Merge tag 'for-5.19-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "Features:

   - subpage:
      - support for PAGE_SIZE > 4K (previously only 64K)
      - make it work with raid56

   - repair super block num_devices automatically if it does not match
     the number of device items

   - defrag can convert inline extents to regular extents, up to now
     inline files were skipped but the setting of mount option
     max_inline could affect the decision logic

   - zoned:
      - minimal accepted zone size is explicitly set to 4MiB
      - make zone reclaim less aggressive and don't reclaim if there are
        enough free zones
      - add per-profile sysfs tunable of the reclaim threshold

   - allow automatic block group reclaim for non-zoned filesystems, with
     sysfs tunables

   - tree-checker: new check, compare extent buffer owner against owner
     rootid

  Performance:

   - avoid blocking on space reservation when doing nowait direct io
     writes (+7% throughput for reads and writes)

   - NOCOW write throughput improvement due to refined locking (+3%)

   - send: reduce pressure to page cache by dropping extent pages right
     after they're processed

  Core:

   - convert all radix trees to xarray

   - add iterators for b-tree node items

   - support printk message index

   - user bulk page allocation for extent buffers

   - switch to bio_alloc API, use on-stack bios where convenient, other
     bio cleanups

   - use rw lock for block groups to favor concurrent reads

   - simplify workques, don't allocate high priority threads for all
     normal queues as we need only one

   - refactor scrub, process chunks based on their constraints and
     similarity

   - allocate direct io structures on stack and pass around only
     pointers, avoids allocation and reduces potential error handling

  Fixes:

   - fix count of reserved transaction items for various inode
     operations

   - fix deadlock between concurrent dio writes when low on free data
     space

   - fix a few cases when zones need to be finished

  VFS, iomap:

   - add helper to check if sb write has started (usable for assertions)

   - new helper iomap_dio_alloc_bio, export iomap_dio_bio_end_io"

* tag 'for-5.19-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (173 commits)
  btrfs: zoned: introduce a minimal zone size 4M and reject mount
  btrfs: allow defrag to convert inline extents to regular extents
  btrfs: add "0x" prefix for unsupported optional features
  btrfs: do not account twice for inode ref when reserving metadata units
  btrfs: zoned: fix comparison of alloc_offset vs meta_write_pointer
  btrfs: send: avoid trashing the page cache
  btrfs: send: keep the current inode open while processing it
  btrfs: allocate the btrfs_dio_private as part of the iomap dio bio
  btrfs: move struct btrfs_dio_private to inode.c
  btrfs: remove the disk_bytenr in struct btrfs_dio_private
  btrfs: allocate dio_data on stack
  iomap: add per-iomap_iter private data
  iomap: allow the file system to provide a bio_set for direct I/O
  btrfs: add a btrfs_dio_rw wrapper
  btrfs: zoned: zone finish unused block group
  btrfs: zoned: properly finish block group on metadata write
  btrfs: zoned: finish block group when there are no more allocatable bytes left
  btrfs: zoned: consolidate zone finish functions
  btrfs: zoned: introduce btrfs_zoned_bg_is_full
  btrfs: improve error reporting in lookup_inline_extent_backref
  ...

67 files changed, 4472 insertions, 4211 deletions

diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 0a0d0eccee4e..548d6a5477b4 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -55,9 +55,8 @@ struct posix_acl *btrfs_get_acl(struct inode *inode, int type, bool rcu)
 	return acl;
 }
 
-static int __btrfs_set_acl(struct btrfs_trans_handle *trans,
-			   struct user_namespace *mnt_userns,
-			   struct inode *inode, struct posix_acl *acl, int type)
+int __btrfs_set_acl(struct btrfs_trans_handle *trans, struct inode *inode,
+		    struct posix_acl *acl, int type)
 {
 	int ret, size = 0;
 	const char *name;
@@ -123,40 +122,8 @@ int btrfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode,
 		if (ret)
 			return ret;
 	}
-	ret = __btrfs_set_acl(NULL, mnt_userns, inode, acl, type);
+	ret = __btrfs_set_acl(NULL, inode, acl, type);
 	if (ret)
 		inode->i_mode = old_mode;
 	return ret;
 }
-
-int btrfs_init_acl(struct btrfs_trans_handle *trans,
-		   struct inode *inode, struct inode *dir)
-{
-	struct posix_acl *default_acl, *acl;
-	int ret = 0;
-
-	/* this happens with subvols */
-	if (!dir)
-		return 0;
-
-	ret = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
-	if (ret)
-		return ret;
-
-	if (default_acl) {
-		ret = __btrfs_set_acl(trans, &init_user_ns, inode, default_acl,
-				      ACL_TYPE_DEFAULT);
-		posix_acl_release(default_acl);
-	}
-
-	if (acl) {
-		if (!ret)
-			ret = __btrfs_set_acl(trans, &init_user_ns, inode, acl,
-					      ACL_TYPE_ACCESS);
-		posix_acl_release(acl);
-	}
-
-	if (!default_acl && !acl)
-		cache_no_acl(inode);
-	return ret;
-}
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 43c89952b7d2..aac240430efe 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -15,13 +15,12 @@
 enum {
 	WORK_DONE_BIT,
 	WORK_ORDER_DONE_BIT,
-	WORK_HIGH_PRIO_BIT,
 };
 
 #define NO_THRESHOLD (-1)
 #define DFT_THRESHOLD (32)
 
-struct __btrfs_workqueue {
+struct btrfs_workqueue {
 	struct workqueue_struct *normal_wq;
 
 	/* File system this workqueue services */
@@ -48,12 +47,7 @@ struct __btrfs_workqueue {
 	spinlock_t thres_lock;
 };
 
-struct btrfs_workqueue {
-	struct __btrfs_workqueue *normal;
-	struct __btrfs_workqueue *high;
-};
-
-struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct __btrfs_workqueue *wq)
+struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct btrfs_workqueue *wq)
 {
 	return wq->fs_info;
 }
@@ -66,22 +60,22 @@ struct btrfs_fs_info * __pure btrfs_work_owner(const struct btrfs_work *work)
 bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq)
 {
 	/*
-	 * We could compare wq->normal->pending with num_online_cpus()
+	 * We could compare wq->pending with num_online_cpus()
 	 * to support "thresh == NO_THRESHOLD" case, but it requires
 	 * moving up atomic_inc/dec in thresh_queue/exec_hook. Let's
 	 * postpone it until someone needs the support of that case.
 	 */
-	if (wq->normal->thresh == NO_THRESHOLD)
+	if (wq->thresh == NO_THRESHOLD)
 		return false;
 
-	return atomic_read(&wq->normal->pending) > wq->normal->thresh * 2;
+	return atomic_read(&wq->pending) > wq->thresh * 2;
 }
 
-static struct __btrfs_workqueue *
-__btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info, const char *name,
-			unsigned int flags, int limit_active, int thresh)
+struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info,
+					      const char *name, unsigned int flags,
+					      int limit_active, int thresh)
 {
-	struct __btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_KERNEL);
+	struct btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_KERNEL);
 
 	if (!ret)
 		return NULL;
@@ -105,12 +99,8 @@ __btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info, const char *name,
 		ret->thresh = thresh;
 	}
 
-	if (flags & WQ_HIGHPRI)
-		ret->normal_wq = alloc_workqueue("btrfs-%s-high", flags,
-						 ret->current_active, name);
-	else
-		ret->normal_wq = alloc_workqueue("btrfs-%s", flags,
-						 ret->current_active, name);
+	ret->normal_wq = alloc_workqueue("btrfs-%s", flags, ret->current_active,
+					 name);
 	if (!ret->normal_wq) {
 		kfree(ret);
 		return NULL;
@@ -119,41 +109,7 @@ __btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info, const char *name,
 	INIT_LIST_HEAD(&ret->ordered_list);
 	spin_lock_init(&ret->list_lock);
 	spin_lock_init(&ret->thres_lock);
-	trace_btrfs_workqueue_alloc(ret, name, flags & WQ_HIGHPRI);
-	return ret;
-}
-
-static inline void
-__btrfs_destroy_workqueue(struct __btrfs_workqueue *wq);
-
-struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info,
-					      const char *name,
-					      unsigned int flags,
-					      int limit_active,
-					      int thresh)
-{
-	struct btrfs_workqueue *ret = kzalloc(sizeof(*ret), GFP_KERNEL);
-
-	if (!ret)
-		return NULL;
-
-	ret->normal = __btrfs_alloc_workqueue(fs_info, name,
-					      flags & ~WQ_HIGHPRI,
-					      limit_active, thresh);
-	if (!ret->normal) {
-		kfree(ret);
-		return NULL;
-	}
-
-	if (flags & WQ_HIGHPRI) {
-		ret->high = __btrfs_alloc_workqueue(fs_info, name, flags,
-						    limit_active, thresh);
-		if (!ret->high) {
-			__btrfs_destroy_workqueue(ret->normal);
-			kfree(ret);
-			return NULL;
-		}
-	}
+	trace_btrfs_workqueue_alloc(ret, name);
 	return ret;
 }
 
@@ -162,7 +118,7 @@ struct btrfs_workqueue *btrfs_alloc_workqueue(struct btrfs_fs_info *fs_info,
  * This hook WILL be called in IRQ handler context,
  * so workqueue_set_max_active MUST NOT be called in this hook
  */
-static inline void thresh_queue_hook(struct __btrfs_workqueue *wq)
+static inline void thresh_queue_hook(struct btrfs_workqueue *wq)
 {
 	if (wq->thresh == NO_THRESHOLD)
 		return;
@@ -174,7 +130,7 @@ static inline void thresh_queue_hook(struct __btrfs_workqueue *wq)
  * This hook is called in kthread content.
  * So workqueue_set_max_active is called here.
  */
-static inline void thresh_exec_hook(struct __btrfs_workqueue *wq)
+static inline void thresh_exec_hook(struct btrfs_workqueue *wq)
 {
 	int new_current_active;
 	long pending;
@@ -217,7 +173,7 @@ out:
 	}
 }
 
-static void run_ordered_work(struct __btrfs_workqueue *wq,
+static void run_ordered_work(struct btrfs_workqueue *wq,
 			     struct btrfs_work *self)
 {
 	struct list_head *list = &wq->ordered_list;
@@ -305,7 +261,7 @@ static void btrfs_work_helper(struct work_struct *normal_work)
 {
 	struct btrfs_work *work = container_of(normal_work, struct btrfs_work,
 					       normal_work);
-	struct __btrfs_workqueue *wq;
+	struct btrfs_workqueue *wq = work->wq;
 	int need_order = 0;
 
 	/*
@@ -318,7 +274,6 @@ static void btrfs_work_helper(struct work_struct *normal_work)
 	 */
 	if (work->ordered_func)
 		need_order = 1;
-	wq = work->wq;
 
 	trace_btrfs_work_sched(work);
 	thresh_exec_hook(wq);
@@ -350,8 +305,7 @@ void btrfs_init_work(struct btrfs_work *work, btrfs_func_t func,
 	work->flags = 0;
 }
 
-static inline void __btrfs_queue_work(struct __btrfs_workqueue *wq,
-				      struct btrfs_work *work)
+void btrfs_queue_work(struct btrfs_workqueue *wq, struct btrfs_work *work)
 {
 	unsigned long flags;
 
@@ -366,54 +320,22 @@ static inline void __btrfs_queue_work(struct __btrfs_workqueue *wq,
 	queue_work(wq->normal_wq, &work->normal_work);
 }
 
-void btrfs_queue_work(struct btrfs_workqueue *wq,
-		      struct btrfs_work *work)
-{
-	struct __btrfs_workqueue *dest_wq;
-
-	if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags) && wq->high)
-		dest_wq = wq->high;
-	else
-		dest_wq = wq->normal;
-	__btrfs_queue_work(dest_wq, work);
-}
-
-static inline void
-__btrfs_destroy_workqueue(struct __btrfs_workqueue *wq)
-{
-	destroy_workqueue(wq->normal_wq);
-	trace_btrfs_workqueue_destroy(wq);
-	kfree(wq);
-}
-
 void btrfs_destroy_workqueue(struct btrfs_workqueue *wq)
 {
 	if (!wq)
 		return;
-	if (wq->high)
-		__btrfs_destroy_workqueue(wq->high);
-	__btrfs_destroy_workqueue(wq->normal);
+	destroy_workqueue(wq->normal_wq);
+	trace_btrfs_workqueue_destroy(wq);
 	kfree(wq);
 }
 
 void btrfs_workqueue_set_max(struct btrfs_workqueue *wq, int limit_active)
 {
-	if (!wq)
-		return;
-	wq->normal->limit_active = limit_active;
-	if (wq->high)
-		wq->high->limit_active = limit_active;
-}
-
-void btrfs_set_work_high_priority(struct btrfs_work *work)
-{
-	set_bit(WORK_HIGH_PRIO_BIT, &work->flags);
+	if (wq)
+		wq->limit_active = limit_active;
 }
 
 void btrfs_flush_workqueue(struct btrfs_workqueue *wq)
 {
-	if (wq->high)
-		flush_workqueue(wq->high->normal_wq);
-
-	flush_workqueue(wq->normal->normal_wq);
+	flush_workqueue(wq->normal_wq);
 }
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 3204daa51b95..07960529b360 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -11,8 +11,6 @@
 
 struct btrfs_fs_info;
 struct btrfs_workqueue;
-/* Internal use only */
-struct __btrfs_workqueue;
 struct btrfs_work;
 typedef void (*btrfs_func_t)(struct btrfs_work *arg);
 typedef void (*btrfs_work_func_t)(struct work_struct *arg);
@@ -25,7 +23,7 @@ struct btrfs_work {
 	/* Don't touch things below */
 	struct work_struct normal_work;
 	struct list_head ordered_list;
-	struct __btrfs_workqueue *wq;
+	struct btrfs_workqueue *wq;
 	unsigned long flags;
 };
 
@@ -40,9 +38,8 @@ void btrfs_queue_work(struct btrfs_workqueue *wq,
 		      struct btrfs_work *work);
 void btrfs_destroy_workqueue(struct btrfs_workqueue *wq);
 void btrfs_workqueue_set_max(struct btrfs_workqueue *wq, int max);
-void btrfs_set_work_high_priority(struct btrfs_work *work);
 struct btrfs_fs_info * __pure btrfs_work_owner(const struct btrfs_work *work);
-struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct __btrfs_workqueue *wq);
+struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct btrfs_workqueue *wq);
 bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq);
 void btrfs_flush_workqueue(struct btrfs_workqueue *wq);
 
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 0dd6de994199..ede389f2602d 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -168,11 +168,12 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
 	struct rb_node **p;
 	struct rb_node *parent = NULL;
 	struct btrfs_block_group *cache;
+	bool leftmost = true;
 
 	ASSERT(block_group->length != 0);
 
-	spin_lock(&info->block_group_cache_lock);
-	p = &info->block_group_cache_tree.rb_node;
+	write_lock(&info->block_group_cache_lock);
+	p = &info->block_group_cache_tree.rb_root.rb_node;
 
 	while (*p) {
 		parent = *p;
@@ -181,20 +182,18 @@ static int btrfs_add_block_group_cache(struct btrfs_fs_info *info,
 			p = &(*p)->rb_left;
 		} else if (block_group->start > cache->start) {
 			p = &(*p)->rb_right;
+			leftmost = false;
 		} else {
-			spin_unlock(&info->block_group_cache_lock);
+			write_unlock(&info->block_group_cache_lock);
 			return -EEXIST;
 		}
 	}
 
 	rb_link_node(&block_group->cache_node, parent, p);
-	rb_insert_color(&block_group->cache_node,
-			&info->block_group_cache_tree);
+	rb_insert_color_cached(&block_group->cache_node,
+			       &info->block_group_cache_tree, leftmost);
 
-	if (info->first_logical_byte > block_group->start)
-		info->first_logical_byte = block_group->start;
-
-	spin_unlock(&info->block_group_cache_lock);
+	write_unlock(&info->block_group_cache_lock);
 
 	return 0;
 }
@@ -210,8 +209,8 @@ static struct btrfs_block_group *block_group_cache_tree_search(
 	struct rb_node *n;
 	u64 end, start;
 
-	spin_lock(&info->block_group_cache_lock);
-	n = info->block_group_cache_tree.rb_node;
+	read_lock(&info->block_group_cache_lock);
+	n = info->block_group_cache_tree.rb_root.rb_node;
 
 	while (n) {
 		cache = rb_entry(n, struct btrfs_block_group, cache_node);
@@ -233,12 +232,9 @@ static struct btrfs_block_group *block_group_cache_tree_search(
 			break;
 		}
 	}
-	if (ret) {
+	if (ret)
 		btrfs_get_block_group(ret);
-		if (bytenr == 0 && info->first_logical_byte > ret->start)
-			info->first_logical_byte = ret->start;
-	}
-	spin_unlock(&info->block_group_cache_lock);
+	read_unlock(&info->block_group_cache_lock);
 
 	return ret;
 }
@@ -267,15 +263,15 @@ struct btrfs_block_group *btrfs_next_block_group(
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	struct rb_node *node;
 
-	spin_lock(&fs_info->block_group_cache_lock);
+	read_lock(&fs_info->block_group_cache_lock);
 
 	/* If our block group was removed, we need a full search. */
 	if (RB_EMPTY_NODE(&cache->cache_node)) {
 		const u64 next_bytenr = cache->start + cache->length;
 
-		spin_unlock(&fs_info->block_group_cache_lock);
+		read_unlock(&fs_info->block_group_cache_lock);
 		btrfs_put_block_group(cache);
-		cache = btrfs_lookup_first_block_group(fs_info, next_bytenr); return cache;
+		return btrfs_lookup_first_block_group(fs_info, next_bytenr);
 	}
 	node = rb_next(&cache->cache_node);
 	btrfs_put_block_group(cache);
@@ -284,46 +280,70 @@ struct btrfs_block_group *btrfs_next_block_group(
 		btrfs_get_block_group(cache);
 	} else
 		cache = NULL;
-	spin_unlock(&fs_info->block_group_cache_lock);
+	read_unlock(&fs_info->block_group_cache_lock);
 	return cache;
 }
 
-bool btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
+/**
+ * Check if we can do a NOCOW write for a given extent.
+ *
+ * @fs_info:       The filesystem information object.
+ * @bytenr:        Logical start address of the extent.
+ *
+ * Check if we can do a NOCOW write for the given extent, and increments the
+ * number of NOCOW writers in the block group that contains the extent, as long
+ * as the block group exists and it's currently not in read-only mode.
+ *
+ * Returns: A non-NULL block group pointer if we can do a NOCOW write, the caller
+ *          is responsible for calling btrfs_dec_nocow_writers() later.
+ *
+ *          Or NULL if we can not do a NOCOW write
+ */
+struct btrfs_block_group *btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info,
+						  u64 bytenr)
 {
 	struct btrfs_block_group *bg;
-	bool ret = true;
+	bool can_nocow = true;
 
 	bg = btrfs_lookup_block_group(fs_info, bytenr);
 	if (!bg)
-		return false;
+		return NULL;
 
 	spin_lock(&bg->lock);
 	if (bg->ro)
-		ret = false;
+		can_nocow = false;
 	else
 		atomic_inc(&bg->nocow_writers);
 	spin_unlock(&bg->lock);
 
-	/* No put on block group, done by btrfs_dec_nocow_writers */
-	if (!ret)
+	if (!can_nocow) {
 		btrfs_put_block_group(bg);
+		return NULL;
+	}
 
-	return ret;
+	/* No put on block group, done by btrfs_dec_nocow_writers(). */
+	return bg;
 }
 
-void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr)
+/**
+ * Decrement the number of NOCOW writers in a block group.
+ *
+ * @bg:       The block group.
+ *
+ * This is meant to be called after a previous call to btrfs_inc_nocow_writers(),
+ * and on the block group returned by that call. Typically this is called after
+ * creating an ordered extent for a NOCOW write, to prevent races with scrub and
+ * relocation.
+ *
+ * After this call, the caller should not use the block group anymore. It it wants
+ * to use it, then it should get a reference on it before calling this function.
+ */
+void btrfs_dec_nocow_writers(struct btrfs_block_group *bg)
 {
-	struct btrfs_block_group *bg;
-
-	bg = btrfs_lookup_block_group(fs_info, bytenr);
-	ASSERT(bg);
 	if (atomic_dec_and_test(&bg->nocow_writers))
 		wake_up_var(&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);
+
+	/* For the lookup done by a previous call to btrfs_inc_nocow_writers(). */
 	btrfs_put_block_group(bg);
 }
 
@@ -772,10 +792,10 @@ int btrfs_cache_block_group(struct btrfs_block_group *cache, int load_cache_only
 	cache->has_caching_ctl = 1;
 	spin_unlock(&cache->lock);
 
-	spin_lock(&fs_info->block_group_cache_lock);
+	write_lock(&fs_info->block_group_cache_lock);
 	refcount_inc(&caching_ctl->count);
 	list_add_tail(&caching_ctl->list, &fs_info->caching_block_groups);
-	spin_unlock(&fs_info->block_group_cache_lock);
+	write_unlock(&fs_info->block_group_cache_lock);
 
 	btrfs_get_block_group(cache);
 
@@ -957,17 +977,15 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	if (ret)
 		goto out;
 
-	spin_lock(&fs_info->block_group_cache_lock);
-	rb_erase(&block_group->cache_node,
-		 &fs_info->block_group_cache_tree);
+	write_lock(&fs_info->block_group_cache_lock);
+	rb_erase_cached(&block_group->cache_node,
+			&fs_info->block_group_cache_tree);
 	RB_CLEAR_NODE(&block_group->cache_node);
 
 	/* Once for the block groups rbtree */
 	btrfs_put_block_group(block_group);
 
-	if (fs_info->first_logical_byte == block_group->start)
-		fs_info->first_logical_byte = (u64)-1;
-	spin_unlock(&fs_info->block_group_cache_lock);
+	write_unlock(&fs_info->block_group_cache_lock);
 
 	down_write(&block_group->space_info->groups_sem);
 	/*
@@ -992,7 +1010,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	if (block_group->cached == BTRFS_CACHE_STARTED)
 		btrfs_wait_block_group_cache_done(block_group);
 	if (block_group->has_caching_ctl) {
-		spin_lock(&fs_info->block_group_cache_lock);
+		write_lock(&fs_info->block_group_cache_lock);
 		if (!caching_ctl) {
 			struct btrfs_caching_control *ctl;
 
@@ -1006,7 +1024,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 		}
 		if (caching_ctl)
 			list_del_init(&caching_ctl->list);
-		spin_unlock(&fs_info->block_group_cache_lock);
+		write_unlock(&fs_info->block_group_cache_lock);
 		if (caching_ctl) {
 			/* Once for the caching bgs list and once for us. */
 			btrfs_put_caching_control(caching_ctl);
@@ -1367,6 +1385,14 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 			goto next;
 		}
 
+		ret = btrfs_zone_finish(block_group);
+		if (ret < 0) {
+			btrfs_dec_block_group_ro(block_group);
+			if (ret == -EAGAIN)
+				ret = 0;
+			goto next;
+		}
+
 		/*
 		 * Want to do this before we do anything else so we can recover
 		 * properly if we fail to join the transaction.
@@ -1512,6 +1538,13 @@ static int reclaim_bgs_cmp(void *unused, const struct list_head *a,
 	return bg1->used > bg2->used;
 }
 
+static inline bool btrfs_should_reclaim(struct btrfs_fs_info *fs_info)
+{
+	if (btrfs_is_zoned(fs_info))
+		return btrfs_zoned_should_reclaim(fs_info);
+	return true;
+}
+
 void btrfs_reclaim_bgs_work(struct work_struct *work)
 {
 	struct btrfs_fs_info *fs_info =
@@ -1522,6 +1555,9 @@ void btrfs_reclaim_bgs_work(struct work_struct *work)
 	if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags))
 		return;
 
+	if (!btrfs_should_reclaim(fs_info))
+		return;
+
 	sb_start_write(fs_info->sb);
 
 	if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE)) {
@@ -1692,35 +1728,13 @@ static int find_first_block_group(struct btrfs_fs_info *fs_info,
 	struct btrfs_root *root = btrfs_block_group_root(fs_info);
 	int ret;
 	struct btrfs_key found_key;
-	struct extent_buffer *leaf;
-	int slot;
-
-	ret = btrfs_search_slot(NULL, root, key, path, 0, 0);
-	if (ret < 0)
-		return ret;
-
-	while (1) {
-		slot = path->slots[0];
-		leaf = path->nodes[0];
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret == 0)
-				continue;
-			if (ret < 0)
-				goto out;
-			break;
-		}
-		btrfs_item_key_to_cpu(leaf, &found_key, slot);
 
+	btrfs_for_each_slot(root, key, &found_key, path, ret) {
 		if (found_key.objectid >= key->objectid &&
 		    found_key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) {
-			ret = read_bg_from_eb(fs_info, &found_key, path);
-			break;
+			return read_bg_from_eb(fs_info, &found_key, path);
 		}
-
-		path->slots[0]++;
 	}
-out:
 	return ret;
 }
 
@@ -3220,6 +3234,31 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 	return ret;
 }
 
+static inline bool should_reclaim_block_group(struct btrfs_block_group *bg,
+					      u64 bytes_freed)
+{
+	const struct btrfs_space_info *space_info = bg->space_info;
+	const int reclaim_thresh = READ_ONCE(space_info->bg_reclaim_threshold);
+	const u64 new_val = bg->used;
+	const u64 old_val = new_val + bytes_freed;
+	u64 thresh;
+
+	if (reclaim_thresh == 0)
+		return false;
+
+	thresh = div_factor_fine(bg->length, reclaim_thresh);
+
+	/*
+	 * If we were below the threshold before don't reclaim, we are likely a
+	 * brand new block group and we don't want to relocate new block groups.
+	 */
+	if (old_val < thresh)
+		return false;
+	if (new_val >= thresh)
+		return false;
+	return true;
+}
+
 int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 			     u64 bytenr, u64 num_bytes, bool alloc)
 {
@@ -3242,6 +3281,8 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 	spin_unlock(&info->delalloc_root_lock);
 
 	while (total) {
+		bool reclaim;
+
 		cache = btrfs_lookup_block_group(info, bytenr);
 		if (!cache) {
 			ret = -ENOENT;
@@ -3287,6 +3328,8 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 					cache->space_info, num_bytes);
 			cache->space_info->bytes_used -= num_bytes;
 			cache->space_info->disk_used -= num_bytes * factor;
+
+			reclaim = should_reclaim_block_group(cache, num_bytes);
 			spin_unlock(&cache->lock);
 			spin_unlock(&cache->space_info->lock);
 
@@ -3313,6 +3356,8 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 		if (!alloc && old_val == 0) {
 			if (!btrfs_test_opt(info, DISCARD_ASYNC))
 				btrfs_mark_bg_unused(cache);
+		} else if (!alloc && reclaim) {
+			btrfs_mark_bg_to_reclaim(cache);
 		}
 
 		btrfs_put_block_group(cache);
@@ -3957,14 +4002,14 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 	struct btrfs_caching_control *caching_ctl;
 	struct rb_node *n;
 
-	spin_lock(&info->block_group_cache_lock);
+	write_lock(&info->block_group_cache_lock);
 	while (!list_empty(&info->caching_block_groups)) {
 		caching_ctl = list_entry(info->caching_block_groups.next,
 					 struct btrfs_caching_control, list);
 		list_del(&caching_ctl->list);
 		btrfs_put_caching_control(caching_ctl);
 	}
-	spin_unlock(&info->block_group_cache_lock);
+	write_unlock(&info->block_group_cache_lock);
 
 	spin_lock(&info->unused_bgs_lock);
 	while (!list_empty(&info->unused_bgs)) {
@@ -3994,14 +4039,14 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 	}
 	spin_unlock(&info->zone_active_bgs_lock);
 
-	spin_lock(&info->block_group_cache_lock);
-	while ((n = rb_last(&info->block_group_cache_tree)) != NULL) {
+	write_lock(&info->block_group_cache_lock);
+	while ((n = rb_last(&info->block_group_cache_tree.rb_root)) != NULL) {
 		block_group = rb_entry(n, struct btrfs_block_group,
 				       cache_node);
-		rb_erase(&block_group->cache_node,
-			 &info->block_group_cache_tree);
+		rb_erase_cached(&block_group->cache_node,
+				&info->block_group_cache_tree);
 		RB_CLEAR_NODE(&block_group->cache_node);
-		spin_unlock(&info->block_group_cache_lock);
+		write_unlock(&info->block_group_cache_lock);
 
 		down_write(&block_group->space_info->groups_sem);
 		list_del(&block_group->list);
@@ -4024,9 +4069,9 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		ASSERT(block_group->swap_extents == 0);
 		btrfs_put_block_group(block_group);
 
-		spin_lock(&info->block_group_cache_lock);
+		write_lock(&info->block_group_cache_lock);
 	}
-	spin_unlock(&info->block_group_cache_lock);
+	write_unlock(&info->block_group_cache_lock);
 
 	btrfs_release_global_block_rsv(info);
 
diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h
index e8308f2ad07d..3ac668ace50a 100644
--- a/fs/btrfs/block-group.h
+++ b/fs/btrfs/block-group.h
@@ -212,6 +212,8 @@ struct btrfs_block_group {
 	u64 meta_write_pointer;
 	struct map_lookup *physical_map;
 	struct list_head active_bg_list;
+	struct work_struct zone_finish_work;
+	struct extent_buffer *last_eb;
 };
 
 static inline u64 btrfs_block_group_end(struct btrfs_block_group *block_group)
@@ -254,8 +256,9 @@ void btrfs_put_block_group(struct btrfs_block_group *cache);
 void btrfs_dec_block_group_reservations(struct btrfs_fs_info *fs_info,
 					const u64 start);
 void btrfs_wait_block_group_reservations(struct btrfs_block_group *bg);
-bool btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr);
-void btrfs_dec_nocow_writers(struct btrfs_fs_info *fs_info, u64 bytenr);
+struct btrfs_block_group *btrfs_inc_nocow_writers(struct btrfs_fs_info *fs_info,
+						  u64 bytenr);
+void btrfs_dec_nocow_writers(struct btrfs_block_group *bg);
 void btrfs_wait_nocow_writers(struct btrfs_block_group *bg);
 void btrfs_wait_block_group_cache_progress(struct btrfs_block_group *cache,
 				           u64 num_bytes);
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 32131a5d321b..33811e896623 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -395,31 +395,6 @@ static inline bool btrfs_inode_can_compress(const struct btrfs_inode *inode)
 	return true;
 }
 
-struct btrfs_dio_private {
-	struct inode *inode;
-
-	/*
-	 * Since DIO can use anonymous page, we cannot use page_offset() to
-	 * grab the file offset, thus need a dedicated member for file offset.
-	 */
-	u64 file_offset;
-	u64 disk_bytenr;
-	/* Used for bio::bi_size */
-	u32 bytes;
-
-	/*
-	 * References to this structure. There is one reference per in-flight
-	 * bio plus one while we're still setting up.
-	 */
-	refcount_t refs;
-
-	/* dio_bio came from fs/direct-io.c */
-	struct bio *dio_bio;
-
-	/* Array of checksums */
-	u8 csums[];
-};
-
 /*
  * btrfs_inode_item stores flags in a u64, btrfs_inode stores them in two
  * separate u32s. These two functions convert between the two representations.
diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c
index abac86a75840..5d20137b7b67 100644
--- a/fs/btrfs/check-integrity.c
+++ b/fs/btrfs/check-integrity.c
@@ -1552,21 +1552,18 @@ static int btrfsic_read_block(struct btrfsic_state *state,
 		return -ENOMEM;
 	block_ctx->datav = block_ctx->mem_to_free;
 	block_ctx->pagev = (struct page **)(block_ctx->datav + num_pages);
-	for (i = 0; i < num_pages; i++) {
-		block_ctx->pagev[i] = alloc_page(GFP_NOFS);
-		if (!block_ctx->pagev[i])
-			return -1;
-	}
+	ret = btrfs_alloc_page_array(num_pages, block_ctx->pagev);
+	if (ret)
+		return ret;
 
 	dev_bytenr = block_ctx->dev_bytenr;
 	for (i = 0; i < num_pages;) {
 		struct bio *bio;
 		unsigned int j;
 
-		bio = btrfs_bio_alloc(num_pages - i);
-		bio_set_dev(bio, block_ctx->dev->bdev);
+		bio = bio_alloc(block_ctx->dev->bdev, num_pages - i,
+				REQ_OP_READ, GFP_NOFS);
 		bio->bi_iter.bi_sector = dev_bytenr >> 9;
-		bio->bi_opf = REQ_OP_READ;
 
 		for (j = i; j < num_pages; j++) {
 			ret = bio_add_page(bio, block_ctx->pagev[j],
@@ -2033,7 +2030,7 @@ continue_loop:
 
 static void btrfsic_bio_end_io(struct bio *bp)
 {
-	struct btrfsic_block *block = (struct btrfsic_block *)bp->bi_private;
+	struct btrfsic_block *block = bp->bi_private;
 	int iodone_w_error;
 
 	/* mutex is not held! This is not save if IO is not yet completed
@@ -2635,100 +2632,93 @@ static struct btrfsic_dev_state *btrfsic_dev_state_lookup(dev_t dev)
 						  &btrfsic_dev_state_hashtable);
 }
 
-static void __btrfsic_submit_bio(struct bio *bio)
+static void btrfsic_check_write_bio(struct bio *bio, struct btrfsic_dev_state *dev_state)
 {
-	struct btrfsic_dev_state *dev_state;
+	unsigned int segs = bio_segments(bio);
+	u64 dev_bytenr = 512 * bio->bi_iter.bi_sector;
+	u64 cur_bytenr = dev_bytenr;
+	struct bvec_iter iter;
+	struct bio_vec bvec;
+	char **mapped_datav;
+	int bio_is_patched = 0;
+	int i = 0;
+
+	if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
+		pr_info(
+"submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n",
+		       bio_op(bio), bio->bi_opf, segs,
+		       bio->bi_iter.bi_sector, dev_bytenr, bio->bi_bdev);
 
-	if (!btrfsic_is_initialized)
+	mapped_datav = kmalloc_array(segs, sizeof(*mapped_datav), GFP_NOFS);
+	if (!mapped_datav)
 		return;
 
-	mutex_lock(&btrfsic_mutex);
-	/* since btrfsic_submit_bio() is also called before
-	 * btrfsic_mount(), this might return NULL */
-	dev_state = btrfsic_dev_state_lookup(bio->bi_bdev->bd_dev);
-	if (NULL != dev_state &&
-	    (bio_op(bio) == REQ_OP_WRITE) && bio_has_data(bio)) {
-		int i = 0;
-		u64 dev_bytenr;
-		u64 cur_bytenr;
-		struct bio_vec bvec;
-		struct bvec_iter iter;
-		int bio_is_patched;
-		char **mapped_datav;
-		unsigned int segs = bio_segments(bio);
-
-		dev_bytenr = 512 * bio->bi_iter.bi_sector;
-		bio_is_patched = 0;
-		if (dev_state->state->print_mask &
-		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n",
-			       bio_op(bio), bio->bi_opf, segs,
-			       bio->bi_iter.bi_sector, dev_bytenr, bio->bi_bdev);
-
-		mapped_datav = kmalloc_array(segs,
-					     sizeof(*mapped_datav), GFP_NOFS);
-		if (!mapped_datav)
-			goto leave;
-		cur_bytenr = dev_bytenr;
-
-		bio_for_each_segment(bvec, bio, iter) {
-			BUG_ON(bvec.bv_len != PAGE_SIZE);
-			mapped_datav[i] = page_address(bvec.bv_page);
-			i++;
-
-			if (dev_state->state->print_mask &
-			    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH_VERBOSE)
-				pr_info("#%u: bytenr=%llu, len=%u, offset=%u\n",
-				       i, cur_bytenr, bvec.bv_len, bvec.bv_offset);
-			cur_bytenr += bvec.bv_len;
-		}
-		btrfsic_process_written_block(dev_state, dev_bytenr,
-					      mapped_datav, segs,
-					      bio, &bio_is_patched,
-					      bio->bi_opf);
-		kfree(mapped_datav);
-	} else if (NULL != dev_state && (bio->bi_opf & REQ_PREFLUSH)) {
-		if (dev_state->state->print_mask &
-		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
-			pr_info("submit_bio(rw=%d,0x%x FLUSH, bdev=%p)\n",
-			       bio_op(bio), bio->bi_opf, bio->bi_bdev);
-		if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) {
-			if ((dev_state->state->print_mask &
-			     (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH |
-			      BTRFSIC_PRINT_MASK_VERBOSE)))
-				pr_info(
-"btrfsic_submit_bio(%pg) with FLUSH but dummy block already in use (ignored)!\n",
-				       dev_state->bdev);
-		} else {
-			struct btrfsic_block *const block =
-				&dev_state->dummy_block_for_bio_bh_flush;
+	bio_for_each_segment(bvec, bio, iter) {
+		BUG_ON(bvec.bv_len != PAGE_SIZE);
+		mapped_datav[i] = page_address(bvec.bv_page);
+		i++;
 
-			block->is_iodone = 0;
-			block->never_written = 0;
-			block->iodone_w_error = 0;
-			block->flush_gen = dev_state->last_flush_gen + 1;
-			block->submit_bio_bh_rw = bio->bi_opf;
-			block->orig_bio_private = bio->bi_private;
-			block->orig_bio_end_io = bio->bi_end_io;
-			block->next_in_same_bio = NULL;
-			bio->bi_private = block;
-			bio->bi_end_io = btrfsic_bio_end_io;
-		}
+		if (dev_state->state->print_mask &
+		    BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH_VERBOSE)
+			pr_info("#%u: bytenr=%llu, len=%u, offset=%u\n",
+			       i, cur_bytenr, bvec.bv_len, bvec.bv_offset);
+		cur_bytenr += bvec.bv_len;
 	}
-leave:
-	mutex_unlock(&btrfsic_mutex);
+
+	btrfsic_process_written_block(dev_state, dev_bytenr, mapped_datav, segs,
+				      bio, &bio_is_patched, bio->bi_opf);
+	kfree(mapped_datav);
 }
 
-void btrfsic_submit_bio(struct bio *bio)
+static void btrfsic_check_flush_bio(struct bio *bio, struct btrfsic_dev_state *dev_state)
 {
-	__btrfsic_submit_bio(bio);
-	submit_bio(bio);
+	if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH)
+		pr_info("submit_bio(rw=%d,0x%x FLUSH, bdev=%p)\n",
+		       bio_op(bio), bio->bi_opf, bio->bi_bdev);
+
+	if (dev_state->dummy_block_for_bio_bh_flush.is_iodone) {
+		struct btrfsic_block *const block =
+			&dev_state->dummy_block_for_bio_bh_flush;
+
+		block->is_iodone = 0;
+		block->never_written = 0;
+		block->iodone_w_error = 0;
+		block->flush_gen = dev_state->last_flush_gen + 1;
+		block->submit_bio_bh_rw = bio->bi_opf;
+		block->orig_bio_private = bio->bi_private;
+		block->orig_bio_end_io = bio->bi_end_io;
+		block->next_in_same_bio = NULL;
+		bio->bi_private = block;
+		bio->bi_end_io = btrfsic_bio_end_io;
+	} else if ((dev_state->state->print_mask &
+		   (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH |
+		    BTRFSIC_PRINT_MASK_VERBOSE))) {
+		pr_info(
+"btrfsic_submit_bio(%pg) with FLUSH but dummy block already in use (ignored)!\n",
+		       dev_state->bdev);
+	}
 }
 
-int btrfsic_submit_bio_wait(struct bio *bio)
+void btrfsic_check_bio(struct bio *bio)
 {
-	__btrfsic_submit_bio(bio);
-	return submit_bio_wait(bio);
+	struct btrfsic_dev_state *dev_state;
+
+	if (!btrfsic_is_initialized)
+		return;
+
+	/*
+	 * We can be called before btrfsic_mount, so there might not be a
+	 * dev_state.
+	 */
+	dev_state = btrfsic_dev_state_lookup(bio->bi_bdev->bd_dev);
+	mutex_lock(&btrfsic_mutex);
+	if (dev_state) {
+		if (bio_op(bio) == REQ_OP_WRITE && bio_has_data(bio))
+			btrfsic_check_write_bio(bio, dev_state);
+		else if (bio->bi_opf & REQ_PREFLUSH)
+			btrfsic_check_flush_bio(bio, dev_state);
+	}
+	mutex_unlock(&btrfsic_mutex);
 }
 
 int btrfsic_mount(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/check-integrity.h b/fs/btrfs/check-integrity.h
index bcc730a06cb5..e4c8aed7996f 100644
--- a/fs/btrfs/check-integrity.h
+++ b/fs/btrfs/check-integrity.h
@@ -7,11 +7,9 @@
 #define BTRFS_CHECK_INTEGRITY_H
 
 #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY
-void btrfsic_submit_bio(struct bio *bio);
-int btrfsic_submit_bio_wait(struct bio *bio);
+void btrfsic_check_bio(struct bio *bio);
 #else
-#define btrfsic_submit_bio submit_bio
-#define btrfsic_submit_bio_wait submit_bio_wait
+static inline void btrfsic_check_bio(struct bio *bio) { }
 #endif
 
 int btrfsic_mount(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index 19bf36d8ffea..f4564f32f6d9 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -425,7 +425,6 @@ out:
 }
 
 static blk_status_t submit_compressed_bio(struct btrfs_fs_info *fs_info,
-					  struct compressed_bio *cb,
 					  struct bio *bio, int mirror_num)
 {
 	blk_status_t ret;
@@ -604,7 +603,7 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 					goto finish_cb;
 			}
 
-			ret = submit_compressed_bio(fs_info, cb, bio, 0);
+			ret = submit_compressed_bio(fs_info, bio, 0);
 			if (ret)
 				goto finish_cb;
 			bio = NULL;
@@ -802,15 +801,13 @@ static noinline int add_ra_bio_pages(struct inode *inode,
  * After the compressed pages are read, we copy the bytes into the
  * bio we were passed and then call the bio end_io calls
  */
-blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
-				 int mirror_num, unsigned long bio_flags)
+void btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
+				  int mirror_num)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map_tree *em_tree;
 	struct compressed_bio *cb;
 	unsigned int compressed_len;
-	unsigned int nr_pages;
-	unsigned int pg_index;
 	struct bio *comp_bio = NULL;
 	const u64 disk_bytenr = bio->bi_iter.bi_sector << SECTOR_SHIFT;
 	u64 cur_disk_byte = disk_bytenr;
@@ -820,7 +817,8 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	u64 em_start;
 	struct extent_map *em;
 	blk_status_t ret;
-	int faili = 0;
+	int ret2;
+	int i;
 	u8 *sums;
 
 	em_tree = &BTRFS_I(inode)->extent_tree;
@@ -855,32 +853,26 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	em_len = em->len;
 	em_start = em->start;
 
-	free_extent_map(em);
-	em = NULL;
-
 	cb->len = bio->bi_iter.bi_size;
 	cb->compressed_len = compressed_len;
-	cb->compress_type = extent_compress_type(bio_flags);
+	cb->compress_type = em->compress_type;
 	cb->orig_bio = bio;
 
-	nr_pages = DIV_ROUND_UP(compressed_len, PAGE_SIZE);
-	cb->compressed_pages = kcalloc(nr_pages, sizeof(struct page *),
-				       GFP_NOFS);
+	free_extent_map(em);
+	em = NULL;
+
+	cb->nr_pages = DIV_ROUND_UP(compressed_len, PAGE_SIZE);
+	cb->compressed_pages = kcalloc(cb->nr_pages, sizeof(struct page *), GFP_NOFS);
 	if (!cb->compressed_pages) {
 		ret = BLK_STS_RESOURCE;
-		goto fail1;
+		goto fail;
 	}
 
-	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
-		cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS);
-		if (!cb->compressed_pages[pg_index]) {
-			faili = pg_index - 1;
-			ret = BLK_STS_RESOURCE;
-			goto fail2;
-		}
+	ret2 = btrfs_alloc_page_array(cb->nr_pages, cb->compressed_pages);
+	if (ret2) {
+		ret = BLK_STS_RESOURCE;
+		goto fail;
 	}
-	faili = nr_pages - 1;
-	cb->nr_pages = nr_pages;
 
 	add_ra_bio_pages(inode, em_start + em_len, cb);
 
@@ -949,28 +941,29 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 						  fs_info->sectorsize);
 			sums += fs_info->csum_size * nr_sectors;
 
-			ret = submit_compressed_bio(fs_info, cb, comp_bio, mirror_num);
+			ret = submit_compressed_bio(fs_info, comp_bio, mirror_num);
 			if (ret)
 				goto finish_cb;
 			comp_bio = NULL;
 		}
 	}
-	return BLK_STS_OK;
+	return;
 
-fail2:
-	while (faili >= 0) {
-		__free_page(cb->compressed_pages[faili]);
-		faili--;
+fail:
+	if (cb->compressed_pages) {
+		for (i = 0; i < cb->nr_pages; i++) {
+			if (cb->compressed_pages[i])
+				__free_page(cb->compressed_pages[i]);
+		}
 	}
 
 	kfree(cb->compressed_pages);
-fail1:
 	kfree(cb);
 out:
 	free_extent_map(em);
 	bio->bi_status = ret;
 	bio_endio(bio);
-	return ret;
+	return;
 finish_cb:
 	if (comp_bio) {
 		comp_bio->bi_status = ret;
@@ -978,7 +971,7 @@ finish_cb:
 	}
 	/* All bytes of @cb is submitted, endio will free @cb */
 	if (cur_disk_byte == disk_bytenr + compressed_len)
-		return ret;
+		return;
 
 	wait_var_event(cb, refcount_read(&cb->pending_sectors) ==
 			   (disk_bytenr + compressed_len - cur_disk_byte) >>
@@ -990,7 +983,6 @@ finish_cb:
 	ASSERT(refcount_read(&cb->pending_sectors));
 	/* Now we are the only one referring @cb, can finish it safely. */
 	finish_compressed_bio_read(cb);
-	return ret;
 }
 
 /*
diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h
index ac5b20731d2a..2707404389a5 100644
--- a/fs/btrfs/compression.h
+++ b/fs/btrfs/compression.h
@@ -102,8 +102,8 @@ blk_status_t btrfs_submit_compressed_write(struct btrfs_inode *inode, u64 start,
 				  unsigned int write_flags,
 				  struct cgroup_subsys_state *blkcg_css,
 				  bool writeback);
-blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
-				 int mirror_num, unsigned long bio_flags);
+void btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
+				  int mirror_num);
 
 unsigned int btrfs_compress_str2level(unsigned int type, const char *str);
 
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 0eecf98d0abb..6e556031a8f3 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -16,6 +16,7 @@
 #include "volumes.h"
 #include "qgroup.h"
 #include "tree-mod-log.h"
+#include "tree-checker.h"
 
 static int split_node(struct btrfs_trans_handle *trans, struct btrfs_root
 		      *root, struct btrfs_path *path, int level);
@@ -342,7 +343,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 			int level = btrfs_header_level(buf);
 
 			ret = btrfs_set_disk_extent_flags(trans, buf,
-							  new_flags, level, 0);
+							  new_flags, level);
 			if (ret)
 				return ret;
 		}
@@ -1390,12 +1391,13 @@ static noinline void unlock_up(struct btrfs_path *path, int level,
 }
 
 /*
- * helper function for btrfs_search_slot.  The goal is to find a block
- * in cache without setting the path to blocking.  If we find the block
- * we return zero and the path is unchanged.
+ * Helper function for btrfs_search_slot() and other functions that do a search
+ * on a btree. The goal is to find a tree block in the cache (the radix tree at
+ * fs_info->buffer_radix), but if we can't find it, or it's not up to date, read
+ * its pages from disk.
  *
- * If we can't find the block, we set the path blocking and do some
- * reada.  -EAGAIN is returned and the search must be repeated.
+ * Returns -EAGAIN, with the path unlocked, if the caller needs to repeat the
+ * whole btree search, starting again from the current root node.
  */
 static int
 read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
@@ -1409,12 +1411,21 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 	struct btrfs_key first_key;
 	int ret;
 	int parent_level;
+	bool unlock_up;
 
+	unlock_up = ((level + 1 < BTRFS_MAX_LEVEL) && p->locks[level + 1]);
 	blocknr = btrfs_node_blockptr(*eb_ret, slot);
 	gen = btrfs_node_ptr_generation(*eb_ret, slot);
 	parent_level = btrfs_header_level(*eb_ret);
 	btrfs_node_key_to_cpu(*eb_ret, &first_key, slot);
 
+	/*
+	 * If we need to read an extent buffer from disk and we are holding locks
+	 * on upper level nodes, we unlock all the upper nodes before reading the
+	 * extent buffer, and then return -EAGAIN to the caller as it needs to
+	 * restart the search. We don't release the lock on the current level
+	 * because we need to walk this node to figure out which blocks to read.
+	 */
 	tmp = find_extent_buffer(fs_info, blocknr);
 	if (tmp) {
 		if (p->reada == READA_FORWARD_ALWAYS)
@@ -1436,30 +1447,38 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 			return 0;
 		}
 
+		if (unlock_up)
+			btrfs_unlock_up_safe(p, level + 1);
+
 		/* now we're allowed to do a blocking uptodate check */
-		ret = btrfs_read_buffer(tmp, gen, parent_level - 1, &first_key);
+		ret = btrfs_read_extent_buffer(tmp, gen, parent_level - 1, &first_key);
 		if (ret) {
 			free_extent_buffer(tmp);
 			btrfs_release_path(p);
 			return -EIO;
 		}
-		*eb_ret = tmp;
-		return 0;
+		if (btrfs_check_eb_owner(tmp, root->root_key.objectid)) {
+			free_extent_buffer(tmp);
+			btrfs_release_path(p);
+			return -EUCLEAN;
+		}
+
+		if (unlock_up)
+			ret = -EAGAIN;
+
+		goto out;
 	}
 
-	/*
-	 * reduce lock contention at high levels
-	 * of the btree by dropping locks before
-	 * we read.  Don't release the lock on the current
-	 * level because we need to walk this node to figure
-	 * out which blocks to read.
-	 */
-	btrfs_unlock_up_safe(p, level + 1);
+	if (unlock_up) {
+		btrfs_unlock_up_safe(p, level + 1);
+		ret = -EAGAIN;
+	} else {
+		ret = 0;
+	}
 
 	if (p->reada != READA_NONE)
 		reada_for_search(fs_info, p, level, slot, key->objectid);
 
-	ret = -EAGAIN;
 	tmp = read_tree_block(fs_info, blocknr, root->root_key.objectid,
 			      gen, parent_level - 1, &first_key);
 	if (IS_ERR(tmp)) {
@@ -1474,9 +1493,15 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 	 */
 	if (!extent_buffer_uptodate(tmp))
 		ret = -EIO;
-	free_extent_buffer(tmp);
 
-	btrfs_release_path(p);
+out:
+	if (ret == 0) {
+		*eb_ret = tmp;
+	} else {
+		free_extent_buffer(tmp);
+		btrfs_release_path(p);
+	}
+
 	return ret;
 }
 
@@ -2279,6 +2304,43 @@ int btrfs_search_backwards(struct btrfs_root *root, struct btrfs_key *key,
 	return ret;
 }
 
+/**
+ * Search for a valid slot for the given path.
+ *
+ * @root:	The root node of the tree.
+ * @key:	Will contain a valid item if found.
+ * @path:	The starting point to validate the slot.
+ *
+ * Return: 0  if the item is valid
+ *         1  if not found
+ *         <0 if error.
+ */
+int btrfs_get_next_valid_item(struct btrfs_root *root, struct btrfs_key *key,
+			      struct btrfs_path *path)
+{
+	while (1) {
+		int ret;
+		const int slot = path->slots[0];
+		const struct extent_buffer *leaf = path->nodes[0];
+
+		/* This is where we start walking the path. */
+		if (slot >= btrfs_header_nritems(leaf)) {
+			/*
+			 * If we've reached the last slot in this leaf we need
+			 * to go to the next leaf and reset the path.
+			 */
+			ret = btrfs_next_leaf(root, path);
+			if (ret)
+				return ret;
+			continue;
+		}
+		/* Store the found, valid item in @key. */
+		btrfs_item_key_to_cpu(leaf, key, slot);
+		break;
+	}
+	return 0;
+}
+
 /*
  * adjust the pointers going up the tree, starting at level
  * making sure the right key of each node is points to 'key'.
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 077c95e9baa5..0e49b1a0c071 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -675,13 +675,13 @@ struct btrfs_fs_info {
 	rwlock_t global_root_lock;
 	struct rb_root global_root_tree;
 
-	spinlock_t fs_roots_radix_lock;
-	struct radix_tree_root fs_roots_radix;
+	/* The xarray that holds all the FS roots */
+	spinlock_t fs_roots_lock;
+	struct xarray fs_roots;
 
 	/* block group cache stuff */
-	spinlock_t block_group_cache_lock;
-	u64 first_logical_byte;
-	struct rb_root block_group_cache_tree;
+	rwlock_t block_group_cache_lock;
+	struct rb_root_cached block_group_cache_tree;
 
 	/* keep track of unallocated space */
 	atomic64_t free_chunk_space;
@@ -848,12 +848,13 @@ struct btrfs_fs_info {
 	 * two
 	 */
 	struct btrfs_workqueue *workers;
+	struct btrfs_workqueue *hipri_workers;
 	struct btrfs_workqueue *delalloc_workers;
 	struct btrfs_workqueue *flush_workers;
 	struct btrfs_workqueue *endio_workers;
 	struct btrfs_workqueue *endio_meta_workers;
 	struct btrfs_workqueue *endio_raid56_workers;
-	struct btrfs_workqueue *rmw_workers;
+	struct workqueue_struct *rmw_workers;
 	struct btrfs_workqueue *endio_meta_write_workers;
 	struct btrfs_workqueue *endio_write_workers;
 	struct btrfs_workqueue *endio_freespace_worker;
@@ -946,9 +947,9 @@ struct btrfs_fs_info {
 	 * running.
 	 */
 	refcount_t scrub_workers_refcnt;
-	struct btrfs_workqueue *scrub_workers;
-	struct btrfs_workqueue *scrub_wr_completion_workers;
-	struct btrfs_workqueue *scrub_parity_workers;
+	struct workqueue_struct *scrub_workers;
+	struct workqueue_struct *scrub_wr_completion_workers;
+	struct workqueue_struct *scrub_parity_workers;
 	struct btrfs_subpage_info *subpage_info;
 
 	struct btrfs_discard_ctl discard_ctl;
@@ -994,10 +995,10 @@ struct btrfs_fs_info {
 
 	struct btrfs_delayed_root *delayed_root;
 
-	/* Extent buffer radix tree */
+	/* Extent buffer xarray */
 	spinlock_t buffer_lock;
 	/* Entries are eb->start / sectorsize */
-	struct radix_tree_root buffer_radix;
+	struct xarray extent_buffers;
 
 	/* next backup root to be overwritten */
 	int backup_root_index;
@@ -1045,10 +1046,7 @@ struct btrfs_fs_info {
 	 * Zone size > 0 when in ZONED mode, otherwise it's used for a check
 	 * if the mode is enabled
 	 */
-	union {
-		u64 zone_size;
-		u64 zoned;
-	};
+	u64 zone_size;
 
 	struct mutex zoned_meta_io_lock;
 	spinlock_t treelog_bg_lock;
@@ -1121,7 +1119,8 @@ enum {
 	 */
 	BTRFS_ROOT_SHAREABLE,
 	BTRFS_ROOT_TRACK_DIRTY,
-	BTRFS_ROOT_IN_RADIX,
+	/* The root is tracked in fs_info::fs_roots */
+	BTRFS_ROOT_REGISTERED,
 	BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
 	BTRFS_ROOT_DEFRAG_RUNNING,
 	BTRFS_ROOT_FORCE_COW,
@@ -1225,10 +1224,10 @@ struct btrfs_root {
 	struct rb_root inode_tree;
 
 	/*
-	 * radix tree that keeps track of delayed nodes of every inode,
-	 * protected by inode_lock
+	 * Xarray that keeps track of delayed nodes of every inode, protected
+	 * by inode_lock
 	 */
-	struct radix_tree_root delayed_nodes_tree;
+	struct xarray delayed_nodes;
 	/*
 	 * right now this just gets used so that a root has its own devid
 	 * for stat.  It may be used for more later
@@ -2784,7 +2783,8 @@ int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans,
 				    u64 bytenr, u64 num_bytes);
 int btrfs_exclude_logged_extents(struct extent_buffer *eb);
 int btrfs_cross_ref_exist(struct btrfs_root *root,
-			  u64 objectid, u64 offset, u64 bytenr, bool strict);
+			  u64 objectid, u64 offset, u64 bytenr, bool strict,
+			  struct btrfs_path *path);
 struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 					     struct btrfs_root *root,
 					     u64 parent, u64 root_objectid,
@@ -2811,8 +2811,7 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		  struct extent_buffer *buf, int full_backref);
 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
-				struct extent_buffer *eb, u64 flags,
-				int level, int is_data);
+				struct extent_buffer *eb, u64 flags, int level);
 int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref);
 
 int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info,
@@ -2892,7 +2891,7 @@ void btrfs_subvolume_release_metadata(struct btrfs_root *root,
 void btrfs_delalloc_release_extents(struct btrfs_inode *inode, u64 num_bytes);
 
 int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes,
-				    u64 disk_num_bytes);
+				    u64 disk_num_bytes, bool noflush);
 u64 btrfs_account_ro_block_groups_free_space(struct btrfs_space_info *sinfo);
 int btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info,
 				   u64 start, u64 end);
@@ -3039,6 +3038,35 @@ int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
 int btrfs_search_backwards(struct btrfs_root *root, struct btrfs_key *key,
 			   struct btrfs_path *path);
 
+int btrfs_get_next_valid_item(struct btrfs_root *root, struct btrfs_key *key,
+			      struct btrfs_path *path);
+
+/*
+ * Search in @root for a given @key, and store the slot found in @found_key.
+ *
+ * @root:	The root node of the tree.
+ * @key:	The key we are looking for.
+ * @found_key:	Will hold the found item.
+ * @path:	Holds the current slot/leaf.
+ * @iter_ret:	Contains the value returned from btrfs_search_slot or
+ * 		btrfs_get_next_valid_item, whichever was executed last.
+ *
+ * The @iter_ret is an output variable that will contain the return value of
+ * btrfs_search_slot, if it encountered an error, or the value returned from
+ * btrfs_get_next_valid_item otherwise. That return value can be 0, if a valid
+ * slot was found, 1 if there were no more leaves, and <0 if there was an error.
+ *
+ * It's recommended to use a separate variable for iter_ret and then use it to
+ * set the function return value so there's no confusion of the 0/1/errno
+ * values stemming from btrfs_search_slot.
+ */
+#define btrfs_for_each_slot(root, key, found_key, path, iter_ret)		\
+	for (iter_ret = btrfs_search_slot(NULL, (root), (key), (path), 0, 0);	\
+		(iter_ret) >= 0 &&						\
+		(iter_ret = btrfs_get_next_valid_item((root), (found_key), (path))) == 0; \
+		(path)->slots[0]++						\
+	)
+
 static inline int btrfs_next_old_item(struct btrfs_root *root,
 				      struct btrfs_path *p, u64 time_seq)
 {
@@ -3190,7 +3218,6 @@ int btrfs_del_orphan_item(struct btrfs_trans_handle *trans,
 int btrfs_find_orphan_item(struct btrfs_root *root, u64 offset);
 
 /* file-item.c */
-struct btrfs_dio_private;
 int btrfs_del_csums(struct btrfs_trans_handle *trans,
 		    struct btrfs_root *root, u64 bytenr, u64 len);
 blk_status_t btrfs_lookup_bio_sums(struct inode *inode, struct bio *bio, u8 *dst);
@@ -3224,8 +3251,8 @@ void btrfs_inode_safe_disk_i_size_write(struct btrfs_inode *inode, u64 new_i_siz
 u64 btrfs_file_extent_end(const struct btrfs_path *path);
 
 /* inode.c */
-blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
-				   int mirror_num, unsigned long bio_flags);
+void btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
+			   int mirror_num, enum btrfs_compression_type compress_type);
 unsigned int btrfs_verify_data_csum(struct btrfs_bio *bbio,
 				    u32 bio_offset, struct page *page,
 				    u64 start, u64 end);
@@ -3255,10 +3282,28 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, long nr,
 int btrfs_set_extent_delalloc(struct btrfs_inode *inode, u64 start, u64 end,
 			      unsigned int extra_bits,
 			      struct extent_state **cached_state);
-int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
-			     struct btrfs_root *new_root,
-			     struct btrfs_root *parent_root,
-			     struct user_namespace *mnt_userns);
+struct btrfs_new_inode_args {
+	/* Input */
+	struct inode *dir;
+	struct dentry *dentry;
+	struct inode *inode;
+	bool orphan;
+	bool subvol;
+
+	/*
+	 * Output from btrfs_new_inode_prepare(), input to
+	 * btrfs_create_new_inode().
+	 */
+	struct posix_acl *default_acl;
+	struct posix_acl *acl;
+};
+int btrfs_new_inode_prepare(struct btrfs_new_inode_args *args,
+			    unsigned int *trans_num_items);
+int btrfs_create_new_inode(struct btrfs_trans_handle *trans,
+			   struct btrfs_new_inode_args *args);
+void btrfs_new_inode_args_destroy(struct btrfs_new_inode_args *args);
+struct inode *btrfs_new_subvol_inode(struct user_namespace *mnt_userns,
+				     struct inode *dir);
  void btrfs_set_delalloc_extent(struct inode *inode, struct extent_state *state,
 			       unsigned *bits);
 void btrfs_clear_delalloc_extent(struct inode *inode,
@@ -3269,7 +3314,6 @@ void btrfs_split_delalloc_extent(struct inode *inode,
 				 struct extent_state *orig, u64 split);
 void btrfs_set_range_writeback(struct btrfs_inode *inode, u64 start, u64 end);
 vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf);
-int btrfs_readpage(struct file *file, struct page *page);
 void btrfs_evict_inode(struct inode *inode);
 int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc);
 struct inode *btrfs_alloc_inode(struct super_block *sb);
@@ -3314,9 +3358,9 @@ ssize_t btrfs_encoded_read(struct kiocb *iocb, struct iov_iter *iter,
 ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from,
 			     const struct btrfs_ioctl_encoded_io_args *encoded);
 
+ssize_t btrfs_dio_rw(struct kiocb *iocb, struct iov_iter *iter, size_t done_before);
+
 extern const struct dentry_operations btrfs_dentry_operations;
-extern const struct iomap_ops btrfs_dio_iomap_ops;
-extern const struct iomap_dio_ops btrfs_dio_ops;
 
 /* Inode locking type flags, by default the exclusive lock is taken */
 #define BTRFS_ILOCK_SHARED	(1U << 0)
@@ -3328,6 +3372,7 @@ void btrfs_inode_unlock(struct inode *inode, unsigned int ilock_flags);
 void btrfs_update_inode_bytes(struct btrfs_inode *inode,
 			      const u64 add_bytes,
 			      const u64 del_bytes);
+void btrfs_assert_inode_range_clean(struct btrfs_inode *inode, u64 start, u64 end);
 
 /* ioctl.c */
 long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
@@ -3403,11 +3448,29 @@ void btrfs_no_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 {
 }
 
-#ifdef CONFIG_PRINTK
+#ifdef CONFIG_PRINTK_INDEX
+
+#define btrfs_printk(fs_info, fmt, args...)					\
+do {										\
+	printk_index_subsys_emit("%sBTRFS %s (device %s): ", NULL, fmt);	\
+	_btrfs_printk(fs_info, fmt, ##args);					\
+} while (0)
+
 __printf(2, 3)
 __cold
-void btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...);
+void _btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...);
+
+#elif defined(CONFIG_PRINTK)
+
+#define btrfs_printk(fs_info, fmt, args...)				\
+	_btrfs_printk(fs_info, fmt, ##args)
+
+__printf(2, 3)
+__cold
+void _btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...);
+
 #else
+
 #define btrfs_printk(fs_info, fmt, args...) \
 	btrfs_no_printk(fs_info, fmt, ##args)
 #endif
@@ -3658,12 +3721,25 @@ do {								\
 				  __LINE__, (errno));		\
 } while (0)
 
+#ifdef CONFIG_PRINTK_INDEX
+
 #define btrfs_handle_fs_error(fs_info, errno, fmt, args...)		\
-do {								\
-	__btrfs_handle_fs_error((fs_info), __func__, __LINE__,	\
-			  (errno), fmt, ##args);		\
+do {									\
+	printk_index_subsys_emit(					\
+		"BTRFS: error (device %s%s) in %s:%d: errno=%d %s",	\
+		KERN_CRIT, fmt);					\
+	__btrfs_handle_fs_error((fs_info), __func__, __LINE__,		\
+				(errno), fmt, ##args);			\
 } while (0)
 
+#else
+
+#define btrfs_handle_fs_error(fs_info, errno, fmt, args...)		\
+	__btrfs_handle_fs_error((fs_info), __func__, __LINE__,		\
+				(errno), fmt, ##args)
+
+#endif
+
 #define BTRFS_FS_ERROR(fs_info)	(unlikely(test_bit(BTRFS_FS_STATE_ERROR, \
 						   &(fs_info)->fs_state)))
 #define BTRFS_FS_LOG_CLEANUP_ERROR(fs_info)				\
@@ -3816,15 +3892,16 @@ static inline int __btrfs_fs_compat_ro(struct btrfs_fs_info *fs_info, u64 flag)
 struct posix_acl *btrfs_get_acl(struct inode *inode, int type, bool rcu);
 int btrfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode,
 		  struct posix_acl *acl, int type);
-int btrfs_init_acl(struct btrfs_trans_handle *trans,
-		   struct inode *inode, struct inode *dir);
+int __btrfs_set_acl(struct btrfs_trans_handle *trans, struct inode *inode,
+		    struct posix_acl *acl, int type);
 #else
 #define btrfs_get_acl NULL
 #define btrfs_set_acl NULL
-static inline int btrfs_init_acl(struct btrfs_trans_handle *trans,
-				 struct inode *inode, struct inode *dir)
+static inline int __btrfs_set_acl(struct btrfs_trans_handle *trans,
+				  struct inode *inode, struct posix_acl *acl,
+				  int type)
 {
-	return 0;
+	return -EOPNOTSUPP;
 }
 #endif
 
@@ -3929,7 +4006,7 @@ static inline int btrfs_is_testing(struct btrfs_fs_info *fs_info)
 
 static inline bool btrfs_is_zoned(const struct btrfs_fs_info *fs_info)
 {
-	return fs_info->zoned != 0;
+	return fs_info->zone_size > 0;
 }
 
 static inline bool btrfs_is_data_reloc_root(const struct btrfs_root *root)
diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c
index bd8267c4687d..36ab0859a263 100644
--- a/fs/btrfs/delalloc-space.c
+++ b/fs/btrfs/delalloc-space.c
@@ -289,7 +289,7 @@ static void calc_inode_reservations(struct btrfs_fs_info *fs_info,
 }
 
 int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes,
-				    u64 disk_num_bytes)
+				    u64 disk_num_bytes, bool noflush)
 {
 	struct btrfs_root *root = inode->root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -308,7 +308,7 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes,
 	 * If we have a transaction open (can happen if we call truncate_block
 	 * from truncate), then we need FLUSH_LIMIT so we don't deadlock.
 	 */
-	if (btrfs_is_free_space_inode(inode)) {
+	if (noflush || btrfs_is_free_space_inode(inode)) {
 		flush = BTRFS_RESERVE_NO_FLUSH;
 	} else {
 		if (current->journal_info)
@@ -333,7 +333,8 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes,
 	 */
 	calc_inode_reservations(fs_info, num_bytes, disk_num_bytes,
 				&meta_reserve, &qgroup_reserve);
-	ret = btrfs_qgroup_reserve_meta_prealloc(root, qgroup_reserve, true);
+	ret = btrfs_qgroup_reserve_meta_prealloc(root, qgroup_reserve, true,
+						 noflush);
 	if (ret)
 		return ret;
 	ret = btrfs_reserve_metadata_bytes(fs_info, block_rsv, meta_reserve, flush);
@@ -456,7 +457,7 @@ int btrfs_delalloc_reserve_space(struct btrfs_inode *inode,
 	ret = btrfs_check_data_free_space(inode, reserved, start, len);
 	if (ret < 0)
 		return ret;
-	ret = btrfs_delalloc_reserve_metadata(inode, len, len);
+	ret = btrfs_delalloc_reserve_metadata(inode, len, len, false);
 	if (ret < 0) {
 		btrfs_free_reserved_data_space(inode, *reserved, start, len);
 		extent_changeset_free(*reserved);
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 748bf6b0d860..66779ab3ed4a 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -78,7 +78,7 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(
 	}
 
 	spin_lock(&root->inode_lock);
-	node = radix_tree_lookup(&root->delayed_nodes_tree, ino);
+	node = xa_load(&root->delayed_nodes, ino);
 
 	if (node) {
 		if (btrfs_inode->delayed_node) {
@@ -90,9 +90,9 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(
 
 		/*
 		 * It's possible that we're racing into the middle of removing
-		 * this node from the radix tree.  In this case, the refcount
+		 * this node from the xarray.  In this case, the refcount
 		 * was zero and it should never go back to one.  Just return
-		 * NULL like it was never in the radix at all; our release
+		 * NULL like it was never in the xarray at all; our release
 		 * function is in the process of removing it.
 		 *
 		 * Some implementations of refcount_inc refuse to bump the
@@ -100,7 +100,7 @@ static struct btrfs_delayed_node *btrfs_get_delayed_node(
 		 * here, refcount_inc() may decide to just WARN_ONCE() instead
 		 * of actually bumping the refcount.
 		 *
-		 * If this node is properly in the radix, we want to bump the
+		 * If this node is properly in the xarray, we want to bump the
 		 * refcount twice, once for the inode and once for this get
 		 * operation.
 		 */
@@ -128,36 +128,30 @@ static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node(
 	u64 ino = btrfs_ino(btrfs_inode);
 	int ret;
 
-again:
-	node = btrfs_get_delayed_node(btrfs_inode);
-	if (node)
-		return node;
+	do {
+		node = btrfs_get_delayed_node(btrfs_inode);
+		if (node)
+			return node;
 
-	node = kmem_cache_zalloc(delayed_node_cache, GFP_NOFS);
-	if (!node)
-		return ERR_PTR(-ENOMEM);
-	btrfs_init_delayed_node(node, root, ino);
+		node = kmem_cache_zalloc(delayed_node_cache, GFP_NOFS);
+		if (!node)
+			return ERR_PTR(-ENOMEM);
+		btrfs_init_delayed_node(node, root, ino);
 
-	/* cached in the btrfs inode and can be accessed */
-	refcount_set(&node->refs, 2);
+		/* Cached in the inode and can be accessed */
+		refcount_set(&node->refs, 2);
 
-	ret = radix_tree_preload(GFP_NOFS);
-	if (ret) {
-		kmem_cache_free(delayed_node_cache, node);
-		return ERR_PTR(ret);
-	}
-
-	spin_lock(&root->inode_lock);
-	ret = radix_tree_insert(&root->delayed_nodes_tree, ino, node);
-	if (ret == -EEXIST) {
-		spin_unlock(&root->inode_lock);
-		kmem_cache_free(delayed_node_cache, node);
-		radix_tree_preload_end();
-		goto again;
-	}
+		spin_lock(&root->inode_lock);
+		ret = xa_insert(&root->delayed_nodes, ino, node, GFP_NOFS);
+		if (ret) {
+			spin_unlock(&root->inode_lock);
+			kmem_cache_free(delayed_node_cache, node);
+			if (ret != -EBUSY)
+				return ERR_PTR(ret);
+		}
+	} while (ret);
 	btrfs_inode->delayed_node = node;
 	spin_unlock(&root->inode_lock);
-	radix_tree_preload_end();
 
 	return node;
 }
@@ -276,8 +270,7 @@ static void __btrfs_release_delayed_node(
 		 * back up.  We can delete it now.
 		 */
 		ASSERT(refcount_read(&delayed_node->refs) == 0);
-		radix_tree_delete(&root->delayed_nodes_tree,
-				  delayed_node->inode_id);
+		xa_erase(&root->delayed_nodes, delayed_node->inode_id);
 		spin_unlock(&root->inode_lock);
 		kmem_cache_free(delayed_node_cache, delayed_node);
 	}
@@ -1870,34 +1863,35 @@ void btrfs_kill_delayed_inode_items(struct btrfs_inode *inode)
 
 void btrfs_kill_all_delayed_nodes(struct btrfs_root *root)
 {
-	u64 inode_id = 0;
+	unsigned long index = 0;
+	struct btrfs_delayed_node *delayed_node;
 	struct btrfs_delayed_node *delayed_nodes[8];
-	int i, n;
 
 	while (1) {
+		int n = 0;
+
 		spin_lock(&root->inode_lock);
-		n = radix_tree_gang_lookup(&root->delayed_nodes_tree,
-					   (void **)delayed_nodes, inode_id,
-					   ARRAY_SIZE(delayed_nodes));
-		if (!n) {
+		if (xa_empty(&root->delayed_nodes)) {
 			spin_unlock(&root->inode_lock);
-			break;
+			return;
 		}
 
-		inode_id = delayed_nodes[n - 1]->inode_id + 1;
-		for (i = 0; i < n; i++) {
+		xa_for_each_start(&root->delayed_nodes, index, delayed_node, index) {
 			/*
 			 * Don't increase refs in case the node is dead and
 			 * about to be removed from the tree in the loop below
 			 */
-			if (!refcount_inc_not_zero(&delayed_nodes[i]->refs))
-				delayed_nodes[i] = NULL;
+			if (refcount_inc_not_zero(&delayed_node->refs)) {
+				delayed_nodes[n] = delayed_node;
+				n++;
+			}
+			if (n >= ARRAY_SIZE(delayed_nodes))
+				break;
 		}
+		index++;
 		spin_unlock(&root->inode_lock);
 
-		for (i = 0; i < n; i++) {
-			if (!delayed_nodes[i])
-				continue;
+		for (int i = 0; i < n; i++) {
 			__btrfs_kill_delayed_node(delayed_nodes[i]);
 			btrfs_release_delayed_node(delayed_nodes[i]);
 		}
diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c
index 4176df149d04..99f37fca2e96 100644
--- a/fs/btrfs/delayed-ref.c
+++ b/fs/btrfs/delayed-ref.c
@@ -930,7 +930,6 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans,
 	is_system = (generic_ref->tree_ref.owning_root == BTRFS_CHUNK_TREE_OBJECTID);
 
 	ASSERT(generic_ref->type == BTRFS_REF_METADATA && generic_ref->action);
-	BUG_ON(extent_op && extent_op->is_data);
 	ref = kmem_cache_alloc(btrfs_delayed_tree_ref_cachep, GFP_NOFS);
 	if (!ref)
 		return -ENOMEM;
@@ -1103,8 +1102,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 
 	init_delayed_ref_head(head_ref, NULL, bytenr, num_bytes, 0, 0,
-			      BTRFS_UPDATE_DELAYED_HEAD, extent_op->is_data,
-			      false);
+			      BTRFS_UPDATE_DELAYED_HEAD, false, false);
 	head_ref->extent_op = extent_op;
 
 	delayed_refs = &trans->transaction->delayed_refs;
diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h
index 91a3aabad150..d6304b690ec4 100644
--- a/fs/btrfs/delayed-ref.h
+++ b/fs/btrfs/delayed-ref.h
@@ -58,7 +58,6 @@ struct btrfs_delayed_extent_op {
 	u8 level;
 	bool update_key;
 	bool update_flags;
-	bool is_data;
 	u64 flags_to_set;
 };
 
diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c
index f26202621989..a7dd6ba25e99 100644
--- a/fs/btrfs/dev-replace.c
+++ b/fs/btrfs/dev-replace.c
@@ -474,6 +474,7 @@ static int mark_block_group_to_copy(struct btrfs_fs_info *fs_info,
 	struct btrfs_dev_extent *dev_extent = NULL;
 	struct btrfs_block_group *cache;
 	struct btrfs_trans_handle *trans;
+	int iter_ret = 0;
 	int ret = 0;
 	u64 chunk_offset;
 
@@ -524,29 +525,8 @@ static int mark_block_group_to_copy(struct btrfs_fs_info *fs_info,
 	key.type = BTRFS_DEV_EXTENT_KEY;
 	key.offset = 0;
 
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto free_path;
-	if (ret > 0) {
-		if (path->slots[0] >=
-		    btrfs_header_nritems(path->nodes[0])) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto free_path;
-			if (ret > 0) {
-				ret = 0;
-				goto free_path;
-			}
-		} else {
-			ret = 0;
-		}
-	}
-
-	while (1) {
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
 		struct extent_buffer *leaf = path->nodes[0];
-		int slot = path->slots[0];
-
-		btrfs_item_key_to_cpu(leaf, &found_key, slot);
 
 		if (found_key.objectid != src_dev->devid)
 			break;
@@ -557,30 +537,23 @@ static int mark_block_group_to_copy(struct btrfs_fs_info *fs_info,
 		if (found_key.offset < key.offset)
 			break;
 
-		dev_extent = btrfs_item_ptr(leaf, slot, struct btrfs_dev_extent);
+		dev_extent = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dev_extent);
 
 		chunk_offset = btrfs_dev_extent_chunk_offset(leaf, dev_extent);
 
 		cache = btrfs_lookup_block_group(fs_info, chunk_offset);
 		if (!cache)
-			goto skip;
+			continue;
 
 		spin_lock(&cache->lock);
 		cache->to_copy = 1;
 		spin_unlock(&cache->lock);
 
 		btrfs_put_block_group(cache);
-
-skip:
-		ret = btrfs_next_item(root, path);
-		if (ret != 0) {
-			if (ret > 0)
-				ret = 0;
-			break;
-		}
 	}
+	if (iter_ret < 0)
+		ret = iter_ret;
 
-free_path:
 	btrfs_free_path(path);
 unlock:
 	mutex_unlock(&fs_info->chunk_mutex);
@@ -881,6 +854,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 				       int scrub_ret)
 {
 	struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
 	struct btrfs_device *tgt_device;
 	struct btrfs_device *src_device;
 	struct btrfs_root *root = fs_info->tree_root;
@@ -930,12 +904,12 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 		WARN_ON(ret);
 
 		/* Prevent write_all_supers() during the finishing procedure */
-		mutex_lock(&fs_info->fs_devices->device_list_mutex);
+		mutex_lock(&fs_devices->device_list_mutex);
 		/* Prevent new chunks being allocated on the source device */
 		mutex_lock(&fs_info->chunk_mutex);
 
 		if (!list_empty(&src_device->post_commit_list)) {
-			mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+			mutex_unlock(&fs_devices->device_list_mutex);
 			mutex_unlock(&fs_info->chunk_mutex);
 		} else {
 			break;
@@ -972,7 +946,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
 error:
 		up_write(&dev_replace->rwsem);
 		mutex_unlock(&fs_info->chunk_mutex);
-		mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+		mutex_unlock(&fs_devices->device_list_mutex);
 		btrfs_rm_dev_replace_blocked(fs_info);
 		if (tgt_device)
 			btrfs_destroy_dev_replace_tgtdev(tgt_device);
@@ -1001,8 +975,8 @@ error:
 
 	btrfs_assign_next_active_device(src_device, tgt_device);
 
-	list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
-	fs_info->fs_devices->rw_devices++;
+	list_add(&tgt_device->dev_alloc_list, &fs_devices->alloc_list);
+	fs_devices->rw_devices++;
 
 	up_write(&dev_replace->rwsem);
 	btrfs_rm_dev_replace_blocked(fs_info);
@@ -1025,7 +999,7 @@ error:
 	 * belong to this filesystem.
 	 */
 	mutex_unlock(&fs_info->chunk_mutex);
-	mutex_unlock(&fs_info->fs_devices->device_list_mutex);
+	mutex_unlock(&fs_devices->device_list_mutex);
 
 	/* replace the sysfs entry */
 	btrfs_sysfs_remove_device(src_device);
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index 3b532bab0755..72fb2c518a2b 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -325,36 +325,15 @@ btrfs_search_dir_index_item(struct btrfs_root *root,
 			    struct btrfs_path *path, u64 dirid,
 			    const char *name, int name_len)
 {
-	struct extent_buffer *leaf;
 	struct btrfs_dir_item *di;
 	struct btrfs_key key;
-	u32 nritems;
 	int ret;
 
 	key.objectid = dirid;
 	key.type = BTRFS_DIR_INDEX_KEY;
 	key.offset = 0;
 
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		return ERR_PTR(ret);
-
-	leaf = path->nodes[0];
-	nritems = btrfs_header_nritems(leaf);
-
-	while (1) {
-		if (path->slots[0] >= nritems) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				return ERR_PTR(ret);
-			if (ret > 0)
-				break;
-			leaf = path->nodes[0];
-			nritems = btrfs_header_nritems(leaf);
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+	btrfs_for_each_slot(root, &key, &key, path, ret) {
 		if (key.objectid != dirid || key.type != BTRFS_DIR_INDEX_KEY)
 			break;
 
@@ -362,10 +341,12 @@ btrfs_search_dir_index_item(struct btrfs_root *root,
 					       name, name_len);
 		if (di)
 			return di;
-
-		path->slots[0]++;
 	}
-	return NULL;
+	/* Adjust return code if the key was not found in the next leaf. */
+	if (ret > 0)
+		ret = 0;
+
+	return ERR_PTR(ret);
 }
 
 struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans,
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 84795d831282..14f8a90df321 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -5,7 +5,6 @@
 
 #include <linux/fs.h>
 #include <linux/blkdev.h>
-#include <linux/radix-tree.h>
 #include <linux/writeback.h>
 #include <linux/workqueue.h>
 #include <linux/kthread.h>
@@ -374,9 +373,9 @@ int btrfs_verify_level_key(struct extent_buffer *eb, int level,
  * @level:		expected level, mandatory check
  * @first_key:		expected key of first slot, skip check if NULL
  */
-static int btree_read_extent_buffer_pages(struct extent_buffer *eb,
-					  u64 parent_transid, int level,
-					  struct btrfs_key *first_key)
+int btrfs_read_extent_buffer(struct extent_buffer *eb,
+			     u64 parent_transid, int level,
+			     struct btrfs_key *first_key)
 {
 	struct btrfs_fs_info *fs_info = eb->fs_info;
 	struct extent_io_tree *io_tree;
@@ -486,7 +485,7 @@ static int csum_dirty_subpage_buffers(struct btrfs_fs_info *fs_info,
 		uptodate = btrfs_subpage_test_uptodate(fs_info, page, cur,
 						       fs_info->nodesize);
 
-		/* A dirty eb shouldn't disappear from buffer_radix */
+		/* A dirty eb shouldn't disappear from extent_buffers */
 		if (WARN_ON(!eb))
 			return -EUCLEAN;
 
@@ -519,7 +518,7 @@ static int csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct bio_vec *bvec
 	u64 found_start;
 	struct extent_buffer *eb;
 
-	if (fs_info->sectorsize < PAGE_SIZE)
+	if (fs_info->nodesize < PAGE_SIZE)
 		return csum_dirty_subpage_buffers(fs_info, bvec);
 
 	eb = (struct extent_buffer *)page->private;
@@ -704,7 +703,7 @@ int btrfs_validate_metadata_buffer(struct btrfs_bio *bbio,
 
 	ASSERT(page->private);
 
-	if (btrfs_sb(page->mapping->host->i_sb)->sectorsize < PAGE_SIZE)
+	if (btrfs_sb(page->mapping->host->i_sb)->nodesize < PAGE_SIZE)
 		return validate_subpage_buffer(page, start, end, mirror);
 
 	eb = (struct extent_buffer *)page->private;
@@ -850,8 +849,7 @@ static void run_one_async_free(struct btrfs_work *work)
 }
 
 blk_status_t btrfs_wq_submit_bio(struct inode *inode, struct bio *bio,
-				 int mirror_num, unsigned long bio_flags,
-				 u64 dio_file_offset,
+				 int mirror_num, u64 dio_file_offset,
 				 extent_submit_bio_start_t *submit_bio_start)
 {
 	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
@@ -874,9 +872,9 @@ blk_status_t btrfs_wq_submit_bio(struct inode *inode, struct bio *bio,
 	async->status = 0;
 
 	if (op_is_sync(bio->bi_opf))
-		btrfs_set_work_high_priority(&async->work);
-
-	btrfs_queue_work(fs_info->workers, &async->work);
+		btrfs_queue_work(fs_info->hipri_workers, &async->work);
+	else
+		btrfs_queue_work(fs_info->workers, &async->work);
 	return 0;
 }
 
@@ -920,8 +918,7 @@ static bool should_async_write(struct btrfs_fs_info *fs_info,
 	return true;
 }
 
-blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio,
-				       int mirror_num, unsigned long bio_flags)
+void btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio, int mirror_num)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	blk_status_t ret;
@@ -933,31 +930,25 @@ blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio,
 		 */
 		ret = btrfs_bio_wq_end_io(fs_info, bio,
 					  BTRFS_WQ_ENDIO_METADATA);
-		if (ret)
-			goto out_w_error;
-		ret = btrfs_map_bio(fs_info, bio, mirror_num);
+		if (!ret)
+			ret = btrfs_map_bio(fs_info, bio, mirror_num);
 	} else if (!should_async_write(fs_info, BTRFS_I(inode))) {
 		ret = btree_csum_one_bio(bio);
-		if (ret)
-			goto out_w_error;
-		ret = btrfs_map_bio(fs_info, bio, mirror_num);
+		if (!ret)
+			ret = btrfs_map_bio(fs_info, bio, mirror_num);
 	} else {
 		/*
 		 * kthread helpers are used to submit writes so that
 		 * checksumming can happen in parallel across all CPUs
 		 */
 		ret = btrfs_wq_submit_bio(inode, bio, mirror_num, 0,
-					  0, btree_submit_bio_start);
+					  btree_submit_bio_start);
 	}
 
-	if (ret)
-		goto out_w_error;
-	return 0;
-
-out_w_error:
-	bio->bi_status = ret;
-	bio_endio(bio);
-	return ret;
+	if (ret) {
+		bio->bi_status = ret;
+		bio_endio(bio);
+	}
 }
 
 #ifdef CONFIG_MIGRATION
@@ -1118,12 +1109,15 @@ struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
 	if (IS_ERR(buf))
 		return buf;
 
-	ret = btree_read_extent_buffer_pages(buf, parent_transid,
-					     level, first_key);
+	ret = btrfs_read_extent_buffer(buf, parent_transid, level, first_key);
 	if (ret) {
 		free_extent_buffer_stale(buf);
 		return ERR_PTR(ret);
 	}
+	if (btrfs_check_eb_owner(buf, owner_root)) {
+		free_extent_buffer_stale(buf);
+		return ERR_PTR(-EUCLEAN);
+	}
 	return buf;
 
 }
@@ -1164,7 +1158,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->nr_delalloc_inodes = 0;
 	root->nr_ordered_extents = 0;
 	root->inode_tree = RB_ROOT;
-	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
+	xa_init_flags(&root->delayed_nodes, GFP_ATOMIC);
 
 	btrfs_init_root_block_rsv(root);
 
@@ -1216,9 +1210,9 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	btrfs_qgroup_init_swapped_blocks(&root->swapped_blocks);
 #ifdef CONFIG_BTRFS_DEBUG
 	INIT_LIST_HEAD(&root->leak_list);
-	spin_lock(&fs_info->fs_roots_radix_lock);
+	spin_lock(&fs_info->fs_roots_lock);
 	list_add_tail(&root->leak_list, &fs_info->allocated_roots);
-	spin_unlock(&fs_info->fs_roots_radix_lock);
+	spin_unlock(&fs_info->fs_roots_lock);
 #endif
 }
 
@@ -1563,6 +1557,23 @@ static struct btrfs_root *read_tree_root_path(struct btrfs_root *tree_root,
 		ret = -EIO;
 		goto fail;
 	}
+
+	/*
+	 * For real fs, and not log/reloc trees, root owner must
+	 * match its root node owner
+	 */
+	if (!test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state) &&
+	    root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID &&
+	    root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID &&
+	    root->root_key.objectid != btrfs_header_owner(root->node)) {
+		btrfs_crit(fs_info,
+"root=%llu block=%llu, tree root owner mismatch, have %llu expect %llu",
+			   root->root_key.objectid, root->node->start,
+			   btrfs_header_owner(root->node),
+			   root->root_key.objectid);
+		ret = -EUCLEAN;
+		goto fail;
+	}
 	root->commit_root = btrfs_root_node(root);
 	return root;
 fail:
@@ -1648,12 +1659,11 @@ static struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_root *root;
 
-	spin_lock(&fs_info->fs_roots_radix_lock);
-	root = radix_tree_lookup(&fs_info->fs_roots_radix,
-				 (unsigned long)root_id);
+	spin_lock(&fs_info->fs_roots_lock);
+	root = xa_load(&fs_info->fs_roots, (unsigned long)root_id);
 	if (root)
 		root = btrfs_grab_root(root);
-	spin_unlock(&fs_info->fs_roots_radix_lock);
+	spin_unlock(&fs_info->fs_roots_lock);
 	return root;
 }
 
@@ -1695,20 +1705,14 @@ int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info,
 {
 	int ret;
 
-	ret = radix_tree_preload(GFP_NOFS);
-	if (ret)
-		return ret;
-
-	spin_lock(&fs_info->fs_roots_radix_lock);
-	ret = radix_tree_insert(&fs_info->fs_roots_radix,
-				(unsigned long)root->root_key.objectid,
-				root);
+	spin_lock(&fs_info->fs_roots_lock);
+	ret = xa_insert(&fs_info->fs_roots, (unsigned long)root->root_key.objectid,
+			root, GFP_NOFS);
 	if (ret == 0) {
 		btrfs_grab_root(root);
-		set_bit(BTRFS_ROOT_IN_RADIX, &root->state);
+		set_bit(BTRFS_ROOT_REGISTERED, &root->state);
 	}
-	spin_unlock(&fs_info->fs_roots_radix_lock);
-	radix_tree_preload_end();
+	spin_unlock(&fs_info->fs_roots_lock);
 
 	return ret;
 }
@@ -1964,7 +1968,7 @@ static void end_workqueue_fn(struct btrfs_work *work)
 
 static int cleaner_kthread(void *arg)
 {
-	struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)arg;
+	struct btrfs_fs_info *fs_info = arg;
 	int again;
 
 	while (1) {
@@ -2266,10 +2270,12 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
 {
 	btrfs_destroy_workqueue(fs_info->fixup_workers);
 	btrfs_destroy_workqueue(fs_info->delalloc_workers);
+	btrfs_destroy_workqueue(fs_info->hipri_workers);
 	btrfs_destroy_workqueue(fs_info->workers);
 	btrfs_destroy_workqueue(fs_info->endio_workers);
 	btrfs_destroy_workqueue(fs_info->endio_raid56_workers);
-	btrfs_destroy_workqueue(fs_info->rmw_workers);
+	if (fs_info->rmw_workers)
+		destroy_workqueue(fs_info->rmw_workers);
 	btrfs_destroy_workqueue(fs_info->endio_write_workers);
 	btrfs_destroy_workqueue(fs_info->endio_freespace_worker);
 	btrfs_destroy_workqueue(fs_info->delayed_workers);
@@ -2336,9 +2342,9 @@ void btrfs_put_root(struct btrfs_root *root)
 		btrfs_drew_lock_destroy(&root->snapshot_lock);
 		free_root_extent_buffers(root);
 #ifdef CONFIG_BTRFS_DEBUG
-		spin_lock(&root->fs_info->fs_roots_radix_lock);
+		spin_lock(&root->fs_info->fs_roots_lock);
 		list_del_init(&root->leak_list);
-		spin_unlock(&root->fs_info->fs_roots_radix_lock);
+		spin_unlock(&root->fs_info->fs_roots_lock);
 #endif
 		kfree(root);
 	}
@@ -2346,28 +2352,21 @@ void btrfs_put_root(struct btrfs_root *root)
 
 void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info)
 {
-	int ret;
-	struct btrfs_root *gang[8];
-	int i;
+	struct btrfs_root *root;
+	unsigned long index = 0;
 
 	while (!list_empty(&fs_info->dead_roots)) {
-		gang[0] = list_entry(fs_info->dead_roots.next,
-				     struct btrfs_root, root_list);
-		list_del(&gang[0]->root_list);
+		root = list_entry(fs_info->dead_roots.next,
+				  struct btrfs_root, root_list);
+		list_del(&root->root_list);
 
-		if (test_bit(BTRFS_ROOT_IN_RADIX, &gang[0]->state))
-			btrfs_drop_and_free_fs_root(fs_info, gang[0]);
-		btrfs_put_root(gang[0]);
+		if (test_bit(BTRFS_ROOT_REGISTERED, &root->state))
+			btrfs_drop_and_free_fs_root(fs_info, root);
+		btrfs_put_root(root);
 	}
 
-	while (1) {
-		ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
-					     (void **)gang, 0,
-					     ARRAY_SIZE(gang));
-		if (!ret)
-			break;
-		for (i = 0; i < ret; i++)
-			btrfs_drop_and_free_fs_root(fs_info, gang[i]);
+	xa_for_each(&fs_info->fs_roots, index, root) {
+		btrfs_drop_and_free_fs_root(fs_info, root);
 	}
 }
 
@@ -2444,7 +2443,9 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info)
 	unsigned int flags = WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_UNBOUND;
 
 	fs_info->workers =
-		btrfs_alloc_workqueue(fs_info, "worker",
+		btrfs_alloc_workqueue(fs_info, "worker", flags, max_active, 16);
+	fs_info->hipri_workers =
+		btrfs_alloc_workqueue(fs_info, "worker-high",
 				      flags | WQ_HIGHPRI, max_active, 16);
 
 	fs_info->delalloc_workers =
@@ -2476,8 +2477,7 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info)
 	fs_info->endio_raid56_workers =
 		btrfs_alloc_workqueue(fs_info, "endio-raid56", flags,
 				      max_active, 4);
-	fs_info->rmw_workers =
-		btrfs_alloc_workqueue(fs_info, "rmw", flags, max_active, 2);
+	fs_info->rmw_workers = alloc_workqueue("btrfs-rmw", flags, max_active);
 	fs_info->endio_write_workers =
 		btrfs_alloc_workqueue(fs_info, "endio-write", flags,
 				      max_active, 2);
@@ -2492,8 +2492,8 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info)
 	fs_info->discard_ctl.discard_workers =
 		alloc_workqueue("btrfs_discard", WQ_UNBOUND | WQ_FREEZABLE, 1);
 
-	if (!(fs_info->workers && fs_info->delalloc_workers &&
-	      fs_info->flush_workers &&
+	if (!(fs_info->workers && fs_info->hipri_workers &&
+	      fs_info->delalloc_workers && fs_info->flush_workers &&
 	      fs_info->endio_workers && fs_info->endio_meta_workers &&
 	      fs_info->endio_meta_write_workers &&
 	      fs_info->endio_write_workers && fs_info->endio_raid56_workers &&
@@ -2815,12 +2815,14 @@ static int validate_super(struct btrfs_fs_info *fs_info,
 	}
 
 	/*
-	 * For 4K page size, we only support 4K sector size.
-	 * For 64K page size, we support 64K and 4K sector sizes.
+	 * We only support at most two sectorsizes: 4K and PAGE_SIZE.
+	 *
+	 * We can support 16K sectorsize with 64K page size without problem,
+	 * but such sectorsize/pagesize combination doesn't make much sense.
+	 * 4K will be our future standard, PAGE_SIZE is supported from the very
+	 * beginning.
 	 */
-	if ((PAGE_SIZE == SZ_4K && sectorsize != PAGE_SIZE) ||
-	    (PAGE_SIZE == SZ_64K && (sectorsize != SZ_4K &&
-				     sectorsize != SZ_64K))) {
+	if (sectorsize > PAGE_SIZE || (sectorsize != SZ_4K && sectorsize != PAGE_SIZE)) {
 		btrfs_err(fs_info,
 			"sectorsize %llu not yet supported for page size %lu",
 			sectorsize, PAGE_SIZE);
@@ -3132,8 +3134,8 @@ static int __cold init_tree_roots(struct btrfs_fs_info *fs_info)
 
 void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
 {
-	INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
-	INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC);
+	xa_init_flags(&fs_info->fs_roots, GFP_ATOMIC);
+	xa_init_flags(&fs_info->extent_buffers, GFP_ATOMIC);
 	INIT_LIST_HEAD(&fs_info->trans_list);
 	INIT_LIST_HEAD(&fs_info->dead_roots);
 	INIT_LIST_HEAD(&fs_info->delayed_iputs);
@@ -3141,7 +3143,7 @@ void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
 	INIT_LIST_HEAD(&fs_info->caching_block_groups);
 	spin_lock_init(&fs_info->delalloc_root_lock);
 	spin_lock_init(&fs_info->trans_lock);
-	spin_lock_init(&fs_info->fs_roots_radix_lock);
+	spin_lock_init(&fs_info->fs_roots_lock);
 	spin_lock_init(&fs_info->delayed_iput_lock);
 	spin_lock_init(&fs_info->defrag_inodes_lock);
 	spin_lock_init(&fs_info->super_lock);
@@ -3209,9 +3211,8 @@ void btrfs_init_fs_info(struct btrfs_fs_info *fs_info)
 	btrfs_init_balance(fs_info);
 	btrfs_init_async_reclaim_work(fs_info);
 
-	spin_lock_init(&fs_info->block_group_cache_lock);
-	fs_info->block_group_cache_tree = RB_ROOT;
-	fs_info->first_logical_byte = (u64)-1;
+	rwlock_init(&fs_info->block_group_cache_lock);
+	fs_info->block_group_cache_tree = RB_ROOT_CACHED;
 
 	extent_io_tree_init(fs_info, &fs_info->excluded_extents,
 			    IO_TREE_FS_EXCLUDED_EXTENTS, NULL);
@@ -3295,7 +3296,7 @@ static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block
 
 static int btrfs_uuid_rescan_kthread(void *data)
 {
-	struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)data;
+	struct btrfs_fs_info *fs_info = data;
 	int ret;
 
 	/*
@@ -3373,7 +3374,7 @@ int btrfs_start_pre_rw_mount(struct btrfs_fs_info *fs_info)
 	/*
 	 * btrfs_find_orphan_roots() is responsible for finding all the dead
 	 * roots (with 0 refs), flag them with BTRFS_ROOT_DEAD_TREE and load
-	 * them into the fs_info->fs_roots_radix tree. This must be done before
+	 * them into the fs_info->fs_roots. This must be done before
 	 * calling btrfs_orphan_cleanup() on the tree root. If we don't do it
 	 * first, then btrfs_orphan_cleanup() will delete a dead root's orphan
 	 * item before the root's tree is deleted - this means that if we unmount
@@ -3611,7 +3612,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 		~BTRFS_FEATURE_INCOMPAT_SUPP;
 	if (features) {
 		btrfs_err(fs_info,
-		    "cannot mount because of unsupported optional features (%llx)",
+		    "cannot mount because of unsupported optional features (0x%llx)",
 		    features);
 		err = -EINVAL;
 		goto fail_alloc;
@@ -3649,7 +3650,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 		~BTRFS_FEATURE_COMPAT_RO_SUPP;
 	if (!sb_rdonly(sb) && features) {
 		btrfs_err(fs_info,
-	"cannot mount read-write because of unsupported optional features (%llx)",
+	"cannot mount read-write because of unsupported optional features (0x%llx)",
 		       features);
 		err = -EINVAL;
 		goto fail_alloc;
@@ -3672,14 +3673,6 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device
 		btrfs_warn(fs_info,
 		"read-write for sector size %u with page size %lu is experimental",
 			   sectorsize, PAGE_SIZE);
-		if (btrfs_super_incompat_flags(fs_info->super_copy) &
-			BTRFS_FEATURE_INCOMPAT_RAID56) {
-			btrfs_err(fs_info,
-		"RAID56 is not yet supported for sector size %u with page size %lu",
-				sectorsize, PAGE_SIZE);
-			err = -EINVAL;
-			goto fail_alloc;
-		}
 		subpage_info = kzalloc(sizeof(*subpage_info), GFP_KERNEL);
 		if (!subpage_info)
 			goto fail_alloc;
@@ -4157,7 +4150,8 @@ static int write_dev_supers(struct btrfs_device *device,
 		if (i == 0 && !btrfs_test_opt(device->fs_info, NOBARRIER))
 			bio->bi_opf |= REQ_FUA;
 
-		btrfsic_submit_bio(bio);
+		btrfsic_check_bio(bio);
+		submit_bio(bio);
 
 		if (btrfs_advance_sb_log(device, i))
 			errors++;
@@ -4271,7 +4265,8 @@ static void write_dev_flush(struct btrfs_device *device)
 	init_completion(&device->flush_wait);
 	bio->bi_private = &device->flush_wait;
 
-	btrfsic_submit_bio(bio);
+	btrfsic_check_bio(bio);
+	submit_bio(bio);
 	set_bit(BTRFS_DEV_STATE_FLUSH_SENT, &device->dev_state);
 }
 
@@ -4504,12 +4499,11 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 {
 	bool drop_ref = false;
 
-	spin_lock(&fs_info->fs_roots_radix_lock);
-	radix_tree_delete(&fs_info->fs_roots_radix,
-			  (unsigned long)root->root_key.objectid);
-	if (test_and_clear_bit(BTRFS_ROOT_IN_RADIX, &root->state))
+	spin_lock(&fs_info->fs_roots_lock);
+	xa_erase(&fs_info->fs_roots, (unsigned long)root->root_key.objectid);
+	if (test_and_clear_bit(BTRFS_ROOT_REGISTERED, &root->state))
 		drop_ref = true;
-	spin_unlock(&fs_info->fs_roots_radix_lock);
+	spin_unlock(&fs_info->fs_roots_lock);
 
 	if (BTRFS_FS_ERROR(fs_info)) {
 		ASSERT(root->log_root == NULL);
@@ -4525,50 +4519,48 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 
 int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info)
 {
-	u64 root_objectid = 0;
-	struct btrfs_root *gang[8];
-	int i = 0;
+	struct btrfs_root *roots[8];
+	unsigned long index = 0;
+	int i;
 	int err = 0;
-	unsigned int ret = 0;
+	int grabbed;
 
 	while (1) {
-		spin_lock(&fs_info->fs_roots_radix_lock);
-		ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
-					     (void **)gang, root_objectid,
-					     ARRAY_SIZE(gang));
-		if (!ret) {
-			spin_unlock(&fs_info->fs_roots_radix_lock);
-			break;
+		struct btrfs_root *root;
+
+		spin_lock(&fs_info->fs_roots_lock);
+		if (!xa_find(&fs_info->fs_roots, &index, ULONG_MAX, XA_PRESENT)) {
+			spin_unlock(&fs_info->fs_roots_lock);
+			return err;
 		}
-		root_objectid = gang[ret - 1]->root_key.objectid + 1;
 
-		for (i = 0; i < ret; i++) {
-			/* Avoid to grab roots in dead_roots */
-			if (btrfs_root_refs(&gang[i]->root_item) == 0) {
-				gang[i] = NULL;
-				continue;
-			}
-			/* grab all the search result for later use */
-			gang[i] = btrfs_grab_root(gang[i]);
+		grabbed = 0;
+		xa_for_each_start(&fs_info->fs_roots, index, root, index) {
+			/* Avoid grabbing roots in dead_roots */
+			if (btrfs_root_refs(&root->root_item) > 0)
+				roots[grabbed++] = btrfs_grab_root(root);
+			if (grabbed >= ARRAY_SIZE(roots))
+				break;
 		}
-		spin_unlock(&fs_info->fs_roots_radix_lock);
+		spin_unlock(&fs_info->fs_roots_lock);
 
-		for (i = 0; i < ret; i++) {
-			if (!gang[i])
+		for (i = 0; i < grabbed; i++) {
+			if (!roots[i])
 				continue;
-			root_objectid = gang[i]->root_key.objectid;
-			err = btrfs_orphan_cleanup(gang[i]);
+			index = roots[i]->root_key.objectid;
+			err = btrfs_orphan_cleanup(roots[i]);
 			if (err)
-				break;
-			btrfs_put_root(gang[i]);
+				goto out;
+			btrfs_put_root(roots[i]);
 		}
-		root_objectid++;
+		index++;
 	}
 
-	/* release the uncleaned roots due to error */
-	for (; i < ret; i++) {
-		if (gang[i])
-			btrfs_put_root(gang[i]);
+out:
+	/* Release the roots that remain uncleaned due to error */
+	for (; i < grabbed; i++) {
+		if (roots[i])
+			btrfs_put_root(roots[i]);
 	}
 	return err;
 }
@@ -4863,13 +4855,6 @@ void btrfs_btree_balance_dirty_nodelay(struct btrfs_fs_info *fs_info)
 	__btrfs_btree_balance_dirty(fs_info, 0);
 }
 
-int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid, int level,
-		      struct btrfs_key *first_key)
-{
-	return btree_read_extent_buffer_pages(buf, parent_transid,
-					      level, first_key);
-}
-
 static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info)
 {
 	/* cleanup FS via transaction */
@@ -4885,31 +4870,28 @@ static void btrfs_error_commit_super(struct btrfs_fs_info *fs_info)
 
 static void btrfs_drop_all_logs(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_root *gang[8];
-	u64 root_objectid = 0;
-	int ret;
-
-	spin_lock(&fs_info->fs_roots_radix_lock);
-	while ((ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix,
-					     (void **)gang, root_objectid,
-					     ARRAY_SIZE(gang))) != 0) {
-		int i;
+	unsigned long index = 0;
+	int grabbed = 0;
+	struct btrfs_root *roots[8];
 
-		for (i = 0; i < ret; i++)
-			gang[i] = btrfs_grab_root(gang[i]);
-		spin_unlock(&fs_info->fs_roots_radix_lock);
+	spin_lock(&fs_info->fs_roots_lock);
+	while ((grabbed = xa_extract(&fs_info->fs_roots, (void **)roots, index,
+				     ULONG_MAX, 8, XA_PRESENT))) {
+		for (int i = 0; i < grabbed; i++)
+			roots[i] = btrfs_grab_root(roots[i]);
+		spin_unlock(&fs_info->fs_roots_lock);
 
-		for (i = 0; i < ret; i++) {
-			if (!gang[i])
+		for (int i = 0; i < grabbed; i++) {
+			if (!roots[i])
 				continue;
-			root_objectid = gang[i]->root_key.objectid;
-			btrfs_free_log(NULL, gang[i]);
-			btrfs_put_root(gang[i]);
+			index = roots[i]->root_key.objectid;
+			btrfs_free_log(NULL, roots[i]);
+			btrfs_put_root(roots[i]);
 		}
-		root_objectid++;
-		spin_lock(&fs_info->fs_roots_radix_lock);
+		index++;
+		spin_lock(&fs_info->fs_roots_lock);
 	}
-	spin_unlock(&fs_info->fs_roots_radix_lock);
+	spin_unlock(&fs_info->fs_roots_lock);
 	btrfs_free_log_root_tree(NULL, fs_info);
 }
 
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 2e10514ecda8..4ee8c42c9f78 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -87,8 +87,7 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info,
 int btrfs_validate_metadata_buffer(struct btrfs_bio *bbio,
 				   struct page *page, u64 start, u64 end,
 				   int mirror);
-blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio,
-				       int mirror_num, unsigned long bio_flags);
+void btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio, int mirror_num);
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info);
 #endif
@@ -120,13 +119,12 @@ void btrfs_put_root(struct btrfs_root *root);
 void btrfs_mark_buffer_dirty(struct extent_buffer *buf);
 int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid,
 			  int atomic);
-int btrfs_read_buffer(struct extent_buffer *buf, u64 parent_transid, int level,
-		      struct btrfs_key *first_key);
+int btrfs_read_extent_buffer(struct extent_buffer *buf, u64 parent_transid,
+			     int level, struct btrfs_key *first_key);
 blk_status_t btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
 			enum btrfs_wq_endio_type metadata);
 blk_status_t btrfs_wq_submit_bio(struct inode *inode, struct bio *bio,
-				 int mirror_num, unsigned long bio_flags,
-				 u64 dio_file_offset,
+				 int mirror_num, u64 dio_file_offset,
 				 extent_submit_bio_start_t *submit_bio_start);
 blk_status_t btrfs_submit_bio_done(void *private_data, struct bio *bio,
 			  int mirror_num);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 6260784e74b5..0867c5cd6e01 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -895,7 +895,13 @@ again:
 	err = -ENOENT;
 	while (1) {
 		if (ptr >= end) {
-			WARN_ON(ptr > end);
+			if (ptr > end) {
+				err = -EUCLEAN;
+				btrfs_print_leaf(path->nodes[0]);
+				btrfs_crit(fs_info,
+"overrun extent record at slot %d while looking for inline extent for root %llu owner %llu offset %llu parent %llu",
+					path->slots[0], root_objectid, owner, offset, parent);
+			}
 			break;
 		}
 		iref = (struct btrfs_extent_inline_ref *)ptr;
@@ -1577,12 +1583,12 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans,
 	u32 item_size;
 	int ret;
 	int err = 0;
-	int metadata = !extent_op->is_data;
+	int metadata = 1;
 
 	if (TRANS_ABORTED(trans))
 		return 0;
 
-	if (metadata && !btrfs_fs_incompat(fs_info, SKINNY_METADATA))
+	if (!btrfs_fs_incompat(fs_info, SKINNY_METADATA))
 		metadata = 0;
 
 	path = btrfs_alloc_path();
@@ -2180,7 +2186,7 @@ out:
 
 int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
 				struct extent_buffer *eb, u64 flags,
-				int level, int is_data)
+				int level)
 {
 	struct btrfs_delayed_extent_op *extent_op;
 	int ret;
@@ -2192,7 +2198,6 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
 	extent_op->flags_to_set = flags;
 	extent_op->update_flags = true;
 	extent_op->update_key = false;
-	extent_op->is_data = is_data ? true : false;
 	extent_op->level = level;
 
 	ret = btrfs_add_delayed_extent_op(trans, eb->start, eb->len, extent_op);
@@ -2357,15 +2362,10 @@ out:
 }
 
 int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
-			  u64 bytenr, bool strict)
+			  u64 bytenr, bool strict, struct btrfs_path *path)
 {
-	struct btrfs_path *path;
 	int ret;
 
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-
 	do {
 		ret = check_committed_ref(root, path, objectid,
 					  offset, bytenr, strict);
@@ -2376,7 +2376,7 @@ int btrfs_cross_ref_exist(struct btrfs_root *root, u64 objectid, u64 offset,
 	} while (ret == -EAGAIN);
 
 out:
-	btrfs_free_path(path);
+	btrfs_release_path(path);
 	if (btrfs_is_data_reloc_root(root))
 		WARN_ON(ret > 0);
 	return ret;
@@ -2497,24 +2497,21 @@ static u64 get_alloc_profile_by_root(struct btrfs_root *root, int data)
 	return ret;
 }
 
-static u64 first_logical_byte(struct btrfs_fs_info *fs_info, u64 search_start)
+static u64 first_logical_byte(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_block_group *cache;
-	u64 bytenr;
-
-	spin_lock(&fs_info->block_group_cache_lock);
-	bytenr = fs_info->first_logical_byte;
-	spin_unlock(&fs_info->block_group_cache_lock);
-
-	if (bytenr < (u64)-1)
-		return bytenr;
+	struct rb_node *leftmost;
+	u64 bytenr = 0;
 
-	cache = btrfs_lookup_first_block_group(fs_info, search_start);
-	if (!cache)
-		return 0;
+	read_lock(&fs_info->block_group_cache_lock);
+	/* Get the block group with the lowest logical start address. */
+	leftmost = rb_first_cached(&fs_info->block_group_cache_tree);
+	if (leftmost) {
+		struct btrfs_block_group *bg;
 
-	bytenr = cache->start;
-	btrfs_put_block_group(cache);
+		bg = rb_entry(leftmost, struct btrfs_block_group, cache_node);
+		bytenr = bg->start;
+	}
+	read_unlock(&fs_info->block_group_cache_lock);
 
 	return bytenr;
 }
@@ -3803,8 +3800,7 @@ static int do_allocation_zoned(struct btrfs_block_group *block_group,
 
 	/* Check RO and no space case before trying to activate it */
 	spin_lock(&block_group->lock);
-	if (block_group->ro ||
-	    block_group->alloc_offset == block_group->zone_capacity) {
+	if (block_group->ro || btrfs_zoned_bg_is_full(block_group)) {
 		ret = 1;
 		/*
 		 * May need to clear fs_info->{treelog,data_reloc}_bg.
@@ -4272,7 +4268,7 @@ static noinline int find_free_extent(struct btrfs_root *root,
 		return ret;
 
 	ffe_ctl->search_start = max(ffe_ctl->search_start,
-				    first_logical_byte(fs_info, 0));
+				    first_logical_byte(fs_info));
 	ffe_ctl->search_start = max(ffe_ctl->search_start, ffe_ctl->hint_byte);
 	if (ffe_ctl->search_start == ffe_ctl->hint_byte) {
 		block_group = btrfs_lookup_block_group(fs_info,
@@ -4959,7 +4955,6 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans,
 		extent_op->flags_to_set = flags;
 		extent_op->update_key = skinny_metadata ? false : true;
 		extent_op->update_flags = true;
-		extent_op->is_data = false;
 		extent_op->level = level;
 
 		btrfs_init_generic_ref(&generic_ref, BTRFS_ADD_DELAYED_EXTENT,
@@ -5144,7 +5139,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
 		ret = btrfs_dec_ref(trans, root, eb, 0);
 		BUG_ON(ret); /* -ENOMEM */
 		ret = btrfs_set_disk_extent_flags(trans, eb, flag,
-						  btrfs_header_level(eb), 0);
+						  btrfs_header_level(eb));
 		BUG_ON(ret); /* -ENOMEM */
 		wc->flags[level] |= flag;
 	}
@@ -5818,7 +5813,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, int for_reloc)
 	btrfs_qgroup_convert_reserved_meta(root, INT_MAX);
 	btrfs_qgroup_free_meta_all_pertrans(root);
 
-	if (test_bit(BTRFS_ROOT_IN_RADIX, &root->state))
+	if (test_bit(BTRFS_ROOT_REGISTERED, &root->state))
 		btrfs_add_dropped_root(trans, root);
 	else
 		btrfs_put_root(root);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 33c19f51d79b..588c7c606a2c 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -6,6 +6,7 @@
 #include <linux/mm.h>
 #include <linux/pagemap.h>
 #include <linux/page-flags.h>
+#include <linux/sched/mm.h>
 #include <linux/spinlock.h>
 #include <linux/blkdev.h>
 #include <linux/swap.h>
@@ -28,6 +29,7 @@
 #include "subpage.h"
 #include "zoned.h"
 #include "block-group.h"
+#include "compression.h"
 
 static struct kmem_cache *extent_state_cache;
 static struct kmem_cache *extent_buffer_cache;
@@ -75,6 +77,7 @@ void btrfs_extent_buffer_leak_debug_check(struct btrfs_fs_info *fs_info)
 	if (!fs_info->allocated_ebs.next)
 		return;
 
+	WARN_ON(!list_empty(&fs_info->allocated_ebs));
 	spin_lock_irqsave(&fs_info->eb_leak_lock, flags);
 	while (!list_empty(&fs_info->allocated_ebs)) {
 		eb = list_first_entry(&fs_info->allocated_ebs,
@@ -135,6 +138,17 @@ struct tree_entry {
 	struct rb_node rb_node;
 };
 
+/*
+ * Structure to record info about the bio being assembled, and other info like
+ * how many bytes are there before stripe/ordered extent boundary.
+ */
+struct btrfs_bio_ctrl {
+	struct bio *bio;
+	enum btrfs_compression_type compress_type;
+	u32 len_to_stripe_boundary;
+	u32 len_to_oe_boundary;
+};
+
 struct extent_page_data {
 	struct btrfs_bio_ctrl bio_ctrl;
 	/* tells writepage not to lock the state bits for this range
@@ -164,24 +178,27 @@ static int add_extent_changeset(struct extent_state *state, u32 bits,
 	return ret;
 }
 
-int __must_check submit_one_bio(struct bio *bio, int mirror_num,
-				unsigned long bio_flags)
+static void submit_one_bio(struct bio *bio, int mirror_num,
+			   enum btrfs_compression_type compress_type)
 {
-	blk_status_t ret = 0;
 	struct extent_io_tree *tree = bio->bi_private;
 
 	bio->bi_private = NULL;
 
 	/* Caller should ensure the bio has at least some range added */
 	ASSERT(bio->bi_iter.bi_size);
+
 	if (is_data_inode(tree->private_data))
-		ret = btrfs_submit_data_bio(tree->private_data, bio, mirror_num,
-					    bio_flags);
+		btrfs_submit_data_bio(tree->private_data, bio, mirror_num,
+					    compress_type);
 	else
-		ret = btrfs_submit_metadata_bio(tree->private_data, bio,
-						mirror_num, bio_flags);
-
-	return blk_status_to_errno(ret);
+		btrfs_submit_metadata_bio(tree->private_data, bio, mirror_num);
+	/*
+	 * Above submission hooks will handle the error by ending the bio,
+	 * which will do the cleanup properly.  So here we should not return
+	 * any error, or the caller of submit_extent_page() will do cleanup
+	 * again, causing problems.
+	 */
 }
 
 /* Cleanup unsubmitted bios */
@@ -202,13 +219,12 @@ static void end_write_bio(struct extent_page_data *epd, int ret)
  * Return 0 if everything is OK.
  * Return <0 for error.
  */
-static int __must_check flush_write_bio(struct extent_page_data *epd)
+static void flush_write_bio(struct extent_page_data *epd)
 {
-	int ret = 0;
 	struct bio *bio = epd->bio_ctrl.bio;
 
 	if (bio) {
-		ret = submit_one_bio(bio, 0, 0);
+		submit_one_bio(bio, 0, 0);
 		/*
 		 * Clean up of epd->bio is handled by its endio function.
 		 * And endio is either triggered by successful bio execution
@@ -218,7 +234,6 @@ static int __must_check flush_write_bio(struct extent_page_data *epd)
 		 */
 		epd->bio_ctrl.bio = NULL;
 	}
-	return ret;
 }
 
 int __init extent_state_cache_init(void)
@@ -2303,12 +2318,13 @@ static int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 			     u64 length, u64 logical, struct page *page,
 			     unsigned int pg_offset, int mirror_num)
 {
-	struct bio *bio;
 	struct btrfs_device *dev;
+	struct bio_vec bvec;
+	struct bio bio;
 	u64 map_length = 0;
 	u64 sector;
 	struct btrfs_io_context *bioc = NULL;
-	int ret;
+	int ret = 0;
 
 	ASSERT(!(fs_info->sb->s_flags & SB_RDONLY));
 	BUG_ON(!mirror_num);
@@ -2316,8 +2332,6 @@ static int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 	if (btrfs_repair_one_zone(fs_info, logical))
 		return 0;
 
-	bio = btrfs_bio_alloc(1);
-	bio->bi_iter.bi_size = 0;
 	map_length = length;
 
 	/*
@@ -2335,52 +2349,50 @@ static int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start,
 		 */
 		ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, logical,
 				      &map_length, &bioc, 0);
-		if (ret) {
-			btrfs_bio_counter_dec(fs_info);
-			bio_put(bio);
-			return -EIO;
-		}
+		if (ret)
+			goto out_counter_dec;
 		ASSERT(bioc->mirror_num == 1);
 	} else {
 		ret = btrfs_map_block(fs_info, BTRFS_MAP_WRITE, logical,
 				      &map_length, &bioc, mirror_num);
-		if (ret) {
-			btrfs_bio_counter_dec(fs_info);
-			bio_put(bio);
-			return -EIO;
-		}
+		if (ret)
+			goto out_counter_dec;
 		BUG_ON(mirror_num != bioc->mirror_num);
 	}
 
 	sector = bioc->stripes[bioc->mirror_num - 1].physical >> 9;
-	bio->bi_iter.bi_sector = sector;
 	dev = bioc->stripes[bioc->mirror_num - 1].dev;
 	btrfs_put_bioc(bioc);
+
 	if (!dev || !dev->bdev ||
 	    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) {
-		btrfs_bio_counter_dec(fs_info);
-		bio_put(bio);
-		return -EIO;
+		ret = -EIO;
+		goto out_counter_dec;
 	}
-	bio_set_dev(bio, dev->bdev);
-	bio->bi_opf = REQ_OP_WRITE | REQ_SYNC;
-	bio_add_page(bio, page, length, pg_offset);
 
-	if (btrfsic_submit_bio_wait(bio)) {
+	bio_init(&bio, dev->bdev, &bvec, 1, REQ_OP_WRITE | REQ_SYNC);
+	bio.bi_iter.bi_sector = sector;
+	__bio_add_page(&bio, page, length, pg_offset);
+
+	btrfsic_check_bio(&bio);
+	ret = submit_bio_wait(&bio);
+	if (ret) {
 		/* try to remap that extent elsewhere? */
-		btrfs_bio_counter_dec(fs_info);
-		bio_put(bio);
 		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
-		return -EIO;
+		goto out_bio_uninit;
 	}
 
 	btrfs_info_rl_in_rcu(fs_info,
 		"read error corrected: ino %llu off %llu (dev %s sector %llu)",
 				  ino, start,
 				  rcu_str_deref(dev->name), sector);
+	ret = 0;
+
+out_bio_uninit:
+	bio_uninit(&bio);
+out_counter_dec:
 	btrfs_bio_counter_dec(fs_info);
-	bio_put(bio);
-	return 0;
+	return ret;
 }
 
 int btrfs_repair_eb_io_failure(const struct extent_buffer *eb, int mirror_num)
@@ -2527,7 +2539,7 @@ static struct io_failure_record *btrfs_get_io_failure_record(struct inode *inode
 	failrec->start = start;
 	failrec->len = sectorsize;
 	failrec->this_mirror = 0;
-	failrec->bio_flags = 0;
+	failrec->compress_type = BTRFS_COMPRESS_NONE;
 
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree, start, failrec->len);
@@ -2551,8 +2563,7 @@ static struct io_failure_record *btrfs_get_io_failure_record(struct inode *inode
 	logical = em->block_start + logical;
 	if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
 		logical = em->block_start;
-		failrec->bio_flags = EXTENT_BIO_COMPRESSED;
-		extent_set_compress_type(&failrec->bio_flags, em->compress_type);
+		failrec->compress_type = em->compress_type;
 	}
 
 	btrfs_debug(fs_info,
@@ -2684,7 +2695,7 @@ int btrfs_repair_one_sector(struct inode *inode,
 	 * will be handled by the endio on the repair_bio, so we can't return an
 	 * error here.
 	 */
-	submit_bio_hook(inode, repair_bio, failrec->this_mirror, failrec->bio_flags);
+	submit_bio_hook(inode, repair_bio, failrec->this_mirror, failrec->compress_type);
 	return BLK_STS_OK;
 }
 
@@ -2710,18 +2721,19 @@ static void end_page_read(struct page *page, bool uptodate, u64 start, u32 len)
 		btrfs_page_set_error(fs_info, page, start, len);
 	}
 
-	if (fs_info->sectorsize == PAGE_SIZE)
+	if (!btrfs_is_subpage(fs_info, page))
 		unlock_page(page);
 	else
 		btrfs_subpage_end_reader(fs_info, page, start, len);
 }
 
-static blk_status_t submit_read_repair(struct inode *inode,
-				      struct bio *failed_bio, u32 bio_offset,
-				      struct page *page, unsigned int pgoff,
-				      u64 start, u64 end, int failed_mirror,
-				      unsigned int error_bitmap,
-				      submit_bio_hook_t *submit_bio_hook)
+static blk_status_t submit_data_read_repair(struct inode *inode,
+					    struct bio *failed_bio,
+					    u32 bio_offset, struct page *page,
+					    unsigned int pgoff,
+					    u64 start, u64 end,
+					    int failed_mirror,
+					    unsigned int error_bitmap)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	const u32 sectorsize = fs_info->sectorsize;
@@ -2731,6 +2743,9 @@ static blk_status_t submit_read_repair(struct inode *inode,
 
 	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
+	/* This repair is only for data */
+	ASSERT(is_data_inode(inode));
+
 	/* We're here because we had some read errors or csum mismatch */
 	ASSERT(error_bitmap);
 
@@ -2759,7 +2774,7 @@ static blk_status_t submit_read_repair(struct inode *inode,
 		ret = btrfs_repair_one_sector(inode, failed_bio,
 				bio_offset + offset,
 				page, pgoff + offset, start + offset,
-				failed_mirror, submit_bio_hook);
+				failed_mirror, btrfs_submit_data_bio);
 		if (!ret) {
 			/*
 			 * We have submitted the read repair, the page release
@@ -2943,7 +2958,7 @@ update:
 static void begin_page_read(struct btrfs_fs_info *fs_info, struct page *page)
 {
 	ASSERT(PageLocked(page));
-	if (fs_info->sectorsize == PAGE_SIZE)
+	if (!btrfs_is_subpage(fs_info, page))
 		return;
 
 	ASSERT(PagePrivate(page));
@@ -2951,7 +2966,7 @@ static void begin_page_read(struct btrfs_fs_info *fs_info, struct page *page)
 }
 
 /*
- * Find extent buffer for a givne bytenr.
+ * Find extent buffer for a given bytenr.
  *
  * This is for end_bio_extent_readpage(), thus we can't do any unsafe locking
  * in endio context.
@@ -2965,16 +2980,14 @@ static struct extent_buffer *find_extent_buffer_readpage(
 	 * For regular sectorsize, we can use page->private to grab extent
 	 * buffer
 	 */
-	if (fs_info->sectorsize == PAGE_SIZE) {
+	if (fs_info->nodesize >= PAGE_SIZE) {
 		ASSERT(PagePrivate(page) && page->private);
 		return (struct extent_buffer *)page->private;
 	}
 
-	/* For subpage case, we need to lookup buffer radix tree */
-	rcu_read_lock();
-	eb = radix_tree_lookup(&fs_info->buffer_radix,
-			       bytenr >> fs_info->sectorsize_bits);
-	rcu_read_unlock();
+	/* For subpage case, we need to lookup extent buffer xarray */
+	eb = xa_load(&fs_info->extent_buffers,
+		     bytenr >> fs_info->sectorsize_bits);
 	ASSERT(eb);
 	return eb;
 }
@@ -3077,13 +3090,13 @@ static void end_bio_extent_readpage(struct bio *bio)
 				goto readpage_ok;
 
 			/*
-			 * btrfs_submit_read_repair() will handle all the good
+			 * submit_data_read_repair() will handle all the good
 			 * and bad sectors, we just continue to the next bvec.
 			 */
-			submit_read_repair(inode, bio, bio_offset, page,
-					   start - page_offset(page), start,
-					   end, mirror, error_bitmap,
-					   btrfs_submit_data_bio);
+			submit_data_read_repair(inode, bio, bio_offset, page,
+						start - page_offset(page),
+						start, end, mirror,
+						error_bitmap);
 
 			ASSERT(bio_offset + len > bio_offset);
 			bio_offset += len;
@@ -3132,6 +3145,42 @@ readpage_ok:
 	bio_put(bio);
 }
 
+/**
+ * Populate every free slot in a provided array with pages.
+ *
+ * @nr_pages:   number of pages to allocate
+ * @page_array: the array to fill with pages; any existing non-null entries in
+ * 		the array will be skipped
+ *
+ * Return: 0        if all pages were able to be allocated;
+ *         -ENOMEM  otherwise, and the caller is responsible for freeing all
+ *                  non-null page pointers in the array.
+ */
+int btrfs_alloc_page_array(unsigned int nr_pages, struct page **page_array)
+{
+	unsigned int allocated;
+
+	for (allocated = 0; allocated < nr_pages;) {
+		unsigned int last = allocated;
+
+		allocated = alloc_pages_bulk_array(GFP_NOFS, nr_pages, page_array);
+
+		if (allocated == nr_pages)
+			return 0;
+
+		/*
+		 * During this iteration, no page could be allocated, even
+		 * though alloc_pages_bulk_array() falls back to alloc_page()
+		 * if  it could not bulk-allocate. So we must be out of memory.
+		 */
+		if (allocated == last)
+			return -ENOMEM;
+
+		memalloc_retry_wait(GFP_NOFS);
+	}
+	return 0;
+}
+
 /*
  * Initialize the members up to but not including 'bio'. Use after allocating a
  * new bio by bio_alloc_bioset as it does not initialize the bytes outside of
@@ -3157,13 +3206,13 @@ struct bio *btrfs_bio_alloc(unsigned int nr_iovecs)
 	return bio;
 }
 
-struct bio *btrfs_bio_clone(struct bio *bio)
+struct bio *btrfs_bio_clone(struct block_device *bdev, struct bio *bio)
 {
 	struct btrfs_bio *bbio;
 	struct bio *new;
 
 	/* Bio allocation backed by a bioset does not fail */
-	new = bio_alloc_clone(bio->bi_bdev, bio, GFP_NOFS, &btrfs_bioset);
+	new = bio_alloc_clone(bdev, bio, GFP_NOFS, &btrfs_bioset);
 	bbio = btrfs_bio(new);
 	btrfs_bio_init(bbio);
 	bbio->iter = bio->bi_iter;
@@ -3198,7 +3247,7 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, u64 offset, u64 size)
  *                a contiguous page to the previous one
  * @size:	portion of page that we want to write
  * @pg_offset:	starting offset in the page
- * @bio_flags:	flags of the current bio to see if we can merge them
+ * @compress_type:   compression type of the current bio to see if we can merge them
  *
  * Attempt to add a page to bio considering stripe alignment etc.
  *
@@ -3210,7 +3259,7 @@ static int btrfs_bio_add_page(struct btrfs_bio_ctrl *bio_ctrl,
 			      struct page *page,
 			      u64 disk_bytenr, unsigned int size,
 			      unsigned int pg_offset,
-			      unsigned long bio_flags)
+			      enum btrfs_compression_type compress_type)
 {
 	struct bio *bio = bio_ctrl->bio;
 	u32 bio_size = bio->bi_iter.bi_size;
@@ -3222,10 +3271,10 @@ static int btrfs_bio_add_page(struct btrfs_bio_ctrl *bio_ctrl,
 	ASSERT(bio);
 	/* The limit should be calculated when bio_ctrl->bio is allocated */
 	ASSERT(bio_ctrl->len_to_oe_boundary && bio_ctrl->len_to_stripe_boundary);
-	if (bio_ctrl->bio_flags != bio_flags)
+	if (bio_ctrl->compress_type != compress_type)
 		return 0;
 
-	if (bio_ctrl->bio_flags & EXTENT_BIO_COMPRESSED)
+	if (bio_ctrl->compress_type != BTRFS_COMPRESS_NONE)
 		contig = bio->bi_iter.bi_sector == sector;
 	else
 		contig = bio_end_sector(bio) == sector;
@@ -3268,7 +3317,7 @@ static int calc_bio_boundaries(struct btrfs_bio_ctrl *bio_ctrl,
 	 * The split happens for real compressed bio, which happens in
 	 * btrfs_submit_compressed_read/write().
 	 */
-	if (bio_ctrl->bio_flags & EXTENT_BIO_COMPRESSED) {
+	if (bio_ctrl->compress_type != BTRFS_COMPRESS_NONE) {
 		bio_ctrl->len_to_oe_boundary = U32_MAX;
 		bio_ctrl->len_to_stripe_boundary = U32_MAX;
 		return 0;
@@ -3311,7 +3360,7 @@ static int alloc_new_bio(struct btrfs_inode *inode,
 			 unsigned int opf,
 			 bio_end_io_t end_io_func,
 			 u64 disk_bytenr, u32 offset, u64 file_offset,
-			 unsigned long bio_flags)
+			 enum btrfs_compression_type compress_type)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct bio *bio;
@@ -3322,12 +3371,12 @@ static int alloc_new_bio(struct btrfs_inode *inode,
 	 * For compressed page range, its disk_bytenr is always @disk_bytenr
 	 * passed in, no matter if we have added any range into previous bio.
 	 */
-	if (bio_flags & EXTENT_BIO_COMPRESSED)
+	if (compress_type != BTRFS_COMPRESS_NONE)
 		bio->bi_iter.bi_sector = disk_bytenr >> SECTOR_SHIFT;
 	else
 		bio->bi_iter.bi_sector = (disk_bytenr + offset) >> SECTOR_SHIFT;
 	bio_ctrl->bio = bio;
-	bio_ctrl->bio_flags = bio_flags;
+	bio_ctrl->compress_type = compress_type;
 	bio->bi_end_io = end_io_func;
 	bio->bi_private = &inode->io_tree;
 	bio->bi_opf = opf;
@@ -3386,7 +3435,7 @@ error:
  * @end_io_func:     end_io callback for new bio
  * @mirror_num:	     desired mirror to read/write
  * @prev_bio_flags:  flags of previous bio to see if we can merge the current one
- * @bio_flags:	flags of the current bio to see if we can merge them
+ * @compress_type:   compress type for current bio
  */
 static int submit_extent_page(unsigned int opf,
 			      struct writeback_control *wbc,
@@ -3395,7 +3444,7 @@ static int submit_extent_page(unsigned int opf,
 			      size_t size, unsigned long pg_offset,
 			      bio_end_io_t end_io_func,
 			      int mirror_num,
-			      unsigned long bio_flags,
+			      enum btrfs_compression_type compress_type,
 			      bool force_bio_submit)
 {
 	int ret = 0;
@@ -3407,10 +3456,8 @@ static int submit_extent_page(unsigned int opf,
 	ASSERT(pg_offset < PAGE_SIZE && size <= PAGE_SIZE &&
 	       pg_offset + size <= PAGE_SIZE);
 	if (force_bio_submit && bio_ctrl->bio) {
-		ret = submit_one_bio(bio_ctrl->bio, mirror_num, bio_ctrl->bio_flags);
+		submit_one_bio(bio_ctrl->bio, mirror_num, bio_ctrl->compress_type);
 		bio_ctrl->bio = NULL;
-		if (ret < 0)
-			return ret;
 	}
 
 	while (cur < pg_offset + size) {
@@ -3422,7 +3469,7 @@ static int submit_extent_page(unsigned int opf,
 			ret = alloc_new_bio(inode, bio_ctrl, wbc, opf,
 					    end_io_func, disk_bytenr, offset,
 					    page_offset(page) + cur,
-					    bio_flags);
+					    compress_type);
 			if (ret < 0)
 				return ret;
 		}
@@ -3430,14 +3477,14 @@ static int submit_extent_page(unsigned int opf,
 		 * We must go through btrfs_bio_add_page() to ensure each
 		 * page range won't cross various boundaries.
 		 */
-		if (bio_flags & EXTENT_BIO_COMPRESSED)
+		if (compress_type != BTRFS_COMPRESS_NONE)
 			added = btrfs_bio_add_page(bio_ctrl, page, disk_bytenr,
 					size - offset, pg_offset + offset,
-					bio_flags);
+					compress_type);
 		else
 			added = btrfs_bio_add_page(bio_ctrl, page,
 					disk_bytenr + offset, size - offset,
-					pg_offset + offset, bio_flags);
+					pg_offset + offset, compress_type);
 
 		/* Metadata page range should never be split */
 		if (!is_data_inode(&inode->vfs_inode))
@@ -3451,11 +3498,8 @@ static int submit_extent_page(unsigned int opf,
 		if (added < size - offset) {
 			/* The bio should contain some page(s) */
 			ASSERT(bio_ctrl->bio->bi_iter.bi_size);
-			ret = submit_one_bio(bio_ctrl->bio, mirror_num,
-					bio_ctrl->bio_flags);
+			submit_one_bio(bio_ctrl->bio, mirror_num, bio_ctrl->compress_type);
 			bio_ctrl->bio = NULL;
-			if (ret < 0)
-				return ret;
 		}
 		cur += added;
 	}
@@ -3478,7 +3522,7 @@ static int attach_extent_buffer_page(struct extent_buffer *eb,
 	if (page->mapping)
 		lockdep_assert_held(&page->mapping->private_lock);
 
-	if (fs_info->sectorsize == PAGE_SIZE) {
+	if (fs_info->nodesize >= PAGE_SIZE) {
 		if (!PagePrivate(page))
 			attach_page_private(page, eb);
 		else
@@ -3513,7 +3557,7 @@ int set_page_extent_mapped(struct page *page)
 
 	fs_info = btrfs_sb(page->mapping->host->i_sb);
 
-	if (fs_info->sectorsize < PAGE_SIZE)
+	if (btrfs_is_subpage(fs_info, page))
 		return btrfs_attach_subpage(fs_info, page, BTRFS_SUBPAGE_DATA);
 
 	attach_page_private(page, (void *)EXTENT_PAGE_PRIVATE);
@@ -3530,7 +3574,7 @@ void clear_page_extent_mapped(struct page *page)
 		return;
 
 	fs_info = btrfs_sb(page->mapping->host->i_sb);
-	if (fs_info->sectorsize < PAGE_SIZE)
+	if (btrfs_is_subpage(fs_info, page))
 		return btrfs_detach_subpage(fs_info, page);
 
 	detach_page_private(page);
@@ -3569,7 +3613,7 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset,
  * XXX JDM: This needs looking at to ensure proper page locking
  * return 0 on success, otherwise return error
  */
-int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
+static int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
 		      struct btrfs_bio_ctrl *bio_ctrl,
 		      unsigned int read_flags, u64 *prev_em_start)
 {
@@ -3638,16 +3682,13 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
 		BUG_ON(extent_map_end(em) <= cur);
 		BUG_ON(end < cur);
 
-		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
-			this_bio_flag |= EXTENT_BIO_COMPRESSED;
-			extent_set_compress_type(&this_bio_flag,
-						 em->compress_type);
-		}
+		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
+			this_bio_flag = em->compress_type;
 
 		iosize = min(extent_map_end(em) - cur, end - cur + 1);
 		cur_end = min(extent_map_end(em) - 1, end);
 		iosize = ALIGN(iosize, blocksize);
-		if (this_bio_flag & EXTENT_BIO_COMPRESSED)
+		if (this_bio_flag != BTRFS_COMPRESS_NONE)
 			disk_bytenr = em->block_start;
 		else
 			disk_bytenr = em->block_start + extent_offset;
@@ -3743,8 +3784,12 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
 					 this_bio_flag,
 					 force_bio_submit);
 		if (ret) {
-			unlock_extent(tree, cur, cur + iosize - 1);
-			end_page_read(page, false, cur, iosize);
+			/*
+			 * We have to unlock the remaining range, or the page
+			 * will never be unlocked.
+			 */
+			unlock_extent(tree, cur, end);
+			end_page_read(page, false, cur, end + 1 - cur);
 			goto out;
 		}
 		cur = cur + iosize;
@@ -3754,6 +3799,26 @@ out:
 	return ret;
 }
 
+int btrfs_readpage(struct file *file, struct page *page)
+{
+	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
+	u64 start = page_offset(page);
+	u64 end = start + PAGE_SIZE - 1;
+	struct btrfs_bio_ctrl bio_ctrl = { 0 };
+	int ret;
+
+	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
+
+	ret = btrfs_do_readpage(page, NULL, &bio_ctrl, 0, NULL);
+	/*
+	 * If btrfs_do_readpage() failed we will want to submit the assembled
+	 * bio to do the cleanup.
+	 */
+	if (bio_ctrl.bio)
+		submit_one_bio(bio_ctrl.bio, 0, bio_ctrl.compress_type);
+	return ret;
+}
+
 static inline void contiguous_readpages(struct page *pages[], int nr_pages,
 					u64 start, u64 end,
 					struct extent_map **em_cached,
@@ -3772,12 +3837,6 @@ static inline void contiguous_readpages(struct page *pages[], int nr_pages,
 	}
 }
 
-static void update_nr_written(struct writeback_control *wbc,
-			      unsigned long nr_written)
-{
-	wbc->nr_to_write -= nr_written;
-}
-
 /*
  * helper for __extent_writepage, doing all of the delayed allocation setup.
  *
@@ -3877,7 +3936,7 @@ static void find_next_dirty_byte(struct btrfs_fs_info *fs_info,
 	 * For regular sector size == page size case, since one page only
 	 * contains one sector, we return the page offset directly.
 	 */
-	if (fs_info->sectorsize == PAGE_SIZE) {
+	if (!btrfs_is_subpage(fs_info, page)) {
 		*start = page_offset(page);
 		*end = page_offset(page) + PAGE_SIZE;
 		return;
@@ -3920,10 +3979,12 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode,
 	u64 extent_offset;
 	u64 block_start;
 	struct extent_map *em;
+	int saved_ret = 0;
 	int ret = 0;
 	int nr = 0;
 	u32 opf = REQ_OP_WRITE;
 	const unsigned int write_flags = wbc_to_write_flags(wbc);
+	bool has_error = false;
 	bool compressed;
 
 	ret = btrfs_writepage_cow_fixup(page);
@@ -3938,7 +3999,7 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode,
 	 * we don't want to touch the inode after unlocking the page,
 	 * so we update the mapping writeback index now
 	 */
-	update_nr_written(wbc, 1);
+	wbc->nr_to_write--;
 
 	while (cur <= end) {
 		u64 disk_bytenr;
@@ -3973,6 +4034,9 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode,
 		if (IS_ERR(em)) {
 			btrfs_page_set_error(fs_info, page, cur, end - cur + 1);
 			ret = PTR_ERR_OR_ZERO(em);
+			has_error = true;
+			if (!saved_ret)
+				saved_ret = ret;
 			break;
 		}
 
@@ -4036,6 +4100,10 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode,
 					 end_bio_extent_writepage,
 					 0, 0, false);
 		if (ret) {
+			has_error = true;
+			if (!saved_ret)
+				saved_ret = ret;
+
 			btrfs_page_set_error(fs_info, page, cur, iosize);
 			if (PageWriteback(page))
 				btrfs_page_clear_writeback(fs_info, page, cur,
@@ -4049,8 +4117,10 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode,
 	 * If we finish without problem, we should not only clear page dirty,
 	 * but also empty subpage dirty bits
 	 */
-	if (!ret)
+	if (!has_error)
 		btrfs_page_assert_not_dirty(fs_info, page);
+	else
+		ret = saved_ret;
 	*nr_ret = nr;
 	return ret;
 }
@@ -4181,9 +4251,6 @@ void wait_on_extent_buffer_writeback(struct extent_buffer *eb)
 
 static void end_extent_buffer_writeback(struct extent_buffer *eb)
 {
-	if (test_bit(EXTENT_BUFFER_ZONE_FINISH, &eb->bflags))
-		btrfs_zone_finish_endio(eb->fs_info, eb->start, eb->len);
-
 	clear_bit(EXTENT_BUFFER_WRITEBACK, &eb->bflags);
 	smp_mb__after_atomic();
 	wake_up_bit(&eb->bflags, EXTENT_BUFFER_WRITEBACK);
@@ -4203,14 +4270,12 @@ static noinline_for_stack int lock_extent_buffer_for_io(struct extent_buffer *eb
 			  struct extent_page_data *epd)
 {
 	struct btrfs_fs_info *fs_info = eb->fs_info;
-	int i, num_pages, failed_page_nr;
+	int i, num_pages;
 	int flush = 0;
 	int ret = 0;
 
 	if (!btrfs_try_tree_write_lock(eb)) {
-		ret = flush_write_bio(epd);
-		if (ret < 0)
-			return ret;
+		flush_write_bio(epd);
 		flush = 1;
 		btrfs_tree_lock(eb);
 	}
@@ -4220,9 +4285,7 @@ static noinline_for_stack int lock_extent_buffer_for_io(struct extent_buffer *eb
 		if (!epd->sync_io)
 			return 0;
 		if (!flush) {
-			ret = flush_write_bio(epd);
-			if (ret < 0)
-				return ret;
+			flush_write_bio(epd);
 			flush = 1;
 		}
 		while (1) {
@@ -4260,7 +4323,7 @@ static noinline_for_stack int lock_extent_buffer_for_io(struct extent_buffer *eb
 	 * Subpage metadata doesn't use page locking at all, so we can skip
 	 * the page locking.
 	 */
-	if (!ret || fs_info->sectorsize < PAGE_SIZE)
+	if (!ret || fs_info->nodesize < PAGE_SIZE)
 		return ret;
 
 	num_pages = num_extent_pages(eb);
@@ -4269,14 +4332,7 @@ static noinline_for_stack int lock_extent_buffer_for_io(struct extent_buffer *eb
 
 		if (!trylock_page(p)) {
 			if (!flush) {
-				int err;
-
-				err = flush_write_bio(epd);
-				if (err < 0) {
-					ret = err;
-					failed_page_nr = i;
-					goto err_unlock;
-				}
+				flush_write_bio(epd);
 				flush = 1;
 			}
 			lock_page(p);
@@ -4284,25 +4340,6 @@ static noinline_for_stack int lock_extent_buffer_for_io(struct extent_buffer *eb
 	}
 
 	return ret;
-err_unlock:
-	/* Unlock already locked pages */
-	for (i = 0; i < failed_page_nr; i++)
-		unlock_page(eb->pages[i]);
-	/*
-	 * Clear EXTENT_BUFFER_WRITEBACK and wake up anyone waiting on it.
-	 * Also set back EXTENT_BUFFER_DIRTY so future attempts to this eb can
-	 * be made and undo everything done before.
-	 */
-	btrfs_tree_lock(eb);
-	spin_lock(&eb->refs_lock);
-	set_bit(EXTENT_BUFFER_DIRTY, &eb->bflags);
-	end_extent_buffer_writeback(eb);
-	spin_unlock(&eb->refs_lock);
-	percpu_counter_add_batch(&fs_info->dirty_metadata_bytes, eb->len,
-				 fs_info->dirty_metadata_batch);
-	btrfs_clear_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN);
-	btrfs_tree_unlock(eb);
-	return ret;
 }
 
 static void set_btree_ioerr(struct page *page, struct extent_buffer *eb)
@@ -4397,8 +4434,8 @@ static struct extent_buffer *find_extent_buffer_nolock(
 	struct extent_buffer *eb;
 
 	rcu_read_lock();
-	eb = radix_tree_lookup(&fs_info->buffer_radix,
-			       start >> fs_info->sectorsize_bits);
+	eb = xa_load(&fs_info->extent_buffers,
+		     start >> fs_info->sectorsize_bits);
 	if (eb && atomic_inc_not_zero(&eb->refs)) {
 		rcu_read_unlock();
 		return eb;
@@ -4420,7 +4457,7 @@ static void end_bio_subpage_eb_writepage(struct bio *bio)
 	struct bvec_iter_all iter_all;
 
 	fs_info = btrfs_sb(bio_first_page_all(bio)->mapping->host->i_sb);
-	ASSERT(fs_info->sectorsize < PAGE_SIZE);
+	ASSERT(fs_info->nodesize < PAGE_SIZE);
 
 	ASSERT(!bio_flagged(bio, BIO_CLONED));
 	bio_for_each_segment_all(bvec, bio, iter_all) {
@@ -4572,7 +4609,7 @@ static int write_one_subpage_eb(struct extent_buffer *eb,
 	 * dirty anymore, we have submitted a page.  Update nr_written in wbc.
 	 */
 	if (no_dirty_ebs)
-		update_nr_written(wbc, 1);
+		wbc->nr_to_write--;
 	return ret;
 }
 
@@ -4608,7 +4645,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb,
 			break;
 		}
 		disk_bytenr += PAGE_SIZE;
-		update_nr_written(wbc, 1);
+		wbc->nr_to_write--;
 		unlock_page(p);
 	}
 
@@ -4747,7 +4784,7 @@ static int submit_eb_page(struct page *page, struct writeback_control *wbc,
 	if (!PagePrivate(page))
 		return 0;
 
-	if (btrfs_sb(page->mapping->host->i_sb)->sectorsize < PAGE_SIZE)
+	if (btrfs_sb(page->mapping->host->i_sb)->nodesize < PAGE_SIZE)
 		return submit_eb_subpage(page, wbc, epd);
 
 	spin_lock(&mapping->private_lock);
@@ -4803,8 +4840,7 @@ static int submit_eb_page(struct page *page, struct writeback_control *wbc,
 		/*
 		 * Implies write in zoned mode. Mark the last eb in a block group.
 		 */
-		if (cache->seq_zone && eb->start + eb->len == cache->zone_capacity)
-			set_bit(EXTENT_BUFFER_ZONE_FINISH, &eb->bflags);
+		btrfs_schedule_zone_finish_bg(cache, eb);
 		btrfs_put_block_group(cache);
 	}
 	ret = write_one_eb(eb, wbc, epd);
@@ -4923,13 +4959,19 @@ retry:
 	 *   if the fs already has error.
 	 */
 	if (!BTRFS_FS_ERROR(fs_info)) {
-		ret = flush_write_bio(&epd);
+		flush_write_bio(&epd);
 	} else {
 		ret = -EROFS;
 		end_write_bio(&epd, ret);
 	}
 out:
 	btrfs_zoned_meta_io_unlock(fs_info);
+	/*
+	 * We can get ret > 0 from submit_extent_page() indicating how many ebs
+	 * were submitted. Reset it to 0 to avoid false alerts for the caller.
+	 */
+	if (ret > 0)
+		ret = 0;
 	return ret;
 }
 
@@ -5031,8 +5073,7 @@ retry:
 			 * tmpfs file mapping
 			 */
 			if (!trylock_page(page)) {
-				ret = flush_write_bio(epd);
-				BUG_ON(ret < 0);
+				flush_write_bio(epd);
 				lock_page(page);
 			}
 
@@ -5042,10 +5083,8 @@ retry:
 			}
 
 			if (wbc->sync_mode != WB_SYNC_NONE) {
-				if (PageWriteback(page)) {
-					ret = flush_write_bio(epd);
-					BUG_ON(ret < 0);
-				}
+				if (PageWriteback(page))
+					flush_write_bio(epd);
 				wait_on_page_writeback(page);
 			}
 
@@ -5085,9 +5124,8 @@ retry:
 		 * page in our current bio, and thus deadlock, so flush the
 		 * write bio here.
 		 */
-		ret = flush_write_bio(epd);
-		if (!ret)
-			goto retry;
+		flush_write_bio(epd);
+		goto retry;
 	}
 
 	if (wbc->range_cyclic || (wbc->nr_to_write > 0 && range_whole))
@@ -5113,8 +5151,7 @@ int extent_write_full_page(struct page *page, struct writeback_control *wbc)
 		return ret;
 	}
 
-	ret = flush_write_bio(&epd);
-	ASSERT(ret <= 0);
+	flush_write_bio(&epd);
 	return ret;
 }
 
@@ -5176,7 +5213,7 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end)
 	}
 
 	if (!found_error)
-		ret = flush_write_bio(&epd);
+		flush_write_bio(&epd);
 	else
 		end_write_bio(&epd, ret);
 
@@ -5209,7 +5246,7 @@ int extent_writepages(struct address_space *mapping,
 		end_write_bio(&epd, ret);
 		return ret;
 	}
-	ret = flush_write_bio(&epd);
+	flush_write_bio(&epd);
 	return ret;
 }
 
@@ -5232,10 +5269,8 @@ void extent_readahead(struct readahead_control *rac)
 	if (em_cached)
 		free_extent_map(em_cached);
 
-	if (bio_ctrl.bio) {
-		if (submit_one_bio(bio_ctrl.bio, 0, bio_ctrl.bio_flags))
-			return;
-	}
+	if (bio_ctrl.bio)
+		submit_one_bio(bio_ctrl.bio, 0, bio_ctrl.compress_type);
 }
 
 /*
@@ -5804,7 +5839,7 @@ static void detach_extent_buffer_page(struct extent_buffer *eb, struct page *pag
 		return;
 	}
 
-	if (fs_info->sectorsize == PAGE_SIZE) {
+	if (fs_info->nodesize >= PAGE_SIZE) {
 		/*
 		 * We do this since we'll remove the pages after we've
 		 * removed the eb from the radix tree, so we could race
@@ -5911,9 +5946,9 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start,
 struct extent_buffer *btrfs_clone_extent_buffer(const struct extent_buffer *src)
 {
 	int i;
-	struct page *p;
 	struct extent_buffer *new;
 	int num_pages = num_extent_pages(src);
+	int ret;
 
 	new = __alloc_extent_buffer(src->fs_info, src->start, src->len);
 	if (new == NULL)
@@ -5926,22 +5961,23 @@ struct extent_buffer *btrfs_clone_extent_buffer(const struct extent_buffer *src)
 	 */
 	set_bit(EXTENT_BUFFER_UNMAPPED, &new->bflags);
 
+	memset(new->pages, 0, sizeof(*new->pages) * num_pages);
+	ret = btrfs_alloc_page_array(num_pages, new->pages);
+	if (ret) {
+		btrfs_release_extent_buffer(new);
+		return NULL;
+	}
+
 	for (i = 0; i < num_pages; i++) {
 		int ret;
+		struct page *p = new->pages[i];
 
-		p = alloc_page(GFP_NOFS);
-		if (!p) {
-			btrfs_release_extent_buffer(new);
-			return NULL;
-		}
 		ret = attach_extent_buffer_page(new, p, NULL);
 		if (ret < 0) {
-			put_page(p);
 			btrfs_release_extent_buffer(new);
 			return NULL;
 		}
 		WARN_ON(PageDirty(p));
-		new->pages[i] = p;
 		copy_page(page_address(p), page_address(src->pages[i]));
 	}
 	set_extent_buffer_uptodate(new);
@@ -5955,31 +5991,36 @@ struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 	struct extent_buffer *eb;
 	int num_pages;
 	int i;
+	int ret;
 
 	eb = __alloc_extent_buffer(fs_info, start, len);
 	if (!eb)
 		return NULL;
 
 	num_pages = num_extent_pages(eb);
+	ret = btrfs_alloc_page_array(num_pages, eb->pages);
+	if (ret)
+		goto err;
+
 	for (i = 0; i < num_pages; i++) {
-		int ret;
+		struct page *p = eb->pages[i];
 
-		eb->pages[i] = alloc_page(GFP_NOFS);
-		if (!eb->pages[i])
-			goto err;
-		ret = attach_extent_buffer_page(eb, eb->pages[i], NULL);
+		ret = attach_extent_buffer_page(eb, p, NULL);
 		if (ret < 0)
 			goto err;
 	}
+
 	set_extent_buffer_uptodate(eb);
 	btrfs_set_header_nritems(eb, 0);
 	set_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags);
 
 	return eb;
 err:
-	for (; i > 0; i--) {
-		detach_extent_buffer_page(eb, eb->pages[i - 1]);
-		__free_page(eb->pages[i - 1]);
+	for (i = 0; i < num_pages; i++) {
+		if (eb->pages[i]) {
+			detach_extent_buffer_page(eb, eb->pages[i]);
+			__free_page(eb->pages[i]);
+		}
 	}
 	__free_extent_buffer(eb);
 	return NULL;
@@ -6086,24 +6127,22 @@ struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,
 	if (!eb)
 		return ERR_PTR(-ENOMEM);
 	eb->fs_info = fs_info;
-again:
-	ret = radix_tree_preload(GFP_NOFS);
-	if (ret) {
-		exists = ERR_PTR(ret);
-		goto free_eb;
-	}
-	spin_lock(&fs_info->buffer_lock);
-	ret = radix_tree_insert(&fs_info->buffer_radix,
-				start >> fs_info->sectorsize_bits, eb);
-	spin_unlock(&fs_info->buffer_lock);
-	radix_tree_preload_end();
-	if (ret == -EEXIST) {
-		exists = find_extent_buffer(fs_info, start);
-		if (exists)
+
+	do {
+		ret = xa_insert(&fs_info->extent_buffers,
+				start >> fs_info->sectorsize_bits,
+				eb, GFP_NOFS);
+		if (ret == -ENOMEM) {
+			exists = ERR_PTR(ret);
 			goto free_eb;
-		else
-			goto again;
-	}
+		}
+		if (ret == -EBUSY) {
+			exists = find_extent_buffer(fs_info, start);
+			if (exists)
+				goto free_eb;
+		}
+	} while (ret);
+
 	check_buffer_tree_ref(eb);
 	set_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags);
 
@@ -6124,7 +6163,7 @@ static struct extent_buffer *grab_extent_buffer(
 	 * don't try to insert two ebs for the same bytenr.  So here we always
 	 * return NULL and just continue.
 	 */
-	if (fs_info->sectorsize < PAGE_SIZE)
+	if (fs_info->nodesize < PAGE_SIZE)
 		return NULL;
 
 	/* Page not yet attached to an extent buffer */
@@ -6146,6 +6185,30 @@ static struct extent_buffer *grab_extent_buffer(
 	return NULL;
 }
 
+static int check_eb_alignment(struct btrfs_fs_info *fs_info, u64 start)
+{
+	if (!IS_ALIGNED(start, fs_info->sectorsize)) {
+		btrfs_err(fs_info, "bad tree block start %llu", start);
+		return -EINVAL;
+	}
+
+	if (fs_info->nodesize < PAGE_SIZE &&
+	    offset_in_page(start) + fs_info->nodesize > PAGE_SIZE) {
+		btrfs_err(fs_info,
+		"tree block crosses page boundary, start %llu nodesize %u",
+			  start, fs_info->nodesize);
+		return -EINVAL;
+	}
+	if (fs_info->nodesize >= PAGE_SIZE &&
+	    !IS_ALIGNED(start, PAGE_SIZE)) {
+		btrfs_err(fs_info,
+		"tree block is not page aligned, start %llu nodesize %u",
+			  start, fs_info->nodesize);
+		return -EINVAL;
+	}
+	return 0;
+}
+
 struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 					  u64 start, u64 owner_root, int level)
 {
@@ -6160,10 +6223,8 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 	int uptodate = 1;
 	int ret;
 
-	if (!IS_ALIGNED(start, fs_info->sectorsize)) {
-		btrfs_err(fs_info, "bad tree block start %llu", start);
+	if (check_eb_alignment(fs_info, start))
 		return ERR_PTR(-EINVAL);
-	}
 
 #if BITS_PER_LONG == 32
 	if (start >= MAX_LFS_FILESIZE) {
@@ -6176,14 +6237,6 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 		btrfs_warn_32bit_limit(fs_info);
 #endif
 
-	if (fs_info->sectorsize < PAGE_SIZE &&
-	    offset_in_page(start) + len > PAGE_SIZE) {
-		btrfs_err(fs_info,
-		"tree block crosses page boundary, start %llu nodesize %lu",
-			  start, len);
-		return ERR_PTR(-EINVAL);
-	}
-
 	eb = find_extent_buffer(fs_info, start);
 	if (eb)
 		return eb;
@@ -6213,7 +6266,7 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 		 * page, but it may change in the future for 16K page size
 		 * support, so we still preallocate the memory in the loop.
 		 */
-		if (fs_info->sectorsize < PAGE_SIZE) {
+		if (fs_info->nodesize < PAGE_SIZE) {
 			prealloc = btrfs_alloc_subpage(fs_info, BTRFS_SUBPAGE_METADATA);
 			if (IS_ERR(prealloc)) {
 				ret = PTR_ERR(prealloc);
@@ -6264,25 +6317,22 @@ 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);
-	if (ret) {
-		exists = ERR_PTR(ret);
-		goto free_eb;
-	}
-
-	spin_lock(&fs_info->buffer_lock);
-	ret = radix_tree_insert(&fs_info->buffer_radix,
-				start >> fs_info->sectorsize_bits, eb);
-	spin_unlock(&fs_info->buffer_lock);
-	radix_tree_preload_end();
-	if (ret == -EEXIST) {
-		exists = find_extent_buffer(fs_info, start);
-		if (exists)
+
+	do {
+		ret = xa_insert(&fs_info->extent_buffers,
+				start >> fs_info->sectorsize_bits,
+				eb, GFP_NOFS);
+		if (ret == -ENOMEM) {
+			exists = ERR_PTR(ret);
 			goto free_eb;
-		else
-			goto again;
-	}
+		}
+		if (ret == -EBUSY) {
+			exists = find_extent_buffer(fs_info, start);
+			if (exists)
+				goto free_eb;
+		}
+	} while (ret);
+
 	/* add one reference for the tree */
 	check_buffer_tree_ref(eb);
 	set_bit(EXTENT_BUFFER_IN_TREE, &eb->bflags);
@@ -6327,10 +6377,8 @@ static int release_extent_buffer(struct extent_buffer *eb)
 
 			spin_unlock(&eb->refs_lock);
 
-			spin_lock(&fs_info->buffer_lock);
-			radix_tree_delete(&fs_info->buffer_radix,
-					  eb->start >> fs_info->sectorsize_bits);
-			spin_unlock(&fs_info->buffer_lock);
+			xa_erase(&fs_info->extent_buffers,
+				 eb->start >> fs_info->sectorsize_bits);
 		} else {
 			spin_unlock(&eb->refs_lock);
 		}
@@ -6432,7 +6480,7 @@ void clear_extent_buffer_dirty(const struct extent_buffer *eb)
 	int num_pages;
 	struct page *page;
 
-	if (eb->fs_info->sectorsize < PAGE_SIZE)
+	if (eb->fs_info->nodesize < PAGE_SIZE)
 		return clear_subpage_extent_buffer_dirty(eb);
 
 	num_pages = num_extent_pages(eb);
@@ -6464,7 +6512,7 @@ bool set_extent_buffer_dirty(struct extent_buffer *eb)
 	WARN_ON(!test_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags));
 
 	if (!was_dirty) {
-		bool subpage = eb->fs_info->sectorsize < PAGE_SIZE;
+		bool subpage = eb->fs_info->nodesize < PAGE_SIZE;
 
 		/*
 		 * For subpage case, we can have other extent buffers in the
@@ -6504,9 +6552,18 @@ void clear_extent_buffer_uptodate(struct extent_buffer *eb)
 	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
-		if (page)
-			btrfs_page_clear_uptodate(fs_info, page,
-						  eb->start, eb->len);
+		if (!page)
+			continue;
+
+		/*
+		 * This is special handling for metadata subpage, as regular
+		 * btrfs_is_subpage() can not handle cloned/dummy metadata.
+		 */
+		if (fs_info->nodesize >= PAGE_SIZE)
+			ClearPageUptodate(page);
+		else
+			btrfs_subpage_clear_uptodate(fs_info, page, eb->start,
+						     eb->len);
 	}
 }
 
@@ -6521,7 +6578,16 @@ void set_extent_buffer_uptodate(struct extent_buffer *eb)
 	num_pages = num_extent_pages(eb);
 	for (i = 0; i < num_pages; i++) {
 		page = eb->pages[i];
-		btrfs_page_set_uptodate(fs_info, page, eb->start, eb->len);
+
+		/*
+		 * This is special handling for metadata subpage, as regular
+		 * btrfs_is_subpage() can not handle cloned/dummy metadata.
+		 */
+		if (fs_info->nodesize >= PAGE_SIZE)
+			SetPageUptodate(page);
+		else
+			btrfs_subpage_set_uptodate(fs_info, page, eb->start,
+						   eb->len);
 	}
 }
 
@@ -6577,12 +6643,8 @@ static int read_extent_buffer_subpage(struct extent_buffer *eb, int wait,
 		atomic_dec(&eb->io_pages);
 	}
 	if (bio_ctrl.bio) {
-		int tmp;
-
-		tmp = submit_one_bio(bio_ctrl.bio, mirror_num, 0);
+		submit_one_bio(bio_ctrl.bio, mirror_num, 0);
 		bio_ctrl.bio = NULL;
-		if (tmp < 0)
-			return tmp;
 	}
 	if (ret || wait != WAIT_COMPLETE)
 		return ret;
@@ -6616,7 +6678,7 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num)
 	if (unlikely(test_bit(EXTENT_BUFFER_WRITE_ERR, &eb->bflags)))
 		return -EIO;
 
-	if (eb->fs_info->sectorsize < PAGE_SIZE)
+	if (eb->fs_info->nodesize < PAGE_SIZE)
 		return read_extent_buffer_subpage(eb, wait, mirror_num);
 
 	num_pages = num_extent_pages(eb);
@@ -6695,10 +6757,8 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num)
 	}
 
 	if (bio_ctrl.bio) {
-		err = submit_one_bio(bio_ctrl.bio, mirror_num, bio_ctrl.bio_flags);
+		submit_one_bio(bio_ctrl.bio, mirror_num, bio_ctrl.compress_type);
 		bio_ctrl.bio = NULL;
-		if (err)
-			return err;
 	}
 
 	if (ret || wait != WAIT_COMPLETE)
@@ -6871,7 +6931,7 @@ static void assert_eb_page_uptodate(const struct extent_buffer *eb,
 	 * would have !PageUptodate && !PageError, as we clear PageError before
 	 * reading.
 	 */
-	if (fs_info->sectorsize < PAGE_SIZE) {
+	if (fs_info->nodesize < PAGE_SIZE) {
 		bool uptodate, error;
 
 		uptodate = btrfs_subpage_test_uptodate(fs_info, page,
@@ -6973,7 +7033,7 @@ void copy_extent_buffer_full(const struct extent_buffer *dst,
 
 	ASSERT(dst->len == src->len);
 
-	if (dst->fs_info->sectorsize == PAGE_SIZE) {
+	if (dst->fs_info->nodesize >= PAGE_SIZE) {
 		num_pages = num_extent_pages(dst);
 		for (i = 0; i < num_pages; i++)
 			copy_page(page_address(dst->pages[i]),
@@ -6982,7 +7042,7 @@ void copy_extent_buffer_full(const struct extent_buffer *dst,
 		size_t src_offset = get_eb_offset_in_page(src, 0);
 		size_t dst_offset = get_eb_offset_in_page(dst, 0);
 
-		ASSERT(src->fs_info->sectorsize < PAGE_SIZE);
+		ASSERT(src->fs_info->nodesize < PAGE_SIZE);
 		memcpy(page_address(dst->pages[0]) + dst_offset,
 		       page_address(src->pages[0]) + src_offset,
 		       src->len);
@@ -7263,42 +7323,25 @@ void memmove_extent_buffer(const struct extent_buffer *dst,
 	}
 }
 
-#define GANG_LOOKUP_SIZE	16
 static struct extent_buffer *get_next_extent_buffer(
 		struct btrfs_fs_info *fs_info, struct page *page, u64 bytenr)
 {
-	struct extent_buffer *gang[GANG_LOOKUP_SIZE];
-	struct extent_buffer *found = NULL;
+	struct extent_buffer *eb;
+	unsigned long index;
 	u64 page_start = page_offset(page);
-	u64 cur = page_start;
 
 	ASSERT(in_range(bytenr, page_start, PAGE_SIZE));
 	lockdep_assert_held(&fs_info->buffer_lock);
 
-	while (cur < page_start + PAGE_SIZE) {
-		int ret;
-		int i;
-
-		ret = radix_tree_gang_lookup(&fs_info->buffer_radix,
-				(void **)gang, cur >> fs_info->sectorsize_bits,
-				min_t(unsigned int, GANG_LOOKUP_SIZE,
-				      PAGE_SIZE / fs_info->nodesize));
-		if (ret == 0)
-			goto out;
-		for (i = 0; i < ret; i++) {
-			/* Already beyond page end */
-			if (gang[i]->start >= page_start + PAGE_SIZE)
-				goto out;
-			/* Found one */
-			if (gang[i]->start >= bytenr) {
-				found = gang[i];
-				goto out;
-			}
-		}
-		cur = gang[ret - 1]->start + gang[ret - 1]->len;
+	xa_for_each_start(&fs_info->extent_buffers, index, eb,
+			  page_start >> fs_info->sectorsize_bits) {
+		if (in_range(eb->start, page_start, PAGE_SIZE))
+			return eb;
+		else if (eb->start >= page_start + PAGE_SIZE)
+		        /* Already beyond page end */
+			return NULL;
 	}
-out:
-	return found;
+	return NULL;
 }
 
 static int try_release_subpage_extent_buffer(struct page *page)
@@ -7375,7 +7418,7 @@ int try_release_extent_buffer(struct page *page)
 {
 	struct extent_buffer *eb;
 
-	if (btrfs_sb(page->mapping->host->i_sb)->sectorsize < PAGE_SIZE)
+	if (btrfs_sb(page->mapping->host->i_sb)->nodesize < PAGE_SIZE)
 		return try_release_subpage_extent_buffer(page);
 
 	/*
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 151e9da5da2d..956fa434df43 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -7,15 +7,9 @@
 #include <linux/refcount.h>
 #include <linux/fiemap.h>
 #include <linux/btrfs_tree.h>
+#include "compression.h"
 #include "ulist.h"
 
-/*
- * flags for bio submission. The high bits indicate the compression
- * type for this bio
- */
-#define EXTENT_BIO_COMPRESSED 1
-#define EXTENT_BIO_FLAG_SHIFT 16
-
 enum {
 	EXTENT_BUFFER_UPTODATE,
 	EXTENT_BUFFER_DIRTY,
@@ -32,7 +26,6 @@ enum {
 	/* write IO error */
 	EXTENT_BUFFER_WRITE_ERR,
 	EXTENT_BUFFER_NO_CHECK,
-	EXTENT_BUFFER_ZONE_FINISH,
 };
 
 /* these are flags for __process_pages_contig */
@@ -71,9 +64,9 @@ struct btrfs_fs_info;
 struct io_failure_record;
 struct extent_io_tree;
 
-typedef blk_status_t (submit_bio_hook_t)(struct inode *inode, struct bio *bio,
+typedef void (submit_bio_hook_t)(struct inode *inode, struct bio *bio,
 					 int mirror_num,
-					 unsigned long bio_flags);
+					 enum btrfs_compression_type compress_type);
 
 typedef blk_status_t (extent_submit_bio_start_t)(struct inode *inode,
 		struct bio *bio, u64 dio_file_offset);
@@ -103,17 +96,6 @@ struct extent_buffer {
 };
 
 /*
- * Structure to record info about the bio being assembled, and other info like
- * how many bytes are there before stripe/ordered extent boundary.
- */
-struct btrfs_bio_ctrl {
-	struct bio *bio;
-	unsigned long bio_flags;
-	u32 len_to_stripe_boundary;
-	u32 len_to_oe_boundary;
-};
-
-/*
  * Structure to record how many bytes and which ranges are set/cleared
  */
 struct extent_changeset {
@@ -158,17 +140,6 @@ static inline void extent_changeset_free(struct extent_changeset *changeset)
 	kfree(changeset);
 }
 
-static inline void extent_set_compress_type(unsigned long *bio_flags,
-					    int compress_type)
-{
-	*bio_flags |= compress_type << EXTENT_BIO_FLAG_SHIFT;
-}
-
-static inline int extent_compress_type(unsigned long bio_flags)
-{
-	return bio_flags >> EXTENT_BIO_FLAG_SHIFT;
-}
-
 struct extent_map_tree;
 
 typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
@@ -178,11 +149,7 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode,
 int try_release_extent_mapping(struct page *page, gfp_t mask);
 int try_release_extent_buffer(struct page *page);
 
-int __must_check submit_one_bio(struct bio *bio, int mirror_num,
-				unsigned long bio_flags);
-int btrfs_do_readpage(struct page *page, struct extent_map **em_cached,
-		      struct btrfs_bio_ctrl *bio_ctrl,
-		      unsigned int read_flags, u64 *prev_em_start);
+int btrfs_readpage(struct file *file, struct page *page);
 int extent_write_full_page(struct page *page, struct writeback_control *wbc);
 int extent_write_locked_range(struct inode *inode, u64 start, u64 end);
 int extent_writepages(struct address_space *mapping,
@@ -277,8 +244,10 @@ void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end);
 void extent_clear_unlock_delalloc(struct btrfs_inode *inode, u64 start, u64 end,
 				  struct page *locked_page,
 				  u32 bits_to_clear, unsigned long page_ops);
+
+int btrfs_alloc_page_array(unsigned int nr_pages, struct page **page_array);
 struct bio *btrfs_bio_alloc(unsigned int nr_iovecs);
-struct bio *btrfs_bio_clone(struct bio *bio);
+struct bio *btrfs_bio_clone(struct block_device *bdev, struct bio *bio);
 struct bio *btrfs_bio_clone_partial(struct bio *orig, u64 offset, u64 size);
 
 void end_extent_writepage(struct page *page, int err, u64 start, u64 end);
@@ -297,7 +266,7 @@ struct io_failure_record {
 	u64 start;
 	u64 len;
 	u64 logical;
-	unsigned long bio_flags;
+	enum btrfs_compression_type compress_type;
 	int this_mirror;
 	int failed_mirror;
 };
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 380054c94e4b..46c2baa8fdf5 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1460,8 +1460,27 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages,
 	return ret;
 }
 
-static int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
-			   size_t *write_bytes, bool nowait)
+/*
+ * Check if we can do nocow write into the range [@pos, @pos + @write_bytes)
+ *
+ * @pos:         File offset.
+ * @write_bytes: The length to write, will be updated to the nocow writeable
+ *               range.
+ *
+ * This function will flush ordered extents in the range to ensure proper
+ * nocow checks.
+ *
+ * Return:
+ * > 0          If we can nocow, and updates @write_bytes.
+ *  0           If we can't do a nocow write.
+ * -EAGAIN      If we can't do a nocow write because snapshoting of the inode's
+ *              root is in progress.
+ * < 0          If an error happened.
+ *
+ * NOTE: Callers need to call btrfs_check_nocow_unlock() if we return > 0.
+ */
+int btrfs_check_nocow_lock(struct btrfs_inode *inode, loff_t pos,
+			   size_t *write_bytes)
 {
 	struct btrfs_fs_info *fs_info = inode->root->fs_info;
 	struct btrfs_root *root = inode->root;
@@ -1472,7 +1491,7 @@ static int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
 	if (!(inode->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)))
 		return 0;
 
-	if (!nowait && !btrfs_drew_try_write_lock(&root->snapshot_lock))
+	if (!btrfs_drew_try_write_lock(&root->snapshot_lock))
 		return -EAGAIN;
 
 	lockstart = round_down(pos, fs_info->sectorsize);
@@ -1480,71 +1499,21 @@ static int check_can_nocow(struct btrfs_inode *inode, loff_t pos,
 			   fs_info->sectorsize) - 1;
 	num_bytes = lockend - lockstart + 1;
 
-	if (nowait) {
-		struct btrfs_ordered_extent *ordered;
-
-		if (!try_lock_extent(&inode->io_tree, lockstart, lockend))
-			return -EAGAIN;
-
-		ordered = btrfs_lookup_ordered_range(inode, lockstart,
-						     num_bytes);
-		if (ordered) {
-			btrfs_put_ordered_extent(ordered);
-			ret = -EAGAIN;
-			goto out_unlock;
-		}
-	} else {
-		btrfs_lock_and_flush_ordered_range(inode, lockstart,
-						   lockend, NULL);
-	}
-
+	btrfs_lock_and_flush_ordered_range(inode, lockstart, lockend, NULL);
 	ret = can_nocow_extent(&inode->vfs_inode, lockstart, &num_bytes,
 			NULL, NULL, NULL, false);
 	if (ret <= 0) {
 		ret = 0;
-		if (!nowait)
-			btrfs_drew_write_unlock(&root->snapshot_lock);
+		btrfs_drew_write_unlock(&root->snapshot_lock);
 	} else {
 		*write_bytes = min_t(size_t, *write_bytes ,
 				     num_bytes - pos + lockstart);
 	}
-out_unlock:
 	unlock_extent(&inode->io_tree, lockstart, lockend);
 
 	return ret;
 }
 
-static int check_nocow_nolock(struct btrfs_inode *inode, loff_t pos,
-			      size_t *write_bytes)
-{
-	return check_can_nocow(inode, pos, write_bytes, true);
-}
-
-/*
- * Check if we can do nocow write into the range [@pos, @pos + @write_bytes)
- *
- * @pos:	 File offset
- * @write_bytes: The length to write, will be updated to the nocow writeable
- *		 range
- *
- * This function will flush ordered extents in the range to ensure proper
- * nocow checks.
- *
- * Return:
- * >0		and update @write_bytes if we can do nocow write
- *  0		if we can't do nocow write
- * -EAGAIN	if we can't get the needed lock or there are ordered extents
- * 		for * (nowait == true) case
- * <0		if other error happened
- *
- * NOTE: Callers need to release the lock by btrfs_check_nocow_unlock().
- */
-int btrfs_check_nocow_lock(struct btrfs_inode *inode, loff_t pos,
-			   size_t *write_bytes)
-{
-	return check_can_nocow(inode, pos, write_bytes, false);
-}
-
 void btrfs_check_nocow_unlock(struct btrfs_inode *inode)
 {
 	btrfs_drew_write_unlock(&inode->root->snapshot_lock);
@@ -1579,20 +1548,15 @@ static int btrfs_write_check(struct kiocb *iocb, struct iov_iter *from,
 	loff_t oldsize;
 	loff_t start_pos;
 
-	if (iocb->ki_flags & IOCB_NOWAIT) {
-		size_t nocow_bytes = count;
-
-		/* We will allocate space in case nodatacow is not set, so bail */
-		if (check_nocow_nolock(BTRFS_I(inode), pos, &nocow_bytes) <= 0)
-			return -EAGAIN;
-		/*
-		 * There are holes in the range or parts of the range that must
-		 * be COWed (shared extents, RO block groups, etc), so just bail
-		 * out.
-		 */
-		if (nocow_bytes < count)
-			return -EAGAIN;
-	}
+	/*
+	 * Quickly bail out on NOWAIT writes if we don't have the nodatacow or
+	 * prealloc flags, as without those flags we always have to COW. We will
+	 * later check if we can really COW into the target range (using
+	 * can_nocow_extent() at btrfs_get_blocks_direct_write()).
+	 */
+	if ((iocb->ki_flags & IOCB_NOWAIT) &&
+	    !(BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)))
+		return -EAGAIN;
 
 	current->backing_dev_info = inode_to_bdi(inode);
 	ret = file_remove_privs(file);
@@ -1720,7 +1684,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb,
 		WARN_ON(reserve_bytes == 0);
 		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode),
 						      reserve_bytes,
-						      reserve_bytes);
+						      reserve_bytes, false);
 		if (ret) {
 			if (!only_release_metadata)
 				btrfs_free_reserved_data_space(BTRFS_I(inode),
@@ -1965,8 +1929,7 @@ relock:
 	 */
 again:
 	from->nofault = true;
-	err = iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
-			   IOMAP_DIO_PARTIAL, written);
+	err = btrfs_dio_rw(iocb, from, written);
 	from->nofault = false;
 
 	/* No increment (+=) because iomap returns a cumulative value. */
@@ -2570,10 +2533,10 @@ static int find_first_non_hole(struct btrfs_inode *inode, u64 *start, u64 *len)
 	return ret;
 }
 
-static int btrfs_punch_hole_lock_range(struct inode *inode,
-				       const u64 lockstart,
-				       const u64 lockend,
-				       struct extent_state **cached_state)
+static void btrfs_punch_hole_lock_range(struct inode *inode,
+					const u64 lockstart,
+					const u64 lockend,
+					struct extent_state **cached_state)
 {
 	/*
 	 * For subpage case, if the range is not at page boundary, we could
@@ -2587,40 +2550,29 @@ static int btrfs_punch_hole_lock_range(struct inode *inode,
 	const u64 page_lockend = round_down(lockend + 1, PAGE_SIZE) - 1;
 
 	while (1) {
-		struct btrfs_ordered_extent *ordered;
-		int ret;
-
 		truncate_pagecache_range(inode, lockstart, lockend);
 
 		lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 				 cached_state);
-		ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode),
-							    lockend);
-
 		/*
-		 * We need to make sure we have no ordered extents in this range
-		 * and nobody raced in and read a page in this range, if we did
-		 * we need to try again.
+		 * We can't have ordered extents in the range, nor dirty/writeback
+		 * pages, because we have locked the inode's VFS lock in exclusive
+		 * mode, we have locked the inode's i_mmap_lock in exclusive mode,
+		 * we have flushed all delalloc in the range and we have waited
+		 * for any ordered extents in the range to complete.
+		 * We can race with anyone reading pages from this range, so after
+		 * locking the range check if we have pages in the range, and if
+		 * we do, unlock the range and retry.
 		 */
-		if ((!ordered ||
-		    (ordered->file_offset + ordered->num_bytes <= lockstart ||
-		     ordered->file_offset > lockend)) &&
-		     !filemap_range_has_page(inode->i_mapping,
-					     page_lockstart, page_lockend)) {
-			if (ordered)
-				btrfs_put_ordered_extent(ordered);
+		if (!filemap_range_has_page(inode->i_mapping, page_lockstart,
+					    page_lockend))
 			break;
-		}
-		if (ordered)
-			btrfs_put_ordered_extent(ordered);
+
 		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart,
 				     lockend, cached_state);
-		ret = btrfs_wait_ordered_range(inode, lockstart,
-					       lockend - lockstart + 1);
-		if (ret)
-			return ret;
 	}
-	return 0;
+
+	btrfs_assert_inode_range_clean(BTRFS_I(inode), lockstart, lockend);
 }
 
 static int btrfs_insert_replace_extent(struct btrfs_trans_handle *trans,
@@ -2976,11 +2928,12 @@ static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len)
 	bool truncated_block = false;
 	bool updated_inode = false;
 
+	btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP);
+
 	ret = btrfs_wait_ordered_range(inode, offset, len);
 	if (ret)
-		return ret;
+		goto out_only_mutex;
 
-	btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP);
 	ino_size = round_up(inode->i_size, fs_info->sectorsize);
 	ret = find_first_non_hole(BTRFS_I(inode), &offset, &len);
 	if (ret < 0)
@@ -3072,10 +3025,7 @@ static int btrfs_punch_hole(struct file *file, loff_t offset, loff_t len)
 		goto out_only_mutex;
 	}
 
-	ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend,
-					  &cached_state);
-	if (ret)
-		goto out_only_mutex;
+	btrfs_punch_hole_lock_range(inode, lockstart, lockend, &cached_state);
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -3237,8 +3187,6 @@ static int btrfs_zero_range(struct inode *inode,
 	u64 bytes_to_reserve = 0;
 	bool space_reserved = false;
 
-	inode_dio_wait(inode);
-
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, alloc_start,
 			      alloc_end - alloc_start);
 	if (IS_ERR(em)) {
@@ -3368,10 +3316,8 @@ reserve_space:
 		if (ret < 0)
 			goto out;
 		space_reserved = true;
-		ret = btrfs_punch_hole_lock_range(inode, lockstart, lockend,
-						  &cached_state);
-		if (ret)
-			goto out;
+		btrfs_punch_hole_lock_range(inode, lockstart, lockend,
+					    &cached_state);
 		ret = btrfs_qgroup_reserve_data(BTRFS_I(inode), &data_reserved,
 						alloc_start, bytes_to_reserve);
 		if (ret) {
@@ -3417,6 +3363,9 @@ static long btrfs_fallocate(struct file *file, int mode,
 	u64 alloc_hint = 0;
 	u64 locked_end;
 	u64 actual_end = 0;
+	u64 data_space_needed = 0;
+	u64 data_space_reserved = 0;
+	u64 qgroup_reserved = 0;
 	struct extent_map *em;
 	int blocksize = btrfs_inode_sectorsize(BTRFS_I(inode));
 	int ret;
@@ -3437,18 +3386,6 @@ static long btrfs_fallocate(struct file *file, int mode,
 	if (mode & FALLOC_FL_PUNCH_HOLE)
 		return btrfs_punch_hole(file, offset, len);
 
-	/*
-	 * Only trigger disk allocation, don't trigger qgroup reserve
-	 *
-	 * For qgroup space, it will be checked later.
-	 */
-	if (!(mode & FALLOC_FL_ZERO_RANGE)) {
-		ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode),
-						      alloc_end - alloc_start);
-		if (ret < 0)
-			return ret;
-	}
-
 	btrfs_inode_lock(inode, BTRFS_ILOCK_MMAP);
 
 	if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size) {
@@ -3485,8 +3422,12 @@ static long btrfs_fallocate(struct file *file, int mode,
 	}
 
 	/*
-	 * wait for ordered IO before we have any locks.  We'll loop again
-	 * below with the locks held.
+	 * We have locked the inode at the VFS level (in exclusive mode) and we
+	 * have locked the i_mmap_lock lock (in exclusive mode). Now before
+	 * locking the file range, flush all dealloc in the range and wait for
+	 * all ordered extents in the range to complete. After this we can lock
+	 * the file range and, due to the previous locking we did, we know there
+	 * can't be more delalloc or ordered extents in the range.
 	 */
 	ret = btrfs_wait_ordered_range(inode, alloc_start,
 				       alloc_end - alloc_start);
@@ -3500,38 +3441,10 @@ static long btrfs_fallocate(struct file *file, int mode,
 	}
 
 	locked_end = alloc_end - 1;
-	while (1) {
-		struct btrfs_ordered_extent *ordered;
+	lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
+			 &cached_state);
 
-		/* the extent lock is ordered inside the running
-		 * transaction
-		 */
-		lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start,
-				 locked_end, &cached_state);
-		ordered = btrfs_lookup_first_ordered_extent(BTRFS_I(inode),
-							    locked_end);
-
-		if (ordered &&
-		    ordered->file_offset + ordered->num_bytes > alloc_start &&
-		    ordered->file_offset < alloc_end) {
-			btrfs_put_ordered_extent(ordered);
-			unlock_extent_cached(&BTRFS_I(inode)->io_tree,
-					     alloc_start, locked_end,
-					     &cached_state);
-			/*
-			 * we can't wait on the range with the transaction
-			 * running or with the extent lock held
-			 */
-			ret = btrfs_wait_ordered_range(inode, alloc_start,
-						       alloc_end - alloc_start);
-			if (ret)
-				goto out;
-		} else {
-			if (ordered)
-				btrfs_put_ordered_extent(ordered);
-			break;
-		}
-	}
+	btrfs_assert_inode_range_clean(BTRFS_I(inode), alloc_start, locked_end);
 
 	/* First, check if we exceed the qgroup limit */
 	INIT_LIST_HEAD(&reserve_list);
@@ -3548,48 +3461,64 @@ static long btrfs_fallocate(struct file *file, int mode,
 		if (em->block_start == EXTENT_MAP_HOLE ||
 		    (cur_offset >= inode->i_size &&
 		     !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
-			ret = add_falloc_range(&reserve_list, cur_offset,
-					       last_byte - cur_offset);
+			const u64 range_len = last_byte - cur_offset;
+
+			ret = add_falloc_range(&reserve_list, cur_offset, range_len);
 			if (ret < 0) {
 				free_extent_map(em);
 				break;
 			}
 			ret = btrfs_qgroup_reserve_data(BTRFS_I(inode),
-					&data_reserved, cur_offset,
-					last_byte - cur_offset);
+					&data_reserved, cur_offset, range_len);
 			if (ret < 0) {
-				cur_offset = last_byte;
 				free_extent_map(em);
 				break;
 			}
-		} else {
-			/*
-			 * Do not need to reserve unwritten extent for this
-			 * range, free reserved data space first, otherwise
-			 * it'll result in false ENOSPC error.
-			 */
-			btrfs_free_reserved_data_space(BTRFS_I(inode),
-				data_reserved, cur_offset,
-				last_byte - cur_offset);
+			qgroup_reserved += range_len;
+			data_space_needed += range_len;
 		}
 		free_extent_map(em);
 		cur_offset = last_byte;
 	}
 
+	if (!ret && data_space_needed > 0) {
+		/*
+		 * We are safe to reserve space here as we can't have delalloc
+		 * in the range, see above.
+		 */
+		ret = btrfs_alloc_data_chunk_ondemand(BTRFS_I(inode),
+						      data_space_needed);
+		if (!ret)
+			data_space_reserved = data_space_needed;
+	}
+
 	/*
 	 * If ret is still 0, means we're OK to fallocate.
 	 * Or just cleanup the list and exit.
 	 */
 	list_for_each_entry_safe(range, tmp, &reserve_list, list) {
-		if (!ret)
+		if (!ret) {
 			ret = btrfs_prealloc_file_range(inode, mode,
 					range->start,
 					range->len, i_blocksize(inode),
 					offset + len, &alloc_hint);
-		else
+			/*
+			 * btrfs_prealloc_file_range() releases space even
+			 * if it returns an error.
+			 */
+			data_space_reserved -= range->len;
+			qgroup_reserved -= range->len;
+		} else if (data_space_reserved > 0) {
 			btrfs_free_reserved_data_space(BTRFS_I(inode),
-					data_reserved, range->start,
-					range->len);
+					       data_reserved, range->start,
+					       range->len);
+			data_space_reserved -= range->len;
+			qgroup_reserved -= range->len;
+		} else if (qgroup_reserved > 0) {
+			btrfs_qgroup_free_data(BTRFS_I(inode), data_reserved,
+					       range->start, range->len);
+			qgroup_reserved -= range->len;
+		}
 		list_del(&range->list);
 		kfree(range);
 	}
@@ -3606,10 +3535,6 @@ out_unlock:
 			     &cached_state);
 out:
 	btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP);
-	/* Let go of our reservation. */
-	if (ret != 0 && !(mode & FALLOC_FL_ZERO_RANGE))
-		btrfs_free_reserved_data_space(BTRFS_I(inode), data_reserved,
-				cur_offset, alloc_end - cur_offset);
 	extent_changeset_free(data_reserved);
 	return ret;
 }
@@ -3767,8 +3692,7 @@ again:
 	 */
 	pagefault_disable();
 	to->nofault = true;
-	ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
-			   IOMAP_DIO_PARTIAL, read);
+	ret = btrfs_dio_rw(iocb, to, read);
 	to->nofault = false;
 	pagefault_enable();
 
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 01a408db5683..f7adee6fa05e 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -2630,16 +2630,19 @@ out:
 static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group,
 					u64 bytenr, u64 size, bool used)
 {
-	struct btrfs_fs_info *fs_info = block_group->fs_info;
+	struct btrfs_space_info *sinfo = block_group->space_info;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	u64 offset = bytenr - block_group->start;
 	u64 to_free, to_unusable;
-	const int bg_reclaim_threshold = READ_ONCE(fs_info->bg_reclaim_threshold);
+	int bg_reclaim_threshold = 0;
 	bool initial = (size == block_group->length);
 	u64 reclaimable_unusable;
 
 	WARN_ON(!initial && offset + size > block_group->zone_capacity);
 
+	if (!initial)
+		bg_reclaim_threshold = READ_ONCE(sinfo->bg_reclaim_threshold);
+
 	spin_lock(&ctl->tree_lock);
 	if (!used)
 		to_free = size;
@@ -4069,7 +4072,7 @@ static int cleanup_free_space_cache_v1(struct btrfs_fs_info *fs_info,
 
 	btrfs_info(fs_info, "cleaning free space cache v1");
 
-	node = rb_first(&fs_info->block_group_cache_tree);
+	node = rb_first_cached(&fs_info->block_group_cache_tree);
 	while (node) {
 		block_group = rb_entry(node, struct btrfs_block_group, cache_node);
 		ret = btrfs_remove_free_space_inode(trans, NULL, block_group);
diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c
index 0ae54d8c10d6..1bf89aa67216 100644
--- a/fs/btrfs/free-space-tree.c
+++ b/fs/btrfs/free-space-tree.c
@@ -1178,7 +1178,7 @@ int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
 		goto abort;
 	}
 
-	node = rb_first(&fs_info->block_group_cache_tree);
+	node = rb_first_cached(&fs_info->block_group_cache_tree);
 	while (node) {
 		block_group = rb_entry(node, struct btrfs_block_group,
 				       cache_node);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 95c499b8424e..da13bd0d10f1 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -64,8 +64,36 @@ struct btrfs_iget_args {
 struct btrfs_dio_data {
 	ssize_t submitted;
 	struct extent_changeset *data_reserved;
+	bool data_space_reserved;
+	bool nocow_done;
 };
 
+struct btrfs_dio_private {
+	struct inode *inode;
+
+	/*
+	 * Since DIO can use anonymous page, we cannot use page_offset() to
+	 * grab the file offset, thus need a dedicated member for file offset.
+	 */
+	u64 file_offset;
+	/* Used for bio::bi_size */
+	u32 bytes;
+
+	/*
+	 * References to this structure. There is one reference per in-flight
+	 * bio plus one while we're still setting up.
+	 */
+	refcount_t refs;
+
+	/* Array of checksums */
+	u8 *csums;
+
+	/* This must be last */
+	struct bio bio;
+};
+
+static struct bio_set btrfs_dio_bioset;
+
 struct btrfs_rename_ctx {
 	/* Output field. Stores the index number of the old directory entry. */
 	u64 index;
@@ -222,15 +250,25 @@ static inline void btrfs_cleanup_ordered_extents(struct btrfs_inode *inode,
 static int btrfs_dirty_inode(struct inode *inode);
 
 static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
-				     struct inode *inode,  struct inode *dir,
-				     const struct qstr *qstr)
+				     struct btrfs_new_inode_args *args)
 {
 	int err;
 
-	err = btrfs_init_acl(trans, inode, dir);
-	if (!err)
-		err = btrfs_xattr_security_init(trans, inode, dir, qstr);
-	return err;
+	if (args->default_acl) {
+		err = __btrfs_set_acl(trans, args->inode, args->default_acl,
+				      ACL_TYPE_DEFAULT);
+		if (err)
+			return err;
+	}
+	if (args->acl) {
+		err = __btrfs_set_acl(trans, args->inode, args->acl, ACL_TYPE_ACCESS);
+		if (err)
+			return err;
+	}
+	if (!args->default_acl && !args->acl)
+		cache_no_acl(args->inode);
+	return btrfs_xattr_security_init(trans, args->inode, args->dir,
+					 &args->dentry->d_name);
 }
 
 /*
@@ -1607,6 +1645,141 @@ static int fallback_to_cow(struct btrfs_inode *inode, struct page *locked_page,
 			      nr_written, 1);
 }
 
+struct can_nocow_file_extent_args {
+	/* Input fields. */
+
+	/* Start file offset of the range we want to NOCOW. */
+	u64 start;
+	/* End file offset (inclusive) of the range we want to NOCOW. */
+	u64 end;
+	bool writeback_path;
+	bool strict;
+	/*
+	 * Free the path passed to can_nocow_file_extent() once it's not needed
+	 * anymore.
+	 */
+	bool free_path;
+
+	/* Output fields. Only set when can_nocow_file_extent() returns 1. */
+
+	u64 disk_bytenr;
+	u64 disk_num_bytes;
+	u64 extent_offset;
+	/* Number of bytes that can be written to in NOCOW mode. */
+	u64 num_bytes;
+};
+
+/*
+ * Check if we can NOCOW the file extent that the path points to.
+ * This function may return with the path released, so the caller should check
+ * if path->nodes[0] is NULL or not if it needs to use the path afterwards.
+ *
+ * Returns: < 0 on error
+ *            0 if we can not NOCOW
+ *            1 if we can NOCOW
+ */
+static int can_nocow_file_extent(struct btrfs_path *path,
+				 struct btrfs_key *key,
+				 struct btrfs_inode *inode,
+				 struct can_nocow_file_extent_args *args)
+{
+	const bool is_freespace_inode = btrfs_is_free_space_inode(inode);
+	struct extent_buffer *leaf = path->nodes[0];
+	struct btrfs_root *root = inode->root;
+	struct btrfs_file_extent_item *fi;
+	u64 extent_end;
+	u8 extent_type;
+	int can_nocow = 0;
+	int ret = 0;
+
+	fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item);
+	extent_type = btrfs_file_extent_type(leaf, fi);
+
+	if (extent_type == BTRFS_FILE_EXTENT_INLINE)
+		goto out;
+
+	/* Can't access these fields unless we know it's not an inline extent. */
+	args->disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
+	args->disk_num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
+	args->extent_offset = btrfs_file_extent_offset(leaf, fi);
+
+	if (!(inode->flags & BTRFS_INODE_NODATACOW) &&
+	    extent_type == BTRFS_FILE_EXTENT_REG)
+		goto out;
+
+	/*
+	 * If the extent was created before the generation where the last snapshot
+	 * for its subvolume was created, then this implies the extent is shared,
+	 * hence we must COW.
+	 */
+	if (!args->strict &&
+	    btrfs_file_extent_generation(leaf, fi) <=
+	    btrfs_root_last_snapshot(&root->root_item))
+		goto out;
+
+	/* An explicit hole, must COW. */
+	if (args->disk_bytenr == 0)
+		goto out;
+
+	/* Compressed/encrypted/encoded extents must be COWed. */
+	if (btrfs_file_extent_compression(leaf, fi) ||
+	    btrfs_file_extent_encryption(leaf, fi) ||
+	    btrfs_file_extent_other_encoding(leaf, fi))
+		goto out;
+
+	extent_end = btrfs_file_extent_end(path);
+
+	/*
+	 * The following checks can be expensive, as they need to take other
+	 * locks and do btree or rbtree searches, so release the path to avoid
+	 * blocking other tasks for too long.
+	 */
+	btrfs_release_path(path);
+
+	ret = btrfs_cross_ref_exist(root, btrfs_ino(inode),
+				    key->offset - args->extent_offset,
+				    args->disk_bytenr, false, path);
+	WARN_ON_ONCE(ret > 0 && is_freespace_inode);
+	if (ret != 0)
+		goto out;
+
+	if (args->free_path) {
+		/*
+		 * We don't need the path anymore, plus through the
+		 * csum_exist_in_range() call below we will end up allocating
+		 * another path. So free the path to avoid unnecessary extra
+		 * memory usage.
+		 */
+		btrfs_free_path(path);
+		path = NULL;
+	}
+
+	/* If there are pending snapshots for this root, we must COW. */
+	if (args->writeback_path && !is_freespace_inode &&
+	    atomic_read(&root->snapshot_force_cow))
+		goto out;
+
+	args->disk_bytenr += args->extent_offset;
+	args->disk_bytenr += args->start - key->offset;
+	args->num_bytes = min(args->end + 1, extent_end) - args->start;
+
+	/*
+	 * Force COW if csums exist in the range. This ensures that csums for a
+	 * given extent are either valid or do not exist.
+	 */
+	ret = csum_exist_in_range(root->fs_info, args->disk_bytenr, args->num_bytes);
+	WARN_ON_ONCE(ret > 0 && is_freespace_inode);
+	if (ret != 0)
+		goto out;
+
+	can_nocow = 1;
+ out:
+	if (args->free_path && path)
+		btrfs_free_path(path);
+
+	return ret < 0 ? ret : can_nocow;
+}
+
 /*
  * when nowcow writeback call back.  This checks for snapshots or COW copies
  * of the extents that exist in the file, and COWs the file as required.
@@ -1627,11 +1800,10 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 	u64 cur_offset = start;
 	int ret;
 	bool check_prev = true;
-	const bool freespace_inode = btrfs_is_free_space_inode(inode);
 	u64 ino = btrfs_ino(inode);
+	struct btrfs_block_group *bg;
 	bool nocow = false;
-	u64 disk_bytenr = 0;
-	const bool force = inode->flags & BTRFS_INODE_NODATACOW;
+	struct can_nocow_file_extent_args nocow_args = { 0 };
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -1644,15 +1816,16 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 		return -ENOMEM;
 	}
 
+	nocow_args.end = end;
+	nocow_args.writeback_path = true;
+
 	while (1) {
 		struct btrfs_key found_key;
 		struct btrfs_file_extent_item *fi;
 		struct extent_buffer *leaf;
 		u64 extent_end;
-		u64 extent_offset;
-		u64 num_bytes = 0;
-		u64 disk_num_bytes;
 		u64 ram_bytes;
+		u64 nocow_end;
 		int extent_type;
 
 		nocow = false;
@@ -1728,116 +1901,38 @@ next_slot:
 		fi = btrfs_item_ptr(leaf, path->slots[0],
 				    struct btrfs_file_extent_item);
 		extent_type = btrfs_file_extent_type(leaf, fi);
-
+		/* If this is triggered then we have a memory corruption. */
+		ASSERT(extent_type < BTRFS_NR_FILE_EXTENT_TYPES);
+		if (WARN_ON(extent_type >= BTRFS_NR_FILE_EXTENT_TYPES)) {
+			ret = -EUCLEAN;
+			goto error;
+		}
 		ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
-		if (extent_type == BTRFS_FILE_EXTENT_REG ||
-		    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
-			disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
-			extent_offset = btrfs_file_extent_offset(leaf, fi);
-			extent_end = found_key.offset +
-				btrfs_file_extent_num_bytes(leaf, fi);
-			disk_num_bytes =
-				btrfs_file_extent_disk_num_bytes(leaf, fi);
-			/*
-			 * If the extent we got ends before our current offset,
-			 * skip to the next extent.
-			 */
-			if (extent_end <= cur_offset) {
-				path->slots[0]++;
-				goto next_slot;
-			}
-			/* Skip holes */
-			if (disk_bytenr == 0)
-				goto out_check;
-			/* Skip compressed/encrypted/encoded extents */
-			if (btrfs_file_extent_compression(leaf, fi) ||
-			    btrfs_file_extent_encryption(leaf, fi) ||
-			    btrfs_file_extent_other_encoding(leaf, fi))
-				goto out_check;
-			/*
-			 * If extent is created before the last volume's snapshot
-			 * this implies the extent is shared, hence we can't do
-			 * nocow. This is the same check as in
-			 * btrfs_cross_ref_exist but without calling
-			 * btrfs_search_slot.
-			 */
-			if (!freespace_inode &&
-			    btrfs_file_extent_generation(leaf, fi) <=
-			    btrfs_root_last_snapshot(&root->root_item))
-				goto out_check;
-			if (extent_type == BTRFS_FILE_EXTENT_REG && !force)
-				goto out_check;
+		extent_end = btrfs_file_extent_end(path);
 
-			/*
-			 * The following checks can be expensive, as they need to
-			 * take other locks and do btree or rbtree searches, so
-			 * release the path to avoid blocking other tasks for too
-			 * long.
-			 */
-			btrfs_release_path(path);
+		/*
+		 * If the extent we got ends before our current offset, skip to
+		 * the next extent.
+		 */
+		if (extent_end <= cur_offset) {
+			path->slots[0]++;
+			goto next_slot;
+		}
 
-			ret = btrfs_cross_ref_exist(root, ino,
-						    found_key.offset -
-						    extent_offset, disk_bytenr, false);
-			if (ret) {
-				/*
-				 * ret could be -EIO if the above fails to read
-				 * metadata.
-				 */
-				if (ret < 0) {
-					if (cow_start != (u64)-1)
-						cur_offset = cow_start;
-					goto error;
-				}
+		nocow_args.start = cur_offset;
+		ret = can_nocow_file_extent(path, &found_key, inode, &nocow_args);
+		if (ret < 0) {
+			if (cow_start != (u64)-1)
+				cur_offset = cow_start;
+			goto error;
+		} else if (ret == 0) {
+			goto out_check;
+		}
 
-				WARN_ON_ONCE(freespace_inode);
-				goto out_check;
-			}
-			disk_bytenr += extent_offset;
-			disk_bytenr += cur_offset - found_key.offset;
-			num_bytes = min(end + 1, extent_end) - cur_offset;
-			/*
-			 * If there are pending snapshots for this root, we
-			 * fall into common COW way
-			 */
-			if (!freespace_inode && atomic_read(&root->snapshot_force_cow))
-				goto out_check;
-			/*
-			 * force cow if csum exists in the range.
-			 * this ensure that csum for a given extent are
-			 * either valid or do not exist.
-			 */
-			ret = csum_exist_in_range(fs_info, disk_bytenr,
-						  num_bytes);
-			if (ret) {
-				/*
-				 * ret could be -EIO if the above fails to read
-				 * metadata.
-				 */
-				if (ret < 0) {
-					if (cow_start != (u64)-1)
-						cur_offset = cow_start;
-					goto error;
-				}
-				WARN_ON_ONCE(freespace_inode);
-				goto out_check;
-			}
-			/* If the extent's block group is RO, we must COW */
-			if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr))
-				goto out_check;
+		ret = 0;
+		bg = btrfs_inc_nocow_writers(fs_info, nocow_args.disk_bytenr);
+		if (bg)
 			nocow = true;
-		} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
-			extent_end = found_key.offset + ram_bytes;
-			extent_end = ALIGN(extent_end, fs_info->sectorsize);
-			/* Skip extents outside of our requested range */
-			if (extent_end <= start) {
-				path->slots[0]++;
-				goto next_slot;
-			}
-		} else {
-			/* If this triggers then we have a memory corruption */
-			BUG();
-		}
 out_check:
 		/*
 		 * If nocow is false then record the beginning of the range
@@ -1869,15 +1964,17 @@ out_check:
 			cow_start = (u64)-1;
 		}
 
+		nocow_end = cur_offset + nocow_args.num_bytes - 1;
+
 		if (extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
-			u64 orig_start = found_key.offset - extent_offset;
+			u64 orig_start = found_key.offset - nocow_args.extent_offset;
 			struct extent_map *em;
 
-			em = create_io_em(inode, cur_offset, num_bytes,
+			em = create_io_em(inode, cur_offset, nocow_args.num_bytes,
 					  orig_start,
-					  disk_bytenr, /* block_start */
-					  num_bytes, /* block_len */
-					  disk_num_bytes, /* orig_block_len */
+					  nocow_args.disk_bytenr, /* block_start */
+					  nocow_args.num_bytes, /* block_len */
+					  nocow_args.disk_num_bytes, /* orig_block_len */
 					  ram_bytes, BTRFS_COMPRESS_NONE,
 					  BTRFS_ORDERED_PREALLOC);
 			if (IS_ERR(em)) {
@@ -1886,20 +1983,23 @@ out_check:
 			}
 			free_extent_map(em);
 			ret = btrfs_add_ordered_extent(inode,
-					cur_offset, num_bytes, num_bytes,
-					disk_bytenr, num_bytes, 0,
+					cur_offset, nocow_args.num_bytes,
+					nocow_args.num_bytes,
+					nocow_args.disk_bytenr,
+					nocow_args.num_bytes, 0,
 					1 << BTRFS_ORDERED_PREALLOC,
 					BTRFS_COMPRESS_NONE);
 			if (ret) {
 				btrfs_drop_extent_cache(inode, cur_offset,
-							cur_offset + num_bytes - 1,
-							0);
+							nocow_end, 0);
 				goto error;
 			}
 		} else {
 			ret = btrfs_add_ordered_extent(inode, cur_offset,
-						       num_bytes, num_bytes,
-						       disk_bytenr, num_bytes,
+						       nocow_args.num_bytes,
+						       nocow_args.num_bytes,
+						       nocow_args.disk_bytenr,
+						       nocow_args.num_bytes,
 						       0,
 						       1 << BTRFS_ORDERED_NOCOW,
 						       BTRFS_COMPRESS_NONE);
@@ -1907,9 +2007,10 @@ out_check:
 				goto error;
 		}
 
-		if (nocow)
-			btrfs_dec_nocow_writers(fs_info, disk_bytenr);
-		nocow = false;
+		if (nocow) {
+			btrfs_dec_nocow_writers(bg);
+			nocow = false;
+		}
 
 		if (btrfs_is_data_reloc_root(root))
 			/*
@@ -1918,10 +2019,9 @@ out_check:
 			 * from freeing metadata of created ordered extent.
 			 */
 			ret = btrfs_reloc_clone_csums(inode, cur_offset,
-						      num_bytes);
+						      nocow_args.num_bytes);
 
-		extent_clear_unlock_delalloc(inode, cur_offset,
-					     cur_offset + num_bytes - 1,
+		extent_clear_unlock_delalloc(inode, cur_offset, nocow_end,
 					     locked_page, EXTENT_LOCKED |
 					     EXTENT_DELALLOC |
 					     EXTENT_CLEAR_DATA_RESV,
@@ -1954,7 +2054,7 @@ out_check:
 
 error:
 	if (nocow)
-		btrfs_dec_nocow_writers(fs_info, disk_bytenr);
+		btrfs_dec_nocow_writers(bg);
 
 	if (ret && cur_offset < end)
 		extent_clear_unlock_delalloc(inode, cur_offset, end,
@@ -2498,9 +2598,8 @@ out:
  *
  *    c-3) otherwise:			async submit
  */
-blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
-				   int mirror_num, unsigned long bio_flags)
-
+void btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
+			   int mirror_num, enum btrfs_compression_type compress_type)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -2529,16 +2628,14 @@ blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
 		if (ret)
 			goto out;
 
-		if (bio_flags & EXTENT_BIO_COMPRESSED) {
+		if (compress_type != BTRFS_COMPRESS_NONE) {
 			/*
 			 * btrfs_submit_compressed_read will handle completing
 			 * the bio if there were any errors, so just return
 			 * here.
 			 */
-			ret = btrfs_submit_compressed_read(inode, bio,
-							   mirror_num,
-							   bio_flags);
-			goto out_no_endio;
+			btrfs_submit_compressed_read(inode, bio, mirror_num);
+			return;
 		} else {
 			/*
 			 * Lookup bio sums does extra checks around whether we
@@ -2555,7 +2652,7 @@ blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio,
 		if (btrfs_is_data_reloc_root(root))
 			goto mapit;
 		/* we're doing a write, do the async checksumming */
-		ret = btrfs_wq_submit_bio(inode, bio, mirror_num, bio_flags,
+		ret = btrfs_wq_submit_bio(inode, bio, mirror_num,
 					  0, btrfs_submit_bio_start);
 		goto out;
 	} else if (!skip_sum) {
@@ -2572,8 +2669,6 @@ out:
 		bio->bi_status = ret;
 		bio_endio(bio);
 	}
-out_no_endio:
-	return ret;
 }
 
 /*
@@ -3264,11 +3359,11 @@ static int check_data_csum(struct inode *inode, struct btrfs_bio *bbio,
 	shash->tfm = fs_info->csum_shash;
 
 	crypto_shash_digest(shash, kaddr + pgoff, len, csum);
+	kunmap_atomic(kaddr);
 
 	if (memcmp(csum, csum_expected, csum_size))
 		goto zeroit;
 
-	kunmap_atomic(kaddr);
 	return 0;
 zeroit:
 	btrfs_print_data_csum_error(BTRFS_I(inode), start, csum, csum_expected,
@@ -3276,9 +3371,7 @@ zeroit:
 	if (bbio->device)
 		btrfs_dev_stat_inc_and_print(bbio->device,
 					     BTRFS_DEV_STAT_CORRUPTION_ERRS);
-	memset(kaddr + pgoff, 1, len);
-	flush_dcache_page(page);
-	kunmap_atomic(kaddr);
+	memzero_page(page, pgoff, len);
 	return -EIO;
 }
 
@@ -3483,6 +3576,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 	u64 last_objectid = 0;
 	int ret = 0, nr_unlink = 0;
 
+	/* Bail out if the cleanup is already running. */
 	if (test_and_set_bit(BTRFS_ROOT_ORPHAN_CLEANUP, &root->state))
 		return 0;
 
@@ -3565,17 +3659,17 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 			 *
 			 * btrfs_find_orphan_roots() ran before us, which has
 			 * found all deleted roots and loaded them into
-			 * fs_info->fs_roots_radix. So here we can find if an
+			 * fs_info->fs_roots. So here we can find if an
 			 * orphan item corresponds to a deleted root by looking
-			 * up the root from that radix tree.
+			 * up the root from that xarray.
 			 */
 
-			spin_lock(&fs_info->fs_roots_radix_lock);
-			dead_root = radix_tree_lookup(&fs_info->fs_roots_radix,
-							 (unsigned long)found_key.objectid);
+			spin_lock(&fs_info->fs_roots_lock);
+			dead_root = xa_load(&fs_info->fs_roots,
+					    (unsigned long)found_key.objectid);
 			if (dead_root && btrfs_root_refs(&dead_root->root_item) == 0)
 				is_dead_root = 1;
-			spin_unlock(&fs_info->fs_roots_radix_lock);
+			spin_unlock(&fs_info->fs_roots_lock);
 
 			if (is_dead_root) {
 				/* prevent this orphan from being found again */
@@ -3815,7 +3909,7 @@ cache_index:
 	 * cache.
 	 *
 	 * This is required for both inode re-read from disk and delayed inode
-	 * in delayed_nodes_tree.
+	 * in the delayed_nodes xarray.
 	 */
 	if (BTRFS_I(inode)->last_trans == fs_info->generation)
 		set_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
@@ -4199,8 +4293,9 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir)
 	 * 1 for the dir index
 	 * 1 for the inode ref
 	 * 1 for the inode
+	 * 1 for the parent inode
 	 */
-	return btrfs_start_transaction_fallback_global_rsv(root, 5);
+	return btrfs_start_transaction_fallback_global_rsv(root, 6);
 }
 
 static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
@@ -4693,7 +4788,7 @@ int btrfs_truncate_block(struct btrfs_inode *inode, loff_t from, loff_t len,
 			goto out;
 		}
 	}
-	ret = btrfs_delalloc_reserve_metadata(inode, blocksize, blocksize);
+	ret = btrfs_delalloc_reserve_metadata(inode, blocksize, blocksize, false);
 	if (ret < 0) {
 		if (!only_release_metadata)
 			btrfs_free_reserved_data_space(inode, data_reserved,
@@ -5780,8 +5875,6 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 	struct list_head ins_list;
 	struct list_head del_list;
 	int ret;
-	struct extent_buffer *leaf;
-	int slot;
 	char *name_ptr;
 	int name_len;
 	int entries = 0;
@@ -5808,35 +5901,19 @@ again:
 	key.offset = ctx->pos;
 	key.objectid = btrfs_ino(BTRFS_I(inode));
 
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto err;
-
-	while (1) {
+	btrfs_for_each_slot(root, &key, &found_key, path, ret) {
 		struct dir_entry *entry;
-
-		leaf = path->nodes[0];
-		slot = path->slots[0];
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto err;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(leaf, &found_key, slot);
+		struct extent_buffer *leaf = path->nodes[0];
 
 		if (found_key.objectid != key.objectid)
 			break;
 		if (found_key.type != BTRFS_DIR_INDEX_KEY)
 			break;
 		if (found_key.offset < ctx->pos)
-			goto next;
+			continue;
 		if (btrfs_should_delete_dir_index(&del_list, found_key.offset))
-			goto next;
-		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
+			continue;
+		di = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dir_item);
 		name_len = btrfs_dir_name_len(leaf, di);
 		if ((total_len + sizeof(struct dir_entry) + name_len) >=
 		    PAGE_SIZE) {
@@ -5863,9 +5940,11 @@ again:
 		entries++;
 		addr += sizeof(struct dir_entry) + name_len;
 		total_len += sizeof(struct dir_entry) + name_len;
-next:
-		path->slots[0]++;
 	}
+	/* Catch error encountered during iteration */
+	if (ret < 0)
+		goto err;
+
 	btrfs_release_path(path);
 
 	ret = btrfs_filldir(private->filldir_buf, entries, ctx);
@@ -6053,6 +6132,57 @@ static int btrfs_insert_inode_locked(struct inode *inode)
 		   btrfs_find_actor, &args);
 }
 
+int btrfs_new_inode_prepare(struct btrfs_new_inode_args *args,
+			    unsigned int *trans_num_items)
+{
+	struct inode *dir = args->dir;
+	struct inode *inode = args->inode;
+	int ret;
+
+	ret = posix_acl_create(dir, &inode->i_mode, &args->default_acl, &args->acl);
+	if (ret)
+		return ret;
+
+	/* 1 to add inode item */
+	*trans_num_items = 1;
+	/* 1 to add compression property */
+	if (BTRFS_I(dir)->prop_compress)
+		(*trans_num_items)++;
+	/* 1 to add default ACL xattr */
+	if (args->default_acl)
+		(*trans_num_items)++;
+	/* 1 to add access ACL xattr */
+	if (args->acl)
+		(*trans_num_items)++;
+#ifdef CONFIG_SECURITY
+	/* 1 to add LSM xattr */
+	if (dir->i_security)
+		(*trans_num_items)++;
+#endif
+	if (args->orphan) {
+		/* 1 to add orphan item */
+		(*trans_num_items)++;
+	} else {
+		/*
+		 * 1 to add dir item
+		 * 1 to add dir index
+		 * 1 to update parent inode item
+		 *
+		 * No need for 1 unit for the inode ref item because it is
+		 * inserted in a batch together with the inode item at
+		 * btrfs_create_new_inode().
+		 */
+		*trans_num_items += 3;
+	}
+	return 0;
+}
+
+void btrfs_new_inode_args_destroy(struct btrfs_new_inode_args *args)
+{
+	posix_acl_release(args->acl);
+	posix_acl_release(args->default_acl);
+}
+
 /*
  * Inherit flags from the parent inode.
  *
@@ -6062,9 +6192,6 @@ static void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
 {
 	unsigned int flags;
 
-	if (!dir)
-		return;
-
 	flags = BTRFS_I(dir)->flags;
 
 	if (flags & BTRFS_INODE_NOCOMPRESS) {
@@ -6084,76 +6211,86 @@ static void btrfs_inherit_iflags(struct inode *inode, struct inode *dir)
 	btrfs_sync_inode_flags_to_i_flags(inode);
 }
 
-static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
-				     struct btrfs_root *root,
-				     struct user_namespace *mnt_userns,
-				     struct inode *dir,
-				     const char *name, int name_len,
-				     u64 ref_objectid, u64 objectid,
-				     umode_t mode, u64 *index)
+int btrfs_create_new_inode(struct btrfs_trans_handle *trans,
+			   struct btrfs_new_inode_args *args)
 {
-	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct inode *inode;
+	struct inode *dir = args->dir;
+	struct inode *inode = args->inode;
+	const char *name = args->orphan ? NULL : args->dentry->d_name.name;
+	int name_len = args->orphan ? 0 : args->dentry->d_name.len;
+	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
+	struct btrfs_root *root;
 	struct btrfs_inode_item *inode_item;
 	struct btrfs_key *location;
 	struct btrfs_path *path;
+	u64 objectid;
 	struct btrfs_inode_ref *ref;
 	struct btrfs_key key[2];
 	u32 sizes[2];
 	struct btrfs_item_batch batch;
 	unsigned long ptr;
-	unsigned int nofs_flag;
 	int ret;
 
 	path = btrfs_alloc_path();
 	if (!path)
-		return ERR_PTR(-ENOMEM);
-
-	nofs_flag = memalloc_nofs_save();
-	inode = new_inode(fs_info->sb);
-	memalloc_nofs_restore(nofs_flag);
-	if (!inode) {
-		btrfs_free_path(path);
-		return ERR_PTR(-ENOMEM);
-	}
+		return -ENOMEM;
 
-	/*
-	 * O_TMPFILE, set link count to 0, so that after this point,
-	 * we fill in an inode item with the correct link count.
-	 */
-	if (!name)
-		set_nlink(inode, 0);
+	if (!args->subvol)
+		BTRFS_I(inode)->root = btrfs_grab_root(BTRFS_I(dir)->root);
+	root = BTRFS_I(inode)->root;
 
-	/*
-	 * we have to initialize this early, so we can reclaim the inode
-	 * number if we fail afterwards in this function.
-	 */
+	ret = btrfs_get_free_objectid(root, &objectid);
+	if (ret)
+		goto out;
 	inode->i_ino = objectid;
 
-	if (dir && name) {
+	if (args->orphan) {
+		/*
+		 * O_TMPFILE, set link count to 0, so that after this point, we
+		 * fill in an inode item with the correct link count.
+		 */
+		set_nlink(inode, 0);
+	} else {
 		trace_btrfs_inode_request(dir);
 
-		ret = btrfs_set_inode_index(BTRFS_I(dir), index);
-		if (ret) {
-			btrfs_free_path(path);
-			iput(inode);
-			return ERR_PTR(ret);
-		}
-	} else if (dir) {
-		*index = 0;
+		ret = btrfs_set_inode_index(BTRFS_I(dir), &BTRFS_I(inode)->dir_index);
+		if (ret)
+			goto out;
 	}
-	/*
-	 * index_cnt is ignored for everything but a dir,
-	 * btrfs_set_inode_index_count has an explanation for the magic
-	 * number
-	 */
-	BTRFS_I(inode)->index_cnt = 2;
-	BTRFS_I(inode)->dir_index = *index;
-	BTRFS_I(inode)->root = btrfs_grab_root(root);
+	/* index_cnt is ignored for everything but a dir. */
+	BTRFS_I(inode)->index_cnt = BTRFS_DIR_START_INDEX;
 	BTRFS_I(inode)->generation = trans->transid;
 	inode->i_generation = BTRFS_I(inode)->generation;
 
 	/*
+	 * Subvolumes don't inherit flags from their parent directory.
+	 * Originally this was probably by accident, but we probably can't
+	 * change it now without compatibility issues.
+	 */
+	if (!args->subvol)
+		btrfs_inherit_iflags(inode, dir);
+
+	if (S_ISREG(inode->i_mode)) {
+		if (btrfs_test_opt(fs_info, NODATASUM))
+			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
+		if (btrfs_test_opt(fs_info, NODATACOW))
+			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW |
+				BTRFS_INODE_NODATASUM;
+	}
+
+	location = &BTRFS_I(inode)->location;
+	location->objectid = objectid;
+	location->offset = 0;
+	location->type = BTRFS_INODE_ITEM_KEY;
+
+	ret = btrfs_insert_inode_locked(inode);
+	if (ret < 0) {
+		if (!args->orphan)
+			BTRFS_I(dir)->index_cnt--;
+		goto out;
+	}
+
+	/*
 	 * We could have gotten an inode number from somebody who was fsynced
 	 * and then removed in this same transaction, so let's just set full
 	 * sync since it will be a full sync anyway and this will blow away the
@@ -6167,7 +6304,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 
 	sizes[0] = sizeof(struct btrfs_inode_item);
 
-	if (name) {
+	if (!args->orphan) {
 		/*
 		 * Start new inodes with an inode_ref. This is slightly more
 		 * efficient for small numbers of hard links since they will
@@ -6176,64 +6313,95 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 		 */
 		key[1].objectid = objectid;
 		key[1].type = BTRFS_INODE_REF_KEY;
-		key[1].offset = ref_objectid;
-
-		sizes[1] = name_len + sizeof(*ref);
-	}
-
-	location = &BTRFS_I(inode)->location;
-	location->objectid = objectid;
-	location->offset = 0;
-	location->type = BTRFS_INODE_ITEM_KEY;
-
-	ret = btrfs_insert_inode_locked(inode);
-	if (ret < 0) {
-		iput(inode);
-		goto fail;
+		if (args->subvol) {
+			key[1].offset = objectid;
+			sizes[1] = 2 + sizeof(*ref);
+		} else {
+			key[1].offset = btrfs_ino(BTRFS_I(dir));
+			sizes[1] = name_len + sizeof(*ref);
+		}
 	}
 
 	batch.keys = &key[0];
 	batch.data_sizes = &sizes[0];
-	batch.total_data_size = sizes[0] + (name ? sizes[1] : 0);
-	batch.nr = name ? 2 : 1;
+	batch.total_data_size = sizes[0] + (args->orphan ? 0 : sizes[1]);
+	batch.nr = args->orphan ? 1 : 2;
 	ret = btrfs_insert_empty_items(trans, root, path, &batch);
-	if (ret != 0)
-		goto fail_unlock;
-
-	inode_init_owner(mnt_userns, inode, dir, mode);
-	inode_set_bytes(inode, 0);
+	if (ret != 0) {
+		btrfs_abort_transaction(trans, ret);
+		goto discard;
+	}
 
 	inode->i_mtime = current_time(inode);
 	inode->i_atime = inode->i_mtime;
 	inode->i_ctime = inode->i_mtime;
 	BTRFS_I(inode)->i_otime = inode->i_mtime;
 
+	/*
+	 * We're going to fill the inode item now, so at this point the inode
+	 * must be fully initialized.
+	 */
+
 	inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
 				  struct btrfs_inode_item);
 	memzero_extent_buffer(path->nodes[0], (unsigned long)inode_item,
 			     sizeof(*inode_item));
 	fill_inode_item(trans, path->nodes[0], inode_item, inode);
 
-	if (name) {
+	if (!args->orphan) {
 		ref = btrfs_item_ptr(path->nodes[0], path->slots[0] + 1,
 				     struct btrfs_inode_ref);
-		btrfs_set_inode_ref_name_len(path->nodes[0], ref, name_len);
-		btrfs_set_inode_ref_index(path->nodes[0], ref, *index);
 		ptr = (unsigned long)(ref + 1);
-		write_extent_buffer(path->nodes[0], name, ptr, name_len);
+		if (args->subvol) {
+			btrfs_set_inode_ref_name_len(path->nodes[0], ref, 2);
+			btrfs_set_inode_ref_index(path->nodes[0], ref, 0);
+			write_extent_buffer(path->nodes[0], "..", ptr, 2);
+		} else {
+			btrfs_set_inode_ref_name_len(path->nodes[0], ref, name_len);
+			btrfs_set_inode_ref_index(path->nodes[0], ref,
+						  BTRFS_I(inode)->dir_index);
+			write_extent_buffer(path->nodes[0], name, ptr, name_len);
+		}
 	}
 
 	btrfs_mark_buffer_dirty(path->nodes[0]);
-	btrfs_free_path(path);
+	btrfs_release_path(path);
 
-	btrfs_inherit_iflags(inode, dir);
+	if (args->subvol) {
+		struct inode *parent;
 
-	if (S_ISREG(mode)) {
-		if (btrfs_test_opt(fs_info, NODATASUM))
-			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
-		if (btrfs_test_opt(fs_info, NODATACOW))
-			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW |
-				BTRFS_INODE_NODATASUM;
+		/*
+		 * Subvolumes inherit properties from their parent subvolume,
+		 * not the directory they were created in.
+		 */
+		parent = btrfs_iget(fs_info->sb, BTRFS_FIRST_FREE_OBJECTID,
+				    BTRFS_I(dir)->root);
+		if (IS_ERR(parent)) {
+			ret = PTR_ERR(parent);
+		} else {
+			ret = btrfs_inode_inherit_props(trans, inode, parent);
+			iput(parent);
+		}
+	} else {
+		ret = btrfs_inode_inherit_props(trans, inode, dir);
+	}
+	if (ret) {
+		btrfs_err(fs_info,
+			  "error inheriting props for ino %llu (root %llu): %d",
+			  btrfs_ino(BTRFS_I(inode)), root->root_key.objectid,
+			  ret);
+	}
+
+	/*
+	 * Subvolumes don't inherit ACLs or get passed to the LSM. This is
+	 * probably a bug.
+	 */
+	if (!args->subvol) {
+		ret = btrfs_init_inode_security(trans, args);
+		if (ret) {
+			btrfs_abort_transaction(trans, ret);
+			goto discard;
+		}
 	}
 
 	inode_tree_add(inode);
@@ -6243,21 +6411,30 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 
 	btrfs_update_root_times(trans, root);
 
-	ret = btrfs_inode_inherit_props(trans, inode, dir);
-	if (ret)
-		btrfs_err(fs_info,
-			  "error inheriting props for ino %llu (root %llu): %d",
-			btrfs_ino(BTRFS_I(inode)), root->root_key.objectid, ret);
+	if (args->orphan) {
+		ret = btrfs_orphan_add(trans, BTRFS_I(inode));
+	} else {
+		ret = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode), name,
+				     name_len, 0, BTRFS_I(inode)->dir_index);
+	}
+	if (ret) {
+		btrfs_abort_transaction(trans, ret);
+		goto discard;
+	}
 
-	return inode;
+	ret = 0;
+	goto out;
 
-fail_unlock:
+discard:
+	/*
+	 * discard_new_inode() calls iput(), but the caller owns the reference
+	 * to the inode.
+	 */
+	ihold(inode);
 	discard_new_inode(inode);
-fail:
-	if (dir && name)
-		BTRFS_I(dir)->index_cnt--;
+out:
 	btrfs_free_path(path);
-	return ERR_PTR(ret);
+	return ret;
 }
 
 /*
@@ -6349,147 +6526,71 @@ fail_dir_item:
 	return ret;
 }
 
-static int btrfs_add_nondir(struct btrfs_trans_handle *trans,
-			    struct btrfs_inode *dir, struct dentry *dentry,
-			    struct btrfs_inode *inode, int backref, u64 index)
-{
-	int err = btrfs_add_link(trans, dir, inode,
-				 dentry->d_name.name, dentry->d_name.len,
-				 backref, index);
-	if (err > 0)
-		err = -EEXIST;
-	return err;
-}
-
-static int btrfs_mknod(struct user_namespace *mnt_userns, struct inode *dir,
-		       struct dentry *dentry, umode_t mode, dev_t rdev)
+static int btrfs_create_common(struct inode *dir, struct dentry *dentry,
+			       struct inode *inode)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
-	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
-	struct inode *inode = NULL;
+	struct btrfs_new_inode_args new_inode_args = {
+		.dir = dir,
+		.dentry = dentry,
+		.inode = inode,
+	};
+	unsigned int trans_num_items;
+	struct btrfs_trans_handle *trans;
 	int err;
-	u64 objectid;
-	u64 index = 0;
 
-	/*
-	 * 2 for inode item and ref
-	 * 2 for dir items
-	 * 1 for xattr if selinux is on
-	 */
-	trans = btrfs_start_transaction(root, 5);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
-
-	err = btrfs_get_free_objectid(root, &objectid);
+	err = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items);
 	if (err)
-		goto out_unlock;
+		goto out_inode;
 
-	inode = btrfs_new_inode(trans, root, mnt_userns, dir,
-			dentry->d_name.name, dentry->d_name.len,
-			btrfs_ino(BTRFS_I(dir)), objectid, mode, &index);
-	if (IS_ERR(inode)) {
-		err = PTR_ERR(inode);
-		inode = NULL;
-		goto out_unlock;
+	trans = btrfs_start_transaction(root, trans_num_items);
+	if (IS_ERR(trans)) {
+		err = PTR_ERR(trans);
+		goto out_new_inode_args;
 	}
 
-	/*
-	* If the active LSM wants to access the inode during
-	* d_instantiate it needs these. Smack checks to see
-	* if the filesystem supports xattrs by looking at the
-	* ops vector.
-	*/
-	inode->i_op = &btrfs_special_inode_operations;
-	init_special_inode(inode, inode->i_mode, rdev);
-
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
-	if (err)
-		goto out_unlock;
-
-	err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode),
-			0, index);
-	if (err)
-		goto out_unlock;
-
-	btrfs_update_inode(trans, root, BTRFS_I(inode));
-	d_instantiate_new(dentry, inode);
+	err = btrfs_create_new_inode(trans, &new_inode_args);
+	if (!err)
+		d_instantiate_new(dentry, inode);
 
-out_unlock:
 	btrfs_end_transaction(trans);
 	btrfs_btree_balance_dirty(fs_info);
-	if (err && inode) {
-		inode_dec_link_count(inode);
-		discard_new_inode(inode);
-	}
+out_new_inode_args:
+	btrfs_new_inode_args_destroy(&new_inode_args);
+out_inode:
+	if (err)
+		iput(inode);
 	return err;
 }
 
-static int btrfs_create(struct user_namespace *mnt_userns, struct inode *dir,
-			struct dentry *dentry, umode_t mode, bool excl)
+static int btrfs_mknod(struct user_namespace *mnt_userns, struct inode *dir,
+		       struct dentry *dentry, umode_t mode, dev_t rdev)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
-	struct btrfs_trans_handle *trans;
-	struct btrfs_root *root = BTRFS_I(dir)->root;
-	struct inode *inode = NULL;
-	int err;
-	u64 objectid;
-	u64 index = 0;
+	struct inode *inode;
 
-	/*
-	 * 2 for inode item and ref
-	 * 2 for dir items
-	 * 1 for xattr if selinux is on
-	 */
-	trans = btrfs_start_transaction(root, 5);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		return -ENOMEM;
+	inode_init_owner(mnt_userns, inode, dir, mode);
+	inode->i_op = &btrfs_special_inode_operations;
+	init_special_inode(inode, inode->i_mode, rdev);
+	return btrfs_create_common(dir, dentry, inode);
+}
 
-	err = btrfs_get_free_objectid(root, &objectid);
-	if (err)
-		goto out_unlock;
+static int btrfs_create(struct user_namespace *mnt_userns, struct inode *dir,
+			struct dentry *dentry, umode_t mode, bool excl)
+{
+	struct inode *inode;
 
-	inode = btrfs_new_inode(trans, root, mnt_userns, dir,
-			dentry->d_name.name, dentry->d_name.len,
-			btrfs_ino(BTRFS_I(dir)), objectid, mode, &index);
-	if (IS_ERR(inode)) {
-		err = PTR_ERR(inode);
-		inode = NULL;
-		goto out_unlock;
-	}
-	/*
-	* If the active LSM wants to access the inode during
-	* d_instantiate it needs these. Smack checks to see
-	* if the filesystem supports xattrs by looking at the
-	* ops vector.
-	*/
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		return -ENOMEM;
+	inode_init_owner(mnt_userns, inode, dir, mode);
 	inode->i_fop = &btrfs_file_operations;
 	inode->i_op = &btrfs_file_inode_operations;
 	inode->i_mapping->a_ops = &btrfs_aops;
-
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
-	if (err)
-		goto out_unlock;
-
-	err = btrfs_update_inode(trans, root, BTRFS_I(inode));
-	if (err)
-		goto out_unlock;
-
-	err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode),
-			0, index);
-	if (err)
-		goto out_unlock;
-
-	d_instantiate_new(dentry, inode);
-
-out_unlock:
-	btrfs_end_transaction(trans);
-	if (err && inode) {
-		inode_dec_link_count(inode);
-		discard_new_inode(inode);
-	}
-	btrfs_btree_balance_dirty(fs_info);
-	return err;
+	return btrfs_create_common(dir, dentry, inode);
 }
 
 static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
@@ -6535,8 +6636,8 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	ihold(inode);
 	set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
 
-	err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry, BTRFS_I(inode),
-			1, index);
+	err = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode),
+			     dentry->d_name.name, dentry->d_name.len, 1, index);
 
 	if (err) {
 		drop_inode = 1;
@@ -6573,66 +6674,15 @@ fail:
 static int btrfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir,
 		       struct dentry *dentry, umode_t mode)
 {
-	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
-	struct inode *inode = NULL;
-	struct btrfs_trans_handle *trans;
-	struct btrfs_root *root = BTRFS_I(dir)->root;
-	int err = 0;
-	u64 objectid = 0;
-	u64 index = 0;
-
-	/*
-	 * 2 items for inode and ref
-	 * 2 items for dir items
-	 * 1 for xattr if selinux is on
-	 */
-	trans = btrfs_start_transaction(root, 5);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
-
-	err = btrfs_get_free_objectid(root, &objectid);
-	if (err)
-		goto out_fail;
-
-	inode = btrfs_new_inode(trans, root, mnt_userns, dir,
-			dentry->d_name.name, dentry->d_name.len,
-			btrfs_ino(BTRFS_I(dir)), objectid,
-			S_IFDIR | mode, &index);
-	if (IS_ERR(inode)) {
-		err = PTR_ERR(inode);
-		inode = NULL;
-		goto out_fail;
-	}
+	struct inode *inode;
 
-	/* these must be set before we unlock the inode */
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		return -ENOMEM;
+	inode_init_owner(mnt_userns, inode, dir, S_IFDIR | mode);
 	inode->i_op = &btrfs_dir_inode_operations;
 	inode->i_fop = &btrfs_dir_file_operations;
-
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
-	if (err)
-		goto out_fail;
-
-	btrfs_i_size_write(BTRFS_I(inode), 0);
-	err = btrfs_update_inode(trans, root, BTRFS_I(inode));
-	if (err)
-		goto out_fail;
-
-	err = btrfs_add_link(trans, BTRFS_I(dir), BTRFS_I(inode),
-			dentry->d_name.name,
-			dentry->d_name.len, 0, index);
-	if (err)
-		goto out_fail;
-
-	d_instantiate_new(dentry, inode);
-
-out_fail:
-	btrfs_end_transaction(trans);
-	if (err && inode) {
-		inode_dec_link_count(inode);
-		discard_new_inode(inode);
-	}
-	btrfs_btree_balance_dirty(fs_info);
-	return err;
+	return btrfs_create_common(dir, dentry, inode);
 }
 
 static noinline int uncompress_inline(struct btrfs_path *path,
@@ -7141,6 +7191,7 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
 			      u64 *ram_bytes, bool strict)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct can_nocow_file_extent_args nocow_args = { 0 };
 	struct btrfs_path *path;
 	int ret;
 	struct extent_buffer *leaf;
@@ -7148,13 +7199,7 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
 	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
-	u64 disk_bytenr;
-	u64 backref_offset;
-	u64 extent_end;
-	u64 num_bytes;
-	int slot;
 	int found_type;
-	bool nocow = (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW);
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -7165,18 +7210,17 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
 	if (ret < 0)
 		goto out;
 
-	slot = path->slots[0];
 	if (ret == 1) {
-		if (slot == 0) {
+		if (path->slots[0] == 0) {
 			/* can't find the item, must cow */
 			ret = 0;
 			goto out;
 		}
-		slot--;
+		path->slots[0]--;
 	}
 	ret = 0;
 	leaf = path->nodes[0];
-	btrfs_item_key_to_cpu(leaf, &key, slot);
+	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
 	if (key.objectid != btrfs_ino(BTRFS_I(inode)) ||
 	    key.type != BTRFS_EXTENT_DATA_KEY) {
 		/* not our file or wrong item type, must cow */
@@ -7188,55 +7232,38 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
 		goto out;
 	}
 
-	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
-	found_type = btrfs_file_extent_type(leaf, fi);
-	if (found_type != BTRFS_FILE_EXTENT_REG &&
-	    found_type != BTRFS_FILE_EXTENT_PREALLOC) {
-		/* not a regular extent, must cow */
-		goto out;
-	}
-
-	if (!nocow && found_type == BTRFS_FILE_EXTENT_REG)
+	if (btrfs_file_extent_end(path) <= offset)
 		goto out;
 
-	extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
-	if (extent_end <= offset)
-		goto out;
+	fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item);
+	found_type = btrfs_file_extent_type(leaf, fi);
+	if (ram_bytes)
+		*ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
 
-	disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
-	if (disk_bytenr == 0)
-		goto out;
+	nocow_args.start = offset;
+	nocow_args.end = offset + *len - 1;
+	nocow_args.strict = strict;
+	nocow_args.free_path = true;
 
-	if (btrfs_file_extent_compression(leaf, fi) ||
-	    btrfs_file_extent_encryption(leaf, fi) ||
-	    btrfs_file_extent_other_encoding(leaf, fi))
-		goto out;
+	ret = can_nocow_file_extent(path, &key, BTRFS_I(inode), &nocow_args);
+	/* can_nocow_file_extent() has freed the path. */
+	path = NULL;
 
-	/*
-	 * Do the same check as in btrfs_cross_ref_exist but without the
-	 * unnecessary search.
-	 */
-	if (!strict &&
-	    (btrfs_file_extent_generation(leaf, fi) <=
-	     btrfs_root_last_snapshot(&root->root_item)))
+	if (ret != 1) {
+		/* Treat errors as not being able to NOCOW. */
+		ret = 0;
 		goto out;
-
-	backref_offset = btrfs_file_extent_offset(leaf, fi);
-
-	if (orig_start) {
-		*orig_start = key.offset - backref_offset;
-		*orig_block_len = btrfs_file_extent_disk_num_bytes(leaf, fi);
-		*ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi);
 	}
 
-	if (btrfs_extent_readonly(fs_info, disk_bytenr))
+	ret = 0;
+	if (btrfs_extent_readonly(fs_info, nocow_args.disk_bytenr))
 		goto out;
 
-	num_bytes = min(offset + *len, extent_end) - offset;
-	if (!nocow && found_type == BTRFS_FILE_EXTENT_PREALLOC) {
+	if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW) &&
+	    found_type == BTRFS_FILE_EXTENT_PREALLOC) {
 		u64 range_end;
 
-		range_end = round_up(offset + num_bytes,
+		range_end = round_up(offset + nocow_args.num_bytes,
 				     root->fs_info->sectorsize) - 1;
 		ret = test_range_bit(io_tree, offset, range_end,
 				     EXTENT_DELALLOC, 0, NULL);
@@ -7246,36 +7273,12 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len,
 		}
 	}
 
-	btrfs_release_path(path);
-
-	/*
-	 * look for other files referencing this extent, if we
-	 * find any we must cow
-	 */
-
-	ret = btrfs_cross_ref_exist(root, btrfs_ino(BTRFS_I(inode)),
-				    key.offset - backref_offset, disk_bytenr,
-				    strict);
-	if (ret) {
-		ret = 0;
-		goto out;
-	}
+	if (orig_start)
+		*orig_start = key.offset - nocow_args.extent_offset;
+	if (orig_block_len)
+		*orig_block_len = nocow_args.disk_num_bytes;
 
-	/*
-	 * adjust disk_bytenr and num_bytes to cover just the bytes
-	 * in this extent we are about to write.  If there
-	 * are any csums in that range we have to cow in order
-	 * to keep the csums correct
-	 */
-	disk_bytenr += backref_offset;
-	disk_bytenr += offset - key.offset;
-	if (csum_exist_in_range(fs_info, disk_bytenr, num_bytes))
-		goto out;
-	/*
-	 * all of the above have passed, it is safe to overwrite this extent
-	 * without cow
-	 */
-	*len = num_bytes;
+	*len = nocow_args.num_bytes;
 	ret = 1;
 out:
 	btrfs_free_path(path);
@@ -7283,14 +7286,22 @@ out:
 }
 
 static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
-			      struct extent_state **cached_state, bool writing)
+			      struct extent_state **cached_state,
+			      unsigned int iomap_flags)
 {
+	const bool writing = (iomap_flags & IOMAP_WRITE);
+	const bool nowait = (iomap_flags & IOMAP_NOWAIT);
+	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
 	struct btrfs_ordered_extent *ordered;
 	int ret = 0;
 
 	while (1) {
-		lock_extent_bits(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-				 cached_state);
+		if (nowait) {
+			if (!try_lock_extent(io_tree, lockstart, lockend))
+				return -EAGAIN;
+		} else {
+			lock_extent_bits(io_tree, lockstart, lockend, cached_state);
+		}
 		/*
 		 * We're concerned with the entire range that we're going to be
 		 * doing DIO to, so we need to make sure there's no ordered
@@ -7311,10 +7322,14 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
 							 lockstart, lockend)))
 			break;
 
-		unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
-				     cached_state);
+		unlock_extent_cached(io_tree, lockstart, lockend, cached_state);
 
 		if (ordered) {
+			if (nowait) {
+				btrfs_put_ordered_extent(ordered);
+				ret = -EAGAIN;
+				break;
+			}
 			/*
 			 * If we are doing a DIO read and the ordered extent we
 			 * found is for a buffered write, we can not wait for it
@@ -7334,7 +7349,7 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
 			    test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags))
 				btrfs_start_ordered_extent(ordered, 1);
 			else
-				ret = -ENOTBLK;
+				ret = nowait ? -EAGAIN : -ENOTBLK;
 			btrfs_put_ordered_extent(ordered);
 		} else {
 			/*
@@ -7350,7 +7365,7 @@ static int lock_extent_direct(struct inode *inode, u64 lockstart, u64 lockend,
 			 * ordered extent to complete while holding a lock on
 			 * that page.
 			 */
-			ret = -ENOTBLK;
+			ret = nowait ? -EAGAIN : -ENOTBLK;
 		}
 
 		if (ret)
@@ -7424,12 +7439,15 @@ static struct extent_map *create_io_em(struct btrfs_inode *inode, u64 start,
 static int btrfs_get_blocks_direct_write(struct extent_map **map,
 					 struct inode *inode,
 					 struct btrfs_dio_data *dio_data,
-					 u64 start, u64 len)
+					 u64 start, u64 len,
+					 unsigned int iomap_flags)
 {
+	const bool nowait = (iomap_flags & IOMAP_NOWAIT);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map *em = *map;
 	int type;
 	u64 block_start, orig_start, orig_block_len, ram_bytes;
+	struct btrfs_block_group *bg;
 	bool can_nocow = false;
 	bool space_reserved = false;
 	u64 prev_len;
@@ -7455,9 +7473,11 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 		block_start = em->block_start + (start - em->start);
 
 		if (can_nocow_extent(inode, start, &len, &orig_start,
-				     &orig_block_len, &ram_bytes, false) == 1 &&
-		    btrfs_inc_nocow_writers(fs_info, block_start))
-			can_nocow = true;
+				     &orig_block_len, &ram_bytes, false) == 1) {
+			bg = btrfs_inc_nocow_writers(fs_info, block_start);
+			if (bg)
+				can_nocow = true;
+		}
 	}
 
 	prev_len = len;
@@ -7465,12 +7485,15 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 		struct extent_map *em2;
 
 		/* We can NOCOW, so only need to reserve metadata space. */
-		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len, len);
+		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len, len,
+						      nowait);
 		if (ret < 0) {
 			/* Our caller expects us to free the input extent map. */
 			free_extent_map(em);
 			*map = NULL;
-			btrfs_dec_nocow_writers(fs_info, block_start);
+			btrfs_dec_nocow_writers(bg);
+			if (nowait && (ret == -ENOSPC || ret == -EDQUOT))
+				ret = -EAGAIN;
 			goto out;
 		}
 		space_reserved = true;
@@ -7479,7 +7502,7 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 					      orig_start, block_start,
 					      len, orig_block_len,
 					      ram_bytes, type);
-		btrfs_dec_nocow_writers(fs_info, block_start);
+		btrfs_dec_nocow_writers(bg);
 		if (type == BTRFS_ORDERED_PREALLOC) {
 			free_extent_map(em);
 			*map = em = em2;
@@ -7489,15 +7512,29 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 			ret = PTR_ERR(em2);
 			goto out;
 		}
+
+		dio_data->nocow_done = true;
 	} else {
 		/* Our caller expects us to free the input extent map. */
 		free_extent_map(em);
 		*map = NULL;
 
-		/* We have to COW, so need to reserve metadata and data space. */
-		ret = btrfs_delalloc_reserve_space(BTRFS_I(inode),
-						   &dio_data->data_reserved,
-						   start, len);
+		if (nowait)
+			return -EAGAIN;
+
+		/*
+		 * If we could not allocate data space before locking the file
+		 * range and we can't do a NOCOW write, then we have to fail.
+		 */
+		if (!dio_data->data_space_reserved)
+			return -ENOSPC;
+
+		/*
+		 * We have to COW and we have already reserved data space before,
+		 * so now we reserve only metadata.
+		 */
+		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode), len, len,
+						      false);
 		if (ret < 0)
 			goto out;
 		space_reserved = true;
@@ -7510,10 +7547,8 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 		*map = em;
 		len = min(len, em->len - (start - em->start));
 		if (len < prev_len)
-			btrfs_delalloc_release_space(BTRFS_I(inode),
-						     dio_data->data_reserved,
-						     start + len, prev_len - len,
-						     true);
+			btrfs_delalloc_release_metadata(BTRFS_I(inode),
+							prev_len - len, true);
 	}
 
 	/*
@@ -7531,15 +7566,7 @@ static int btrfs_get_blocks_direct_write(struct extent_map **map,
 out:
 	if (ret && space_reserved) {
 		btrfs_delalloc_release_extents(BTRFS_I(inode), len);
-		if (can_nocow) {
-			btrfs_delalloc_release_metadata(BTRFS_I(inode), len, true);
-		} else {
-			btrfs_delalloc_release_space(BTRFS_I(inode),
-						     dio_data->data_reserved,
-						     start, len, true);
-			extent_changeset_free(dio_data->data_reserved);
-			dio_data->data_reserved = NULL;
-		}
+		btrfs_delalloc_release_metadata(BTRFS_I(inode), len, true);
 	}
 	return ret;
 }
@@ -7548,14 +7575,16 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 		loff_t length, unsigned int flags, struct iomap *iomap,
 		struct iomap *srcmap)
 {
+	struct iomap_iter *iter = container_of(iomap, struct iomap_iter, iomap);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_map *em;
 	struct extent_state *cached_state = NULL;
-	struct btrfs_dio_data *dio_data = NULL;
+	struct btrfs_dio_data *dio_data = iter->private;
 	u64 lockstart, lockend;
 	const bool write = !!(flags & IOMAP_WRITE);
 	int ret = 0;
 	u64 len = length;
+	const u64 data_alloc_len = length;
 	bool unlock_extents = false;
 
 	if (!write)
@@ -7565,34 +7594,67 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 	lockend = start + len - 1;
 
 	/*
-	 * The generic stuff only does filemap_write_and_wait_range, which
-	 * isn't enough if we've written compressed pages to this area, so we
-	 * need to flush the dirty pages again to make absolutely sure that any
-	 * outstanding dirty pages are on disk.
+	 * iomap_dio_rw() only does filemap_write_and_wait_range(), which isn't
+	 * enough if we've written compressed pages to this area, so we need to
+	 * flush the dirty pages again to make absolutely sure that any
+	 * outstanding dirty pages are on disk - the first flush only starts
+	 * compression on the data, while keeping the pages locked, so by the
+	 * time the second flush returns we know bios for the compressed pages
+	 * were submitted and finished, and the pages no longer under writeback.
+	 *
+	 * If we have a NOWAIT request and we have any pages in the range that
+	 * are locked, likely due to compression still in progress, we don't want
+	 * to block on page locks. We also don't want to block on pages marked as
+	 * dirty or under writeback (same as for the non-compression case).
+	 * iomap_dio_rw() did the same check, but after that and before we got
+	 * here, mmap'ed writes may have happened or buffered reads started
+	 * (readpage() and readahead(), which lock pages), as we haven't locked
+	 * the file range yet.
 	 */
 	if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
 		     &BTRFS_I(inode)->runtime_flags)) {
-		ret = filemap_fdatawrite_range(inode->i_mapping, start,
-					       start + length - 1);
-		if (ret)
-			return ret;
+		if (flags & IOMAP_NOWAIT) {
+			if (filemap_range_needs_writeback(inode->i_mapping,
+							  lockstart, lockend))
+				return -EAGAIN;
+		} else {
+			ret = filemap_fdatawrite_range(inode->i_mapping, start,
+						       start + length - 1);
+			if (ret)
+				return ret;
+		}
 	}
 
-	dio_data = kzalloc(sizeof(*dio_data), GFP_NOFS);
-	if (!dio_data)
-		return -ENOMEM;
-
-	iomap->private = dio_data;
+	memset(dio_data, 0, sizeof(*dio_data));
 
+	/*
+	 * We always try to allocate data space and must do it before locking
+	 * the file range, to avoid deadlocks with concurrent writes to the same
+	 * range if the range has several extents and the writes don't expand the
+	 * current i_size (the inode lock is taken in shared mode). If we fail to
+	 * allocate data space here we continue and later, after locking the
+	 * file range, we fail with ENOSPC only if we figure out we can not do a
+	 * NOCOW write.
+	 */
+	if (write && !(flags & IOMAP_NOWAIT)) {
+		ret = btrfs_check_data_free_space(BTRFS_I(inode),
+						  &dio_data->data_reserved,
+						  start, data_alloc_len);
+		if (!ret)
+			dio_data->data_space_reserved = true;
+		else if (ret && !(BTRFS_I(inode)->flags &
+				  (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)))
+			goto err;
+	}
 
 	/*
 	 * If this errors out it's because we couldn't invalidate pagecache for
-	 * this range and we need to fallback to buffered.
+	 * this range and we need to fallback to buffered IO, or we are doing a
+	 * NOWAIT read/write and we need to block.
 	 */
-	if (lock_extent_direct(inode, lockstart, lockend, &cached_state, write)) {
-		ret = -ENOTBLK;
+	ret = lock_extent_direct(inode, lockstart, lockend, &cached_state, flags);
+	if (ret < 0)
 		goto err;
-	}
 
 	em = btrfs_get_extent(BTRFS_I(inode), NULL, 0, start, len);
 	if (IS_ERR(em)) {
@@ -7652,12 +7714,30 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start,
 
 	if (write) {
 		ret = btrfs_get_blocks_direct_write(&em, inode, dio_data,
-						    start, len);
+						    start, len, flags);
 		if (ret < 0)
 			goto unlock_err;
 		unlock_extents = true;
 		/* Recalc len in case the new em is smaller than requested */
 		len = min(len, em->len - (start - em->start));
+		if (dio_data->data_space_reserved) {
+			u64 release_offset;
+			u64 release_len = 0;
+
+			if (dio_data->nocow_done) {
+				release_offset = start;
+				release_len = data_alloc_len;
+			} else if (len < data_alloc_len) {
+				release_offset = start + len;
+				release_len = data_alloc_len - len;
+			}
+
+			if (release_len > 0)
+				btrfs_free_reserved_data_space(BTRFS_I(inode),
+							       dio_data->data_reserved,
+							       release_offset,
+							       release_len);
+		}
 	} else {
 		/*
 		 * We need to unlock only the end area that we aren't using.
@@ -7702,7 +7782,12 @@ unlock_err:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend,
 			     &cached_state);
 err:
-	kfree(dio_data);
+	if (dio_data->data_space_reserved) {
+		btrfs_free_reserved_data_space(BTRFS_I(inode),
+					       dio_data->data_reserved,
+					       start, data_alloc_len);
+		extent_changeset_free(dio_data->data_reserved);
+	}
 
 	return ret;
 }
@@ -7710,15 +7795,16 @@ err:
 static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length,
 		ssize_t written, unsigned int flags, struct iomap *iomap)
 {
-	int ret = 0;
-	struct btrfs_dio_data *dio_data = iomap->private;
+	struct iomap_iter *iter = container_of(iomap, struct iomap_iter, iomap);
+	struct btrfs_dio_data *dio_data = iter->private;
 	size_t submitted = dio_data->submitted;
 	const bool write = !!(flags & IOMAP_WRITE);
+	int ret = 0;
 
 	if (!write && (iomap->type == IOMAP_HOLE)) {
 		/* If reading from a hole, unlock and return */
 		unlock_extent(&BTRFS_I(inode)->io_tree, pos, pos + length - 1);
-		goto out;
+		return 0;
 	}
 
 	if (submitted < length) {
@@ -7735,10 +7821,6 @@ static int btrfs_dio_iomap_end(struct inode *inode, loff_t pos, loff_t length,
 
 	if (write)
 		extent_changeset_free(dio_data->data_reserved);
-out:
-	kfree(dio_data);
-	iomap->private = NULL;
-
 	return ret;
 }
 
@@ -7751,40 +7833,36 @@ static void btrfs_dio_private_put(struct btrfs_dio_private *dip)
 	if (!refcount_dec_and_test(&dip->refs))
 		return;
 
-	if (btrfs_op(dip->dio_bio) == BTRFS_MAP_WRITE) {
+	if (btrfs_op(&dip->bio) == BTRFS_MAP_WRITE) {
 		__endio_write_update_ordered(BTRFS_I(dip->inode),
 					     dip->file_offset,
 					     dip->bytes,
-					     !dip->dio_bio->bi_status);
+					     !dip->bio.bi_status);
 	} else {
 		unlock_extent(&BTRFS_I(dip->inode)->io_tree,
 			      dip->file_offset,
 			      dip->file_offset + dip->bytes - 1);
 	}
 
-	bio_endio(dip->dio_bio);
-	kfree(dip);
+	kfree(dip->csums);
+	bio_endio(&dip->bio);
 }
 
-static blk_status_t submit_dio_repair_bio(struct inode *inode, struct bio *bio,
-					  int mirror_num,
-					  unsigned long bio_flags)
+static void submit_dio_repair_bio(struct inode *inode, struct bio *bio,
+				  int mirror_num,
+				  enum btrfs_compression_type compress_type)
 {
 	struct btrfs_dio_private *dip = bio->bi_private;
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	blk_status_t ret;
 
 	BUG_ON(bio_op(bio) == REQ_OP_WRITE);
 
-	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA);
-	if (ret)
-		return ret;
+	if (btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA))
+		return;
 
 	refcount_inc(&dip->refs);
-	ret = btrfs_map_bio(fs_info, bio, mirror_num);
-	if (ret)
+	if (btrfs_map_bio(fs_info, bio, mirror_num))
 		refcount_dec(&dip->refs);
-	return ret;
 }
 
 static blk_status_t btrfs_check_read_dio_bio(struct btrfs_dio_private *dip,
@@ -7869,7 +7947,7 @@ static void btrfs_end_dio_bio(struct bio *bio)
 		err = btrfs_check_read_dio_bio(dip, bbio, !err);
 
 	if (err)
-		dip->dio_bio->bi_status = err;
+		dip->bio.bi_status = err;
 
 	btrfs_record_physical_zoned(dip->inode, bbio->file_offset, bio);
 
@@ -7899,7 +7977,7 @@ static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio,
 		goto map;
 
 	if (write && async_submit) {
-		ret = btrfs_wq_submit_bio(inode, bio, 0, 0, file_offset,
+		ret = btrfs_wq_submit_bio(inode, bio, 0, file_offset,
 					  btrfs_submit_bio_start_direct_io);
 		goto err;
 	} else if (write) {
@@ -7924,50 +8002,16 @@ err:
 	return ret;
 }
 
-/*
- * If this succeeds, the btrfs_dio_private is responsible for cleaning up locked
- * or ordered extents whether or not we submit any bios.
- */
-static struct btrfs_dio_private *btrfs_create_dio_private(struct bio *dio_bio,
-							  struct inode *inode,
-							  loff_t file_offset)
-{
-	const bool write = (btrfs_op(dio_bio) == BTRFS_MAP_WRITE);
-	const bool csum = !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM);
-	size_t dip_size;
-	struct btrfs_dio_private *dip;
-
-	dip_size = sizeof(*dip);
-	if (!write && csum) {
-		struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-		size_t nblocks;
-
-		nblocks = dio_bio->bi_iter.bi_size >> fs_info->sectorsize_bits;
-		dip_size += fs_info->csum_size * nblocks;
-	}
-
-	dip = kzalloc(dip_size, GFP_NOFS);
-	if (!dip)
-		return NULL;
-
-	dip->inode = inode;
-	dip->file_offset = file_offset;
-	dip->bytes = dio_bio->bi_iter.bi_size;
-	dip->disk_bytenr = dio_bio->bi_iter.bi_sector << 9;
-	dip->dio_bio = dio_bio;
-	refcount_set(&dip->refs, 1);
-	return dip;
-}
-
 static void btrfs_submit_direct(const struct iomap_iter *iter,
 		struct bio *dio_bio, loff_t file_offset)
 {
+	struct btrfs_dio_private *dip =
+		container_of(dio_bio, struct btrfs_dio_private, bio);
 	struct inode *inode = iter->inode;
 	const bool write = (btrfs_op(dio_bio) == BTRFS_MAP_WRITE);
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	const bool raid56 = (btrfs_data_alloc_profile(fs_info) &
 			     BTRFS_BLOCK_GROUP_RAID56_MASK);
-	struct btrfs_dio_private *dip;
 	struct bio *bio;
 	u64 start_sector;
 	int async_submit = 0;
@@ -7978,27 +8022,28 @@ static void btrfs_submit_direct(const struct iomap_iter *iter,
 	int ret;
 	blk_status_t status;
 	struct btrfs_io_geometry geom;
-	struct btrfs_dio_data *dio_data = iter->iomap.private;
+	struct btrfs_dio_data *dio_data = iter->private;
 	struct extent_map *em = NULL;
 
-	dip = btrfs_create_dio_private(dio_bio, inode, file_offset);
-	if (!dip) {
-		if (!write) {
-			unlock_extent(&BTRFS_I(inode)->io_tree, file_offset,
-				file_offset + dio_bio->bi_iter.bi_size - 1);
-		}
-		dio_bio->bi_status = BLK_STS_RESOURCE;
-		bio_endio(dio_bio);
-		return;
-	}
+	dip->inode = inode;
+	dip->file_offset = file_offset;
+	dip->bytes = dio_bio->bi_iter.bi_size;
+	refcount_set(&dip->refs, 1);
+	dip->csums = NULL;
+
+	if (!write && !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
+		unsigned int nr_sectors =
+			(dio_bio->bi_iter.bi_size >> fs_info->sectorsize_bits);
 
-	if (!write) {
 		/*
 		 * Load the csums up front to reduce csum tree searches and
 		 * contention when submitting bios.
-		 *
-		 * If we have csums disabled this will do nothing.
 		 */
+		status = BLK_STS_RESOURCE;
+		dip->csums = kcalloc(nr_sectors, fs_info->csum_size, GFP_NOFS);
+		if (!dip)
+			goto out_err;
+
 		status = btrfs_lookup_bio_sums(inode, dio_bio, dip->csums);
 		if (status != BLK_STS_OK)
 			goto out_err;
@@ -8088,19 +8133,28 @@ static void btrfs_submit_direct(const struct iomap_iter *iter,
 out_err_em:
 	free_extent_map(em);
 out_err:
-	dip->dio_bio->bi_status = status;
+	dio_bio->bi_status = status;
 	btrfs_dio_private_put(dip);
 }
 
-const struct iomap_ops btrfs_dio_iomap_ops = {
+static const struct iomap_ops btrfs_dio_iomap_ops = {
 	.iomap_begin            = btrfs_dio_iomap_begin,
 	.iomap_end              = btrfs_dio_iomap_end,
 };
 
-const struct iomap_dio_ops btrfs_dio_ops = {
+static const struct iomap_dio_ops btrfs_dio_ops = {
 	.submit_io		= btrfs_submit_direct,
+	.bio_set		= &btrfs_dio_bioset,
 };
 
+ssize_t btrfs_dio_rw(struct kiocb *iocb, struct iov_iter *iter, size_t done_before)
+{
+	struct btrfs_dio_data data;
+
+	return iomap_dio_rw(iocb, iter, &btrfs_dio_iomap_ops, &btrfs_dio_ops,
+			    IOMAP_DIO_PARTIAL, &data, done_before);
+}
+
 static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			u64 start, u64 len)
 {
@@ -8113,27 +8167,6 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	return extent_fiemap(BTRFS_I(inode), fieinfo, start, len);
 }
 
-int btrfs_readpage(struct file *file, struct page *page)
-{
-	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
-	u64 start = page_offset(page);
-	u64 end = start + PAGE_SIZE - 1;
-	struct btrfs_bio_ctrl bio_ctrl = { 0 };
-	int ret;
-
-	btrfs_lock_and_flush_ordered_range(inode, start, end, NULL);
-
-	ret = btrfs_do_readpage(page, NULL, &bio_ctrl, 0, NULL);
-	if (bio_ctrl.bio) {
-		int ret2;
-
-		ret2 = submit_one_bio(bio_ctrl.bio, 0, bio_ctrl.bio_flags);
-		if (ret == 0)
-			ret = ret2;
-	}
-	return ret;
-}
-
 static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
 {
 	struct inode *inode = page->mapping->host;
@@ -8182,7 +8215,7 @@ static void wait_subpage_spinlock(struct page *page)
 	struct btrfs_fs_info *fs_info = btrfs_sb(page->mapping->host->i_sb);
 	struct btrfs_subpage *subpage;
 
-	if (fs_info->sectorsize == PAGE_SIZE)
+	if (!btrfs_is_subpage(fs_info, page))
 		return;
 
 	ASSERT(PagePrivate(page) && page->private);
@@ -8764,46 +8797,23 @@ out:
 	return ret;
 }
 
-/*
- * create a new subvolume directory/inode (helper for the ioctl).
- */
-int btrfs_create_subvol_root(struct btrfs_trans_handle *trans,
-			     struct btrfs_root *new_root,
-			     struct btrfs_root *parent_root,
-			     struct user_namespace *mnt_userns)
+struct inode *btrfs_new_subvol_inode(struct user_namespace *mnt_userns,
+				     struct inode *dir)
 {
 	struct inode *inode;
-	int err;
-	u64 index = 0;
-	u64 ino;
-
-	err = btrfs_get_free_objectid(new_root, &ino);
-	if (err < 0)
-		return err;
-
-	inode = btrfs_new_inode(trans, new_root, mnt_userns, NULL, "..", 2,
-				ino, ino,
-				S_IFDIR | (~current_umask() & S_IRWXUGO),
-				&index);
-	if (IS_ERR(inode))
-		return PTR_ERR(inode);
-	inode->i_op = &btrfs_dir_inode_operations;
-	inode->i_fop = &btrfs_dir_file_operations;
-
-	set_nlink(inode, 1);
-	btrfs_i_size_write(BTRFS_I(inode), 0);
-	unlock_new_inode(inode);
-
-	err = btrfs_subvol_inherit_props(trans, new_root, parent_root);
-	if (err)
-		btrfs_err(new_root->fs_info,
-			  "error inheriting subvolume %llu properties: %d",
-			  new_root->root_key.objectid, err);
 
-	err = btrfs_update_inode(trans, new_root, BTRFS_I(inode));
-
-	iput(inode);
-	return err;
+	inode = new_inode(dir->i_sb);
+	if (inode) {
+		/*
+		 * Subvolumes don't inherit the sgid bit or the parent's gid if
+		 * the parent's sgid bit is set. This is probably a bug.
+		 */
+		inode_init_owner(mnt_userns, inode, NULL,
+				 S_IFDIR | (~current_umask() & S_IRWXUGO));
+		inode->i_op = &btrfs_dir_inode_operations;
+		inode->i_fop = &btrfs_dir_file_operations;
+	}
+	return inode;
 }
 
 struct inode *btrfs_alloc_inode(struct super_block *sb)
@@ -8943,7 +8953,7 @@ int btrfs_drop_inode(struct inode *inode)
 
 static void init_once(void *foo)
 {
-	struct btrfs_inode *ei = (struct btrfs_inode *) foo;
+	struct btrfs_inode *ei = foo;
 
 	inode_init_once(&ei->vfs_inode);
 }
@@ -8955,6 +8965,7 @@ void __cold btrfs_destroy_cachep(void)
 	 * destroy cache.
 	 */
 	rcu_barrier();
+	bioset_exit(&btrfs_dio_bioset);
 	kmem_cache_destroy(btrfs_inode_cachep);
 	kmem_cache_destroy(btrfs_trans_handle_cachep);
 	kmem_cache_destroy(btrfs_path_cachep);
@@ -8995,6 +9006,11 @@ int __init btrfs_init_cachep(void)
 	if (!btrfs_free_space_bitmap_cachep)
 		goto fail;
 
+	if (bioset_init(&btrfs_dio_bioset, BIO_POOL_SIZE,
+			offsetof(struct btrfs_dio_private, bio),
+			BIOSET_NEED_BVECS))
+		goto fail;
+
 	return 0;
 fail:
 	btrfs_destroy_cachep();
@@ -9050,6 +9066,7 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(old_dir->i_sb);
 	struct btrfs_trans_handle *trans;
+	unsigned int trans_num_items;
 	struct btrfs_root *root = BTRFS_I(old_dir)->root;
 	struct btrfs_root *dest = BTRFS_I(new_dir)->root;
 	struct inode *new_inode = new_dentry->d_inode;
@@ -9081,14 +9098,37 @@ static int btrfs_rename_exchange(struct inode *old_dir,
 		down_read(&fs_info->subvol_sem);
 
 	/*
-	 * We want to reserve the absolute worst case amount of items.  So if
-	 * both inodes are subvols and we need to unlink them then that would
-	 * require 4 item modifications, but if they are both normal inodes it
-	 * would require 5 item modifications, so we'll assume their normal
-	 * inodes.  So 5 * 2 is 10, plus 2 for the new links, so 12 total items
-	 * should cover the worst case number of items we'll modify.
+	 * For each inode:
+	 * 1 to remove old dir item
+	 * 1 to remove old dir index
+	 * 1 to add new dir item
+	 * 1 to add new dir index
+	 * 1 to update parent inode
+	 *
+	 * If the parents are the same, we only need to account for one
 	 */
-	trans = btrfs_start_transaction(root, 12);
+	trans_num_items = (old_dir == new_dir ? 9 : 10);
+	if (old_ino == BTRFS_FIRST_FREE_OBJECTID) {
+		/*
+		 * 1 to remove old root ref
+		 * 1 to remove old root backref
+		 * 1 to add new root ref
+		 * 1 to add new root backref
+		 */
+		trans_num_items += 4;
+	} else {
+		/*
+		 * 1 to update inode item
+		 * 1 to remove old inode ref
+		 * 1 to add new inode ref
+		 */
+		trans_num_items += 3;
+	}
+	if (new_ino == BTRFS_FIRST_FREE_OBJECTID)
+		trans_num_items += 4;
+	else
+		trans_num_items += 3;
+	trans = btrfs_start_transaction(root, trans_num_items);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
 		goto out_notrans;
@@ -9255,56 +9295,19 @@ out_notrans:
 	return ret;
 }
 
-static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans,
-				     struct btrfs_root *root,
-				     struct user_namespace *mnt_userns,
-				     struct inode *dir,
-				     struct dentry *dentry)
+static struct inode *new_whiteout_inode(struct user_namespace *mnt_userns,
+					struct inode *dir)
 {
-	int ret;
 	struct inode *inode;
-	u64 objectid;
-	u64 index;
-
-	ret = btrfs_get_free_objectid(root, &objectid);
-	if (ret)
-		return ret;
 
-	inode = btrfs_new_inode(trans, root, mnt_userns, dir,
-				dentry->d_name.name,
-				dentry->d_name.len,
-				btrfs_ino(BTRFS_I(dir)),
-				objectid,
-				S_IFCHR | WHITEOUT_MODE,
-				&index);
-
-	if (IS_ERR(inode)) {
-		ret = PTR_ERR(inode);
-		return ret;
+	inode = new_inode(dir->i_sb);
+	if (inode) {
+		inode_init_owner(mnt_userns, inode, dir,
+				 S_IFCHR | WHITEOUT_MODE);
+		inode->i_op = &btrfs_special_inode_operations;
+		init_special_inode(inode, inode->i_mode, WHITEOUT_DEV);
 	}
-
-	inode->i_op = &btrfs_special_inode_operations;
-	init_special_inode(inode, inode->i_mode,
-		WHITEOUT_DEV);
-
-	ret = btrfs_init_inode_security(trans, inode, dir,
-				&dentry->d_name);
-	if (ret)
-		goto out;
-
-	ret = btrfs_add_nondir(trans, BTRFS_I(dir), dentry,
-				BTRFS_I(inode), 0, index);
-	if (ret)
-		goto out;
-
-	ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
-out:
-	unlock_new_inode(inode);
-	if (ret)
-		inode_dec_link_count(inode);
-	iput(inode);
-
-	return ret;
+	return inode;
 }
 
 static int btrfs_rename(struct user_namespace *mnt_userns,
@@ -9313,6 +9316,10 @@ static int btrfs_rename(struct user_namespace *mnt_userns,
 			unsigned int flags)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(old_dir->i_sb);
+	struct btrfs_new_inode_args whiteout_args = {
+		.dir = old_dir,
+		.dentry = old_dentry,
+	};
 	struct btrfs_trans_handle *trans;
 	unsigned int trans_num_items;
 	struct btrfs_root *root = BTRFS_I(old_dir)->root;
@@ -9367,23 +9374,56 @@ static int btrfs_rename(struct user_namespace *mnt_userns,
 	if (new_inode && S_ISREG(old_inode->i_mode) && new_inode->i_size)
 		filemap_flush(old_inode->i_mapping);
 
-	/* close the racy window with snapshot create/destroy ioctl */
-	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
+	if (flags & RENAME_WHITEOUT) {
+		whiteout_args.inode = new_whiteout_inode(mnt_userns, old_dir);
+		if (!whiteout_args.inode)
+			return -ENOMEM;
+		ret = btrfs_new_inode_prepare(&whiteout_args, &trans_num_items);
+		if (ret)
+			goto out_whiteout_inode;
+	} else {
+		/* 1 to update the old parent inode. */
+		trans_num_items = 1;
+	}
+
+	if (old_ino == BTRFS_FIRST_FREE_OBJECTID) {
+		/* Close the race window with snapshot create/destroy ioctl */
 		down_read(&fs_info->subvol_sem);
+		/*
+		 * 1 to remove old root ref
+		 * 1 to remove old root backref
+		 * 1 to add new root ref
+		 * 1 to add new root backref
+		 */
+		trans_num_items += 4;
+	} else {
+		/*
+		 * 1 to update inode
+		 * 1 to remove old inode ref
+		 * 1 to add new inode ref
+		 */
+		trans_num_items += 3;
+	}
 	/*
-	 * We want to reserve the absolute worst case amount of items.  So if
-	 * both inodes are subvols and we need to unlink them then that would
-	 * require 4 item modifications, but if they are both normal inodes it
-	 * would require 5 item modifications, so we'll assume they are normal
-	 * inodes.  So 5 * 2 is 10, plus 1 for the new link, so 11 total items
-	 * should cover the worst case number of items we'll modify.
-	 * If our rename has the whiteout flag, we need more 5 units for the
-	 * new inode (1 inode item, 1 inode ref, 2 dir items and 1 xattr item
-	 * when selinux is enabled).
+	 * 1 to remove old dir item
+	 * 1 to remove old dir index
+	 * 1 to add new dir item
+	 * 1 to add new dir index
 	 */
-	trans_num_items = 11;
-	if (flags & RENAME_WHITEOUT)
+	trans_num_items += 4;
+	/* 1 to update new parent inode if it's not the same as the old parent */
+	if (new_dir != old_dir)
+		trans_num_items++;
+	if (new_inode) {
+		/*
+		 * 1 to update inode
+		 * 1 to remove inode ref
+		 * 1 to remove dir item
+		 * 1 to remove dir index
+		 * 1 to possibly add orphan item
+		 */
 		trans_num_items += 5;
+	}
 	trans = btrfs_start_transaction(root, trans_num_items);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
@@ -9479,12 +9519,14 @@ static int btrfs_rename(struct user_namespace *mnt_userns,
 				   rename_ctx.index, new_dentry->d_parent);
 
 	if (flags & RENAME_WHITEOUT) {
-		ret = btrfs_whiteout_for_rename(trans, root, mnt_userns,
-						old_dir, old_dentry);
-
+		ret = btrfs_create_new_inode(trans, &whiteout_args);
 		if (ret) {
 			btrfs_abort_transaction(trans, ret);
 			goto out_fail;
+		} else {
+			unlock_new_inode(whiteout_args.inode);
+			iput(whiteout_args.inode);
+			whiteout_args.inode = NULL;
 		}
 	}
 out_fail:
@@ -9493,7 +9535,11 @@ out_fail:
 out_notrans:
 	if (old_ino == BTRFS_FIRST_FREE_OBJECTID)
 		up_read(&fs_info->subvol_sem);
-
+	if (flags & RENAME_WHITEOUT)
+		btrfs_new_inode_args_destroy(&whiteout_args);
+out_whiteout_inode:
+	if (flags & RENAME_WHITEOUT)
+		iput(whiteout_args.inode);
 	return ret;
 }
 
@@ -9712,10 +9758,13 @@ static int btrfs_symlink(struct user_namespace *mnt_userns, struct inode *dir,
 	struct btrfs_root *root = BTRFS_I(dir)->root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
-	struct inode *inode = NULL;
+	struct inode *inode;
+	struct btrfs_new_inode_args new_inode_args = {
+		.dir = dir,
+		.dentry = dentry,
+	};
+	unsigned int trans_num_items;
 	int err;
-	u64 objectid;
-	u64 index = 0;
 	int name_len;
 	int datasize;
 	unsigned long ptr;
@@ -9726,49 +9775,40 @@ static int btrfs_symlink(struct user_namespace *mnt_userns, struct inode *dir,
 	if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(fs_info))
 		return -ENAMETOOLONG;
 
-	/*
-	 * 2 items for inode item and ref
-	 * 2 items for dir items
-	 * 1 item for updating parent inode item
-	 * 1 item for the inline extent item
-	 * 1 item for xattr if selinux is on
-	 */
-	trans = btrfs_start_transaction(root, 7);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		return -ENOMEM;
+	inode_init_owner(mnt_userns, inode, dir, S_IFLNK | S_IRWXUGO);
+	inode->i_op = &btrfs_symlink_inode_operations;
+	inode_nohighmem(inode);
+	inode->i_mapping->a_ops = &btrfs_aops;
+	btrfs_i_size_write(BTRFS_I(inode), name_len);
+	inode_set_bytes(inode, name_len);
 
-	err = btrfs_get_free_objectid(root, &objectid);
+	new_inode_args.inode = inode;
+	err = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items);
 	if (err)
-		goto out_unlock;
+		goto out_inode;
+	/* 1 additional item for the inline extent */
+	trans_num_items++;
 
-	inode = btrfs_new_inode(trans, root, mnt_userns, dir,
-				dentry->d_name.name, dentry->d_name.len,
-				btrfs_ino(BTRFS_I(dir)), objectid,
-				S_IFLNK | S_IRWXUGO, &index);
-	if (IS_ERR(inode)) {
-		err = PTR_ERR(inode);
-		inode = NULL;
-		goto out_unlock;
+	trans = btrfs_start_transaction(root, trans_num_items);
+	if (IS_ERR(trans)) {
+		err = PTR_ERR(trans);
+		goto out_new_inode_args;
 	}
 
-	/*
-	* If the active LSM wants to access the inode during
-	* d_instantiate it needs these. Smack checks to see
-	* if the filesystem supports xattrs by looking at the
-	* ops vector.
-	*/
-	inode->i_fop = &btrfs_file_operations;
-	inode->i_op = &btrfs_file_inode_operations;
-	inode->i_mapping->a_ops = &btrfs_aops;
-
-	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
+	err = btrfs_create_new_inode(trans, &new_inode_args);
 	if (err)
-		goto out_unlock;
+		goto out;
 
 	path = btrfs_alloc_path();
 	if (!path) {
 		err = -ENOMEM;
-		goto out_unlock;
+		btrfs_abort_transaction(trans, err);
+		discard_new_inode(inode);
+		inode = NULL;
+		goto out;
 	}
 	key.objectid = btrfs_ino(BTRFS_I(inode));
 	key.offset = 0;
@@ -9777,8 +9817,11 @@ static int btrfs_symlink(struct user_namespace *mnt_userns, struct inode *dir,
 	err = btrfs_insert_empty_item(trans, root, path, &key,
 				      datasize);
 	if (err) {
+		btrfs_abort_transaction(trans, err);
 		btrfs_free_path(path);
-		goto out_unlock;
+		discard_new_inode(inode);
+		inode = NULL;
+		goto out;
 	}
 	leaf = path->nodes[0];
 	ei = btrfs_item_ptr(leaf, path->slots[0],
@@ -9796,31 +9839,16 @@ static int btrfs_symlink(struct user_namespace *mnt_userns, struct inode *dir,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_free_path(path);
 
-	inode->i_op = &btrfs_symlink_inode_operations;
-	inode_nohighmem(inode);
-	inode_set_bytes(inode, name_len);
-	btrfs_i_size_write(BTRFS_I(inode), name_len);
-	err = btrfs_update_inode(trans, root, BTRFS_I(inode));
-	/*
-	 * Last step, add directory indexes for our symlink inode. This is the
-	 * last step to avoid extra cleanup of these indexes if an error happens
-	 * elsewhere above.
-	 */
-	if (!err)
-		err = btrfs_add_nondir(trans, BTRFS_I(dir), dentry,
-				BTRFS_I(inode), 0, index);
-	if (err)
-		goto out_unlock;
-
 	d_instantiate_new(dentry, inode);
-
-out_unlock:
+	err = 0;
+out:
 	btrfs_end_transaction(trans);
-	if (err && inode) {
-		inode_dec_link_count(inode);
-		discard_new_inode(inode);
-	}
 	btrfs_btree_balance_dirty(fs_info);
+out_new_inode_args:
+	btrfs_new_inode_args_destroy(&new_inode_args);
+out_inode:
+	if (err)
+		iput(inode);
 	return err;
 }
 
@@ -10071,62 +10099,58 @@ static int btrfs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
 	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(dir)->root;
-	struct inode *inode = NULL;
-	u64 objectid;
-	u64 index;
-	int ret = 0;
-
-	/*
-	 * 5 units required for adding orphan entry
-	 */
-	trans = btrfs_start_transaction(root, 5);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
-
-	ret = btrfs_get_free_objectid(root, &objectid);
-	if (ret)
-		goto out;
-
-	inode = btrfs_new_inode(trans, root, mnt_userns, dir, NULL, 0,
-			btrfs_ino(BTRFS_I(dir)), objectid, mode, &index);
-	if (IS_ERR(inode)) {
-		ret = PTR_ERR(inode);
-		inode = NULL;
-		goto out;
-	}
+	struct inode *inode;
+	struct btrfs_new_inode_args new_inode_args = {
+		.dir = dir,
+		.dentry = dentry,
+		.orphan = true,
+	};
+	unsigned int trans_num_items;
+	int ret;
 
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		return -ENOMEM;
+	inode_init_owner(mnt_userns, inode, dir, mode);
 	inode->i_fop = &btrfs_file_operations;
 	inode->i_op = &btrfs_file_inode_operations;
-
 	inode->i_mapping->a_ops = &btrfs_aops;
 
-	ret = btrfs_init_inode_security(trans, inode, dir, NULL);
+	new_inode_args.inode = inode;
+	ret = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items);
 	if (ret)
-		goto out;
+		goto out_inode;
 
-	ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
-	if (ret)
-		goto out;
-	ret = btrfs_orphan_add(trans, BTRFS_I(inode));
-	if (ret)
-		goto out;
+	trans = btrfs_start_transaction(root, trans_num_items);
+	if (IS_ERR(trans)) {
+		ret = PTR_ERR(trans);
+		goto out_new_inode_args;
+	}
+
+	ret = btrfs_create_new_inode(trans, &new_inode_args);
 
 	/*
-	 * We set number of links to 0 in btrfs_new_inode(), and here we set
-	 * it to 1 because d_tmpfile() will issue a warning if the count is 0,
-	 * through:
+	 * We set number of links to 0 in btrfs_create_new_inode(), and here we
+	 * set it to 1 because d_tmpfile() will issue a warning if the count is
+	 * 0, through:
 	 *
 	 *    d_tmpfile() -> inode_dec_link_count() -> drop_nlink()
 	 */
 	set_nlink(inode, 1);
-	d_tmpfile(dentry, inode);
-	unlock_new_inode(inode);
-	mark_inode_dirty(inode);
-out:
+
+	if (!ret) {
+		d_tmpfile(dentry, inode);
+		unlock_new_inode(inode);
+		mark_inode_dirty(inode);
+	}
+
 	btrfs_end_transaction(trans);
-	if (ret && inode)
-		discard_new_inode(inode);
 	btrfs_btree_balance_dirty(fs_info);
+out_new_inode_args:
+	btrfs_new_inode_args_destroy(&new_inode_args);
+out_inode:
+	if (ret)
+		iput(inode);
 	return ret;
 }
 
@@ -10458,13 +10482,11 @@ static ssize_t btrfs_encoded_read_regular(struct kiocb *iocb,
 	pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS);
 	if (!pages)
 		return -ENOMEM;
-	for (i = 0; i < nr_pages; i++) {
-		pages[i] = alloc_page(GFP_NOFS);
-		if (!pages[i]) {
-			ret = -ENOMEM;
-			goto out;
+	ret = btrfs_alloc_page_array(nr_pages, pages);
+	if (ret) {
+		ret = -ENOMEM;
+		goto out;
 		}
-	}
 
 	ret = btrfs_encoded_read_regular_fill_pages(inode, start, disk_bytenr,
 						    disk_io_size, pages);
@@ -10800,7 +10822,8 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from,
 	ret = btrfs_qgroup_reserve_data(inode, &data_reserved, start, num_bytes);
 	if (ret)
 		goto out_free_data_space;
-	ret = btrfs_delalloc_reserve_metadata(inode, num_bytes, disk_num_bytes);
+	ret = btrfs_delalloc_reserve_metadata(inode, num_bytes, disk_num_bytes,
+					      false);
 	if (ret)
 		goto out_qgroup_free_data;
 
@@ -11309,6 +11332,41 @@ void btrfs_update_inode_bytes(struct btrfs_inode *inode,
 	spin_unlock(&inode->lock);
 }
 
+/**
+ * Verify that there are no ordered extents for a given file range.
+ *
+ * @inode:   The target inode.
+ * @start:   Start offset of the file range, should be sector size aligned.
+ * @end:     End offset (inclusive) of the file range, its value +1 should be
+ *           sector size aligned.
+ *
+ * This should typically be used for cases where we locked an inode's VFS lock in
+ * exclusive mode, we have also locked the inode's i_mmap_lock in exclusive mode,
+ * we have flushed all delalloc in the range, we have waited for all ordered
+ * extents in the range to complete and finally we have locked the file range in
+ * the inode's io_tree.
+ */
+void btrfs_assert_inode_range_clean(struct btrfs_inode *inode, u64 start, u64 end)
+{
+	struct btrfs_root *root = inode->root;
+	struct btrfs_ordered_extent *ordered;
+
+	if (!IS_ENABLED(CONFIG_BTRFS_ASSERT))
+		return;
+
+	ordered = btrfs_lookup_first_ordered_range(inode, start, end + 1 - start);
+	if (ordered) {
+		btrfs_err(root->fs_info,
+"found unexpected ordered extent in file range [%llu, %llu] for inode %llu root %llu (ordered range [%llu, %llu])",
+			  start, end, btrfs_ino(inode), root->root_key.objectid,
+			  ordered->file_offset,
+			  ordered->file_offset + ordered->num_bytes - 1);
+		btrfs_put_ordered_extent(ordered);
+	}
+
+	ASSERT(ordered == NULL);
+}
+
 static const struct inode_operations btrfs_dir_inode_operations = {
 	.getattr	= btrfs_getattr,
 	.lookup		= btrfs_lookup,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index b2c692b2fd8d..43b6f23bbd89 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -540,9 +540,35 @@ int __pure btrfs_is_empty_uuid(u8 *uuid)
 	return 1;
 }
 
+/*
+ * Calculate the number of transaction items to reserve for creating a subvolume
+ * or snapshot, not including the inode, directory entries, or parent directory.
+ */
+static unsigned int create_subvol_num_items(struct btrfs_qgroup_inherit *inherit)
+{
+	/*
+	 * 1 to add root block
+	 * 1 to add root item
+	 * 1 to add root ref
+	 * 1 to add root backref
+	 * 1 to add UUID item
+	 * 1 to add qgroup info
+	 * 1 to add qgroup limit
+	 *
+	 * Ideally the last two would only be accounted if qgroups are enabled,
+	 * but that can change between now and the time we would insert them.
+	 */
+	unsigned int num_items = 7;
+
+	if (inherit) {
+		/* 2 to add qgroup relations for each inherited qgroup */
+		num_items += 2 * inherit->num_qgroups;
+	}
+	return num_items;
+}
+
 static noinline int create_subvol(struct user_namespace *mnt_userns,
 				  struct inode *dir, struct dentry *dentry,
-				  const char *name, int namelen,
 				  struct btrfs_qgroup_inherit *inherit)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
@@ -555,11 +581,15 @@ static noinline int create_subvol(struct user_namespace *mnt_userns,
 	struct btrfs_root *new_root;
 	struct btrfs_block_rsv block_rsv;
 	struct timespec64 cur_time = current_time(dir);
-	struct inode *inode;
+	struct btrfs_new_inode_args new_inode_args = {
+		.dir = dir,
+		.dentry = dentry,
+		.subvol = true,
+	};
+	unsigned int trans_num_items;
 	int ret;
-	dev_t anon_dev = 0;
+	dev_t anon_dev;
 	u64 objectid;
-	u64 index = 0;
 
 	root_item = kzalloc(sizeof(*root_item), GFP_KERNEL);
 	if (!root_item)
@@ -567,11 +597,7 @@ static noinline int create_subvol(struct user_namespace *mnt_userns,
 
 	ret = btrfs_get_free_objectid(fs_info->tree_root, &objectid);
 	if (ret)
-		goto fail_free;
-
-	ret = get_anon_bdev(&anon_dev);
-	if (ret < 0)
-		goto fail_free;
+		goto out_root_item;
 
 	/*
 	 * Don't create subvolume whose level is not zero. Or qgroup will be
@@ -579,36 +605,47 @@ static noinline int create_subvol(struct user_namespace *mnt_userns,
 	 */
 	if (btrfs_qgroup_level(objectid)) {
 		ret = -ENOSPC;
-		goto fail_free;
+		goto out_root_item;
 	}
 
+	ret = get_anon_bdev(&anon_dev);
+	if (ret < 0)
+		goto out_root_item;
+
+	new_inode_args.inode = btrfs_new_subvol_inode(mnt_userns, dir);
+	if (!new_inode_args.inode) {
+		ret = -ENOMEM;
+		goto out_anon_dev;
+	}
+	ret = btrfs_new_inode_prepare(&new_inode_args, &trans_num_items);
+	if (ret)
+		goto out_inode;
+	trans_num_items += create_subvol_num_items(inherit);
+
 	btrfs_init_block_rsv(&block_rsv, BTRFS_BLOCK_RSV_TEMP);
-	/*
-	 * The same as the snapshot creation, please see the comment
-	 * of create_snapshot().
-	 */
-	ret = btrfs_subvolume_reserve_metadata(root, &block_rsv, 8, false);
+	ret = btrfs_subvolume_reserve_metadata(root, &block_rsv,
+					       trans_num_items, false);
 	if (ret)
-		goto fail_free;
+		goto out_new_inode_args;
 
 	trans = btrfs_start_transaction(root, 0);
 	if (IS_ERR(trans)) {
 		ret = PTR_ERR(trans);
 		btrfs_subvolume_release_metadata(root, &block_rsv);
-		goto fail_free;
+		goto out_new_inode_args;
 	}
 	trans->block_rsv = &block_rsv;
 	trans->bytes_reserved = block_rsv.size;
 
 	ret = btrfs_qgroup_inherit(trans, 0, objectid, inherit);
 	if (ret)
-		goto fail;
+		goto out;
 
 	leaf = btrfs_alloc_tree_block(trans, root, 0, objectid, NULL, 0, 0, 0,
 				      BTRFS_NESTING_NORMAL);
 	if (IS_ERR(leaf)) {
 		ret = PTR_ERR(leaf);
-		goto fail;
+		goto out;
 	}
 
 	btrfs_mark_buffer_dirty(leaf);
@@ -663,75 +700,46 @@ static noinline int create_subvol(struct user_namespace *mnt_userns,
 		btrfs_tree_unlock(leaf);
 		btrfs_free_tree_block(trans, objectid, leaf, 0, 1);
 		free_extent_buffer(leaf);
-		goto fail;
+		goto out;
 	}
 
 	free_extent_buffer(leaf);
 	leaf = NULL;
 
-	key.offset = (u64)-1;
 	new_root = btrfs_get_new_fs_root(fs_info, objectid, anon_dev);
 	if (IS_ERR(new_root)) {
-		free_anon_bdev(anon_dev);
 		ret = PTR_ERR(new_root);
 		btrfs_abort_transaction(trans, ret);
-		goto fail;
+		goto out;
 	}
-	/* Freeing will be done in btrfs_put_root() of new_root */
+	/* anon_dev is owned by new_root now. */
 	anon_dev = 0;
+	BTRFS_I(new_inode_args.inode)->root = new_root;
+	/* ... and new_root is owned by new_inode_args.inode now. */
 
 	ret = btrfs_record_root_in_trans(trans, new_root);
 	if (ret) {
-		btrfs_put_root(new_root);
-		btrfs_abort_transaction(trans, ret);
-		goto fail;
-	}
-
-	ret = btrfs_create_subvol_root(trans, new_root, root, mnt_userns);
-	btrfs_put_root(new_root);
-	if (ret) {
-		/* We potentially lose an unused inode item here */
 		btrfs_abort_transaction(trans, ret);
-		goto fail;
-	}
-
-	/*
-	 * insert the directory item
-	 */
-	ret = btrfs_set_inode_index(BTRFS_I(dir), &index);
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		goto fail;
-	}
-
-	ret = btrfs_insert_dir_item(trans, name, namelen, BTRFS_I(dir), &key,
-				    BTRFS_FT_DIR, index);
-	if (ret) {
-		btrfs_abort_transaction(trans, ret);
-		goto fail;
+		goto out;
 	}
 
-	btrfs_i_size_write(BTRFS_I(dir), dir->i_size + namelen * 2);
-	ret = btrfs_update_inode(trans, root, BTRFS_I(dir));
+	ret = btrfs_uuid_tree_add(trans, root_item->uuid,
+				  BTRFS_UUID_KEY_SUBVOL, objectid);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
-		goto fail;
+		goto out;
 	}
 
-	ret = btrfs_add_root_ref(trans, objectid, root->root_key.objectid,
-				 btrfs_ino(BTRFS_I(dir)), index, name, namelen);
+	ret = btrfs_create_new_inode(trans, &new_inode_args);
 	if (ret) {
 		btrfs_abort_transaction(trans, ret);
-		goto fail;
+		goto out;
 	}
 
-	ret = btrfs_uuid_tree_add(trans, root_item->uuid,
-				  BTRFS_UUID_KEY_SUBVOL, objectid);
-	if (ret)
-		btrfs_abort_transaction(trans, ret);
+	d_instantiate_new(dentry, new_inode_args.inode);
+	new_inode_args.inode = NULL;
 
-fail:
-	kfree(root_item);
+out:
 	trans->block_rsv = NULL;
 	trans->bytes_reserved = 0;
 	btrfs_subvolume_release_metadata(root, &block_rsv);
@@ -740,18 +748,14 @@ fail:
 		btrfs_end_transaction(trans);
 	else
 		ret = btrfs_commit_transaction(trans);
-
-	if (!ret) {
-		inode = btrfs_lookup_dentry(dir, dentry);
-		if (IS_ERR(inode))
-			return PTR_ERR(inode);
-		d_instantiate(dentry, inode);
-	}
-	return ret;
-
-fail_free:
+out_new_inode_args:
+	btrfs_new_inode_args_destroy(&new_inode_args);
+out_inode:
+	iput(new_inode_args.inode);
+out_anon_dev:
 	if (anon_dev)
 		free_anon_bdev(anon_dev);
+out_root_item:
 	kfree(root_item);
 	return ret;
 }
@@ -763,6 +767,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb);
 	struct inode *inode;
 	struct btrfs_pending_snapshot *pending_snapshot;
+	unsigned int trans_num_items;
 	struct btrfs_trans_handle *trans;
 	int ret;
 
@@ -800,16 +805,14 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir,
 	btrfs_init_block_rsv(&pending_snapshot->block_rsv,
 			     BTRFS_BLOCK_RSV_TEMP);
 	/*
-	 * 1 - parent dir inode
-	 * 2 - dir entries
-	 * 1 - root item
-	 * 2 - root ref/backref
-	 * 1 - root of snapshot
-	 * 1 - UUID item
+	 * 1 to add dir item
+	 * 1 to add dir index
+	 * 1 to update parent inode item
 	 */
+	trans_num_items = create_subvol_num_items(inherit) + 3;
 	ret = btrfs_subvolume_reserve_metadata(BTRFS_I(dir)->root,
-					&pending_snapshot->block_rsv, 8,
-					false);
+					       &pending_snapshot->block_rsv,
+					       trans_num_items, false);
 	if (ret)
 		goto free_pending;
 
@@ -979,7 +982,7 @@ static noinline int btrfs_mksubvol(const struct path *parent,
 	if (snap_src)
 		error = create_snapshot(snap_src, dir, dentry, readonly, inherit);
 	else
-		error = create_subvol(mnt_userns, dir, dentry, name, namelen, inherit);
+		error = create_subvol(mnt_userns, dir, dentry, inherit);
 
 	if (!error)
 		fsnotify_mkdir(dir, dentry);
@@ -1413,8 +1416,19 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 		if (!em)
 			break;
 
-		/* Skip hole/inline/preallocated extents */
-		if (em->block_start >= EXTENT_MAP_LAST_BYTE ||
+		/*
+		 * If the file extent is an inlined one, we may still want to
+		 * defrag it (fallthrough) if it will cause a regular extent.
+		 * This is for users who want to convert inline extents to
+		 * regular ones through max_inline= mount option.
+		 */
+		if (em->block_start == EXTENT_MAP_INLINE &&
+		    em->len <= inode->root->fs_info->max_inline)
+			goto next;
+
+		/* Skip hole/delalloc/preallocated extents */
+		if (em->block_start == EXTENT_MAP_HOLE ||
+		    em->block_start == EXTENT_MAP_DELALLOC ||
 		    test_bit(EXTENT_FLAG_PREALLOC, &em->flags))
 			goto next;
 
@@ -1473,6 +1487,15 @@ static int defrag_collect_targets(struct btrfs_inode *inode,
 		if (em->len >= get_extent_max_capacity(em))
 			goto next;
 
+		/*
+		 * Normally there are no more extents after an inline one, thus
+		 * @next_mergeable will normally be false and not defragged.
+		 * So if an inline extent passed all above checks, just add it
+		 * for defrag, and be converted to regular extents.
+		 */
+		if (em->block_start == EXTENT_MAP_INLINE)
+			goto add;
+
 		next_mergeable = defrag_check_next_extent(&inode->vfs_inode, em,
 						extent_thresh, newer_than, locked);
 		if (!next_mergeable) {
@@ -2594,7 +2617,7 @@ err:
 static noinline int btrfs_ioctl_tree_search(struct inode *inode,
 					    void __user *argp)
 {
-	struct btrfs_ioctl_search_args __user *uargs;
+	struct btrfs_ioctl_search_args __user *uargs = argp;
 	struct btrfs_ioctl_search_key sk;
 	int ret;
 	size_t buf_size;
@@ -2602,8 +2625,6 @@ static noinline int btrfs_ioctl_tree_search(struct inode *inode,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	uargs = (struct btrfs_ioctl_search_args __user *)argp;
-
 	if (copy_from_user(&sk, &uargs->key, sizeof(sk)))
 		return -EFAULT;
 
@@ -2626,7 +2647,7 @@ static noinline int btrfs_ioctl_tree_search(struct inode *inode,
 static noinline int btrfs_ioctl_tree_search_v2(struct inode *inode,
 					       void __user *argp)
 {
-	struct btrfs_ioctl_search_args_v2 __user *uarg;
+	struct btrfs_ioctl_search_args_v2 __user *uarg = argp;
 	struct btrfs_ioctl_search_args_v2 args;
 	int ret;
 	size_t buf_size;
@@ -2636,7 +2657,6 @@ static noinline int btrfs_ioctl_tree_search_v2(struct inode *inode,
 		return -EPERM;
 
 	/* copy search header and buffer size */
-	uarg = (struct btrfs_ioctl_search_args_v2 __user *)argp;
 	if (copy_from_user(&args, uarg, sizeof(args)))
 		return -EFAULT;
 
@@ -4344,10 +4364,6 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg)
 	bool need_unlock; /* for mut. excl. ops lock */
 	int ret;
 
-	if (!arg)
-		btrfs_warn(fs_info,
-	"IOC_BALANCE ioctl (v1) is deprecated and will be removed in kernel 5.18");
-
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
@@ -4355,6 +4371,13 @@ static long btrfs_ioctl_balance(struct file *file, void __user *arg)
 	if (ret)
 		return ret;
 
+	bargs = memdup_user(arg, sizeof(*bargs));
+	if (IS_ERR(bargs)) {
+		ret = PTR_ERR(bargs);
+		bargs = NULL;
+		goto out;
+	}
+
 again:
 	if (btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE)) {
 		mutex_lock(&fs_info->balance_mutex);
@@ -4402,59 +4425,42 @@ again:
 	}
 
 locked:
-
-	if (arg) {
-		bargs = memdup_user(arg, sizeof(*bargs));
-		if (IS_ERR(bargs)) {
-			ret = PTR_ERR(bargs);
+	if (bargs->flags & BTRFS_BALANCE_RESUME) {
+		if (!fs_info->balance_ctl) {
+			ret = -ENOTCONN;
 			goto out_unlock;
 		}
 
-		if (bargs->flags & BTRFS_BALANCE_RESUME) {
-			if (!fs_info->balance_ctl) {
-				ret = -ENOTCONN;
-				goto out_bargs;
-			}
+		bctl = fs_info->balance_ctl;
+		spin_lock(&fs_info->balance_lock);
+		bctl->flags |= BTRFS_BALANCE_RESUME;
+		spin_unlock(&fs_info->balance_lock);
+		btrfs_exclop_balance(fs_info, BTRFS_EXCLOP_BALANCE);
 
-			bctl = fs_info->balance_ctl;
-			spin_lock(&fs_info->balance_lock);
-			bctl->flags |= BTRFS_BALANCE_RESUME;
-			spin_unlock(&fs_info->balance_lock);
-			btrfs_exclop_balance(fs_info, BTRFS_EXCLOP_BALANCE);
+		goto do_balance;
+	}
 
-			goto do_balance;
-		}
-	} else {
-		bargs = NULL;
+	if (bargs->flags & ~(BTRFS_BALANCE_ARGS_MASK | BTRFS_BALANCE_TYPE_MASK)) {
+		ret = -EINVAL;
+		goto out_unlock;
 	}
 
 	if (fs_info->balance_ctl) {
 		ret = -EINPROGRESS;
-		goto out_bargs;
+		goto out_unlock;
 	}
 
 	bctl = kzalloc(sizeof(*bctl), GFP_KERNEL);
 	if (!bctl) {
 		ret = -ENOMEM;
-		goto out_bargs;
-	}
-
-	if (arg) {
-		memcpy(&bctl->data, &bargs->data, sizeof(bctl->data));
-		memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta));
-		memcpy(&bctl->sys, &bargs->sys, sizeof(bctl->sys));
-
-		bctl->flags = bargs->flags;
-	} else {
-		/* balance everything - no filters */
-		bctl->flags |= BTRFS_BALANCE_TYPE_MASK;
+		goto out_unlock;
 	}
 
-	if (bctl->flags & ~(BTRFS_BALANCE_ARGS_MASK | BTRFS_BALANCE_TYPE_MASK)) {
-		ret = -EINVAL;
-		goto out_bctl;
-	}
+	memcpy(&bctl->data, &bargs->data, sizeof(bctl->data));
+	memcpy(&bctl->meta, &bargs->meta, sizeof(bctl->meta));
+	memcpy(&bctl->sys, &bargs->sys, sizeof(bctl->sys));
 
+	bctl->flags = bargs->flags;
 do_balance:
 	/*
 	 * Ownership of bctl and exclusive operation goes to btrfs_balance.
@@ -4467,21 +4473,19 @@ do_balance:
 	ret = btrfs_balance(fs_info, bctl, bargs);
 	bctl = NULL;
 
-	if ((ret == 0 || ret == -ECANCELED) && arg) {
+	if (ret == 0 || ret == -ECANCELED) {
 		if (copy_to_user(arg, bargs, sizeof(*bargs)))
 			ret = -EFAULT;
 	}
 
-out_bctl:
 	kfree(bctl);
-out_bargs:
-	kfree(bargs);
 out_unlock:
 	mutex_unlock(&fs_info->balance_mutex);
 	if (need_unlock)
 		btrfs_exclop_finish(fs_info);
 out:
 	mnt_drop_write_file(file);
+	kfree(bargs);
 	return ret;
 }
 
diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c
index 1b31481f9e72..a2ec8ecae8de 100644
--- a/fs/btrfs/props.c
+++ b/fs/btrfs/props.c
@@ -380,9 +380,8 @@ static struct prop_handler prop_handlers[] = {
 	},
 };
 
-static int inherit_props(struct btrfs_trans_handle *trans,
-			 struct inode *inode,
-			 struct inode *parent)
+int btrfs_inode_inherit_props(struct btrfs_trans_handle *trans,
+			      struct inode *inode, struct inode *parent)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
@@ -457,41 +456,6 @@ static int inherit_props(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-int btrfs_inode_inherit_props(struct btrfs_trans_handle *trans,
-			      struct inode *inode,
-			      struct inode *dir)
-{
-	if (!dir)
-		return 0;
-
-	return inherit_props(trans, inode, dir);
-}
-
-int btrfs_subvol_inherit_props(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root,
-			       struct btrfs_root *parent_root)
-{
-	struct super_block *sb = root->fs_info->sb;
-	struct inode *parent_inode, *child_inode;
-	int ret;
-
-	parent_inode = btrfs_iget(sb, BTRFS_FIRST_FREE_OBJECTID, parent_root);
-	if (IS_ERR(parent_inode))
-		return PTR_ERR(parent_inode);
-
-	child_inode = btrfs_iget(sb, BTRFS_FIRST_FREE_OBJECTID, root);
-	if (IS_ERR(child_inode)) {
-		iput(parent_inode);
-		return PTR_ERR(child_inode);
-	}
-
-	ret = inherit_props(trans, child_inode, parent_inode);
-	iput(child_inode);
-	iput(parent_inode);
-
-	return ret;
-}
-
 void __init btrfs_props_init(void)
 {
 	int i;
diff --git a/fs/btrfs/props.h b/fs/btrfs/props.h
index 59bea741cfcf..ca9dd3df129b 100644
--- a/fs/btrfs/props.h
+++ b/fs/btrfs/props.h
@@ -23,8 +23,4 @@ int btrfs_inode_inherit_props(struct btrfs_trans_handle *trans,
 			      struct inode *inode,
 			      struct inode *dir);
 
-int btrfs_subvol_inherit_props(struct btrfs_trans_handle *trans,
-			       struct btrfs_root *root,
-			       struct btrfs_root *parent_root);
-
 #endif
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index 1866b1f0da01..db723c0026bd 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -2290,7 +2290,7 @@ int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 		return 0;
 
 	if (!extent_buffer_uptodate(root_eb)) {
-		ret = btrfs_read_buffer(root_eb, root_gen, root_level, NULL);
+		ret = btrfs_read_extent_buffer(root_eb, root_gen, root_level, NULL);
 		if (ret)
 			goto out;
 	}
@@ -3939,12 +3939,13 @@ int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
 }
 
 int __btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
-				enum btrfs_qgroup_rsv_type type, bool enforce)
+				enum btrfs_qgroup_rsv_type type, bool enforce,
+				bool noflush)
 {
 	int ret;
 
 	ret = btrfs_qgroup_reserve_meta(root, num_bytes, type, enforce);
-	if (ret <= 0 && ret != -EDQUOT)
+	if ((ret <= 0 && ret != -EDQUOT) || noflush)
 		return ret;
 
 	ret = try_flush_qgroup(root);
diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index 880e9df0dac1..0c4dd2a9af96 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -364,19 +364,23 @@ int btrfs_qgroup_free_data(struct btrfs_inode *inode,
 int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
 			      enum btrfs_qgroup_rsv_type type, bool enforce);
 int __btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes,
-				enum btrfs_qgroup_rsv_type type, bool enforce);
+				enum btrfs_qgroup_rsv_type type, bool enforce,
+				bool noflush);
 /* Reserve metadata space for pertrans and prealloc type */
 static inline int btrfs_qgroup_reserve_meta_pertrans(struct btrfs_root *root,
 				int num_bytes, bool enforce)
 {
 	return __btrfs_qgroup_reserve_meta(root, num_bytes,
-			BTRFS_QGROUP_RSV_META_PERTRANS, enforce);
+					   BTRFS_QGROUP_RSV_META_PERTRANS,
+					   enforce, false);
 }
 static inline int btrfs_qgroup_reserve_meta_prealloc(struct btrfs_root *root,
-				int num_bytes, bool enforce)
+						     int num_bytes, bool enforce,
+						     bool noflush)
 {
 	return __btrfs_qgroup_reserve_meta(root, num_bytes,
-			BTRFS_QGROUP_RSV_META_PREALLOC, enforce);
+					   BTRFS_QGROUP_RSV_META_PREALLOC,
+					   enforce, noflush);
 }
 
 void __btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes,
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index 0e239a4c3b26..a5b623ee6fac 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -52,6 +52,17 @@ struct btrfs_stripe_hash_table {
 	struct btrfs_stripe_hash table[];
 };
 
+/*
+ * A bvec like structure to present a sector inside a page.
+ *
+ * Unlike bvec we don't need bvlen, as it's fixed to sectorsize.
+ */
+struct sector_ptr {
+	struct page *page;
+	unsigned int pgoff:24;
+	unsigned int uptodate:8;
+};
+
 enum btrfs_rbio_ops {
 	BTRFS_RBIO_WRITE,
 	BTRFS_RBIO_READ_REBUILD,
@@ -77,7 +88,7 @@ struct btrfs_raid_bio {
 	/*
 	 * for scheduling work in the helper threads
 	 */
-	struct btrfs_work work;
+	struct work_struct work;
 
 	/*
 	 * bio list and bio_list_lock are used
@@ -101,15 +112,6 @@ struct btrfs_raid_bio {
 	 */
 	unsigned long flags;
 
-	/* size of each individual stripe on disk */
-	int stripe_len;
-
-	/* number of data stripes (no p/q) */
-	int nr_data;
-
-	int real_stripes;
-
-	int stripe_npages;
 	/*
 	 * set if we're doing a parity rebuild
 	 * for a read from higher up, which is handled
@@ -118,18 +120,35 @@ struct btrfs_raid_bio {
 	 */
 	enum btrfs_rbio_ops operation;
 
-	/* first bad stripe */
-	int faila;
+	/* Size of each individual stripe on disk */
+	u32 stripe_len;
 
-	/* second bad stripe (for raid6 use) */
-	int failb;
+	/* How many pages there are for the full stripe including P/Q */
+	u16 nr_pages;
 
-	int scrubp;
-	/*
-	 * number of pages needed to represent the full
-	 * stripe
-	 */
-	int nr_pages;
+	/* How many sectors there are for the full stripe including P/Q */
+	u16 nr_sectors;
+
+	/* Number of data stripes (no p/q) */
+	u8 nr_data;
+
+	/* Numer of all stripes (including P/Q) */
+	u8 real_stripes;
+
+	/* How many pages there are for each stripe */
+	u8 stripe_npages;
+
+	/* How many sectors there are for each stripe */
+	u8 stripe_nsectors;
+
+	/* First bad stripe, -1 means no corruption */
+	s8 faila;
+
+	/* Second bad stripe (for RAID6 use) */
+	s8 failb;
+
+	/* Stripe number that we're scrubbing  */
+	u8 scrubp;
 
 	/*
 	 * size of all the bios in the bio_list.  This
@@ -156,28 +175,29 @@ struct btrfs_raid_bio {
 	 */
 	struct page **stripe_pages;
 
-	/*
-	 * pointers to the pages in the bio_list.  Stored
-	 * here for faster lookup
-	 */
-	struct page **bio_pages;
+	/* Pointers to the sectors in the bio_list, for faster lookup */
+	struct sector_ptr *bio_sectors;
 
 	/*
-	 * bitmap to record which horizontal stripe has data
+	 * For subpage support, we need to map each sector to above
+	 * stripe_pages.
 	 */
+	struct sector_ptr *stripe_sectors;
+
+	/* Bitmap to record which horizontal stripe has data */
 	unsigned long *dbitmap;
 
 	/* allocated with real_stripes-many pointers for finish_*() calls */
 	void **finish_pointers;
 
-	/* allocated with stripe_npages-many bits for finish_*() calls */
+	/* Allocated with stripe_nsectors-many bits for finish_*() calls */
 	unsigned long *finish_pbitmap;
 };
 
 static int __raid56_parity_recover(struct btrfs_raid_bio *rbio);
 static noinline void finish_rmw(struct btrfs_raid_bio *rbio);
-static void rmw_work(struct btrfs_work *work);
-static void read_rebuild_work(struct btrfs_work *work);
+static void rmw_work(struct work_struct *work);
+static void read_rebuild_work(struct work_struct *work);
 static int fail_bio_stripe(struct btrfs_raid_bio *rbio, struct bio *bio);
 static int fail_rbio_index(struct btrfs_raid_bio *rbio, int failed);
 static void __free_raid_bio(struct btrfs_raid_bio *rbio);
@@ -186,12 +206,12 @@ static int alloc_rbio_pages(struct btrfs_raid_bio *rbio);
 
 static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 					 int need_check);
-static void scrub_parity_work(struct btrfs_work *work);
+static void scrub_parity_work(struct work_struct *work);
 
-static void start_async_work(struct btrfs_raid_bio *rbio, btrfs_func_t work_func)
+static void start_async_work(struct btrfs_raid_bio *rbio, work_func_t work_func)
 {
-	btrfs_init_work(&rbio->work, work_func, NULL, NULL);
-	btrfs_queue_work(rbio->bioc->fs_info->rmw_workers, &rbio->work);
+	INIT_WORK(&rbio->work, work_func);
+	queue_work(rbio->bioc->fs_info->rmw_workers, &rbio->work);
 }
 
 /*
@@ -239,7 +259,7 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info)
 
 /*
  * caching an rbio means to copy anything from the
- * bio_pages array into the stripe_pages array.  We
+ * bio_sectors array into the stripe_pages array.  We
  * use the page uptodate bit in the stripe cache array
  * to indicate if it has valid data
  *
@@ -255,12 +275,18 @@ static void cache_rbio_pages(struct btrfs_raid_bio *rbio)
 	if (ret)
 		return;
 
-	for (i = 0; i < rbio->nr_pages; i++) {
-		if (!rbio->bio_pages[i])
+	for (i = 0; i < rbio->nr_sectors; i++) {
+		/* Some range not covered by bio (partial write), skip it */
+		if (!rbio->bio_sectors[i].page)
 			continue;
 
-		copy_highpage(rbio->stripe_pages[i], rbio->bio_pages[i]);
-		SetPageUptodate(rbio->stripe_pages[i]);
+		ASSERT(rbio->stripe_sectors[i].page);
+		memcpy_page(rbio->stripe_sectors[i].page,
+			    rbio->stripe_sectors[i].pgoff,
+			    rbio->bio_sectors[i].page,
+			    rbio->bio_sectors[i].pgoff,
+			    rbio->bioc->fs_info->sectorsize);
+		rbio->stripe_sectors[i].uptodate = 1;
 	}
 	set_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
 }
@@ -283,9 +309,50 @@ static int rbio_bucket(struct btrfs_raid_bio *rbio)
 	return hash_64(num >> 16, BTRFS_STRIPE_HASH_TABLE_BITS);
 }
 
+static bool full_page_sectors_uptodate(struct btrfs_raid_bio *rbio,
+				       unsigned int page_nr)
+{
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
+	const u32 sectors_per_page = PAGE_SIZE / sectorsize;
+	int i;
+
+	ASSERT(page_nr < rbio->nr_pages);
+
+	for (i = sectors_per_page * page_nr;
+	     i < sectors_per_page * page_nr + sectors_per_page;
+	     i++) {
+		if (!rbio->stripe_sectors[i].uptodate)
+			return false;
+	}
+	return true;
+}
+
 /*
- * stealing an rbio means taking all the uptodate pages from the stripe
- * array in the source rbio and putting them into the destination rbio
+ * Update the stripe_sectors[] array to use correct page and pgoff
+ *
+ * Should be called every time any page pointer in stripes_pages[] got modified.
+ */
+static void index_stripe_sectors(struct btrfs_raid_bio *rbio)
+{
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
+	u32 offset;
+	int i;
+
+	for (i = 0, offset = 0; i < rbio->nr_sectors; i++, offset += sectorsize) {
+		int page_index = offset >> PAGE_SHIFT;
+
+		ASSERT(page_index < rbio->nr_pages);
+		rbio->stripe_sectors[i].page = rbio->stripe_pages[page_index];
+		rbio->stripe_sectors[i].pgoff = offset_in_page(offset);
+	}
+}
+
+/*
+ * Stealing an rbio means taking all the uptodate pages from the stripe array
+ * in the source rbio and putting them into the destination rbio.
+ *
+ * This will also update the involved stripe_sectors[] which are referring to
+ * the old pages.
  */
 static void steal_rbio(struct btrfs_raid_bio *src, struct btrfs_raid_bio *dest)
 {
@@ -298,9 +365,8 @@ static void steal_rbio(struct btrfs_raid_bio *src, struct btrfs_raid_bio *dest)
 
 	for (i = 0; i < dest->nr_pages; i++) {
 		s = src->stripe_pages[i];
-		if (!s || !PageUptodate(s)) {
+		if (!s || !full_page_sectors_uptodate(src, i))
 			continue;
-		}
 
 		d = dest->stripe_pages[i];
 		if (d)
@@ -309,6 +375,8 @@ static void steal_rbio(struct btrfs_raid_bio *src, struct btrfs_raid_bio *dest)
 		dest->stripe_pages[i] = s;
 		src->stripe_pages[i] = NULL;
 	}
+	index_stripe_sectors(dest);
+	index_stripe_sectors(src);
 }
 
 /*
@@ -600,39 +668,39 @@ static int rbio_can_merge(struct btrfs_raid_bio *last,
 	return 1;
 }
 
-static int rbio_stripe_page_index(struct btrfs_raid_bio *rbio, int stripe,
-				  int index)
+static unsigned int rbio_stripe_sector_index(const struct btrfs_raid_bio *rbio,
+					     unsigned int stripe_nr,
+					     unsigned int sector_nr)
 {
-	return stripe * rbio->stripe_npages + index;
+	ASSERT(stripe_nr < rbio->real_stripes);
+	ASSERT(sector_nr < rbio->stripe_nsectors);
+
+	return stripe_nr * rbio->stripe_nsectors + sector_nr;
 }
 
-/*
- * these are just the pages from the rbio array, not from anything
- * the FS sent down to us
- */
-static struct page *rbio_stripe_page(struct btrfs_raid_bio *rbio, int stripe,
-				     int index)
+/* Return a sector from rbio->stripe_sectors, not from the bio list */
+static struct sector_ptr *rbio_stripe_sector(const struct btrfs_raid_bio *rbio,
+					     unsigned int stripe_nr,
+					     unsigned int sector_nr)
 {
-	return rbio->stripe_pages[rbio_stripe_page_index(rbio, stripe, index)];
+	return &rbio->stripe_sectors[rbio_stripe_sector_index(rbio, stripe_nr,
+							      sector_nr)];
 }
 
-/*
- * helper to index into the pstripe
- */
-static struct page *rbio_pstripe_page(struct btrfs_raid_bio *rbio, int index)
+/* Grab a sector inside P stripe */
+static struct sector_ptr *rbio_pstripe_sector(const struct btrfs_raid_bio *rbio,
+					      unsigned int sector_nr)
 {
-	return rbio_stripe_page(rbio, rbio->nr_data, index);
+	return rbio_stripe_sector(rbio, rbio->nr_data, sector_nr);
 }
 
-/*
- * helper to index into the qstripe, returns null
- * if there is no qstripe
- */
-static struct page *rbio_qstripe_page(struct btrfs_raid_bio *rbio, int index)
+/* Grab a sector inside Q stripe, return NULL if not RAID6 */
+static struct sector_ptr *rbio_qstripe_sector(const struct btrfs_raid_bio *rbio,
+					      unsigned int sector_nr)
 {
 	if (rbio->nr_data + 1 == rbio->real_stripes)
 		return NULL;
-	return rbio_stripe_page(rbio, rbio->nr_data + 1, index);
+	return rbio_stripe_sector(rbio, rbio->nr_data + 1, sector_nr);
 }
 
 /*
@@ -911,47 +979,43 @@ static void raid_write_end_io(struct bio *bio)
 	rbio_orig_end_io(rbio, err);
 }
 
-/*
- * the read/modify/write code wants to use the original bio for
- * any pages it included, and then use the rbio for everything
- * else.  This function decides if a given index (stripe number)
- * and page number in that stripe fall inside the original bio
- * or the rbio.
- *
- * if you set bio_list_only, you'll get a NULL back for any ranges
- * that are outside the bio_list
+/**
+ * Get a sector pointer specified by its @stripe_nr and @sector_nr
  *
- * This doesn't take any refs on anything, you get a bare page pointer
- * and the caller must bump refs as required.
+ * @rbio:               The raid bio
+ * @stripe_nr:          Stripe number, valid range [0, real_stripe)
+ * @sector_nr:		Sector number inside the stripe,
+ *			valid range [0, stripe_nsectors)
+ * @bio_list_only:      Whether to use sectors inside the bio list only.
  *
- * You must call index_rbio_pages once before you can trust
- * the answers from this function.
+ * The read/modify/write code wants to reuse the original bio page as much
+ * as possible, and only use stripe_sectors as fallback.
  */
-static struct page *page_in_rbio(struct btrfs_raid_bio *rbio,
-				 int index, int pagenr, int bio_list_only)
+static struct sector_ptr *sector_in_rbio(struct btrfs_raid_bio *rbio,
+					 int stripe_nr, int sector_nr,
+					 bool bio_list_only)
 {
-	int chunk_page;
-	struct page *p = NULL;
+	struct sector_ptr *sector;
+	int index;
+
+	ASSERT(stripe_nr >= 0 && stripe_nr < rbio->real_stripes);
+	ASSERT(sector_nr >= 0 && sector_nr < rbio->stripe_nsectors);
 
-	chunk_page = index * (rbio->stripe_len >> PAGE_SHIFT) + pagenr;
+	index = stripe_nr * rbio->stripe_nsectors + sector_nr;
+	ASSERT(index >= 0 && index < rbio->nr_sectors);
 
 	spin_lock_irq(&rbio->bio_list_lock);
-	p = rbio->bio_pages[chunk_page];
+	sector = &rbio->bio_sectors[index];
+	if (sector->page || bio_list_only) {
+		/* Don't return sector without a valid page pointer */
+		if (!sector->page)
+			sector = NULL;
+		spin_unlock_irq(&rbio->bio_list_lock);
+		return sector;
+	}
 	spin_unlock_irq(&rbio->bio_list_lock);
 
-	if (p || bio_list_only)
-		return p;
-
-	return rbio->stripe_pages[chunk_page];
-}
-
-/*
- * number of pages we need for the entire stripe across all the
- * drives
- */
-static unsigned long rbio_nr_pages(unsigned long stripe_len, int nr_stripes)
-{
-	return DIV_ROUND_UP(stripe_len, PAGE_SIZE) * nr_stripes;
+	return &rbio->stripe_sectors[index];
 }
 
 /*
@@ -960,22 +1024,28 @@ static unsigned long rbio_nr_pages(unsigned long stripe_len, int nr_stripes)
  */
 static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 					 struct btrfs_io_context *bioc,
-					 u64 stripe_len)
+					 u32 stripe_len)
 {
+	const unsigned int real_stripes = bioc->num_stripes - bioc->num_tgtdevs;
+	const unsigned int stripe_npages = stripe_len >> PAGE_SHIFT;
+	const unsigned int num_pages = stripe_npages * real_stripes;
+	const unsigned int stripe_nsectors = stripe_len >> fs_info->sectorsize_bits;
+	const unsigned int num_sectors = stripe_nsectors * real_stripes;
 	struct btrfs_raid_bio *rbio;
 	int nr_data = 0;
-	int real_stripes = bioc->num_stripes - bioc->num_tgtdevs;
-	int num_pages = rbio_nr_pages(stripe_len, real_stripes);
-	int stripe_npages = DIV_ROUND_UP(stripe_len, PAGE_SIZE);
 	void *p;
 
+	ASSERT(IS_ALIGNED(stripe_len, PAGE_SIZE));
+	/* PAGE_SIZE must also be aligned to sectorsize for subpage support */
+	ASSERT(IS_ALIGNED(PAGE_SIZE, fs_info->sectorsize));
+
 	rbio = kzalloc(sizeof(*rbio) +
 		       sizeof(*rbio->stripe_pages) * num_pages +
-		       sizeof(*rbio->bio_pages) * num_pages +
+		       sizeof(*rbio->bio_sectors) * num_sectors +
+		       sizeof(*rbio->stripe_sectors) * num_sectors +
 		       sizeof(*rbio->finish_pointers) * real_stripes +
-		       sizeof(*rbio->dbitmap) * BITS_TO_LONGS(stripe_npages) +
-		       sizeof(*rbio->finish_pbitmap) *
-				BITS_TO_LONGS(stripe_npages),
+		       sizeof(*rbio->dbitmap) * BITS_TO_LONGS(stripe_nsectors) +
+		       sizeof(*rbio->finish_pbitmap) * BITS_TO_LONGS(stripe_nsectors),
 		       GFP_NOFS);
 	if (!rbio)
 		return ERR_PTR(-ENOMEM);
@@ -988,8 +1058,10 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 	rbio->bioc = bioc;
 	rbio->stripe_len = stripe_len;
 	rbio->nr_pages = num_pages;
+	rbio->nr_sectors = num_sectors;
 	rbio->real_stripes = real_stripes;
 	rbio->stripe_npages = stripe_npages;
+	rbio->stripe_nsectors = stripe_nsectors;
 	rbio->faila = -1;
 	rbio->failb = -1;
 	refcount_set(&rbio->refs, 1);
@@ -997,8 +1069,8 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 	atomic_set(&rbio->stripes_pending, 0);
 
 	/*
-	 * the stripe_pages, bio_pages, etc arrays point to the extra
-	 * memory we allocated past the end of the rbio
+	 * The stripe_pages, bio_sectors, etc arrays point to the extra memory
+	 * we allocated past the end of the rbio.
 	 */
 	p = rbio + 1;
 #define CONSUME_ALLOC(ptr, count)	do {				\
@@ -1006,10 +1078,11 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 		p = (unsigned char *)p + sizeof(*(ptr)) * (count);	\
 	} while (0)
 	CONSUME_ALLOC(rbio->stripe_pages, num_pages);
-	CONSUME_ALLOC(rbio->bio_pages, num_pages);
+	CONSUME_ALLOC(rbio->bio_sectors, num_sectors);
+	CONSUME_ALLOC(rbio->stripe_sectors, num_sectors);
 	CONSUME_ALLOC(rbio->finish_pointers, real_stripes);
-	CONSUME_ALLOC(rbio->dbitmap, BITS_TO_LONGS(stripe_npages));
-	CONSUME_ALLOC(rbio->finish_pbitmap, BITS_TO_LONGS(stripe_npages));
+	CONSUME_ALLOC(rbio->dbitmap, BITS_TO_LONGS(stripe_nsectors));
+	CONSUME_ALLOC(rbio->finish_pbitmap, BITS_TO_LONGS(stripe_nsectors));
 #undef  CONSUME_ALLOC
 
 	if (bioc->map_type & BTRFS_BLOCK_GROUP_RAID5)
@@ -1026,59 +1099,63 @@ static struct btrfs_raid_bio *alloc_rbio(struct btrfs_fs_info *fs_info,
 /* allocate pages for all the stripes in the bio, including parity */
 static int alloc_rbio_pages(struct btrfs_raid_bio *rbio)
 {
-	int i;
-	struct page *page;
+	int ret;
 
-	for (i = 0; i < rbio->nr_pages; i++) {
-		if (rbio->stripe_pages[i])
-			continue;
-		page = alloc_page(GFP_NOFS);
-		if (!page)
-			return -ENOMEM;
-		rbio->stripe_pages[i] = page;
-	}
+	ret = btrfs_alloc_page_array(rbio->nr_pages, rbio->stripe_pages);
+	if (ret < 0)
+		return ret;
+	/* Mapping all sectors */
+	index_stripe_sectors(rbio);
 	return 0;
 }
 
 /* only allocate pages for p/q stripes */
 static int alloc_rbio_parity_pages(struct btrfs_raid_bio *rbio)
 {
-	int i;
-	struct page *page;
+	const int data_pages = rbio->nr_data * rbio->stripe_npages;
+	int ret;
 
-	i = rbio_stripe_page_index(rbio, rbio->nr_data, 0);
+	ret = btrfs_alloc_page_array(rbio->nr_pages - data_pages,
+				     rbio->stripe_pages + data_pages);
+	if (ret < 0)
+		return ret;
 
-	for (; i < rbio->nr_pages; i++) {
-		if (rbio->stripe_pages[i])
-			continue;
-		page = alloc_page(GFP_NOFS);
-		if (!page)
-			return -ENOMEM;
-		rbio->stripe_pages[i] = page;
-	}
+	index_stripe_sectors(rbio);
 	return 0;
 }
 
 /*
- * add a single page from a specific stripe into our list of bios for IO
- * this will try to merge into existing bios if possible, and returns
- * zero if all went well.
+ * Add a single sector @sector into our list of bios for IO.
+ *
+ * Return 0 if everything went well.
+ * Return <0 for error.
  */
-static int rbio_add_io_page(struct btrfs_raid_bio *rbio,
-			    struct bio_list *bio_list,
-			    struct page *page,
-			    int stripe_nr,
-			    unsigned long page_index,
-			    unsigned long bio_max_len)
-{
+static int rbio_add_io_sector(struct btrfs_raid_bio *rbio,
+			      struct bio_list *bio_list,
+			      struct sector_ptr *sector,
+			      unsigned int stripe_nr,
+			      unsigned int sector_nr,
+			      unsigned long bio_max_len,
+			      unsigned int opf)
+{
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
 	struct bio *last = bio_list->tail;
 	int ret;
 	struct bio *bio;
 	struct btrfs_io_stripe *stripe;
 	u64 disk_start;
 
+	/*
+	 * Note: here stripe_nr has taken device replace into consideration,
+	 * thus it can be larger than rbio->real_stripe.
+	 * So here we check against bioc->num_stripes, not rbio->real_stripes.
+	 */
+	ASSERT(stripe_nr >= 0 && stripe_nr < rbio->bioc->num_stripes);
+	ASSERT(sector_nr >= 0 && sector_nr < rbio->stripe_nsectors);
+	ASSERT(sector->page);
+
 	stripe = &rbio->bioc->stripes[stripe_nr];
-	disk_start = stripe->physical + (page_index << PAGE_SHIFT);
+	disk_start = stripe->physical + sector_nr * sectorsize;
 
 	/* if the device is missing, just fail this stripe */
 	if (!stripe->dev->bdev)
@@ -1095,20 +1172,20 @@ static int rbio_add_io_page(struct btrfs_raid_bio *rbio,
 		 */
 		if (last_end == disk_start && !last->bi_status &&
 		    last->bi_bdev == stripe->dev->bdev) {
-			ret = bio_add_page(last, page, PAGE_SIZE, 0);
-			if (ret == PAGE_SIZE)
+			ret = bio_add_page(last, sector->page, sectorsize,
+					   sector->pgoff);
+			if (ret == sectorsize)
 				return 0;
 		}
 	}
 
 	/* put a new bio on the list */
-	bio = btrfs_bio_alloc(bio_max_len >> PAGE_SHIFT ?: 1);
-	btrfs_bio(bio)->device = stripe->dev;
-	bio->bi_iter.bi_size = 0;
-	bio_set_dev(bio, stripe->dev->bdev);
+	bio = bio_alloc(stripe->dev->bdev, max(bio_max_len >> PAGE_SHIFT, 1UL),
+			opf, GFP_NOFS);
 	bio->bi_iter.bi_sector = disk_start >> 9;
+	bio->bi_private = rbio;
 
-	bio_add_page(bio, page, PAGE_SIZE, 0);
+	bio_add_page(bio, sector->page, sectorsize, sector->pgoff);
 	bio_list_add(bio_list, bio);
 	return 0;
 }
@@ -1130,6 +1207,32 @@ static void validate_rbio_for_rmw(struct btrfs_raid_bio *rbio)
 	}
 }
 
+static void index_one_bio(struct btrfs_raid_bio *rbio, struct bio *bio)
+{
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
+	struct bio_vec bvec;
+	struct bvec_iter iter;
+	u32 offset = (bio->bi_iter.bi_sector << SECTOR_SHIFT) -
+		     rbio->bioc->raid_map[0];
+
+	if (bio_flagged(bio, BIO_CLONED))
+		bio->bi_iter = btrfs_bio(bio)->iter;
+
+	bio_for_each_segment(bvec, bio, iter) {
+		u32 bvec_offset;
+
+		for (bvec_offset = 0; bvec_offset < bvec.bv_len;
+		     bvec_offset += sectorsize, offset += sectorsize) {
+			int index = offset / sectorsize;
+			struct sector_ptr *sector = &rbio->bio_sectors[index];
+
+			sector->page = bvec.bv_page;
+			sector->pgoff = bvec.bv_offset + bvec_offset;
+			ASSERT(sector->pgoff < PAGE_SIZE);
+		}
+	}
+}
+
 /*
  * helper function to walk our bio list and populate the bio_pages array with
  * the result.  This seems expensive, but it is faster than constantly
@@ -1141,28 +1244,11 @@ static void validate_rbio_for_rmw(struct btrfs_raid_bio *rbio)
 static void index_rbio_pages(struct btrfs_raid_bio *rbio)
 {
 	struct bio *bio;
-	u64 start;
-	unsigned long stripe_offset;
-	unsigned long page_index;
 
 	spin_lock_irq(&rbio->bio_list_lock);
-	bio_list_for_each(bio, &rbio->bio_list) {
-		struct bio_vec bvec;
-		struct bvec_iter iter;
-		int i = 0;
+	bio_list_for_each(bio, &rbio->bio_list)
+		index_one_bio(rbio, bio);
 
-		start = bio->bi_iter.bi_sector << 9;
-		stripe_offset = start - rbio->bioc->raid_map[0];
-		page_index = stripe_offset >> PAGE_SHIFT;
-
-		if (bio_flagged(bio, BIO_CLONED))
-			bio->bi_iter = btrfs_bio(bio)->iter;
-
-		bio_for_each_segment(bvec, bio, iter) {
-			rbio->bio_pages[page_index + i] = bvec.bv_page;
-			i++;
-		}
-	}
 	spin_unlock_irq(&rbio->bio_list_lock);
 }
 
@@ -1177,10 +1263,11 @@ static void index_rbio_pages(struct btrfs_raid_bio *rbio)
 static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 {
 	struct btrfs_io_context *bioc = rbio->bioc;
+	const u32 sectorsize = bioc->fs_info->sectorsize;
 	void **pointers = rbio->finish_pointers;
 	int nr_data = rbio->nr_data;
 	int stripe;
-	int pagenr;
+	int sectornr;
 	bool has_qstripe;
 	struct bio_list bio_list;
 	struct bio *bio;
@@ -1224,35 +1311,37 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 	else
 		clear_bit(RBIO_CACHE_READY_BIT, &rbio->flags);
 
-	for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
-		struct page *p;
-		/* first collect one page from each data stripe */
+	for (sectornr = 0; sectornr < rbio->stripe_nsectors; sectornr++) {
+		struct sector_ptr *sector;
+
+		/* First collect one sector from each data stripe */
 		for (stripe = 0; stripe < nr_data; stripe++) {
-			p = page_in_rbio(rbio, stripe, pagenr, 0);
-			pointers[stripe] = kmap_local_page(p);
+			sector = sector_in_rbio(rbio, stripe, sectornr, 0);
+			pointers[stripe] = kmap_local_page(sector->page) +
+					   sector->pgoff;
 		}
 
-		/* then add the parity stripe */
-		p = rbio_pstripe_page(rbio, pagenr);
-		SetPageUptodate(p);
-		pointers[stripe++] = kmap_local_page(p);
+		/* Then add the parity stripe */
+		sector = rbio_pstripe_sector(rbio, sectornr);
+		sector->uptodate = 1;
+		pointers[stripe++] = kmap_local_page(sector->page) + sector->pgoff;
 
 		if (has_qstripe) {
-
 			/*
-			 * raid6, add the qstripe and call the
-			 * library function to fill in our p/q
+			 * RAID6, add the qstripe and call the library function
+			 * to fill in our p/q
 			 */
-			p = rbio_qstripe_page(rbio, pagenr);
-			SetPageUptodate(p);
-			pointers[stripe++] = kmap_local_page(p);
+			sector = rbio_qstripe_sector(rbio, sectornr);
+			sector->uptodate = 1;
+			pointers[stripe++] = kmap_local_page(sector->page) +
+					     sector->pgoff;
 
-			raid6_call.gen_syndrome(rbio->real_stripes, PAGE_SIZE,
+			raid6_call.gen_syndrome(rbio->real_stripes, sectorsize,
 						pointers);
 		} else {
 			/* raid5 */
-			copy_page(pointers[nr_data], pointers[0]);
-			run_xor(pointers + 1, nr_data - 1, PAGE_SIZE);
+			memcpy(pointers[nr_data], pointers[0], sectorsize);
+			run_xor(pointers + 1, nr_data - 1, sectorsize);
 		}
 		for (stripe = stripe - 1; stripe >= 0; stripe--)
 			kunmap_local(pointers[stripe]);
@@ -1264,18 +1353,20 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 	 * everything else.
 	 */
 	for (stripe = 0; stripe < rbio->real_stripes; stripe++) {
-		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
-			struct page *page;
+		for (sectornr = 0; sectornr < rbio->stripe_nsectors; sectornr++) {
+			struct sector_ptr *sector;
+
 			if (stripe < rbio->nr_data) {
-				page = page_in_rbio(rbio, stripe, pagenr, 1);
-				if (!page)
+				sector = sector_in_rbio(rbio, stripe, sectornr, 1);
+				if (!sector)
 					continue;
 			} else {
-			       page = rbio_stripe_page(rbio, stripe, pagenr);
+				sector = rbio_stripe_sector(rbio, stripe, sectornr);
 			}
 
-			ret = rbio_add_io_page(rbio, &bio_list,
-				       page, stripe, pagenr, rbio->stripe_len);
+			ret = rbio_add_io_sector(rbio, &bio_list, sector, stripe,
+						 sectornr, rbio->stripe_len,
+						 REQ_OP_WRITE);
 			if (ret)
 				goto cleanup;
 		}
@@ -1288,19 +1379,21 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio)
 		if (!bioc->tgtdev_map[stripe])
 			continue;
 
-		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
-			struct page *page;
+		for (sectornr = 0; sectornr < rbio->stripe_nsectors; sectornr++) {
+			struct sector_ptr *sector;
+
 			if (stripe < rbio->nr_data) {
-				page = page_in_rbio(rbio, stripe, pagenr, 1);
-				if (!page)
+				sector = sector_in_rbio(rbio, stripe, sectornr, 1);
+				if (!sector)
 					continue;
 			} else {
-			       page = rbio_stripe_page(rbio, stripe, pagenr);
+				sector = rbio_stripe_sector(rbio, stripe, sectornr);
 			}
 
-			ret = rbio_add_io_page(rbio, &bio_list, page,
+			ret = rbio_add_io_sector(rbio, &bio_list, sector,
 					       rbio->bioc->tgtdev_map[stripe],
-					       pagenr, rbio->stripe_len);
+					       sectornr, rbio->stripe_len,
+					       REQ_OP_WRITE);
 			if (ret)
 				goto cleanup;
 		}
@@ -1311,9 +1404,7 @@ write_data:
 	BUG_ON(atomic_read(&rbio->stripes_pending) == 0);
 
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_private = rbio;
 		bio->bi_end_io = raid_write_end_io;
-		bio->bi_opf = REQ_OP_WRITE;
 
 		submit_bio(bio);
 	}
@@ -1417,18 +1508,48 @@ static int fail_bio_stripe(struct btrfs_raid_bio *rbio,
 }
 
 /*
+ * For subpage case, we can no longer set page Uptodate directly for
+ * stripe_pages[], thus we need to locate the sector.
+ */
+static struct sector_ptr *find_stripe_sector(struct btrfs_raid_bio *rbio,
+					     struct page *page,
+					     unsigned int pgoff)
+{
+	int i;
+
+	for (i = 0; i < rbio->nr_sectors; i++) {
+		struct sector_ptr *sector = &rbio->stripe_sectors[i];
+
+		if (sector->page == page && sector->pgoff == pgoff)
+			return sector;
+	}
+	return NULL;
+}
+
+/*
  * this sets each page in the bio uptodate.  It should only be used on private
  * rbio pages, nothing that comes in from the higher layers
  */
-static void set_bio_pages_uptodate(struct bio *bio)
+static void set_bio_pages_uptodate(struct btrfs_raid_bio *rbio, struct bio *bio)
 {
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
 	struct bio_vec *bvec;
 	struct bvec_iter_all iter_all;
 
 	ASSERT(!bio_flagged(bio, BIO_CLONED));
 
-	bio_for_each_segment_all(bvec, bio, iter_all)
-		SetPageUptodate(bvec->bv_page);
+	bio_for_each_segment_all(bvec, bio, iter_all) {
+		struct sector_ptr *sector;
+		int pgoff;
+
+		for (pgoff = bvec->bv_offset; pgoff - bvec->bv_offset < bvec->bv_len;
+		     pgoff += sectorsize) {
+			sector = find_stripe_sector(rbio, bvec->bv_page, pgoff);
+			ASSERT(sector);
+			if (sector)
+				sector->uptodate = 1;
+		}
+	}
 }
 
 /*
@@ -1446,7 +1567,7 @@ static void raid_rmw_end_io(struct bio *bio)
 	if (bio->bi_status)
 		fail_bio_stripe(rbio, bio);
 	else
-		set_bio_pages_uptodate(bio);
+		set_bio_pages_uptodate(rbio, bio);
 
 	bio_put(bio);
 
@@ -1478,7 +1599,7 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
 	int bios_to_read = 0;
 	struct bio_list bio_list;
 	int ret;
-	int pagenr;
+	int sectornr;
 	int stripe;
 	struct bio *bio;
 
@@ -1496,28 +1617,30 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
 	 * stripe
 	 */
 	for (stripe = 0; stripe < rbio->nr_data; stripe++) {
-		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
-			struct page *page;
+		for (sectornr = 0; sectornr < rbio->stripe_nsectors; sectornr++) {
+			struct sector_ptr *sector;
+
 			/*
-			 * we want to find all the pages missing from
-			 * the rbio and read them from the disk.  If
-			 * page_in_rbio finds a page in the bio list
-			 * we don't need to read it off the stripe.
+			 * We want to find all the sectors missing from the
+			 * rbio and read them from the disk.  If * sector_in_rbio()
+			 * finds a page in the bio list we don't need to read
+			 * it off the stripe.
 			 */
-			page = page_in_rbio(rbio, stripe, pagenr, 1);
-			if (page)
+			sector = sector_in_rbio(rbio, stripe, sectornr, 1);
+			if (sector)
 				continue;
 
-			page = rbio_stripe_page(rbio, stripe, pagenr);
+			sector = rbio_stripe_sector(rbio, stripe, sectornr);
 			/*
-			 * the bio cache may have handed us an uptodate
-			 * page.  If so, be happy and use it
+			 * The bio cache may have handed us an uptodate page.
+			 * If so, be happy and use it.
 			 */
-			if (PageUptodate(page))
+			if (sector->uptodate)
 				continue;
 
-			ret = rbio_add_io_page(rbio, &bio_list, page,
-				       stripe, pagenr, rbio->stripe_len);
+			ret = rbio_add_io_sector(rbio, &bio_list, sector,
+				       stripe, sectornr, rbio->stripe_len,
+				       REQ_OP_READ);
 			if (ret)
 				goto cleanup;
 		}
@@ -1540,9 +1663,7 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
 	 */
 	atomic_set(&rbio->stripes_pending, bios_to_read);
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_private = rbio;
 		bio->bi_end_io = raid_rmw_end_io;
-		bio->bi_opf = REQ_OP_READ;
 
 		btrfs_bio_wq_end_io(rbio->bioc->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
 
@@ -1624,7 +1745,7 @@ struct btrfs_plug_cb {
 	struct blk_plug_cb cb;
 	struct btrfs_fs_info *info;
 	struct list_head rbio_list;
-	struct btrfs_work work;
+	struct work_struct work;
 };
 
 /*
@@ -1692,7 +1813,7 @@ static void run_plug(struct btrfs_plug_cb *plug)
  * if the unplug comes from schedule, we have to push the
  * work off to a helper thread
  */
-static void unplug_work(struct btrfs_work *work)
+static void unplug_work(struct work_struct *work)
 {
 	struct btrfs_plug_cb *plug;
 	plug = container_of(work, struct btrfs_plug_cb, work);
@@ -1705,9 +1826,8 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
 	plug = container_of(cb, struct btrfs_plug_cb, cb);
 
 	if (from_schedule) {
-		btrfs_init_work(&plug->work, unplug_work, NULL, NULL);
-		btrfs_queue_work(plug->info->rmw_workers,
-				 &plug->work);
+		INIT_WORK(&plug->work, unplug_work);
+		queue_work(plug->info->rmw_workers, &plug->work);
 		return;
 	}
 	run_plug(plug);
@@ -1716,8 +1836,7 @@ static void btrfs_raid_unplug(struct blk_plug_cb *cb, bool from_schedule)
 /*
  * our main entry point for writes from the rest of the FS.
  */
-int raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc,
-			u64 stripe_len)
+int raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc, u32 stripe_len)
 {
 	struct btrfs_fs_info *fs_info = bioc->fs_info;
 	struct btrfs_raid_bio *rbio;
@@ -1772,14 +1891,18 @@ int raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc,
  */
 static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 {
-	int pagenr, stripe;
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
+	int sectornr, stripe;
 	void **pointers;
 	void **unmap_array;
 	int faila = -1, failb = -1;
-	struct page *page;
 	blk_status_t err;
 	int i;
 
+	/*
+	 * This array stores the pointer for each sector, thus it has the extra
+	 * pgoff value added from each sector
+	 */
 	pointers = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS);
 	if (!pointers) {
 		err = BLK_STS_RESOURCE;
@@ -1808,43 +1931,44 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 
 	index_rbio_pages(rbio);
 
-	for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
+	for (sectornr = 0; sectornr < rbio->stripe_nsectors; sectornr++) {
+		struct sector_ptr *sector;
+
 		/*
 		 * Now we just use bitmap to mark the horizontal stripes in
 		 * which we have data when doing parity scrub.
 		 */
 		if (rbio->operation == BTRFS_RBIO_PARITY_SCRUB &&
-		    !test_bit(pagenr, rbio->dbitmap))
+		    !test_bit(sectornr, rbio->dbitmap))
 			continue;
 
 		/*
-		 * Setup our array of pointers with pages from each stripe
+		 * Setup our array of pointers with sectors from each stripe
 		 *
 		 * NOTE: store a duplicate array of pointers to preserve the
 		 * pointer order
 		 */
 		for (stripe = 0; stripe < rbio->real_stripes; stripe++) {
 			/*
-			 * if we're rebuilding a read, we have to use
+			 * If we're rebuilding a read, we have to use
 			 * pages from the bio list
 			 */
 			if ((rbio->operation == BTRFS_RBIO_READ_REBUILD ||
 			     rbio->operation == BTRFS_RBIO_REBUILD_MISSING) &&
 			    (stripe == faila || stripe == failb)) {
-				page = page_in_rbio(rbio, stripe, pagenr, 0);
+				sector = sector_in_rbio(rbio, stripe, sectornr, 0);
 			} else {
-				page = rbio_stripe_page(rbio, stripe, pagenr);
+				sector = rbio_stripe_sector(rbio, stripe, sectornr);
 			}
-			pointers[stripe] = kmap_local_page(page);
+			ASSERT(sector->page);
+			pointers[stripe] = kmap_local_page(sector->page) +
+					   sector->pgoff;
 			unmap_array[stripe] = pointers[stripe];
 		}
 
-		/* all raid6 handling here */
+		/* All raid6 handling here */
 		if (rbio->bioc->map_type & BTRFS_BLOCK_GROUP_RAID6) {
-			/*
-			 * single failure, rebuild from parity raid5
-			 * style
-			 */
+			/* Single failure, rebuild from parity raid5 style */
 			if (failb < 0) {
 				if (faila == rbio->nr_data) {
 					/*
@@ -1887,10 +2011,10 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 
 			if (rbio->bioc->raid_map[failb] == RAID5_P_STRIPE) {
 				raid6_datap_recov(rbio->real_stripes,
-						  PAGE_SIZE, faila, pointers);
+						  sectorsize, faila, pointers);
 			} else {
 				raid6_2data_recov(rbio->real_stripes,
-						  PAGE_SIZE, faila, failb,
+						  sectorsize, faila, failb,
 						  pointers);
 			}
 		} else {
@@ -1900,7 +2024,7 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 			BUG_ON(failb != -1);
 pstripe:
 			/* Copy parity block into failed block to start with */
-			copy_page(pointers[faila], pointers[rbio->nr_data]);
+			memcpy(pointers[faila], pointers[rbio->nr_data], sectorsize);
 
 			/* rearrange the pointer array */
 			p = pointers[faila];
@@ -1909,7 +2033,7 @@ pstripe:
 			pointers[rbio->nr_data - 1] = p;
 
 			/* xor in the rest */
-			run_xor(pointers, rbio->nr_data - 1, PAGE_SIZE);
+			run_xor(pointers, rbio->nr_data - 1, sectorsize);
 		}
 		/* if we're doing this rebuild as part of an rmw, go through
 		 * and set all of our private rbio pages in the
@@ -1918,14 +2042,14 @@ pstripe:
 		 * other endio functions will fiddle the uptodate bits
 		 */
 		if (rbio->operation == BTRFS_RBIO_WRITE) {
-			for (i = 0;  i < rbio->stripe_npages; i++) {
+			for (i = 0;  i < rbio->stripe_nsectors; i++) {
 				if (faila != -1) {
-					page = rbio_stripe_page(rbio, faila, i);
-					SetPageUptodate(page);
+					sector = rbio_stripe_sector(rbio, faila, i);
+					sector->uptodate = 1;
 				}
 				if (failb != -1) {
-					page = rbio_stripe_page(rbio, failb, i);
-					SetPageUptodate(page);
+					sector = rbio_stripe_sector(rbio, failb, i);
+					sector->uptodate = 1;
 				}
 			}
 		}
@@ -1998,7 +2122,7 @@ static void raid_recover_end_io(struct bio *bio)
 	if (bio->bi_status)
 		fail_bio_stripe(rbio, bio);
 	else
-		set_bio_pages_uptodate(bio);
+		set_bio_pages_uptodate(rbio, bio);
 	bio_put(bio);
 
 	if (!atomic_dec_and_test(&rbio->stripes_pending))
@@ -2023,7 +2147,7 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
 	int bios_to_read = 0;
 	struct bio_list bio_list;
 	int ret;
-	int pagenr;
+	int sectornr;
 	int stripe;
 	struct bio *bio;
 
@@ -2046,20 +2170,20 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
 			continue;
 		}
 
-		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
-			struct page *p;
+		for (sectornr = 0; sectornr < rbio->stripe_nsectors; sectornr++) {
+			struct sector_ptr *sector;
 
 			/*
 			 * the rmw code may have already read this
 			 * page in
 			 */
-			p = rbio_stripe_page(rbio, stripe, pagenr);
-			if (PageUptodate(p))
+			sector = rbio_stripe_sector(rbio, stripe, sectornr);
+			if (sector->uptodate)
 				continue;
 
-			ret = rbio_add_io_page(rbio, &bio_list,
-				       rbio_stripe_page(rbio, stripe, pagenr),
-				       stripe, pagenr, rbio->stripe_len);
+			ret = rbio_add_io_sector(rbio, &bio_list, sector,
+						 stripe, sectornr, rbio->stripe_len,
+						 REQ_OP_READ);
 			if (ret < 0)
 				goto cleanup;
 		}
@@ -2086,9 +2210,7 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
 	 */
 	atomic_set(&rbio->stripes_pending, bios_to_read);
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_private = rbio;
 		bio->bi_end_io = raid_recover_end_io;
-		bio->bi_opf = REQ_OP_READ;
 
 		btrfs_bio_wq_end_io(rbio->bioc->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
 
@@ -2115,7 +2237,7 @@ cleanup:
  * of the drive.
  */
 int raid56_parity_recover(struct bio *bio, struct btrfs_io_context *bioc,
-			  u64 stripe_len, int mirror_num, int generic_io)
+			  u32 stripe_len, int mirror_num, int generic_io)
 {
 	struct btrfs_fs_info *fs_info = bioc->fs_info;
 	struct btrfs_raid_bio *rbio;
@@ -2193,7 +2315,7 @@ int raid56_parity_recover(struct bio *bio, struct btrfs_io_context *bioc,
 
 }
 
-static void rmw_work(struct btrfs_work *work)
+static void rmw_work(struct work_struct *work)
 {
 	struct btrfs_raid_bio *rbio;
 
@@ -2201,7 +2323,7 @@ static void rmw_work(struct btrfs_work *work)
 	raid56_rmw_stripe(rbio);
 }
 
-static void read_rebuild_work(struct btrfs_work *work)
+static void read_rebuild_work(struct work_struct *work)
 {
 	struct btrfs_raid_bio *rbio;
 
@@ -2221,7 +2343,7 @@ static void read_rebuild_work(struct btrfs_work *work)
 
 struct btrfs_raid_bio *raid56_parity_alloc_scrub_rbio(struct bio *bio,
 				struct btrfs_io_context *bioc,
-				u64 stripe_len, struct btrfs_device *scrub_dev,
+				u32 stripe_len, struct btrfs_device *scrub_dev,
 				unsigned long *dbitmap, int stripe_nsectors)
 {
 	struct btrfs_fs_info *fs_info = bioc->fs_info;
@@ -2252,9 +2374,6 @@ struct btrfs_raid_bio *raid56_parity_alloc_scrub_rbio(struct bio *bio,
 	}
 	ASSERT(i < rbio->real_stripes);
 
-	/* Now we just support the sectorsize equals to page size */
-	ASSERT(fs_info->sectorsize == PAGE_SIZE);
-	ASSERT(rbio->stripe_npages == stripe_nsectors);
 	bitmap_copy(rbio->dbitmap, dbitmap, stripe_nsectors);
 
 	/*
@@ -2268,17 +2387,19 @@ struct btrfs_raid_bio *raid56_parity_alloc_scrub_rbio(struct bio *bio,
 
 /* Used for both parity scrub and missing. */
 void raid56_add_scrub_pages(struct btrfs_raid_bio *rbio, struct page *page,
-			    u64 logical)
+			    unsigned int pgoff, u64 logical)
 {
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
 	int stripe_offset;
 	int index;
 
 	ASSERT(logical >= rbio->bioc->raid_map[0]);
-	ASSERT(logical + PAGE_SIZE <= rbio->bioc->raid_map[0] +
+	ASSERT(logical + sectorsize <= rbio->bioc->raid_map[0] +
 				rbio->stripe_len * rbio->nr_data);
 	stripe_offset = (int)(logical - rbio->bioc->raid_map[0]);
-	index = stripe_offset >> PAGE_SHIFT;
-	rbio->bio_pages[index] = page;
+	index = stripe_offset / sectorsize;
+	rbio->bio_sectors[index].page = page;
+	rbio->bio_sectors[index].pgoff = pgoff;
 }
 
 /*
@@ -2287,14 +2408,16 @@ void raid56_add_scrub_pages(struct btrfs_raid_bio *rbio, struct page *page,
  */
 static int alloc_rbio_essential_pages(struct btrfs_raid_bio *rbio)
 {
-	int i;
-	int bit;
-	int index;
-	struct page *page;
+	const u32 sectorsize = rbio->bioc->fs_info->sectorsize;
+	int stripe;
+	int sectornr;
+
+	for_each_set_bit(sectornr, rbio->dbitmap, rbio->stripe_nsectors) {
+		for (stripe = 0; stripe < rbio->real_stripes; stripe++) {
+			struct page *page;
+			int index = (stripe * rbio->stripe_nsectors + sectornr) *
+				    sectorsize >> PAGE_SHIFT;
 
-	for_each_set_bit(bit, rbio->dbitmap, rbio->stripe_npages) {
-		for (i = 0; i < rbio->real_stripes; i++) {
-			index = i * rbio->stripe_npages + bit;
 			if (rbio->stripe_pages[index])
 				continue;
 
@@ -2304,6 +2427,7 @@ static int alloc_rbio_essential_pages(struct btrfs_raid_bio *rbio)
 			rbio->stripe_pages[index] = page;
 		}
 	}
+	index_stripe_sectors(rbio);
 	return 0;
 }
 
@@ -2311,14 +2435,15 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 					 int need_check)
 {
 	struct btrfs_io_context *bioc = rbio->bioc;
+	const u32 sectorsize = bioc->fs_info->sectorsize;
 	void **pointers = rbio->finish_pointers;
 	unsigned long *pbitmap = rbio->finish_pbitmap;
 	int nr_data = rbio->nr_data;
 	int stripe;
-	int pagenr;
+	int sectornr;
 	bool has_qstripe;
-	struct page *p_page = NULL;
-	struct page *q_page = NULL;
+	struct sector_ptr p_sector = { 0 };
+	struct sector_ptr q_sector = { 0 };
 	struct bio_list bio_list;
 	struct bio *bio;
 	int is_replace = 0;
@@ -2335,7 +2460,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 
 	if (bioc->num_tgtdevs && bioc->tgtdev_map[rbio->scrubp]) {
 		is_replace = 1;
-		bitmap_copy(pbitmap, rbio->dbitmap, rbio->stripe_npages);
+		bitmap_copy(pbitmap, rbio->dbitmap, rbio->stripe_nsectors);
 	}
 
 	/*
@@ -2348,54 +2473,59 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 	if (!need_check)
 		goto writeback;
 
-	p_page = alloc_page(GFP_NOFS);
-	if (!p_page)
+	p_sector.page = alloc_page(GFP_NOFS);
+	if (!p_sector.page)
 		goto cleanup;
-	SetPageUptodate(p_page);
+	p_sector.pgoff = 0;
+	p_sector.uptodate = 1;
 
 	if (has_qstripe) {
 		/* RAID6, allocate and map temp space for the Q stripe */
-		q_page = alloc_page(GFP_NOFS);
-		if (!q_page) {
-			__free_page(p_page);
+		q_sector.page = alloc_page(GFP_NOFS);
+		if (!q_sector.page) {
+			__free_page(p_sector.page);
+			p_sector.page = NULL;
 			goto cleanup;
 		}
-		SetPageUptodate(q_page);
-		pointers[rbio->real_stripes - 1] = kmap_local_page(q_page);
+		q_sector.pgoff = 0;
+		q_sector.uptodate = 1;
+		pointers[rbio->real_stripes - 1] = kmap_local_page(q_sector.page);
 	}
 
 	atomic_set(&rbio->error, 0);
 
 	/* Map the parity stripe just once */
-	pointers[nr_data] = kmap_local_page(p_page);
+	pointers[nr_data] = kmap_local_page(p_sector.page);
 
-	for_each_set_bit(pagenr, rbio->dbitmap, rbio->stripe_npages) {
-		struct page *p;
+	for_each_set_bit(sectornr, rbio->dbitmap, rbio->stripe_nsectors) {
+		struct sector_ptr *sector;
 		void *parity;
+
 		/* first collect one page from each data stripe */
 		for (stripe = 0; stripe < nr_data; stripe++) {
-			p = page_in_rbio(rbio, stripe, pagenr, 0);
-			pointers[stripe] = kmap_local_page(p);
+			sector = sector_in_rbio(rbio, stripe, sectornr, 0);
+			pointers[stripe] = kmap_local_page(sector->page) +
+					   sector->pgoff;
 		}
 
 		if (has_qstripe) {
 			/* RAID6, call the library function to fill in our P/Q */
-			raid6_call.gen_syndrome(rbio->real_stripes, PAGE_SIZE,
+			raid6_call.gen_syndrome(rbio->real_stripes, sectorsize,
 						pointers);
 		} else {
 			/* raid5 */
-			copy_page(pointers[nr_data], pointers[0]);
-			run_xor(pointers + 1, nr_data - 1, PAGE_SIZE);
+			memcpy(pointers[nr_data], pointers[0], sectorsize);
+			run_xor(pointers + 1, nr_data - 1, sectorsize);
 		}
 
 		/* Check scrubbing parity and repair it */
-		p = rbio_stripe_page(rbio, rbio->scrubp, pagenr);
-		parity = kmap_local_page(p);
-		if (memcmp(parity, pointers[rbio->scrubp], PAGE_SIZE))
-			copy_page(parity, pointers[rbio->scrubp]);
+		sector = rbio_stripe_sector(rbio, rbio->scrubp, sectornr);
+		parity = kmap_local_page(sector->page) + sector->pgoff;
+		if (memcmp(parity, pointers[rbio->scrubp], sectorsize) != 0)
+			memcpy(parity, pointers[rbio->scrubp], sectorsize);
 		else
 			/* Parity is right, needn't writeback */
-			bitmap_clear(rbio->dbitmap, pagenr, 1);
+			bitmap_clear(rbio->dbitmap, sectornr, 1);
 		kunmap_local(parity);
 
 		for (stripe = nr_data - 1; stripe >= 0; stripe--)
@@ -2403,10 +2533,12 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 	}
 
 	kunmap_local(pointers[nr_data]);
-	__free_page(p_page);
-	if (q_page) {
+	__free_page(p_sector.page);
+	p_sector.page = NULL;
+	if (q_sector.page) {
 		kunmap_local(pointers[rbio->real_stripes - 1]);
-		__free_page(q_page);
+		__free_page(q_sector.page);
+		q_sector.page = NULL;
 	}
 
 writeback:
@@ -2415,12 +2547,12 @@ writeback:
 	 * higher layers (the bio_list in our rbio) and our p/q.  Ignore
 	 * everything else.
 	 */
-	for_each_set_bit(pagenr, rbio->dbitmap, rbio->stripe_npages) {
-		struct page *page;
+	for_each_set_bit(sectornr, rbio->dbitmap, rbio->stripe_nsectors) {
+		struct sector_ptr *sector;
 
-		page = rbio_stripe_page(rbio, rbio->scrubp, pagenr);
-		ret = rbio_add_io_page(rbio, &bio_list,
-			       page, rbio->scrubp, pagenr, rbio->stripe_len);
+		sector = rbio_stripe_sector(rbio, rbio->scrubp, sectornr);
+		ret = rbio_add_io_sector(rbio, &bio_list, sector, rbio->scrubp,
+					 sectornr, rbio->stripe_len, REQ_OP_WRITE);
 		if (ret)
 			goto cleanup;
 	}
@@ -2428,13 +2560,13 @@ writeback:
 	if (!is_replace)
 		goto submit_write;
 
-	for_each_set_bit(pagenr, pbitmap, rbio->stripe_npages) {
-		struct page *page;
+	for_each_set_bit(sectornr, pbitmap, rbio->stripe_nsectors) {
+		struct sector_ptr *sector;
 
-		page = rbio_stripe_page(rbio, rbio->scrubp, pagenr);
-		ret = rbio_add_io_page(rbio, &bio_list, page,
+		sector = rbio_stripe_sector(rbio, rbio->scrubp, sectornr);
+		ret = rbio_add_io_sector(rbio, &bio_list, sector,
 				       bioc->tgtdev_map[rbio->scrubp],
-				       pagenr, rbio->stripe_len);
+				       sectornr, rbio->stripe_len, REQ_OP_WRITE);
 		if (ret)
 			goto cleanup;
 	}
@@ -2450,9 +2582,7 @@ submit_write:
 	atomic_set(&rbio->stripes_pending, nr_data);
 
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_private = rbio;
 		bio->bi_end_io = raid_write_end_io;
-		bio->bi_opf = REQ_OP_WRITE;
 
 		submit_bio(bio);
 	}
@@ -2548,7 +2678,7 @@ static void raid56_parity_scrub_end_io(struct bio *bio)
 	if (bio->bi_status)
 		fail_bio_stripe(rbio, bio);
 	else
-		set_bio_pages_uptodate(bio);
+		set_bio_pages_uptodate(rbio, bio);
 
 	bio_put(bio);
 
@@ -2568,7 +2698,7 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio)
 	int bios_to_read = 0;
 	struct bio_list bio_list;
 	int ret;
-	int pagenr;
+	int sectornr;
 	int stripe;
 	struct bio *bio;
 
@@ -2584,28 +2714,29 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio)
 	 * stripe
 	 */
 	for (stripe = 0; stripe < rbio->real_stripes; stripe++) {
-		for_each_set_bit(pagenr, rbio->dbitmap, rbio->stripe_npages) {
-			struct page *page;
+		for_each_set_bit(sectornr , rbio->dbitmap, rbio->stripe_nsectors) {
+			struct sector_ptr *sector;
 			/*
-			 * we want to find all the pages missing from
-			 * the rbio and read them from the disk.  If
-			 * page_in_rbio finds a page in the bio list
-			 * we don't need to read it off the stripe.
+			 * We want to find all the sectors missing from the
+			 * rbio and read them from the disk.  If * sector_in_rbio()
+			 * finds a sector in the bio list we don't need to read
+			 * it off the stripe.
 			 */
-			page = page_in_rbio(rbio, stripe, pagenr, 1);
-			if (page)
+			sector = sector_in_rbio(rbio, stripe, sectornr, 1);
+			if (sector)
 				continue;
 
-			page = rbio_stripe_page(rbio, stripe, pagenr);
+			sector = rbio_stripe_sector(rbio, stripe, sectornr);
 			/*
-			 * the bio cache may have handed us an uptodate
-			 * page.  If so, be happy and use it
+			 * The bio cache may have handed us an uptodate sector.
+			 * If so, be happy and use it.
 			 */
-			if (PageUptodate(page))
+			if (sector->uptodate)
 				continue;
 
-			ret = rbio_add_io_page(rbio, &bio_list, page,
-				       stripe, pagenr, rbio->stripe_len);
+			ret = rbio_add_io_sector(rbio, &bio_list, sector,
+						 stripe, sectornr, rbio->stripe_len,
+						 REQ_OP_READ);
 			if (ret)
 				goto cleanup;
 		}
@@ -2628,9 +2759,7 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio)
 	 */
 	atomic_set(&rbio->stripes_pending, bios_to_read);
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_private = rbio;
 		bio->bi_end_io = raid56_parity_scrub_end_io;
-		bio->bi_opf = REQ_OP_READ;
 
 		btrfs_bio_wq_end_io(rbio->bioc->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
 
@@ -2651,7 +2780,7 @@ finish:
 	validate_rbio_for_parity_scrub(rbio);
 }
 
-static void scrub_parity_work(struct btrfs_work *work)
+static void scrub_parity_work(struct work_struct *work)
 {
 	struct btrfs_raid_bio *rbio;
 
diff --git a/fs/btrfs/raid56.h b/fs/btrfs/raid56.h
index 72c00fc284b5..aaad08aefd7d 100644
--- a/fs/btrfs/raid56.h
+++ b/fs/btrfs/raid56.h
@@ -31,15 +31,14 @@ struct btrfs_raid_bio;
 struct btrfs_device;
 
 int raid56_parity_recover(struct bio *bio, struct btrfs_io_context *bioc,
-			  u64 stripe_len, int mirror_num, int generic_io);
-int raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc,
-			u64 stripe_len);
+			  u32 stripe_len, int mirror_num, int generic_io);
+int raid56_parity_write(struct bio *bio, struct btrfs_io_context *bioc, u32 stripe_len);
 
 void raid56_add_scrub_pages(struct btrfs_raid_bio *rbio, struct page *page,
-			    u64 logical);
+			    unsigned int pgoff, u64 logical);
 
 struct btrfs_raid_bio *raid56_parity_alloc_scrub_rbio(struct bio *bio,
-				struct btrfs_io_context *bioc, u64 stripe_len,
+				struct btrfs_io_context *bioc, u32 stripe_len,
 				struct btrfs_device *scrub_dev,
 				unsigned long *dbitmap, int stripe_nsectors);
 void raid56_parity_submit_scrub_rbio(struct btrfs_raid_bio *rbio);
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
index 998e3f180d90..c39f8b3a5a4a 100644
--- a/fs/btrfs/reflink.c
+++ b/fs/btrfs/reflink.c
@@ -614,14 +614,23 @@ static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1,
 static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1,
 				     struct inode *inode2, u64 loff2, u64 len)
 {
+	u64 range1_end = loff1 + len - 1;
+	u64 range2_end = loff2 + len - 1;
+
 	if (inode1 < inode2) {
 		swap(inode1, inode2);
 		swap(loff1, loff2);
+		swap(range1_end, range2_end);
 	} else if (inode1 == inode2 && loff2 < loff1) {
 		swap(loff1, loff2);
+		swap(range1_end, range2_end);
 	}
-	lock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1);
-	lock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
+
+	lock_extent(&BTRFS_I(inode1)->io_tree, loff1, range1_end);
+	lock_extent(&BTRFS_I(inode2)->io_tree, loff2, range2_end);
+
+	btrfs_assert_inode_range_clean(BTRFS_I(inode1), loff1, range1_end);
+	btrfs_assert_inode_range_clean(BTRFS_I(inode2), loff2, range2_end);
 }
 
 static void btrfs_double_mmap_lock(struct inode *inode1, struct inode *inode2)
@@ -771,7 +780,6 @@ static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in,
 	struct inode *inode_in = file_inode(file_in);
 	struct inode *inode_out = file_inode(file_out);
 	u64 bs = BTRFS_I(inode_out)->root->fs_info->sb->s_blocksize;
-	bool same_inode = inode_out == inode_in;
 	u64 wb_len;
 	int ret;
 
@@ -810,15 +818,6 @@ static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in,
 		wb_len = ALIGN(*len, bs);
 
 	/*
-	 * Since we don't lock ranges, wait for ongoing lockless dio writes (as
-	 * any in progress could create its ordered extents after we wait for
-	 * existing ordered extents below).
-	 */
-	inode_dio_wait(inode_in);
-	if (!same_inode)
-		inode_dio_wait(inode_out);
-
-	/*
 	 * Workaround to make sure NOCOW buffered write reach disk as NOCOW.
 	 *
 	 * Btrfs' back references do not have a block level granularity, they
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index fdc2c4b411f0..edddd93d2118 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -362,7 +362,7 @@ struct btrfs_root *find_reloc_root(struct btrfs_fs_info *fs_info, u64 bytenr)
 	rb_node = rb_simple_search(&rc->reloc_root_tree.rb_root, bytenr);
 	if (rb_node) {
 		node = rb_entry(rb_node, struct mapping_node, rb_node);
-		root = (struct btrfs_root *)node->data;
+		root = node->data;
 	}
 	spin_unlock(&rc->reloc_root_tree.lock);
 	return btrfs_grab_root(root);
@@ -2997,7 +2997,8 @@ static int relocate_one_page(struct inode *inode, struct file_ra_state *ra,
 
 		/* Reserve metadata for this range */
 		ret = btrfs_delalloc_reserve_metadata(BTRFS_I(inode),
-						      clamped_len, clamped_len);
+						      clamped_len, clamped_len,
+						      false);
 		if (ret)
 			goto release_page;
 
@@ -3845,8 +3846,7 @@ out:
 	btrfs_end_transaction(trans);
 	btrfs_btree_balance_dirty(fs_info);
 	if (err) {
-		if (inode)
-			iput(inode);
+		iput(inode);
 		inode = ERR_PTR(err);
 	}
 	return inode;
@@ -3977,6 +3977,17 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
 	if (!bg)
 		return -ENOENT;
 
+	/*
+	 * Relocation of a data block group creates ordered extents.  Without
+	 * sb_start_write(), we can freeze the filesystem while unfinished
+	 * ordered extents are left. Such ordered extents can cause a deadlock
+	 * e.g. when syncfs() is waiting for their completion but they can't
+	 * finish because they block when joining a transaction, due to the
+	 * fact that the freeze locks are being held in write mode.
+	 */
+	if (bg->flags & BTRFS_BLOCK_GROUP_DATA)
+		ASSERT(sb_write_started(fs_info->sb));
+
 	if (btrfs_pinned_by_swapfile(fs_info, bg)) {
 		btrfs_put_block_group(bg);
 		return -ETXTBSY;
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index ca7426ef61c8..a64b26b16904 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -509,7 +509,8 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 		/* One for parent inode, two for dir entries */
 		qgroup_num_bytes = 3 * fs_info->nodesize;
 		ret = btrfs_qgroup_reserve_meta_prealloc(root,
-				qgroup_num_bytes, true);
+							 qgroup_num_bytes, true,
+							 false);
 		if (ret)
 			return ret;
 	}
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 8cd713d37ad2..e7b0323e6efd 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -45,14 +45,14 @@ struct scrub_ctx;
  * operations. The first one configures an upper limit for the number
  * of (dynamically allocated) pages that are added to a bio.
  */
-#define SCRUB_PAGES_PER_BIO	32	/* 128KiB per bio for x86 */
-#define SCRUB_BIOS_PER_SCTX	64	/* 8MiB per device in flight for x86 */
+#define SCRUB_SECTORS_PER_BIO	32	/* 128KiB per bio for 4KiB pages */
+#define SCRUB_BIOS_PER_SCTX	64	/* 8MiB per device in flight for 4KiB pages */
 
 /*
  * The following value times PAGE_SIZE needs to be large enough to match the
  * largest node/leaf/sector size that shall be supported.
  */
-#define SCRUB_MAX_PAGES_PER_BLOCK	(BTRFS_MAX_METADATA_BLOCKSIZE / SZ_4K)
+#define SCRUB_MAX_SECTORS_PER_BLOCK	(BTRFS_MAX_METADATA_BLOCKSIZE / SZ_4K)
 
 struct scrub_recover {
 	refcount_t		refs;
@@ -60,7 +60,7 @@ struct scrub_recover {
 	u64			map_length;
 };
 
-struct scrub_page {
+struct scrub_sector {
 	struct scrub_block	*sblock;
 	struct page		*page;
 	struct btrfs_device	*dev;
@@ -87,16 +87,16 @@ struct scrub_bio {
 	blk_status_t		status;
 	u64			logical;
 	u64			physical;
-	struct scrub_page	*pagev[SCRUB_PAGES_PER_BIO];
-	int			page_count;
+	struct scrub_sector	*sectors[SCRUB_SECTORS_PER_BIO];
+	int			sector_count;
 	int			next_free;
-	struct btrfs_work	work;
+	struct work_struct	work;
 };
 
 struct scrub_block {
-	struct scrub_page	*pagev[SCRUB_MAX_PAGES_PER_BLOCK];
-	int			page_count;
-	atomic_t		outstanding_pages;
+	struct scrub_sector	*sectors[SCRUB_MAX_SECTORS_PER_BLOCK];
+	int			sector_count;
+	atomic_t		outstanding_sectors;
 	refcount_t		refs; /* free mem on transition to zero */
 	struct scrub_ctx	*sctx;
 	struct scrub_parity	*sparity;
@@ -110,7 +110,7 @@ struct scrub_block {
 		/* It is for the data with checksum */
 		unsigned int	data_corrected:1;
 	};
-	struct btrfs_work	work;
+	struct work_struct	work;
 };
 
 /* Used for the chunks with parity stripe such RAID5/6 */
@@ -129,10 +129,10 @@ struct scrub_parity {
 
 	refcount_t		refs;
 
-	struct list_head	spages;
+	struct list_head	sectors_list;
 
 	/* Work of parity check and repair */
-	struct btrfs_work	work;
+	struct work_struct	work;
 
 	/* Mark the parity blocks which have data */
 	unsigned long		*dbitmap;
@@ -158,7 +158,7 @@ struct scrub_ctx {
 	struct list_head	csum_list;
 	atomic_t		cancel_req;
 	int			readonly;
-	int			pages_per_bio;
+	int			sectors_per_bio;
 
 	/* State of IO submission throttling affecting the associated device */
 	ktime_t			throttle_deadline;
@@ -212,43 +212,43 @@ static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
 static void scrub_recheck_block_checksum(struct scrub_block *sblock);
 static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
 					     struct scrub_block *sblock_good);
-static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
+static int scrub_repair_sector_from_good_copy(struct scrub_block *sblock_bad,
 					    struct scrub_block *sblock_good,
-					    int page_num, int force_write);
+					    int sector_num, int force_write);
 static void scrub_write_block_to_dev_replace(struct scrub_block *sblock);
-static int scrub_write_page_to_dev_replace(struct scrub_block *sblock,
-					   int page_num);
+static int scrub_write_sector_to_dev_replace(struct scrub_block *sblock,
+					     int sector_num);
 static int scrub_checksum_data(struct scrub_block *sblock);
 static int scrub_checksum_tree_block(struct scrub_block *sblock);
 static int scrub_checksum_super(struct scrub_block *sblock);
 static void scrub_block_put(struct scrub_block *sblock);
-static void scrub_page_get(struct scrub_page *spage);
-static void scrub_page_put(struct scrub_page *spage);
+static void scrub_sector_get(struct scrub_sector *sector);
+static void scrub_sector_put(struct scrub_sector *sector);
 static void scrub_parity_get(struct scrub_parity *sparity);
 static void scrub_parity_put(struct scrub_parity *sparity);
-static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u32 len,
-		       u64 physical, struct btrfs_device *dev, u64 flags,
-		       u64 gen, int mirror_num, u8 *csum,
-		       u64 physical_for_dev_replace);
+static int scrub_sectors(struct scrub_ctx *sctx, u64 logical, u32 len,
+			 u64 physical, struct btrfs_device *dev, u64 flags,
+			 u64 gen, int mirror_num, u8 *csum,
+			 u64 physical_for_dev_replace);
 static void scrub_bio_end_io(struct bio *bio);
-static void scrub_bio_end_io_worker(struct btrfs_work *work);
+static void scrub_bio_end_io_worker(struct work_struct *work);
 static void scrub_block_complete(struct scrub_block *sblock);
-static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
-			       u64 extent_logical, u32 extent_len,
-			       u64 *extent_physical,
-			       struct btrfs_device **extent_dev,
-			       int *extent_mirror_num);
-static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
-				    struct scrub_page *spage);
+static void scrub_find_good_copy(struct btrfs_fs_info *fs_info,
+				 u64 extent_logical, u32 extent_len,
+				 u64 *extent_physical,
+				 struct btrfs_device **extent_dev,
+				 int *extent_mirror_num);
+static int scrub_add_sector_to_wr_bio(struct scrub_ctx *sctx,
+				      struct scrub_sector *sector);
 static void scrub_wr_submit(struct scrub_ctx *sctx);
 static void scrub_wr_bio_end_io(struct bio *bio);
-static void scrub_wr_bio_end_io_worker(struct btrfs_work *work);
+static void scrub_wr_bio_end_io_worker(struct work_struct *work);
 static void scrub_put_ctx(struct scrub_ctx *sctx);
 
-static inline int scrub_is_page_on_raid56(struct scrub_page *spage)
+static inline int scrub_is_page_on_raid56(struct scrub_sector *sector)
 {
-	return spage->recover &&
-	       (spage->recover->bioc->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK);
+	return sector->recover &&
+	       (sector->recover->bioc->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK);
 }
 
 static void scrub_pending_bio_inc(struct scrub_ctx *sctx)
@@ -535,9 +535,9 @@ static noinline_for_stack void scrub_free_ctx(struct scrub_ctx *sctx)
 	if (sctx->curr != -1) {
 		struct scrub_bio *sbio = sctx->bios[sctx->curr];
 
-		for (i = 0; i < sbio->page_count; i++) {
-			WARN_ON(!sbio->pagev[i]->page);
-			scrub_block_put(sbio->pagev[i]->sblock);
+		for (i = 0; i < sbio->sector_count; i++) {
+			WARN_ON(!sbio->sectors[i]->page);
+			scrub_block_put(sbio->sectors[i]->sblock);
 		}
 		bio_put(sbio->bio);
 	}
@@ -572,7 +572,7 @@ static noinline_for_stack struct scrub_ctx *scrub_setup_ctx(
 		goto nomem;
 	refcount_set(&sctx->refs, 1);
 	sctx->is_dev_replace = is_dev_replace;
-	sctx->pages_per_bio = SCRUB_PAGES_PER_BIO;
+	sctx->sectors_per_bio = SCRUB_SECTORS_PER_BIO;
 	sctx->curr = -1;
 	sctx->fs_info = fs_info;
 	INIT_LIST_HEAD(&sctx->csum_list);
@@ -586,9 +586,8 @@ static noinline_for_stack struct scrub_ctx *scrub_setup_ctx(
 
 		sbio->index = i;
 		sbio->sctx = sctx;
-		sbio->page_count = 0;
-		btrfs_init_work(&sbio->work, scrub_bio_end_io_worker, NULL,
-				NULL);
+		sbio->sector_count = 0;
+		INIT_WORK(&sbio->work, scrub_bio_end_io_worker);
 
 		if (i != SCRUB_BIOS_PER_SCTX - 1)
 			sctx->bios[i]->next_free = i + 1;
@@ -728,16 +727,16 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock)
 	u8 ref_level = 0;
 	int ret;
 
-	WARN_ON(sblock->page_count < 1);
-	dev = sblock->pagev[0]->dev;
+	WARN_ON(sblock->sector_count < 1);
+	dev = sblock->sectors[0]->dev;
 	fs_info = sblock->sctx->fs_info;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return;
 
-	swarn.physical = sblock->pagev[0]->physical;
-	swarn.logical = sblock->pagev[0]->logical;
+	swarn.physical = sblock->sectors[0]->physical;
+	swarn.logical = sblock->sectors[0]->logical;
 	swarn.errstr = errstr;
 	swarn.dev = NULL;
 
@@ -798,8 +797,8 @@ static inline void scrub_put_recover(struct btrfs_fs_info *fs_info,
 
 /*
  * scrub_handle_errored_block gets called when either verification of the
- * pages failed or the bio failed to read, e.g. with EIO. In the latter
- * case, this function handles all pages in the bio, even though only one
+ * sectors failed or the bio failed to read, e.g. with EIO. In the latter
+ * case, this function handles all sectors in the bio, even though only one
  * may be bad.
  * The goal of this function is to repair the errored block by using the
  * contents of one of the mirrors.
@@ -817,16 +816,16 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 	struct scrub_block *sblock_bad;
 	int ret;
 	int mirror_index;
-	int page_num;
+	int sector_num;
 	int success;
 	bool full_stripe_locked;
 	unsigned int nofs_flag;
 	static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL,
 				      DEFAULT_RATELIMIT_BURST);
 
-	BUG_ON(sblock_to_check->page_count < 1);
+	BUG_ON(sblock_to_check->sector_count < 1);
 	fs_info = sctx->fs_info;
-	if (sblock_to_check->pagev[0]->flags & BTRFS_EXTENT_FLAG_SUPER) {
+	if (sblock_to_check->sectors[0]->flags & BTRFS_EXTENT_FLAG_SUPER) {
 		/*
 		 * if we find an error in a super block, we just report it.
 		 * They will get written with the next transaction commit
@@ -837,13 +836,13 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 		spin_unlock(&sctx->stat_lock);
 		return 0;
 	}
-	logical = sblock_to_check->pagev[0]->logical;
-	BUG_ON(sblock_to_check->pagev[0]->mirror_num < 1);
-	failed_mirror_index = sblock_to_check->pagev[0]->mirror_num - 1;
-	is_metadata = !(sblock_to_check->pagev[0]->flags &
+	logical = sblock_to_check->sectors[0]->logical;
+	BUG_ON(sblock_to_check->sectors[0]->mirror_num < 1);
+	failed_mirror_index = sblock_to_check->sectors[0]->mirror_num - 1;
+	is_metadata = !(sblock_to_check->sectors[0]->flags &
 			BTRFS_EXTENT_FLAG_DATA);
-	have_csum = sblock_to_check->pagev[0]->have_csum;
-	dev = sblock_to_check->pagev[0]->dev;
+	have_csum = sblock_to_check->sectors[0]->have_csum;
+	dev = sblock_to_check->sectors[0]->dev;
 
 	if (!sctx->is_dev_replace && btrfs_repair_one_zone(fs_info, logical))
 		return 0;
@@ -854,7 +853,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 	 * might be waiting the scrub task to pause (which needs to wait for all
 	 * the worker tasks to complete before pausing).
 	 * We do allocations in the workers through insert_full_stripe_lock()
-	 * and scrub_add_page_to_wr_bio(), which happens down the call chain of
+	 * and scrub_add_sector_to_wr_bio(), which happens down the call chain of
 	 * this function.
 	 */
 	nofs_flag = memalloc_nofs_save();
@@ -918,7 +917,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 		goto out;
 	}
 
-	/* setup the context, map the logical blocks and alloc the pages */
+	/* Setup the context, map the logical blocks and alloc the sectors */
 	ret = scrub_setup_recheck_block(sblock_to_check, sblocks_for_recheck);
 	if (ret) {
 		spin_lock(&sctx->stat_lock);
@@ -937,7 +936,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 	if (!sblock_bad->header_error && !sblock_bad->checksum_error &&
 	    sblock_bad->no_io_error_seen) {
 		/*
-		 * the error disappeared after reading page by page, or
+		 * The error disappeared after reading sector by sector, or
 		 * the area was part of a huge bio and other parts of the
 		 * bio caused I/O errors, or the block layer merged several
 		 * read requests into one and the error is caused by a
@@ -998,10 +997,10 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 	 * that is known to contain an error is rewritten. Afterwards
 	 * the block is known to be corrected.
 	 * If a mirror is found which is completely correct, and no
-	 * checksum is present, only those pages are rewritten that had
+	 * checksum is present, only those sectors are rewritten that had
 	 * an I/O error in the block to be repaired, since it cannot be
-	 * determined, which copy of the other pages is better (and it
-	 * could happen otherwise that a correct page would be
+	 * determined, which copy of the other sectors is better (and it
+	 * could happen otherwise that a correct sector would be
 	 * overwritten by a bad one).
 	 */
 	for (mirror_index = 0; ;mirror_index++) {
@@ -1011,25 +1010,25 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 			continue;
 
 		/* raid56's mirror can be more than BTRFS_MAX_MIRRORS */
-		if (!scrub_is_page_on_raid56(sblock_bad->pagev[0])) {
+		if (!scrub_is_page_on_raid56(sblock_bad->sectors[0])) {
 			if (mirror_index >= BTRFS_MAX_MIRRORS)
 				break;
-			if (!sblocks_for_recheck[mirror_index].page_count)
+			if (!sblocks_for_recheck[mirror_index].sector_count)
 				break;
 
 			sblock_other = sblocks_for_recheck + mirror_index;
 		} else {
-			struct scrub_recover *r = sblock_bad->pagev[0]->recover;
+			struct scrub_recover *r = sblock_bad->sectors[0]->recover;
 			int max_allowed = r->bioc->num_stripes - r->bioc->num_tgtdevs;
 
 			if (mirror_index >= max_allowed)
 				break;
-			if (!sblocks_for_recheck[1].page_count)
+			if (!sblocks_for_recheck[1].sector_count)
 				break;
 
 			ASSERT(failed_mirror_index == 0);
 			sblock_other = sblocks_for_recheck + 1;
-			sblock_other->pagev[0]->mirror_num = 1 + mirror_index;
+			sblock_other->sectors[0]->mirror_num = 1 + mirror_index;
 		}
 
 		/* build and submit the bios, check checksums */
@@ -1078,16 +1077,16 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 	 * area are unreadable.
 	 */
 	success = 1;
-	for (page_num = 0; page_num < sblock_bad->page_count;
-	     page_num++) {
-		struct scrub_page *spage_bad = sblock_bad->pagev[page_num];
+	for (sector_num = 0; sector_num < sblock_bad->sector_count;
+	     sector_num++) {
+		struct scrub_sector *sector_bad = sblock_bad->sectors[sector_num];
 		struct scrub_block *sblock_other = NULL;
 
-		/* skip no-io-error page in scrub */
-		if (!spage_bad->io_error && !sctx->is_dev_replace)
+		/* Skip no-io-error sectors in scrub */
+		if (!sector_bad->io_error && !sctx->is_dev_replace)
 			continue;
 
-		if (scrub_is_page_on_raid56(sblock_bad->pagev[0])) {
+		if (scrub_is_page_on_raid56(sblock_bad->sectors[0])) {
 			/*
 			 * In case of dev replace, if raid56 rebuild process
 			 * didn't work out correct data, then copy the content
@@ -1096,14 +1095,14 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 			 * sblock_for_recheck array to target device.
 			 */
 			sblock_other = NULL;
-		} else if (spage_bad->io_error) {
-			/* try to find no-io-error page in mirrors */
+		} else if (sector_bad->io_error) {
+			/* Try to find no-io-error sector in mirrors */
 			for (mirror_index = 0;
 			     mirror_index < BTRFS_MAX_MIRRORS &&
-			     sblocks_for_recheck[mirror_index].page_count > 0;
+			     sblocks_for_recheck[mirror_index].sector_count > 0;
 			     mirror_index++) {
 				if (!sblocks_for_recheck[mirror_index].
-				    pagev[page_num]->io_error) {
+				    sectors[sector_num]->io_error) {
 					sblock_other = sblocks_for_recheck +
 						       mirror_index;
 					break;
@@ -1115,27 +1114,26 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check)
 
 		if (sctx->is_dev_replace) {
 			/*
-			 * did not find a mirror to fetch the page
-			 * from. scrub_write_page_to_dev_replace()
-			 * handles this case (page->io_error), by
-			 * filling the block with zeros before
-			 * submitting the write request
+			 * Did not find a mirror to fetch the sector from.
+			 * scrub_write_sector_to_dev_replace() handles this
+			 * case (sector->io_error), by filling the block with
+			 * zeros before submitting the write request
 			 */
 			if (!sblock_other)
 				sblock_other = sblock_bad;
 
-			if (scrub_write_page_to_dev_replace(sblock_other,
-							    page_num) != 0) {
+			if (scrub_write_sector_to_dev_replace(sblock_other,
+							      sector_num) != 0) {
 				atomic64_inc(
 					&fs_info->dev_replace.num_write_errors);
 				success = 0;
 			}
 		} else if (sblock_other) {
-			ret = scrub_repair_page_from_good_copy(sblock_bad,
-							       sblock_other,
-							       page_num, 0);
+			ret = scrub_repair_sector_from_good_copy(sblock_bad,
+								 sblock_other,
+								 sector_num, 0);
 			if (0 == ret)
-				spage_bad->io_error = 0;
+				sector_bad->io_error = 0;
 			else
 				success = 0;
 		}
@@ -1186,18 +1184,16 @@ out:
 			struct scrub_block *sblock = sblocks_for_recheck +
 						     mirror_index;
 			struct scrub_recover *recover;
-			int page_index;
+			int i;
 
-			for (page_index = 0; page_index < sblock->page_count;
-			     page_index++) {
-				sblock->pagev[page_index]->sblock = NULL;
-				recover = sblock->pagev[page_index]->recover;
+			for (i = 0; i < sblock->sector_count; i++) {
+				sblock->sectors[i]->sblock = NULL;
+				recover = sblock->sectors[i]->recover;
 				if (recover) {
 					scrub_put_recover(fs_info, recover);
-					sblock->pagev[page_index]->recover =
-									NULL;
+					sblock->sectors[i]->recover = NULL;
 				}
-				scrub_page_put(sblock->pagev[page_index]);
+				scrub_sector_put(sblock->sectors[i]);
 			}
 		}
 		kfree(sblocks_for_recheck);
@@ -1255,26 +1251,25 @@ static int scrub_setup_recheck_block(struct scrub_block *original_sblock,
 {
 	struct scrub_ctx *sctx = original_sblock->sctx;
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
-	u64 length = original_sblock->page_count * fs_info->sectorsize;
-	u64 logical = original_sblock->pagev[0]->logical;
-	u64 generation = original_sblock->pagev[0]->generation;
-	u64 flags = original_sblock->pagev[0]->flags;
-	u64 have_csum = original_sblock->pagev[0]->have_csum;
+	u64 length = original_sblock->sector_count << fs_info->sectorsize_bits;
+	u64 logical = original_sblock->sectors[0]->logical;
+	u64 generation = original_sblock->sectors[0]->generation;
+	u64 flags = original_sblock->sectors[0]->flags;
+	u64 have_csum = original_sblock->sectors[0]->have_csum;
 	struct scrub_recover *recover;
 	struct btrfs_io_context *bioc;
 	u64 sublen;
 	u64 mapped_length;
 	u64 stripe_offset;
 	int stripe_index;
-	int page_index = 0;
+	int sector_index = 0;
 	int mirror_index;
 	int nmirrors;
 	int ret;
 
 	/*
-	 * note: the two members refs and outstanding_pages
-	 * are not used (and not set) in the blocks that are used for
-	 * the recheck procedure
+	 * Note: the two members refs and outstanding_sectors are not used (and
+	 * not set) in the blocks that are used for the recheck procedure.
 	 */
 
 	while (length > 0) {
@@ -1306,20 +1301,20 @@ static int scrub_setup_recheck_block(struct scrub_block *original_sblock,
 		recover->bioc = bioc;
 		recover->map_length = mapped_length;
 
-		ASSERT(page_index < SCRUB_MAX_PAGES_PER_BLOCK);
+		ASSERT(sector_index < SCRUB_MAX_SECTORS_PER_BLOCK);
 
 		nmirrors = min(scrub_nr_raid_mirrors(bioc), BTRFS_MAX_MIRRORS);
 
 		for (mirror_index = 0; mirror_index < nmirrors;
 		     mirror_index++) {
 			struct scrub_block *sblock;
-			struct scrub_page *spage;
+			struct scrub_sector *sector;
 
 			sblock = sblocks_for_recheck + mirror_index;
 			sblock->sctx = sctx;
 
-			spage = kzalloc(sizeof(*spage), GFP_NOFS);
-			if (!spage) {
+			sector = kzalloc(sizeof(*sector), GFP_NOFS);
+			if (!sector) {
 leave_nomem:
 				spin_lock(&sctx->stat_lock);
 				sctx->stat.malloc_errors++;
@@ -1327,16 +1322,16 @@ leave_nomem:
 				scrub_put_recover(fs_info, recover);
 				return -ENOMEM;
 			}
-			scrub_page_get(spage);
-			sblock->pagev[page_index] = spage;
-			spage->sblock = sblock;
-			spage->flags = flags;
-			spage->generation = generation;
-			spage->logical = logical;
-			spage->have_csum = have_csum;
+			scrub_sector_get(sector);
+			sblock->sectors[sector_index] = sector;
+			sector->sblock = sblock;
+			sector->flags = flags;
+			sector->generation = generation;
+			sector->logical = logical;
+			sector->have_csum = have_csum;
 			if (have_csum)
-				memcpy(spage->csum,
-				       original_sblock->pagev[0]->csum,
+				memcpy(sector->csum,
+				       original_sblock->sectors[0]->csum,
 				       sctx->fs_info->csum_size);
 
 			scrub_stripe_index_and_offset(logical,
@@ -1348,28 +1343,28 @@ leave_nomem:
 						      mirror_index,
 						      &stripe_index,
 						      &stripe_offset);
-			spage->physical = bioc->stripes[stripe_index].physical +
+			sector->physical = bioc->stripes[stripe_index].physical +
 					 stripe_offset;
-			spage->dev = bioc->stripes[stripe_index].dev;
+			sector->dev = bioc->stripes[stripe_index].dev;
 
-			BUG_ON(page_index >= original_sblock->page_count);
-			spage->physical_for_dev_replace =
-				original_sblock->pagev[page_index]->
+			BUG_ON(sector_index >= original_sblock->sector_count);
+			sector->physical_for_dev_replace =
+				original_sblock->sectors[sector_index]->
 				physical_for_dev_replace;
-			/* for missing devices, dev->bdev is NULL */
-			spage->mirror_num = mirror_index + 1;
-			sblock->page_count++;
-			spage->page = alloc_page(GFP_NOFS);
-			if (!spage->page)
+			/* For missing devices, dev->bdev is NULL */
+			sector->mirror_num = mirror_index + 1;
+			sblock->sector_count++;
+			sector->page = alloc_page(GFP_NOFS);
+			if (!sector->page)
 				goto leave_nomem;
 
 			scrub_get_recover(recover);
-			spage->recover = recover;
+			sector->recover = recover;
 		}
 		scrub_put_recover(fs_info, recover);
 		length -= sublen;
 		logical += sublen;
-		page_index++;
+		sector_index++;
 	}
 
 	return 0;
@@ -1382,19 +1377,19 @@ static void scrub_bio_wait_endio(struct bio *bio)
 
 static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info,
 					struct bio *bio,
-					struct scrub_page *spage)
+					struct scrub_sector *sector)
 {
 	DECLARE_COMPLETION_ONSTACK(done);
 	int ret;
 	int mirror_num;
 
-	bio->bi_iter.bi_sector = spage->logical >> 9;
+	bio->bi_iter.bi_sector = sector->logical >> 9;
 	bio->bi_private = &done;
 	bio->bi_end_io = scrub_bio_wait_endio;
 
-	mirror_num = spage->sblock->pagev[0]->mirror_num;
-	ret = raid56_parity_recover(bio, spage->recover->bioc,
-				    spage->recover->map_length,
+	mirror_num = sector->sblock->sectors[0]->mirror_num;
+	ret = raid56_parity_recover(bio, sector->recover->bioc,
+				    sector->recover->map_length,
 				    mirror_num, 0);
 	if (ret)
 		return ret;
@@ -1406,26 +1401,25 @@ static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info,
 static void scrub_recheck_block_on_raid56(struct btrfs_fs_info *fs_info,
 					  struct scrub_block *sblock)
 {
-	struct scrub_page *first_page = sblock->pagev[0];
+	struct scrub_sector *first_sector = sblock->sectors[0];
 	struct bio *bio;
-	int page_num;
+	int i;
 
-	/* All pages in sblock belong to the same stripe on the same device. */
-	ASSERT(first_page->dev);
-	if (!first_page->dev->bdev)
+	/* All sectors in sblock belong to the same stripe on the same device. */
+	ASSERT(first_sector->dev);
+	if (!first_sector->dev->bdev)
 		goto out;
 
-	bio = btrfs_bio_alloc(BIO_MAX_VECS);
-	bio_set_dev(bio, first_page->dev->bdev);
+	bio = bio_alloc(first_sector->dev->bdev, BIO_MAX_VECS, REQ_OP_READ, GFP_NOFS);
 
-	for (page_num = 0; page_num < sblock->page_count; page_num++) {
-		struct scrub_page *spage = sblock->pagev[page_num];
+	for (i = 0; i < sblock->sector_count; i++) {
+		struct scrub_sector *sector = sblock->sectors[i];
 
-		WARN_ON(!spage->page);
-		bio_add_page(bio, spage->page, PAGE_SIZE, 0);
+		WARN_ON(!sector->page);
+		bio_add_page(bio, sector->page, PAGE_SIZE, 0);
 	}
 
-	if (scrub_submit_raid56_bio_wait(fs_info, bio, first_page)) {
+	if (scrub_submit_raid56_bio_wait(fs_info, bio, first_sector)) {
 		bio_put(bio);
 		goto out;
 	}
@@ -1436,65 +1430,63 @@ static void scrub_recheck_block_on_raid56(struct btrfs_fs_info *fs_info,
 
 	return;
 out:
-	for (page_num = 0; page_num < sblock->page_count; page_num++)
-		sblock->pagev[page_num]->io_error = 1;
+	for (i = 0; i < sblock->sector_count; i++)
+		sblock->sectors[i]->io_error = 1;
 
 	sblock->no_io_error_seen = 0;
 }
 
 /*
- * this function will check the on disk data for checksum errors, header
- * errors and read I/O errors. If any I/O errors happen, the exact pages
- * which are errored are marked as being bad. The goal is to enable scrub
- * to take those pages that are not errored from all the mirrors so that
- * the pages that are errored in the just handled mirror can be repaired.
+ * This function will check the on disk data for checksum errors, header errors
+ * and read I/O errors. If any I/O errors happen, the exact sectors which are
+ * errored are marked as being bad. The goal is to enable scrub to take those
+ * sectors that are not errored from all the mirrors so that the sectors that
+ * are errored in the just handled mirror can be repaired.
  */
 static void scrub_recheck_block(struct btrfs_fs_info *fs_info,
 				struct scrub_block *sblock,
 				int retry_failed_mirror)
 {
-	int page_num;
+	int i;
 
 	sblock->no_io_error_seen = 1;
 
 	/* short cut for raid56 */
-	if (!retry_failed_mirror && scrub_is_page_on_raid56(sblock->pagev[0]))
+	if (!retry_failed_mirror && scrub_is_page_on_raid56(sblock->sectors[0]))
 		return scrub_recheck_block_on_raid56(fs_info, sblock);
 
-	for (page_num = 0; page_num < sblock->page_count; page_num++) {
-		struct bio *bio;
-		struct scrub_page *spage = sblock->pagev[page_num];
+	for (i = 0; i < sblock->sector_count; i++) {
+		struct scrub_sector *sector = sblock->sectors[i];
+		struct bio bio;
+		struct bio_vec bvec;
 
-		if (spage->dev->bdev == NULL) {
-			spage->io_error = 1;
+		if (sector->dev->bdev == NULL) {
+			sector->io_error = 1;
 			sblock->no_io_error_seen = 0;
 			continue;
 		}
 
-		WARN_ON(!spage->page);
-		bio = btrfs_bio_alloc(1);
-		bio_set_dev(bio, spage->dev->bdev);
-
-		bio_add_page(bio, spage->page, fs_info->sectorsize, 0);
-		bio->bi_iter.bi_sector = spage->physical >> 9;
-		bio->bi_opf = REQ_OP_READ;
+		WARN_ON(!sector->page);
+		bio_init(&bio, sector->dev->bdev, &bvec, 1, REQ_OP_READ);
+		bio_add_page(&bio, sector->page, fs_info->sectorsize, 0);
+		bio.bi_iter.bi_sector = sector->physical >> 9;
 
-		if (btrfsic_submit_bio_wait(bio)) {
-			spage->io_error = 1;
+		btrfsic_check_bio(&bio);
+		if (submit_bio_wait(&bio)) {
+			sector->io_error = 1;
 			sblock->no_io_error_seen = 0;
 		}
 
-		bio_put(bio);
+		bio_uninit(&bio);
 	}
 
 	if (sblock->no_io_error_seen)
 		scrub_recheck_block_checksum(sblock);
 }
 
-static inline int scrub_check_fsid(u8 fsid[],
-				   struct scrub_page *spage)
+static inline int scrub_check_fsid(u8 fsid[], struct scrub_sector *sector)
 {
-	struct btrfs_fs_devices *fs_devices = spage->dev->fs_devices;
+	struct btrfs_fs_devices *fs_devices = sector->dev->fs_devices;
 	int ret;
 
 	ret = memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE);
@@ -1507,7 +1499,7 @@ static void scrub_recheck_block_checksum(struct scrub_block *sblock)
 	sblock->checksum_error = 0;
 	sblock->generation_error = 0;
 
-	if (sblock->pagev[0]->flags & BTRFS_EXTENT_FLAG_DATA)
+	if (sblock->sectors[0]->flags & BTRFS_EXTENT_FLAG_DATA)
 		scrub_checksum_data(sblock);
 	else
 		scrub_checksum_tree_block(sblock);
@@ -1516,15 +1508,14 @@ static void scrub_recheck_block_checksum(struct scrub_block *sblock)
 static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
 					     struct scrub_block *sblock_good)
 {
-	int page_num;
+	int i;
 	int ret = 0;
 
-	for (page_num = 0; page_num < sblock_bad->page_count; page_num++) {
+	for (i = 0; i < sblock_bad->sector_count; i++) {
 		int ret_sub;
 
-		ret_sub = scrub_repair_page_from_good_copy(sblock_bad,
-							   sblock_good,
-							   page_num, 1);
+		ret_sub = scrub_repair_sector_from_good_copy(sblock_bad,
+							     sblock_good, i, 1);
 		if (ret_sub)
 			ret = ret_sub;
 	}
@@ -1532,47 +1523,43 @@ static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad,
 	return ret;
 }
 
-static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
-					    struct scrub_block *sblock_good,
-					    int page_num, int force_write)
+static int scrub_repair_sector_from_good_copy(struct scrub_block *sblock_bad,
+					      struct scrub_block *sblock_good,
+					      int sector_num, int force_write)
 {
-	struct scrub_page *spage_bad = sblock_bad->pagev[page_num];
-	struct scrub_page *spage_good = sblock_good->pagev[page_num];
+	struct scrub_sector *sector_bad = sblock_bad->sectors[sector_num];
+	struct scrub_sector *sector_good = sblock_good->sectors[sector_num];
 	struct btrfs_fs_info *fs_info = sblock_bad->sctx->fs_info;
 	const u32 sectorsize = fs_info->sectorsize;
 
-	BUG_ON(spage_bad->page == NULL);
-	BUG_ON(spage_good->page == NULL);
+	BUG_ON(sector_bad->page == NULL);
+	BUG_ON(sector_good->page == NULL);
 	if (force_write || sblock_bad->header_error ||
-	    sblock_bad->checksum_error || spage_bad->io_error) {
-		struct bio *bio;
+	    sblock_bad->checksum_error || sector_bad->io_error) {
+		struct bio bio;
+		struct bio_vec bvec;
 		int ret;
 
-		if (!spage_bad->dev->bdev) {
+		if (!sector_bad->dev->bdev) {
 			btrfs_warn_rl(fs_info,
 				"scrub_repair_page_from_good_copy(bdev == NULL) is unexpected");
 			return -EIO;
 		}
 
-		bio = btrfs_bio_alloc(1);
-		bio_set_dev(bio, spage_bad->dev->bdev);
-		bio->bi_iter.bi_sector = spage_bad->physical >> 9;
-		bio->bi_opf = REQ_OP_WRITE;
+		bio_init(&bio, sector_bad->dev->bdev, &bvec, 1, REQ_OP_WRITE);
+		bio.bi_iter.bi_sector = sector_bad->physical >> 9;
+		__bio_add_page(&bio, sector_good->page, sectorsize, 0);
 
-		ret = bio_add_page(bio, spage_good->page, sectorsize, 0);
-		if (ret != sectorsize) {
-			bio_put(bio);
-			return -EIO;
-		}
+		btrfsic_check_bio(&bio);
+		ret = submit_bio_wait(&bio);
+		bio_uninit(&bio);
 
-		if (btrfsic_submit_bio_wait(bio)) {
-			btrfs_dev_stat_inc_and_print(spage_bad->dev,
+		if (ret) {
+			btrfs_dev_stat_inc_and_print(sector_bad->dev,
 				BTRFS_DEV_STAT_WRITE_ERRS);
 			atomic64_inc(&fs_info->dev_replace.num_write_errors);
-			bio_put(bio);
 			return -EIO;
 		}
-		bio_put(bio);
 	}
 
 	return 0;
@@ -1581,7 +1568,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad,
 static void scrub_write_block_to_dev_replace(struct scrub_block *sblock)
 {
 	struct btrfs_fs_info *fs_info = sblock->sctx->fs_info;
-	int page_num;
+	int i;
 
 	/*
 	 * This block is used for the check of the parity on the source device,
@@ -1590,25 +1577,24 @@ static void scrub_write_block_to_dev_replace(struct scrub_block *sblock)
 	if (sblock->sparity)
 		return;
 
-	for (page_num = 0; page_num < sblock->page_count; page_num++) {
+	for (i = 0; i < sblock->sector_count; i++) {
 		int ret;
 
-		ret = scrub_write_page_to_dev_replace(sblock, page_num);
+		ret = scrub_write_sector_to_dev_replace(sblock, i);
 		if (ret)
 			atomic64_inc(&fs_info->dev_replace.num_write_errors);
 	}
 }
 
-static int scrub_write_page_to_dev_replace(struct scrub_block *sblock,
-					   int page_num)
+static int scrub_write_sector_to_dev_replace(struct scrub_block *sblock, int sector_num)
 {
-	struct scrub_page *spage = sblock->pagev[page_num];
+	struct scrub_sector *sector = sblock->sectors[sector_num];
 
-	BUG_ON(spage->page == NULL);
-	if (spage->io_error)
-		clear_page(page_address(spage->page));
+	BUG_ON(sector->page == NULL);
+	if (sector->io_error)
+		clear_page(page_address(sector->page));
 
-	return scrub_add_page_to_wr_bio(sblock->sctx, spage);
+	return scrub_add_sector_to_wr_bio(sblock->sctx, sector);
 }
 
 static int fill_writer_pointer_gap(struct scrub_ctx *sctx, u64 physical)
@@ -1633,8 +1619,8 @@ static int fill_writer_pointer_gap(struct scrub_ctx *sctx, u64 physical)
 	return ret;
 }
 
-static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx,
-				    struct scrub_page *spage)
+static int scrub_add_sector_to_wr_bio(struct scrub_ctx *sctx,
+				      struct scrub_sector *sector)
 {
 	struct scrub_bio *sbio;
 	int ret;
@@ -1650,45 +1636,38 @@ again:
 			return -ENOMEM;
 		}
 		sctx->wr_curr_bio->sctx = sctx;
-		sctx->wr_curr_bio->page_count = 0;
+		sctx->wr_curr_bio->sector_count = 0;
 	}
 	sbio = sctx->wr_curr_bio;
-	if (sbio->page_count == 0) {
-		struct bio *bio;
-
-		ret = fill_writer_pointer_gap(sctx,
-					      spage->physical_for_dev_replace);
+	if (sbio->sector_count == 0) {
+		ret = fill_writer_pointer_gap(sctx, sector->physical_for_dev_replace);
 		if (ret) {
 			mutex_unlock(&sctx->wr_lock);
 			return ret;
 		}
 
-		sbio->physical = spage->physical_for_dev_replace;
-		sbio->logical = spage->logical;
+		sbio->physical = sector->physical_for_dev_replace;
+		sbio->logical = sector->logical;
 		sbio->dev = sctx->wr_tgtdev;
-		bio = sbio->bio;
-		if (!bio) {
-			bio = btrfs_bio_alloc(sctx->pages_per_bio);
-			sbio->bio = bio;
+		if (!sbio->bio) {
+			sbio->bio = bio_alloc(sbio->dev->bdev, sctx->sectors_per_bio,
+					      REQ_OP_WRITE, GFP_NOFS);
 		}
-
-		bio->bi_private = sbio;
-		bio->bi_end_io = scrub_wr_bio_end_io;
-		bio_set_dev(bio, sbio->dev->bdev);
-		bio->bi_iter.bi_sector = sbio->physical >> 9;
-		bio->bi_opf = REQ_OP_WRITE;
+		sbio->bio->bi_private = sbio;
+		sbio->bio->bi_end_io = scrub_wr_bio_end_io;
+		sbio->bio->bi_iter.bi_sector = sbio->physical >> 9;
 		sbio->status = 0;
-	} else if (sbio->physical + sbio->page_count * sectorsize !=
-		   spage->physical_for_dev_replace ||
-		   sbio->logical + sbio->page_count * sectorsize !=
-		   spage->logical) {
+	} else if (sbio->physical + sbio->sector_count * sectorsize !=
+		   sector->physical_for_dev_replace ||
+		   sbio->logical + sbio->sector_count * sectorsize !=
+		   sector->logical) {
 		scrub_wr_submit(sctx);
 		goto again;
 	}
 
-	ret = bio_add_page(sbio->bio, spage->page, sectorsize, 0);
+	ret = bio_add_page(sbio->bio, sector->page, sectorsize, 0);
 	if (ret != sectorsize) {
-		if (sbio->page_count < 1) {
+		if (sbio->sector_count < 1) {
 			bio_put(sbio->bio);
 			sbio->bio = NULL;
 			mutex_unlock(&sctx->wr_lock);
@@ -1698,10 +1677,10 @@ again:
 		goto again;
 	}
 
-	sbio->pagev[sbio->page_count] = spage;
-	scrub_page_get(spage);
-	sbio->page_count++;
-	if (sbio->page_count == sctx->pages_per_bio)
+	sbio->sectors[sbio->sector_count] = sector;
+	scrub_sector_get(sector);
+	sbio->sector_count++;
+	if (sbio->sector_count == sctx->sectors_per_bio)
 		scrub_wr_submit(sctx);
 	mutex_unlock(&sctx->wr_lock);
 
@@ -1717,16 +1696,16 @@ static void scrub_wr_submit(struct scrub_ctx *sctx)
 
 	sbio = sctx->wr_curr_bio;
 	sctx->wr_curr_bio = NULL;
-	WARN_ON(!sbio->bio->bi_bdev);
 	scrub_pending_bio_inc(sctx);
 	/* process all writes in a single worker thread. Then the block layer
 	 * orders the requests before sending them to the driver which
 	 * doubled the write performance on spinning disks when measured
 	 * with Linux 3.5 */
-	btrfsic_submit_bio(sbio->bio);
+	btrfsic_check_bio(sbio->bio);
+	submit_bio(sbio->bio);
 
 	if (btrfs_is_zoned(sctx->fs_info))
-		sctx->write_pointer = sbio->physical + sbio->page_count *
+		sctx->write_pointer = sbio->physical + sbio->sector_count *
 			sctx->fs_info->sectorsize;
 }
 
@@ -1738,31 +1717,31 @@ static void scrub_wr_bio_end_io(struct bio *bio)
 	sbio->status = bio->bi_status;
 	sbio->bio = bio;
 
-	btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL);
-	btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work);
+	INIT_WORK(&sbio->work, scrub_wr_bio_end_io_worker);
+	queue_work(fs_info->scrub_wr_completion_workers, &sbio->work);
 }
 
-static void scrub_wr_bio_end_io_worker(struct btrfs_work *work)
+static void scrub_wr_bio_end_io_worker(struct work_struct *work)
 {
 	struct scrub_bio *sbio = container_of(work, struct scrub_bio, work);
 	struct scrub_ctx *sctx = sbio->sctx;
 	int i;
 
-	ASSERT(sbio->page_count <= SCRUB_PAGES_PER_BIO);
+	ASSERT(sbio->sector_count <= SCRUB_SECTORS_PER_BIO);
 	if (sbio->status) {
 		struct btrfs_dev_replace *dev_replace =
 			&sbio->sctx->fs_info->dev_replace;
 
-		for (i = 0; i < sbio->page_count; i++) {
-			struct scrub_page *spage = sbio->pagev[i];
+		for (i = 0; i < sbio->sector_count; i++) {
+			struct scrub_sector *sector = sbio->sectors[i];
 
-			spage->io_error = 1;
+			sector->io_error = 1;
 			atomic64_inc(&dev_replace->num_write_errors);
 		}
 	}
 
-	for (i = 0; i < sbio->page_count; i++)
-		scrub_page_put(sbio->pagev[i]);
+	for (i = 0; i < sbio->sector_count; i++)
+		scrub_sector_put(sbio->sectors[i]);
 
 	bio_put(sbio->bio);
 	kfree(sbio);
@@ -1786,8 +1765,8 @@ static int scrub_checksum(struct scrub_block *sblock)
 	sblock->generation_error = 0;
 	sblock->checksum_error = 0;
 
-	WARN_ON(sblock->page_count < 1);
-	flags = sblock->pagev[0]->flags;
+	WARN_ON(sblock->sector_count < 1);
+	flags = sblock->sectors[0]->flags;
 	ret = 0;
 	if (flags & BTRFS_EXTENT_FLAG_DATA)
 		ret = scrub_checksum_data(sblock);
@@ -1809,26 +1788,26 @@ static int scrub_checksum_data(struct scrub_block *sblock)
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
 	u8 csum[BTRFS_CSUM_SIZE];
-	struct scrub_page *spage;
+	struct scrub_sector *sector;
 	char *kaddr;
 
-	BUG_ON(sblock->page_count < 1);
-	spage = sblock->pagev[0];
-	if (!spage->have_csum)
+	BUG_ON(sblock->sector_count < 1);
+	sector = sblock->sectors[0];
+	if (!sector->have_csum)
 		return 0;
 
-	kaddr = page_address(spage->page);
+	kaddr = page_address(sector->page);
 
 	shash->tfm = fs_info->csum_shash;
 	crypto_shash_init(shash);
 
 	/*
-	 * In scrub_pages() and scrub_pages_for_parity() we ensure each spage
+	 * In scrub_sectors() and scrub_sectors_for_parity() we ensure each sector
 	 * only contains one sector of data.
 	 */
 	crypto_shash_digest(shash, kaddr, fs_info->sectorsize, csum);
 
-	if (memcmp(csum, spage->csum, fs_info->csum_size))
+	if (memcmp(csum, sector->csum, fs_info->csum_size))
 		sblock->checksum_error = 1;
 	return sblock->checksum_error;
 }
@@ -1849,16 +1828,16 @@ static int scrub_checksum_tree_block(struct scrub_block *sblock)
 	const u32 sectorsize = sctx->fs_info->sectorsize;
 	const int num_sectors = fs_info->nodesize >> fs_info->sectorsize_bits;
 	int i;
-	struct scrub_page *spage;
+	struct scrub_sector *sector;
 	char *kaddr;
 
-	BUG_ON(sblock->page_count < 1);
+	BUG_ON(sblock->sector_count < 1);
 
-	/* Each member in pagev is just one block, not a full page */
-	ASSERT(sblock->page_count == num_sectors);
+	/* Each member in sectors is just one sector */
+	ASSERT(sblock->sector_count == num_sectors);
 
-	spage = sblock->pagev[0];
-	kaddr = page_address(spage->page);
+	sector = sblock->sectors[0];
+	kaddr = page_address(sector->page);
 	h = (struct btrfs_header *)kaddr;
 	memcpy(on_disk_csum, h->csum, sctx->fs_info->csum_size);
 
@@ -1867,15 +1846,15 @@ static int scrub_checksum_tree_block(struct scrub_block *sblock)
 	 * a) don't have an extent buffer and
 	 * b) the page is already kmapped
 	 */
-	if (spage->logical != btrfs_stack_header_bytenr(h))
+	if (sector->logical != btrfs_stack_header_bytenr(h))
 		sblock->header_error = 1;
 
-	if (spage->generation != btrfs_stack_header_generation(h)) {
+	if (sector->generation != btrfs_stack_header_generation(h)) {
 		sblock->header_error = 1;
 		sblock->generation_error = 1;
 	}
 
-	if (!scrub_check_fsid(h->fsid, spage))
+	if (!scrub_check_fsid(h->fsid, sector))
 		sblock->header_error = 1;
 
 	if (memcmp(h->chunk_tree_uuid, fs_info->chunk_tree_uuid,
@@ -1888,7 +1867,7 @@ static int scrub_checksum_tree_block(struct scrub_block *sblock)
 			    sectorsize - BTRFS_CSUM_SIZE);
 
 	for (i = 1; i < num_sectors; i++) {
-		kaddr = page_address(sblock->pagev[i]->page);
+		kaddr = page_address(sblock->sectors[i]->page);
 		crypto_shash_update(shash, kaddr, sectorsize);
 	}
 
@@ -1906,23 +1885,23 @@ static int scrub_checksum_super(struct scrub_block *sblock)
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
 	u8 calculated_csum[BTRFS_CSUM_SIZE];
-	struct scrub_page *spage;
+	struct scrub_sector *sector;
 	char *kaddr;
 	int fail_gen = 0;
 	int fail_cor = 0;
 
-	BUG_ON(sblock->page_count < 1);
-	spage = sblock->pagev[0];
-	kaddr = page_address(spage->page);
+	BUG_ON(sblock->sector_count < 1);
+	sector = sblock->sectors[0];
+	kaddr = page_address(sector->page);
 	s = (struct btrfs_super_block *)kaddr;
 
-	if (spage->logical != btrfs_super_bytenr(s))
+	if (sector->logical != btrfs_super_bytenr(s))
 		++fail_cor;
 
-	if (spage->generation != btrfs_super_generation(s))
+	if (sector->generation != btrfs_super_generation(s))
 		++fail_gen;
 
-	if (!scrub_check_fsid(s->fsid, spage))
+	if (!scrub_check_fsid(s->fsid, sector))
 		++fail_cor;
 
 	shash->tfm = fs_info->csum_shash;
@@ -1943,10 +1922,10 @@ static int scrub_checksum_super(struct scrub_block *sblock)
 		++sctx->stat.super_errors;
 		spin_unlock(&sctx->stat_lock);
 		if (fail_cor)
-			btrfs_dev_stat_inc_and_print(spage->dev,
+			btrfs_dev_stat_inc_and_print(sector->dev,
 				BTRFS_DEV_STAT_CORRUPTION_ERRS);
 		else
-			btrfs_dev_stat_inc_and_print(spage->dev,
+			btrfs_dev_stat_inc_and_print(sector->dev,
 				BTRFS_DEV_STAT_GENERATION_ERRS);
 	}
 
@@ -1966,23 +1945,23 @@ static void scrub_block_put(struct scrub_block *sblock)
 		if (sblock->sparity)
 			scrub_parity_put(sblock->sparity);
 
-		for (i = 0; i < sblock->page_count; i++)
-			scrub_page_put(sblock->pagev[i]);
+		for (i = 0; i < sblock->sector_count; i++)
+			scrub_sector_put(sblock->sectors[i]);
 		kfree(sblock);
 	}
 }
 
-static void scrub_page_get(struct scrub_page *spage)
+static void scrub_sector_get(struct scrub_sector *sector)
 {
-	atomic_inc(&spage->refs);
+	atomic_inc(&sector->refs);
 }
 
-static void scrub_page_put(struct scrub_page *spage)
+static void scrub_sector_put(struct scrub_sector *sector)
 {
-	if (atomic_dec_and_test(&spage->refs)) {
-		if (spage->page)
-			__free_page(spage->page);
-		kfree(spage);
+	if (atomic_dec_and_test(&sector->refs)) {
+		if (sector->page)
+			__free_page(sector->page);
+		kfree(sector);
 	}
 }
 
@@ -2057,13 +2036,14 @@ static void scrub_submit(struct scrub_ctx *sctx)
 	sbio = sctx->bios[sctx->curr];
 	sctx->curr = -1;
 	scrub_pending_bio_inc(sctx);
-	btrfsic_submit_bio(sbio->bio);
+	btrfsic_check_bio(sbio->bio);
+	submit_bio(sbio->bio);
 }
 
-static int scrub_add_page_to_rd_bio(struct scrub_ctx *sctx,
-				    struct scrub_page *spage)
+static int scrub_add_sector_to_rd_bio(struct scrub_ctx *sctx,
+				      struct scrub_sector *sector)
 {
-	struct scrub_block *sblock = spage->sblock;
+	struct scrub_block *sblock = sector->sblock;
 	struct scrub_bio *sbio;
 	const u32 sectorsize = sctx->fs_info->sectorsize;
 	int ret;
@@ -2078,7 +2058,7 @@ again:
 		if (sctx->curr != -1) {
 			sctx->first_free = sctx->bios[sctx->curr]->next_free;
 			sctx->bios[sctx->curr]->next_free = -1;
-			sctx->bios[sctx->curr]->page_count = 0;
+			sctx->bios[sctx->curr]->sector_count = 0;
 			spin_unlock(&sctx->list_lock);
 		} else {
 			spin_unlock(&sctx->list_lock);
@@ -2086,37 +2066,31 @@ again:
 		}
 	}
 	sbio = sctx->bios[sctx->curr];
-	if (sbio->page_count == 0) {
-		struct bio *bio;
-
-		sbio->physical = spage->physical;
-		sbio->logical = spage->logical;
-		sbio->dev = spage->dev;
-		bio = sbio->bio;
-		if (!bio) {
-			bio = btrfs_bio_alloc(sctx->pages_per_bio);
-			sbio->bio = bio;
+	if (sbio->sector_count == 0) {
+		sbio->physical = sector->physical;
+		sbio->logical = sector->logical;
+		sbio->dev = sector->dev;
+		if (!sbio->bio) {
+			sbio->bio = bio_alloc(sbio->dev->bdev, sctx->sectors_per_bio,
+					      REQ_OP_READ, GFP_NOFS);
 		}
-
-		bio->bi_private = sbio;
-		bio->bi_end_io = scrub_bio_end_io;
-		bio_set_dev(bio, sbio->dev->bdev);
-		bio->bi_iter.bi_sector = sbio->physical >> 9;
-		bio->bi_opf = REQ_OP_READ;
+		sbio->bio->bi_private = sbio;
+		sbio->bio->bi_end_io = scrub_bio_end_io;
+		sbio->bio->bi_iter.bi_sector = sbio->physical >> 9;
 		sbio->status = 0;
-	} else if (sbio->physical + sbio->page_count * sectorsize !=
-		   spage->physical ||
-		   sbio->logical + sbio->page_count * sectorsize !=
-		   spage->logical ||
-		   sbio->dev != spage->dev) {
+	} else if (sbio->physical + sbio->sector_count * sectorsize !=
+		   sector->physical ||
+		   sbio->logical + sbio->sector_count * sectorsize !=
+		   sector->logical ||
+		   sbio->dev != sector->dev) {
 		scrub_submit(sctx);
 		goto again;
 	}
 
-	sbio->pagev[sbio->page_count] = spage;
-	ret = bio_add_page(sbio->bio, spage->page, sectorsize, 0);
+	sbio->sectors[sbio->sector_count] = sector;
+	ret = bio_add_page(sbio->bio, sector->page, sectorsize, 0);
 	if (ret != sectorsize) {
-		if (sbio->page_count < 1) {
+		if (sbio->sector_count < 1) {
 			bio_put(sbio->bio);
 			sbio->bio = NULL;
 			return -EIO;
@@ -2126,9 +2100,9 @@ again:
 	}
 
 	scrub_block_get(sblock); /* one for the page added to the bio */
-	atomic_inc(&sblock->outstanding_pages);
-	sbio->page_count++;
-	if (sbio->page_count == sctx->pages_per_bio)
+	atomic_inc(&sblock->outstanding_sectors);
+	sbio->sector_count++;
+	if (sbio->sector_count == sctx->sectors_per_bio)
 		scrub_submit(sctx);
 
 	return 0;
@@ -2144,10 +2118,10 @@ static void scrub_missing_raid56_end_io(struct bio *bio)
 
 	bio_put(bio);
 
-	btrfs_queue_work(fs_info->scrub_workers, &sblock->work);
+	queue_work(fs_info->scrub_workers, &sblock->work);
 }
 
-static void scrub_missing_raid56_worker(struct btrfs_work *work)
+static void scrub_missing_raid56_worker(struct work_struct *work)
 {
 	struct scrub_block *sblock = container_of(work, struct scrub_block, work);
 	struct scrub_ctx *sctx = sblock->sctx;
@@ -2155,8 +2129,8 @@ static void scrub_missing_raid56_worker(struct btrfs_work *work)
 	u64 logical;
 	struct btrfs_device *dev;
 
-	logical = sblock->pagev[0]->logical;
-	dev = sblock->pagev[0]->dev;
+	logical = sblock->sectors[0]->logical;
+	dev = sblock->sectors[0]->dev;
 
 	if (sblock->no_io_error_seen)
 		scrub_recheck_block_checksum(sblock);
@@ -2193,8 +2167,8 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
 {
 	struct scrub_ctx *sctx = sblock->sctx;
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
-	u64 length = sblock->page_count * PAGE_SIZE;
-	u64 logical = sblock->pagev[0]->logical;
+	u64 length = sblock->sector_count << fs_info->sectorsize_bits;
+	u64 logical = sblock->sectors[0]->logical;
 	struct btrfs_io_context *bioc = NULL;
 	struct bio *bio;
 	struct btrfs_raid_bio *rbio;
@@ -2213,12 +2187,12 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
 		 * We shouldn't be scrubbing a missing device. Even for dev
 		 * replace, we should only get here for RAID 5/6. We either
 		 * managed to mount something with no mirrors remaining or
-		 * there's a bug in scrub_remap_extent()/btrfs_map_block().
+		 * there's a bug in scrub_find_good_copy()/btrfs_map_block().
 		 */
 		goto bioc_out;
 	}
 
-	bio = btrfs_bio_alloc(BIO_MAX_VECS);
+	bio = bio_alloc(NULL, BIO_MAX_VECS, REQ_OP_READ, GFP_NOFS);
 	bio->bi_iter.bi_sector = logical >> 9;
 	bio->bi_private = sblock;
 	bio->bi_end_io = scrub_missing_raid56_end_io;
@@ -2227,13 +2201,17 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock)
 	if (!rbio)
 		goto rbio_out;
 
-	for (i = 0; i < sblock->page_count; i++) {
-		struct scrub_page *spage = sblock->pagev[i];
+	for (i = 0; i < sblock->sector_count; i++) {
+		struct scrub_sector *sector = sblock->sectors[i];
 
-		raid56_add_scrub_pages(rbio, spage->page, spage->logical);
+		/*
+		 * For now, our scrub is still one page per sector, so pgoff
+		 * is always 0.
+		 */
+		raid56_add_scrub_pages(rbio, sector->page, 0, sector->logical);
 	}
 
-	btrfs_init_work(&sblock->work, scrub_missing_raid56_worker, NULL, NULL);
+	INIT_WORK(&sblock->work, scrub_missing_raid56_worker);
 	scrub_block_get(sblock);
 	scrub_pending_bio_inc(sctx);
 	raid56_submit_missing_rbio(rbio);
@@ -2249,7 +2227,7 @@ bioc_out:
 	spin_unlock(&sctx->stat_lock);
 }
 
-static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u32 len,
+static int scrub_sectors(struct scrub_ctx *sctx, u64 logical, u32 len,
 		       u64 physical, struct btrfs_device *dev, u64 flags,
 		       u64 gen, int mirror_num, u8 *csum,
 		       u64 physical_for_dev_replace)
@@ -2273,7 +2251,7 @@ static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u32 len,
 	sblock->no_io_error_seen = 1;
 
 	for (index = 0; len > 0; index++) {
-		struct scrub_page *spage;
+		struct scrub_sector *sector;
 		/*
 		 * Here we will allocate one page for one sector to scrub.
 		 * This is fine if PAGE_SIZE == sectorsize, but will cost
@@ -2281,8 +2259,8 @@ static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u32 len,
 		 */
 		u32 l = min(sectorsize, len);
 
-		spage = kzalloc(sizeof(*spage), GFP_KERNEL);
-		if (!spage) {
+		sector = kzalloc(sizeof(*sector), GFP_KERNEL);
+		if (!sector) {
 leave_nomem:
 			spin_lock(&sctx->stat_lock);
 			sctx->stat.malloc_errors++;
@@ -2290,26 +2268,26 @@ leave_nomem:
 			scrub_block_put(sblock);
 			return -ENOMEM;
 		}
-		ASSERT(index < SCRUB_MAX_PAGES_PER_BLOCK);
-		scrub_page_get(spage);
-		sblock->pagev[index] = spage;
-		spage->sblock = sblock;
-		spage->dev = dev;
-		spage->flags = flags;
-		spage->generation = gen;
-		spage->logical = logical;
-		spage->physical = physical;
-		spage->physical_for_dev_replace = physical_for_dev_replace;
-		spage->mirror_num = mirror_num;
+		ASSERT(index < SCRUB_MAX_SECTORS_PER_BLOCK);
+		scrub_sector_get(sector);
+		sblock->sectors[index] = sector;
+		sector->sblock = sblock;
+		sector->dev = dev;
+		sector->flags = flags;
+		sector->generation = gen;
+		sector->logical = logical;
+		sector->physical = physical;
+		sector->physical_for_dev_replace = physical_for_dev_replace;
+		sector->mirror_num = mirror_num;
 		if (csum) {
-			spage->have_csum = 1;
-			memcpy(spage->csum, csum, sctx->fs_info->csum_size);
+			sector->have_csum = 1;
+			memcpy(sector->csum, csum, sctx->fs_info->csum_size);
 		} else {
-			spage->have_csum = 0;
+			sector->have_csum = 0;
 		}
-		sblock->page_count++;
-		spage->page = alloc_page(GFP_KERNEL);
-		if (!spage->page)
+		sblock->sector_count++;
+		sector->page = alloc_page(GFP_KERNEL);
+		if (!sector->page)
 			goto leave_nomem;
 		len -= l;
 		logical += l;
@@ -2317,7 +2295,7 @@ leave_nomem:
 		physical_for_dev_replace += l;
 	}
 
-	WARN_ON(sblock->page_count == 0);
+	WARN_ON(sblock->sector_count == 0);
 	if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state)) {
 		/*
 		 * This case should only be hit for RAID 5/6 device replace. See
@@ -2325,11 +2303,11 @@ leave_nomem:
 		 */
 		scrub_missing_raid56_pages(sblock);
 	} else {
-		for (index = 0; index < sblock->page_count; index++) {
-			struct scrub_page *spage = sblock->pagev[index];
+		for (index = 0; index < sblock->sector_count; index++) {
+			struct scrub_sector *sector = sblock->sectors[index];
 			int ret;
 
-			ret = scrub_add_page_to_rd_bio(sctx, spage);
+			ret = scrub_add_sector_to_rd_bio(sctx, sector);
 			if (ret) {
 				scrub_block_put(sblock);
 				return ret;
@@ -2353,31 +2331,31 @@ static void scrub_bio_end_io(struct bio *bio)
 	sbio->status = bio->bi_status;
 	sbio->bio = bio;
 
-	btrfs_queue_work(fs_info->scrub_workers, &sbio->work);
+	queue_work(fs_info->scrub_workers, &sbio->work);
 }
 
-static void scrub_bio_end_io_worker(struct btrfs_work *work)
+static void scrub_bio_end_io_worker(struct work_struct *work)
 {
 	struct scrub_bio *sbio = container_of(work, struct scrub_bio, work);
 	struct scrub_ctx *sctx = sbio->sctx;
 	int i;
 
-	ASSERT(sbio->page_count <= SCRUB_PAGES_PER_BIO);
+	ASSERT(sbio->sector_count <= SCRUB_SECTORS_PER_BIO);
 	if (sbio->status) {
-		for (i = 0; i < sbio->page_count; i++) {
-			struct scrub_page *spage = sbio->pagev[i];
+		for (i = 0; i < sbio->sector_count; i++) {
+			struct scrub_sector *sector = sbio->sectors[i];
 
-			spage->io_error = 1;
-			spage->sblock->no_io_error_seen = 0;
+			sector->io_error = 1;
+			sector->sblock->no_io_error_seen = 0;
 		}
 	}
 
-	/* now complete the scrub_block items that have all pages completed */
-	for (i = 0; i < sbio->page_count; i++) {
-		struct scrub_page *spage = sbio->pagev[i];
-		struct scrub_block *sblock = spage->sblock;
+	/* Now complete the scrub_block items that have all pages completed */
+	for (i = 0; i < sbio->sector_count; i++) {
+		struct scrub_sector *sector = sbio->sectors[i];
+		struct scrub_block *sblock = sector->sblock;
 
-		if (atomic_dec_and_test(&sblock->outstanding_pages))
+		if (atomic_dec_and_test(&sblock->outstanding_sectors))
 			scrub_block_complete(sblock);
 		scrub_block_put(sblock);
 	}
@@ -2456,8 +2434,8 @@ static void scrub_block_complete(struct scrub_block *sblock)
 	}
 
 	if (sblock->sparity && corrupted && !sblock->data_corrected) {
-		u64 start = sblock->pagev[0]->logical;
-		u64 end = sblock->pagev[sblock->page_count - 1]->logical +
+		u64 start = sblock->sectors[0]->logical;
+		u64 end = sblock->sectors[sblock->sector_count - 1]->logical +
 			  sblock->sctx->fs_info->sectorsize;
 
 		ASSERT(end - start <= U32_MAX);
@@ -2532,8 +2510,11 @@ static int scrub_find_csum(struct scrub_ctx *sctx, u64 logical, u8 *csum)
 static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map,
 			u64 logical, u32 len,
 			u64 physical, struct btrfs_device *dev, u64 flags,
-			u64 gen, int mirror_num, u64 physical_for_dev_replace)
+			u64 gen, int mirror_num)
 {
+	struct btrfs_device *src_dev = dev;
+	u64 src_physical = physical;
+	int src_mirror = mirror_num;
 	int ret;
 	u8 csum[BTRFS_CSUM_SIZE];
 	u32 blocksize;
@@ -2561,6 +2542,18 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map,
 		WARN_ON(1);
 	}
 
+	/*
+	 * For dev-replace case, we can have @dev being a missing device.
+	 * Regular scrub will avoid its execution on missing device at all,
+	 * as that would trigger tons of read error.
+	 *
+	 * Reading from missing device will cause read error counts to
+	 * increase unnecessarily.
+	 * So here we change the read source to a good mirror.
+	 */
+	if (sctx->is_dev_replace && !dev->bdev)
+		scrub_find_good_copy(sctx->fs_info, logical, len, &src_physical,
+				     &src_dev, &src_mirror);
 	while (len) {
 		u32 l = min(len, blocksize);
 		int have_csum = 0;
@@ -2571,20 +2564,20 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map,
 			if (have_csum == 0)
 				++sctx->stat.no_csum;
 		}
-		ret = scrub_pages(sctx, logical, l, physical, dev, flags, gen,
-				  mirror_num, have_csum ? csum : NULL,
-				  physical_for_dev_replace);
+		ret = scrub_sectors(sctx, logical, l, src_physical, src_dev,
+				    flags, gen, src_mirror,
+				    have_csum ? csum : NULL, physical);
 		if (ret)
 			return ret;
 		len -= l;
 		logical += l;
 		physical += l;
-		physical_for_dev_replace += l;
+		src_physical += l;
 	}
 	return 0;
 }
 
-static int scrub_pages_for_parity(struct scrub_parity *sparity,
+static int scrub_sectors_for_parity(struct scrub_parity *sparity,
 				  u64 logical, u32 len,
 				  u64 physical, struct btrfs_device *dev,
 				  u64 flags, u64 gen, int mirror_num, u8 *csum)
@@ -2613,10 +2606,10 @@ static int scrub_pages_for_parity(struct scrub_parity *sparity,
 	scrub_parity_get(sparity);
 
 	for (index = 0; len > 0; index++) {
-		struct scrub_page *spage;
+		struct scrub_sector *sector;
 
-		spage = kzalloc(sizeof(*spage), GFP_KERNEL);
-		if (!spage) {
+		sector = kzalloc(sizeof(*sector), GFP_KERNEL);
+		if (!sector) {
 leave_nomem:
 			spin_lock(&sctx->stat_lock);
 			sctx->stat.malloc_errors++;
@@ -2624,29 +2617,29 @@ leave_nomem:
 			scrub_block_put(sblock);
 			return -ENOMEM;
 		}
-		ASSERT(index < SCRUB_MAX_PAGES_PER_BLOCK);
+		ASSERT(index < SCRUB_MAX_SECTORS_PER_BLOCK);
 		/* For scrub block */
-		scrub_page_get(spage);
-		sblock->pagev[index] = spage;
+		scrub_sector_get(sector);
+		sblock->sectors[index] = sector;
 		/* For scrub parity */
-		scrub_page_get(spage);
-		list_add_tail(&spage->list, &sparity->spages);
-		spage->sblock = sblock;
-		spage->dev = dev;
-		spage->flags = flags;
-		spage->generation = gen;
-		spage->logical = logical;
-		spage->physical = physical;
-		spage->mirror_num = mirror_num;
+		scrub_sector_get(sector);
+		list_add_tail(&sector->list, &sparity->sectors_list);
+		sector->sblock = sblock;
+		sector->dev = dev;
+		sector->flags = flags;
+		sector->generation = gen;
+		sector->logical = logical;
+		sector->physical = physical;
+		sector->mirror_num = mirror_num;
 		if (csum) {
-			spage->have_csum = 1;
-			memcpy(spage->csum, csum, sctx->fs_info->csum_size);
+			sector->have_csum = 1;
+			memcpy(sector->csum, csum, sctx->fs_info->csum_size);
 		} else {
-			spage->have_csum = 0;
+			sector->have_csum = 0;
 		}
-		sblock->page_count++;
-		spage->page = alloc_page(GFP_KERNEL);
-		if (!spage->page)
+		sblock->sector_count++;
+		sector->page = alloc_page(GFP_KERNEL);
+		if (!sector->page)
 			goto leave_nomem;
 
 
@@ -2656,19 +2649,19 @@ leave_nomem:
 		physical += sectorsize;
 	}
 
-	WARN_ON(sblock->page_count == 0);
-	for (index = 0; index < sblock->page_count; index++) {
-		struct scrub_page *spage = sblock->pagev[index];
+	WARN_ON(sblock->sector_count == 0);
+	for (index = 0; index < sblock->sector_count; index++) {
+		struct scrub_sector *sector = sblock->sectors[index];
 		int ret;
 
-		ret = scrub_add_page_to_rd_bio(sctx, spage);
+		ret = scrub_add_sector_to_rd_bio(sctx, sector);
 		if (ret) {
 			scrub_block_put(sblock);
 			return ret;
 		}
 	}
 
-	/* last one frees, either here or in bio completion for last page */
+	/* Last one frees, either here or in bio completion for last sector */
 	scrub_block_put(sblock);
 	return 0;
 }
@@ -2707,7 +2700,7 @@ static int scrub_extent_for_parity(struct scrub_parity *sparity,
 			if (have_csum == 0)
 				goto skip;
 		}
-		ret = scrub_pages_for_parity(sparity, logical, l, physical, dev,
+		ret = scrub_sectors_for_parity(sparity, logical, l, physical, dev,
 					     flags, gen, mirror_num,
 					     have_csum ? csum : NULL);
 		if (ret)
@@ -2767,7 +2760,7 @@ static int get_raid56_logic_offset(u64 physical, int num,
 static void scrub_free_parity(struct scrub_parity *sparity)
 {
 	struct scrub_ctx *sctx = sparity->sctx;
-	struct scrub_page *curr, *next;
+	struct scrub_sector *curr, *next;
 	int nbits;
 
 	nbits = bitmap_weight(sparity->ebitmap, sparity->nsectors);
@@ -2778,15 +2771,15 @@ static void scrub_free_parity(struct scrub_parity *sparity)
 		spin_unlock(&sctx->stat_lock);
 	}
 
-	list_for_each_entry_safe(curr, next, &sparity->spages, list) {
+	list_for_each_entry_safe(curr, next, &sparity->sectors_list, list) {
 		list_del_init(&curr->list);
-		scrub_page_put(curr);
+		scrub_sector_put(curr);
 	}
 
 	kfree(sparity);
 }
 
-static void scrub_parity_bio_endio_worker(struct btrfs_work *work)
+static void scrub_parity_bio_endio_worker(struct work_struct *work)
 {
 	struct scrub_parity *sparity = container_of(work, struct scrub_parity,
 						    work);
@@ -2798,7 +2791,7 @@ static void scrub_parity_bio_endio_worker(struct btrfs_work *work)
 
 static void scrub_parity_bio_endio(struct bio *bio)
 {
-	struct scrub_parity *sparity = (struct scrub_parity *)bio->bi_private;
+	struct scrub_parity *sparity = bio->bi_private;
 	struct btrfs_fs_info *fs_info = sparity->sctx->fs_info;
 
 	if (bio->bi_status)
@@ -2807,9 +2800,8 @@ static void scrub_parity_bio_endio(struct bio *bio)
 
 	bio_put(bio);
 
-	btrfs_init_work(&sparity->work, scrub_parity_bio_endio_worker, NULL,
-			NULL);
-	btrfs_queue_work(fs_info->scrub_parity_workers, &sparity->work);
+	INIT_WORK(&sparity->work, scrub_parity_bio_endio_worker);
+	queue_work(fs_info->scrub_parity_workers, &sparity->work);
 }
 
 static void scrub_parity_check_and_repair(struct scrub_parity *sparity)
@@ -2834,7 +2826,7 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity)
 	if (ret || !bioc || !bioc->raid_map)
 		goto bioc_out;
 
-	bio = btrfs_bio_alloc(BIO_MAX_VECS);
+	bio = bio_alloc(NULL, BIO_MAX_VECS, REQ_OP_READ, GFP_NOFS);
 	bio->bi_iter.bi_sector = sparity->logic_start >> 9;
 	bio->bi_private = sparity;
 	bio->bi_end_io = scrub_parity_bio_endio;
@@ -2882,6 +2874,251 @@ static void scrub_parity_put(struct scrub_parity *sparity)
 	scrub_parity_check_and_repair(sparity);
 }
 
+/*
+ * Return 0 if the extent item range covers any byte of the range.
+ * Return <0 if the extent item is before @search_start.
+ * Return >0 if the extent item is after @start_start + @search_len.
+ */
+static int compare_extent_item_range(struct btrfs_path *path,
+				     u64 search_start, u64 search_len)
+{
+	struct btrfs_fs_info *fs_info = path->nodes[0]->fs_info;
+	u64 len;
+	struct btrfs_key key;
+
+	btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+	ASSERT(key.type == BTRFS_EXTENT_ITEM_KEY ||
+	       key.type == BTRFS_METADATA_ITEM_KEY);
+	if (key.type == BTRFS_METADATA_ITEM_KEY)
+		len = fs_info->nodesize;
+	else
+		len = key.offset;
+
+	if (key.objectid + len <= search_start)
+		return -1;
+	if (key.objectid >= search_start + search_len)
+		return 1;
+	return 0;
+}
+
+/*
+ * Locate one extent item which covers any byte in range
+ * [@search_start, @search_start + @search_length)
+ *
+ * If the path is not initialized, we will initialize the search by doing
+ * a btrfs_search_slot().
+ * If the path is already initialized, we will use the path as the initial
+ * slot, to avoid duplicated btrfs_search_slot() calls.
+ *
+ * NOTE: If an extent item starts before @search_start, we will still
+ * return the extent item. This is for data extent crossing stripe boundary.
+ *
+ * Return 0 if we found such extent item, and @path will point to the extent item.
+ * Return >0 if no such extent item can be found, and @path will be released.
+ * Return <0 if hit fatal error, and @path will be released.
+ */
+static int find_first_extent_item(struct btrfs_root *extent_root,
+				  struct btrfs_path *path,
+				  u64 search_start, u64 search_len)
+{
+	struct btrfs_fs_info *fs_info = extent_root->fs_info;
+	struct btrfs_key key;
+	int ret;
+
+	/* Continue using the existing path */
+	if (path->nodes[0])
+		goto search_forward;
+
+	if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
+		key.type = BTRFS_METADATA_ITEM_KEY;
+	else
+		key.type = BTRFS_EXTENT_ITEM_KEY;
+	key.objectid = search_start;
+	key.offset = (u64)-1;
+
+	ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0);
+	if (ret < 0)
+		return ret;
+
+	ASSERT(ret > 0);
+	/*
+	 * Here we intentionally pass 0 as @min_objectid, as there could be
+	 * an extent item starting before @search_start.
+	 */
+	ret = btrfs_previous_extent_item(extent_root, path, 0);
+	if (ret < 0)
+		return ret;
+	/*
+	 * No matter whether we have found an extent item, the next loop will
+	 * properly do every check on the key.
+	 */
+search_forward:
+	while (true) {
+		btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+		if (key.objectid >= search_start + search_len)
+			break;
+		if (key.type != BTRFS_METADATA_ITEM_KEY &&
+		    key.type != BTRFS_EXTENT_ITEM_KEY)
+			goto next;
+
+		ret = compare_extent_item_range(path, search_start, search_len);
+		if (ret == 0)
+			return ret;
+		if (ret > 0)
+			break;
+next:
+		path->slots[0]++;
+		if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
+			ret = btrfs_next_leaf(extent_root, path);
+			if (ret) {
+				/* Either no more item or fatal error */
+				btrfs_release_path(path);
+				return ret;
+			}
+		}
+	}
+	btrfs_release_path(path);
+	return 1;
+}
+
+static void get_extent_info(struct btrfs_path *path, u64 *extent_start_ret,
+			    u64 *size_ret, u64 *flags_ret, u64 *generation_ret)
+{
+	struct btrfs_key key;
+	struct btrfs_extent_item *ei;
+
+	btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]);
+	ASSERT(key.type == BTRFS_METADATA_ITEM_KEY ||
+	       key.type == BTRFS_EXTENT_ITEM_KEY);
+	*extent_start_ret = key.objectid;
+	if (key.type == BTRFS_METADATA_ITEM_KEY)
+		*size_ret = path->nodes[0]->fs_info->nodesize;
+	else
+		*size_ret = key.offset;
+	ei = btrfs_item_ptr(path->nodes[0], path->slots[0], struct btrfs_extent_item);
+	*flags_ret = btrfs_extent_flags(path->nodes[0], ei);
+	*generation_ret = btrfs_extent_generation(path->nodes[0], ei);
+}
+
+static bool does_range_cross_boundary(u64 extent_start, u64 extent_len,
+				      u64 boundary_start, u64 boudary_len)
+{
+	return (extent_start < boundary_start &&
+		extent_start + extent_len > boundary_start) ||
+	       (extent_start < boundary_start + boudary_len &&
+		extent_start + extent_len > boundary_start + boudary_len);
+}
+
+static int scrub_raid56_data_stripe_for_parity(struct scrub_ctx *sctx,
+					       struct scrub_parity *sparity,
+					       struct map_lookup *map,
+					       struct btrfs_device *sdev,
+					       struct btrfs_path *path,
+					       u64 logical)
+{
+	struct btrfs_fs_info *fs_info = sctx->fs_info;
+	struct btrfs_root *extent_root = btrfs_extent_root(fs_info, logical);
+	struct btrfs_root *csum_root = btrfs_csum_root(fs_info, logical);
+	u64 cur_logical = logical;
+	int ret;
+
+	ASSERT(map->type & BTRFS_BLOCK_GROUP_RAID56_MASK);
+
+	/* Path must not be populated */
+	ASSERT(!path->nodes[0]);
+
+	while (cur_logical < logical + map->stripe_len) {
+		struct btrfs_io_context *bioc = NULL;
+		struct btrfs_device *extent_dev;
+		u64 extent_start;
+		u64 extent_size;
+		u64 mapped_length;
+		u64 extent_flags;
+		u64 extent_gen;
+		u64 extent_physical;
+		u64 extent_mirror_num;
+
+		ret = find_first_extent_item(extent_root, path, cur_logical,
+					     logical + map->stripe_len - cur_logical);
+		/* No more extent item in this data stripe */
+		if (ret > 0) {
+			ret = 0;
+			break;
+		}
+		if (ret < 0)
+			break;
+		get_extent_info(path, &extent_start, &extent_size, &extent_flags,
+				&extent_gen);
+
+		/* Metadata should not cross stripe boundaries */
+		if ((extent_flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) &&
+		    does_range_cross_boundary(extent_start, extent_size,
+					      logical, map->stripe_len)) {
+			btrfs_err(fs_info,
+	"scrub: tree block %llu spanning stripes, ignored. logical=%llu",
+				  extent_start, logical);
+			spin_lock(&sctx->stat_lock);
+			sctx->stat.uncorrectable_errors++;
+			spin_unlock(&sctx->stat_lock);
+			cur_logical += extent_size;
+			continue;
+		}
+
+		/* Skip hole range which doesn't have any extent */
+		cur_logical = max(extent_start, cur_logical);
+
+		/* Truncate the range inside this data stripe */
+		extent_size = min(extent_start + extent_size,
+				  logical + map->stripe_len) - cur_logical;
+		extent_start = cur_logical;
+		ASSERT(extent_size <= U32_MAX);
+
+		scrub_parity_mark_sectors_data(sparity, extent_start, extent_size);
+
+		mapped_length = extent_size;
+		ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, extent_start,
+				      &mapped_length, &bioc, 0);
+		if (!ret && (!bioc || mapped_length < extent_size))
+			ret = -EIO;
+		if (ret) {
+			btrfs_put_bioc(bioc);
+			scrub_parity_mark_sectors_error(sparity, extent_start,
+							extent_size);
+			break;
+		}
+		extent_physical = bioc->stripes[0].physical;
+		extent_mirror_num = bioc->mirror_num;
+		extent_dev = bioc->stripes[0].dev;
+		btrfs_put_bioc(bioc);
+
+		ret = btrfs_lookup_csums_range(csum_root, extent_start,
+					       extent_start + extent_size - 1,
+					       &sctx->csum_list, 1);
+		if (ret) {
+			scrub_parity_mark_sectors_error(sparity, extent_start,
+							extent_size);
+			break;
+		}
+
+		ret = scrub_extent_for_parity(sparity, extent_start,
+					      extent_size, extent_physical,
+					      extent_dev, extent_flags,
+					      extent_gen, extent_mirror_num);
+		scrub_free_csums(sctx);
+
+		if (ret) {
+			scrub_parity_mark_sectors_error(sparity, extent_start,
+							extent_size);
+			break;
+		}
+
+		cond_resched();
+		cur_logical += extent_size;
+	}
+	btrfs_release_path(path);
+	return ret;
+}
+
 static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx,
 						  struct map_lookup *map,
 						  struct btrfs_device *sdev,
@@ -2889,28 +3126,12 @@ static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx,
 						  u64 logic_end)
 {
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
-	struct btrfs_root *root = btrfs_extent_root(fs_info, logic_start);
-	struct btrfs_root *csum_root;
-	struct btrfs_extent_item *extent;
-	struct btrfs_io_context *bioc = NULL;
 	struct btrfs_path *path;
-	u64 flags;
+	u64 cur_logical;
 	int ret;
-	int slot;
-	struct extent_buffer *l;
-	struct btrfs_key key;
-	u64 generation;
-	u64 extent_logical;
-	u64 extent_physical;
-	/* Check the comment in scrub_stripe() for why u32 is enough here */
-	u32 extent_len;
-	u64 mapped_length;
-	struct btrfs_device *extent_dev;
 	struct scrub_parity *sparity;
 	int nsectors;
 	int bitmap_len;
-	int extent_mirror_num;
-	int stop_loop = 0;
 
 	path = btrfs_alloc_path();
 	if (!path) {
@@ -2943,178 +3164,19 @@ static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx,
 	sparity->logic_start = logic_start;
 	sparity->logic_end = logic_end;
 	refcount_set(&sparity->refs, 1);
-	INIT_LIST_HEAD(&sparity->spages);
+	INIT_LIST_HEAD(&sparity->sectors_list);
 	sparity->dbitmap = sparity->bitmap;
 	sparity->ebitmap = (void *)sparity->bitmap + bitmap_len;
 
 	ret = 0;
-	while (logic_start < logic_end) {
-		if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
-			key.type = BTRFS_METADATA_ITEM_KEY;
-		else
-			key.type = BTRFS_EXTENT_ITEM_KEY;
-		key.objectid = logic_start;
-		key.offset = (u64)-1;
-
-		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+	for (cur_logical = logic_start; cur_logical < logic_end;
+	     cur_logical += map->stripe_len) {
+		ret = scrub_raid56_data_stripe_for_parity(sctx, sparity, map,
+							  sdev, path, cur_logical);
 		if (ret < 0)
-			goto out;
-
-		if (ret > 0) {
-			ret = btrfs_previous_extent_item(root, path, 0);
-			if (ret < 0)
-				goto out;
-			if (ret > 0) {
-				btrfs_release_path(path);
-				ret = btrfs_search_slot(NULL, root, &key,
-							path, 0, 0);
-				if (ret < 0)
-					goto out;
-			}
-		}
-
-		stop_loop = 0;
-		while (1) {
-			u64 bytes;
-
-			l = path->nodes[0];
-			slot = path->slots[0];
-			if (slot >= btrfs_header_nritems(l)) {
-				ret = btrfs_next_leaf(root, path);
-				if (ret == 0)
-					continue;
-				if (ret < 0)
-					goto out;
-
-				stop_loop = 1;
-				break;
-			}
-			btrfs_item_key_to_cpu(l, &key, slot);
-
-			if (key.type != BTRFS_EXTENT_ITEM_KEY &&
-			    key.type != BTRFS_METADATA_ITEM_KEY)
-				goto next;
-
-			if (key.type == BTRFS_METADATA_ITEM_KEY)
-				bytes = fs_info->nodesize;
-			else
-				bytes = key.offset;
-
-			if (key.objectid + bytes <= logic_start)
-				goto next;
-
-			if (key.objectid >= logic_end) {
-				stop_loop = 1;
-				break;
-			}
-
-			while (key.objectid >= logic_start + map->stripe_len)
-				logic_start += map->stripe_len;
-
-			extent = btrfs_item_ptr(l, slot,
-						struct btrfs_extent_item);
-			flags = btrfs_extent_flags(l, extent);
-			generation = btrfs_extent_generation(l, extent);
-
-			if ((flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) &&
-			    (key.objectid < logic_start ||
-			     key.objectid + bytes >
-			     logic_start + map->stripe_len)) {
-				btrfs_err(fs_info,
-					  "scrub: tree block %llu spanning stripes, ignored. logical=%llu",
-					  key.objectid, logic_start);
-				spin_lock(&sctx->stat_lock);
-				sctx->stat.uncorrectable_errors++;
-				spin_unlock(&sctx->stat_lock);
-				goto next;
-			}
-again:
-			extent_logical = key.objectid;
-			ASSERT(bytes <= U32_MAX);
-			extent_len = bytes;
-
-			if (extent_logical < logic_start) {
-				extent_len -= logic_start - extent_logical;
-				extent_logical = logic_start;
-			}
-
-			if (extent_logical + extent_len >
-			    logic_start + map->stripe_len)
-				extent_len = logic_start + map->stripe_len -
-					     extent_logical;
-
-			scrub_parity_mark_sectors_data(sparity, extent_logical,
-						       extent_len);
-
-			mapped_length = extent_len;
-			bioc = NULL;
-			ret = btrfs_map_block(fs_info, BTRFS_MAP_READ,
-					extent_logical, &mapped_length, &bioc,
-					0);
-			if (!ret) {
-				if (!bioc || mapped_length < extent_len)
-					ret = -EIO;
-			}
-			if (ret) {
-				btrfs_put_bioc(bioc);
-				goto out;
-			}
-			extent_physical = bioc->stripes[0].physical;
-			extent_mirror_num = bioc->mirror_num;
-			extent_dev = bioc->stripes[0].dev;
-			btrfs_put_bioc(bioc);
-
-			csum_root = btrfs_csum_root(fs_info, extent_logical);
-			ret = btrfs_lookup_csums_range(csum_root,
-						extent_logical,
-						extent_logical + extent_len - 1,
-						&sctx->csum_list, 1);
-			if (ret)
-				goto out;
-
-			ret = scrub_extent_for_parity(sparity, extent_logical,
-						      extent_len,
-						      extent_physical,
-						      extent_dev, flags,
-						      generation,
-						      extent_mirror_num);
-
-			scrub_free_csums(sctx);
-
-			if (ret)
-				goto out;
-
-			if (extent_logical + extent_len <
-			    key.objectid + bytes) {
-				logic_start += map->stripe_len;
-
-				if (logic_start >= logic_end) {
-					stop_loop = 1;
-					break;
-				}
-
-				if (logic_start < key.objectid + bytes) {
-					cond_resched();
-					goto again;
-				}
-			}
-next:
-			path->slots[0]++;
-		}
-
-		btrfs_release_path(path);
-
-		if (stop_loop)
 			break;
-
-		logic_start += map->stripe_len;
-	}
-out:
-	if (ret < 0) {
-		ASSERT(logic_end - logic_start <= U32_MAX);
-		scrub_parity_mark_sectors_error(sparity, logic_start,
-						logic_end - logic_start);
 	}
+
 	scrub_parity_put(sparity);
 	scrub_submit(sctx);
 	mutex_lock(&sctx->wr_lock);
@@ -3165,6 +3227,206 @@ static int sync_write_pointer_for_zoned(struct scrub_ctx *sctx, u64 logical,
 	return ret;
 }
 
+/*
+ * Scrub one range which can only has simple mirror based profile.
+ * (Including all range in SINGLE/DUP/RAID1/RAID1C*, and each stripe in
+ *  RAID0/RAID10).
+ *
+ * Since we may need to handle a subset of block group, we need @logical_start
+ * and @logical_length parameter.
+ */
+static int scrub_simple_mirror(struct scrub_ctx *sctx,
+			       struct btrfs_root *extent_root,
+			       struct btrfs_root *csum_root,
+			       struct btrfs_block_group *bg,
+			       struct map_lookup *map,
+			       u64 logical_start, u64 logical_length,
+			       struct btrfs_device *device,
+			       u64 physical, int mirror_num)
+{
+	struct btrfs_fs_info *fs_info = sctx->fs_info;
+	const u64 logical_end = logical_start + logical_length;
+	/* An artificial limit, inherit from old scrub behavior */
+	const u32 max_length = SZ_64K;
+	struct btrfs_path path = { 0 };
+	u64 cur_logical = logical_start;
+	int ret;
+
+	/* The range must be inside the bg */
+	ASSERT(logical_start >= bg->start && logical_end <= bg->start + bg->length);
+
+	path.search_commit_root = 1;
+	path.skip_locking = 1;
+	/* Go through each extent items inside the logical range */
+	while (cur_logical < logical_end) {
+		u64 extent_start;
+		u64 extent_len;
+		u64 extent_flags;
+		u64 extent_gen;
+		u64 scrub_len;
+
+		/* Canceled? */
+		if (atomic_read(&fs_info->scrub_cancel_req) ||
+		    atomic_read(&sctx->cancel_req)) {
+			ret = -ECANCELED;
+			break;
+		}
+		/* Paused? */
+		if (atomic_read(&fs_info->scrub_pause_req)) {
+			/* Push queued extents */
+			sctx->flush_all_writes = true;
+			scrub_submit(sctx);
+			mutex_lock(&sctx->wr_lock);
+			scrub_wr_submit(sctx);
+			mutex_unlock(&sctx->wr_lock);
+			wait_event(sctx->list_wait,
+				   atomic_read(&sctx->bios_in_flight) == 0);
+			sctx->flush_all_writes = false;
+			scrub_blocked_if_needed(fs_info);
+		}
+		/* Block group removed? */
+		spin_lock(&bg->lock);
+		if (bg->removed) {
+			spin_unlock(&bg->lock);
+			ret = 0;
+			break;
+		}
+		spin_unlock(&bg->lock);
+
+		ret = find_first_extent_item(extent_root, &path, cur_logical,
+					     logical_end - cur_logical);
+		if (ret > 0) {
+			/* No more extent, just update the accounting */
+			sctx->stat.last_physical = physical + logical_length;
+			ret = 0;
+			break;
+		}
+		if (ret < 0)
+			break;
+		get_extent_info(&path, &extent_start, &extent_len,
+				&extent_flags, &extent_gen);
+		/* Skip hole range which doesn't have any extent */
+		cur_logical = max(extent_start, cur_logical);
+
+		/*
+		 * Scrub len has three limits:
+		 * - Extent size limit
+		 * - Scrub range limit
+		 *   This is especially imporatant for RAID0/RAID10 to reuse
+		 *   this function
+		 * - Max scrub size limit
+		 */
+		scrub_len = min(min(extent_start + extent_len,
+				    logical_end), cur_logical + max_length) -
+			    cur_logical;
+
+		if (extent_flags & BTRFS_EXTENT_FLAG_DATA) {
+			ret = btrfs_lookup_csums_range(csum_root, cur_logical,
+					cur_logical + scrub_len - 1,
+					&sctx->csum_list, 1);
+			if (ret)
+				break;
+		}
+		if ((extent_flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) &&
+		    does_range_cross_boundary(extent_start, extent_len,
+					      logical_start, logical_length)) {
+			btrfs_err(fs_info,
+"scrub: tree block %llu spanning boundaries, ignored. boundary=[%llu, %llu)",
+				  extent_start, logical_start, logical_end);
+			spin_lock(&sctx->stat_lock);
+			sctx->stat.uncorrectable_errors++;
+			spin_unlock(&sctx->stat_lock);
+			cur_logical += scrub_len;
+			continue;
+		}
+		ret = scrub_extent(sctx, map, cur_logical, scrub_len,
+				   cur_logical - logical_start + physical,
+				   device, extent_flags, extent_gen,
+				   mirror_num);
+		scrub_free_csums(sctx);
+		if (ret)
+			break;
+		if (sctx->is_dev_replace)
+			sync_replace_for_zoned(sctx);
+		cur_logical += scrub_len;
+		/* Don't hold CPU for too long time */
+		cond_resched();
+	}
+	btrfs_release_path(&path);
+	return ret;
+}
+
+/* Calculate the full stripe length for simple stripe based profiles */
+static u64 simple_stripe_full_stripe_len(const struct map_lookup *map)
+{
+	ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 |
+			    BTRFS_BLOCK_GROUP_RAID10));
+
+	return map->num_stripes / map->sub_stripes * map->stripe_len;
+}
+
+/* Get the logical bytenr for the stripe */
+static u64 simple_stripe_get_logical(struct map_lookup *map,
+				     struct btrfs_block_group *bg,
+				     int stripe_index)
+{
+	ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 |
+			    BTRFS_BLOCK_GROUP_RAID10));
+	ASSERT(stripe_index < map->num_stripes);
+
+	/*
+	 * (stripe_index / sub_stripes) gives how many data stripes we need to
+	 * skip.
+	 */
+	return (stripe_index / map->sub_stripes) * map->stripe_len + bg->start;
+}
+
+/* Get the mirror number for the stripe */
+static int simple_stripe_mirror_num(struct map_lookup *map, int stripe_index)
+{
+	ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 |
+			    BTRFS_BLOCK_GROUP_RAID10));
+	ASSERT(stripe_index < map->num_stripes);
+
+	/* For RAID0, it's fixed to 1, for RAID10 it's 0,1,0,1... */
+	return stripe_index % map->sub_stripes + 1;
+}
+
+static int scrub_simple_stripe(struct scrub_ctx *sctx,
+			       struct btrfs_root *extent_root,
+			       struct btrfs_root *csum_root,
+			       struct btrfs_block_group *bg,
+			       struct map_lookup *map,
+			       struct btrfs_device *device,
+			       int stripe_index)
+{
+	const u64 logical_increment = simple_stripe_full_stripe_len(map);
+	const u64 orig_logical = simple_stripe_get_logical(map, bg, stripe_index);
+	const u64 orig_physical = map->stripes[stripe_index].physical;
+	const int mirror_num = simple_stripe_mirror_num(map, stripe_index);
+	u64 cur_logical = orig_logical;
+	u64 cur_physical = orig_physical;
+	int ret = 0;
+
+	while (cur_logical < bg->start + bg->length) {
+		/*
+		 * Inside each stripe, RAID0 is just SINGLE, and RAID10 is
+		 * just RAID1, so we can reuse scrub_simple_mirror() to scrub
+		 * this stripe.
+		 */
+		ret = scrub_simple_mirror(sctx, extent_root, csum_root, bg, map,
+					  cur_logical, map->stripe_len, device,
+					  cur_physical, mirror_num);
+		if (ret)
+			return ret;
+		/* Skip to next stripe which belongs to the target device */
+		cur_logical += logical_increment;
+		/* For physical offset, we just go to next stripe */
+		cur_physical += map->stripe_len;
+	}
+	return ret;
+}
+
 static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 					   struct btrfs_block_group *bg,
 					   struct map_lookup *map,
@@ -3175,59 +3437,22 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	struct btrfs_fs_info *fs_info = sctx->fs_info;
 	struct btrfs_root *root;
 	struct btrfs_root *csum_root;
-	struct btrfs_extent_item *extent;
 	struct blk_plug plug;
+	const u64 profile = map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK;
 	const u64 chunk_logical = bg->start;
-	u64 flags;
 	int ret;
-	int slot;
-	u64 nstripes;
-	struct extent_buffer *l;
-	u64 physical;
+	u64 physical = map->stripes[stripe_index].physical;
+	const u64 physical_end = physical + dev_extent_len;
 	u64 logical;
 	u64 logic_end;
-	u64 physical_end;
-	u64 generation;
-	int mirror_num;
-	struct btrfs_key key;
+	/* The logical increment after finishing one stripe */
 	u64 increment;
+	/* Offset inside the chunk */
 	u64 offset;
-	u64 extent_logical;
-	u64 extent_physical;
-	/*
-	 * Unlike chunk length, extent length should never go beyond
-	 * BTRFS_MAX_EXTENT_SIZE, thus u32 is enough here.
-	 */
-	u32 extent_len;
 	u64 stripe_logical;
 	u64 stripe_end;
-	struct btrfs_device *extent_dev;
-	int extent_mirror_num;
 	int stop_loop = 0;
 
-	physical = map->stripes[stripe_index].physical;
-	offset = 0;
-	nstripes = div64_u64(dev_extent_len, map->stripe_len);
-	mirror_num = 1;
-	increment = map->stripe_len;
-	if (map->type & BTRFS_BLOCK_GROUP_RAID0) {
-		offset = map->stripe_len * stripe_index;
-		increment = map->stripe_len * map->num_stripes;
-	} else if (map->type & BTRFS_BLOCK_GROUP_RAID10) {
-		int factor = map->num_stripes / map->sub_stripes;
-		offset = map->stripe_len * (stripe_index / map->sub_stripes);
-		increment = map->stripe_len * factor;
-		mirror_num = stripe_index % map->sub_stripes + 1;
-	} else if (map->type & BTRFS_BLOCK_GROUP_RAID1_MASK) {
-		mirror_num = stripe_index % map->num_stripes + 1;
-	} else if (map->type & BTRFS_BLOCK_GROUP_DUP) {
-		mirror_num = stripe_index % map->num_stripes + 1;
-	} else if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-		get_raid56_logic_offset(physical, stripe_index, map, &offset,
-					NULL);
-		increment = map->stripe_len * nr_data_stripes(map);
-	}
-
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
@@ -3241,21 +3466,12 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	path->skip_locking = 1;
 	path->reada = READA_FORWARD;
 
-	logical = chunk_logical + offset;
-	physical_end = physical + nstripes * map->stripe_len;
-	if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-		get_raid56_logic_offset(physical_end, stripe_index,
-					map, &logic_end, NULL);
-		logic_end += chunk_logical;
-	} else {
-		logic_end = logical + increment * nstripes;
-	}
 	wait_event(sctx->list_wait,
 		   atomic_read(&sctx->bios_in_flight) == 0);
 	scrub_blocked_if_needed(fs_info);
 
-	root = btrfs_extent_root(fs_info, logical);
-	csum_root = btrfs_csum_root(fs_info, logical);
+	root = btrfs_extent_root(fs_info, bg->start);
+	csum_root = btrfs_csum_root(fs_info, bg->start);
 
 	/*
 	 * collect all data csums for the stripe to avoid seeking during
@@ -3272,241 +3488,83 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx,
 	}
 
 	/*
-	 * now find all extents for each stripe and scrub them
+	 * There used to be a big double loop to handle all profiles using the
+	 * same routine, which grows larger and more gross over time.
+	 *
+	 * So here we handle each profile differently, so simpler profiles
+	 * have simpler scrubbing function.
 	 */
-	ret = 0;
-	while (physical < physical_end) {
-		/*
-		 * canceled?
-		 */
-		if (atomic_read(&fs_info->scrub_cancel_req) ||
-		    atomic_read(&sctx->cancel_req)) {
-			ret = -ECANCELED;
-			goto out;
-		}
+	if (!(profile & (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID10 |
+			 BTRFS_BLOCK_GROUP_RAID56_MASK))) {
 		/*
-		 * check to see if we have to pause
+		 * Above check rules out all complex profile, the remaining
+		 * profiles are SINGLE|DUP|RAID1|RAID1C*, which is simple
+		 * mirrored duplication without stripe.
+		 *
+		 * Only @physical and @mirror_num needs to calculated using
+		 * @stripe_index.
 		 */
-		if (atomic_read(&fs_info->scrub_pause_req)) {
-			/* push queued extents */
-			sctx->flush_all_writes = true;
-			scrub_submit(sctx);
-			mutex_lock(&sctx->wr_lock);
-			scrub_wr_submit(sctx);
-			mutex_unlock(&sctx->wr_lock);
-			wait_event(sctx->list_wait,
-				   atomic_read(&sctx->bios_in_flight) == 0);
-			sctx->flush_all_writes = false;
-			scrub_blocked_if_needed(fs_info);
-		}
-
-		if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-			ret = get_raid56_logic_offset(physical, stripe_index,
-						      map, &logical,
-						      &stripe_logical);
-			logical += chunk_logical;
-			if (ret) {
-				/* it is parity strip */
-				stripe_logical += chunk_logical;
-				stripe_end = stripe_logical + increment;
-				ret = scrub_raid56_parity(sctx, map, scrub_dev,
-							  stripe_logical,
-							  stripe_end);
-				if (ret)
-					goto out;
-				goto skip;
-			}
-		}
-
-		if (btrfs_fs_incompat(fs_info, SKINNY_METADATA))
-			key.type = BTRFS_METADATA_ITEM_KEY;
-		else
-			key.type = BTRFS_EXTENT_ITEM_KEY;
-		key.objectid = logical;
-		key.offset = (u64)-1;
-
-		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-		if (ret < 0)
-			goto out;
-
-		if (ret > 0) {
-			ret = btrfs_previous_extent_item(root, path, 0);
-			if (ret < 0)
-				goto out;
-			if (ret > 0) {
-				/* there's no smaller item, so stick with the
-				 * larger one */
-				btrfs_release_path(path);
-				ret = btrfs_search_slot(NULL, root, &key,
-							path, 0, 0);
-				if (ret < 0)
-					goto out;
-			}
-		}
-
-		stop_loop = 0;
-		while (1) {
-			u64 bytes;
-
-			l = path->nodes[0];
-			slot = path->slots[0];
-			if (slot >= btrfs_header_nritems(l)) {
-				ret = btrfs_next_leaf(root, path);
-				if (ret == 0)
-					continue;
-				if (ret < 0)
-					goto out;
-
-				stop_loop = 1;
-				break;
-			}
-			btrfs_item_key_to_cpu(l, &key, slot);
-
-			if (key.type != BTRFS_EXTENT_ITEM_KEY &&
-			    key.type != BTRFS_METADATA_ITEM_KEY)
-				goto next;
-
-			if (key.type == BTRFS_METADATA_ITEM_KEY)
-				bytes = fs_info->nodesize;
-			else
-				bytes = key.offset;
-
-			if (key.objectid + bytes <= logical)
-				goto next;
-
-			if (key.objectid >= logical + map->stripe_len) {
-				/* out of this device extent */
-				if (key.objectid >= logic_end)
-					stop_loop = 1;
-				break;
-			}
-
-			/*
-			 * If our block group was removed in the meanwhile, just
-			 * stop scrubbing since there is no point in continuing.
-			 * Continuing would prevent reusing its device extents
-			 * for new block groups for a long time.
-			 */
-			spin_lock(&bg->lock);
-			if (bg->removed) {
-				spin_unlock(&bg->lock);
-				ret = 0;
-				goto out;
-			}
-			spin_unlock(&bg->lock);
-
-			extent = btrfs_item_ptr(l, slot,
-						struct btrfs_extent_item);
-			flags = btrfs_extent_flags(l, extent);
-			generation = btrfs_extent_generation(l, extent);
-
-			if ((flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) &&
-			    (key.objectid < logical ||
-			     key.objectid + bytes >
-			     logical + map->stripe_len)) {
-				btrfs_err(fs_info,
-					   "scrub: tree block %llu spanning stripes, ignored. logical=%llu",
-				       key.objectid, logical);
-				spin_lock(&sctx->stat_lock);
-				sctx->stat.uncorrectable_errors++;
-				spin_unlock(&sctx->stat_lock);
-				goto next;
-			}
-
-again:
-			extent_logical = key.objectid;
-			ASSERT(bytes <= U32_MAX);
-			extent_len = bytes;
-
-			/*
-			 * trim extent to this stripe
-			 */
-			if (extent_logical < logical) {
-				extent_len -= logical - extent_logical;
-				extent_logical = logical;
-			}
-			if (extent_logical + extent_len >
-			    logical + map->stripe_len) {
-				extent_len = logical + map->stripe_len -
-					     extent_logical;
-			}
+		ret = scrub_simple_mirror(sctx, root, csum_root, bg, map,
+				bg->start, bg->length, scrub_dev,
+				map->stripes[stripe_index].physical,
+				stripe_index + 1);
+		offset = 0;
+		goto out;
+	}
+	if (profile & (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID10)) {
+		ret = scrub_simple_stripe(sctx, root, csum_root, bg, map,
+					  scrub_dev, stripe_index);
+		offset = map->stripe_len * (stripe_index / map->sub_stripes);
+		goto out;
+	}
 
-			extent_physical = extent_logical - logical + physical;
-			extent_dev = scrub_dev;
-			extent_mirror_num = mirror_num;
-			if (sctx->is_dev_replace)
-				scrub_remap_extent(fs_info, extent_logical,
-						   extent_len, &extent_physical,
-						   &extent_dev,
-						   &extent_mirror_num);
-
-			if (flags & BTRFS_EXTENT_FLAG_DATA) {
-				ret = btrfs_lookup_csums_range(csum_root,
-						extent_logical,
-						extent_logical + extent_len - 1,
-						&sctx->csum_list, 1);
-				if (ret)
-					goto out;
-			}
+	/* Only RAID56 goes through the old code */
+	ASSERT(map->type & BTRFS_BLOCK_GROUP_RAID56_MASK);
+	ret = 0;
 
-			ret = scrub_extent(sctx, map, extent_logical, extent_len,
-					   extent_physical, extent_dev, flags,
-					   generation, extent_mirror_num,
-					   extent_logical - logical + physical);
+	/* Calculate the logical end of the stripe */
+	get_raid56_logic_offset(physical_end, stripe_index,
+				map, &logic_end, NULL);
+	logic_end += chunk_logical;
 
-			scrub_free_csums(sctx);
+	/* Initialize @offset in case we need to go to out: label */
+	get_raid56_logic_offset(physical, stripe_index, map, &offset, NULL);
+	increment = map->stripe_len * nr_data_stripes(map);
 
+	/*
+	 * Due to the rotation, for RAID56 it's better to iterate each stripe
+	 * using their physical offset.
+	 */
+	while (physical < physical_end) {
+		ret = get_raid56_logic_offset(physical, stripe_index, map,
+					      &logical, &stripe_logical);
+		logical += chunk_logical;
+		if (ret) {
+			/* it is parity strip */
+			stripe_logical += chunk_logical;
+			stripe_end = stripe_logical + increment;
+			ret = scrub_raid56_parity(sctx, map, scrub_dev,
+						  stripe_logical,
+						  stripe_end);
 			if (ret)
 				goto out;
+			goto next;
+		}
 
-			if (sctx->is_dev_replace)
-				sync_replace_for_zoned(sctx);
-
-			if (extent_logical + extent_len <
-			    key.objectid + bytes) {
-				if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-					/*
-					 * loop until we find next data stripe
-					 * or we have finished all stripes.
-					 */
-loop:
-					physical += map->stripe_len;
-					ret = get_raid56_logic_offset(physical,
-							stripe_index, map,
-							&logical, &stripe_logical);
-					logical += chunk_logical;
-
-					if (ret && physical < physical_end) {
-						stripe_logical += chunk_logical;
-						stripe_end = stripe_logical +
-								increment;
-						ret = scrub_raid56_parity(sctx,
-							map, scrub_dev,
-							stripe_logical,
-							stripe_end);
-						if (ret)
-							goto out;
-						goto loop;
-					}
-				} else {
-					physical += map->stripe_len;
-					logical += increment;
-				}
-				if (logical < key.objectid + bytes) {
-					cond_resched();
-					goto again;
-				}
-
-				if (physical >= physical_end) {
-					stop_loop = 1;
-					break;
-				}
-			}
+		/*
+		 * Now we're at a data stripe, scrub each extents in the range.
+		 *
+		 * At this stage, if we ignore the repair part, inside each data
+		 * stripe it is no different than SINGLE profile.
+		 * We can reuse scrub_simple_mirror() here, as the repair part
+		 * is still based on @mirror_num.
+		 */
+		ret = scrub_simple_mirror(sctx, root, csum_root, bg, map,
+					  logical, map->stripe_len,
+					  scrub_dev, physical, 1);
+		if (ret < 0)
+			goto out;
 next:
-			path->slots[0]++;
-		}
-		btrfs_release_path(path);
-skip:
 		logical += increment;
 		physical += map->stripe_len;
 		spin_lock(&sctx->stat_lock);
@@ -3964,9 +4022,9 @@ static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx,
 		if (!btrfs_check_super_location(scrub_dev, bytenr))
 			continue;
 
-		ret = scrub_pages(sctx, bytenr, BTRFS_SUPER_INFO_SIZE, bytenr,
-				  scrub_dev, BTRFS_EXTENT_FLAG_SUPER, gen, i,
-				  NULL, bytenr);
+		ret = scrub_sectors(sctx, bytenr, BTRFS_SUPER_INFO_SIZE, bytenr,
+				    scrub_dev, BTRFS_EXTENT_FLAG_SUPER, gen, i,
+				    NULL, bytenr);
 		if (ret)
 			return ret;
 	}
@@ -3979,22 +4037,23 @@ static void scrub_workers_put(struct btrfs_fs_info *fs_info)
 {
 	if (refcount_dec_and_mutex_lock(&fs_info->scrub_workers_refcnt,
 					&fs_info->scrub_lock)) {
-		struct btrfs_workqueue *scrub_workers = NULL;
-		struct btrfs_workqueue *scrub_wr_comp = NULL;
-		struct btrfs_workqueue *scrub_parity = NULL;
-
-		scrub_workers = fs_info->scrub_workers;
-		scrub_wr_comp = fs_info->scrub_wr_completion_workers;
-		scrub_parity = fs_info->scrub_parity_workers;
+		struct workqueue_struct *scrub_workers = fs_info->scrub_workers;
+		struct workqueue_struct *scrub_wr_comp =
+						fs_info->scrub_wr_completion_workers;
+		struct workqueue_struct *scrub_parity =
+						fs_info->scrub_parity_workers;
 
 		fs_info->scrub_workers = NULL;
 		fs_info->scrub_wr_completion_workers = NULL;
 		fs_info->scrub_parity_workers = NULL;
 		mutex_unlock(&fs_info->scrub_lock);
 
-		btrfs_destroy_workqueue(scrub_workers);
-		btrfs_destroy_workqueue(scrub_wr_comp);
-		btrfs_destroy_workqueue(scrub_parity);
+		if (scrub_workers)
+			destroy_workqueue(scrub_workers);
+		if (scrub_wr_comp)
+			destroy_workqueue(scrub_wr_comp);
+		if (scrub_parity)
+			destroy_workqueue(scrub_parity);
 	}
 }
 
@@ -4004,9 +4063,9 @@ static void scrub_workers_put(struct btrfs_fs_info *fs_info)
 static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 						int is_dev_replace)
 {
-	struct btrfs_workqueue *scrub_workers = NULL;
-	struct btrfs_workqueue *scrub_wr_comp = NULL;
-	struct btrfs_workqueue *scrub_parity = NULL;
+	struct workqueue_struct *scrub_workers = NULL;
+	struct workqueue_struct *scrub_wr_comp = NULL;
+	struct workqueue_struct *scrub_parity = NULL;
 	unsigned int flags = WQ_FREEZABLE | WQ_UNBOUND;
 	int max_active = fs_info->thread_pool_size;
 	int ret = -ENOMEM;
@@ -4014,18 +4073,16 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 	if (refcount_inc_not_zero(&fs_info->scrub_workers_refcnt))
 		return 0;
 
-	scrub_workers = btrfs_alloc_workqueue(fs_info, "scrub", flags,
-					      is_dev_replace ? 1 : max_active, 4);
+	scrub_workers = alloc_workqueue("btrfs-scrub", flags,
+					is_dev_replace ? 1 : max_active);
 	if (!scrub_workers)
 		goto fail_scrub_workers;
 
-	scrub_wr_comp = btrfs_alloc_workqueue(fs_info, "scrubwrc", flags,
-					      max_active, 2);
+	scrub_wr_comp = alloc_workqueue("btrfs-scrubwrc", flags, max_active);
 	if (!scrub_wr_comp)
 		goto fail_scrub_wr_completion_workers;
 
-	scrub_parity = btrfs_alloc_workqueue(fs_info, "scrubparity", flags,
-					     max_active, 2);
+	scrub_parity = alloc_workqueue("btrfs-scrubparity", flags, max_active);
 	if (!scrub_parity)
 		goto fail_scrub_parity_workers;
 
@@ -4046,11 +4103,11 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info,
 	mutex_unlock(&fs_info->scrub_lock);
 
 	ret = 0;
-	btrfs_destroy_workqueue(scrub_parity);
+	destroy_workqueue(scrub_parity);
 fail_scrub_parity_workers:
-	btrfs_destroy_workqueue(scrub_wr_comp);
+	destroy_workqueue(scrub_wr_comp);
 fail_scrub_wr_completion_workers:
-	btrfs_destroy_workqueue(scrub_workers);
+	destroy_workqueue(scrub_workers);
 fail_scrub_workers:
 	return ret;
 }
@@ -4082,18 +4139,16 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	}
 
 	if (fs_info->nodesize >
-	    PAGE_SIZE * SCRUB_MAX_PAGES_PER_BLOCK ||
-	    fs_info->sectorsize > PAGE_SIZE * SCRUB_MAX_PAGES_PER_BLOCK) {
+	    SCRUB_MAX_SECTORS_PER_BLOCK << fs_info->sectorsize_bits ||
+	    fs_info->sectorsize > PAGE_SIZE * SCRUB_MAX_SECTORS_PER_BLOCK) {
 		/*
-		 * would exhaust the array bounds of pagev member in
+		 * Would exhaust the array bounds of sectorv member in
 		 * struct scrub_block
 		 */
 		btrfs_err(fs_info,
-			  "scrub: size assumption nodesize and sectorsize <= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails",
-		       fs_info->nodesize,
-		       SCRUB_MAX_PAGES_PER_BLOCK,
-		       fs_info->sectorsize,
-		       SCRUB_MAX_PAGES_PER_BLOCK);
+"scrub: nodesize and sectorsize <= SCRUB_MAX_SECTORS_PER_BLOCK (%d <= %d && %d <= %d) fails",
+		       fs_info->nodesize, SCRUB_MAX_SECTORS_PER_BLOCK,
+		       fs_info->sectorsize, SCRUB_MAX_SECTORS_PER_BLOCK);
 		return -EINVAL;
 	}
 
@@ -4161,7 +4216,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start,
 	/*
 	 * In order to avoid deadlock with reclaim when there is a transaction
 	 * trying to pause scrub, make sure we use GFP_NOFS for all the
-	 * allocations done at btrfs_scrub_pages() and scrub_pages_for_parity()
+	 * allocations done at btrfs_scrub_sectors() and scrub_sectors_for_parity()
 	 * invoked by our callees. The pausing request is done when the
 	 * transaction commit starts, and it blocks the transaction until scrub
 	 * is paused (done at specific points at scrub_stripe() or right above
@@ -4295,11 +4350,11 @@ int btrfs_scrub_progress(struct btrfs_fs_info *fs_info, u64 devid,
 	return dev ? (sctx ? 0 : -ENOTCONN) : -ENODEV;
 }
 
-static void scrub_remap_extent(struct btrfs_fs_info *fs_info,
-			       u64 extent_logical, u32 extent_len,
-			       u64 *extent_physical,
-			       struct btrfs_device **extent_dev,
-			       int *extent_mirror_num)
+static void scrub_find_good_copy(struct btrfs_fs_info *fs_info,
+				 u64 extent_logical, u32 extent_len,
+				 u64 *extent_physical,
+				 struct btrfs_device **extent_dev,
+				 int *extent_mirror_num)
 {
 	u64 mapped_length;
 	struct btrfs_io_context *bioc = NULL;
diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index 7d1642937274..5a05beabf0c3 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -10,7 +10,6 @@
 #include <linux/mount.h>
 #include <linux/xattr.h>
 #include <linux/posix_acl_xattr.h>
-#include <linux/radix-tree.h>
 #include <linux/vmalloc.h>
 #include <linux/string.h>
 #include <linux/compat.h>
@@ -128,11 +127,18 @@ struct send_ctx {
 	struct list_head new_refs;
 	struct list_head deleted_refs;
 
-	struct radix_tree_root name_cache;
+	struct xarray name_cache;
 	struct list_head name_cache_list;
 	int name_cache_size;
 
+	/*
+	 * The inode we are currently processing. It's not NULL only when we
+	 * need to issue write commands for data extents from this inode.
+	 */
+	struct inode *cur_inode;
 	struct file_ra_state ra;
+	u64 page_cache_clear_start;
+	bool clean_page_cache;
 
 	/*
 	 * We process inodes by their increasing order, so if before an
@@ -262,14 +268,13 @@ struct orphan_dir_info {
 struct name_cache_entry {
 	struct list_head list;
 	/*
-	 * radix_tree has only 32bit entries but we need to handle 64bit inums.
-	 * We use the lower 32bit of the 64bit inum to store it in the tree. If
-	 * more then one inum would fall into the same entry, we use radix_list
-	 * to store the additional entries. radix_list is also used to store
-	 * entries where two entries have the same inum but different
-	 * generations.
+	 * On 32bit kernels, xarray has only 32bit indices, but we need to
+	 * handle 64bit inums. We use the lower 32bit of the 64bit inum to store
+	 * it in the tree. If more than one inum would fall into the same entry,
+	 * we use inum_aliases to store the additional entries. inum_aliases is
+	 * also used to store entries with the same inum but different generations.
 	 */
-	struct list_head radix_list;
+	struct list_head inum_aliases;
 	u64 ino;
 	u64 gen;
 	u64 parent_ino;
@@ -2019,9 +2024,9 @@ out:
 }
 
 /*
- * Insert a name cache entry. On 32bit kernels the radix tree index is 32bit,
+ * Insert a name cache entry. On 32bit kernels the xarray index is 32bit,
  * so we need to do some special handling in case we have clashes. This function
- * takes care of this with the help of name_cache_entry::radix_list.
+ * takes care of this with the help of name_cache_entry::inum_aliases.
  * In case of error, nce is kfreed.
  */
 static int name_cache_insert(struct send_ctx *sctx,
@@ -2030,8 +2035,7 @@ static int name_cache_insert(struct send_ctx *sctx,
 	int ret = 0;
 	struct list_head *nce_head;
 
-	nce_head = radix_tree_lookup(&sctx->name_cache,
-			(unsigned long)nce->ino);
+	nce_head = xa_load(&sctx->name_cache, (unsigned long)nce->ino);
 	if (!nce_head) {
 		nce_head = kmalloc(sizeof(*nce_head), GFP_KERNEL);
 		if (!nce_head) {
@@ -2040,14 +2044,14 @@ static int name_cache_insert(struct send_ctx *sctx,
 		}
 		INIT_LIST_HEAD(nce_head);
 
-		ret = radix_tree_insert(&sctx->name_cache, nce->ino, nce_head);
+		ret = xa_insert(&sctx->name_cache, nce->ino, nce_head, GFP_KERNEL);
 		if (ret < 0) {
 			kfree(nce_head);
 			kfree(nce);
 			return ret;
 		}
 	}
-	list_add_tail(&nce->radix_list, nce_head);
+	list_add_tail(&nce->inum_aliases, nce_head);
 	list_add_tail(&nce->list, &sctx->name_cache_list);
 	sctx->name_cache_size++;
 
@@ -2059,15 +2063,14 @@ static void name_cache_delete(struct send_ctx *sctx,
 {
 	struct list_head *nce_head;
 
-	nce_head = radix_tree_lookup(&sctx->name_cache,
-			(unsigned long)nce->ino);
+	nce_head = xa_load(&sctx->name_cache, (unsigned long)nce->ino);
 	if (!nce_head) {
 		btrfs_err(sctx->send_root->fs_info,
 	      "name_cache_delete lookup failed ino %llu cache size %d, leaking memory",
 			nce->ino, sctx->name_cache_size);
 	}
 
-	list_del(&nce->radix_list);
+	list_del(&nce->inum_aliases);
 	list_del(&nce->list);
 	sctx->name_cache_size--;
 
@@ -2075,7 +2078,7 @@ static void name_cache_delete(struct send_ctx *sctx,
 	 * We may not get to the final release of nce_head if the lookup fails
 	 */
 	if (nce_head && list_empty(nce_head)) {
-		radix_tree_delete(&sctx->name_cache, (unsigned long)nce->ino);
+		xa_erase(&sctx->name_cache, (unsigned long)nce->ino);
 		kfree(nce_head);
 	}
 }
@@ -2086,11 +2089,11 @@ static struct name_cache_entry *name_cache_search(struct send_ctx *sctx,
 	struct list_head *nce_head;
 	struct name_cache_entry *cur;
 
-	nce_head = radix_tree_lookup(&sctx->name_cache, (unsigned long)ino);
+	nce_head = xa_load(&sctx->name_cache, (unsigned long)ino);
 	if (!nce_head)
 		return NULL;
 
-	list_for_each_entry(cur, nce_head, radix_list) {
+	list_for_each_entry(cur, nce_head, inum_aliases) {
 		if (cur->ino == ino && cur->gen == gen)
 			return cur;
 	}
@@ -2675,61 +2678,43 @@ out:
 static int did_create_dir(struct send_ctx *sctx, u64 dir)
 {
 	int ret = 0;
+	int iter_ret = 0;
 	struct btrfs_path *path = NULL;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
 	struct btrfs_key di_key;
-	struct extent_buffer *eb;
 	struct btrfs_dir_item *di;
-	int slot;
 
 	path = alloc_path_for_send();
-	if (!path) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!path)
+		return -ENOMEM;
 
 	key.objectid = dir;
 	key.type = BTRFS_DIR_INDEX_KEY;
 	key.offset = 0;
-	ret = btrfs_search_slot(NULL, sctx->send_root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
 
-	while (1) {
-		eb = path->nodes[0];
-		slot = path->slots[0];
-		if (slot >= btrfs_header_nritems(eb)) {
-			ret = btrfs_next_leaf(sctx->send_root, path);
-			if (ret < 0) {
-				goto out;
-			} else if (ret > 0) {
-				ret = 0;
-				break;
-			}
-			continue;
-		}
+	btrfs_for_each_slot(sctx->send_root, &key, &found_key, path, iter_ret) {
+		struct extent_buffer *eb = path->nodes[0];
 
-		btrfs_item_key_to_cpu(eb, &found_key, slot);
 		if (found_key.objectid != key.objectid ||
 		    found_key.type != key.type) {
 			ret = 0;
-			goto out;
+			break;
 		}
 
-		di = btrfs_item_ptr(eb, slot, struct btrfs_dir_item);
+		di = btrfs_item_ptr(eb, path->slots[0], struct btrfs_dir_item);
 		btrfs_dir_item_key_to_cpu(eb, di, &di_key);
 
 		if (di_key.type != BTRFS_ROOT_ITEM_KEY &&
 		    di_key.objectid < sctx->send_progress) {
 			ret = 1;
-			goto out;
+			break;
 		}
-
-		path->slots[0]++;
 	}
+	/* Catch error found during iteration */
+	if (iter_ret < 0)
+		ret = iter_ret;
 
-out:
 	btrfs_free_path(path);
 	return ret;
 }
@@ -2933,6 +2918,7 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 		     u64 send_progress)
 {
 	int ret = 0;
+	int iter_ret = 0;
 	struct btrfs_root *root = sctx->parent_root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
@@ -2959,23 +2945,9 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 	if (odi)
 		key.offset = odi->last_dir_index_offset;
 
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-
-	while (1) {
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
 		struct waiting_dir_move *dm;
 
-		if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto out;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-		btrfs_item_key_to_cpu(path->nodes[0], &found_key,
-				      path->slots[0]);
 		if (found_key.objectid != key.objectid ||
 		    found_key.type != key.type)
 			break;
@@ -3010,8 +2982,10 @@ static int can_rmdir(struct send_ctx *sctx, u64 dir, u64 dir_gen,
 			ret = 0;
 			goto out;
 		}
-
-		path->slots[0]++;
+	}
+	if (iter_ret < 0) {
+		ret = iter_ret;
+		goto out;
 	}
 	free_orphan_dir_info(sctx, odi);
 
@@ -3579,7 +3553,7 @@ static int check_ino_in_path(struct btrfs_root *root,
 }
 
 /*
- * Check if ino ino1 is an ancestor of inode ino2 in the given root for any
+ * Check if inode ino1 is an ancestor of inode ino2 in the given root for any
  * possible path (in case ino2 is not a directory and has multiple hard links).
  * Return 1 if true, 0 if false and < 0 on error.
  */
@@ -3591,6 +3565,7 @@ static int is_ancestor(struct btrfs_root *root,
 {
 	bool free_fs_path = false;
 	int ret = 0;
+	int iter_ret = 0;
 	struct btrfs_path *path = NULL;
 	struct btrfs_key key;
 
@@ -3611,26 +3586,12 @@ static int is_ancestor(struct btrfs_root *root,
 	key.type = BTRFS_INODE_REF_KEY;
 	key.offset = 0;
 
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-
-	while (true) {
+	btrfs_for_each_slot(root, &key, &key, path, iter_ret) {
 		struct extent_buffer *leaf = path->nodes[0];
 		int slot = path->slots[0];
 		u32 cur_offset = 0;
 		u32 item_size;
 
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto out;
-			if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(leaf, &key, slot);
 		if (key.objectid != ino2)
 			break;
 		if (key.type != BTRFS_INODE_REF_KEY &&
@@ -3668,10 +3629,12 @@ static int is_ancestor(struct btrfs_root *root,
 			if (ret)
 				goto out;
 		}
-		path->slots[0]++;
 	}
 	ret = 0;
- out:
+	if (iter_ret < 0)
+		ret = iter_ret;
+
+out:
 	btrfs_free_path(path);
 	if (free_fs_path)
 		fs_path_free(fs_path);
@@ -4551,13 +4514,12 @@ out:
 static int process_all_refs(struct send_ctx *sctx,
 			    enum btrfs_compare_tree_result cmd)
 {
-	int ret;
+	int ret = 0;
+	int iter_ret = 0;
 	struct btrfs_root *root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
-	struct extent_buffer *eb;
-	int slot;
 	iterate_inode_ref_t cb;
 	int pending_move = 0;
 
@@ -4581,24 +4543,7 @@ static int process_all_refs(struct send_ctx *sctx,
 	key.objectid = sctx->cmp_key->objectid;
 	key.type = BTRFS_INODE_REF_KEY;
 	key.offset = 0;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-
-	while (1) {
-		eb = path->nodes[0];
-		slot = path->slots[0];
-		if (slot >= btrfs_header_nritems(eb)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto out;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(eb, &found_key, slot);
-
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
 		if (found_key.objectid != key.objectid ||
 		    (found_key.type != BTRFS_INODE_REF_KEY &&
 		     found_key.type != BTRFS_INODE_EXTREF_KEY))
@@ -4607,8 +4552,11 @@ static int process_all_refs(struct send_ctx *sctx,
 		ret = iterate_inode_ref(root, path, &found_key, 0, cb, sctx);
 		if (ret < 0)
 			goto out;
-
-		path->slots[0]++;
+	}
+	/* Catch error found during iteration */
+	if (iter_ret < 0) {
+		ret = iter_ret;
+		goto out;
 	}
 	btrfs_release_path(path);
 
@@ -4870,13 +4818,12 @@ out:
 
 static int process_all_new_xattrs(struct send_ctx *sctx)
 {
-	int ret;
+	int ret = 0;
+	int iter_ret = 0;
 	struct btrfs_root *root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
-	struct extent_buffer *eb;
-	int slot;
 
 	path = alloc_path_for_send();
 	if (!path)
@@ -4887,39 +4834,21 @@ static int process_all_new_xattrs(struct send_ctx *sctx)
 	key.objectid = sctx->cmp_key->objectid;
 	key.type = BTRFS_XATTR_ITEM_KEY;
 	key.offset = 0;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-
-	while (1) {
-		eb = path->nodes[0];
-		slot = path->slots[0];
-		if (slot >= btrfs_header_nritems(eb)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				goto out;
-			} else if (ret > 0) {
-				ret = 0;
-				break;
-			}
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(eb, &found_key, slot);
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
 		if (found_key.objectid != key.objectid ||
 		    found_key.type != key.type) {
 			ret = 0;
-			goto out;
+			break;
 		}
 
 		ret = iterate_dir_item(root, path, __process_new_xattr, sctx);
 		if (ret < 0)
-			goto out;
-
-		path->slots[0]++;
+			break;
 	}
+	/* Catch error found during iteration */
+	if (iter_ret < 0)
+		ret = iter_ret;
 
-out:
 	btrfs_free_path(path);
 	return ret;
 }
@@ -4946,7 +4875,6 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len)
 {
 	struct btrfs_root *root = sctx->send_root;
 	struct btrfs_fs_info *fs_info = root->fs_info;
-	struct inode *inode;
 	struct page *page;
 	pgoff_t index = offset >> PAGE_SHIFT;
 	pgoff_t last_index;
@@ -4957,37 +4885,30 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len)
 	if (ret)
 		return ret;
 
-	inode = btrfs_iget(fs_info->sb, sctx->cur_ino, root);
-	if (IS_ERR(inode))
-		return PTR_ERR(inode);
-
 	last_index = (offset + len - 1) >> PAGE_SHIFT;
 
-	/* initial readahead */
-	memset(&sctx->ra, 0, sizeof(struct file_ra_state));
-	file_ra_state_init(&sctx->ra, inode->i_mapping);
-
 	while (index <= last_index) {
 		unsigned cur_len = min_t(unsigned, len,
 					 PAGE_SIZE - pg_offset);
 
-		page = find_lock_page(inode->i_mapping, index);
+		page = find_lock_page(sctx->cur_inode->i_mapping, index);
 		if (!page) {
-			page_cache_sync_readahead(inode->i_mapping, &sctx->ra,
-				NULL, index, last_index + 1 - index);
+			page_cache_sync_readahead(sctx->cur_inode->i_mapping,
+						  &sctx->ra, NULL, index,
+						  last_index + 1 - index);
 
-			page = find_or_create_page(inode->i_mapping, index,
-					GFP_KERNEL);
+			page = find_or_create_page(sctx->cur_inode->i_mapping,
+						   index, GFP_KERNEL);
 			if (!page) {
 				ret = -ENOMEM;
 				break;
 			}
 		}
 
-		if (PageReadahead(page)) {
-			page_cache_async_readahead(inode->i_mapping, &sctx->ra,
-				NULL, page, index, last_index + 1 - index);
-		}
+		if (PageReadahead(page))
+			page_cache_async_readahead(sctx->cur_inode->i_mapping,
+						   &sctx->ra, NULL, page, index,
+						   last_index + 1 - index);
 
 		if (!PageUptodate(page)) {
 			btrfs_readpage(NULL, page);
@@ -5013,7 +4934,7 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len)
 		len -= cur_len;
 		sctx->send_size += cur_len;
 	}
-	iput(inode);
+
 	return ret;
 }
 
@@ -5220,12 +5141,49 @@ static int send_extent_data(struct send_ctx *sctx,
 			    const u64 offset,
 			    const u64 len)
 {
+	const u64 end = offset + len;
 	u64 read_size = max_send_read_size(sctx);
 	u64 sent = 0;
 
 	if (sctx->flags & BTRFS_SEND_FLAG_NO_FILE_DATA)
 		return send_update_extent(sctx, offset, len);
 
+	if (sctx->cur_inode == NULL) {
+		struct btrfs_root *root = sctx->send_root;
+
+		sctx->cur_inode = btrfs_iget(root->fs_info->sb, sctx->cur_ino, root);
+		if (IS_ERR(sctx->cur_inode)) {
+			int err = PTR_ERR(sctx->cur_inode);
+
+			sctx->cur_inode = NULL;
+			return err;
+		}
+		memset(&sctx->ra, 0, sizeof(struct file_ra_state));
+		file_ra_state_init(&sctx->ra, sctx->cur_inode->i_mapping);
+
+		/*
+		 * It's very likely there are no pages from this inode in the page
+		 * cache, so after reading extents and sending their data, we clean
+		 * the page cache to avoid trashing the page cache (adding pressure
+		 * to the page cache and forcing eviction of other data more useful
+		 * for applications).
+		 *
+		 * We decide if we should clean the page cache simply by checking
+		 * if the inode's mapping nrpages is 0 when we first open it, and
+		 * not by using something like filemap_range_has_page() before
+		 * reading an extent because when we ask the readahead code to
+		 * read a given file range, it may (and almost always does) read
+		 * pages from beyond that range (see the documentation for
+		 * page_cache_sync_readahead()), so it would not be reliable,
+		 * because after reading the first extent future calls to
+		 * filemap_range_has_page() would return true because the readahead
+		 * on the previous extent resulted in reading pages of the current
+		 * extent as well.
+		 */
+		sctx->clean_page_cache = (sctx->cur_inode->i_mapping->nrpages == 0);
+		sctx->page_cache_clear_start = round_down(offset, PAGE_SIZE);
+	}
+
 	while (sent < len) {
 		u64 size = min(len - sent, read_size);
 		int ret;
@@ -5235,6 +5193,37 @@ static int send_extent_data(struct send_ctx *sctx,
 			return ret;
 		sent += size;
 	}
+
+	if (sctx->clean_page_cache && IS_ALIGNED(end, PAGE_SIZE)) {
+		/*
+		 * Always operate only on ranges that are a multiple of the page
+		 * size. This is not only to prevent zeroing parts of a page in
+		 * the case of subpage sector size, but also to guarantee we evict
+		 * pages, as passing a range that is smaller than page size does
+		 * not evict the respective page (only zeroes part of its content).
+		 *
+		 * Always start from the end offset of the last range cleared.
+		 * This is because the readahead code may (and very often does)
+		 * reads pages beyond the range we request for readahead. So if
+		 * we have an extent layout like this:
+		 *
+		 *            [ extent A ] [ extent B ] [ extent C ]
+		 *
+		 * When we ask page_cache_sync_readahead() to read extent A, it
+		 * may also trigger reads for pages of extent B. If we are doing
+		 * an incremental send and extent B has not changed between the
+		 * parent and send snapshots, some or all of its pages may end
+		 * up being read and placed in the page cache. So when truncating
+		 * the page cache we always start from the end offset of the
+		 * previously processed extent up to the end of the current
+		 * extent.
+		 */
+		truncate_inode_pages_range(&sctx->cur_inode->i_data,
+					   sctx->page_cache_clear_start,
+					   end - 1);
+		sctx->page_cache_clear_start = end;
+	}
+
 	return 0;
 }
 
@@ -5965,13 +5954,12 @@ out:
 
 static int process_all_extents(struct send_ctx *sctx)
 {
-	int ret;
+	int ret = 0;
+	int iter_ret = 0;
 	struct btrfs_root *root;
 	struct btrfs_path *path;
 	struct btrfs_key key;
 	struct btrfs_key found_key;
-	struct extent_buffer *eb;
-	int slot;
 
 	root = sctx->send_root;
 	path = alloc_path_for_send();
@@ -5981,41 +5969,21 @@ static int process_all_extents(struct send_ctx *sctx)
 	key.objectid = sctx->cmp_key->objectid;
 	key.type = BTRFS_EXTENT_DATA_KEY;
 	key.offset = 0;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
-
-	while (1) {
-		eb = path->nodes[0];
-		slot = path->slots[0];
-
-		if (slot >= btrfs_header_nritems(eb)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0) {
-				goto out;
-			} else if (ret > 0) {
-				ret = 0;
-				break;
-			}
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(eb, &found_key, slot);
-
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
 		if (found_key.objectid != key.objectid ||
 		    found_key.type != key.type) {
 			ret = 0;
-			goto out;
+			break;
 		}
 
 		ret = process_extent(sctx, path, &found_key);
 		if (ret < 0)
-			goto out;
-
-		path->slots[0]++;
+			break;
 	}
+	/* Catch error found during iteration */
+	if (iter_ret < 0)
+		ret = iter_ret;
 
-out:
 	btrfs_free_path(path);
 	return ret;
 }
@@ -6205,8 +6173,11 @@ static int btrfs_unlink_all_paths(struct send_ctx *sctx)
 {
 	LIST_HEAD(deleted_refs);
 	struct btrfs_path *path;
+	struct btrfs_root *root = sctx->parent_root;
 	struct btrfs_key key;
+	struct btrfs_key found_key;
 	struct parent_paths_ctx ctx;
+	int iter_ret = 0;
 	int ret;
 
 	path = alloc_path_for_send();
@@ -6216,39 +6187,26 @@ static int btrfs_unlink_all_paths(struct send_ctx *sctx)
 	key.objectid = sctx->cur_ino;
 	key.type = BTRFS_INODE_REF_KEY;
 	key.offset = 0;
-	ret = btrfs_search_slot(NULL, sctx->parent_root, &key, path, 0, 0);
-	if (ret < 0)
-		goto out;
 
 	ctx.refs = &deleted_refs;
 	ctx.sctx = sctx;
 
-	while (true) {
-		struct extent_buffer *eb = path->nodes[0];
-		int slot = path->slots[0];
-
-		if (slot >= btrfs_header_nritems(eb)) {
-			ret = btrfs_next_leaf(sctx->parent_root, path);
-			if (ret < 0)
-				goto out;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(eb, &key, slot);
-		if (key.objectid != sctx->cur_ino)
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
+		if (found_key.objectid != key.objectid)
 			break;
-		if (key.type != BTRFS_INODE_REF_KEY &&
-		    key.type != BTRFS_INODE_EXTREF_KEY)
+		if (found_key.type != key.type &&
+		    found_key.type != BTRFS_INODE_EXTREF_KEY)
 			break;
 
-		ret = iterate_inode_ref(sctx->parent_root, path, &key, 1,
+		ret = iterate_inode_ref(root, path, &found_key, 1,
 					record_parent_ref, &ctx);
 		if (ret < 0)
 			goto out;
-
-		path->slots[0]++;
+	}
+	/* Catch error found during iteration */
+	if (iter_ret < 0) {
+		ret = iter_ret;
+		goto out;
 	}
 
 	while (!list_empty(&deleted_refs)) {
@@ -6270,6 +6228,30 @@ out:
 	return ret;
 }
 
+static void close_current_inode(struct send_ctx *sctx)
+{
+	u64 i_size;
+
+	if (sctx->cur_inode == NULL)
+		return;
+
+	i_size = i_size_read(sctx->cur_inode);
+
+	/*
+	 * If we are doing an incremental send, we may have extents between the
+	 * last processed extent and the i_size that have not been processed
+	 * because they haven't changed but we may have read some of their pages
+	 * through readahead, see the comments at send_extent_data().
+	 */
+	if (sctx->clean_page_cache && sctx->page_cache_clear_start < i_size)
+		truncate_inode_pages_range(&sctx->cur_inode->i_data,
+					   sctx->page_cache_clear_start,
+					   round_up(i_size, PAGE_SIZE) - 1);
+
+	iput(sctx->cur_inode);
+	sctx->cur_inode = NULL;
+}
+
 static int changed_inode(struct send_ctx *sctx,
 			 enum btrfs_compare_tree_result result)
 {
@@ -6280,6 +6262,8 @@ static int changed_inode(struct send_ctx *sctx,
 	u64 left_gen = 0;
 	u64 right_gen = 0;
 
+	close_current_inode(sctx);
+
 	sctx->cur_ino = key->objectid;
 	sctx->cur_inode_new_gen = 0;
 	sctx->cur_inode_last_extent = (u64)-1;
@@ -7534,7 +7518,7 @@ long btrfs_ioctl_send(struct inode *inode, struct btrfs_ioctl_send_args *arg)
 
 	INIT_LIST_HEAD(&sctx->new_refs);
 	INIT_LIST_HEAD(&sctx->deleted_refs);
-	INIT_RADIX_TREE(&sctx->name_cache, GFP_KERNEL);
+	xa_init_flags(&sctx->name_cache, GFP_KERNEL);
 	INIT_LIST_HEAD(&sctx->name_cache_list);
 
 	sctx->flags = arg->flags;
@@ -7766,6 +7750,8 @@ out:
 
 		name_cache_free(sctx);
 
+		close_current_inode(sctx);
+
 		kfree(sctx);
 	}
 
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index b87931a458eb..2dd8754cb990 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -181,6 +181,12 @@ void btrfs_clear_space_info_full(struct btrfs_fs_info *info)
 		found->full = 0;
 }
 
+/*
+ * Block groups with more than this value (percents) of unusable space will be
+ * scheduled for background reclaim.
+ */
+#define BTRFS_DEFAULT_ZONED_RECLAIM_THRESH			(75)
+
 static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 {
 
@@ -203,6 +209,9 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 	INIT_LIST_HEAD(&space_info->priority_tickets);
 	space_info->clamp = 1;
 
+	if (btrfs_is_zoned(info))
+		space_info->bg_reclaim_threshold = BTRFS_DEFAULT_ZONED_RECLAIM_THRESH;
+
 	ret = btrfs_sysfs_add_space_info_type(info, space_info);
 	if (ret)
 		return ret;
@@ -519,7 +528,7 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 		items = calc_reclaim_items_nr(fs_info, to_reclaim) * 2;
 	}
 
-	trans = (struct btrfs_trans_handle *)current->journal_info;
+	trans = current->journal_info;
 
 	/*
 	 * If we are doing more ordered than delalloc we need to just wait on
diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h
index d841fed73492..c096695598c1 100644
--- a/fs/btrfs/space-info.h
+++ b/fs/btrfs/space-info.h
@@ -3,6 +3,8 @@
 #ifndef BTRFS_SPACE_INFO_H
 #define BTRFS_SPACE_INFO_H
 
+#include "volumes.h"
+
 struct btrfs_space_info {
 	spinlock_t lock;
 
@@ -24,6 +26,12 @@ struct btrfs_space_info {
 				   the space info if we had an ENOSPC in the
 				   allocator. */
 
+	/*
+	 * Once a block group drops below this threshold (percents) we'll
+	 * schedule it for reclaim.
+	 */
+	int bg_reclaim_threshold;
+
 	int clamp;		/* Used to scale our threshold for preemptive
 				   flushing. The value is >> clamp, so turns
 				   out to be a 2^clamp divisor. */
diff --git a/fs/btrfs/subpage.c b/fs/btrfs/subpage.c
index ef7ae20d2b77..a105b291444f 100644
--- a/fs/btrfs/subpage.c
+++ b/fs/btrfs/subpage.c
@@ -63,6 +63,29 @@
  *   This means a slightly higher tree locking latency.
  */
 
+bool btrfs_is_subpage(const struct btrfs_fs_info *fs_info, struct page *page)
+{
+	if (fs_info->sectorsize >= PAGE_SIZE)
+		return false;
+
+	/*
+	 * Only data pages (either through DIO or compression) can have no
+	 * mapping. And if page->mapping->host is data inode, it's subpage.
+	 * As we have ruled our sectorsize >= PAGE_SIZE case already.
+	 */
+	if (!page->mapping || !page->mapping->host ||
+	    is_data_inode(page->mapping->host))
+		return true;
+
+	/*
+	 * Now the only remaining case is metadata, which we only go subpage
+	 * routine if nodesize < PAGE_SIZE.
+	 */
+	if (fs_info->nodesize < PAGE_SIZE)
+		return true;
+	return false;
+}
+
 void btrfs_init_subpage_info(struct btrfs_subpage_info *subpage_info, u32 sectorsize)
 {
 	unsigned int cur = 0;
@@ -107,7 +130,7 @@ int btrfs_attach_subpage(const struct btrfs_fs_info *fs_info,
 		ASSERT(PageLocked(page));
 
 	/* Either not subpage, or the page already has private attached */
-	if (fs_info->sectorsize == PAGE_SIZE || PagePrivate(page))
+	if (!btrfs_is_subpage(fs_info, page) || PagePrivate(page))
 		return 0;
 
 	subpage = btrfs_alloc_subpage(fs_info, type);
@@ -124,10 +147,10 @@ void btrfs_detach_subpage(const struct btrfs_fs_info *fs_info,
 	struct btrfs_subpage *subpage;
 
 	/* Either not subpage, or already detached */
-	if (fs_info->sectorsize == PAGE_SIZE || !PagePrivate(page))
+	if (!btrfs_is_subpage(fs_info, page) || !PagePrivate(page))
 		return;
 
-	subpage = (struct btrfs_subpage *)detach_page_private(page);
+	subpage = detach_page_private(page);
 	ASSERT(subpage);
 	btrfs_free_subpage(subpage);
 }
@@ -175,7 +198,7 @@ void btrfs_page_inc_eb_refs(const struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_subpage *subpage;
 
-	if (fs_info->sectorsize == PAGE_SIZE)
+	if (!btrfs_is_subpage(fs_info, page))
 		return;
 
 	ASSERT(PagePrivate(page) && page->mapping);
@@ -190,7 +213,7 @@ void btrfs_page_dec_eb_refs(const struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_subpage *subpage;
 
-	if (fs_info->sectorsize == PAGE_SIZE)
+	if (!btrfs_is_subpage(fs_info, page))
 		return;
 
 	ASSERT(PagePrivate(page) && page->mapping);
@@ -319,7 +342,7 @@ bool btrfs_subpage_end_and_test_writer(const struct btrfs_fs_info *fs_info,
 int btrfs_page_start_writer_lock(const struct btrfs_fs_info *fs_info,
 		struct page *page, u64 start, u32 len)
 {
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) {
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page)) {
 		lock_page(page);
 		return 0;
 	}
@@ -336,7 +359,7 @@ int btrfs_page_start_writer_lock(const struct btrfs_fs_info *fs_info,
 void btrfs_page_end_writer_lock(const struct btrfs_fs_info *fs_info,
 		struct page *page, u64 start, u32 len)
 {
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE)
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page))
 		return unlock_page(page);
 	btrfs_subpage_clamp_range(page, &start, &len);
 	if (btrfs_subpage_end_and_test_writer(fs_info, page, start, len))
@@ -620,7 +643,7 @@ IMPLEMENT_BTRFS_SUBPAGE_TEST_OP(checked);
 void btrfs_page_set_##name(const struct btrfs_fs_info *fs_info,		\
 		struct page *page, u64 start, u32 len)			\
 {									\
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) {	\
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page)) {	\
 		set_page_func(page);					\
 		return;							\
 	}								\
@@ -629,7 +652,7 @@ void btrfs_page_set_##name(const struct btrfs_fs_info *fs_info,		\
 void btrfs_page_clear_##name(const struct btrfs_fs_info *fs_info,	\
 		struct page *page, u64 start, u32 len)			\
 {									\
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) {	\
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page)) {	\
 		clear_page_func(page);					\
 		return;							\
 	}								\
@@ -638,14 +661,14 @@ void btrfs_page_clear_##name(const struct btrfs_fs_info *fs_info,	\
 bool btrfs_page_test_##name(const struct btrfs_fs_info *fs_info,	\
 		struct page *page, u64 start, u32 len)			\
 {									\
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE)	\
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page))	\
 		return test_page_func(page);				\
 	return btrfs_subpage_test_##name(fs_info, page, start, len);	\
 }									\
 void btrfs_page_clamp_set_##name(const struct btrfs_fs_info *fs_info,	\
 		struct page *page, u64 start, u32 len)			\
 {									\
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) {	\
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page)) {	\
 		set_page_func(page);					\
 		return;							\
 	}								\
@@ -655,7 +678,7 @@ void btrfs_page_clamp_set_##name(const struct btrfs_fs_info *fs_info,	\
 void btrfs_page_clamp_clear_##name(const struct btrfs_fs_info *fs_info, \
 		struct page *page, u64 start, u32 len)			\
 {									\
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) {	\
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page)) {	\
 		clear_page_func(page);					\
 		return;							\
 	}								\
@@ -665,7 +688,7 @@ void btrfs_page_clamp_clear_##name(const struct btrfs_fs_info *fs_info, \
 bool btrfs_page_clamp_test_##name(const struct btrfs_fs_info *fs_info,	\
 		struct page *page, u64 start, u32 len)			\
 {									\
-	if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE)	\
+	if (unlikely(!fs_info) || !btrfs_is_subpage(fs_info, page))	\
 		return test_page_func(page);				\
 	btrfs_subpage_clamp_range(page, &start, &len);			\
 	return btrfs_subpage_test_##name(fs_info, page, start, len);	\
@@ -694,7 +717,7 @@ void btrfs_page_assert_not_dirty(const struct btrfs_fs_info *fs_info,
 		return;
 
 	ASSERT(!PageDirty(page));
-	if (fs_info->sectorsize == PAGE_SIZE)
+	if (!btrfs_is_subpage(fs_info, page))
 		return;
 
 	ASSERT(PagePrivate(page) && page->private);
@@ -722,8 +745,8 @@ void btrfs_page_unlock_writer(struct btrfs_fs_info *fs_info, struct page *page,
 	struct btrfs_subpage *subpage;
 
 	ASSERT(PageLocked(page));
-	/* For regular page size case, we just unlock the page */
-	if (fs_info->sectorsize == PAGE_SIZE)
+	/* For non-subpage case, we just unlock the page */
+	if (!btrfs_is_subpage(fs_info, page))
 		return unlock_page(page);
 
 	ASSERT(PagePrivate(page) && page->private);
diff --git a/fs/btrfs/subpage.h b/fs/btrfs/subpage.h
index 7accb5c40d33..0e80ad336904 100644
--- a/fs/btrfs/subpage.h
+++ b/fs/btrfs/subpage.h
@@ -74,6 +74,8 @@ enum btrfs_subpage_type {
 	BTRFS_SUBPAGE_DATA,
 };
 
+bool btrfs_is_subpage(const struct btrfs_fs_info *fs_info, struct page *page);
+
 void btrfs_init_subpage_info(struct btrfs_subpage_info *subpage_info, u32 sectorsize);
 int btrfs_attach_subpage(const struct btrfs_fs_info *fs_info,
 			 struct page *page, enum btrfs_subpage_type type);
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index b228efe8ab6e..b1fdc6a26c76 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -261,7 +261,7 @@ static struct ratelimit_state printk_limits[] = {
 	RATELIMIT_STATE_INIT(printk_limits[7], DEFAULT_RATELIMIT_INTERVAL, 100),
 };
 
-void __cold btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
+void __cold _btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...)
 {
 	char lvl[PRINTK_MAX_SINGLE_HEADER_LEN + 1] = "\0";
 	struct va_format vaf;
@@ -292,10 +292,10 @@ void __cold btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, .
 			char statestr[STATE_STRING_BUF_LEN];
 
 			btrfs_state_to_string(fs_info, statestr);
-			printk("%sBTRFS %s (device %s%s): %pV\n", lvl, type,
+			_printk("%sBTRFS %s (device %s%s): %pV\n", lvl, type,
 				fs_info->sb->s_id, statestr, &vaf);
 		} else {
-			printk("%sBTRFS %s: %pV\n", lvl, type, &vaf);
+			_printk("%sBTRFS %s: %pV\n", lvl, type, &vaf);
 		}
 	}
 
@@ -1903,6 +1903,7 @@ static void btrfs_resize_thread_pool(struct btrfs_fs_info *fs_info,
 	       old_pool_size, new_pool_size);
 
 	btrfs_workqueue_set_max(fs_info->workers, new_pool_size);
+	btrfs_workqueue_set_max(fs_info->hipri_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->delalloc_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->caching_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->endio_workers, new_pool_size);
@@ -1912,8 +1913,6 @@ static void btrfs_resize_thread_pool(struct btrfs_fs_info *fs_info,
 	btrfs_workqueue_set_max(fs_info->endio_write_workers, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->endio_freespace_worker, new_pool_size);
 	btrfs_workqueue_set_max(fs_info->delayed_workers, new_pool_size);
-	btrfs_workqueue_set_max(fs_info->scrub_wr_completion_workers,
-				new_pool_size);
 }
 
 static inline void btrfs_remount_begin(struct btrfs_fs_info *fs_info,
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index ba78ca5aabbb..92a1fa8e3da6 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -394,11 +394,9 @@ static ssize_t supported_sectorsizes_show(struct kobject *kobj,
 {
 	ssize_t ret = 0;
 
-	/* 4K sector size is also supported with 64K page size */
-	if (PAGE_SIZE == SZ_64K)
+	/* An artificial limit to only support 4K and PAGE_SIZE */
+	if (PAGE_SIZE > SZ_4K)
 		ret += sysfs_emit_at(buf, ret, "%u ", SZ_4K);
-
-	/* Only sectorsize == PAGE_SIZE is now supported */
 	ret += sysfs_emit_at(buf, ret, "%lu\n", PAGE_SIZE);
 
 	return ret;
@@ -722,6 +720,42 @@ SPACE_INFO_ATTR(bytes_zone_unusable);
 SPACE_INFO_ATTR(disk_used);
 SPACE_INFO_ATTR(disk_total);
 
+static ssize_t btrfs_sinfo_bg_reclaim_threshold_show(struct kobject *kobj,
+						     struct kobj_attribute *a,
+						     char *buf)
+{
+	struct btrfs_space_info *space_info = to_space_info(kobj);
+	ssize_t ret;
+
+	ret = sysfs_emit(buf, "%d\n", READ_ONCE(space_info->bg_reclaim_threshold));
+
+	return ret;
+}
+
+static ssize_t btrfs_sinfo_bg_reclaim_threshold_store(struct kobject *kobj,
+						      struct kobj_attribute *a,
+						      const char *buf, size_t len)
+{
+	struct btrfs_space_info *space_info = to_space_info(kobj);
+	int thresh;
+	int ret;
+
+	ret = kstrtoint(buf, 10, &thresh);
+	if (ret)
+		return ret;
+
+	if (thresh < 0 || thresh > 100)
+		return -EINVAL;
+
+	WRITE_ONCE(space_info->bg_reclaim_threshold, thresh);
+
+	return len;
+}
+
+BTRFS_ATTR_RW(space_info, bg_reclaim_threshold,
+	      btrfs_sinfo_bg_reclaim_threshold_show,
+	      btrfs_sinfo_bg_reclaim_threshold_store);
+
 /*
  * Allocation information about block group types.
  *
@@ -738,6 +772,7 @@ static struct attribute *space_info_attrs[] = {
 	BTRFS_ATTR_PTR(space_info, bytes_zone_unusable),
 	BTRFS_ATTR_PTR(space_info, disk_used),
 	BTRFS_ATTR_PTR(space_info, disk_total),
+	BTRFS_ATTR_PTR(space_info, bg_reclaim_threshold),
 	NULL,
 };
 ATTRIBUTE_GROUPS(space_info);
diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c
index d8e56edd6991..1591bfa55bcc 100644
--- a/fs/btrfs/tests/btrfs-tests.c
+++ b/fs/btrfs/tests/btrfs-tests.c
@@ -150,8 +150,8 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize)
 
 void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
 {
-	struct radix_tree_iter iter;
-	void **slot;
+	unsigned long index;
+	struct extent_buffer *eb;
 	struct btrfs_device *dev, *tmp;
 
 	if (!fs_info)
@@ -163,25 +163,9 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info)
 
 	test_mnt->mnt_sb->s_fs_info = NULL;
 
-	spin_lock(&fs_info->buffer_lock);
-	radix_tree_for_each_slot(slot, &fs_info->buffer_radix, &iter, 0) {
-		struct extent_buffer *eb;
-
-		eb = radix_tree_deref_slot_protected(slot, &fs_info->buffer_lock);
-		if (!eb)
-			continue;
-		/* Shouldn't happen but that kind of thinking creates CVE's */
-		if (radix_tree_exception(eb)) {
-			if (radix_tree_deref_retry(eb))
-				slot = radix_tree_iter_retry(&iter);
-			continue;
-		}
-		slot = radix_tree_iter_resume(slot, &iter);
-		spin_unlock(&fs_info->buffer_lock);
+	xa_for_each(&fs_info->extent_buffers, index, eb) {
 		free_extent_buffer_stale(eb);
-		spin_lock(&fs_info->buffer_lock);
 	}
-	spin_unlock(&fs_info->buffer_lock);
 
 	btrfs_mapping_tree_free(&fs_info->mapping_tree);
 	list_for_each_entry_safe(dev, tmp, &fs_info->fs_devices->devices,
@@ -202,7 +186,7 @@ void btrfs_free_dummy_root(struct btrfs_root *root)
 	if (!root)
 		return;
 	/* Will be freed by btrfs_free_fs_roots */
-	if (WARN_ON(test_bit(BTRFS_ROOT_IN_RADIX, &root->state)))
+	if (WARN_ON(test_bit(BTRFS_ROOT_REGISTERED, &root->state)))
 		return;
 	btrfs_global_root_delete(root);
 	btrfs_put_root(root);
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index b008c5110958..06c0a958d114 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -23,7 +23,7 @@
 #include "space-info.h"
 #include "zoned.h"
 
-#define BTRFS_ROOT_TRANS_TAG 0
+#define BTRFS_ROOT_TRANS_TAG				XA_MARK_0
 
 /*
  * Transaction states and transitions
@@ -221,7 +221,7 @@ static noinline void switch_commit_roots(struct btrfs_trans_handle *trans)
 	 * the caching thread will re-start it's search from 3, and thus find
 	 * the hole from [4,6) to add to the free space cache.
 	 */
-	spin_lock(&fs_info->block_group_cache_lock);
+	write_lock(&fs_info->block_group_cache_lock);
 	list_for_each_entry_safe(caching_ctl, next,
 				 &fs_info->caching_block_groups, list) {
 		struct btrfs_block_group *cache = caching_ctl->block_group;
@@ -234,7 +234,7 @@ static noinline void switch_commit_roots(struct btrfs_trans_handle *trans)
 			cache->last_byte_to_unpin = caching_ctl->progress;
 		}
 	}
-	spin_unlock(&fs_info->block_group_cache_lock);
+	write_unlock(&fs_info->block_group_cache_lock);
 	up_write(&fs_info->commit_root_sem);
 }
 
@@ -437,15 +437,15 @@ static int record_root_in_trans(struct btrfs_trans_handle *trans,
 		 */
 		smp_wmb();
 
-		spin_lock(&fs_info->fs_roots_radix_lock);
+		spin_lock(&fs_info->fs_roots_lock);
 		if (root->last_trans == trans->transid && !force) {
-			spin_unlock(&fs_info->fs_roots_radix_lock);
+			spin_unlock(&fs_info->fs_roots_lock);
 			return 0;
 		}
-		radix_tree_tag_set(&fs_info->fs_roots_radix,
-				   (unsigned long)root->root_key.objectid,
-				   BTRFS_ROOT_TRANS_TAG);
-		spin_unlock(&fs_info->fs_roots_radix_lock);
+		xa_set_mark(&fs_info->fs_roots,
+			    (unsigned long)root->root_key.objectid,
+			    BTRFS_ROOT_TRANS_TAG);
+		spin_unlock(&fs_info->fs_roots_lock);
 		root->last_trans = trans->transid;
 
 		/* this is pretty tricky.  We don't want to
@@ -487,11 +487,9 @@ void btrfs_add_dropped_root(struct btrfs_trans_handle *trans,
 	spin_unlock(&cur_trans->dropped_roots_lock);
 
 	/* Make sure we don't try to update the root at commit time */
-	spin_lock(&fs_info->fs_roots_radix_lock);
-	radix_tree_tag_clear(&fs_info->fs_roots_radix,
-			     (unsigned long)root->root_key.objectid,
-			     BTRFS_ROOT_TRANS_TAG);
-	spin_unlock(&fs_info->fs_roots_radix_lock);
+	xa_clear_mark(&fs_info->fs_roots,
+		      (unsigned long)root->root_key.objectid,
+		      BTRFS_ROOT_TRANS_TAG);
 }
 
 int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
@@ -1404,9 +1402,8 @@ void btrfs_add_dead_root(struct btrfs_root *root)
 static noinline int commit_fs_roots(struct btrfs_trans_handle *trans)
 {
 	struct btrfs_fs_info *fs_info = trans->fs_info;
-	struct btrfs_root *gang[8];
-	int i;
-	int ret;
+	struct btrfs_root *root;
+	unsigned long index;
 
 	/*
 	 * At this point no one can be using this transaction to modify any tree
@@ -1414,57 +1411,46 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans)
 	 */
 	ASSERT(trans->transaction->state == TRANS_STATE_COMMIT_DOING);
 
-	spin_lock(&fs_info->fs_roots_radix_lock);
-	while (1) {
-		ret = radix_tree_gang_lookup_tag(&fs_info->fs_roots_radix,
-						 (void **)gang, 0,
-						 ARRAY_SIZE(gang),
-						 BTRFS_ROOT_TRANS_TAG);
-		if (ret == 0)
-			break;
-		for (i = 0; i < ret; i++) {
-			struct btrfs_root *root = gang[i];
-			int ret2;
-
-			/*
-			 * At this point we can neither have tasks logging inodes
-			 * from a root nor trying to commit a log tree.
-			 */
-			ASSERT(atomic_read(&root->log_writers) == 0);
-			ASSERT(atomic_read(&root->log_commit[0]) == 0);
-			ASSERT(atomic_read(&root->log_commit[1]) == 0);
-
-			radix_tree_tag_clear(&fs_info->fs_roots_radix,
-					(unsigned long)root->root_key.objectid,
-					BTRFS_ROOT_TRANS_TAG);
-			spin_unlock(&fs_info->fs_roots_radix_lock);
-
-			btrfs_free_log(trans, root);
-			ret2 = btrfs_update_reloc_root(trans, root);
-			if (ret2)
-				return ret2;
-
-			/* see comments in should_cow_block() */
-			clear_bit(BTRFS_ROOT_FORCE_COW, &root->state);
-			smp_mb__after_atomic();
-
-			if (root->commit_root != root->node) {
-				list_add_tail(&root->dirty_list,
-					&trans->transaction->switch_commits);
-				btrfs_set_root_node(&root->root_item,
-						    root->node);
-			}
+	spin_lock(&fs_info->fs_roots_lock);
+	xa_for_each_marked(&fs_info->fs_roots, index, root, BTRFS_ROOT_TRANS_TAG) {
+		int ret;
+
+		/*
+		 * At this point we can neither have tasks logging inodes
+		 * from a root nor trying to commit a log tree.
+		 */
+		ASSERT(atomic_read(&root->log_writers) == 0);
+		ASSERT(atomic_read(&root->log_commit[0]) == 0);
+		ASSERT(atomic_read(&root->log_commit[1]) == 0);
+
+		xa_clear_mark(&fs_info->fs_roots,
+			      (unsigned long)root->root_key.objectid,
+			      BTRFS_ROOT_TRANS_TAG);
+		spin_unlock(&fs_info->fs_roots_lock);
 
-			ret2 = btrfs_update_root(trans, fs_info->tree_root,
-						&root->root_key,
-						&root->root_item);
-			if (ret2)
-				return ret2;
-			spin_lock(&fs_info->fs_roots_radix_lock);
-			btrfs_qgroup_free_meta_all_pertrans(root);
+		btrfs_free_log(trans, root);
+		ret = btrfs_update_reloc_root(trans, root);
+		if (ret)
+			return ret;
+
+		/* See comments in should_cow_block() */
+		clear_bit(BTRFS_ROOT_FORCE_COW, &root->state);
+		smp_mb__after_atomic();
+
+		if (root->commit_root != root->node) {
+			list_add_tail(&root->dirty_list,
+				      &trans->transaction->switch_commits);
+			btrfs_set_root_node(&root->root_item, root->node);
 		}
+
+		ret = btrfs_update_root(trans, fs_info->tree_root,
+					&root->root_key, &root->root_item);
+		if (ret)
+			return ret;
+		spin_lock(&fs_info->fs_roots_lock);
+		btrfs_qgroup_free_meta_all_pertrans(root);
 	}
-	spin_unlock(&fs_info->fs_roots_radix_lock);
+	spin_unlock(&fs_info->fs_roots_lock);
 	return 0;
 }
 
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index e56c0107eea3..9e0e0ae2288c 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -1855,3 +1855,58 @@ out:
 	return ret;
 }
 ALLOW_ERROR_INJECTION(btrfs_check_node, ERRNO);
+
+int btrfs_check_eb_owner(const struct extent_buffer *eb, u64 root_owner)
+{
+	const bool is_subvol = is_fstree(root_owner);
+	const u64 eb_owner = btrfs_header_owner(eb);
+
+	/*
+	 * Skip dummy fs, as selftests don't create unique ebs for each dummy
+	 * root.
+	 */
+	if (test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &eb->fs_info->fs_state))
+		return 0;
+	/*
+	 * There are several call sites (backref walking, qgroup, and data
+	 * reloc) passing 0 as @root_owner, as they are not holding the
+	 * tree root.  In that case, we can not do a reliable ownership check,
+	 * so just exit.
+	 */
+	if (root_owner == 0)
+		return 0;
+	/*
+	 * These trees use key.offset as their owner, our callers don't have
+	 * the extra capacity to pass key.offset here.  So we just skip them.
+	 */
+	if (root_owner == BTRFS_TREE_LOG_OBJECTID ||
+	    root_owner == BTRFS_TREE_RELOC_OBJECTID)
+		return 0;
+
+	if (!is_subvol) {
+		/* For non-subvolume trees, the eb owner should match root owner */
+		if (unlikely(root_owner != eb_owner)) {
+			btrfs_crit(eb->fs_info,
+"corrupted %s, root=%llu block=%llu owner mismatch, have %llu expect %llu",
+				btrfs_header_level(eb) == 0 ? "leaf" : "node",
+				root_owner, btrfs_header_bytenr(eb), eb_owner,
+				root_owner);
+			return -EUCLEAN;
+		}
+		return 0;
+	}
+
+	/*
+	 * For subvolume trees, owners can mismatch, but they should all belong
+	 * to subvolume trees.
+	 */
+	if (unlikely(is_subvol != is_fstree(eb_owner))) {
+		btrfs_crit(eb->fs_info,
+"corrupted %s, root=%llu block=%llu owner mismatch, have %llu expect [%llu, %llu]",
+			btrfs_header_level(eb) == 0 ? "leaf" : "node",
+			root_owner, btrfs_header_bytenr(eb), eb_owner,
+			BTRFS_FIRST_FREE_OBJECTID, BTRFS_LAST_FREE_OBJECTID);
+		return -EUCLEAN;
+	}
+	return 0;
+}
diff --git a/fs/btrfs/tree-checker.h b/fs/btrfs/tree-checker.h
index 32fecc9dc1dd..ece497e26558 100644
--- a/fs/btrfs/tree-checker.h
+++ b/fs/btrfs/tree-checker.h
@@ -25,5 +25,6 @@ int btrfs_check_node(struct extent_buffer *node);
 
 int btrfs_check_chunk_valid(struct extent_buffer *leaf,
 			    struct btrfs_chunk *chunk, u64 logical);
+int btrfs_check_eb_owner(const struct extent_buffer *eb, u64 root_owner);
 
 #endif
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index e65633686378..370388fadf96 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -333,7 +333,7 @@ static int process_one_buffer(struct btrfs_root *log,
 	 * pin down any logged extents, so we have to read the block.
 	 */
 	if (btrfs_fs_incompat(fs_info, MIXED_GROUPS)) {
-		ret = btrfs_read_buffer(eb, gen, level, NULL);
+		ret = btrfs_read_extent_buffer(eb, gen, level, NULL);
 		if (ret)
 			return ret;
 	}
@@ -894,8 +894,7 @@ update_inode:
 	btrfs_update_inode_bytes(BTRFS_I(inode), nbytes, drop_args.bytes_found);
 	ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
 out:
-	if (inode)
-		iput(inode);
+	iput(inode);
 	return ret;
 }
 
@@ -2575,7 +2574,7 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 	int i;
 	int ret;
 
-	ret = btrfs_read_buffer(eb, gen, level, NULL);
+	ret = btrfs_read_extent_buffer(eb, gen, level, NULL);
 	if (ret)
 		return ret;
 
@@ -2786,7 +2785,7 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 
 			path->slots[*level]++;
 			if (wc->free) {
-				ret = btrfs_read_buffer(next, ptr_gen,
+				ret = btrfs_read_extent_buffer(next, ptr_gen,
 							*level - 1, &first_key);
 				if (ret) {
 					free_extent_buffer(next);
@@ -2815,7 +2814,7 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans,
 			free_extent_buffer(next);
 			continue;
 		}
-		ret = btrfs_read_buffer(next, ptr_gen, *level - 1, &first_key);
+		ret = btrfs_read_extent_buffer(next, ptr_gen, *level - 1, &first_key);
 		if (ret) {
 			free_extent_buffer(next);
 			return ret;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index b6b00338037c..9c20049d1fec 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -164,24 +164,12 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
  */
 enum btrfs_raid_types __attribute_const__ btrfs_bg_flags_to_raid_index(u64 flags)
 {
-	if (flags & BTRFS_BLOCK_GROUP_RAID10)
-		return BTRFS_RAID_RAID10;
-	else if (flags & BTRFS_BLOCK_GROUP_RAID1)
-		return BTRFS_RAID_RAID1;
-	else if (flags & BTRFS_BLOCK_GROUP_RAID1C3)
-		return BTRFS_RAID_RAID1C3;
-	else if (flags & BTRFS_BLOCK_GROUP_RAID1C4)
-		return BTRFS_RAID_RAID1C4;
-	else if (flags & BTRFS_BLOCK_GROUP_DUP)
-		return BTRFS_RAID_DUP;
-	else if (flags & BTRFS_BLOCK_GROUP_RAID0)
-		return BTRFS_RAID_RAID0;
-	else if (flags & BTRFS_BLOCK_GROUP_RAID5)
-		return BTRFS_RAID_RAID5;
-	else if (flags & BTRFS_BLOCK_GROUP_RAID6)
-		return BTRFS_RAID_RAID6;
-
-	return BTRFS_RAID_SINGLE; /* BTRFS_BLOCK_GROUP_SINGLE */
+	const u64 profile = (flags & BTRFS_BLOCK_GROUP_PROFILE_MASK);
+
+	if (!profile)
+		return BTRFS_RAID_SINGLE;
+
+	return BTRFS_BG_FLAG_TO_INDEX(profile);
 }
 
 const char *btrfs_bg_type_to_raid_name(u64 flags)
@@ -4062,13 +4050,6 @@ static inline int validate_convert_profile(struct btrfs_fs_info *fs_info,
 	if (!(bargs->flags & BTRFS_BALANCE_ARGS_CONVERT))
 		return true;
 
-	if (fs_info->sectorsize < PAGE_SIZE &&
-		bargs->target & BTRFS_BLOCK_GROUP_RAID56_MASK) {
-		btrfs_err(fs_info,
-		"RAID56 is not yet supported for sectorsize %u with page size %lu",
-			  fs_info->sectorsize, PAGE_SIZE);
-		return false;
-	}
 	/* Profile is valid and does not have bits outside of the allowed set */
 	if (alloc_profile_is_valid(bargs->target, 1) &&
 	    (bargs->target & ~allowed) == 0)
@@ -6312,7 +6293,7 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, struct extent_map *em,
 	u64 offset;
 	u64 stripe_offset;
 	u64 stripe_nr;
-	u64 stripe_len;
+	u32 stripe_len;
 	u64 raid56_full_stripe_start = (u64)-1;
 	int data_stripes;
 
@@ -6323,19 +6304,13 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, struct extent_map *em,
 	offset = logical - em->start;
 	/* Len of a stripe in a chunk */
 	stripe_len = map->stripe_len;
-	/* Stripe where this block falls in */
-	stripe_nr = div64_u64(offset, stripe_len);
-	/* Offset of stripe in the chunk */
-	stripe_offset = stripe_nr * stripe_len;
-	if (offset < stripe_offset) {
-		btrfs_crit(fs_info,
-"stripe math has gone wrong, stripe_offset=%llu offset=%llu start=%llu logical=%llu stripe_len=%llu",
-			stripe_offset, offset, em->start, logical, stripe_len);
-		return -EINVAL;
-	}
+	/*
+	 * Stripe_nr is where this block falls in
+	 * stripe_offset is the offset of this block in its stripe.
+	 */
+	stripe_nr = div64_u64_rem(offset, stripe_len, &stripe_offset);
+	ASSERT(stripe_offset < U32_MAX);
 
-	/* stripe_offset is the offset of this block in its stripe */
-	stripe_offset = offset - stripe_offset;
 	data_stripes = nr_data_stripes(map);
 
 	/* Only stripe based profiles needs to check against stripe length. */
@@ -6737,11 +6712,11 @@ static void submit_stripe_bio(struct btrfs_io_context *bioc, struct bio *bio,
 		bio_op(bio), bio->bi_opf, bio->bi_iter.bi_sector,
 		(unsigned long)dev->bdev->bd_dev, rcu_str_deref(dev->name),
 		dev->devid, bio->bi_iter.bi_size);
-	bio_set_dev(bio, dev->bdev);
 
 	btrfs_bio_counter_inc_noblocked(fs_info);
 
-	btrfsic_submit_bio(bio);
+	btrfsic_check_bio(bio);
+	submit_bio(bio);
 }
 
 static void bioc_error(struct btrfs_io_context *bioc, struct bio *bio, u64 logical)
@@ -6823,10 +6798,12 @@ blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 			continue;
 		}
 
-		if (dev_nr < total_devs - 1)
-			bio = btrfs_bio_clone(first_bio);
-		else
+		if (dev_nr < total_devs - 1) {
+			bio = btrfs_bio_clone(dev->bdev, first_bio);
+		} else {
 			bio = first_bio;
+			bio_set_dev(bio, dev->bdev);
+		}
 
 		submit_stripe_bio(bioc, bio, bioc->stripes[dev_nr].physical, dev);
 	}
@@ -7359,7 +7336,6 @@ static int read_one_dev(struct extent_buffer *leaf,
 
 int btrfs_read_sys_array(struct btrfs_fs_info *fs_info)
 {
-	struct btrfs_root *root = fs_info->tree_root;
 	struct btrfs_super_block *super_copy = fs_info->super_copy;
 	struct extent_buffer *sb;
 	struct btrfs_disk_key *disk_key;
@@ -7375,30 +7351,16 @@ int btrfs_read_sys_array(struct btrfs_fs_info *fs_info)
 	struct btrfs_key key;
 
 	ASSERT(BTRFS_SUPER_INFO_SIZE <= fs_info->nodesize);
+
 	/*
-	 * This will create extent buffer of nodesize, superblock size is
-	 * fixed to BTRFS_SUPER_INFO_SIZE. If nodesize > sb size, this will
-	 * overallocate but we can keep it as-is, only the first page is used.
+	 * We allocated a dummy extent, just to use extent buffer accessors.
+	 * There will be unused space after BTRFS_SUPER_INFO_SIZE, but
+	 * that's fine, we will not go beyond system chunk array anyway.
 	 */
-	sb = btrfs_find_create_tree_block(fs_info, BTRFS_SUPER_INFO_OFFSET,
-					  root->root_key.objectid, 0);
-	if (IS_ERR(sb))
-		return PTR_ERR(sb);
+	sb = alloc_dummy_extent_buffer(fs_info, BTRFS_SUPER_INFO_OFFSET);
+	if (!sb)
+		return -ENOMEM;
 	set_extent_buffer_uptodate(sb);
-	/*
-	 * The sb extent buffer is artificial and just used to read the system array.
-	 * set_extent_buffer_uptodate() call does not properly mark all it's
-	 * pages up-to-date when the page is larger: extent does not cover the
-	 * whole page and consequently check_page_uptodate does not find all
-	 * the page's extents up-to-date (the hole beyond sb),
-	 * write_extent_buffer then triggers a WARN_ON.
-	 *
-	 * Regular short extents go through mark_extent_buffer_dirty/writeback cycle,
-	 * but sb spans only this function. Add an explicit SetPageUptodate call
-	 * to silence the warning eg. on PowerPC 64.
-	 */
-	if (PAGE_SIZE > BTRFS_SUPER_INFO_SIZE)
-		SetPageUptodate(sb->pages[0]);
 
 	write_extent_buffer(sb, super_copy, 0, BTRFS_SUPER_INFO_SIZE);
 	array_size = btrfs_super_sys_array_size(super_copy);
@@ -7561,6 +7523,7 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info)
 	struct btrfs_key found_key;
 	int ret;
 	int slot;
+	int iter_ret = 0;
 	u64 total_dev = 0;
 	u64 last_ra_node = 0;
 
@@ -7604,30 +7567,18 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info)
 	key.objectid = BTRFS_DEV_ITEMS_OBJECTID;
 	key.offset = 0;
 	key.type = 0;
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto error;
-	while (1) {
-		struct extent_buffer *node;
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
+		struct extent_buffer *node = path->nodes[1];
 
 		leaf = path->nodes[0];
 		slot = path->slots[0];
-		if (slot >= btrfs_header_nritems(leaf)) {
-			ret = btrfs_next_leaf(root, path);
-			if (ret == 0)
-				continue;
-			if (ret < 0)
-				goto error;
-			break;
-		}
-		node = path->nodes[1];
+
 		if (node) {
 			if (last_ra_node != node->start) {
 				readahead_tree_node_children(node);
 				last_ra_node = node->start;
 			}
 		}
-		btrfs_item_key_to_cpu(leaf, &found_key, slot);
 		if (found_key.type == BTRFS_DEV_ITEM_KEY) {
 			struct btrfs_dev_item *dev_item;
 			dev_item = btrfs_item_ptr(leaf, slot,
@@ -7652,7 +7603,11 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info)
 			if (ret)
 				goto error;
 		}
-		path->slots[0]++;
+	}
+	/* Catch error found during iteration */
+	if (iter_ret < 0) {
+		ret = iter_ret;
+		goto error;
 	}
 
 	/*
@@ -7660,12 +7615,12 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info)
 	 * do another round of validation checks.
 	 */
 	if (total_dev != fs_info->fs_devices->total_devices) {
-		btrfs_err(fs_info,
-	   "super_num_devices %llu mismatch with num_devices %llu found here",
+		btrfs_warn(fs_info,
+"super block num_devices %llu mismatch with DEV_ITEM count %llu, will be repaired on next transaction commit",
 			  btrfs_super_num_devices(fs_info->super_copy),
 			  total_dev);
-		ret = -EINVAL;
-		goto error;
+		fs_info->fs_devices->total_devices = total_dev;
+		btrfs_set_super_num_devices(fs_info->super_copy, total_dev);
 	}
 	if (btrfs_super_total_bytes(fs_info->super_copy) <
 	    fs_info->fs_devices->total_rw_bytes) {
@@ -8277,7 +8232,7 @@ bool btrfs_pinned_by_swapfile(struct btrfs_fs_info *fs_info, void *ptr)
 
 static int relocating_repair_kthread(void *data)
 {
-	struct btrfs_block_group *cache = (struct btrfs_block_group *)data;
+	struct btrfs_block_group *cache = data;
 	struct btrfs_fs_info *fs_info = cache->fs_info;
 	u64 target;
 	int ret = 0;
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index b11c563d2025..6721002000ee 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -17,17 +17,51 @@ extern struct mutex uuid_mutex;
 
 #define BTRFS_STRIPE_LEN	SZ_64K
 
+/* Used by sanity check for btrfs_raid_types. */
+#define const_ffs(n) (__builtin_ctzll(n) + 1)
+
+/*
+ * The conversion from BTRFS_BLOCK_GROUP_* bits to btrfs_raid_type requires
+ * RAID0 always to be the lowest profile bit.
+ * Although it's part of on-disk format and should never change, do extra
+ * compile-time sanity checks.
+ */
+static_assert(const_ffs(BTRFS_BLOCK_GROUP_RAID0) <
+	      const_ffs(BTRFS_BLOCK_GROUP_PROFILE_MASK & ~BTRFS_BLOCK_GROUP_RAID0));
+static_assert(const_ilog2(BTRFS_BLOCK_GROUP_RAID0) >
+	      ilog2(BTRFS_BLOCK_GROUP_TYPE_MASK));
+
+/* ilog2() can handle both constants and variables */
+#define BTRFS_BG_FLAG_TO_INDEX(profile)					\
+	ilog2((profile) >> (ilog2(BTRFS_BLOCK_GROUP_RAID0) - 1))
+
+enum btrfs_raid_types {
+	/* SINGLE is the special one as it doesn't have on-disk bit. */
+	BTRFS_RAID_SINGLE  = 0,
+
+	BTRFS_RAID_RAID0   = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID0),
+	BTRFS_RAID_RAID1   = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID1),
+	BTRFS_RAID_DUP	   = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_DUP),
+	BTRFS_RAID_RAID10  = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID10),
+	BTRFS_RAID_RAID5   = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID5),
+	BTRFS_RAID_RAID6   = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID6),
+	BTRFS_RAID_RAID1C3 = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID1C3),
+	BTRFS_RAID_RAID1C4 = BTRFS_BG_FLAG_TO_INDEX(BTRFS_BLOCK_GROUP_RAID1C4),
+
+	BTRFS_NR_RAID_TYPES
+};
+
 struct btrfs_io_geometry {
 	/* remaining bytes before crossing a stripe */
 	u64 len;
 	/* offset of logical address in chunk */
 	u64 offset;
 	/* length of single IO stripe */
-	u64 stripe_len;
+	u32 stripe_len;
+	/* offset of address in stripe */
+	u32 stripe_offset;
 	/* number of stripe where address falls */
 	u64 stripe_nr;
-	/* offset of address in stripe */
-	u64 stripe_offset;
 	/* offset of raid56 stripe into the chunk */
 	u64 raid56_stripe_offset;
 };
@@ -430,7 +464,7 @@ struct map_lookup {
 	u64 type;
 	int io_align;
 	int io_width;
-	u64 stripe_len;
+	u32 stripe_len;
 	int num_stripes;
 	int sub_stripes;
 	int verified_stripes; /* For mount time dev extent verification */
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index 85691dc2232f..7421abcf325a 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -272,10 +272,12 @@ out:
 
 ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 {
+	struct btrfs_key found_key;
 	struct btrfs_key key;
 	struct inode *inode = d_inode(dentry);
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_path *path;
+	int iter_ret = 0;
 	int ret = 0;
 	size_t total_size = 0, size_left = size;
 
@@ -294,44 +296,23 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 	path->reada = READA_FORWARD;
 
 	/* search for our xattrs */
-	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-	if (ret < 0)
-		goto err;
-
-	while (1) {
+	btrfs_for_each_slot(root, &key, &found_key, path, iter_ret) {
 		struct extent_buffer *leaf;
 		int slot;
 		struct btrfs_dir_item *di;
-		struct btrfs_key found_key;
 		u32 item_size;
 		u32 cur;
 
 		leaf = path->nodes[0];
 		slot = path->slots[0];
 
-		/* this is where we start walking through the path */
-		if (slot >= btrfs_header_nritems(leaf)) {
-			/*
-			 * if we've reached the last slot in this leaf we need
-			 * to go to the next leaf and reset everything
-			 */
-			ret = btrfs_next_leaf(root, path);
-			if (ret < 0)
-				goto err;
-			else if (ret > 0)
-				break;
-			continue;
-		}
-
-		btrfs_item_key_to_cpu(leaf, &found_key, slot);
-
 		/* check to make sure this item is what we want */
 		if (found_key.objectid != key.objectid)
 			break;
 		if (found_key.type > BTRFS_XATTR_ITEM_KEY)
 			break;
 		if (found_key.type < BTRFS_XATTR_ITEM_KEY)
-			goto next_item;
+			continue;
 
 		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
 		item_size = btrfs_item_size(leaf, slot);
@@ -351,8 +332,8 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 				goto next;
 
 			if (!buffer || (name_len + 1) > size_left) {
-				ret = -ERANGE;
-				goto err;
+			        iter_ret = -ERANGE;
+				break;
 			}
 
 			read_extent_buffer(leaf, buffer, name_ptr, name_len);
@@ -364,12 +345,13 @@ next:
 			cur += this_len;
 			di = (struct btrfs_dir_item *)((char *)di + this_len);
 		}
-next_item:
-		path->slots[0]++;
 	}
-	ret = total_size;
 
-err:
+	if (iter_ret < 0)
+		ret = iter_ret;
+	else
+		ret = total_size;
+
 	btrfs_free_path(path);
 
 	return ret;
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index 29b54fd9c128..11237a913bee 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -51,11 +51,13 @@
 #define BTRFS_MIN_ACTIVE_ZONES		(BTRFS_SUPER_MIRROR_MAX + 5)
 
 /*
- * Maximum supported zone size. Currently, SMR disks have a zone size of
- * 256MiB, and we are expecting ZNS drives to be in the 1-4GiB range. We do not
- * expect the zone size to become larger than 8GiB in the near future.
+ * Minimum / maximum supported zone size. Currently, SMR disks have a zone
+ * size of 256MiB, and we are expecting ZNS drives to be in the 1-4GiB range.
+ * We do not expect the zone size to become larger than 8GiB or smaller than
+ * 4MiB in the near future.
  */
 #define BTRFS_MAX_ZONE_SIZE		SZ_8G
+#define BTRFS_MIN_ZONE_SIZE		SZ_4M
 
 #define SUPER_INFO_SECTORS	((u64)BTRFS_SUPER_INFO_SIZE >> SECTOR_SHIFT)
 
@@ -401,6 +403,13 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache)
 				 zone_info->zone_size, BTRFS_MAX_ZONE_SIZE);
 		ret = -EINVAL;
 		goto out;
+	} else if (zone_info->zone_size < BTRFS_MIN_ZONE_SIZE) {
+		btrfs_err_in_rcu(fs_info,
+		"zoned: %s: zone size %llu smaller than supported minimum %u",
+				 rcu_str_deref(device->name),
+				 zone_info->zone_size, BTRFS_MIN_ZONE_SIZE);
+		ret = -EINVAL;
+		goto out;
 	}
 
 	nr_sectors = bdev_nr_sectors(bdev);
@@ -1835,7 +1844,7 @@ bool btrfs_zone_activate(struct btrfs_block_group *block_group)
 	}
 
 	/* No space left */
-	if (block_group->alloc_offset == block_group->zone_capacity) {
+	if (btrfs_zoned_bg_is_full(block_group)) {
 		ret = false;
 		goto out_unlock;
 	}
@@ -1872,20 +1881,14 @@ out_unlock:
 	return ret;
 }
 
-int btrfs_zone_finish(struct btrfs_block_group *block_group)
+static int do_zone_finish(struct btrfs_block_group *block_group, bool fully_written)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	struct map_lookup *map;
-	struct btrfs_device *device;
-	u64 physical;
+	bool need_zone_finish;
 	int ret = 0;
 	int i;
 
-	if (!btrfs_is_zoned(fs_info))
-		return 0;
-
-	map = block_group->physical_map;
-
 	spin_lock(&block_group->lock);
 	if (!block_group->zone_is_active) {
 		spin_unlock(&block_group->lock);
@@ -1895,40 +1898,56 @@ int btrfs_zone_finish(struct btrfs_block_group *block_group)
 	/* Check if we have unwritten allocated space */
 	if ((block_group->flags &
 	     (BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_SYSTEM)) &&
-	    block_group->alloc_offset > block_group->meta_write_pointer) {
+	    block_group->start + block_group->alloc_offset > block_group->meta_write_pointer) {
 		spin_unlock(&block_group->lock);
 		return -EAGAIN;
 	}
-	spin_unlock(&block_group->lock);
-
-	ret = btrfs_inc_block_group_ro(block_group, false);
-	if (ret)
-		return ret;
-
-	/* Ensure all writes in this block group finish */
-	btrfs_wait_block_group_reservations(block_group);
-	/* No need to wait for NOCOW writers. Zoned mode does not allow that. */
-	btrfs_wait_ordered_roots(fs_info, U64_MAX, block_group->start,
-				 block_group->length);
-
-	spin_lock(&block_group->lock);
 
 	/*
-	 * Bail out if someone already deactivated the block group, or
-	 * allocated space is left in the block group.
+	 * If we are sure that the block group is full (= no more room left for
+	 * new allocation) and the IO for the last usable block is completed, we
+	 * don't need to wait for the other IOs. This holds because we ensure
+	 * the sequential IO submissions using the ZONE_APPEND command for data
+	 * and block_group->meta_write_pointer for metadata.
 	 */
-	if (!block_group->zone_is_active) {
+	if (!fully_written) {
 		spin_unlock(&block_group->lock);
-		btrfs_dec_block_group_ro(block_group);
-		return 0;
-	}
 
-	if (block_group->reserved) {
-		spin_unlock(&block_group->lock);
-		btrfs_dec_block_group_ro(block_group);
-		return -EAGAIN;
+		ret = btrfs_inc_block_group_ro(block_group, false);
+		if (ret)
+			return ret;
+
+		/* Ensure all writes in this block group finish */
+		btrfs_wait_block_group_reservations(block_group);
+		/* No need to wait for NOCOW writers. Zoned mode does not allow that */
+		btrfs_wait_ordered_roots(fs_info, U64_MAX, block_group->start,
+					 block_group->length);
+
+		spin_lock(&block_group->lock);
+
+		/*
+		 * Bail out if someone already deactivated the block group, or
+		 * allocated space is left in the block group.
+		 */
+		if (!block_group->zone_is_active) {
+			spin_unlock(&block_group->lock);
+			btrfs_dec_block_group_ro(block_group);
+			return 0;
+		}
+
+		if (block_group->reserved) {
+			spin_unlock(&block_group->lock);
+			btrfs_dec_block_group_ro(block_group);
+			return -EAGAIN;
+		}
 	}
 
+	/*
+	 * The block group is not fully allocated, so not fully written yet. We
+	 * need to send ZONE_FINISH command to free up an active zone.
+	 */
+	need_zone_finish = !btrfs_zoned_bg_is_full(block_group);
+
 	block_group->zone_is_active = 0;
 	block_group->alloc_offset = block_group->zone_capacity;
 	block_group->free_space_ctl->free_space = 0;
@@ -1936,24 +1955,29 @@ int btrfs_zone_finish(struct btrfs_block_group *block_group)
 	btrfs_clear_data_reloc_bg(block_group);
 	spin_unlock(&block_group->lock);
 
+	map = block_group->physical_map;
 	for (i = 0; i < map->num_stripes; i++) {
-		device = map->stripes[i].dev;
-		physical = map->stripes[i].physical;
+		struct btrfs_device *device = map->stripes[i].dev;
+		const u64 physical = map->stripes[i].physical;
 
 		if (device->zone_info->max_active_zones == 0)
 			continue;
 
-		ret = blkdev_zone_mgmt(device->bdev, REQ_OP_ZONE_FINISH,
-				       physical >> SECTOR_SHIFT,
-				       device->zone_info->zone_size >> SECTOR_SHIFT,
-				       GFP_NOFS);
+		if (need_zone_finish) {
+			ret = blkdev_zone_mgmt(device->bdev, REQ_OP_ZONE_FINISH,
+					       physical >> SECTOR_SHIFT,
+					       device->zone_info->zone_size >> SECTOR_SHIFT,
+					       GFP_NOFS);
 
-		if (ret)
-			return ret;
+			if (ret)
+				return ret;
+		}
 
 		btrfs_dev_clear_active_zone(device, physical);
 	}
-	btrfs_dec_block_group_ro(block_group);
+
+	if (!fully_written)
+		btrfs_dec_block_group_ro(block_group);
 
 	spin_lock(&fs_info->zone_active_bgs_lock);
 	ASSERT(!list_empty(&block_group->active_bg_list));
@@ -1966,6 +1990,14 @@ int btrfs_zone_finish(struct btrfs_block_group *block_group)
 	return 0;
 }
 
+int btrfs_zone_finish(struct btrfs_block_group *block_group)
+{
+	if (!btrfs_is_zoned(block_group->fs_info))
+		return 0;
+
+	return do_zone_finish(block_group, false);
+}
+
 bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices, u64 flags)
 {
 	struct btrfs_fs_info *fs_info = fs_devices->fs_info;
@@ -1997,9 +2029,7 @@ bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices, u64 flags)
 void btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical, u64 length)
 {
 	struct btrfs_block_group *block_group;
-	struct map_lookup *map;
-	struct btrfs_device *device;
-	u64 physical;
+	u64 min_alloc_bytes;
 
 	if (!btrfs_is_zoned(fs_info))
 		return;
@@ -2007,42 +2037,52 @@ void btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical, u64 len
 	block_group = btrfs_lookup_block_group(fs_info, logical);
 	ASSERT(block_group);
 
-	if (logical + length < block_group->start + block_group->zone_capacity)
-		goto out;
-
-	spin_lock(&block_group->lock);
+	/* No MIXED_BG on zoned btrfs. */
+	if (block_group->flags & BTRFS_BLOCK_GROUP_DATA)
+		min_alloc_bytes = fs_info->sectorsize;
+	else
+		min_alloc_bytes = fs_info->nodesize;
 
-	if (!block_group->zone_is_active) {
-		spin_unlock(&block_group->lock);
+	/* Bail out if we can allocate more data from this block group. */
+	if (logical + length + min_alloc_bytes <=
+	    block_group->start + block_group->zone_capacity)
 		goto out;
-	}
 
-	block_group->zone_is_active = 0;
-	/* We should have consumed all the free space */
-	ASSERT(block_group->alloc_offset == block_group->zone_capacity);
-	ASSERT(block_group->free_space_ctl->free_space == 0);
-	btrfs_clear_treelog_bg(block_group);
-	btrfs_clear_data_reloc_bg(block_group);
-	spin_unlock(&block_group->lock);
+	do_zone_finish(block_group, true);
 
-	map = block_group->physical_map;
-	device = map->stripes[0].dev;
-	physical = map->stripes[0].physical;
+out:
+	btrfs_put_block_group(block_group);
+}
 
-	if (!device->zone_info->max_active_zones)
-		goto out;
+static void btrfs_zone_finish_endio_workfn(struct work_struct *work)
+{
+	struct btrfs_block_group *bg =
+		container_of(work, struct btrfs_block_group, zone_finish_work);
 
-	btrfs_dev_clear_active_zone(device, physical);
+	wait_on_extent_buffer_writeback(bg->last_eb);
+	free_extent_buffer(bg->last_eb);
+	btrfs_zone_finish_endio(bg->fs_info, bg->start, bg->length);
+	btrfs_put_block_group(bg);
+}
 
-	spin_lock(&fs_info->zone_active_bgs_lock);
-	ASSERT(!list_empty(&block_group->active_bg_list));
-	list_del_init(&block_group->active_bg_list);
-	spin_unlock(&fs_info->zone_active_bgs_lock);
+void btrfs_schedule_zone_finish_bg(struct btrfs_block_group *bg,
+				   struct extent_buffer *eb)
+{
+	if (!bg->seq_zone || eb->start + eb->len * 2 <= bg->start + bg->zone_capacity)
+		return;
 
-	btrfs_put_block_group(block_group);
+	if (WARN_ON(bg->zone_finish_work.func == btrfs_zone_finish_endio_workfn)) {
+		btrfs_err(bg->fs_info, "double scheduling of bg %llu zone finishing",
+			  bg->start);
+		return;
+	}
 
-out:
-	btrfs_put_block_group(block_group);
+	/* For the work */
+	btrfs_get_block_group(bg);
+	atomic_inc(&eb->refs);
+	bg->last_eb = eb;
+	INIT_WORK(&bg->zone_finish_work, btrfs_zone_finish_endio_workfn);
+	queue_work(system_unbound_wq, &bg->zone_finish_work);
 }
 
 void btrfs_clear_data_reloc_bg(struct btrfs_block_group *bg)
@@ -2072,3 +2112,30 @@ void btrfs_free_zone_cache(struct btrfs_fs_info *fs_info)
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
 }
+
+bool btrfs_zoned_should_reclaim(struct btrfs_fs_info *fs_info)
+{
+	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
+	struct btrfs_device *device;
+	u64 used = 0;
+	u64 total = 0;
+	u64 factor;
+
+	ASSERT(btrfs_is_zoned(fs_info));
+
+	if (fs_info->bg_reclaim_threshold == 0)
+		return false;
+
+	mutex_lock(&fs_devices->device_list_mutex);
+	list_for_each_entry(device, &fs_devices->devices, dev_list) {
+		if (!device->bdev)
+			continue;
+
+		total += device->disk_total_bytes;
+		used += device->bytes_used;
+	}
+	mutex_unlock(&fs_devices->device_list_mutex);
+
+	factor = div64_u64(used * 100, total);
+	return factor >= fs_info->bg_reclaim_threshold;
+}
diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h
index 6dee76248cb4..bb1a189e11f9 100644
--- a/fs/btrfs/zoned.h
+++ b/fs/btrfs/zoned.h
@@ -10,11 +10,7 @@
 #include "block-group.h"
 #include "btrfs_inode.h"
 
-/*
- * Block groups with more than this value (percents) of unusable space will be
- * scheduled for background reclaim.
- */
-#define BTRFS_DEFAULT_RECLAIM_THRESH		75
+#define BTRFS_DEFAULT_RECLAIM_THRESH           			(75)
 
 struct btrfs_zoned_device_info {
 	/*
@@ -76,8 +72,11 @@ int btrfs_zone_finish(struct btrfs_block_group *block_group);
 bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices, u64 flags);
 void btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info, u64 logical,
 			     u64 length);
+void btrfs_schedule_zone_finish_bg(struct btrfs_block_group *bg,
+				   struct extent_buffer *eb);
 void btrfs_clear_data_reloc_bg(struct btrfs_block_group *bg);
 void btrfs_free_zone_cache(struct btrfs_fs_info *fs_info);
+bool btrfs_zoned_should_reclaim(struct btrfs_fs_info *fs_info);
 #else /* CONFIG_BLK_DEV_ZONED */
 static inline int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos,
 				     struct blk_zone *zone)
@@ -233,9 +232,17 @@ static inline bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices,
 static inline void btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info,
 					   u64 logical, u64 length) { }
 
+static inline void btrfs_schedule_zone_finish_bg(struct btrfs_block_group *bg,
+						 struct extent_buffer *eb) { }
+
 static inline void btrfs_clear_data_reloc_bg(struct btrfs_block_group *bg) { }
 
 static inline void btrfs_free_zone_cache(struct btrfs_fs_info *fs_info) { }
+
+static inline bool btrfs_zoned_should_reclaim(struct btrfs_fs_info *fs_info)
+{
+	return false;
+}
 #endif
 
 static inline bool btrfs_dev_is_sequential(struct btrfs_device *device, u64 pos)
@@ -370,4 +377,10 @@ static inline void btrfs_zoned_data_reloc_unlock(struct btrfs_inode *inode)
 		mutex_unlock(&root->fs_info->zoned_data_reloc_io_lock);
 }
 
+static inline bool btrfs_zoned_bg_is_full(const struct btrfs_block_group *bg)
+{
+	ASSERT(btrfs_is_zoned(bg->fs_info));
+	return (bg->alloc_offset == bg->zone_capacity);
+}
+
 #endif
diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c
index fc42dd0badd7..0fe31a6f6e68 100644
--- a/fs/btrfs/zstd.c
+++ b/fs/btrfs/zstd.c
@@ -93,22 +93,26 @@ static inline struct workspace *list_to_workspace(struct list_head *list)
 
 void zstd_free_workspace(struct list_head *ws);
 struct list_head *zstd_alloc_workspace(unsigned int level);
-/*
- * zstd_reclaim_timer_fn - reclaim timer
+
+/**
+ * Timer callback to free unused workspaces.
+ *
  * @t: timer
  *
  * This scans the lru_list and attempts to reclaim any workspace that hasn't
  * been used for ZSTD_BTRFS_RECLAIM_JIFFIES.
+ *
+ * The context is softirq and does not need the _bh locking primitives.
  */
 static void zstd_reclaim_timer_fn(struct timer_list *timer)
 {
 	unsigned long reclaim_threshold = jiffies - ZSTD_BTRFS_RECLAIM_JIFFIES;
 	struct list_head *pos, *next;
 
-	spin_lock_bh(&wsm.lock);
+	spin_lock(&wsm.lock);
 
 	if (list_empty(&wsm.lru_list)) {
-		spin_unlock_bh(&wsm.lock);
+		spin_unlock(&wsm.lock);
 		return;
 	}
 
@@ -137,7 +141,7 @@ static void zstd_reclaim_timer_fn(struct timer_list *timer)
 	if (!list_empty(&wsm.lru_list))
 		mod_timer(&wsm.timer, jiffies + ZSTD_BTRFS_RECLAIM_JIFFIES);
 
-	spin_unlock_bh(&wsm.lock);
+	spin_unlock(&wsm.lock);
 }
 
 /*
diff --git a/fs/erofs/data.c b/fs/erofs/data.c
index bb9c1fd48c19..252f4ee977d5 100644
--- a/fs/erofs/data.c
+++ b/fs/erofs/data.c
@@ -399,7 +399,7 @@ static ssize_t erofs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 
 		if (!err)
 			return iomap_dio_rw(iocb, to, &erofs_iomap_ops,
-					    NULL, 0, 0);
+					    NULL, 0, NULL, 0);
 		if (err < 0)
 			return err;
 	}
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 6feb07e3e1eb..109d07629f81 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -76,7 +76,7 @@ static ssize_t ext4_dio_read_iter(struct kiocb *iocb, struct iov_iter *to)
 		return generic_file_read_iter(iocb, to);
 	}
 
-	ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL, 0, 0);
+	ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL, 0, NULL, 0);
 	inode_unlock_shared(inode);
 
 	file_accessed(iocb->ki_filp);
@@ -565,7 +565,7 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		iomap_ops = &ext4_iomap_overwrite_ops;
 	ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops,
 			   (unaligned_io || extend) ? IOMAP_DIO_FORCE_WAIT : 0,
-			   0);
+			   NULL, 0);
 	if (ret == -ENOTBLK)
 		ret = 0;
 
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 35b6c720c2bc..100637b1adb3 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -4308,7 +4308,7 @@ static ssize_t f2fs_dio_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	 */
 	inc_page_count(sbi, F2FS_DIO_READ);
 	dio = __iomap_dio_rw(iocb, to, &f2fs_iomap_ops,
-			     &f2fs_iomap_dio_read_ops, 0, 0);
+			     &f2fs_iomap_dio_read_ops, 0, NULL, 0);
 	if (IS_ERR_OR_NULL(dio)) {
 		ret = PTR_ERR_OR_ZERO(dio);
 		if (ret != -EIOCBQUEUED)
@@ -4526,7 +4526,7 @@ static ssize_t f2fs_dio_write_iter(struct kiocb *iocb, struct iov_iter *from,
 	if (pos + count > inode->i_size)
 		dio_flags |= IOMAP_DIO_FORCE_WAIT;
 	dio = __iomap_dio_rw(iocb, from, &f2fs_iomap_ops,
-			     &f2fs_iomap_dio_write_ops, dio_flags, 0);
+			     &f2fs_iomap_dio_write_ops, dio_flags, NULL, 0);
 	if (IS_ERR_OR_NULL(dio)) {
 		ret = PTR_ERR_OR_ZERO(dio);
 		if (ret == -ENOTBLK)
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 2556ae1f92ea..0b07d5a7bb81 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -835,7 +835,7 @@ retry:
 	pagefault_disable();
 	to->nofault = true;
 	ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL,
-			   IOMAP_DIO_PARTIAL, read);
+			   IOMAP_DIO_PARTIAL, NULL, read);
 	to->nofault = false;
 	pagefault_enable();
 	if (ret <= 0 && ret != -EFAULT)
@@ -898,7 +898,7 @@ retry:
 
 	from->nofault = true;
 	ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL,
-			   IOMAP_DIO_PARTIAL, written);
+			   IOMAP_DIO_PARTIAL, NULL, written);
 	from->nofault = false;
 	if (ret <= 0) {
 		if (ret == -ENOTBLK)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 80f9b047aa1b..370c3241618a 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -51,6 +51,15 @@ struct iomap_dio {
 	};
 };
 
+static struct bio *iomap_dio_alloc_bio(const struct iomap_iter *iter,
+		struct iomap_dio *dio, unsigned short nr_vecs, unsigned int opf)
+{
+	if (dio->dops && dio->dops->bio_set)
+		return bio_alloc_bioset(iter->iomap.bdev, nr_vecs, opf,
+					GFP_KERNEL, dio->dops->bio_set);
+	return bio_alloc(iter->iomap.bdev, nr_vecs, opf, GFP_KERNEL);
+}
+
 static void iomap_dio_submit_bio(const struct iomap_iter *iter,
 		struct iomap_dio *dio, struct bio *bio, loff_t pos)
 {
@@ -145,7 +154,7 @@ static inline void iomap_dio_set_error(struct iomap_dio *dio, int ret)
 	cmpxchg(&dio->error, 0, ret);
 }
 
-static void iomap_dio_bio_end_io(struct bio *bio)
+void iomap_dio_bio_end_io(struct bio *bio)
 {
 	struct iomap_dio *dio = bio->bi_private;
 	bool should_dirty = (dio->flags & IOMAP_DIO_DIRTY);
@@ -177,16 +186,16 @@ static void iomap_dio_bio_end_io(struct bio *bio)
 		bio_put(bio);
 	}
 }
+EXPORT_SYMBOL_GPL(iomap_dio_bio_end_io);
 
 static void iomap_dio_zero(const struct iomap_iter *iter, struct iomap_dio *dio,
 		loff_t pos, unsigned len)
 {
 	struct inode *inode = file_inode(dio->iocb->ki_filp);
 	struct page *page = ZERO_PAGE(0);
-	int flags = REQ_SYNC | REQ_IDLE;
 	struct bio *bio;
 
-	bio = bio_alloc(iter->iomap.bdev, 1, REQ_OP_WRITE | flags, GFP_KERNEL);
+	bio = iomap_dio_alloc_bio(iter, dio, 1, REQ_OP_WRITE | REQ_SYNC | REQ_IDLE);
 	fscrypt_set_bio_crypt_ctx(bio, inode, pos >> inode->i_blkbits,
 				  GFP_KERNEL);
 	bio->bi_iter.bi_sector = iomap_sector(&iter->iomap, pos);
@@ -311,7 +320,7 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
 			goto out;
 		}
 
-		bio = bio_alloc(iomap->bdev, nr_pages, bio_opf, GFP_KERNEL);
+		bio = iomap_dio_alloc_bio(iter, dio, nr_pages, bio_opf);
 		fscrypt_set_bio_crypt_ctx(bio, inode, pos >> inode->i_blkbits,
 					  GFP_KERNEL);
 		bio->bi_iter.bi_sector = iomap_sector(iomap, pos);
@@ -474,7 +483,7 @@ static loff_t iomap_dio_iter(const struct iomap_iter *iter,
 struct iomap_dio *
 __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
-		unsigned int dio_flags, size_t done_before)
+		unsigned int dio_flags, void *private, size_t done_before)
 {
 	struct address_space *mapping = iocb->ki_filp->f_mapping;
 	struct inode *inode = file_inode(iocb->ki_filp);
@@ -483,6 +492,7 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		.pos		= iocb->ki_pos,
 		.len		= iov_iter_count(iter),
 		.flags		= IOMAP_DIRECT,
+		.private	= private,
 	};
 	loff_t end = iomi.pos + iomi.len - 1, ret = 0;
 	bool wait_for_completion =
@@ -672,11 +682,12 @@ EXPORT_SYMBOL_GPL(__iomap_dio_rw);
 ssize_t
 iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
-		unsigned int dio_flags, size_t done_before)
+		unsigned int dio_flags, void *private, size_t done_before)
 {
 	struct iomap_dio *dio;
 
-	dio = __iomap_dio_rw(iocb, iter, ops, dops, dio_flags, done_before);
+	dio = __iomap_dio_rw(iocb, iter, ops, dops, dio_flags, private,
+			     done_before);
 	if (IS_ERR_OR_NULL(dio))
 		return PTR_ERR_OR_ZERO(dio);
 	return iomap_dio_complete(dio);
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 5bddb1e9e0b3..85c412107a10 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -225,7 +225,7 @@ xfs_file_dio_read(
 	ret = xfs_ilock_iocb(iocb, XFS_IOLOCK_SHARED);
 	if (ret)
 		return ret;
-	ret = iomap_dio_rw(iocb, to, &xfs_read_iomap_ops, NULL, 0, 0);
+	ret = iomap_dio_rw(iocb, to, &xfs_read_iomap_ops, NULL, 0, NULL, 0);
 	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
 	return ret;
@@ -534,7 +534,7 @@ xfs_file_dio_write_aligned(
 	}
 	trace_xfs_file_direct_write(iocb, from);
 	ret = iomap_dio_rw(iocb, from, &xfs_direct_write_iomap_ops,
-			   &xfs_dio_write_ops, 0, 0);
+			   &xfs_dio_write_ops, 0, NULL, 0);
 out_unlock:
 	if (iolock)
 		xfs_iunlock(ip, iolock);
@@ -612,7 +612,7 @@ retry_exclusive:
 
 	trace_xfs_file_direct_write(iocb, from);
 	ret = iomap_dio_rw(iocb, from, &xfs_direct_write_iomap_ops,
-			   &xfs_dio_write_ops, flags, 0);
+			   &xfs_dio_write_ops, flags, NULL, 0);
 
 	/*
 	 * Retry unaligned I/O with exclusive blocking semantics if the DIO
diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c
index 652752df1a2f..8f306485c953 100644
--- a/fs/zonefs/super.c
+++ b/fs/zonefs/super.c
@@ -900,7 +900,7 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
 		ret = zonefs_file_dio_append(iocb, from);
 	else
 		ret = iomap_dio_rw(iocb, from, &zonefs_iomap_ops,
-				   &zonefs_write_dio_ops, 0, 0);
+				   &zonefs_write_dio_ops, 0, NULL, 0);
 	if (zi->i_ztype == ZONEFS_ZTYPE_SEQ &&
 	    (ret > 0 || ret == -EIOCBQUEUED)) {
 		if (ret > 0)
@@ -1042,7 +1042,7 @@ static ssize_t zonefs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 		}
 		file_accessed(iocb->ki_filp);
 		ret = iomap_dio_rw(iocb, to, &zonefs_iomap_ops,
-				   &zonefs_read_dio_ops, 0, 0);
+				   &zonefs_read_dio_ops, 0, NULL, 0);
 	} else {
 		ret = generic_file_read_iter(iocb, to);
 		if (ret == -EIO)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 87b5af1d9fbe..02e7f60638b8 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1708,6 +1708,11 @@ static inline bool __sb_start_write_trylock(struct super_block *sb, int level)
 #define __sb_writers_release(sb, lev)	\
 	percpu_rwsem_release(&(sb)->s_writers.rw_sem[(lev)-1], 1, _THIS_IP_)
 
+static inline bool sb_write_started(const struct super_block *sb)
+{
+	return lockdep_is_held_type(sb->s_writers.rw_sem + SB_FREEZE_WRITE - 1, 1);
+}
+
 /**
  * sb_end_write - drop write access to a superblock
  * @sb: the super we wrote to
diff --git a/include/linux/iomap.h b/include/linux/iomap.h
index b76f0dd149fb..5b6f64f4d771 100644
--- a/include/linux/iomap.h
+++ b/include/linux/iomap.h
@@ -188,6 +188,7 @@ struct iomap_iter {
 	unsigned flags;
 	struct iomap iomap;
 	struct iomap srcmap;
+	void *private;
 };
 
 int iomap_iter(struct iomap_iter *iter, const struct iomap_ops *ops);
@@ -320,6 +321,16 @@ struct iomap_dio_ops {
 		      unsigned flags);
 	void (*submit_io)(const struct iomap_iter *iter, struct bio *bio,
 		          loff_t file_offset);
+
+	/*
+	 * Filesystems wishing to attach private information to a direct io bio
+	 * must provide a ->submit_io method that attaches the additional
+	 * information to the bio and changes the ->bi_end_io callback to a
+	 * custom function.  This function should, at a minimum, perform any
+	 * relevant post-processing of the bio and end with a call to
+	 * iomap_dio_bio_end_io.
+	 */
+	struct bio_set *bio_set;
 };
 
 /*
@@ -344,11 +355,12 @@ struct iomap_dio_ops {
 
 ssize_t iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
-		unsigned int dio_flags, size_t done_before);
+		unsigned int dio_flags, void *private, size_t done_before);
 struct iomap_dio *__iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 		const struct iomap_ops *ops, const struct iomap_dio_ops *dops,
-		unsigned int dio_flags, size_t done_before);
+		unsigned int dio_flags, void *private, size_t done_before);
 ssize_t iomap_dio_complete(struct iomap_dio *dio);
+void iomap_dio_bio_end_io(struct bio *bio);
 
 #ifdef CONFIG_SWAP
 struct file;
diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h
index f068ff30d654..290f07eb050a 100644
--- a/include/trace/events/btrfs.h
+++ b/include/trace/events/btrfs.h
@@ -24,7 +24,7 @@ struct btrfs_free_cluster;
 struct map_lookup;
 struct extent_buffer;
 struct btrfs_work;
-struct __btrfs_workqueue;
+struct btrfs_workqueue;
 struct btrfs_qgroup_extent_record;
 struct btrfs_qgroup;
 struct extent_io_tree;
@@ -1457,42 +1457,36 @@ DEFINE_EVENT(btrfs__work, btrfs_ordered_sched,
 	TP_ARGS(work)
 );
 
-DECLARE_EVENT_CLASS(btrfs__workqueue,
+DECLARE_EVENT_CLASS(btrfs_workqueue,
 
-	TP_PROTO(const struct __btrfs_workqueue *wq,
-		 const char *name, int high),
+	TP_PROTO(const struct btrfs_workqueue *wq, const char *name),
 
-	TP_ARGS(wq, name, high),
+	TP_ARGS(wq, name),
 
 	TP_STRUCT__entry_btrfs(
 		__field(	const void *,	wq			)
 		__string(	name,	name			)
-		__field(	int ,	high			)
 	),
 
 	TP_fast_assign_btrfs(btrfs_workqueue_owner(wq),
 		__entry->wq		= wq;
 		__assign_str(name, name);
-		__entry->high		= high;
 	),
 
-	TP_printk_btrfs("name=%s%s wq=%p", __get_str(name),
-		  __print_flags(__entry->high, "",
-				{(WQ_HIGHPRI),	"-high"}),
+	TP_printk_btrfs("name=%s wq=%p", __get_str(name),
 		  __entry->wq)
 );
 
-DEFINE_EVENT(btrfs__workqueue, btrfs_workqueue_alloc,
+DEFINE_EVENT(btrfs_workqueue, btrfs_workqueue_alloc,
 
-	TP_PROTO(const struct __btrfs_workqueue *wq,
-		 const char *name, int high),
+	TP_PROTO(const struct btrfs_workqueue *wq, const char *name),
 
-	TP_ARGS(wq, name, high)
+	TP_ARGS(wq, name)
 );
 
-DECLARE_EVENT_CLASS(btrfs__workqueue_done,
+DECLARE_EVENT_CLASS(btrfs_workqueue_done,
 
-	TP_PROTO(const struct __btrfs_workqueue *wq),
+	TP_PROTO(const struct btrfs_workqueue *wq),
 
 	TP_ARGS(wq),
 
@@ -1507,9 +1501,9 @@ DECLARE_EVENT_CLASS(btrfs__workqueue_done,
 	TP_printk_btrfs("wq=%p", __entry->wq)
 );
 
-DEFINE_EVENT(btrfs__workqueue_done, btrfs_workqueue_destroy,
+DEFINE_EVENT(btrfs_workqueue_done, btrfs_workqueue_destroy,
 
-	TP_PROTO(const struct __btrfs_workqueue *wq),
+	TP_PROTO(const struct btrfs_workqueue *wq),
 
 	TP_ARGS(wq)
 );
diff --git a/include/uapi/linux/btrfs_tree.h b/include/uapi/linux/btrfs_tree.h
index b069752a8ecf..d4117152d907 100644
--- a/include/uapi/linux/btrfs_tree.h
+++ b/include/uapi/linux/btrfs_tree.h
@@ -880,19 +880,6 @@ struct btrfs_dev_replace_item {
 #define BTRFS_BLOCK_GROUP_RESERVED	(BTRFS_AVAIL_ALLOC_BIT_SINGLE | \
 					 BTRFS_SPACE_INFO_GLOBAL_RSV)
 
-enum btrfs_raid_types {
-	BTRFS_RAID_RAID10,
-	BTRFS_RAID_RAID1,
-	BTRFS_RAID_DUP,
-	BTRFS_RAID_RAID0,
-	BTRFS_RAID_SINGLE,
-	BTRFS_RAID_RAID5,
-	BTRFS_RAID_RAID6,
-	BTRFS_RAID_RAID1C3,
-	BTRFS_RAID_RAID1C4,
-	BTRFS_NR_RAID_TYPES
-};
-
 #define BTRFS_BLOCK_GROUP_TYPE_MASK	(BTRFS_BLOCK_GROUP_DATA |    \
 					 BTRFS_BLOCK_GROUP_SYSTEM |  \
 					 BTRFS_BLOCK_GROUP_METADATA)