -EINVAL;
 		goto fail_alloc;
 	}
@@ -2781,7 +2782,7 @@ int open_ctree(struct super_block *sb,
 		features |= BTRFS_FEATURE_INCOMPAT_COMPRESS_LZO;
 
 	if (features & BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA)
-		printk(KERN_INFO "BTRFS: has skinny extents\n");
+		btrfs_info(fs_info, "has skinny extents");
 
 	/*
 	 * flag our filesystem as having big metadata blocks if
@@ -2789,7 +2790,8 @@ int open_ctree(struct super_block *sb,
 	 */
 	if (btrfs_super_nodesize(disk_super) > PAGE_SIZE) {
 		if (!(features & BTRFS_FEATURE_INCOMPAT_BIG_METADATA))
-			printk(KERN_INFO "BTRFS: flagging fs with big metadata feature\n");
+			btrfs_info(fs_info,
+				"flagging fs with big metadata feature");
 		features |= BTRFS_FEATURE_INCOMPAT_BIG_METADATA;
 	}
 
@@ -2805,9 +2807,9 @@ int open_ctree(struct super_block *sb,
 	 */
 	if ((features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS) &&
 	    (sectorsize != nodesize)) {
-		printk(KERN_ERR "BTRFS: unequal leaf/node/sector sizes "
-				"are not allowed for mixed block groups on %s\n",
-				sb->s_id);
+		btrfs_err(fs_info,
+"unequal nodesize/sectorsize (%u != %u) are not allowed for mixed block groups",
+			nodesize, sectorsize);
 		goto fail_alloc;
 	}
 
@@ -2820,8 +2822,8 @@ int open_ctree(struct super_block *sb,
 	features = btrfs_super_compat_ro_flags(disk_super) &
 		~BTRFS_FEATURE_COMPAT_RO_SUPP;
 	if (!(sb->s_flags & MS_RDONLY) && features) {
-		printk(KERN_ERR "BTRFS: couldn't mount RDWR because of "
-		       "unsupported option features (%Lx).\n",
+		btrfs_err(fs_info,
+	"cannot mount read-write because of unsupported optional features (%llx)",
 		       features);
 		err = -EINVAL;
 		goto fail_alloc;
@@ -2850,8 +2852,7 @@ int open_ctree(struct super_block *sb,
 	ret = btrfs_read_sys_array(tree_root);
 	mutex_unlock(&fs_info->chunk_mutex);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: failed to read the system "
-		       "array on %s\n", sb->s_id);
+		btrfs_err(fs_info, "failed to read the system array: %d", ret);
 		goto fail_sb_buffer;
 	}
 
@@ -2865,8 +2866,7 @@ int open_ctree(struct super_block *sb,
 					   generation);
 	if (IS_ERR(chunk_root->node) ||
 	    !extent_buffer_uptodate(chunk_root->node)) {
-		printk(KERN_ERR "BTRFS: failed to read chunk root on %s\n",
-		       sb->s_id);
+		btrfs_err(fs_info, "failed to read chunk root");
 		if (!IS_ERR(chunk_root->node))
 			free_extent_buffer(chunk_root->node);
 		chunk_root->node = NULL;
@@ -2880,8 +2880,7 @@ int open_ctree(struct super_block *sb,
 
 	ret = btrfs_read_chunk_tree(chunk_root);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: failed to read chunk tree on %s\n",
-		       sb->s_id);
+		btrfs_err(fs_info, "failed to read chunk tree: %d", ret);
 		goto fail_tree_roots;
 	}
 
@@ -2892,8 +2891,7 @@ int open_ctree(struct super_block *sb,
 	btrfs_close_extra_devices(fs_devices, 0);
 
 	if (!fs_devices->latest_bdev) {
-		printk(KERN_ERR "BTRFS: failed to read devices on %s\n",
-		       sb->s_id);
+		btrfs_err(fs_info, "failed to read devices");
 		goto fail_tree_roots;
 	}
 
@@ -2905,8 +2903,7 @@ int open_ctree(struct super_block *sb,
 					  generation);
 	if (IS_ERR(tree_root->node) ||
 	    !extent_buffer_uptodate(tree_root->node)) {
-		printk(KERN_WARNING "BTRFS: failed to read tree root on %s\n",
-		       sb->s_id);
+		btrfs_warn(fs_info, "failed to read tree root");
 		if (!IS_ERR(tree_root->node))
 			free_extent_buffer(tree_root->node);
 		tree_root->node = NULL;
@@ -2938,20 +2935,19 @@ int open_ctree(struct super_block *sb,
 
 	ret = btrfs_recover_balance(fs_info);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: failed to recover balance\n");
+		btrfs_err(fs_info, "failed to recover balance: %d", ret);
 		goto fail_block_groups;
 	}
 
 	ret = btrfs_init_dev_stats(fs_info);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: failed to init dev_stats: %d\n",
-		       ret);
+		btrfs_err(fs_info, "failed to init dev_stats: %d", ret);
 		goto fail_block_groups;
 	}
 
 	ret = btrfs_init_dev_replace(fs_info);
 	if (ret) {
-		pr_err("BTRFS: failed to init dev_replace: %d\n", ret);
+		btrfs_err(fs_info, "failed to init dev_replace: %d", ret);
 		goto fail_block_groups;
 	}
 
@@ -2959,31 +2955,33 @@ int open_ctree(struct super_block *sb,
 
 	ret = btrfs_sysfs_add_fsid(fs_devices, NULL);
 	if (ret) {
-		pr_err("BTRFS: failed to init sysfs fsid interface: %d\n", ret);
+		btrfs_err(fs_info, "failed to init sysfs fsid interface: %d",
+				ret);
 		goto fail_block_groups;
 	}
 
 	ret = btrfs_sysfs_add_device(fs_devices);
 	if (ret) {
-		pr_err("BTRFS: failed to init sysfs device interface: %d\n", ret);
+		btrfs_err(fs_info, "failed to init sysfs device interface: %d",
+				ret);
 		goto fail_fsdev_sysfs;
 	}
 
 	ret = btrfs_sysfs_add_mounted(fs_info);
 	if (ret) {
-		pr_err("BTRFS: failed to init sysfs interface: %d\n", ret);
+		btrfs_err(fs_info, "failed to init sysfs interface: %d", ret);
 		goto fail_fsdev_sysfs;
 	}
 
 	ret = btrfs_init_space_info(fs_info);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: Failed to initial space info: %d\n", ret);
+		btrfs_err(fs_info, "failed to initialize space info: %d", ret);
 		goto fail_sysfs;
 	}
 
 	ret = btrfs_read_block_groups(fs_info->extent_root);
 	if (ret) {
-		printk(KERN_ERR "BTRFS: Failed to read block groups: %d\n", ret);
+		btrfs_err(fs_info, "failed to read block groups: %d", ret);
 		goto fail_sysfs;
 	}
 	fs_info->num_tolerated_disk_barrier_failures =
@@ -2991,12 +2989,20 @@ int open_ctree(struct super_block *sb,
 	if (fs_info->fs_devices->missing_devices >
 	     fs_info->num_tolerated_disk_barrier_failures &&
 	    !(sb->s_flags & MS_RDONLY)) {
-		pr_warn("BTRFS: missing devices(%llu) exceeds the limit(%d), writeable mount is not allowed\n",
+		btrfs_warn(fs_info,
+"missing devices (%llu) exceeds the limit (%d), writeable mount is not allowed",
 			fs_info->fs_devices->missing_devices,
 			fs_info->num_tolerated_disk_barrier_failures);
 		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))
@@ -3011,8 +3017,7 @@ int open_ctree(struct super_block *sb,
 	if (!btrfs_test_opt(tree_root, SSD) &&
 	    !btrfs_test_opt(tree_root, NOSSD) &&
 	    !fs_info->fs_devices->rotating) {
-		printk(KERN_INFO "BTRFS: detected SSD devices, enabling SSD "
-		       "mode\n");
+		btrfs_info(fs_info, "detected SSD devices, enabling SSD mode");
 		btrfs_set_opt(fs_info->mount_opt, SSD);
 	}
 
@@ -3030,8 +3035,9 @@ int open_ctree(struct super_block *sb,
 				    1 : 0,
 				    fs_info->check_integrity_print_mask);
 		if (ret)
-			printk(KERN_WARNING "BTRFS: failed to initialize"
-			       " integrity check module %s\n", sb->s_id);
+			btrfs_warn(fs_info,
+				"failed to initialize integrity check module: %d",
+				ret);
 	}
 #endif
 	ret = btrfs_read_qgroup_config(fs_info);
@@ -3056,17 +3062,17 @@ int open_ctree(struct super_block *sb,
 		ret = btrfs_cleanup_fs_roots(fs_info);
 		if (ret)
 			goto fail_qgroup;
-
-		mutex_lock(&fs_info->cleaner_mutex);
+		/* We locked cleaner_mutex before creating cleaner_kthread. */
 		ret = btrfs_recover_relocation(tree_root);
-		mutex_unlock(&fs_info->cleaner_mutex);
 		if (ret < 0) {
-			printk(KERN_WARNING
-			       "BTRFS: failed to recover relocation\n");
+			btrfs_warn(fs_info, "failed to recover relocation: %d",
+					ret);
 			err = -EINVAL;
 			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;
@@ -3083,11 +3089,11 @@ int open_ctree(struct super_block *sb,
 
 	if (btrfs_test_opt(tree_root, FREE_SPACE_TREE) &&
 	    !btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
-		pr_info("BTRFS: creating free space tree\n");
+		btrfs_info(fs_info, "creating free space tree");
 		ret = btrfs_create_free_space_tree(fs_info);
 		if (ret) {
-			pr_warn("BTRFS: failed to create free space tree %d\n",
-				ret);
+			btrfs_warn(fs_info,
+				"failed to create free space tree: %d", ret);
 			close_ctree(tree_root);
 			return ret;
 		}
@@ -3104,14 +3110,14 @@ int open_ctree(struct super_block *sb,
 
 	ret = btrfs_resume_balance_async(fs_info);
 	if (ret) {
-		printk(KERN_WARNING "BTRFS: failed to resume balance\n");
+		btrfs_warn(fs_info, "failed to resume balance: %d", ret);
 		close_ctree(tree_root);
 		return ret;
 	}
 
 	ret = btrfs_resume_dev_replace_async(fs_info);
 	if (ret) {
-		pr_warn("BTRFS: failed to resume dev_replace\n");
+		btrfs_warn(fs_info, "failed to resume device replace: %d", ret);
 		close_ctree(tree_root);
 		return ret;
 	}
@@ -3120,33 +3126,33 @@ int open_ctree(struct super_block *sb,
 
 	if (btrfs_test_opt(tree_root, CLEAR_CACHE) &&
 	    btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) {
-		pr_info("BTRFS: clearing free space tree\n");
+		btrfs_info(fs_info, "clearing free space tree");
 		ret = btrfs_clear_free_space_tree(fs_info);
 		if (ret) {
-			pr_warn("BTRFS: failed to clear free space tree %d\n",
-				ret);
+			btrfs_warn(fs_info,
+				"failed to clear free space tree: %d", ret);
 			close_ctree(tree_root);
 			return ret;
 		}
 	}
 
 	if (!fs_info->uuid_root) {
-		pr_info("BTRFS: creating UUID tree\n");
+		btrfs_info(fs_info, "creating UUID tree");
 		ret = btrfs_create_uuid_tree(fs_info);
 		if (ret) {
-			pr_warn("BTRFS: failed to create the UUID tree %d\n",
-				ret);
+			btrfs_warn(fs_info,
+				"failed to create the UUID tree: %d", ret);
 			close_ctree(tree_root);
 			return ret;
 		}
 	} else if (btrfs_test_opt(tree_root, RESCAN_UUID_TREE) ||
 		   fs_info->generation !=
 				btrfs_super_uuid_tree_generation(disk_super)) {
-		pr_info("BTRFS: checking UUID tree\n");
+		btrfs_info(fs_info, "checking UUID tree");
 		ret = btrfs_check_uuid_tree(fs_info);
 		if (ret) {
-			pr_warn("BTRFS: failed to check the UUID tree %d\n",
-				ret);
+			btrfs_warn(fs_info,
+				"failed to check the UUID tree: %d", ret);
 			close_ctree(tree_root);
 			return ret;
 		}
@@ -3180,6 +3186,10 @@ int open_ctree(struct super_block *sb,
 	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:
@@ -3646,7 +3656,7 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors)
 		if (ret) {
 			mutex_unlock(
 				&root->fs_info->fs_devices->device_list_mutex);
-			btrfs_std_error(root->fs_info, ret,
+			btrfs_handle_fs_error(root->fs_info, ret,
 				    "errors while submitting device barriers.");
 			return ret;
 		}
@@ -3686,7 +3696,7 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors)
 		mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
 
 		/* FUA is masked off if unsupported and can't be the reason */
-		btrfs_std_error(root->fs_info, -EIO,
+		btrfs_handle_fs_error(root->fs_info, -EIO,
 			    "%d errors while writing supers", total_errors);
 		return -EIO;
 	}
@@ -3704,7 +3714,7 @@ static int write_all_supers(struct btrfs_root *root, int max_mirrors)
 	}
 	mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
 	if (total_errors > max_errors) {
-		btrfs_std_error(root->fs_info, -EIO,
+		btrfs_handle_fs_error(root->fs_info, -EIO,
 			    "%d errors while writing supers", total_errors);
 		return -EIO;
 	}

fs/btrfs/extent-tree.c

@@ -3824,6 +3824,59 @@ int btrfs_extent_readonly(struct btrfs_root *root, u64 bytenr)
 	return readonly;
 }
 
+bool btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
+{
+	struct btrfs_block_group_cache *bg;
+	bool ret = true;
+
+	bg = btrfs_lookup_block_group(fs_info, bytenr);
+	if (!bg)
+		return false;
+
+	spin_lock(&bg->lock);
+	if (bg->ro)
+		ret = false;
+	else
+		atomic_inc(&bg->nocow_writers);
+	spin_unlock(&bg->lock);
+
+	/* no put on block group, done by btrfs_dec_nocow_writers */
+	if (!ret)
+		btrfs_put_block_group(bg);
+
+	return ret;
+
+}
+
+void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
+{
+	struct btrfs_block_group_cache *bg;
+
+	bg = btrfs_lookup_block_group(fs_info, bytenr);
+	ASSERT(bg);
+	if (atomic_dec_and_test(&bg->nocow_writers))
+		wake_up_atomic_t(&bg->nocow_writers);
+	/*
+	 * Once for our lookup and once for the lookup done by a previous call
+	 * to btrfs_inc_nocow_writers()
+	 */
+	btrfs_put_block_group(bg);
+	btrfs_put_block_group(bg);
+}
+
+static int btrfs_wait_nocow_writers_atomic_t(atomic_t *a)
+{
+	schedule();
+	return 0;
+}
+
+void btrfs_wait_nocow_writers(struct btrfs_block_group_cache *bg)
+{
+	wait_on_atomic_t(&bg->nocow_writers,
+			 btrfs_wait_nocow_writers_atomic_t,
+			 TASK_UNINTERRUPTIBLE);
+}
+
 static const char *alloc_name(u64 flags)
 {
 	switch (flags) {
@@ -4141,7 +4194,7 @@ int btrfs_alloc_data_chunk_ondemand(struct inode *inode, u64 bytes)
 
 			if (need_commit > 0) {
 				btrfs_start_delalloc_roots(fs_info, 0, -1);
-				btrfs_wait_ordered_roots(fs_info, -1);
+				btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
 			}
 
 			trans = btrfs_join_transaction(root);
@@ -4583,7 +4636,8 @@ static void btrfs_writeback_inodes_sb_nr(struct btrfs_root *root,
 		 */
 		btrfs_start_delalloc_roots(root->fs_info, 0, nr_items);
 		if (!current->journal_info)
-			btrfs_wait_ordered_roots(root->fs_info, nr_items);
+			btrfs_wait_ordered_roots(root->fs_info, nr_items,
+						 0, (u64)-1);
 	}
 }
 
@@ -4620,7 +4674,7 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 
 	/* Calc the number of the pages we need flush for space reservation */
 	items = calc_reclaim_items_nr(root, to_reclaim);
-	to_reclaim = items * EXTENT_SIZE_PER_ITEM;
+	to_reclaim = (u64)items * EXTENT_SIZE_PER_ITEM;
 
 	trans = (struct btrfs_trans_handle *)current->journal_info;
 	block_rsv = &root->fs_info->delalloc_block_rsv;
@@ -4632,7 +4686,8 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 		if (trans)
 			return;
 		if (wait_ordered)
-			btrfs_wait_ordered_roots(root->fs_info, items);
+			btrfs_wait_ordered_roots(root->fs_info, items,
+						 0, (u64)-1);
 		return;
 	}
 
@@ -4671,7 +4726,8 @@ static void shrink_delalloc(struct btrfs_root *root, u64 to_reclaim, u64 orig,
 
 		loops++;
 		if (wait_ordered && !trans) {
-			btrfs_wait_ordered_roots(root->fs_info, items);
+			btrfs_wait_ordered_roots(root->fs_info, items,
+						 0, (u64)-1);
 		} else {
 			time_left = schedule_timeout_killable(1);
 			if (time_left)
@@ -6172,6 +6228,57 @@ int btrfs_exclude_logged_extents(struct btrfs_root *log,
 	return 0;
 }
 
+static void
+btrfs_inc_block_group_reservations(struct btrfs_block_group_cache *bg)
+{
+	atomic_inc(&bg->reservations);
+}
+
+void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
+					const u64 start)
+{
+	struct btrfs_block_group_cache *bg;
+
+	bg = btrfs_lookup_block_group(fs_info, start);
+	ASSERT(bg);
+	if (atomic_dec_and_test(&bg->reservations))
+		wake_up_atomic_t(&bg->reservations);
+	btrfs_put_block_group(bg);
+}
+
+static int btrfs_wait_bg_reservations_atomic_t(atomic_t *a)
+{
+	schedule();
+	return 0;
+}
+
+void btrfs_wait_block_group_reservations(struct btrfs_block_group_cache *bg)
+{
+	struct btrfs_space_info *space_info = bg->space_info;
+
+	ASSERT(bg->ro);
+
+	if (!(bg->flags & BTRFS_BLOCK_GROUP_DATA))
+		return;
+
+	/*
+	 * Our block group is read only but before we set it to read only,
+	 * some task might have had allocated an extent from it already, but it
+	 * has not yet created a respective ordered extent (and added it to a
+	 * root's list of ordered extents).
+	 * Therefore wait for any task currently allocating extents, since the
+	 * block group's reservations counter is incremented while a read lock
+	 * on the groups' semaphore is held and decremented after releasing
+	 * the read access on that semaphore and creating the ordered extent.
+	 */
+	down_write(&space_info->groups_sem);
+	up_write(&space_info->groups_sem);
+
+	wait_on_atomic_t(&bg->reservations,
+			 btrfs_wait_bg_reservations_atomic_t,
+			 TASK_UNINTERRUPTIBLE);
+}
+
 /**
  * btrfs_update_reserved_bytes - update the block_group and space info counters
  * @cache:	The cache we are manipulating
@@ -7025,36 +7132,35 @@ btrfs_lock_cluster(struct btrfs_block_group_cache *block_group,
 		   int delalloc)
 {
 	struct btrfs_block_group_cache *used_bg = NULL;
-	bool locked = false;
-again:
+
 	spin_lock(&cluster->refill_lock);
-	if (locked) {
-		if (used_bg == cluster->block_group)
+	while (1) {
+		used_bg = cluster->block_group;
+		if (!used_bg)
+			return NULL;
+
+		if (used_bg == block_group)
 			return used_bg;
 
-		up_read(&used_bg->data_rwsem);
-		btrfs_put_block_group(used_bg);
-	}
+		btrfs_get_block_group(used_bg);
 
-	used_bg = cluster->block_group;
-	if (!used_bg)
-		return NULL;
+		if (!delalloc)
+			return used_bg;
 
-	if (used_bg == block_group)
-		return used_bg;
+		if (down_read_trylock(&used_bg->data_rwsem))
+			return used_bg;
 
-	btrfs_get_block_group(used_bg);
+		spin_unlock(&cluster->refill_lock);
 
-	if (!delalloc)
-		return used_bg;
+		down_read(&used_bg->data_rwsem);
 
-	if (down_read_trylock(&used_bg->data_rwsem))
-		return used_bg;
+		spin_lock(&cluster->refill_lock);
+		if (used_bg == cluster->block_group)
+			return used_bg;
 
-	spin_unlock(&cluster->refill_lock);
-	down_read(&used_bg->data_rwsem);
-	locked = true;
-	goto again;
+		up_read(&used_bg->data_rwsem);
+		btrfs_put_block_group(used_bg);
+	}
 }
 
 static inline void
@@ -7431,6 +7537,7 @@ static noinline int find_free_extent(struct btrfs_root *orig_root,
 			btrfs_add_free_space(block_group, offset, num_bytes);
 			goto loop;
 		}
+		btrfs_inc_block_group_reservations(block_group);
 
 		/* we are all good, lets return */
 		ins->objectid = search_start;
@@ -7612,8 +7719,10 @@ int btrfs_reserve_extent(struct btrfs_root *root,
 	WARN_ON(num_bytes < root->sectorsize);
 	ret = find_free_extent(root, num_bytes, empty_size, hint_byte, ins,
 			       flags, delalloc);
-
-	if (ret == -ENOSPC) {
+	if (!ret && !is_data) {
+		btrfs_dec_block_group_reservations(root->fs_info,
+						   ins->objectid);
+	} else if (ret == -ENOSPC) {
 		if (!final_tried && ins->offset) {
 			num_bytes = min(num_bytes >> 1, ins->offset);
 			num_bytes = round_down(num_bytes, root->sectorsize);
@@ -9058,7 +9167,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 	if (!for_reloc && root_dropped == false)
 		btrfs_add_dead_root(root);
 	if (err && err != -EAGAIN)
-		btrfs_std_error(root->fs_info, err, NULL);
+		btrfs_handle_fs_error(root->fs_info, err, NULL);
 	return err;
 }
 

fs/btrfs/extent_io.c

@@ -3200,14 +3200,10 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
 	return ret;
 }
 
-static noinline void update_nr_written(struct page *page,
-				      struct writeback_control *wbc,
-				      unsigned long nr_written)
+static void update_nr_written(struct page *page, struct writeback_control *wbc,
+			      unsigned long nr_written)
 {
 	wbc->nr_to_write -= nr_written;
-	if (wbc->range_cyclic || (wbc->nr_to_write > 0 &&
-	    wbc->range_start == 0 && wbc->range_end == LLONG_MAX))
-		page->mapping->writeback_index = page->index + nr_written;
 }
 
 /*
@@ -3368,6 +3364,8 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 
 	while (cur <= end) {
 		u64 em_end;
+		unsigned long max_nr;
+
 		if (cur >= i_size) {
 			if (tree->ops && tree->ops->writepage_end_io_hook)
 				tree->ops->writepage_end_io_hook(page, cur,
@@ -3423,32 +3421,23 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode,
 			continue;
 		}
 
-		if (tree->ops && tree->ops->writepage_io_hook) {
-			ret = tree->ops->writepage_io_hook(page, cur,
-						cur + iosize - 1);
-		} else {
-			ret = 0;
+		max_nr = (i_size >> PAGE_SHIFT) + 1;
+
+		set_range_writeback(tree, cur, cur + iosize - 1);
+		if (!PageWriteback(page)) {
+			btrfs_err(BTRFS_I(inode)->root->fs_info,
+				   "page %lu not writeback, cur %llu end %llu",
+			       page->index, cur, end);
 		}
-		if (ret) {
-			SetPageError(page);
-		} else {
-			unsigned long max_nr = (i_size >> PAGE_SHIFT) + 1;
 
-			set_range_writeback(tree, cur, cur + iosize - 1);
-			if (!PageWriteback(page)) {
-				btrfs_err(BTRFS_I(inode)->root->fs_info,
-					   "page %lu not writeback, cur %llu end %llu",
-				       page->index, cur, end);
-			}
+		ret = submit_extent_page(write_flags, tree, wbc, page,
+					 sector, iosize, pg_offset,
+					 bdev, &epd->bio, max_nr,
+					 end_bio_extent_writepage,
+					 0, 0, 0, false);
+		if (ret)
+			SetPageError(page);
 
-			ret = submit_extent_page(write_flags, tree, wbc, page,
-						 sector, iosize, pg_offset,
-						 bdev, &epd->bio, max_nr,
-						 end_bio_extent_writepage,
-						 0, 0, 0, false);
-			if (ret)
-				SetPageError(page);
-		}
 		cur = cur + iosize;
 		pg_offset += iosize;
 		nr++;
@@ -3920,12 +3909,13 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 	struct inode *inode = mapping->host;
 	int ret = 0;
 	int done = 0;
-	int err = 0;
 	int nr_to_write_done = 0;
 	struct pagevec pvec;
 	int nr_pages;
 	pgoff_t index;
 	pgoff_t end;		/* Inclusive */
+	pgoff_t done_index;
+	int range_whole = 0;
 	int scanned = 0;
 	int tag;
 
@@ -3948,6 +3938,8 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 	} else {
 		index = wbc->range_start >> PAGE_SHIFT;
 		end = wbc->range_end >> PAGE_SHIFT;
+		if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
+			range_whole = 1;
 		scanned = 1;
 	}
 	if (wbc->sync_mode == WB_SYNC_ALL)
@@ -3957,6 +3949,7 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 retry:
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		tag_pages_for_writeback(mapping, index, end);
+	done_index = index;
 	while (!done && !nr_to_write_done && (index <= end) &&
 	       (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, tag,
 			min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1))) {
@@ -3966,6 +3959,7 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
 
+			done_index = page->index;
 			/*
 			 * At this point we hold neither mapping->tree_lock nor
 			 * lock on the page itself: the page may be truncated or
@@ -4007,8 +4001,20 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 				unlock_page(page);
 				ret = 0;
 			}
-			if (!err && ret < 0)
-				err = ret;
+			if (ret < 0) {
+				/*
+				 * done_index is set past this page,
+				 * so media errors will not choke
+				 * background writeout for the entire
+				 * file. This has consequences for
+				 * range_cyclic semantics (ie. it may
+				 * not be suitable for data integrity
+				 * writeout).
+				 */
+				done_index = page->index + 1;
+				done = 1;
+				break;
+			}
 
 			/*
 			 * the filesystem may choose to bump up nr_to_write.
@@ -4020,7 +4026,7 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 		pagevec_release(&pvec);
 		cond_resched();
 	}
-	if (!scanned && !done && !err) {
+	if (!scanned && !done) {
 		/*
 		 * We hit the last page and there is more work to be done: wrap
 		 * back to the start of the file
@@ -4029,8 +4035,12 @@ static int extent_write_cache_pages(struct extent_io_tree *tree,
 		index = 0;
 		goto retry;
 	}
+
+	if (wbc->range_cyclic || (wbc->nr_to_write > 0 && range_whole))
+		mapping->writeback_index = done_index;
+
 	btrfs_add_delayed_iput(inode);
-	return err;
+	return ret;
 }
 
 static void flush_epd_write_bio(struct extent_page_data *epd)
@@ -4822,7 +4832,7 @@ struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 		return NULL;
 	eb->fs_info = fs_info;
 again:
-	ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
+	ret = radix_tree_preload(GFP_NOFS);
 	if (ret)
 		goto free_eb;
 	spin_lock(&fs_info->buffer_lock);
@@ -4923,7 +4933,7 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 	if (uptodate)
 		set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
 again:
-	ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
+	ret = radix_tree_preload(GFP_NOFS);
 	if (ret)
 		goto free_eb;
 

fs/btrfs/extent_io.h

@@ -71,7 +71,6 @@ struct extent_io_ops {
 			     u64 start, u64 end, int *page_started,
 			     unsigned long *nr_written);
 	int (*writepage_start_hook)(struct page *page, u64 start, u64 end);
-	int (*writepage_io_hook)(struct page *page, u64 start, u64 end);
 	extent_submit_bio_hook_t *submit_bio_hook;
 	int (*merge_bio_hook)(int rw, struct page *page, unsigned long offset,
 			      size_t size, struct bio *bio,

fs/btrfs/file.c

@@ -1696,7 +1696,9 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 			btrfs_end_write_no_snapshoting(root);
 			btrfs_delalloc_release_metadata(inode, release_bytes);
 		} else {
-			btrfs_delalloc_release_space(inode, pos, release_bytes);
+			btrfs_delalloc_release_space(inode,
+						round_down(pos, root->sectorsize),
+						release_bytes);
 		}
 	}
 
@@ -2952,7 +2954,7 @@ const struct file_operations btrfs_file_operations = {
 	.fallocate	= btrfs_fallocate,
 	.unlocked_ioctl	= btrfs_ioctl,
 #ifdef CONFIG_COMPAT
-	.compat_ioctl	= btrfs_ioctl,
+	.compat_ioctl	= btrfs_compat_ioctl,
 #endif
 	.copy_file_range = btrfs_copy_file_range,
 	.clone_file_range = btrfs_clone_file_range,

fs/btrfs/inode-item.c

@@ -157,7 +157,7 @@ static int btrfs_del_inode_extref(struct btrfs_trans_handle *trans,
 	 */
 	if (!btrfs_find_name_in_ext_backref(path, ref_objectid,
 					    name, name_len, &extref)) {
-		btrfs_std_error(root->fs_info, -ENOENT, NULL);
+		btrfs_handle_fs_error(root->fs_info, -ENOENT, NULL);
 		ret = -EROFS;
 		goto out;
 	}

fs/btrfs/ioctl.c

@@ -125,10 +125,10 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags)
 	if (flags & BTRFS_INODE_NODATACOW)
 		iflags |= FS_NOCOW_FL;
 
-	if ((flags & BTRFS_INODE_COMPRESS) && !(flags & BTRFS_INODE_NOCOMPRESS))
-		iflags |= FS_COMPR_FL;
-	else if (flags & BTRFS_INODE_NOCOMPRESS)
+	if (flags & BTRFS_INODE_NOCOMPRESS)
 		iflags |= FS_NOCOMP_FL;
+	else if (flags & BTRFS_INODE_COMPRESS)
+		iflags |= FS_COMPR_FL;
 
 	return iflags;
 }
@@ -439,7 +439,7 @@ static noinline int create_subvol(struct inode *dir,
 {
 	struct btrfs_trans_handle *trans;
 	struct btrfs_key key;
-	struct btrfs_root_item root_item;
+	struct btrfs_root_item *root_item;
 	struct btrfs_inode_item *inode_item;
 	struct extent_buffer *leaf;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
@@ -455,16 +455,22 @@ static noinline int create_subvol(struct inode *dir,
 	u64 qgroup_reserved;
 	uuid_le new_uuid;
 
+	root_item = kzalloc(sizeof(*root_item), GFP_KERNEL);
+	if (!root_item)
+		return -ENOMEM;
+
 	ret = btrfs_find_free_objectid(root->fs_info->tree_root, &objectid);
 	if (ret)
-		return ret;
+		goto fail_free;
 
 	/*
 	 * Don't create subvolume whose level is not zero. Or qgroup will be
 	 * screwed up since it assume subvolme qgroup's level to be 0.
 	 */
-	if (btrfs_qgroup_level(objectid))
-		return -ENOSPC;
+	if (btrfs_qgroup_level(objectid)) {
+		ret = -ENOSPC;
+		goto fail_free;
+	}
 
 	btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
 	/*
@@ -474,14 +480,14 @@ static noinline int create_subvol(struct inode *dir,
 	ret = btrfs_subvolume_reserve_metadata(root, &block_rsv,
 					       8, &qgroup_reserved, false);
 	if (ret)
-		return ret;
+		goto fail_free;
 
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
 		btrfs_subvolume_release_metadata(root, &block_rsv,
 						 qgroup_reserved);
-		return ret;
+		goto fail_free;
 	}
 	trans->block_rsv = &block_rsv;
 	trans->bytes_reserved = block_rsv.size;
@@ -509,47 +515,45 @@ static noinline int create_subvol(struct inode *dir,
 			    BTRFS_UUID_SIZE);
 	btrfs_mark_buffer_dirty(leaf);
 
-	memset(&root_item, 0, sizeof(root_item));
-
-	inode_item = &root_item.inode;
+	inode_item = &root_item->inode;
 	btrfs_set_stack_inode_generation(inode_item, 1);
 	btrfs_set_stack_inode_size(inode_item, 3);
 	btrfs_set_stack_inode_nlink(inode_item, 1);
 	btrfs_set_stack_inode_nbytes(inode_item, root->nodesize);
 	btrfs_set_stack_inode_mode(inode_item, S_IFDIR | 0755);
 
-	btrfs_set_root_flags(&root_item, 0);
-	btrfs_set_root_limit(&root_item, 0);
+	btrfs_set_root_flags(root_item, 0);
+	btrfs_set_root_limit(root_item, 0);
 	btrfs_set_stack_inode_flags(inode_item, BTRFS_INODE_ROOT_ITEM_INIT);
 
-	btrfs_set_root_bytenr(&root_item, leaf->start);
-	btrfs_set_root_generation(&root_item, trans->transid);
-	btrfs_set_root_level(&root_item, 0);
-	btrfs_set_root_refs(&root_item, 1);
-	btrfs_set_root_used(&root_item, leaf->len);
-	btrfs_set_root_last_snapshot(&root_item, 0);
+	btrfs_set_root_bytenr(root_item, leaf->start);
+	btrfs_set_root_generation(root_item, trans->transid);
+	btrfs_set_root_level(root_item, 0);
+	btrfs_set_root_refs(root_item, 1);
+	btrfs_set_root_used(root_item, leaf->len);
+	btrfs_set_root_last_snapshot(root_item, 0);
 
-	btrfs_set_root_generation_v2(&root_item,
-			btrfs_root_generation(&root_item));
+	btrfs_set_root_generation_v2(root_item,
+			btrfs_root_generation(root_item));
 	uuid_le_gen(&new_uuid);
-	memcpy(root_item.uuid, new_uuid.b, BTRFS_UUID_SIZE);
-	btrfs_set_stack_timespec_sec(&root_item.otime, cur_time.tv_sec);
-	btrfs_set_stack_timespec_nsec(&root_item.otime, cur_time.tv_nsec);
-	root_item.ctime = root_item.otime;
-	btrfs_set_root_ctransid(&root_item, trans->transid);
-	btrfs_set_root_otransid(&root_item, trans->transid);
+	memcpy(root_item->uuid, new_uuid.b, BTRFS_UUID_SIZE);
+	btrfs_set_stack_timespec_sec(&root_item->otime, cur_time.tv_sec);
+	btrfs_set_stack_timespec_nsec(&root_item->otime, cur_time.tv_nsec);
+	root_item->ctime = root_item->otime;
+	btrfs_set_root_ctransid(root_item, trans->transid);
+	btrfs_set_root_otransid(root_item, trans->transid);
 
 	btrfs_tree_unlock(leaf);
 	free_extent_buffer(leaf);
 	leaf = NULL;
 
-	btrfs_set_root_dirid(&root_item, new_dirid);
+	btrfs_set_root_dirid(root_item, new_dirid);
 
 	key.objectid = objectid;
 	key.offset = 0;
 	key.type = BTRFS_ROOT_ITEM_KEY;
 	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
-				&root_item);
+				root_item);
 	if (ret)
 		goto fail;
 
@@ -601,12 +605,13 @@ static noinline int create_subvol(struct inode *dir,
 	BUG_ON(ret);
 
 	ret = btrfs_uuid_tree_add(trans, root->fs_info->uuid_root,
-				  root_item.uuid, BTRFS_UUID_KEY_SUBVOL,
+				  root_item->uuid, BTRFS_UUID_KEY_SUBVOL,
 				  objectid);
 	if (ret)
 		btrfs_abort_transaction(trans, root, ret);
 
 fail:
+	kfree(root_item);
 	trans->block_rsv = NULL;
 	trans->bytes_reserved = 0;
 	btrfs_subvolume_release_metadata(root, &block_rsv, qgroup_reserved);
@@ -629,6 +634,10 @@ static noinline int create_subvol(struct inode *dir,
 		d_instantiate(dentry, inode);
 	}
 	return ret;
+
+fail_free:
+	kfree(root_item);
+	return ret;
 }
 
 static void btrfs_wait_for_no_snapshoting_writes(struct btrfs_root *root)
@@ -681,7 +690,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	if (ret)
 		goto dec_and_free;
 
-	btrfs_wait_ordered_extents(root, -1);
+	btrfs_wait_ordered_extents(root, -1, 0, (u64)-1);
 
 	btrfs_init_block_rsv(&pending_snapshot->block_rsv,
 			     BTRFS_BLOCK_RSV_TEMP);
@@ -2671,10 +2680,10 @@ static long btrfs_ioctl_add_dev(struct btrfs_root *root, void __user *arg)
 	return ret;
 }
 
-static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
+static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg)
 {
 	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
-	struct btrfs_ioctl_vol_args *vol_args;
+	struct btrfs_ioctl_vol_args_v2 *vol_args;
 	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -2690,7 +2699,9 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 		goto err_drop;
 	}
 
-	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
+	/* Check for compatibility reject unknown flags */
+	if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED)
+		return -EOPNOTSUPP;
 
 	if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
 			1)) {
@@ -2699,13 +2710,23 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	}
 
 	mutex_lock(&root->fs_info->volume_mutex);
-	ret = btrfs_rm_device(root, vol_args->name);
+	if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID) {
+		ret = btrfs_rm_device(root, NULL, vol_args->devid);
+	} else {
+		vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0';
+		ret = btrfs_rm_device(root, vol_args->name, 0);
+	}
 	mutex_unlock(&root->fs_info->volume_mutex);
 	atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0);
 
-	if (!ret)
-		btrfs_info(root->fs_info, "disk deleted %s",vol_args->name);
-
+	if (!ret) {
+		if (vol_args->flags & BTRFS_DEVICE_SPEC_BY_ID)
+			btrfs_info(root->fs_info, "device deleted: id %llu",
+					vol_args->devid);
+		else
+			btrfs_info(root->fs_info, "device deleted: %s",
+					vol_args->name);
+	}
 out:
 	kfree(vol_args);
 err_drop:
@@ -2713,6 +2734,47 @@ static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
 	return ret;
 }
 
+static long btrfs_ioctl_rm_dev(struct file *file, void __user *arg)
+{
+	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
+	struct btrfs_ioctl_vol_args *vol_args;
+	int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	ret = mnt_want_write_file(file);
+	if (ret)
+		return ret;
+
+	if (atomic_xchg(&root->fs_info->mutually_exclusive_operation_running,
+			1)) {
+		ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
+		goto out_drop_write;
+	}
+
+	vol_args = memdup_user(arg, sizeof(*vol_args));
+	if (IS_ERR(vol_args)) {
+		ret = PTR_ERR(vol_args);
+		goto out;
+	}
+
+	vol_args->name[BTRFS_PATH_NAME_MAX] = '\0';
+	mutex_lock(&root->fs_info->volume_mutex);
+	ret = btrfs_rm_device(root, vol_args->name, 0);
+	mutex_unlock(&root->fs_info->volume_mutex);
+
+	if (!ret)
+		btrfs_info(root->fs_info, "disk deleted %s",vol_args->name);
+	kfree(vol_args);
+out:
+	atomic_set(&root->fs_info->mutually_exclusive_operation_running, 0);
+out_drop_write:
+	mnt_drop_write_file(file);
+
+	return ret;
+}
+
 static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg)
 {
 	struct btrfs_ioctl_fs_info_args *fi_args;
@@ -3472,13 +3534,16 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 	u64 last_dest_end = destoff;
 
 	ret = -ENOMEM;
-	buf = vmalloc(root->nodesize);
-	if (!buf)
-		return ret;
+	buf = kmalloc(root->nodesize, GFP_KERNEL | __GFP_NOWARN);
+	if (!buf) {
+		buf = vmalloc(root->nodesize);
+		if (!buf)
+			return ret;
+	}
 
 	path = btrfs_alloc_path();
 	if (!path) {
-		vfree(buf);
+		kvfree(buf);
 		return ret;
 	}
 
@@ -3779,7 +3844,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 
 out:
 	btrfs_free_path(path);
-	vfree(buf);
+	kvfree(buf);
 	return ret;
 }
 
@@ -4380,7 +4445,7 @@ static long btrfs_ioctl_dev_replace(struct btrfs_root *root, void __user *arg)
 			1)) {
 			ret = BTRFS_ERROR_DEV_EXCL_RUN_IN_PROGRESS;
 		} else {
-			ret = btrfs_dev_replace_start(root, p);
+			ret = btrfs_dev_replace_by_ioctl(root, p);
 			atomic_set(
 			 &root->fs_info->mutually_exclusive_operation_running,
 			 0);
@@ -4851,8 +4916,8 @@ static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg)
 	/* update qgroup status and info */
 	err = btrfs_run_qgroups(trans, root->fs_info);
 	if (err < 0)
-		btrfs_std_error(root->fs_info, ret,
-			    "failed to update qgroup status and info\n");
+		btrfs_handle_fs_error(root->fs_info, err,
+			    "failed to update qgroup status and info");
 	err = btrfs_end_transaction(trans, root);
 	if (err && !ret)
 		ret = err;
@@ -5398,9 +5463,15 @@ static int btrfs_ioctl_set_features(struct file *file, void __user *arg)
 	if (ret)
 		return ret;
 
+	ret = mnt_want_write_file(file);
+	if (ret)
+		return ret;
+
 	trans = btrfs_start_transaction(root, 0);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		goto out_drop_write;
+	}
 
 	spin_lock(&root->fs_info->super_lock);
 	newflags = btrfs_super_compat_flags(super_block);
@@ -5419,7 +5490,11 @@ static int btrfs_ioctl_set_features(struct file *file, void __user *arg)
 	btrfs_set_super_incompat_flags(super_block, newflags);
 	spin_unlock(&root->fs_info->super_lock);
 
-	return btrfs_commit_transaction(trans, root);
+	ret = btrfs_commit_transaction(trans, root);
+out_drop_write:
+	mnt_drop_write_file(file);
+
+	return ret;
 }
 
 long btrfs_ioctl(struct file *file, unsigned int
@@ -5463,6 +5538,8 @@ long btrfs_ioctl(struct file *file, unsigned int
 		return btrfs_ioctl_add_dev(root, argp);
 	case BTRFS_IOC_RM_DEV:
 		return btrfs_ioctl_rm_dev(file, argp);
+	case BTRFS_IOC_RM_DEV_V2:
+		return btrfs_ioctl_rm_dev_v2(file, argp);
 	case BTRFS_IOC_FS_INFO:
 		return btrfs_ioctl_fs_info(root, argp);
 	case BTRFS_IOC_DEV_INFO:
@@ -5556,3 +5633,24 @@ long btrfs_ioctl(struct file *file, unsigned int
 
 	return -ENOTTY;
 }
+
+#ifdef CONFIG_COMPAT
+long btrfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case FS_IOC32_GETFLAGS:
+		cmd = FS_IOC_GETFLAGS;
+		break;
+	case FS_IOC32_SETFLAGS:
+		cmd = FS_IOC_SETFLAGS;
+		break;
+	case FS_IOC32_GETVERSION:
+		cmd = FS_IOC_GETVERSION;
+		break;
+	default:
+		return -ENOIOCTLCMD;
+	}
+
+	return btrfs_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
+}
+#endif

fs/btrfs/ordered-data.c

@@ -661,14 +661,15 @@ static void btrfs_run_ordered_extent_work(struct btrfs_work *work)
  * wait for all the ordered extents in a root.  This is done when balancing
  * space between drives.
  */
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
+int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
+			       const u64 range_start, const u64 range_len)
 {
-	struct list_head splice, works;
+	LIST_HEAD(splice);
+	LIST_HEAD(skipped);
+	LIST_HEAD(works);
 	struct btrfs_ordered_extent *ordered, *next;
 	int count = 0;
-
-	INIT_LIST_HEAD(&splice);
-	INIT_LIST_HEAD(&works);
+	const u64 range_end = range_start + range_len;
 
 	mutex_lock(&root->ordered_extent_mutex);
 	spin_lock(&root->ordered_extent_lock);
@@ -676,6 +677,14 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 	while (!list_empty(&splice) && nr) {
 		ordered = list_first_entry(&splice, struct btrfs_ordered_extent,
 					   root_extent_list);
+
+		if (range_end <= ordered->start ||
+		    ordered->start + ordered->disk_len <= range_start) {
+			list_move_tail(&ordered->root_extent_list, &skipped);
+			cond_resched_lock(&root->ordered_extent_lock);
+			continue;
+		}
+
 		list_move_tail(&ordered->root_extent_list,
 			       &root->ordered_extents);
 		atomic_inc(&ordered->refs);
@@ -694,6 +703,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 			nr--;
 		count++;
 	}
+	list_splice_tail(&skipped, &root->ordered_extents);
 	list_splice_tail(&splice, &root->ordered_extents);
 	spin_unlock(&root->ordered_extent_lock);
 
@@ -708,7 +718,8 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 	return count;
 }
 
-void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr)
+void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
+			      const u64 range_start, const u64 range_len)
 {
 	struct btrfs_root *root;
 	struct list_head splice;
@@ -728,7 +739,8 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr)
 			       &fs_info->ordered_roots);
 		spin_unlock(&fs_info->ordered_root_lock);
 
-		done = btrfs_wait_ordered_extents(root, nr);
+		done = btrfs_wait_ordered_extents(root, nr,
+						  range_start, range_len);
 		btrfs_put_fs_root(root);
 
 		spin_lock(&fs_info->ordered_root_lock);

fs/btrfs/ordered-data.h

@@ -197,8 +197,10 @@ int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
 			   u32 *sum, int len);
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr);
-void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr);
+int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr,
+			       const u64 range_start, const u64 range_len);
+void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr,
+			      const u64 range_start, const u64 range_len);
 void btrfs_get_logged_extents(struct inode *inode,
 			      struct list_head *logged_list,
 			      const loff_t start,

fs/btrfs/relocation.c

@@ -2418,7 +2418,7 @@ void merge_reloc_roots(struct reloc_control *rc)
 	}
 out:
 	if (ret) {
-		btrfs_std_error(root->fs_info, ret, NULL);
+		btrfs_handle_fs_error(root->fs_info, ret, NULL);
 		if (!list_empty(&reloc_roots))
 			free_reloc_roots(&reloc_roots);
 
@@ -4254,12 +4254,11 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
 	btrfs_info(extent_root->fs_info, "relocating block group %llu flags %llu",
 	       rc->block_group->key.objectid, rc->block_group->flags);
 
-	ret = btrfs_start_delalloc_roots(fs_info, 0, -1);
-	if (ret < 0) {
-		err = ret;
-		goto out;
-	}
-	btrfs_wait_ordered_roots(fs_info, -1);
+	btrfs_wait_block_group_reservations(rc->block_group);
+	btrfs_wait_nocow_writers(rc->block_group);
+	btrfs_wait_ordered_roots(fs_info, -1,
+				 rc->block_group->key.objectid,
+				 rc->block_group->key.offset);
 
 	while (1) {
 		mutex_lock(&fs_info->cleaner_mutex);

fs/btrfs/root-tree.c

@@ -284,7 +284,7 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 			trans = btrfs_join_transaction(tree_root);
 			if (IS_ERR(trans)) {
 				err = PTR_ERR(trans);
-				btrfs_std_error(tree_root->fs_info, err,
+				btrfs_handle_fs_error(tree_root->fs_info, err,
 					    "Failed to start trans to delete "
 					    "orphan item");
 				break;
@@ -293,7 +293,7 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 						    root_key.objectid);
 			btrfs_end_transaction(trans, tree_root);
 			if (err) {
-				btrfs_std_error(tree_root->fs_info, err,
+				btrfs_handle_fs_error(tree_root->fs_info, err,
 					    "Failed to delete root orphan "
 					    "item");
 				break;

fs/btrfs/scrub.c

@@ -1350,7 +1350,7 @@ static int scrub_setup_recheck_block(struct scrub_block *original_sblock,
 		recover->bbio = bbio;
 		recover->map_length = mapped_length;
 
-		BUG_ON(page_index >= SCRUB_PAGES_PER_RD_BIO);
+		BUG_ON(page_index >= SCRUB_MAX_PAGES_PER_BLOCK);
 
 		nmirrors = min(scrub_nr_raid_mirrors(bbio), BTRFS_MAX_MIRRORS);
 
@@ -2127,6 +2127,8 @@ static void scrub_missing_raid56_end_io(struct bio *bio)
 	if (bio->bi_error)
 		sblock->no_io_error_seen = 0;
 
+	bio_put(bio);
+
 	btrfs_queue_work(fs_info->scrub_workers, &sblock->work);
 }
 
@@ -2860,7 +2862,7 @@ static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx,
 	int extent_mirror_num;
 	int stop_loop = 0;
 
-	nsectors = map->stripe_len / root->sectorsize;
+	nsectors = div_u64(map->stripe_len, root->sectorsize);
 	bitmap_len = scrub_calc_parity_bitmap_len(nsectors);
 	sparity = kzalloc(sizeof(struct scrub_parity) + 2 * bitmap_len,
 			  GFP_NOFS);
@@ -3070,7 +3072,6 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	int slot;
 	u64 nstripes;
 	struct extent_buffer *l;
-	struct btrfs_key key;
 	u64 physical;
 	u64 logical;
 	u64 logic_end;
@@ -3079,7 +3080,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	int mirror_num;
 	struct reada_control *reada1;
 	struct reada_control *reada2;
-	struct btrfs_key key_start;
+	struct btrfs_key key;
 	struct btrfs_key key_end;
 	u64 increment = map->stripe_len;
 	u64 offset;
@@ -3158,21 +3159,21 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	scrub_blocked_if_needed(fs_info);
 
 	/* FIXME it might be better to start readahead at commit root */
-	key_start.objectid = logical;
-	key_start.type = BTRFS_EXTENT_ITEM_KEY;
-	key_start.offset = (u64)0;
+	key.objectid = logical;
+	key.type = BTRFS_EXTENT_ITEM_KEY;
+	key.offset = (u64)0;
 	key_end.objectid = logic_end;
 	key_end.type = BTRFS_METADATA_ITEM_KEY;
 	key_end.offset = (u64)-1;
-	reada1 = btrfs_reada_add(root, &key_start, &key_end);
+	reada1 = btrfs_reada_add(root, &key, &key_end);
 
-	key_start.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
-	key_start.type = BTRFS_EXTENT_CSUM_KEY;
-	key_start.offset = logical;
+	key.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
+	key.type = BTRFS_EXTENT_CSUM_KEY;
+	key.offset = logical;
 	key_end.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
 	key_end.type = BTRFS_EXTENT_CSUM_KEY;
 	key_end.offset = logic_end;
-	reada2 = btrfs_reada_add(csum_root, &key_start, &key_end);
+	reada2 = btrfs_reada_add(csum_root, &key, &key_end);
 
 	if (!IS_ERR(reada1))
 		btrfs_reada_wait(reada1);

fs/btrfs/send.c

@@ -5939,6 +5939,7 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
 	u32 i;
 	u64 *clone_sources_tmp = NULL;
 	int clone_sources_to_rollback = 0;
+	unsigned alloc_size;
 	int sort_clone_roots = 0;
 	int index;
 
@@ -5978,6 +5979,12 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
 		goto out;
 	}
 
+	if (arg->clone_sources_count >
+	    ULLONG_MAX / sizeof(*arg->clone_sources)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
 	if (!access_ok(VERIFY_READ, arg->clone_sources,
 			sizeof(*arg->clone_sources) *
 			arg->clone_sources_count)) {
@@ -6022,40 +6029,53 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
 	sctx->clone_roots_cnt = arg->clone_sources_count;
 
 	sctx->send_max_size = BTRFS_SEND_BUF_SIZE;
-	sctx->send_buf = vmalloc(sctx->send_max_size);
+	sctx->send_buf = kmalloc(sctx->send_max_size, GFP_KERNEL | __GFP_NOWARN);
 	if (!sctx->send_buf) {
-		ret = -ENOMEM;
-		goto out;
+		sctx->send_buf = vmalloc(sctx->send_max_size);
+		if (!sctx->send_buf) {
+			ret = -ENOMEM;
+			goto out;
+		}
 	}
 
-	sctx->read_buf = vmalloc(BTRFS_SEND_READ_SIZE);
+	sctx->read_buf = kmalloc(BTRFS_SEND_READ_SIZE, GFP_KERNEL | __GFP_NOWARN);
 	if (!sctx->read_buf) {
-		ret = -ENOMEM;
-		goto out;
+		sctx->read_buf = vmalloc(BTRFS_SEND_READ_SIZE);
+		if (!sctx->read_buf) {
+			ret = -ENOMEM;
+			goto out;
+		}
 	}
 
 	sctx->pending_dir_moves = RB_ROOT;
 	sctx->waiting_dir_moves = RB_ROOT;
 	sctx->orphan_dirs = RB_ROOT;
 
-	sctx->clone_roots = vzalloc(sizeof(struct clone_root) *
-			(arg->clone_sources_count + 1));
+	alloc_size = sizeof(struct clone_root) * (arg->clone_sources_count + 1);
+
+	sctx->clone_roots = kzalloc(alloc_size, GFP_KERNEL | __GFP_NOWARN);
 	if (!sctx->clone_roots) {
-		ret = -ENOMEM;
-		goto out;
+		sctx->clone_roots = vzalloc(alloc_size);
+		if (!sctx->clone_roots) {
+			ret = -ENOMEM;
+			goto out;
+		}
 	}
 
+	alloc_size = arg->clone_sources_count * sizeof(*arg->clone_sources);
+
 	if (arg->clone_sources_count) {
-		clone_sources_tmp = vmalloc(arg->clone_sources_count *
-				sizeof(*arg->clone_sources));
+		clone_sources_tmp = kmalloc(alloc_size, GFP_KERNEL | __GFP_NOWARN);
 		if (!clone_sources_tmp) {
-			ret = -ENOMEM;
-			goto out;
+			clone_sources_tmp = vmalloc(alloc_size);
+			if (!clone_sources_tmp) {
+				ret = -ENOMEM;
+				goto out;
+			}
 		}
 
 		ret = copy_from_user(clone_sources_tmp, arg->clone_sources,
-				arg->clone_sources_count *
-				sizeof(*arg->clone_sources));
+				alloc_size);
 		if (ret) {
 			ret = -EFAULT;
 			goto out;
@@ -6089,7 +6109,7 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
 			sctx->clone_roots[i].root = clone_root;
 			clone_sources_to_rollback = i + 1;
 		}
-		vfree(clone_sources_tmp);
+		kvfree(clone_sources_tmp);
 		clone_sources_tmp = NULL;
 	}
 
@@ -6207,15 +6227,15 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
 		btrfs_root_dec_send_in_progress(sctx->parent_root);
 
 	kfree(arg);
-	vfree(clone_sources_tmp);
+	kvfree(clone_sources_tmp);
 
 	if (sctx) {
 		if (sctx->send_filp)
 			fput(sctx->send_filp);
 
-		vfree(sctx->clone_roots);
-		vfree(sctx->send_buf);
-		vfree(sctx->read_buf);
+		kvfree(sctx->clone_roots);
+		kvfree(sctx->send_buf);
+		kvfree(sctx->read_buf);
 
 		name_cache_free(sctx);
 

fs/btrfs/super.c

@@ -97,15 +97,6 @@ const char *btrfs_decode_error(int errno)
 	return errstr;
 }
 
-static void save_error_info(struct btrfs_fs_info *fs_info)
-{
-	/*
-	 * today we only save the error info into ram.  Long term we'll
-	 * also send it down to the disk
-	 */
-	set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
-}
-
 /* btrfs handle error by forcing the filesystem readonly */
 static void btrfs_handle_error(struct btrfs_fs_info *fs_info)
 {
@@ -131,11 +122,11 @@ static void btrfs_handle_error(struct btrfs_fs_info *fs_info)
 }
 
 /*
- * __btrfs_std_error decodes expected errors from the caller and
+ * __btrfs_handle_fs_error decodes expected errors from the caller and
  * invokes the approciate error response.
  */
 __cold
-void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
+void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function,
 		       unsigned int line, int errno, const char *fmt, ...)
 {
 	struct super_block *sb = fs_info->sb;
@@ -170,8 +161,13 @@ void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function,
 	}
 #endif
 
+	/*
+	 * Today we only save the error info to memory.  Long term we'll
+	 * also send it down to the disk
+	 */
+	set_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state);
+
 	/* Don't go through full error handling during mount */
-	save_error_info(fs_info);
 	if (sb->s_flags & MS_BORN)
 		btrfs_handle_error(fs_info);
 }
@@ -252,7 +248,7 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans,
 	/* Wake up anybody who may be waiting on this transaction */
 	wake_up(&root->fs_info->transaction_wait);
 	wake_up(&root->fs_info->transaction_blocked_wait);
-	__btrfs_std_error(root->fs_info, function, line, errno, NULL);
+	__btrfs_handle_fs_error(root->fs_info, function, line, errno, NULL);
 }
 /*
  * __btrfs_panic decodes unexpected, fatal errors from the caller,
@@ -1160,7 +1156,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
 		return 0;
 	}
 
-	btrfs_wait_ordered_roots(fs_info, -1);
+	btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
 
 	trans = btrfs_attach_transaction_barrier(root);
 	if (IS_ERR(trans)) {
@@ -1488,10 +1484,10 @@ static int setup_security_options(struct btrfs_fs_info *fs_info,
 		memcpy(&fs_info->security_opts, sec_opts, sizeof(*sec_opts));
 	} else {
 		/*
-		 * Since SELinux(the only one supports security_mnt_opts) does
-		 * NOT support changing context during remount/mount same sb,
-		 * This must be the same or part of the same security options,
-		 * just free it.
+		 * Since SELinux (the only one supporting security_mnt_opts)
+		 * does NOT support changing context during remount/mount of
+		 * the same sb, this must be the same or part of the same
+		 * security options, just free it.
 		 */
 		security_free_mnt_opts(sec_opts);
 	}
@@ -1669,8 +1665,8 @@ static inline void btrfs_remount_cleanup(struct btrfs_fs_info *fs_info,
 					 unsigned long old_opts)
 {
 	/*
-	 * We need cleanup all defragable inodes if the autodefragment is
-	 * close or the fs is R/O.
+	 * We need to cleanup all defragable inodes if the autodefragment is
+	 * close or the filesystem is read only.
 	 */
 	if (btrfs_raw_test_opt(old_opts, AUTO_DEFRAG) &&
 	    (!btrfs_raw_test_opt(fs_info->mount_opt, AUTO_DEFRAG) ||
@@ -2051,9 +2047,10 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
 	int ret;
 	u64 thresh = 0;
+	int mixed = 0;
 
 	/*
-	 * holding chunk_muext to avoid allocating new chunks, holding
+	 * holding chunk_mutex to avoid allocating new chunks, holding
 	 * device_list_mutex to avoid the device being removed
 	 */
 	rcu_read_lock();
@@ -2076,8 +2073,17 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 				}
 			}
 		}
-		if (found->flags & BTRFS_BLOCK_GROUP_METADATA)
-			total_free_meta += found->disk_total - found->disk_used;
+
+		/*
+		 * Metadata in mixed block goup profiles are accounted in data
+		 */
+		if (!mixed && found->flags & BTRFS_BLOCK_GROUP_METADATA) {
+			if (found->flags & BTRFS_BLOCK_GROUP_DATA)
+				mixed = 1;
+			else
+				total_free_meta += found->disk_total -
+					found->disk_used;
+		}
 
 		total_used += found->disk_used;
 	}
@@ -2090,7 +2096,11 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 
 	/* Account global block reserve as used, it's in logical size already */
 	spin_lock(&block_rsv->lock);
-	buf->f_bfree -= block_rsv->size >> bits;
+	/* Mixed block groups accounting is not byte-accurate, avoid overflow */
+	if (buf->f_bfree >= block_rsv->size >> bits)
+		buf->f_bfree -= block_rsv->size >> bits;
+	else
+		buf->f_bfree = 0;
 	spin_unlock(&block_rsv->lock);
 
 	buf->f_bavail = div_u64(total_free_data, factor);
@@ -2115,7 +2125,7 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	 */
 	thresh = 4 * 1024 * 1024;
 
-	if (total_free_meta - thresh < block_rsv->size)
+	if (!mixed && total_free_meta - thresh < block_rsv->size)
 		buf->f_bavail = 0;
 
 	buf->f_type = BTRFS_SUPER_MAGIC;

fs/btrfs/sysfs.c

@@ -120,6 +120,9 @@ static ssize_t btrfs_feature_attr_store(struct kobject *kobj,
 	if (!fs_info)
 		return -EPERM;
 
+	if (fs_info->sb->s_flags & MS_RDONLY)
+		return -EROFS;
+
 	ret = kstrtoul(skip_spaces(buf), 0, &val);
 	if (ret)
 		return ret;
@@ -364,7 +367,13 @@ static ssize_t btrfs_label_show(struct kobject *kobj,
 {
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 	char *label = fs_info->super_copy->label;
-	return snprintf(buf, PAGE_SIZE, label[0] ? "%s\n" : "%s", label);
+	ssize_t ret;
+
+	spin_lock(&fs_info->super_lock);
+	ret = snprintf(buf, PAGE_SIZE, label[0] ? "%s\n" : "%s", label);
+	spin_unlock(&fs_info->super_lock);
+
+	return ret;
 }
 
 static ssize_t btrfs_label_store(struct kobject *kobj,
@@ -374,6 +383,9 @@ static ssize_t btrfs_label_store(struct kobject *kobj,
 	struct btrfs_fs_info *fs_info = to_fs_info(kobj);
 	size_t p_len;
 
+	if (!fs_info)
+		return -EPERM;
+
 	if (fs_info->sb->s_flags & MS_RDONLY)
 		return -EROFS;
 

fs/btrfs/transaction.c

@@ -311,10 +311,11 @@ static noinline int join_transaction(struct btrfs_root *root, unsigned int type)
  * when the transaction commits
  */
 static int record_root_in_trans(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root)
+			       struct btrfs_root *root,
+			       int force)
 {
-	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
-	    root->last_trans < trans->transid) {
+	if ((test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
+	    root->last_trans < trans->transid) || force) {
 		WARN_ON(root == root->fs_info->extent_root);
 		WARN_ON(root->commit_root != root->node);
 
@@ -331,7 +332,7 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans,
 		smp_wmb();
 
 		spin_lock(&root->fs_info->fs_roots_radix_lock);
-		if (root->last_trans == trans->transid) {
+		if (root->last_trans == trans->transid && !force) {
 			spin_unlock(&root->fs_info->fs_roots_radix_lock);
 			return 0;
 		}
@@ -402,7 +403,7 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
 		return 0;
 
 	mutex_lock(&root->fs_info->reloc_mutex);
-	record_root_in_trans(trans, root);
+	record_root_in_trans(trans, root, 0);
 	mutex_unlock(&root->fs_info->reloc_mutex);
 
 	return 0;
@@ -1310,6 +1311,97 @@ 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.
+ *
+ * Will do all needed qgroup inherit and dirty hack like switch commit
+ * roots inside one transaction and write all btree into disk, to make
+ * qgroup works.
+ */
+static int qgroup_account_snapshot(struct btrfs_trans_handle *trans,
+				   struct btrfs_root *src,
+				   struct btrfs_root *parent,
+				   struct btrfs_qgroup_inherit *inherit,
+				   u64 dst_objectid)
+{
+	struct btrfs_fs_info *fs_info = src->fs_info;
+	int ret;
+
+	/*
+	 * Save some performance in the case that qgroups are not
+	 * enabled. If this check races with the ioctl, rescan will
+	 * kick in anyway.
+	 */
+	mutex_lock(&fs_info->qgroup_ioctl_lock);
+	if (!fs_info->quota_enabled) {
+		mutex_unlock(&fs_info->qgroup_ioctl_lock);
+		return 0;
+	}
+	mutex_unlock(&fs_info->qgroup_ioctl_lock);
+
+	/*
+	 * We are going to commit transaction, see btrfs_commit_transaction()
+	 * comment for reason locking tree_log_mutex
+	 */
+	mutex_lock(&fs_info->tree_log_mutex);
+
+	ret = commit_fs_roots(trans, src);
+	if (ret)
+		goto out;
+	ret = btrfs_qgroup_prepare_account_extents(trans, fs_info);
+	if (ret < 0)
+		goto out;
+	ret = btrfs_qgroup_account_extents(trans, fs_info);
+	if (ret < 0)
+		goto out;
+
+	/* Now qgroup are all updated, we can inherit it to new qgroups */
+	ret = btrfs_qgroup_inherit(trans, fs_info,
+				   src->root_key.objectid, dst_objectid,
+				   inherit);
+	if (ret < 0)
+		goto out;
+
+	/*
+	 * Now we do a simplified commit transaction, which will:
+	 * 1) commit all subvolume and extent tree
+	 *    To ensure all subvolume and extent tree have a valid
+	 *    commit_root to accounting later insert_dir_item()
+	 * 2) write all btree blocks onto disk
+	 *    This is to make sure later btree modification will be cowed
+	 *    Or commit_root can be populated and cause wrong qgroup numbers
+	 * In this simplified commit, we don't really care about other trees
+	 * like chunk and root tree, as they won't affect qgroup.
+	 * And we don't write super to avoid half committed status.
+	 */
+	ret = commit_cowonly_roots(trans, src);
+	if (ret)
+		goto out;
+	switch_commit_roots(trans->transaction, fs_info);
+	ret = btrfs_write_and_wait_transaction(trans, src);
+	if (ret)
+		btrfs_std_error(fs_info, ret,
+			"Error while writing out transaction for qgroup");
+
+out:
+	mutex_unlock(&fs_info->tree_log_mutex);
+
+	/*
+	 * Force parent root to be updated, as we recorded it before so its
+	 * last_trans == cur_transid.
+	 * Or it won't be committed again onto disk after later
+	 * insert_dir_item()
+	 */
+	if (!ret)
+		record_root_in_trans(trans, parent, 1);
+	return ret;
+}
+
 /*
  * new snapshots need to be created at a very specific time in the
  * transaction commit.  This does the actual creation.
@@ -1383,7 +1475,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	dentry = pending->dentry;
 	parent_inode = pending->dir;
 	parent_root = BTRFS_I(parent_inode)->root;
-	record_root_in_trans(trans, parent_root);
+	record_root_in_trans(trans, parent_root, 0);
 
 	cur_time = current_fs_time(parent_inode->i_sb);
 
@@ -1420,7 +1512,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 		goto fail;
 	}
 
-	record_root_in_trans(trans, root);
+	record_root_in_trans(trans, root, 0);
 	btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
 	memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));
 	btrfs_check_and_init_root_item(new_root_item);
@@ -1516,6 +1608,17 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 		goto fail;
 	}
 
+	/*
+	 * Do special qgroup accounting for snapshot, as we do some qgroup
+	 * snapshot hack to do fast snapshot.
+	 * To co-operate with that hack, we do hack again.
+	 * Or snapshot will be greatly slowed down by a subtree qgroup rescan
+	 */
+	ret = qgroup_account_snapshot(trans, root, parent_root,
+				      pending->inherit, objectid);
+	if (ret < 0)
+		goto fail;
+
 	ret = btrfs_insert_dir_item(trans, parent_root,
 				    dentry->d_name.name, dentry->d_name.len,
 				    parent_inode, &key,
@@ -1559,23 +1662,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 		goto fail;
 	}
 
-	/*
-	 * account qgroup counters before qgroup_inherit()
-	 */
-	ret = btrfs_qgroup_prepare_account_extents(trans, fs_info);
-	if (ret)
-		goto fail;
-	ret = btrfs_qgroup_account_extents(trans, fs_info);
-	if (ret)
-		goto fail;
-	ret = btrfs_qgroup_inherit(trans, fs_info,
-				   root->root_key.objectid,
-				   objectid, pending->inherit);
-	if (ret) {
-		btrfs_abort_transaction(trans, root, ret);
-		goto fail;
-	}
-
 fail:
 	pending->error = ret;
 dir_item_existed:
@@ -1821,7 +1907,7 @@ static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
 static inline void btrfs_wait_delalloc_flush(struct btrfs_fs_info *fs_info)
 {
 	if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
-		btrfs_wait_ordered_roots(fs_info, -1);
+		btrfs_wait_ordered_roots(fs_info, -1, 0, (u64)-1);
 }
 
 static inline void
@@ -2145,7 +2231,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_write_and_wait_transaction(trans, root);
 	if (ret) {
-		btrfs_std_error(root->fs_info, ret,
+		btrfs_handle_fs_error(root->fs_info, ret,
 			    "Error while writing out transaction");
 		mutex_unlock(&root->fs_info->tree_log_mutex);
 		goto scrub_continue;

fs/btrfs/tree-log.c

@@ -4141,6 +4141,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 
 	INIT_LIST_HEAD(&extents);
 
+	down_write(&BTRFS_I(inode)->dio_sem);
 	write_lock(&tree->lock);
 	test_gen = root->fs_info->last_trans_committed;
 
@@ -4169,13 +4170,20 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 	}
 
 	list_sort(NULL, &extents, extent_cmp);
+	btrfs_get_logged_extents(inode, logged_list, start, end);
 	/*
-	 * Collect any new ordered extents within the range. This is to
-	 * prevent logging file extent items without waiting for the disk
-	 * location they point to being written. We do this only to deal
-	 * with races against concurrent lockless direct IO writes.
+	 * Some ordered extents started by fsync might have completed
+	 * before we could collect them into the list logged_list, which
+	 * means they're gone, not in our logged_list nor in the inode's
+	 * ordered tree. We want the application/user space to know an
+	 * error happened while attempting to persist file data so that
+	 * it can take proper action. If such error happened, we leave
+	 * without writing to the log tree and the fsync must report the
+	 * file data write error and not commit the current transaction.
 	 */
-	btrfs_get_logged_extents(inode, logged_list, start, end);
+	ret = btrfs_inode_check_errors(inode);
+	if (ret)
+		ctx->io_err = ret;
 process:
 	while (!list_empty(&extents)) {
 		em = list_entry(extents.next, struct extent_map, list);
@@ -4202,6 +4210,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 	}
 	WARN_ON(!list_empty(&extents));
 	write_unlock(&tree->lock);
+	up_write(&BTRFS_I(inode)->dio_sem);
 
 	btrfs_release_path(path);
 	return ret;
@@ -4622,23 +4631,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 
 	mutex_lock(&BTRFS_I(inode)->log_mutex);
 
-	/*
-	 * Collect ordered extents only if we are logging data. This is to
-	 * ensure a subsequent request to log this inode in LOG_INODE_ALL mode
-	 * will process the ordered extents if they still exists at the time,
-	 * because when we collect them we test and set for the flag
-	 * BTRFS_ORDERED_LOGGED to prevent multiple log requests to process the
-	 * same ordered extents. The consequence for the LOG_INODE_ALL log mode
-	 * not processing the ordered extents is that we end up logging the
-	 * corresponding file extent items, based on the extent maps in the
-	 * inode's extent_map_tree's modified_list, without logging the
-	 * respective checksums (since the may still be only attached to the
-	 * ordered extents and have not been inserted in the csum tree by
-	 * btrfs_finish_ordered_io() yet).
-	 */
-	if (inode_only == LOG_INODE_ALL)
-		btrfs_get_logged_extents(inode, &logged_list, start, end);
-
 	/*
 	 * a brute force approach to making sure we get the most uptodate
 	 * copies of everything.
@@ -4846,21 +4838,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
 			goto out_unlock;
 	}
 	if (fast_search) {
-		/*
-		 * Some ordered extents started by fsync might have completed
-		 * before we collected the ordered extents in logged_list, which
-		 * means they're gone, not in our logged_list nor in the inode's
-		 * ordered tree. We want the application/user space to know an
-		 * error happened while attempting to persist file data so that
-		 * it can take proper action. If such error happened, we leave
-		 * without writing to the log tree and the fsync must report the
-		 * file data write error and not commit the current transaction.
-		 */
-		err = btrfs_inode_check_errors(inode);
-		if (err) {
-			ctx->io_err = err;
-			goto out_unlock;
-		}
 		ret = btrfs_log_changed_extents(trans, root, inode, dst_path,
 						&logged_list, ctx, start, end);
 		if (ret) {
@@ -5158,7 +5135,7 @@ static int log_new_dir_dentries(struct btrfs_trans_handle *trans,
 			}
 
 			ctx->log_new_dentries = false;
-			if (type == BTRFS_FT_DIR)
+			if (type == BTRFS_FT_DIR || type == BTRFS_FT_SYMLINK)
 				log_mode = LOG_INODE_ALL;
 			btrfs_release_path(path);
 			ret = btrfs_log_inode(trans, root, di_inode,
@@ -5278,11 +5255,16 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans,
 			if (IS_ERR(dir_inode))
 				continue;
 
+			if (ctx)
+				ctx->log_new_dentries = false;
 			ret = btrfs_log_inode(trans, root, dir_inode,
 					      LOG_INODE_ALL, 0, LLONG_MAX, ctx);
 			if (!ret &&
 			    btrfs_must_commit_transaction(trans, dir_inode))
 				ret = 1;
+			if (!ret && ctx && ctx->log_new_dentries)
+				ret = log_new_dir_dentries(trans, root,
+							   dir_inode, ctx);
 			iput(dir_inode);
 			if (ret)
 				goto out;
@@ -5519,7 +5501,7 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 
 	ret = walk_log_tree(trans, log_root_tree, &wc);
 	if (ret) {
-		btrfs_std_error(fs_info, ret, "Failed to pin buffers while "
+		btrfs_handle_fs_error(fs_info, ret, "Failed to pin buffers while "
 			    "recovering log root tree.");
 		goto error;
 	}
@@ -5533,7 +5515,7 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 		ret = btrfs_search_slot(NULL, log_root_tree, &key, path, 0, 0);
 
 		if (ret < 0) {
-			btrfs_std_error(fs_info, ret,
+			btrfs_handle_fs_error(fs_info, ret,
 				    "Couldn't find tree log root.");
 			goto error;
 		}
@@ -5551,7 +5533,7 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 		log = btrfs_read_fs_root(log_root_tree, &found_key);
 		if (IS_ERR(log)) {
 			ret = PTR_ERR(log);
-			btrfs_std_error(fs_info, ret,
+			btrfs_handle_fs_error(fs_info, ret,
 				    "Couldn't read tree log root.");
 			goto error;
 		}
@@ -5566,7 +5548,7 @@ int btrfs_recover_log_trees(struct btrfs_root *log_root_tree)
 			free_extent_buffer(log->node);
 			free_extent_buffer(log->commit_root);
 			kfree(log);
-			btrfs_std_error(fs_info, ret, "Couldn't read target root "
+			btrfs_handle_fs_error(fs_info, ret, "Couldn't read target root "
 				    "for tree log recovery.");
 			goto error;
 		}
@@ -5652,11 +5634,9 @@ void btrfs_record_unlink_dir(struct btrfs_trans_handle *trans,
 	 * into the file.  When the file is logged we check it and
 	 * don't log the parents if the file is fully on disk.
 	 */
-	if (S_ISREG(inode->i_mode)) {
-		mutex_lock(&BTRFS_I(inode)->log_mutex);
-		BTRFS_I(inode)->last_unlink_trans = trans->transid;
-		mutex_unlock(&BTRFS_I(inode)->log_mutex);
-	}
+	mutex_lock(&BTRFS_I(inode)->log_mutex);
+	BTRFS_I(inode)->last_unlink_trans = trans->transid;
+	mutex_unlock(&BTRFS_I(inode)->log_mutex);
 
 	/*
 	 * if this directory was already logged any new