dm thin metadata: Fix ABBA deadlock by resetting dm_bufio_client

As described in commit 8111964f1b85 ("dm thin: Fix ABBA deadlock between
shrink_slab and dm_pool_abort_metadata"), ABBA deadlocks will be
triggered because shrinker_rwsem currently needs to held by
dm_pool_abort_metadata() as a side-effect of thin-pool metadata
operation failure.

The following three problem scenarios have been noticed:

1) Described by commit 8111964f1b85 ("dm thin: Fix ABBA deadlock between
   shrink_slab and dm_pool_abort_metadata")

2) shrinker_rwsem and throttle->lock
          P1(drop cache)                        P2(kworker)
drop_caches_sysctl_handler
 drop_slab
  shrink_slab
   down_read(&shrinker_rwsem)  - LOCK A
   do_shrink_slab
    super_cache_scan
     prune_icache_sb
      dispose_list
       evict
        ext4_evict_inode
         ext4_clear_inode
          ext4_discard_preallocations
           ext4_mb_load_buddy_gfp
            ext4_mb_init_cache
             ext4_wait_block_bitmap
              __ext4_error
               ext4_handle_error
                ext4_commit_super
                 ...
                 dm_submit_bio
                                     do_worker
                                      throttle_work_update
                                       down_write(&t->lock) -- LOCK B
                                      process_deferred_bios
                                       commit
                                        metadata_operation_failed
                                         dm_pool_abort_metadata
                                          dm_block_manager_create
                                           dm_bufio_client_create
                                            register_shrinker
                                             down_write(&shrinker_rwsem)
                                             -- LOCK A
                 thin_map
                  thin_bio_map
                   thin_defer_bio_with_throttle
                    throttle_lock
                     down_read(&t->lock)  - LOCK B

3) shrinker_rwsem and wait_on_buffer
          P1(drop cache)                            P2(kworker)
drop_caches_sysctl_handler
 drop_slab
  shrink_slab
   down_read(&shrinker_rwsem)  - LOCK A
   do_shrink_slab
   ...
    ext4_wait_block_bitmap
     __ext4_error
      ext4_handle_error
       jbd2_journal_abort
        jbd2_journal_update_sb_errno
         jbd2_write_superblock
          submit_bh
           // LOCK B
           // RELEASE B
                             do_worker
                              throttle_work_update
                               down_write(&t->lock) - LOCK B
                              process_deferred_bios
                               process_bio
                               commit
                                metadata_operation_failed
                                 dm_pool_abort_metadata
                                  dm_block_manager_create
                                   dm_bufio_client_create
                                    register_shrinker
                                     register_shrinker_prepared
                                      down_write(&shrinker_rwsem)  - LOCK A
                               bio_endio
      wait_on_buffer
       __wait_on_buffer

Fix these by resetting dm_bufio_client without holding shrinker_rwsem.

Fixes: 8111964f1b85 ("dm thin: Fix ABBA deadlock between shrink_slab and dm_pool_abort_metadata")
Cc: stable@vger.kernel.org
Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com>
Signed-off-by: Mike Snitzer <snitzer@kernel.org>

4 files changed, 12 insertions, 1 deletions

diff --git a/drivers/md/persistent-data/dm-block-manager.c b/drivers/md/persistent-data/dm-block-manager.c
index 7bdfc23f758a..0e010e1204aa 100644
--- a/drivers/md/persistent-data/dm-block-manager.c
+++ b/drivers/md/persistent-data/dm-block-manager.c
@@ -421,6 +421,12 @@ void dm_block_manager_destroy(struct dm_block_manager *bm)
 }
 EXPORT_SYMBOL_GPL(dm_block_manager_destroy);
 
+void dm_block_manager_reset(struct dm_block_manager *bm)
+{
+	dm_bufio_client_reset(bm->bufio);
+}
+EXPORT_SYMBOL_GPL(dm_block_manager_reset);
+
 unsigned int dm_bm_block_size(struct dm_block_manager *bm)
 {
 	return dm_bufio_get_block_size(bm->bufio);
diff --git a/drivers/md/persistent-data/dm-block-manager.h b/drivers/md/persistent-data/dm-block-manager.h
index 5746b0f82a03..f706d3de8d5a 100644
--- a/drivers/md/persistent-data/dm-block-manager.h
+++ b/drivers/md/persistent-data/dm-block-manager.h
@@ -36,6 +36,7 @@ struct dm_block_manager *dm_block_manager_create(
 	struct block_device *bdev, unsigned int block_size,
 	unsigned int max_held_per_thread);
 void dm_block_manager_destroy(struct dm_block_manager *bm);
+void dm_block_manager_reset(struct dm_block_manager *bm);
 
 unsigned int dm_bm_block_size(struct dm_block_manager *bm);
 dm_block_t dm_bm_nr_blocks(struct dm_block_manager *bm);
diff --git a/drivers/md/persistent-data/dm-space-map.h b/drivers/md/persistent-data/dm-space-map.h
index dab490353781..6bf69922b5ad 100644
--- a/drivers/md/persistent-data/dm-space-map.h
+++ b/drivers/md/persistent-data/dm-space-map.h
@@ -77,7 +77,8 @@ struct dm_space_map {
 
 static inline void dm_sm_destroy(struct dm_space_map *sm)
 {
-	sm->destroy(sm);
+	if (sm)
+		sm->destroy(sm);
 }
 
 static inline int dm_sm_extend(struct dm_space_map *sm, dm_block_t extra_blocks)
diff --git a/drivers/md/persistent-data/dm-transaction-manager.c b/drivers/md/persistent-data/dm-transaction-manager.c
index 6dc016248baf..c88fa6266203 100644
--- a/drivers/md/persistent-data/dm-transaction-manager.c
+++ b/drivers/md/persistent-data/dm-transaction-manager.c
@@ -199,6 +199,9 @@ EXPORT_SYMBOL_GPL(dm_tm_create_non_blocking_clone);
 
 void dm_tm_destroy(struct dm_transaction_manager *tm)
 {
+	if (!tm)
+		return;
+
 	if (!tm->is_clone)
 		wipe_shadow_table(tm);