Merge tag 'for-5.1-part1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "This contains usual mix of new features, core changes and fixes; full
  list below. I'm planning second pull request, with a few more fixes
  that arrived recently but too close to merge window, will send it next
  week.

  New features:

   - support zstd compression levels

   - new ioctl to unregister a device from the module (ie. reverse of
     device scan)

   - scrub prints a message to log when it's about to start or finish

  Core changes:

   - qgroups can now skip part of a tree that does not get updated
     during relocation, because this does not affect the quota
     accounting, estimated speedup in run time is about 20%

   - the compression workspace management had to be enhanced due to zstd
     requirements

   - various enospc fixes, when there's high fragmentation the
     over-reservation can cause ENOSPC that might not happen after a
     flush, in such cases try to wait if the situation improves

  Fixes:

   - various ioctls could overwrite previous return value if
     copy_to_user fails, fix this so the original error is reported

   - more reclaim vs GFP_KERNEL fixes

   - other cleanups and refactoring

   - fix a (valid) lockdep warning in a test when device replace is
     destroying worker threads

   - make qgroup async transaction commit more aggressive, this avoids
     some 'quota limit reached' errors if there are not enough data to
     trigger transaction in order to flush

   - fix deadlock between snapshot deletion and quotas when backref
     walking is called from context that already holds the same locks

   - fsync fixes:
      - fix fsync after succession of renames of different files
      - fix fsync after succession of renames and unlink/rmdir"

* tag 'for-5.1-part1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (92 commits)
  btrfs: Remove unnecessary casts in btrfs_read_root_item
  Btrfs: remove assertion when searching for a key in a node/leaf
  Btrfs: add missing error handling after doing leaf/node binary search
  btrfs: drop the lock on error in btrfs_dev_replace_cancel
  btrfs: ensure that a DUP or RAID1 block group has exactly two stripes
  btrfs: init csum_list before possible free
  Btrfs: remove no longer needed range length checks for deduplication
  Btrfs: fix fsync after succession of renames and unlink/rmdir
  Btrfs: fix fsync after succession of renames of different files
  btrfs: honor path->skip_locking in backref code
  btrfs: qgroup: Make qgroup async transaction commit more aggressive
  btrfs: qgroup: Move reserved data accounting from btrfs_delayed_ref_head to btrfs_qgroup_extent_record
  btrfs: scrub: remove unused nocow worker pointer
  btrfs: scrub: add assertions for worker pointers
  btrfs: scrub: convert scrub_workers_refcnt to refcount_t
  btrfs: scrub: add scrub_lock lockdep check in scrub_workers_get
  btrfs: scrub: fix circular locking dependency warning
  btrfs: fix comment its device list mutex not volume lock
  btrfs: extent_io: Kill the forward declaration of flush_write_bio
  btrfs: Fix grossly misleading argument names in extent io search
  ...

1 files changed, 18 insertions, 21 deletions

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 6a2a2a951705..5216e7b3f9ad 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -17,6 +17,7 @@
 #include <linux/semaphore.h>
 #include <linux/error-injection.h>
 #include <linux/crc32c.h>
+#include <linux/sched/mm.h>
 #include <asm/unaligned.h>
 #include "ctree.h"
 #include "disk-io.h"
@@ -341,7 +342,7 @@ static int verify_parent_transid(struct extent_io_tree *io_tree,
 
 	if (need_lock) {
 		btrfs_tree_read_lock(eb);
-		btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+		btrfs_set_lock_blocking_read(eb);
 	}
 
 	lock_extent_bits(io_tree, eb->start, eb->start + eb->len - 1,
@@ -1120,7 +1121,7 @@ void clean_tree_block(struct btrfs_fs_info *fs_info,
 						 -buf->len,
 						 fs_info->dirty_metadata_batch);
 			/* ugh, clear_extent_buffer_dirty needs to lock the page */
-			btrfs_set_lock_blocking(buf);
+			btrfs_set_lock_blocking_write(buf);
 			clear_extent_buffer_dirty(buf);
 		}
 	}
@@ -1175,6 +1176,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	INIT_LIST_HEAD(&root->delalloc_root);
 	INIT_LIST_HEAD(&root->ordered_extents);
 	INIT_LIST_HEAD(&root->ordered_root);
+	INIT_LIST_HEAD(&root->reloc_dirty_list);
 	INIT_LIST_HEAD(&root->logged_list[0]);
 	INIT_LIST_HEAD(&root->logged_list[1]);
 	spin_lock_init(&root->inode_lock);
@@ -1218,6 +1220,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
 	root->anon_dev = 0;
 
 	spin_lock_init(&root->root_item_lock);
+	btrfs_qgroup_init_swapped_blocks(&root->swapped_blocks);
 }
 
 static struct btrfs_root *btrfs_alloc_root(struct btrfs_fs_info *fs_info,
@@ -1258,10 +1261,17 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans,
 	struct btrfs_root *tree_root = fs_info->tree_root;
 	struct btrfs_root *root;
 	struct btrfs_key key;
+	unsigned int nofs_flag;
 	int ret = 0;
 	uuid_le uuid = NULL_UUID_LE;
 
+	/*
+	 * We're holding a transaction handle, so use a NOFS memory allocation
+	 * context to avoid deadlock if reclaim happens.
+	 */
+	nofs_flag = memalloc_nofs_save();
 	root = btrfs_alloc_root(fs_info, GFP_KERNEL);
+	memalloc_nofs_restore(nofs_flag);
 	if (!root)
 		return ERR_PTR(-ENOMEM);
 
@@ -1707,9 +1717,7 @@ static int cleaner_kthread(void *arg)
 			goto sleep;
 		}
 
-		mutex_lock(&fs_info->cleaner_delayed_iput_mutex);
 		btrfs_run_delayed_iputs(fs_info);
-		mutex_unlock(&fs_info->cleaner_delayed_iput_mutex);
 
 		again = btrfs_clean_one_deleted_snapshot(root);
 		mutex_unlock(&fs_info->cleaner_mutex);
@@ -2101,7 +2109,7 @@ static void btrfs_init_scrub(struct btrfs_fs_info *fs_info)
 	atomic_set(&fs_info->scrubs_paused, 0);
 	atomic_set(&fs_info->scrub_cancel_req, 0);
 	init_waitqueue_head(&fs_info->scrub_pause_wait);
-	fs_info->scrub_workers_refcnt = 0;
+	refcount_set(&fs_info->scrub_workers_refcnt, 0);
 }
 
 static void btrfs_init_balance(struct btrfs_fs_info *fs_info)
@@ -2666,7 +2674,6 @@ int open_ctree(struct super_block *sb,
 	mutex_init(&fs_info->delete_unused_bgs_mutex);
 	mutex_init(&fs_info->reloc_mutex);
 	mutex_init(&fs_info->delalloc_root_mutex);
-	mutex_init(&fs_info->cleaner_delayed_iput_mutex);
 	seqlock_init(&fs_info->profiles_lock);
 
 	INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
@@ -2688,6 +2695,7 @@ int open_ctree(struct super_block *sb,
 	atomic_set(&fs_info->defrag_running, 0);
 	atomic_set(&fs_info->qgroup_op_seq, 0);
 	atomic_set(&fs_info->reada_works_cnt, 0);
+	atomic_set(&fs_info->nr_delayed_iputs, 0);
 	atomic64_set(&fs_info->tree_mod_seq, 0);
 	fs_info->sb = sb;
 	fs_info->max_inline = BTRFS_DEFAULT_MAX_INLINE;
@@ -2765,6 +2773,7 @@ int open_ctree(struct super_block *sb,
 	init_waitqueue_head(&fs_info->transaction_wait);
 	init_waitqueue_head(&fs_info->transaction_blocked_wait);
 	init_waitqueue_head(&fs_info->async_submit_wait);
+	init_waitqueue_head(&fs_info->delayed_iputs_wait);
 
 	INIT_LIST_HEAD(&fs_info->pinned_chunks);
 
@@ -4238,16 +4247,9 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 
 		head = rb_entry(node, struct btrfs_delayed_ref_head,
 				href_node);
-		if (!mutex_trylock(&head->mutex)) {
-			refcount_inc(&head->refs);
-			spin_unlock(&delayed_refs->lock);
-
-			mutex_lock(&head->mutex);
-			mutex_unlock(&head->mutex);
-			btrfs_put_delayed_ref_head(head);
-			spin_lock(&delayed_refs->lock);
+		if (btrfs_delayed_ref_lock(delayed_refs, head))
 			continue;
-		}
+
 		spin_lock(&head->lock);
 		while ((n = rb_first_cached(&head->ref_tree)) != NULL) {
 			ref = rb_entry(n, struct btrfs_delayed_ref_node,
@@ -4263,12 +4265,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 		if (head->must_insert_reserved)
 			pin_bytes = true;
 		btrfs_free_delayed_extent_op(head->extent_op);
-		delayed_refs->num_heads--;
-		if (head->processing == 0)
-			delayed_refs->num_heads_ready--;
-		atomic_dec(&delayed_refs->num_entries);
-		rb_erase_cached(&head->href_node, &delayed_refs->href_root);
-		RB_CLEAR_NODE(&head->href_node);
+		btrfs_delete_ref_head(delayed_refs, head);
 		spin_unlock(&head->lock);
 		spin_unlock(&delayed_refs->lock);
 		mutex_unlock(&head->mutex);