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, 105 insertions, 15 deletions

diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h
index 20c6bd5fa701..46ba7bd2961c 100644
--- a/fs/btrfs/qgroup.h
+++ b/fs/btrfs/qgroup.h
@@ -6,6 +6,8 @@
 #ifndef BTRFS_QGROUP_H
 #define BTRFS_QGROUP_H
 
+#include <linux/spinlock.h>
+#include <linux/rbtree.h>
 #include "ulist.h"
 #include "delayed-ref.h"
 
@@ -38,6 +40,66 @@
  */
 
 /*
+ * Special performance optimization for balance.
+ *
+ * For balance, we need to swap subtree of subvolume and reloc trees.
+ * In theory, we need to trace all subtree blocks of both subvolume and reloc
+ * trees, since their owner has changed during such swap.
+ *
+ * However since balance has ensured that both subtrees are containing the
+ * same contents and have the same tree structures, such swap won't cause
+ * qgroup number change.
+ *
+ * But there is a race window between subtree swap and transaction commit,
+ * during that window, if we increase/decrease tree level or merge/split tree
+ * blocks, we still need to trace the original subtrees.
+ *
+ * So for balance, we use a delayed subtree tracing, whose workflow is:
+ *
+ * 1) Record the subtree root block get swapped.
+ *
+ *    During subtree swap:
+ *    O = Old tree blocks
+ *    N = New tree blocks
+ *          reloc tree                     subvolume tree X
+ *             Root                               Root
+ *            /    \                             /    \
+ *          NA     OB                          OA      OB
+ *        /  |     |  \                      /  |      |  \
+ *      NC  ND     OE  OF                   OC  OD     OE  OF
+ *
+ *   In this case, NA and OA are going to be swapped, record (NA, OA) into
+ *   subvolume tree X.
+ *
+ * 2) After subtree swap.
+ *          reloc tree                     subvolume tree X
+ *             Root                               Root
+ *            /    \                             /    \
+ *          OA     OB                          NA      OB
+ *        /  |     |  \                      /  |      |  \
+ *      OC  OD     OE  OF                   NC  ND     OE  OF
+ *
+ * 3a) COW happens for OB
+ *     If we are going to COW tree block OB, we check OB's bytenr against
+ *     tree X's swapped_blocks structure.
+ *     If it doesn't fit any, nothing will happen.
+ *
+ * 3b) COW happens for NA
+ *     Check NA's bytenr against tree X's swapped_blocks, and get a hit.
+ *     Then we do subtree scan on both subtrees OA and NA.
+ *     Resulting 6 tree blocks to be scanned (OA, OC, OD, NA, NC, ND).
+ *
+ *     Then no matter what we do to subvolume tree X, qgroup numbers will
+ *     still be correct.
+ *     Then NA's record gets removed from X's swapped_blocks.
+ *
+ * 4)  Transaction commit
+ *     Any record in X's swapped_blocks gets removed, since there is no
+ *     modification to the swapped subtrees, no need to trigger heavy qgroup
+ *     subtree rescan for them.
+ */
+
+/*
  * Record a dirty extent, and info qgroup to update quota on it
  * TODO: Use kmem cache to alloc it.
  */
@@ -45,9 +107,38 @@ struct btrfs_qgroup_extent_record {
 	struct rb_node node;
 	u64 bytenr;
 	u64 num_bytes;
+
+	/*
+	 * For qgroup reserved data space freeing.
+	 *
+	 * @data_rsv_refroot and @data_rsv will be recorded after
+	 * BTRFS_ADD_DELAYED_EXTENT is called.
+	 * And will be used to free reserved qgroup space at
+	 * transaction commit time.
+	 */
+	u32 data_rsv;		/* reserved data space needs to be freed */
+	u64 data_rsv_refroot;	/* which root the reserved data belongs to */
 	struct ulist *old_roots;
 };
 
+struct btrfs_qgroup_swapped_block {
+	struct rb_node node;
+
+	int level;
+	bool trace_leaf;
+
+	/* bytenr/generation of the tree block in subvolume tree after swap */
+	u64 subvol_bytenr;
+	u64 subvol_generation;
+
+	/* bytenr/generation of the tree block in reloc tree after swap */
+	u64 reloc_bytenr;
+	u64 reloc_generation;
+
+	u64 last_snapshot;
+	struct btrfs_key first_key;
+};
+
 /*
  * Qgroup reservation types:
  *
@@ -236,12 +327,6 @@ int btrfs_qgroup_trace_leaf_items(struct btrfs_trans_handle *trans,
 int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 			       struct extent_buffer *root_eb,
 			       u64 root_gen, int root_level);
-
-int btrfs_qgroup_trace_subtree_swap(struct btrfs_trans_handle *trans,
-				struct btrfs_block_group_cache *bg_cache,
-				struct extent_buffer *src_parent, int src_slot,
-				struct extent_buffer *dst_parent, int dst_slot,
-				u64 last_snapshot);
 int btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, u64 bytenr,
 				u64 num_bytes, struct ulist *old_roots,
 				struct ulist *new_roots);
@@ -252,15 +337,6 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
 void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
 			       u64 ref_root, u64 num_bytes,
 			       enum btrfs_qgroup_rsv_type type);
-static inline void btrfs_qgroup_free_delayed_ref(struct btrfs_fs_info *fs_info,
-						 u64 ref_root, u64 num_bytes)
-{
-	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
-		return;
-	trace_btrfs_qgroup_free_delayed_ref(fs_info, ref_root, num_bytes);
-	btrfs_qgroup_free_refroot(fs_info, ref_root, num_bytes,
-				  BTRFS_QGROUP_RSV_DATA);
-}
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 int btrfs_verify_qgroup_counts(struct btrfs_fs_info *fs_info, u64 qgroupid,
@@ -325,4 +401,18 @@ void btrfs_qgroup_convert_reserved_meta(struct btrfs_root *root, int num_bytes);
 
 void btrfs_qgroup_check_reserved_leak(struct inode *inode);
 
+/* btrfs_qgroup_swapped_blocks related functions */
+void btrfs_qgroup_init_swapped_blocks(
+	struct btrfs_qgroup_swapped_blocks *swapped_blocks);
+
+void btrfs_qgroup_clean_swapped_blocks(struct btrfs_root *root);
+int btrfs_qgroup_add_swapped_blocks(struct btrfs_trans_handle *trans,
+		struct btrfs_root *subvol_root,
+		struct btrfs_block_group_cache *bg,
+		struct extent_buffer *subvol_parent, int subvol_slot,
+		struct extent_buffer *reloc_parent, int reloc_slot,
+		u64 last_snapshot);
+int btrfs_qgroup_trace_subtree_after_cow(struct btrfs_trans_handle *trans,
+		struct btrfs_root *root, struct extent_buffer *eb);
+
 #endif