Merge tag 'for-6.1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "There's a bunch of performance improvements, most notably the FIEMAP
  speedup, the new block group tree to speed up mount on large
  filesystems, more io_uring integration, some sysfs exports and the
  usual fixes and core updates.

  Summary:

  Performance:

   - outstanding FIEMAP speed improvement
      - algorithmic change how extents are enumerated leads to orders of
        magnitude speed boost (uncached and cached)
      - extent sharing check speedup (2.2x uncached, 3x cached)
      - add more cancellation points, allowing to interrupt seeking in
        files with large number of extents
      - more efficient hole and data seeking (4x uncached, 1.3x cached)
      - sample results:
	    256M, 32K extents:   4s ->  29ms  (~150x)
	    512M, 64K extents:  30s ->  59ms  (~550x)
	    1G,  128K extents: 225s -> 120ms (~1800x)

   - improved inode logging, especially for directories (on dbench
     workload throughput +25%, max latency -21%)

   - improved buffered IO, remove redundant extent state tracking,
     lowering memory consumption and avoiding rb tree traversal

   - add sysfs tunable to let qgroup temporarily skip exact accounting
     when deleting snapshot, leading to a speedup but requiring a rescan
     after that, will be used by snapper

   - support io_uring and buffered writes, until now it was just for
     direct IO, with the no-wait semantics implemented in the buffered
     write path it now works and leads to speed improvement in IOPS
     (2x), throughput (2.2x), latency (depends, 2x to 150x)

   - small performance improvements when dropping and searching for
     extent maps as well as when flushing delalloc in COW mode
     (throughput +5MB/s)

  User visible changes:

   - new incompatible feature block-group-tree adding a dedicated tree
     for tracking block groups, this allows a much faster load during
     mount and avoids seeking unlike when it's scattered in the extent
     tree items
      - this reduces mount time for many-terabyte sized filesystems
      - conversion tool will be provided so existing filesystem can also
        be updated in place
      - to reduce test matrix and feature combinations requires no-holes
        and free-space-tree (mkfs defaults since 5.15)

   - improved reporting of super block corruption detected by scrub

   - scrub also tries to repair super block and does not wait until next
     commit

   - discard stats and tunables are exported in sysfs
     (/sys/fs/btrfs/FSID/discard)

   - qgroup status is exported in sysfs
     (/sys/sys/fs/btrfs/FSID/qgroups/)

   - verify that super block was not modified when thawing filesystem

  Fixes:

   - FIEMAP fixes
      - fix extent sharing status, does not depend on the cached status
        where merged
      - flush delalloc so compressed extents are reported correctly

   - fix alignment of VMA for memory mapped files on THP

   - send: fix failures when processing inodes with no links (orphan
     files and directories)

   - fix race between quota enable and quota rescan ioctl

   - handle more corner cases for read-only compat feature verification

   - fix missed extent on fsync after dropping extent maps

  Core:

   - lockdep annotations to validate various transactions states and
     state transitions

   - preliminary support for fs-verity in send

   - more effective memory use in scrub for subpage where sector is
     smaller than page

   - block group caching progress logic has been removed, load is now
     synchronous

   - simplify end IO callbacks and bio handling, use chained bios
     instead of own tracking

   - add no-wait semantics to several functions (tree search, nocow,
     flushing, buffered write

   - cleanups and refactoring

  MM changes:

   - export balance_dirty_pages_ratelimited_flags"

* tag 'for-6.1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (177 commits)
  btrfs: set generation before calling btrfs_clean_tree_block in btrfs_init_new_buffer
  btrfs: drop extent map range more efficiently
  btrfs: avoid pointless extent map tree search when flushing delalloc
  btrfs: remove unnecessary next extent map search
  btrfs: remove unnecessary NULL pointer checks when searching extent maps
  btrfs: assert tree is locked when clearing extent map from logging
  btrfs: remove unnecessary extent map initializations
  btrfs: remove the refcount warning/check at free_extent_map()
  btrfs: add helper to replace extent map range with a new extent map
  btrfs: move open coded extent map tree deletion out of inode eviction
  btrfs: use cond_resched_rwlock_write() during inode eviction
  btrfs: use extent_map_end() at btrfs_drop_extent_map_range()
  btrfs: move btrfs_drop_extent_cache() to extent_map.c
  btrfs: fix missed extent on fsync after dropping extent maps
  btrfs: remove stale prototype of btrfs_write_inode
  btrfs: enable nowait async buffered writes
  btrfs: assert nowait mode is not used for some btree search functions
  btrfs: make btrfs_buffered_write nowait compatible
  btrfs: plumb NOWAIT through the write path
  btrfs: make lock_and_cleanup_extent_if_need nowait compatible
  ...

1 files changed, 74 insertions, 22 deletions

diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index db723c0026bd..9334c3157c22 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -275,7 +275,7 @@ static int __add_relation_rb(struct btrfs_qgroup *member, struct btrfs_qgroup *p
 }
 
 /*
- * Add relation specified by two qgoup ids.
+ * Add relation specified by two qgroup ids.
  *
  * Must be called with qgroup_lock held.
  *
@@ -333,6 +333,13 @@ int btrfs_verify_qgroup_counts(struct btrfs_fs_info *fs_info, u64 qgroupid,
 }
 #endif
 
+static void qgroup_mark_inconsistent(struct btrfs_fs_info *fs_info)
+{
+	fs_info->qgroup_flags |= (BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT |
+				  BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN |
+				  BTRFS_QGROUP_RUNTIME_FLAG_NO_ACCOUNTING);
+}
+
 /*
  * The full config is read in one go, only called from open_ctree()
  * It doesn't use any locking, as at this point we're still single-threaded
@@ -401,7 +408,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
 			}
 			if (btrfs_qgroup_status_generation(l, ptr) !=
 			    fs_info->generation) {
-				flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+				qgroup_mark_inconsistent(fs_info);
 				btrfs_err(fs_info,
 					"qgroup generation mismatch, marked as inconsistent");
 			}
@@ -419,7 +426,7 @@ int btrfs_read_qgroup_config(struct btrfs_fs_info *fs_info)
 		if ((qgroup && found_key.type == BTRFS_QGROUP_INFO_KEY) ||
 		    (!qgroup && found_key.type == BTRFS_QGROUP_LIMIT_KEY)) {
 			btrfs_err(fs_info, "inconsistent qgroup config");
-			flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+			qgroup_mark_inconsistent(fs_info);
 		}
 		if (!qgroup) {
 			qgroup = add_qgroup_rb(fs_info, found_key.offset);
@@ -878,7 +885,8 @@ static int update_qgroup_status_item(struct btrfs_trans_handle *trans)
 	l = path->nodes[0];
 	slot = path->slots[0];
 	ptr = btrfs_item_ptr(l, slot, struct btrfs_qgroup_status_item);
-	btrfs_set_qgroup_status_flags(l, ptr, fs_info->qgroup_flags);
+	btrfs_set_qgroup_status_flags(l, ptr, fs_info->qgroup_flags &
+				      BTRFS_QGROUP_STATUS_FLAGS_MASK);
 	btrfs_set_qgroup_status_generation(l, ptr, trans->transid);
 	btrfs_set_qgroup_status_rescan(l, ptr,
 				fs_info->qgroup_rescan_progress.objectid);
@@ -1052,7 +1060,8 @@ int btrfs_quota_enable(struct btrfs_fs_info *fs_info)
 	btrfs_set_qgroup_status_version(leaf, ptr, BTRFS_QGROUP_STATUS_VERSION);
 	fs_info->qgroup_flags = BTRFS_QGROUP_STATUS_FLAG_ON |
 				BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
-	btrfs_set_qgroup_status_flags(leaf, ptr, fs_info->qgroup_flags);
+	btrfs_set_qgroup_status_flags(leaf, ptr, fs_info->qgroup_flags &
+				      BTRFS_QGROUP_STATUS_FLAGS_MASK);
 	btrfs_set_qgroup_status_rescan(leaf, ptr, 0);
 
 	btrfs_mark_buffer_dirty(leaf);
@@ -1174,6 +1183,21 @@ out_add_root:
 		fs_info->qgroup_rescan_running = true;
 	        btrfs_queue_work(fs_info->qgroup_rescan_workers,
 	                         &fs_info->qgroup_rescan_work);
+	} else {
+		/*
+		 * We have set both BTRFS_FS_QUOTA_ENABLED and
+		 * BTRFS_QGROUP_STATUS_FLAG_ON, so we can only fail with
+		 * -EINPROGRESS. That can happen because someone started the
+		 * rescan worker by calling quota rescan ioctl before we
+		 * attempted to initialize the rescan worker. Failure due to
+		 * quotas disabled in the meanwhile is not possible, because
+		 * we are holding a write lock on fs_info->subvol_sem, which
+		 * is also acquired when disabling quotas.
+		 * Ignore such error, and any other error would need to undo
+		 * everything we did in the transaction we just committed.
+		 */
+		ASSERT(ret == -EINPROGRESS);
+		ret = 0;
 	}
 
 out_free_path:
@@ -1255,6 +1279,7 @@ int btrfs_quota_disable(struct btrfs_fs_info *fs_info)
 	quota_root = fs_info->quota_root;
 	fs_info->quota_root = NULL;
 	fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_ON;
+	fs_info->qgroup_drop_subtree_thres = BTRFS_MAX_LEVEL;
 	spin_unlock(&fs_info->qgroup_lock);
 
 	btrfs_free_qgroup_config(fs_info);
@@ -1717,7 +1742,7 @@ int btrfs_limit_qgroup(struct btrfs_trans_handle *trans, u64 qgroupid,
 
 	ret = update_qgroup_limit_item(trans, qgroup);
 	if (ret) {
-		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(fs_info);
 		btrfs_info(fs_info, "unable to update quota limit for %llu",
 		       qgroupid);
 	}
@@ -1790,10 +1815,13 @@ int btrfs_qgroup_trace_extent_post(struct btrfs_trans_handle *trans,
 	 */
 	ASSERT(trans != NULL);
 
+	if (trans->fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_NO_ACCOUNTING)
+		return 0;
+
 	ret = btrfs_find_all_roots(NULL, trans->fs_info, bytenr, 0, &old_root,
 				   true);
 	if (ret < 0) {
-		trans->fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(trans->fs_info);
 		btrfs_warn(trans->fs_info,
 "error accounting new delayed refs extent (err code: %d), quota inconsistent",
 			ret);
@@ -2269,7 +2297,7 @@ static int qgroup_trace_subtree_swap(struct btrfs_trans_handle *trans,
 out:
 	btrfs_free_path(dst_path);
 	if (ret < 0)
-		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(fs_info);
 	return ret;
 }
 
@@ -2280,6 +2308,7 @@ int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	int ret = 0;
 	int level;
+	u8 drop_subptree_thres;
 	struct extent_buffer *eb = root_eb;
 	struct btrfs_path *path = NULL;
 
@@ -2289,6 +2318,23 @@ int btrfs_qgroup_trace_subtree(struct btrfs_trans_handle *trans,
 	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
 		return 0;
 
+	spin_lock(&fs_info->qgroup_lock);
+	drop_subptree_thres = fs_info->qgroup_drop_subtree_thres;
+	spin_unlock(&fs_info->qgroup_lock);
+
+	/*
+	 * This function only gets called for snapshot drop, if we hit a high
+	 * node here, it means we are going to change ownership for quite a lot
+	 * of extents, which will greatly slow down btrfs_commit_transaction().
+	 *
+	 * So here if we find a high tree here, we just skip the accounting and
+	 * mark qgroup inconsistent.
+	 */
+	if (root_level >= drop_subptree_thres) {
+		qgroup_mark_inconsistent(fs_info);
+		return 0;
+	}
+
 	if (!extent_buffer_uptodate(root_eb)) {
 		ret = btrfs_read_extent_buffer(root_eb, root_gen, root_level, NULL);
 		if (ret)
@@ -2604,7 +2650,8 @@ int btrfs_qgroup_account_extent(struct btrfs_trans_handle *trans, u64 bytenr,
 	 * If quotas get disabled meanwhile, the resources need to be freed and
 	 * we can't just exit here.
 	 */
-	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
+	if (!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
+	    fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_NO_ACCOUNTING)
 		goto out_free;
 
 	if (new_roots) {
@@ -2700,7 +2747,8 @@ int btrfs_qgroup_account_extents(struct btrfs_trans_handle *trans)
 		num_dirty_extents++;
 		trace_btrfs_qgroup_account_extents(fs_info, record);
 
-		if (!ret) {
+		if (!ret && !(fs_info->qgroup_flags &
+			      BTRFS_QGROUP_RUNTIME_FLAG_NO_ACCOUNTING)) {
 			/*
 			 * Old roots should be searched when inserting qgroup
 			 * extent record
@@ -2773,12 +2821,10 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans)
 		spin_unlock(&fs_info->qgroup_lock);
 		ret = update_qgroup_info_item(trans, qgroup);
 		if (ret)
-			fs_info->qgroup_flags |=
-					BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+			qgroup_mark_inconsistent(fs_info);
 		ret = update_qgroup_limit_item(trans, qgroup);
 		if (ret)
-			fs_info->qgroup_flags |=
-					BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+			qgroup_mark_inconsistent(fs_info);
 		spin_lock(&fs_info->qgroup_lock);
 	}
 	if (test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags))
@@ -2789,7 +2835,7 @@ int btrfs_run_qgroups(struct btrfs_trans_handle *trans)
 
 	ret = update_qgroup_status_item(trans);
 	if (ret)
-		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(fs_info);
 
 	return ret;
 }
@@ -2907,7 +2953,7 @@ int btrfs_qgroup_inherit(struct btrfs_trans_handle *trans, u64 srcid,
 
 		ret = update_qgroup_limit_item(trans, dstgroup);
 		if (ret) {
-			fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+			qgroup_mark_inconsistent(fs_info);
 			btrfs_info(fs_info,
 				   "unable to update quota limit for %llu",
 				   dstgroup->qgroupid);
@@ -3013,7 +3059,7 @@ out:
 	if (!committing)
 		mutex_unlock(&fs_info->qgroup_ioctl_lock);
 	if (need_rescan)
-		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(fs_info);
 	return ret;
 }
 
@@ -3286,7 +3332,8 @@ static bool rescan_should_stop(struct btrfs_fs_info *fs_info)
 {
 	return btrfs_fs_closing(fs_info) ||
 		test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state) ||
-		!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags);
+		!test_bit(BTRFS_FS_QUOTA_ENABLED, &fs_info->flags) ||
+			  fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN;
 }
 
 static void btrfs_qgroup_rescan_worker(struct btrfs_work *work)
@@ -3351,7 +3398,8 @@ out:
 	}
 
 	mutex_lock(&fs_info->qgroup_rescan_lock);
-	if (!stopped)
+	if (!stopped ||
+	    fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN)
 		fs_info->qgroup_flags &= ~BTRFS_QGROUP_STATUS_FLAG_RESCAN;
 	if (trans) {
 		ret = update_qgroup_status_item(trans);
@@ -3362,6 +3410,7 @@ out:
 		}
 	}
 	fs_info->qgroup_rescan_running = false;
+	fs_info->qgroup_flags &= ~BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN;
 	complete_all(&fs_info->qgroup_rescan_completion);
 	mutex_unlock(&fs_info->qgroup_rescan_lock);
 
@@ -3372,6 +3421,8 @@ out:
 
 	if (stopped) {
 		btrfs_info(fs_info, "qgroup scan paused");
+	} else if (fs_info->qgroup_flags & BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN) {
+		btrfs_info(fs_info, "qgroup scan cancelled");
 	} else if (err >= 0) {
 		btrfs_info(fs_info, "qgroup scan completed%s",
 			err > 0 ? " (inconsistency flag cleared)" : "");
@@ -3434,6 +3485,8 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid,
 
 	memset(&fs_info->qgroup_rescan_progress, 0,
 		sizeof(fs_info->qgroup_rescan_progress));
+	fs_info->qgroup_flags &= ~(BTRFS_QGROUP_RUNTIME_FLAG_CANCEL_RESCAN |
+				   BTRFS_QGROUP_RUNTIME_FLAG_NO_ACCOUNTING);
 	fs_info->qgroup_rescan_progress.objectid = progress_objectid;
 	init_completion(&fs_info->qgroup_rescan_completion);
 	mutex_unlock(&fs_info->qgroup_rescan_lock);
@@ -4231,8 +4284,7 @@ out_unlock:
 	spin_unlock(&blocks->lock);
 out:
 	if (ret < 0)
-		fs_info->qgroup_flags |=
-			BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(fs_info);
 	return ret;
 }
 
@@ -4319,7 +4371,7 @@ out:
 		btrfs_err_rl(fs_info,
 			     "failed to account subtree at bytenr %llu: %d",
 			     subvol_eb->start, ret);
-		fs_info->qgroup_flags |= BTRFS_QGROUP_STATUS_FLAG_INCONSISTENT;
+		qgroup_mark_inconsistent(fs_info);
 	}
 	return ret;
 }