Merge tag 'for-5.17-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs updates from David Sterba:
 "This end of the year branch is intentionally not that exciting. Most
  of the changes are under the hood, but there are some minor user
  visible improvements and several performance improvements too.

  Features:

   - make send work with concurrent block group relocation.

     We're not allowed to prevent send failing or silently producing
     some bad stream but with more fine grained locking and checks it's
     possible. The send vs deduplication exclusion could reuse the same
     logic in the future.

   - new exclusive operation 'balance paused' to allow adding a device
     to filesystem with paused balance

   - new sysfs file for fsid stored in the per-device directory to help
     distinguish devices when seeding is enabled, the fsid may differ
     from the one reported by the filesystem

  Performance improvements:

   - less metadata needed for directory logging, directory deletion is
     20-40% faster

   - in zoned mode, cache zone information during mount to speed up
     repeated queries (about 50% speedup)

   - free space tree entries get indexed and searched by size (latency
     -30%, search run time -30%)

   - less contention in tree node locking when inserting a key and no
     splits are needed (files/sec in fsmark improves by 1-20%)

  Fixes:

   - fix ENOSPC failure when attempting direct IO write into NOCOW range

   - fix deadlock between quota enable and other quota operations

   - global reserve minimum calculations fixed to account for free space
     tree

   - in zoned mode, fix condition for chunk allocation that may not find
     the right zone for reuse and could lead to early ENOSPC

  Core:

   - global reserve stealing got simplified and cleaned up in evict

   - remove async transaction commit based on manual transaction refs,
     reuse existing kthread and mechanisms to let it commit transaction
     before timeout

   - preparatory work for extent tree v2, add wrappers for global tree
     roots, truncation path cleanups

   - remove readahead framework, it's a bit overengineered and used only
     for scrub, and yet it does not cover all its needs, there is
     another readahead built in the b-tree search that is now used,
     performance drop on HDD is about 5% which is acceptable and scrub
     is often throttled anyway, on SSDs there's no reported drop but
     slight improvement

   - self tests report extent tree state when error occurs

   - replace assert with debugging information when an uncommitted
     transaction is found at unmount time

  Other:

   - error handling improvements

   - other cleanups and refactoring"

* tag 'for-5.17-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (115 commits)
  btrfs: output more debug messages for uncommitted transaction
  btrfs: respect the max size in the header when activating swap file
  btrfs: fix argument list that the kdoc format and script verified
  btrfs: remove unnecessary parameter type from compression_decompress_bio
  btrfs: selftests: dump extent io tree if extent-io-tree test failed
  btrfs: scrub: cleanup the argument list of scrub_stripe()
  btrfs: scrub: cleanup the argument list of scrub_chunk()
  btrfs: remove reada infrastructure
  btrfs: scrub: use btrfs_path::reada for extent tree readahead
  btrfs: scrub: remove the unnecessary path parameter for scrub_raid56_parity()
  btrfs: refactor unlock_up
  btrfs: skip transaction commit after failure to create subvolume
  btrfs: zoned: fix chunk allocation condition for zoned allocator
  btrfs: add extent allocator hook to decide to allocate chunk or not
  btrfs: zoned: unset dedicated block group on allocation failure
  btrfs: zoned: drop redundant check for REQ_OP_ZONE_APPEND and btrfs_is_zoned
  btrfs: zoned: sink zone check into btrfs_repair_one_zone
  btrfs: zoned: simplify btrfs_check_meta_write_pointer
  btrfs: zoned: encapsulate inode locking for zoned relocation
  btrfs: sysfs: add devinfo/fsid to retrieve actual fsid from the device
  ...

1 files changed, 25 insertions, 5 deletions

diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h
index e53ab7b96437..cbf016a7bb5d 100644
--- a/fs/btrfs/zoned.h
+++ b/fs/btrfs/zoned.h
@@ -8,6 +8,7 @@
 #include "volumes.h"
 #include "disk-io.h"
 #include "block-group.h"
+#include "btrfs_inode.h"
 
 /*
  * Block groups with more than this value (percents) of unusable space will be
@@ -28,6 +29,7 @@ struct btrfs_zoned_device_info {
 	unsigned long *seq_zones;
 	unsigned long *empty_zones;
 	unsigned long *active_zones;
+	struct blk_zone *zone_cache;
 	struct blk_zone sb_zones[2 * BTRFS_SUPER_MIRROR_MAX];
 };
 
@@ -35,7 +37,7 @@ struct btrfs_zoned_device_info {
 int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos,
 		       struct blk_zone *zone);
 int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info);
-int btrfs_get_dev_zone_info(struct btrfs_device *device);
+int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache);
 void btrfs_destroy_dev_zone_info(struct btrfs_device *device);
 int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info);
 int btrfs_check_mountopts_zoned(struct btrfs_fs_info *info);
@@ -71,11 +73,11 @@ struct btrfs_device *btrfs_zoned_get_device(struct btrfs_fs_info *fs_info,
 					    u64 logical, u64 length);
 bool btrfs_zone_activate(struct btrfs_block_group *block_group);
 int btrfs_zone_finish(struct btrfs_block_group *block_group);
-bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices,
-			     int raid_index);
+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_clear_data_reloc_bg(struct btrfs_block_group *bg);
+void btrfs_free_zone_cache(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)
@@ -88,7 +90,8 @@ static inline int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_i
 	return 0;
 }
 
-static inline int btrfs_get_dev_zone_info(struct btrfs_device *device)
+static inline int btrfs_get_dev_zone_info(struct btrfs_device *device,
+					  bool populate_cache)
 {
 	return 0;
 }
@@ -222,7 +225,7 @@ static inline int btrfs_zone_finish(struct btrfs_block_group *block_group)
 }
 
 static inline bool btrfs_can_activate_zone(struct btrfs_fs_devices *fs_devices,
-					   int raid_index)
+					   u64 flags)
 {
 	return true;
 }
@@ -232,6 +235,7 @@ static inline void btrfs_zone_finish_endio(struct btrfs_fs_info *fs_info,
 
 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) { }
 #endif
 
 static inline bool btrfs_dev_is_sequential(struct btrfs_device *device, u64 pos)
@@ -350,4 +354,20 @@ static inline void btrfs_clear_treelog_bg(struct btrfs_block_group *bg)
 	spin_unlock(&fs_info->treelog_bg_lock);
 }
 
+static inline void btrfs_zoned_data_reloc_lock(struct btrfs_inode *inode)
+{
+	struct btrfs_root *root = inode->root;
+
+	if (btrfs_is_data_reloc_root(root) && btrfs_is_zoned(root->fs_info))
+		btrfs_inode_lock(&inode->vfs_inode, 0);
+}
+
+static inline void btrfs_zoned_data_reloc_unlock(struct btrfs_inode *inode)
+{
+	struct btrfs_root *root = inode->root;
+
+	if (btrfs_is_data_reloc_root(root) && btrfs_is_zoned(root->fs_info))
+		btrfs_inode_unlock(&inode->vfs_inode, 0);
+}
+
 #endif