[PATCH v4 19/27] btrfs: serialize meta IOs on HMZONED mode

[Naohiro Aota <naohiro.aota@wdc.com>]

As same as in data IO path, we must serialize write IOs for metadata. We
cannot add mutex around allocation and submit because metadata blocks are
allocated in an earlier stage to build up B-trees.

Thus, this commit add hmzoned_meta_io_lock and hold it during metadata IO
submission in btree_write_cache_pages() to serialize IOs. Furthermore, this
commit add per-block grorup metadata IO submission pointer
"meta_write_pointer" to ensure sequential writing.

Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
---
 fs/btrfs/ctree.h     |  3 +++
 fs/btrfs/disk-io.c   |  1 +
 fs/btrfs/extent_io.c | 17 ++++++++++++++++-
 fs/btrfs/hmzoned.c   | 45 ++++++++++++++++++++++++++++++++++++++++++++
 fs/btrfs/hmzoned.h   | 17 +++++++++++++++++
 5 files changed, 82 insertions(+), 1 deletion(-)

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index d4df9624cb04..e974174e12a2 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -619,6 +619,7 @@ struct btrfs_block_group_cache {
 	 */
 	u64 alloc_offset;
 	struct mutex zone_io_lock;
+	u64 meta_write_pointer;
 };
 
 /* delayed seq elem */
@@ -1105,6 +1106,8 @@ struct btrfs_fs_info {
 	spinlock_t ref_verify_lock;
 	struct rb_root block_tree;
 #endif
+
+	struct mutex hmzoned_meta_io_lock;
 };
 
 static inline struct btrfs_fs_info *btrfs_sb(struct super_block *sb)
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index d36cdb1b1421..a9632e455eb5 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -2701,6 +2701,7 @@ 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->hmzoned_meta_io_lock);
 	seqlock_init(&fs_info->profiles_lock);
 
 	INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 4e67b16c9f80..ff963b2214aa 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -3892,7 +3892,9 @@ int btree_write_cache_pages(struct address_space *mapping,
 				   struct writeback_control *wbc)
 {
 	struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree;
+	struct btrfs_fs_info *fs_info = tree->fs_info;
 	struct extent_buffer *eb, *prev_eb = NULL;
+	struct btrfs_block_group_cache *cache = NULL;
 	struct extent_page_data epd = {
 		.bio = NULL,
 		.tree = tree,
@@ -3922,6 +3924,7 @@ int btree_write_cache_pages(struct address_space *mapping,
 		tag = PAGECACHE_TAG_TOWRITE;
 	else
 		tag = PAGECACHE_TAG_DIRTY;
+	btrfs_hmzoned_meta_io_lock(fs_info);
 retry:
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		tag_pages_for_writeback(mapping, index, end);
@@ -3965,6 +3968,14 @@ int btree_write_cache_pages(struct address_space *mapping,
 			if (!ret)
 				continue;
 
+			if (!btrfs_check_meta_write_pointer(fs_info, eb,
+							    &cache)) {
+				ret = 0;
+				done = 1;
+				free_extent_buffer(eb);
+				break;
+			}
+
 			prev_eb = eb;
 			ret = lock_extent_buffer_for_io(eb, &epd);
 			if (!ret) {
@@ -3999,12 +4010,16 @@ int btree_write_cache_pages(struct address_space *mapping,
 		index = 0;
 		goto retry;
 	}
+	if (cache)
+		btrfs_put_block_group(cache);
 	ASSERT(ret <= 0);
 	if (ret < 0) {
 		end_write_bio(&epd, ret);
-		return ret;
+		goto out;
 	}
 	ret = flush_write_bio(&epd);
+out:
+	btrfs_hmzoned_meta_io_unlock(fs_info);
 	return ret;
 }
 
diff --git a/fs/btrfs/hmzoned.c b/fs/btrfs/hmzoned.c
index 81d8037ae7f6..bfc95a0443d0 100644
--- a/fs/btrfs/hmzoned.c
+++ b/fs/btrfs/hmzoned.c
@@ -545,6 +545,9 @@ int btrfs_load_block_group_zone_info(struct btrfs_block_group_cache *cache)
 
 out:
 	cache->alloc_type = alloc_type;
+	if (!ret)
+		cache->meta_write_pointer =
+			cache->alloc_offset + cache->key.objectid;
 	kfree(alloc_offsets);
 	free_extent_map(em);
 
@@ -648,3 +651,45 @@ void btrfs_free_redirty_list(struct btrfs_transaction *trans)
 	}
 	spin_unlock(&trans->releasing_ebs_lock);
 }
+
+bool btrfs_check_meta_write_pointer(struct btrfs_fs_info *fs_info,
+				    struct extent_buffer *eb,
+				    struct btrfs_block_group_cache **cache_ret)
+{
+	struct btrfs_block_group_cache *cache;
+
+	if (!btrfs_fs_incompat(fs_info, HMZONED))
+		return true;
+
+	cache = *cache_ret;
+
+	if (cache &&
+	    (eb->start < cache->key.objectid ||
+	     cache->key.objectid + cache->key.offset <= eb->start)) {
+		btrfs_put_block_group(cache);
+		cache = NULL;
+		*cache_ret = NULL;
+	}
+
+	if (!cache)
+		cache = btrfs_lookup_block_group(fs_info,
+						 eb->start);
+
+	if (cache) {
+		*cache_ret = cache;
+
+		if (cache->alloc_type != BTRFS_ALLOC_SEQ)
+			return true;
+
+		if (cache->meta_write_pointer != eb->start) {
+			btrfs_put_block_group(cache);
+			cache = NULL;
+			*cache_ret = NULL;
+			return false;
+		}
+
+		cache->meta_write_pointer = eb->start + eb->len;
+	}
+
+	return true;
+}
diff --git a/fs/btrfs/hmzoned.h b/fs/btrfs/hmzoned.h
index a8e7286708d4..c68c4b8056a4 100644
--- a/fs/btrfs/hmzoned.h
+++ b/fs/btrfs/hmzoned.h
@@ -40,6 +40,9 @@ void btrfs_redirty_list_add(struct btrfs_transaction *trans,
 			    struct extent_buffer *eb);
 void btrfs_free_redirty_list(struct btrfs_transaction *trans);
 void btrfs_hmzoned_data_io_unlock_at(struct inode *inode, u64 start, u64 len);
+bool btrfs_check_meta_write_pointer(struct btrfs_fs_info *fs_info,
+				    struct extent_buffer *eb,
+				    struct btrfs_block_group_cache **cache_ret);
 
 static inline bool btrfs_dev_is_sequential(struct btrfs_device *device, u64 pos)
 {
@@ -174,4 +177,18 @@ static inline void btrfs_hmzoned_data_io_unlock_logical(
 	btrfs_put_block_group(cache);
 }
 
+static inline void btrfs_hmzoned_meta_io_lock(struct btrfs_fs_info *fs_info)
+{
+	if (!btrfs_fs_incompat(fs_info, HMZONED))
+		return;
+	mutex_lock(&fs_info->hmzoned_meta_io_lock);
+}
+
+static inline void btrfs_hmzoned_meta_io_unlock(struct btrfs_fs_info *fs_info)
+{
+	if (!btrfs_fs_incompat(fs_info, HMZONED))
+		return;
+	mutex_unlock(&fs_info->hmzoned_meta_io_lock);
+}
+
 #endif
-- 
2.23.0