Add STM32MP25 SPI NOR support

Merge series from patrice.chotard@foss.st.com:

This series adds SPI NOR support for STM32MP25 SoCs from STMicroelectronics.

On STM32MP25 SoCs family, an Octo Memory Manager block manages the muxing,
the memory area split, the chip select override and the time constraint
between its 2 Octo SPI children.

Due to these depedencies, this series adds support for:
  - Octo Memory Manager driver (not applied for SPI).
  - Octo SPI driver.
  - yaml schema for Octo Memory Manager and Octo SPI drivers.

The device tree files adds Octo Memory Manager and its 2 associated Octo
SPI chidren in stm32mp251.dtsi and adds SPI NOR support in stm32mp257f-ev1
board.

68 files changed, 486 insertions, 294 deletions

diff --git a/fs/bcachefs/Kconfig b/fs/bcachefs/Kconfig
index 85eea7a4dea3..fc7efd0a7525 100644
--- a/fs/bcachefs/Kconfig
+++ b/fs/bcachefs/Kconfig
@@ -61,6 +61,13 @@ config BCACHEFS_DEBUG
 	The resulting code will be significantly slower than normal; you
 	probably shouldn't select this option unless you're a developer.
 
+config BCACHEFS_INJECT_TRANSACTION_RESTARTS
+	bool "Randomly inject transaction restarts"
+	depends on BCACHEFS_DEBUG
+	help
+	Randomly inject transaction restarts in a few core paths - may have a
+	significant performance penalty
+
 config BCACHEFS_TESTS
 	bool "bcachefs unit and performance tests"
 	depends on BCACHEFS_FS
diff --git a/fs/bcachefs/alloc_background.c b/fs/bcachefs/alloc_background.c
index fc2ef33b67b3..3ea809990ef1 100644
--- a/fs/bcachefs/alloc_background.c
+++ b/fs/bcachefs/alloc_background.c
@@ -1803,7 +1803,6 @@ struct discard_buckets_state {
 	u64		open;
 	u64		need_journal_commit;
 	u64		discarded;
-	u64		need_journal_commit_this_dev;
 };
 
 static int bch2_discard_one_bucket(struct btree_trans *trans,
@@ -1827,11 +1826,11 @@ static int bch2_discard_one_bucket(struct btree_trans *trans,
 		goto out;
 	}
 
-	if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal,
-			c->journal.flushed_seq_ondisk,
-			pos.inode, pos.offset)) {
-		s->need_journal_commit++;
-		s->need_journal_commit_this_dev++;
+	u64 seq_ready = bch2_bucket_journal_seq_ready(&c->buckets_waiting_for_journal,
+						      pos.inode, pos.offset);
+	if (seq_ready > c->journal.flushed_seq_ondisk) {
+		if (seq_ready > c->journal.flushing_seq)
+			s->need_journal_commit++;
 		goto out;
 	}
 
@@ -1865,23 +1864,24 @@ static int bch2_discard_one_bucket(struct btree_trans *trans,
 		discard_locked = true;
 	}
 
-	if (!bkey_eq(*discard_pos_done, iter.pos) &&
-	    ca->mi.discard && !c->opts.nochanges) {
-		/*
-		 * This works without any other locks because this is the only
-		 * thread that removes items from the need_discard tree
-		 */
-		bch2_trans_unlock_long(trans);
-		blkdev_issue_discard(ca->disk_sb.bdev,
-				     k.k->p.offset * ca->mi.bucket_size,
-				     ca->mi.bucket_size,
-				     GFP_KERNEL);
-		*discard_pos_done = iter.pos;
+	if (!bkey_eq(*discard_pos_done, iter.pos)) {
 		s->discarded++;
+		*discard_pos_done = iter.pos;
 
-		ret = bch2_trans_relock_notrace(trans);
-		if (ret)
-			goto out;
+		if (ca->mi.discard && !c->opts.nochanges) {
+			/*
+			 * This works without any other locks because this is the only
+			 * thread that removes items from the need_discard tree
+			 */
+			bch2_trans_unlock_long(trans);
+			blkdev_issue_discard(ca->disk_sb.bdev,
+					     k.k->p.offset * ca->mi.bucket_size,
+					     ca->mi.bucket_size,
+					     GFP_KERNEL);
+			ret = bch2_trans_relock_notrace(trans);
+			if (ret)
+				goto out;
+		}
 	}
 
 	SET_BCH_ALLOC_V4_NEED_DISCARD(&a->v, false);
@@ -1929,6 +1929,9 @@ static void bch2_do_discards_work(struct work_struct *work)
 				   POS(ca->dev_idx, U64_MAX), 0, k,
 			bch2_discard_one_bucket(trans, ca, &iter, &discard_pos_done, &s, false)));
 
+	if (s.need_journal_commit > dev_buckets_available(ca, BCH_WATERMARK_normal))
+		bch2_journal_flush_async(&c->journal, NULL);
+
 	trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded,
 			      bch2_err_str(ret));
 
@@ -2024,7 +2027,7 @@ static void bch2_do_discards_fast_work(struct work_struct *work)
 			break;
 	}
 
-	trace_discard_buckets(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret));
+	trace_discard_buckets_fast(c, s.seen, s.open, s.need_journal_commit, s.discarded, bch2_err_str(ret));
 
 	bch2_trans_put(trans);
 	percpu_ref_put(&ca->io_ref);
diff --git a/fs/bcachefs/alloc_foreground.c b/fs/bcachefs/alloc_foreground.c
index 6df41c331a52..5a781fb4c794 100644
--- a/fs/bcachefs/alloc_foreground.c
+++ b/fs/bcachefs/alloc_foreground.c
@@ -205,8 +205,12 @@ static inline bool may_alloc_bucket(struct bch_fs *c,
 		return false;
 	}
 
-	if (bch2_bucket_needs_journal_commit(&c->buckets_waiting_for_journal,
-			c->journal.flushed_seq_ondisk, bucket.inode, bucket.offset)) {
+	u64 journal_seq_ready =
+		bch2_bucket_journal_seq_ready(&c->buckets_waiting_for_journal,
+					      bucket.inode, bucket.offset);
+	if (journal_seq_ready > c->journal.flushed_seq_ondisk) {
+		if (journal_seq_ready > c->journal.flushing_seq)
+			s->need_journal_commit++;
 		s->skipped_need_journal_commit++;
 		return false;
 	}
@@ -570,7 +574,7 @@ alloc:
 		? bch2_bucket_alloc_freelist(trans, ca, watermark, &s, cl)
 		: bch2_bucket_alloc_early(trans, ca, watermark, &s, cl);
 
-	if (s.skipped_need_journal_commit * 2 > avail)
+	if (s.need_journal_commit * 2 > avail)
 		bch2_journal_flush_async(&c->journal, NULL);
 
 	if (!ob && s.btree_bitmap != BTREE_BITMAP_ANY) {
diff --git a/fs/bcachefs/alloc_types.h b/fs/bcachefs/alloc_types.h
index 9bbb28e90b93..4aa8ee026cb8 100644
--- a/fs/bcachefs/alloc_types.h
+++ b/fs/bcachefs/alloc_types.h
@@ -18,6 +18,7 @@ struct bucket_alloc_state {
 	u64	buckets_seen;
 	u64	skipped_open;
 	u64	skipped_need_journal_commit;
+	u64	need_journal_commit;
 	u64	skipped_nocow;
 	u64	skipped_nouse;
 	u64	skipped_mi_btree_bitmap;
diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c
index 5988219c6908..e32fce4fd258 100644
--- a/fs/bcachefs/btree_iter.c
+++ b/fs/bcachefs/btree_iter.c
@@ -2357,6 +2357,12 @@ struct bkey_s_c bch2_btree_iter_peek_max(struct btree_iter *iter, struct bpos en
 	bch2_btree_iter_verify_entry_exit(iter);
 	EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bkey_eq(end, POS_MAX));
 
+	ret = trans_maybe_inject_restart(trans, _RET_IP_);
+	if (unlikely(ret)) {
+		k = bkey_s_c_err(ret);
+		goto out_no_locked;
+	}
+
 	if (iter->update_path) {
 		bch2_path_put_nokeep(trans, iter->update_path,
 				     iter->flags & BTREE_ITER_intent);
@@ -2622,6 +2628,12 @@ struct bkey_s_c bch2_btree_iter_peek_prev_min(struct btree_iter *iter, struct bp
 	bch2_btree_iter_verify_entry_exit(iter);
 	EBUG_ON((iter->flags & BTREE_ITER_filter_snapshots) && bpos_eq(end, POS_MIN));
 
+	int ret = trans_maybe_inject_restart(trans, _RET_IP_);
+	if (unlikely(ret)) {
+		k = bkey_s_c_err(ret);
+		goto out_no_locked;
+	}
+
 	while (1) {
 		k = __bch2_btree_iter_peek_prev(iter, search_key);
 		if (unlikely(!k.k))
@@ -2749,6 +2761,12 @@ struct bkey_s_c bch2_btree_iter_peek_slot(struct btree_iter *iter)
 	bch2_btree_iter_verify_entry_exit(iter);
 	EBUG_ON(btree_iter_path(trans, iter)->level && (iter->flags & BTREE_ITER_with_key_cache));
 
+	ret = trans_maybe_inject_restart(trans, _RET_IP_);
+	if (unlikely(ret)) {
+		k = bkey_s_c_err(ret);
+		goto out_no_locked;
+	}
+
 	/* extents can't span inode numbers: */
 	if ((iter->flags & BTREE_ITER_is_extents) &&
 	    unlikely(iter->pos.offset == KEY_OFFSET_MAX)) {
@@ -3106,6 +3124,10 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size)
 
 	WARN_ON_ONCE(new_bytes > BTREE_TRANS_MEM_MAX);
 
+	ret = trans_maybe_inject_restart(trans, _RET_IP_);
+	if (ret)
+		return ERR_PTR(ret);
+
 	struct btree_transaction_stats *s = btree_trans_stats(trans);
 	s->max_mem = max(s->max_mem, new_bytes);
 
@@ -3163,7 +3185,8 @@ out_new_mem:
 
 	if (old_bytes) {
 		trace_and_count(c, trans_restart_mem_realloced, trans, _RET_IP_, new_bytes);
-		return ERR_PTR(btree_trans_restart(trans, BCH_ERR_transaction_restart_mem_realloced));
+		return ERR_PTR(btree_trans_restart_ip(trans,
+					BCH_ERR_transaction_restart_mem_realloced, _RET_IP_));
 	}
 out_change_top:
 	p = trans->mem + trans->mem_top;
@@ -3271,6 +3294,14 @@ u32 bch2_trans_begin(struct btree_trans *trans)
 
 	trans->last_begin_ip = _RET_IP_;
 
+#ifdef CONFIG_BCACHEFS_INJECT_TRANSACTION_RESTARTS
+	if (trans->restarted) {
+		trans->restart_count_this_trans++;
+	} else {
+		trans->restart_count_this_trans = 0;
+	}
+#endif
+
 	trans_set_locked(trans, false);
 
 	if (trans->restarted) {
diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h
index b9538e6e6d65..b96157f3dc9c 100644
--- a/fs/bcachefs/btree_iter.h
+++ b/fs/bcachefs/btree_iter.h
@@ -355,6 +355,18 @@ static int btree_trans_restart(struct btree_trans *trans, int err)
 	return btree_trans_restart_ip(trans, err, _THIS_IP_);
 }
 
+static inline int trans_maybe_inject_restart(struct btree_trans *trans, unsigned long ip)
+{
+#ifdef CONFIG_BCACHEFS_INJECT_TRANSACTION_RESTARTS
+	if (!(ktime_get_ns() & ~(~0ULL << min(63, (10 + trans->restart_count_this_trans))))) {
+		trace_and_count(trans->c, trans_restart_injected, trans, ip);
+		return btree_trans_restart_ip(trans,
+					BCH_ERR_transaction_restart_fault_inject, ip);
+	}
+#endif
+	return 0;
+}
+
 bool bch2_btree_node_upgrade(struct btree_trans *,
 			     struct btree_path *, unsigned);
 
@@ -739,7 +751,7 @@ transaction_restart:							\
 	if (!_ret2)							\
 		bch2_trans_verify_not_restarted(_trans, _restart_count);\
 									\
-	_ret2 ?: trans_was_restarted(_trans, _restart_count);		\
+	_ret2 ?: trans_was_restarted(_trans, _orig_restart_count);		\
 })
 
 #define for_each_btree_key_max_continue(_trans, _iter,			\
diff --git a/fs/bcachefs/btree_key_cache.c b/fs/bcachefs/btree_key_cache.c
index c378b97ebeca..1821f40c161a 100644
--- a/fs/bcachefs/btree_key_cache.c
+++ b/fs/bcachefs/btree_key_cache.c
@@ -748,7 +748,6 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc)
 				rcu_read_unlock();
 				mutex_lock(&bc->table.mutex);
 				mutex_unlock(&bc->table.mutex);
-				rcu_read_lock();
 				continue;
 			}
 			for (i = 0; i < tbl->size; i++)
diff --git a/fs/bcachefs/btree_trans_commit.c b/fs/bcachefs/btree_trans_commit.c
index 2760dd9569ed..c4f524b2ca9a 100644
--- a/fs/bcachefs/btree_trans_commit.c
+++ b/fs/bcachefs/btree_trans_commit.c
@@ -999,6 +999,10 @@ int __bch2_trans_commit(struct btree_trans *trans, unsigned flags)
 
 	bch2_trans_verify_not_unlocked_or_in_restart(trans);
 
+	ret = trans_maybe_inject_restart(trans, _RET_IP_);
+	if (unlikely(ret))
+		goto out_reset;
+
 	if (!trans->nr_updates &&
 	    !trans->journal_entries_u64s)
 		goto out_reset;
diff --git a/fs/bcachefs/btree_types.h b/fs/bcachefs/btree_types.h
index a6f251eb4164..a09cbe9cd94f 100644
--- a/fs/bcachefs/btree_types.h
+++ b/fs/bcachefs/btree_types.h
@@ -509,6 +509,9 @@ struct btree_trans {
 	bool			notrace_relock_fail:1;
 	enum bch_errcode	restarted:16;
 	u32			restart_count;
+#ifdef CONFIG_BCACHEFS_INJECT_TRANSACTION_RESTARTS
+	u32			restart_count_this_trans;
+#endif
 
 	u64			last_begin_time;
 	unsigned long		last_begin_ip;
diff --git a/fs/bcachefs/btree_update_interior.h b/fs/bcachefs/btree_update_interior.h
index 7930ffea3075..26d646e1275c 100644
--- a/fs/bcachefs/btree_update_interior.h
+++ b/fs/bcachefs/btree_update_interior.h
@@ -278,12 +278,12 @@ static inline struct btree_node_entry *want_new_bset(struct bch_fs *c, struct bt
 {
 	struct bset_tree *t = bset_tree_last(b);
 	struct btree_node_entry *bne = max(write_block(b),
-			(void *) btree_bkey_last(b, bset_tree_last(b)));
+			(void *) btree_bkey_last(b, t));
 	ssize_t remaining_space =
 		__bch2_btree_u64s_remaining(b, bne->keys.start);
 
 	if (unlikely(bset_written(b, bset(b, t)))) {
-		if (remaining_space > (ssize_t) (block_bytes(c) >> 3))
+		if (b->written + block_sectors(c) <= btree_sectors(c))
 			return bne;
 	} else {
 		if (unlikely(bset_u64s(t) * sizeof(u64) > btree_write_set_buffer(b)) &&
diff --git a/fs/bcachefs/buckets_waiting_for_journal.c b/fs/bcachefs/buckets_waiting_for_journal.c
index f9fb150eda70..c8a488e6b7b8 100644
--- a/fs/bcachefs/buckets_waiting_for_journal.c
+++ b/fs/bcachefs/buckets_waiting_for_journal.c
@@ -22,23 +22,21 @@ static void bucket_table_init(struct buckets_waiting_for_journal_table *t, size_
 	memset(t->d, 0, sizeof(t->d[0]) << t->bits);
 }
 
-bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *b,
-				      u64 flushed_seq,
-				      unsigned dev, u64 bucket)
+u64 bch2_bucket_journal_seq_ready(struct buckets_waiting_for_journal *b,
+				  unsigned dev, u64 bucket)
 {
 	struct buckets_waiting_for_journal_table *t;
 	u64 dev_bucket = (u64) dev << 56 | bucket;
-	bool ret = false;
-	unsigned i;
+	u64 ret = 0;
 
 	mutex_lock(&b->lock);
 	t = b->t;
 
-	for (i = 0; i < ARRAY_SIZE(t->hash_seeds); i++) {
+	for (unsigned i = 0; i < ARRAY_SIZE(t->hash_seeds); i++) {
 		struct bucket_hashed *h = bucket_hash(t, i, dev_bucket);
 
 		if (h->dev_bucket == dev_bucket) {
-			ret = h->journal_seq > flushed_seq;
+			ret = h->journal_seq;
 			break;
 		}
 	}
diff --git a/fs/bcachefs/buckets_waiting_for_journal.h b/fs/bcachefs/buckets_waiting_for_journal.h
index d2ae19cbe18c..365619ca44c8 100644
--- a/fs/bcachefs/buckets_waiting_for_journal.h
+++ b/fs/bcachefs/buckets_waiting_for_journal.h
@@ -4,8 +4,8 @@
 
 #include "buckets_waiting_for_journal_types.h"
 
-bool bch2_bucket_needs_journal_commit(struct buckets_waiting_for_journal *,
-				      u64, unsigned, u64);
+u64 bch2_bucket_journal_seq_ready(struct buckets_waiting_for_journal *,
+				  unsigned, u64);
 int bch2_set_bucket_needs_journal_commit(struct buckets_waiting_for_journal *,
 					 u64, unsigned, u64, u64);
 
diff --git a/fs/bcachefs/disk_accounting.h b/fs/bcachefs/disk_accounting.h
index 5360cbb3ec29..f4372cafea2e 100644
--- a/fs/bcachefs/disk_accounting.h
+++ b/fs/bcachefs/disk_accounting.h
@@ -210,11 +210,13 @@ static inline void bch2_accounting_mem_read_counters(struct bch_accounting_mem *
 static inline void bch2_accounting_mem_read(struct bch_fs *c, struct bpos p,
 					    u64 *v, unsigned nr)
 {
+	percpu_down_read(&c->mark_lock);
 	struct bch_accounting_mem *acc = &c->accounting;
 	unsigned idx = eytzinger0_find(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]),
 				       accounting_pos_cmp, &p);
 
 	bch2_accounting_mem_read_counters(acc, idx, v, nr, false);
+	percpu_up_read(&c->mark_lock);
 }
 
 static inline struct bversion journal_pos_to_bversion(struct journal_res *res, unsigned offset)
diff --git a/fs/bcachefs/inode.h b/fs/bcachefs/inode.h
index d2e134528f0e..428b9be6af34 100644
--- a/fs/bcachefs/inode.h
+++ b/fs/bcachefs/inode.h
@@ -285,12 +285,14 @@ void bch2_inode_opts_get(struct bch_io_opts *, struct bch_fs *,
 			 struct bch_inode_unpacked *);
 int bch2_inum_opts_get(struct btree_trans*, subvol_inum, struct bch_io_opts *);
 
+#include "rebalance.h"
+
 static inline struct bch_extent_rebalance
 bch2_inode_rebalance_opts_get(struct bch_fs *c, struct bch_inode_unpacked *inode)
 {
 	struct bch_io_opts io_opts;
 	bch2_inode_opts_get(&io_opts, c, inode);
-	return io_opts_to_rebalance_opts(&io_opts);
+	return io_opts_to_rebalance_opts(c, &io_opts);
 }
 
 int bch2_inode_rm_snapshot(struct btree_trans *, u64, u32);
diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c
index dd508d93e9fc..03892388832b 100644
--- a/fs/bcachefs/io_write.c
+++ b/fs/bcachefs/io_write.c
@@ -411,6 +411,16 @@ void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op)
 	__bch2_write_op_error(out, op, op->pos.offset);
 }
 
+static void bch2_write_op_error_trans(struct btree_trans *trans, struct printbuf *out,
+				      struct bch_write_op *op, u64 offset)
+{
+	bch2_inum_offset_err_msg_trans(trans, out,
+				       (subvol_inum) { op->subvol, op->pos.inode, },
+				       offset << 9);
+	prt_printf(out, "write error%s: ",
+		   op->flags & BCH_WRITE_MOVE ? "(internal move)" : "");
+}
+
 void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c,
 			       enum bch_data_type type,
 			       const struct bkey_i *k,
@@ -1193,7 +1203,7 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op)
 			struct bkey_i *insert = bch2_keylist_front(&op->insert_keys);
 
 			struct printbuf buf = PRINTBUF;
-			__bch2_write_op_error(&buf, op, bkey_start_offset(&insert->k));
+			bch2_write_op_error_trans(trans, &buf, op, bkey_start_offset(&insert->k));
 			prt_printf(&buf, "btree update error: %s", bch2_err_str(ret));
 			bch_err_ratelimited(c, "%s", buf.buf);
 			printbuf_exit(&buf);
diff --git a/fs/bcachefs/journal.c b/fs/bcachefs/journal.c
index cb2c3722f674..24c294d4634e 100644
--- a/fs/bcachefs/journal.c
+++ b/fs/bcachefs/journal.c
@@ -319,6 +319,16 @@ void bch2_journal_halt(struct journal *j)
 	spin_unlock(&j->lock);
 }
 
+void bch2_journal_halt_locked(struct journal *j)
+{
+	lockdep_assert_held(&j->lock);
+
+	__journal_entry_close(j, JOURNAL_ENTRY_ERROR_VAL, true);
+	if (!j->err_seq)
+		j->err_seq = journal_cur_seq(j);
+	journal_wake(j);
+}
+
 static bool journal_entry_want_write(struct journal *j)
 {
 	bool ret = !journal_entry_is_open(j) ||
@@ -381,9 +391,12 @@ static int journal_entry_open(struct journal *j)
 	if (nr_unwritten_journal_entries(j) == ARRAY_SIZE(j->buf))
 		return JOURNAL_ERR_max_in_flight;
 
-	if (bch2_fs_fatal_err_on(journal_cur_seq(j) >= JOURNAL_SEQ_MAX,
-				 c, "cannot start: journal seq overflow"))
+	if (journal_cur_seq(j) >= JOURNAL_SEQ_MAX) {
+		bch_err(c, "cannot start: journal seq overflow");
+		if (bch2_fs_emergency_read_only_locked(c))
+			bch_err(c, "fatal error - emergency read only");
 		return JOURNAL_ERR_insufficient_devices; /* -EROFS */
+	}
 
 	BUG_ON(!j->cur_entry_sectors);
 
@@ -783,6 +796,7 @@ recheck_need_open:
 	}
 
 	buf->must_flush = true;
+	j->flushing_seq = max(j->flushing_seq, seq);
 
 	if (parent && !closure_wait(&buf->wait, parent))
 		BUG();
diff --git a/fs/bcachefs/journal.h b/fs/bcachefs/journal.h
index dccddd5420ad..107f7f901cd9 100644
--- a/fs/bcachefs/journal.h
+++ b/fs/bcachefs/journal.h
@@ -409,6 +409,7 @@ bool bch2_journal_noflush_seq(struct journal *, u64, u64);
 int bch2_journal_meta(struct journal *);
 
 void bch2_journal_halt(struct journal *);
+void bch2_journal_halt_locked(struct journal *);
 
 static inline int bch2_journal_error(struct journal *j)
 {
diff --git a/fs/bcachefs/journal_reclaim.c b/fs/bcachefs/journal_reclaim.c
index 6a9cefb635d6..d373cd181a7f 100644
--- a/fs/bcachefs/journal_reclaim.c
+++ b/fs/bcachefs/journal_reclaim.c
@@ -384,12 +384,16 @@ void bch2_journal_pin_drop(struct journal *j,
 	spin_unlock(&j->lock);
 }
 
-static enum journal_pin_type journal_pin_type(journal_pin_flush_fn fn)
+static enum journal_pin_type journal_pin_type(struct journal_entry_pin *pin,
+					      journal_pin_flush_fn fn)
 {
 	if (fn == bch2_btree_node_flush0 ||
-	    fn == bch2_btree_node_flush1)
-		return JOURNAL_PIN_TYPE_btree;
-	else if (fn == bch2_btree_key_cache_journal_flush)
+	    fn == bch2_btree_node_flush1) {
+		unsigned idx = fn == bch2_btree_node_flush1;
+		struct btree *b = container_of(pin, struct btree, writes[idx].journal);
+
+		return JOURNAL_PIN_TYPE_btree0 - b->c.level;
+	} else if (fn == bch2_btree_key_cache_journal_flush)
 		return JOURNAL_PIN_TYPE_key_cache;
 	else
 		return JOURNAL_PIN_TYPE_other;
@@ -441,7 +445,7 @@ void bch2_journal_pin_copy(struct journal *j,
 
 	bool reclaim = __journal_pin_drop(j, dst);
 
-	bch2_journal_pin_set_locked(j, seq, dst, flush_fn, journal_pin_type(flush_fn));
+	bch2_journal_pin_set_locked(j, seq, dst, flush_fn, journal_pin_type(dst, flush_fn));
 
 	if (reclaim)
 		bch2_journal_reclaim_fast(j);
@@ -465,7 +469,7 @@ void bch2_journal_pin_set(struct journal *j, u64 seq,
 
 	bool reclaim = __journal_pin_drop(j, pin);
 
-	bch2_journal_pin_set_locked(j, seq, pin, flush_fn, journal_pin_type(flush_fn));
+	bch2_journal_pin_set_locked(j, seq, pin, flush_fn, journal_pin_type(pin, flush_fn));
 
 	if (reclaim)
 		bch2_journal_reclaim_fast(j);
@@ -587,7 +591,7 @@ static size_t journal_flush_pins(struct journal *j,
 		spin_lock(&j->lock);
 		/* Pin might have been dropped or rearmed: */
 		if (likely(!err && !j->flush_in_progress_dropped))
-			list_move(&pin->list, &journal_seq_pin(j, seq)->flushed[journal_pin_type(flush_fn)]);
+			list_move(&pin->list, &journal_seq_pin(j, seq)->flushed[journal_pin_type(pin, flush_fn)]);
 		j->flush_in_progress = NULL;
 		j->flush_in_progress_dropped = false;
 		spin_unlock(&j->lock);
@@ -869,18 +873,13 @@ static int journal_flush_done(struct journal *j, u64 seq_to_flush,
 
 	mutex_lock(&j->reclaim_lock);
 
-	if (journal_flush_pins_or_still_flushing(j, seq_to_flush,
-			       BIT(JOURNAL_PIN_TYPE_key_cache)|
-			       BIT(JOURNAL_PIN_TYPE_other))) {
-		*did_work = true;
-		goto unlock;
-	}
-
-	if (journal_flush_pins_or_still_flushing(j, seq_to_flush,
-			       BIT(JOURNAL_PIN_TYPE_btree))) {
-		*did_work = true;
-		goto unlock;
-	}
+	for (int type = JOURNAL_PIN_TYPE_NR - 1;
+	     type >= 0;
+	     --type)
+		if (journal_flush_pins_or_still_flushing(j, seq_to_flush, BIT(type))) {
+			*did_work = true;
+			goto unlock;
+		}
 
 	if (seq_to_flush > journal_cur_seq(j))
 		bch2_journal_entry_close(j);
diff --git a/fs/bcachefs/journal_types.h b/fs/bcachefs/journal_types.h
index 3ba433a48eb8..1ef3a28ed6ab 100644
--- a/fs/bcachefs/journal_types.h
+++ b/fs/bcachefs/journal_types.h
@@ -53,7 +53,10 @@ struct journal_buf {
  */
 
 enum journal_pin_type {
-	JOURNAL_PIN_TYPE_btree,
+	JOURNAL_PIN_TYPE_btree3,
+	JOURNAL_PIN_TYPE_btree2,
+	JOURNAL_PIN_TYPE_btree1,
+	JOURNAL_PIN_TYPE_btree0,
 	JOURNAL_PIN_TYPE_key_cache,
 	JOURNAL_PIN_TYPE_other,
 	JOURNAL_PIN_TYPE_NR,
@@ -237,6 +240,7 @@ struct journal {
 	/* seq, last_seq from the most recent journal entry successfully written */
 	u64			seq_ondisk;
 	u64			flushed_seq_ondisk;
+	u64			flushing_seq;
 	u64			last_seq_ondisk;
 	u64			err_seq;
 	u64			last_empty_seq;
diff --git a/fs/bcachefs/opts.h b/fs/bcachefs/opts.h
index a182b5d454ba..9d397fc2a1f0 100644
--- a/fs/bcachefs/opts.h
+++ b/fs/bcachefs/opts.h
@@ -659,18 +659,4 @@ static inline void bch2_io_opts_fixups(struct bch_io_opts *opts)
 struct bch_io_opts bch2_opts_to_inode_opts(struct bch_opts);
 bool bch2_opt_is_inode_opt(enum bch_opt_id);
 
-/* rebalance opts: */
-
-static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_io_opts *opts)
-{
-	return (struct bch_extent_rebalance) {
-		.type = BIT(BCH_EXTENT_ENTRY_rebalance),
-#define x(_name)							\
-		._name = opts->_name,					\
-		._name##_from_inode = opts->_name##_from_inode,
-		BCH_REBALANCE_OPTS()
-#undef x
-	};
-};
-
 #endif /* _BCACHEFS_OPTS_H */
diff --git a/fs/bcachefs/rebalance.c b/fs/bcachefs/rebalance.c
index 4adc74cd3f70..d0a1f5cd5c2b 100644
--- a/fs/bcachefs/rebalance.c
+++ b/fs/bcachefs/rebalance.c
@@ -121,12 +121,10 @@ u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *c, struct bkey_s_c k)
 		}
 	}
 incompressible:
-	if (opts->background_target &&
-	    bch2_target_accepts_data(c, BCH_DATA_user, opts->background_target)) {
+	if (opts->background_target)
 		bkey_for_each_ptr_decode(k.k, ptrs, p, entry)
 			if (!p.ptr.cached && !bch2_dev_in_target(c, p.ptr.dev, opts->background_target))
 				sectors += p.crc.compressed_size;
-	}
 
 	return sectors;
 }
@@ -140,7 +138,7 @@ static bool bch2_bkey_rebalance_needs_update(struct bch_fs *c, struct bch_io_opt
 	const struct bch_extent_rebalance *old = bch2_bkey_rebalance_opts(k);
 
 	if (k.k->type == KEY_TYPE_reflink_v || bch2_bkey_ptrs_need_rebalance(c, opts, k)) {
-		struct bch_extent_rebalance new = io_opts_to_rebalance_opts(opts);
+		struct bch_extent_rebalance new = io_opts_to_rebalance_opts(c, opts);
 		return old == NULL || memcmp(old, &new, sizeof(new));
 	} else {
 		return old != NULL;
@@ -163,7 +161,7 @@ int bch2_bkey_set_needs_rebalance(struct bch_fs *c, struct bch_io_opts *opts,
 			k.k->u64s += sizeof(*old) / sizeof(u64);
 		}
 
-		*old = io_opts_to_rebalance_opts(opts);
+		*old = io_opts_to_rebalance_opts(c, opts);
 	} else {
 		if (old)
 			extent_entry_drop(k, (union bch_extent_entry *) old);
diff --git a/fs/bcachefs/rebalance.h b/fs/bcachefs/rebalance.h
index 0a0821ab895d..62a3859d3823 100644
--- a/fs/bcachefs/rebalance.h
+++ b/fs/bcachefs/rebalance.h
@@ -4,8 +4,28 @@
 
 #include "compress.h"
 #include "disk_groups.h"
+#include "opts.h"
 #include "rebalance_types.h"
 
+static inline struct bch_extent_rebalance io_opts_to_rebalance_opts(struct bch_fs *c,
+								    struct bch_io_opts *opts)
+{
+	struct bch_extent_rebalance r = {
+		.type = BIT(BCH_EXTENT_ENTRY_rebalance),
+#define x(_name)							\
+		._name = opts->_name,					\
+		._name##_from_inode = opts->_name##_from_inode,
+		BCH_REBALANCE_OPTS()
+#undef x
+	};
+
+	if (r.background_target &&
+	    !bch2_target_accepts_data(c, BCH_DATA_user, r.background_target))
+		r.background_target = 0;
+
+	return r;
+};
+
 u64 bch2_bkey_sectors_need_rebalance(struct bch_fs *, struct bkey_s_c);
 int bch2_bkey_set_needs_rebalance(struct bch_fs *, struct bch_io_opts *, struct bkey_i *);
 int bch2_get_update_rebalance_opts(struct btree_trans *,
diff --git a/fs/bcachefs/reflink.c b/fs/bcachefs/reflink.c
index 93ba4f4e47ca..376fd0a6e868 100644
--- a/fs/bcachefs/reflink.c
+++ b/fs/bcachefs/reflink.c
@@ -381,6 +381,8 @@ static s64 gc_trigger_reflink_p_segment(struct btree_trans *trans,
 not_found:
 	if (flags & BTREE_TRIGGER_check_repair) {
 		ret = bch2_indirect_extent_missing_error(trans, p, *idx, next_idx, false);
+		if (ret == -BCH_ERR_missing_indirect_extent)
+			ret = 0;
 		if (ret)
 			goto err;
 	}
diff --git a/fs/bcachefs/sb-errors_format.h b/fs/bcachefs/sb-errors_format.h
index ea0a18364751..b86ec013d7d7 100644
--- a/fs/bcachefs/sb-errors_format.h
+++ b/fs/bcachefs/sb-errors_format.h
@@ -180,9 +180,9 @@ enum bch_fsck_flags {
 	x(ptr_crc_nonce_mismatch,				162,	0)		\
 	x(ptr_stripe_redundant,					163,	0)		\
 	x(reservation_key_nr_replicas_invalid,			164,	0)		\
-	x(reflink_v_refcount_wrong,				165,	0)		\
+	x(reflink_v_refcount_wrong,				165,	FSCK_AUTOFIX)	\
 	x(reflink_v_pos_bad,					292,	0)		\
-	x(reflink_p_to_missing_reflink_v,			166,	0)		\
+	x(reflink_p_to_missing_reflink_v,			166,	FSCK_AUTOFIX)	\
 	x(reflink_refcount_underflow,				293,	0)		\
 	x(stripe_pos_bad,					167,	0)		\
 	x(stripe_val_size_bad,					168,	0)		\
diff --git a/fs/bcachefs/subvolume.c b/fs/bcachefs/subvolume.c
index e3d0475232e5..b7b96283c316 100644
--- a/fs/bcachefs/subvolume.c
+++ b/fs/bcachefs/subvolume.c
@@ -428,7 +428,7 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid)
 		bch2_bkey_get_iter_typed(trans, &snapshot_iter,
 				BTREE_ID_snapshots, POS(0, snapid),
 				0, snapshot);
-	ret = bkey_err(subvol);
+	ret = bkey_err(snapshot);
 	bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c,
 				"missing snapshot %u", snapid);
 	if (ret)
@@ -440,6 +440,11 @@ static int __bch2_subvolume_delete(struct btree_trans *trans, u32 subvolid)
 		bch2_bkey_get_iter_typed(trans, &snapshot_tree_iter,
 				BTREE_ID_snapshot_trees, POS(0, treeid),
 				0, snapshot_tree);
+	ret = bkey_err(snapshot_tree);
+	bch2_fs_inconsistent_on(bch2_err_matches(ret, ENOENT), trans->c,
+				"missing snapshot tree %u", treeid);
+	if (ret)
+		goto err;
 
 	if (le32_to_cpu(snapshot_tree.v->master_subvol) == subvolid) {
 		struct bkey_i_snapshot_tree *snapshot_tree_mut =
diff --git a/fs/bcachefs/super.c b/fs/bcachefs/super.c
index d97ea7bd1171..6d97d412fed9 100644
--- a/fs/bcachefs/super.c
+++ b/fs/bcachefs/super.c
@@ -411,6 +411,17 @@ bool bch2_fs_emergency_read_only(struct bch_fs *c)
 	return ret;
 }
 
+bool bch2_fs_emergency_read_only_locked(struct bch_fs *c)
+{
+	bool ret = !test_and_set_bit(BCH_FS_emergency_ro, &c->flags);
+
+	bch2_journal_halt_locked(&c->journal);
+	bch2_fs_read_only_async(c);
+
+	wake_up(&bch2_read_only_wait);
+	return ret;
+}
+
 static int bch2_fs_read_write_late(struct bch_fs *c)
 {
 	int ret;
diff --git a/fs/bcachefs/super.h b/fs/bcachefs/super.h
index fa6d52216510..04f8287eff5c 100644
--- a/fs/bcachefs/super.h
+++ b/fs/bcachefs/super.h
@@ -29,6 +29,7 @@ int bch2_dev_resize(struct bch_fs *, struct bch_dev *, u64);
 struct bch_dev *bch2_dev_lookup(struct bch_fs *, const char *);
 
 bool bch2_fs_emergency_read_only(struct bch_fs *);
+bool bch2_fs_emergency_read_only_locked(struct bch_fs *);
 void bch2_fs_read_only(struct bch_fs *);
 
 int bch2_fs_read_write(struct bch_fs *);
diff --git a/fs/bcachefs/trace.h b/fs/bcachefs/trace.h
index 56a5a7fbc0fd..c1b51009edf6 100644
--- a/fs/bcachefs/trace.h
+++ b/fs/bcachefs/trace.h
@@ -727,7 +727,7 @@ DEFINE_EVENT(fs_str, bucket_alloc_fail,
 	TP_ARGS(c, str)
 );
 
-TRACE_EVENT(discard_buckets,
+DECLARE_EVENT_CLASS(discard_buckets_class,
 	TP_PROTO(struct bch_fs *c, u64 seen, u64 open,
 		 u64 need_journal_commit, u64 discarded, const char *err),
 	TP_ARGS(c, seen, open, need_journal_commit, discarded, err),
@@ -759,6 +759,18 @@ TRACE_EVENT(discard_buckets,
 		  __entry->err)
 );
 
+DEFINE_EVENT(discard_buckets_class, discard_buckets,
+	TP_PROTO(struct bch_fs *c, u64 seen, u64 open,
+		 u64 need_journal_commit, u64 discarded, const char *err),
+	TP_ARGS(c, seen, open, need_journal_commit, discarded, err)
+);
+
+DEFINE_EVENT(discard_buckets_class, discard_buckets_fast,
+	TP_PROTO(struct bch_fs *c, u64 seen, u64 open,
+		 u64 need_journal_commit, u64 discarded, const char *err),
+	TP_ARGS(c, seen, open, need_journal_commit, discarded, err)
+);
+
 TRACE_EVENT(bucket_invalidate,
 	TP_PROTO(struct bch_fs *c, unsigned dev, u64 bucket, u32 sectors),
 	TP_ARGS(c, dev, bucket, sectors),
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 92071ca0655f..3dc5a35dd19b 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -1496,6 +1496,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 
 		if (!p->skip_locking) {
 			btrfs_unlock_up_safe(p, parent_level + 1);
+			btrfs_maybe_reset_lockdep_class(root, tmp);
 			tmp_locked = true;
 			btrfs_tree_read_lock(tmp);
 			btrfs_release_path(p);
@@ -1539,6 +1540,7 @@ read_block_for_search(struct btrfs_root *root, struct btrfs_path *p,
 
 	if (!p->skip_locking) {
 		ASSERT(ret == -EAGAIN);
+		btrfs_maybe_reset_lockdep_class(root, tmp);
 		tmp_locked = true;
 		btrfs_tree_read_lock(tmp);
 		btrfs_release_path(p);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index a2cac9d0a1a9..b2fae67f8fa3 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -523,8 +523,6 @@ static void end_bbio_data_read(struct btrfs_bio *bbio)
 		u64 end;
 		u32 len;
 
-		/* For now only order 0 folios are supported for data. */
-		ASSERT(folio_order(folio) == 0);
 		btrfs_debug(fs_info,
 			"%s: bi_sector=%llu, err=%d, mirror=%u",
 			__func__, bio->bi_iter.bi_sector, bio->bi_status,
@@ -552,7 +550,6 @@ static void end_bbio_data_read(struct btrfs_bio *bbio)
 
 		if (likely(uptodate)) {
 			loff_t i_size = i_size_read(inode);
-			pgoff_t end_index = i_size >> folio_shift(folio);
 
 			/*
 			 * Zero out the remaining part if this range straddles
@@ -561,9 +558,11 @@ static void end_bbio_data_read(struct btrfs_bio *bbio)
 			 * Here we should only zero the range inside the folio,
 			 * not touch anything else.
 			 *
-			 * NOTE: i_size is exclusive while end is inclusive.
+			 * NOTE: i_size is exclusive while end is inclusive and
+			 * folio_contains() takes PAGE_SIZE units.
 			 */
-			if (folio_index(folio) == end_index && i_size <= end) {
+			if (folio_contains(folio, i_size >> PAGE_SHIFT) &&
+			    i_size <= end) {
 				u32 zero_start = max(offset_in_folio(folio, i_size),
 						     offset_in_folio(folio, start));
 				u32 zero_len = offset_in_folio(folio, end) + 1 -
@@ -899,7 +898,6 @@ static struct extent_map *get_extent_map(struct btrfs_inode *inode,
 					 u64 len, struct extent_map **em_cached)
 {
 	struct extent_map *em;
-	struct extent_state *cached_state = NULL;
 
 	ASSERT(em_cached);
 
@@ -915,14 +913,12 @@ static struct extent_map *get_extent_map(struct btrfs_inode *inode,
 		*em_cached = NULL;
 	}
 
-	btrfs_lock_and_flush_ordered_range(inode, start, start + len - 1, &cached_state);
 	em = btrfs_get_extent(inode, folio, start, len);
 	if (!IS_ERR(em)) {
 		BUG_ON(*em_cached);
 		refcount_inc(&em->refs);
 		*em_cached = em;
 	}
-	unlock_extent(&inode->io_tree, start, start + len - 1, &cached_state);
 
 	return em;
 }
@@ -956,7 +952,7 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached,
 		return ret;
 	}
 
-	if (folio->index == last_byte >> folio_shift(folio)) {
+	if (folio_contains(folio, last_byte >> PAGE_SHIFT)) {
 		size_t zero_offset = offset_in_folio(folio, last_byte);
 
 		if (zero_offset) {
@@ -1079,11 +1075,18 @@ static int btrfs_do_readpage(struct folio *folio, struct extent_map **em_cached,
 
 int btrfs_read_folio(struct file *file, struct folio *folio)
 {
+	struct btrfs_inode *inode = folio_to_inode(folio);
+	const u64 start = folio_pos(folio);
+	const u64 end = start + folio_size(folio) - 1;
+	struct extent_state *cached_state = NULL;
 	struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ };
 	struct extent_map *em_cached = NULL;
 	int ret;
 
+	btrfs_lock_and_flush_ordered_range(inode, start, end, &cached_state);
 	ret = btrfs_do_readpage(folio, &em_cached, &bio_ctrl, NULL);
+	unlock_extent(&inode->io_tree, start, end, &cached_state);
+
 	free_extent_map(em_cached);
 
 	/*
@@ -2380,12 +2383,20 @@ void btrfs_readahead(struct readahead_control *rac)
 {
 	struct btrfs_bio_ctrl bio_ctrl = { .opf = REQ_OP_READ | REQ_RAHEAD };
 	struct folio *folio;
+	struct btrfs_inode *inode = BTRFS_I(rac->mapping->host);
+	const u64 start = readahead_pos(rac);
+	const u64 end = start + readahead_length(rac) - 1;
+	struct extent_state *cached_state = NULL;
 	struct extent_map *em_cached = NULL;
 	u64 prev_em_start = (u64)-1;
 
+	btrfs_lock_and_flush_ordered_range(inode, start, end, &cached_state);
+
 	while ((folio = readahead_folio(rac)) != NULL)
 		btrfs_do_readpage(folio, &em_cached, &bio_ctrl, &prev_em_start);
 
+	unlock_extent(&inode->io_tree, start, end, &cached_state);
+
 	if (em_cached)
 		free_extent_map(em_cached);
 	submit_one_bio(&bio_ctrl);
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 36f51c311bb1..ed3c0d6546c5 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1039,7 +1039,6 @@ int btrfs_write_check(struct kiocb *iocb, size_t count)
 	loff_t pos = iocb->ki_pos;
 	int ret;
 	loff_t oldsize;
-	loff_t start_pos;
 
 	/*
 	 * Quickly bail out on NOWAIT writes if we don't have the nodatacow or
@@ -1066,9 +1065,8 @@ int btrfs_write_check(struct kiocb *iocb, size_t count)
 		inode_inc_iversion(inode);
 	}
 
-	start_pos = round_down(pos, fs_info->sectorsize);
 	oldsize = i_size_read(inode);
-	if (start_pos > oldsize) {
+	if (pos > oldsize) {
 		/* Expand hole size to cover write data, preventing empty gap */
 		loff_t end_pos = round_up(pos + count, fs_info->sectorsize);
 
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 30eceaf829a7..4aca7475fd82 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -1229,6 +1229,18 @@ struct btrfs_ordered_extent *btrfs_split_ordered_extent(
 	 */
 	if (WARN_ON_ONCE(len >= ordered->num_bytes))
 		return ERR_PTR(-EINVAL);
+	/*
+	 * If our ordered extent had an error there's no point in continuing.
+	 * The error may have come from a transaction abort done either by this
+	 * task or some other concurrent task, and the transaction abort path
+	 * iterates over all existing ordered extents and sets the flag
+	 * BTRFS_ORDERED_IOERR on them.
+	 */
+	if (unlikely(flags & (1U << BTRFS_ORDERED_IOERR))) {
+		const int fs_error = BTRFS_FS_ERROR(fs_info);
+
+		return fs_error ? ERR_PTR(fs_error) : ERR_PTR(-EIO);
+	}
 	/* We cannot split partially completed ordered extents. */
 	if (ordered->bytes_left) {
 		ASSERT(!(flags & ~BTRFS_ORDERED_TYPE_FLAGS));
diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
index b90fabe302e6..f9d3766c809b 100644
--- a/fs/btrfs/qgroup.c
+++ b/fs/btrfs/qgroup.c
@@ -1880,11 +1880,7 @@ int btrfs_qgroup_cleanup_dropped_subvolume(struct btrfs_fs_info *fs_info, u64 su
 	 * Commit current transaction to make sure all the rfer/excl numbers
 	 * get updated.
 	 */
-	trans = btrfs_start_transaction(fs_info->quota_root, 0);
-	if (IS_ERR(trans))
-		return PTR_ERR(trans);
-
-	ret = btrfs_commit_transaction(trans);
+	ret = btrfs_commit_current_transaction(fs_info->quota_root);
 	if (ret < 0)
 		return ret;
 
@@ -1897,8 +1893,11 @@ int btrfs_qgroup_cleanup_dropped_subvolume(struct btrfs_fs_info *fs_info, u64 su
 	/*
 	 * It's squota and the subvolume still has numbers needed for future
 	 * accounting, in this case we can not delete it.  Just skip it.
+	 *
+	 * Or the qgroup is already removed by a qgroup rescan. For both cases we're
+	 * safe to ignore them.
 	 */
-	if (ret == -EBUSY)
+	if (ret == -EBUSY || ret == -ENOENT)
 		ret = 0;
 	return ret;
 }
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 15312013f2a3..aca83a98b75a 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -274,8 +274,10 @@ loop:
 	cur_trans = fs_info->running_transaction;
 	if (cur_trans) {
 		if (TRANS_ABORTED(cur_trans)) {
+			const int abort_error = cur_trans->aborted;
+
 			spin_unlock(&fs_info->trans_lock);
-			return cur_trans->aborted;
+			return abort_error;
 		}
 		if (btrfs_blocked_trans_types[cur_trans->state] & type) {
 			spin_unlock(&fs_info->trans_lock);
diff --git a/fs/dcache.c b/fs/dcache.c
index 9cc0d47da321..e3634916ffb9 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1700,7 +1700,7 @@ static struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name)
 	smp_store_release(&dentry->d_name.name, dname); /* ^^^ */
 
 	dentry->d_flags = 0;
-	lockref_init(&dentry->d_lockref, 1);
+	lockref_init(&dentry->d_lockref);
 	seqcount_spinlock_init(&dentry->d_seq, &dentry->d_lock);
 	dentry->d_inode = NULL;
 	dentry->d_parent = dentry;
@@ -2966,11 +2966,11 @@ static int __d_unalias(struct dentry *dentry, struct dentry *alias)
 		goto out_err;
 	m2 = &alias->d_parent->d_inode->i_rwsem;
 out_unalias:
-	if (alias->d_op->d_unalias_trylock &&
+	if (alias->d_op && alias->d_op->d_unalias_trylock &&
 	    !alias->d_op->d_unalias_trylock(alias))
 		goto out_err;
 	__d_move(alias, dentry, false);
-	if (alias->d_op->d_unalias_unlock)
+	if (alias->d_op && alias->d_op->d_unalias_unlock)
 		alias->d_op->d_unalias_unlock(alias);
 	ret = 0;
 out_err:
diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
index 29f8963bb523..d771e06db738 100644
--- a/fs/erofs/zdata.c
+++ b/fs/erofs/zdata.c
@@ -726,7 +726,7 @@ static int z_erofs_register_pcluster(struct z_erofs_frontend *fe)
 	if (IS_ERR(pcl))
 		return PTR_ERR(pcl);
 
-	lockref_init(&pcl->lockref, 1); /* one ref for this request */
+	lockref_init(&pcl->lockref); /* one ref for this request */
 	pcl->algorithmformat = map->m_algorithmformat;
 	pcl->length = 0;
 	pcl->partial = true;
diff --git a/fs/file_table.c b/fs/file_table.c
index f0291a66f9db..5c00dc38558d 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -194,6 +194,11 @@ static int init_file(struct file *f, int flags, const struct cred *cred)
 	 * refcount bumps we should reinitialize the reused file first.
 	 */
 	file_ref_init(&f->f_ref, 1);
+	/*
+	 * Disable permission and pre-content events for all files by default.
+	 * They may be enabled later by file_set_fsnotify_mode_from_watchers().
+	 */
+	file_set_fsnotify_mode(f, FMODE_NONOTIFY_PERM);
 	return 0;
 }
 
@@ -375,7 +380,13 @@ struct file *alloc_file_pseudo(struct inode *inode, struct vfsmount *mnt,
 	if (IS_ERR(file)) {
 		ihold(inode);
 		path_put(&path);
+		return file;
 	}
+	/*
+	 * Disable all fsnotify events for pseudo files by default.
+	 * They may be enabled by caller with file_set_fsnotify_mode().
+	 */
+	file_set_fsnotify_mode(file, FMODE_NONOTIFY);
 	return file;
 }
 EXPORT_SYMBOL(alloc_file_pseudo);
@@ -400,6 +411,11 @@ struct file *alloc_file_pseudo_noaccount(struct inode *inode,
 		return file;
 	}
 	file_init_path(file, &path, fops);
+	/*
+	 * Disable all fsnotify events for pseudo files by default.
+	 * They may be enabled by caller with file_set_fsnotify_mode().
+	 */
+	file_set_fsnotify_mode(file, FMODE_NONOTIFY);
 	return file;
 }
 EXPORT_SYMBOL_GPL(alloc_file_pseudo_noaccount);
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 8c4c1f871a88..65c07aa95718 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -1201,8 +1201,8 @@ int gfs2_glock_get(struct gfs2_sbd *sdp, u64 number,
 	if (glops->go_instantiate)
 		gl->gl_flags |= BIT(GLF_INSTANTIATE_NEEDED);
 	gl->gl_name = name;
+	lockref_init(&gl->gl_lockref);
 	lockdep_set_subclass(&gl->gl_lockref.lock, glops->go_subclass);
-	gl->gl_lockref.count = 1;
 	gl->gl_state = LM_ST_UNLOCKED;
 	gl->gl_target = LM_ST_UNLOCKED;
 	gl->gl_demote_state = LM_ST_EXCLUSIVE;
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index 04cadc02e5a6..0727f60ad028 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -51,7 +51,6 @@ static void gfs2_init_glock_once(void *foo)
 {
 	struct gfs2_glock *gl = foo;
 
-	spin_lock_init(&gl->gl_lockref.lock);
 	INIT_LIST_HEAD(&gl->gl_holders);
 	INIT_LIST_HEAD(&gl->gl_lru);
 	INIT_LIST_HEAD(&gl->gl_ail_list);
diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c
index 58bc5013ca49..2298e06797ac 100644
--- a/fs/gfs2/quota.c
+++ b/fs/gfs2/quota.c
@@ -236,7 +236,7 @@ static struct gfs2_quota_data *qd_alloc(unsigned hash, struct gfs2_sbd *sdp, str
 		return NULL;
 
 	qd->qd_sbd = sdp;
-	lockref_init(&qd->qd_lockref, 0);
+	lockref_init(&qd->qd_lockref);
 	qd->qd_id = qid;
 	qd->qd_slot = -1;
 	INIT_LIST_HEAD(&qd->qd_lru);
@@ -297,7 +297,6 @@ static int qd_get(struct gfs2_sbd *sdp, struct kqid qid,
 	spin_lock_bucket(hash);
 	*qdp = qd = gfs2_qd_search_bucket(hash, sdp, qid);
 	if (qd == NULL) {
-		new_qd->qd_lockref.count++;
 		*qdp = new_qd;
 		list_add(&new_qd->qd_list, &sdp->sd_quota_list);
 		hlist_bl_add_head_rcu(&new_qd->qd_hlist, &qd_hash_table[hash]);
@@ -1450,6 +1449,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp)
 			if (qd == NULL)
 				goto fail_brelse;
 
+			qd->qd_lockref.count = 0;
 			set_bit(QDF_CHANGE, &qd->qd_flags);
 			qd->qd_change = qc_change;
 			qd->qd_slot = slot;
diff --git a/fs/namespace.c b/fs/namespace.c
index a3ed3f2980cb..8f1000f9f3df 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -5087,30 +5087,29 @@ static int statmount_mnt_opts(struct kstatmount *s, struct seq_file *seq)
 {
 	struct vfsmount *mnt = s->mnt;
 	struct super_block *sb = mnt->mnt_sb;
+	size_t start = seq->count;
 	int err;
 
-	if (sb->s_op->show_options) {
-		size_t start = seq->count;
-
-		err = security_sb_show_options(seq, sb);
-		if (err)
-			return err;
+	err = security_sb_show_options(seq, sb);
+	if (err)
+		return err;
 
+	if (sb->s_op->show_options) {
 		err = sb->s_op->show_options(seq, mnt->mnt_root);
 		if (err)
 			return err;
+	}
 
-		if (unlikely(seq_has_overflowed(seq)))
-			return -EAGAIN;
+	if (unlikely(seq_has_overflowed(seq)))
+		return -EAGAIN;
 
-		if (seq->count == start)
-			return 0;
+	if (seq->count == start)
+		return 0;
 
-		/* skip leading comma */
-		memmove(seq->buf + start, seq->buf + start + 1,
-			seq->count - start - 1);
-		seq->count--;
-	}
+	/* skip leading comma */
+	memmove(seq->buf + start, seq->buf + start + 1,
+		seq->count - start - 1);
+	seq->count--;
 
 	return 0;
 }
@@ -5191,39 +5190,45 @@ static int statmount_string(struct kstatmount *s, u64 flag)
 	size_t kbufsize;
 	struct seq_file *seq = &s->seq;
 	struct statmount *sm = &s->sm;
-	u32 start = seq->count;
+	u32 start, *offp;
+
+	/* Reserve an empty string at the beginning for any unset offsets */
+	if (!seq->count)
+		seq_putc(seq, 0);
+
+	start = seq->count;
 
 	switch (flag) {
 	case STATMOUNT_FS_TYPE:
-		sm->fs_type = start;
+		offp = &sm->fs_type;
 		ret = statmount_fs_type(s, seq);
 		break;
 	case STATMOUNT_MNT_ROOT:
-		sm->mnt_root = start;
+		offp = &sm->mnt_root;
 		ret = statmount_mnt_root(s, seq);
 		break;
 	case STATMOUNT_MNT_POINT:
-		sm->mnt_point = start;
+		offp = &sm->mnt_point;
 		ret = statmount_mnt_point(s, seq);
 		break;
 	case STATMOUNT_MNT_OPTS:
-		sm->mnt_opts = start;
+		offp = &sm->mnt_opts;
 		ret = statmount_mnt_opts(s, seq);
 		break;
 	case STATMOUNT_OPT_ARRAY:
-		sm->opt_array = start;
+		offp = &sm->opt_array;
 		ret = statmount_opt_array(s, seq);
 		break;
 	case STATMOUNT_OPT_SEC_ARRAY:
-		sm->opt_sec_array = start;
+		offp = &sm->opt_sec_array;
 		ret = statmount_opt_sec_array(s, seq);
 		break;
 	case STATMOUNT_FS_SUBTYPE:
-		sm->fs_subtype = start;
+		offp = &sm->fs_subtype;
 		statmount_fs_subtype(s, seq);
 		break;
 	case STATMOUNT_SB_SOURCE:
-		sm->sb_source = start;
+		offp = &sm->sb_source;
 		ret = statmount_sb_source(s, seq);
 		break;
 	default:
@@ -5251,6 +5256,7 @@ static int statmount_string(struct kstatmount *s, u64 flag)
 
 	seq->buf[seq->count++] = '\0';
 	sm->mask |= flag;
+	*offp = start;
 	return 0;
 }
 
diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c
index 0e552d873eaa..fb9b1656a287 100644
--- a/fs/nfsd/filecache.c
+++ b/fs/nfsd/filecache.c
@@ -446,11 +446,20 @@ nfsd_file_dispose_list_delayed(struct list_head *dispose)
 						struct nfsd_file, nf_gc);
 		struct nfsd_net *nn = net_generic(nf->nf_net, nfsd_net_id);
 		struct nfsd_fcache_disposal *l = nn->fcache_disposal;
+		struct svc_serv *serv;
 
 		spin_lock(&l->lock);
 		list_move_tail(&nf->nf_gc, &l->freeme);
 		spin_unlock(&l->lock);
-		svc_wake_up(nn->nfsd_serv);
+
+		/*
+		 * The filecache laundrette is shut down after the
+		 * nn->nfsd_serv pointer is cleared, but before the
+		 * svc_serv is freed.
+		 */
+		serv = nn->nfsd_serv;
+		if (serv)
+			svc_wake_up(serv);
 	}
 }
 
diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c
index 4e3be7201b1c..5fb202acb0fd 100644
--- a/fs/nfsd/nfs2acl.c
+++ b/fs/nfsd/nfs2acl.c
@@ -84,6 +84,8 @@ out:
 fail:
 	posix_acl_release(resp->acl_access);
 	posix_acl_release(resp->acl_default);
+	resp->acl_access = NULL;
+	resp->acl_default = NULL;
 	goto out;
 }
 
diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c
index 5e34e98db969..7b5433bd3019 100644
--- a/fs/nfsd/nfs3acl.c
+++ b/fs/nfsd/nfs3acl.c
@@ -76,6 +76,8 @@ out:
 fail:
 	posix_acl_release(resp->acl_access);
 	posix_acl_release(resp->acl_default);
+	resp->acl_access = NULL;
+	resp->acl_default = NULL;
 	goto out;
 }
 
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 50e468bdb8d4..484077200c5d 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -679,7 +679,7 @@ static int nfs4_xdr_dec_cb_getattr(struct rpc_rqst *rqstp,
 		return status;
 
 	status = decode_cb_op_status(xdr, OP_CB_GETATTR, &cb->cb_status);
-	if (unlikely(status || cb->cb_seq_status))
+	if (unlikely(status || cb->cb_status))
 		return status;
 	if (xdr_stream_decode_uint32_array(xdr, bitmap, 3) < 0)
 		return -NFSERR_BAD_XDR;
@@ -1583,8 +1583,11 @@ nfsd4_run_cb_work(struct work_struct *work)
 		nfsd4_process_cb_update(cb);
 
 	clnt = clp->cl_cb_client;
-	if (!clnt) {
-		/* Callback channel broken, or client killed; give up: */
+	if (!clnt || clp->cl_state == NFSD4_COURTESY) {
+		/*
+		 * Callback channel broken, client killed or
+		 * nfs4_client in courtesy state; give up.
+		 */
 		nfsd41_destroy_cb(cb);
 		return;
 	}
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index b7a0cfd05401..153eeea2c7c9 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -4459,10 +4459,11 @@ nfsd4_sequence(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 			}
 		} while (slot && --cnt > 0);
 	}
+
+out:
 	seq->maxslots = max(session->se_target_maxslots, seq->maxslots);
 	seq->target_maxslots = session->se_target_maxslots;
 
-out:
 	switch (clp->cl_cb_state) {
 	case NFSD4_CB_DOWN:
 		seq->status_flags = SEQ4_STATUS_CB_PATH_DOWN;
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index 32019751a41e..aef474f1b84b 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -380,8 +380,9 @@ __fh_verify(struct svc_rqst *rqstp,
 	error = check_nfsd_access(exp, rqstp, may_bypass_gss);
 	if (error)
 		goto out;
-
-	svc_xprt_set_valid(rqstp->rq_xprt);
+	/* During LOCALIO call to fh_verify will be called with a NULL rqstp */
+	if (rqstp)
+		svc_xprt_set_valid(rqstp->rq_xprt);
 
 	/* Finally, check access permissions. */
 	error = nfsd_permission(cred, exp, dentry, access);
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index 8ee495a58d0a..fae1b6d397ea 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -648,7 +648,7 @@ EXPORT_SYMBOL_GPL(fsnotify);
  * Later, fsnotify permission hooks do not check if there are permission event
  * watches, but that there were permission event watches at open time.
  */
-void file_set_fsnotify_mode(struct file *file)
+void file_set_fsnotify_mode_from_watchers(struct file *file)
 {
 	struct dentry *dentry = file->f_path.dentry, *parent;
 	struct super_block *sb = dentry->d_sb;
@@ -665,7 +665,7 @@ void file_set_fsnotify_mode(struct file *file)
 	 */
 	if (likely(!fsnotify_sb_has_priority_watchers(sb,
 						FSNOTIFY_PRIO_CONTENT))) {
-		file->f_mode |= FMODE_NONOTIFY_PERM;
+		file_set_fsnotify_mode(file, FMODE_NONOTIFY_PERM);
 		return;
 	}
 
@@ -676,7 +676,7 @@ void file_set_fsnotify_mode(struct file *file)
 	if ((!d_is_dir(dentry) && !d_is_reg(dentry)) ||
 	    likely(!fsnotify_sb_has_priority_watchers(sb,
 						FSNOTIFY_PRIO_PRE_CONTENT))) {
-		file->f_mode |= FMODE_NONOTIFY | FMODE_NONOTIFY_PERM;
+		file_set_fsnotify_mode(file, FMODE_NONOTIFY | FMODE_NONOTIFY_PERM);
 		return;
 	}
 
@@ -686,19 +686,25 @@ void file_set_fsnotify_mode(struct file *file)
 	 */
 	mnt_mask = READ_ONCE(real_mount(file->f_path.mnt)->mnt_fsnotify_mask);
 	if (unlikely(fsnotify_object_watched(d_inode(dentry), mnt_mask,
-				     FSNOTIFY_PRE_CONTENT_EVENTS)))
+				     FSNOTIFY_PRE_CONTENT_EVENTS))) {
+		/* Enable pre-content events */
+		file_set_fsnotify_mode(file, 0);
 		return;
+	}
 
 	/* Is parent watching for pre-content events on this file? */
 	if (dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED) {
 		parent = dget_parent(dentry);
 		p_mask = fsnotify_inode_watches_children(d_inode(parent));
 		dput(parent);
-		if (p_mask & FSNOTIFY_PRE_CONTENT_EVENTS)
+		if (p_mask & FSNOTIFY_PRE_CONTENT_EVENTS) {
+			/* Enable pre-content events */
+			file_set_fsnotify_mode(file, 0);
 			return;
+		}
 	}
 	/* Nobody watching for pre-content events from this file */
-	file->f_mode |= FMODE_NONOTIFY | FMODE_NONOTIFY_PERM;
+	file_set_fsnotify_mode(file, FMODE_NONOTIFY | FMODE_NONOTIFY_PERM);
 }
 #endif
 
diff --git a/fs/open.c b/fs/open.c
index 932e5a6de63b..1be20de9f283 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -905,7 +905,8 @@ static int do_dentry_open(struct file *f,
 	f->f_sb_err = file_sample_sb_err(f);
 
 	if (unlikely(f->f_flags & O_PATH)) {
-		f->f_mode = FMODE_PATH | FMODE_OPENED | FMODE_NONOTIFY;
+		f->f_mode = FMODE_PATH | FMODE_OPENED;
+		file_set_fsnotify_mode(f, FMODE_NONOTIFY);
 		f->f_op = &empty_fops;
 		return 0;
 	}
@@ -935,10 +936,10 @@ static int do_dentry_open(struct file *f,
 
 	/*
 	 * Set FMODE_NONOTIFY_* bits according to existing permission watches.
-	 * If FMODE_NONOTIFY was already set for an fanotify fd, this doesn't
-	 * change anything.
+	 * If FMODE_NONOTIFY mode was already set for an fanotify fd or for a
+	 * pseudo file, this call will not change the mode.
 	 */
-	file_set_fsnotify_mode(f);
+	file_set_fsnotify_mode_from_watchers(f);
 	error = fsnotify_open_perm(f);
 	if (error)
 		goto cleanup_all;
@@ -1122,7 +1123,7 @@ struct file *dentry_open_nonotify(const struct path *path, int flags,
 	if (!IS_ERR(f)) {
 		int error;
 
-		f->f_mode |= FMODE_NONOTIFY;
+		file_set_fsnotify_mode(f, FMODE_NONOTIFY);
 		error = vfs_open(path, f);
 		if (error) {
 			fput(f);
diff --git a/fs/pidfs.c b/fs/pidfs.c
index 049352f973de..63f9699ebac3 100644
--- a/fs/pidfs.c
+++ b/fs/pidfs.c
@@ -287,7 +287,6 @@ static bool pidfs_ioctl_valid(unsigned int cmd)
 	switch (cmd) {
 	case FS_IOC_GETVERSION:
 	case PIDFD_GET_CGROUP_NAMESPACE:
-	case PIDFD_GET_INFO:
 	case PIDFD_GET_IPC_NAMESPACE:
 	case PIDFD_GET_MNT_NAMESPACE:
 	case PIDFD_GET_NET_NAMESPACE:
@@ -300,6 +299,17 @@ static bool pidfs_ioctl_valid(unsigned int cmd)
 		return true;
 	}
 
+	/* Extensible ioctls require some more careful checks. */
+	switch (_IOC_NR(cmd)) {
+	case _IOC_NR(PIDFD_GET_INFO):
+		/*
+		 * Try to prevent performing a pidfd ioctl when someone
+		 * erronously mistook the file descriptor for a pidfd.
+		 * This is not perfect but will catch most cases.
+		 */
+		return (_IOC_TYPE(cmd) == _IOC_TYPE(PIDFD_GET_INFO));
+	}
+
 	return false;
 }
 
diff --git a/fs/pipe.c b/fs/pipe.c
index 94b59045ab44..ce1af7592780 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -960,6 +960,12 @@ int create_pipe_files(struct file **res, int flags)
 	res[1] = f;
 	stream_open(inode, res[0]);
 	stream_open(inode, res[1]);
+	/*
+	 * Disable permission and pre-content events, but enable legacy
+	 * inotify events for legacy users.
+	 */
+	file_set_fsnotify_mode(res[0], FMODE_NONOTIFY_PERM);
+	file_set_fsnotify_mode(res[1], FMODE_NONOTIFY_PERM);
 	return 0;
 }
 
diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h
index a68434ad744a..4bdd6a43e521 100644
--- a/fs/smb/client/cifsglob.h
+++ b/fs/smb/client/cifsglob.h
@@ -357,7 +357,7 @@ struct smb_version_operations {
 	int (*handle_cancelled_mid)(struct mid_q_entry *, struct TCP_Server_Info *);
 	void (*downgrade_oplock)(struct TCP_Server_Info *server,
 				 struct cifsInodeInfo *cinode, __u32 oplock,
-				 unsigned int epoch, bool *purge_cache);
+				 __u16 epoch, bool *purge_cache);
 	/* process transaction2 response */
 	bool (*check_trans2)(struct mid_q_entry *, struct TCP_Server_Info *,
 			     char *, int);
@@ -552,12 +552,12 @@ struct smb_version_operations {
 	/* if we can do cache read operations */
 	bool (*is_read_op)(__u32);
 	/* set oplock level for the inode */
-	void (*set_oplock_level)(struct cifsInodeInfo *, __u32, unsigned int,
-				 bool *);
+	void (*set_oplock_level)(struct cifsInodeInfo *cinode, __u32 oplock, __u16 epoch,
+				 bool *purge_cache);
 	/* create lease context buffer for CREATE request */
 	char * (*create_lease_buf)(u8 *lease_key, u8 oplock);
 	/* parse lease context buffer and return oplock/epoch info */
-	__u8 (*parse_lease_buf)(void *buf, unsigned int *epoch, char *lkey);
+	__u8 (*parse_lease_buf)(void *buf, __u16 *epoch, char *lkey);
 	ssize_t (*copychunk_range)(const unsigned int,
 			struct cifsFileInfo *src_file,
 			struct cifsFileInfo *target_file,
@@ -1447,7 +1447,7 @@ struct cifs_fid {
 	__u8 create_guid[16];
 	__u32 access;
 	struct cifs_pending_open *pending_open;
-	unsigned int epoch;
+	__u16 epoch;
 #ifdef CONFIG_CIFS_DEBUG2
 	__u64 mid;
 #endif /* CIFS_DEBUG2 */
@@ -1480,7 +1480,7 @@ struct cifsFileInfo {
 	bool oplock_break_cancelled:1;
 	bool status_file_deleted:1; /* file has been deleted */
 	bool offload:1; /* offload final part of _put to a wq */
-	unsigned int oplock_epoch; /* epoch from the lease break */
+	__u16 oplock_epoch; /* epoch from the lease break */
 	__u32 oplock_level; /* oplock/lease level from the lease break */
 	int count;
 	spinlock_t file_info_lock; /* protects four flag/count fields above */
@@ -1508,7 +1508,6 @@ struct cifs_io_parms {
 struct cifs_io_request {
 	struct netfs_io_request		rreq;
 	struct cifsFileInfo		*cfile;
-	struct TCP_Server_Info		*server;
 	pid_t				pid;
 };
 
@@ -1577,7 +1576,7 @@ struct cifsInodeInfo {
 	spinlock_t	open_file_lock;	/* protects openFileList */
 	__u32 cifsAttrs; /* e.g. DOS archive bit, sparse, compressed, system */
 	unsigned int oplock;		/* oplock/lease level we have */
-	unsigned int epoch;		/* used to track lease state changes */
+	__u16 epoch;		/* used to track lease state changes */
 #define CIFS_INODE_PENDING_OPLOCK_BREAK   (0) /* oplock break in progress */
 #define CIFS_INODE_PENDING_WRITERS	  (1) /* Writes in progress */
 #define CIFS_INODE_FLAG_UNUSED		  (2) /* Unused flag */
diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c
index dad521336b5e..f65a8a90ba27 100644
--- a/fs/smb/client/dfs.c
+++ b/fs/smb/client/dfs.c
@@ -150,25 +150,27 @@ again:
 			if (rc)
 				continue;
 
-			if (tgt.flags & DFSREF_STORAGE_SERVER) {
-				rc = cifs_mount_get_tcon(mnt_ctx);
-				if (!rc)
-					rc = cifs_is_path_remote(mnt_ctx);
+			rc = cifs_mount_get_tcon(mnt_ctx);
+			if (rc) {
+				if (tgt.server_type == DFS_TYPE_LINK &&
+				    DFS_INTERLINK(tgt.flags))
+					rc = -EREMOTE;
+			} else {
+				rc = cifs_is_path_remote(mnt_ctx);
 				if (!rc) {
 					ref_walk_set_tgt_hint(rw);
 					break;
 				}
-				if (rc != -EREMOTE)
-					continue;
 			}
-
-			rc = ref_walk_advance(rw);
-			if (!rc) {
-				rc = setup_dfs_ref(&tgt, rw);
-				if (rc)
-					break;
-				ref_walk_mark_end(rw);
-				goto again;
+			if (rc == -EREMOTE) {
+				rc = ref_walk_advance(rw);
+				if (!rc) {
+					rc = setup_dfs_ref(&tgt, rw);
+					if (rc)
+						break;
+					ref_walk_mark_end(rw);
+					goto again;
+				}
 			}
 		}
 	} while (rc && ref_walk_descend(rw));
diff --git a/fs/smb/client/dfs.h b/fs/smb/client/dfs.h
index ed4cd7cf1ec6..e60f0a24a8a1 100644
--- a/fs/smb/client/dfs.h
+++ b/fs/smb/client/dfs.h
@@ -188,4 +188,11 @@ static inline void dfs_put_root_smb_sessions(struct list_head *head)
 	}
 }
 
+static inline const char *dfs_ses_refpath(struct cifs_ses *ses)
+{
+	const char *path = ses->server->leaf_fullpath;
+
+	return path ? path + 1 : ERR_PTR(-ENOENT);
+}
+
 #endif /* _CIFS_DFS_H */
diff --git a/fs/smb/client/dfs_cache.c b/fs/smb/client/dfs_cache.c
index 5022bb1f122a..4dada26d56b5 100644
--- a/fs/smb/client/dfs_cache.c
+++ b/fs/smb/client/dfs_cache.c
@@ -1136,33 +1136,19 @@ static bool is_ses_good(struct cifs_ses *ses)
 	return ret;
 }
 
-static char *get_ses_refpath(struct cifs_ses *ses)
-{
-	struct TCP_Server_Info *server = ses->server;
-	char *path = ERR_PTR(-ENOENT);
-
-	if (server->leaf_fullpath) {
-		path = kstrdup(server->leaf_fullpath + 1, GFP_KERNEL);
-		if (!path)
-			path = ERR_PTR(-ENOMEM);
-	}
-	return path;
-}
-
 /* Refresh dfs referral of @ses */
 static void refresh_ses_referral(struct cifs_ses *ses)
 {
 	struct cache_entry *ce;
 	unsigned int xid;
-	char *path;
+	const char *path;
 	int rc = 0;
 
 	xid = get_xid();
 
-	path = get_ses_refpath(ses);
+	path = dfs_ses_refpath(ses);
 	if (IS_ERR(path)) {
 		rc = PTR_ERR(path);
-		path = NULL;
 		goto out;
 	}
 
@@ -1181,7 +1167,6 @@ static void refresh_ses_referral(struct cifs_ses *ses)
 
 out:
 	free_xid(xid);
-	kfree(path);
 }
 
 static int __refresh_tcon_referral(struct cifs_tcon *tcon,
@@ -1231,19 +1216,18 @@ static void refresh_tcon_referral(struct cifs_tcon *tcon, bool force_refresh)
 	struct dfs_info3_param *refs = NULL;
 	struct cache_entry *ce;
 	struct cifs_ses *ses;
-	unsigned int xid;
 	bool needs_refresh;
-	char *path;
+	const char *path;
+	unsigned int xid;
 	int numrefs = 0;
 	int rc = 0;
 
 	xid = get_xid();
 	ses = tcon->ses;
 
-	path = get_ses_refpath(ses);
+	path = dfs_ses_refpath(ses);
 	if (IS_ERR(path)) {
 		rc = PTR_ERR(path);
-		path = NULL;
 		goto out;
 	}
 
@@ -1271,7 +1255,6 @@ static void refresh_tcon_referral(struct cifs_tcon *tcon, bool force_refresh)
 
 out:
 	free_xid(xid);
-	kfree(path);
 	free_dfs_info_array(refs, numrefs);
 }
 
diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c
index 79de2f2f9c41..8582cf61242c 100644
--- a/fs/smb/client/file.c
+++ b/fs/smb/client/file.c
@@ -147,7 +147,7 @@ static int cifs_prepare_read(struct netfs_io_subrequest *subreq)
 	struct netfs_io_request *rreq = subreq->rreq;
 	struct cifs_io_subrequest *rdata = container_of(subreq, struct cifs_io_subrequest, subreq);
 	struct cifs_io_request *req = container_of(subreq->rreq, struct cifs_io_request, rreq);
-	struct TCP_Server_Info *server = req->server;
+	struct TCP_Server_Info *server;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode->i_sb);
 	size_t size;
 	int rc = 0;
@@ -156,6 +156,8 @@ static int cifs_prepare_read(struct netfs_io_subrequest *subreq)
 		rdata->xid = get_xid();
 		rdata->have_xid = true;
 	}
+
+	server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses);
 	rdata->server = server;
 
 	if (cifs_sb->ctx->rsize == 0)
@@ -198,7 +200,7 @@ static void cifs_issue_read(struct netfs_io_subrequest *subreq)
 	struct netfs_io_request *rreq = subreq->rreq;
 	struct cifs_io_subrequest *rdata = container_of(subreq, struct cifs_io_subrequest, subreq);
 	struct cifs_io_request *req = container_of(subreq->rreq, struct cifs_io_request, rreq);
-	struct TCP_Server_Info *server = req->server;
+	struct TCP_Server_Info *server = rdata->server;
 	int rc = 0;
 
 	cifs_dbg(FYI, "%s: op=%08x[%x] mapping=%p len=%zu/%zu\n",
@@ -266,7 +268,6 @@ static int cifs_init_request(struct netfs_io_request *rreq, struct file *file)
 		open_file = file->private_data;
 		rreq->netfs_priv = file->private_data;
 		req->cfile = cifsFileInfo_get(open_file);
-		req->server = cifs_pick_channel(tlink_tcon(req->cfile->tlink)->ses);
 		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD)
 			req->pid = req->cfile->pid;
 	} else if (rreq->origin != NETFS_WRITEBACK) {
diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c
index 9756b876a75e..d6e2fb669c40 100644
--- a/fs/smb/client/smb1ops.c
+++ b/fs/smb/client/smb1ops.c
@@ -377,7 +377,7 @@ coalesce_t2(char *second_buf, struct smb_hdr *target_hdr)
 static void
 cifs_downgrade_oplock(struct TCP_Server_Info *server,
 		      struct cifsInodeInfo *cinode, __u32 oplock,
-		      unsigned int epoch, bool *purge_cache)
+		      __u16 epoch, bool *purge_cache)
 {
 	cifs_set_oplock_level(cinode, oplock);
 }
diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c
index 77309217dab4..ec36bed54b0b 100644
--- a/fs/smb/client/smb2ops.c
+++ b/fs/smb/client/smb2ops.c
@@ -3904,22 +3904,22 @@ static long smb3_fallocate(struct file *file, struct cifs_tcon *tcon, int mode,
 static void
 smb2_downgrade_oplock(struct TCP_Server_Info *server,
 		      struct cifsInodeInfo *cinode, __u32 oplock,
-		      unsigned int epoch, bool *purge_cache)
+		      __u16 epoch, bool *purge_cache)
 {
 	server->ops->set_oplock_level(cinode, oplock, 0, NULL);
 }
 
 static void
 smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
-		       unsigned int epoch, bool *purge_cache);
+		       __u16 epoch, bool *purge_cache);
 
 static void
 smb3_downgrade_oplock(struct TCP_Server_Info *server,
 		       struct cifsInodeInfo *cinode, __u32 oplock,
-		       unsigned int epoch, bool *purge_cache)
+		       __u16 epoch, bool *purge_cache)
 {
 	unsigned int old_state = cinode->oplock;
-	unsigned int old_epoch = cinode->epoch;
+	__u16 old_epoch = cinode->epoch;
 	unsigned int new_state;
 
 	if (epoch > old_epoch) {
@@ -3939,7 +3939,7 @@ smb3_downgrade_oplock(struct TCP_Server_Info *server,
 
 static void
 smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
-		      unsigned int epoch, bool *purge_cache)
+		      __u16 epoch, bool *purge_cache)
 {
 	oplock &= 0xFF;
 	cinode->lease_granted = false;
@@ -3963,7 +3963,7 @@ smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
 
 static void
 smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
-		       unsigned int epoch, bool *purge_cache)
+		       __u16 epoch, bool *purge_cache)
 {
 	char message[5] = {0};
 	unsigned int new_oplock = 0;
@@ -4000,7 +4000,7 @@ smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
 
 static void
 smb3_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock,
-		      unsigned int epoch, bool *purge_cache)
+		      __u16 epoch, bool *purge_cache)
 {
 	unsigned int old_oplock = cinode->oplock;
 
@@ -4114,7 +4114,7 @@ smb3_create_lease_buf(u8 *lease_key, u8 oplock)
 }
 
 static __u8
-smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key)
+smb2_parse_lease_buf(void *buf, __u16 *epoch, char *lease_key)
 {
 	struct create_lease *lc = (struct create_lease *)buf;
 
@@ -4125,7 +4125,7 @@ smb2_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key)
 }
 
 static __u8
-smb3_parse_lease_buf(void *buf, unsigned int *epoch, char *lease_key)
+smb3_parse_lease_buf(void *buf, __u16 *epoch, char *lease_key)
 {
 	struct create_lease_v2 *lc = (struct create_lease_v2 *)buf;
 
diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c
index 40ad9e79437a..ed7812247ebc 100644
--- a/fs/smb/client/smb2pdu.c
+++ b/fs/smb/client/smb2pdu.c
@@ -2169,7 +2169,7 @@ tcon_exit:
 
 tcon_error_exit:
 	if (rsp && rsp->hdr.Status == STATUS_BAD_NETWORK_NAME)
-		cifs_tcon_dbg(VFS, "BAD_NETWORK_NAME: %s\n", tree);
+		cifs_dbg(VFS | ONCE, "BAD_NETWORK_NAME: %s\n", tree);
 	goto tcon_exit;
 }
 
@@ -2329,7 +2329,7 @@ parse_posix_ctxt(struct create_context *cc, struct smb2_file_all_info *info,
 
 int smb2_parse_contexts(struct TCP_Server_Info *server,
 			struct kvec *rsp_iov,
-			unsigned int *epoch,
+			__u16 *epoch,
 			char *lease_key, __u8 *oplock,
 			struct smb2_file_all_info *buf,
 			struct create_posix_rsp *posix)
diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h
index 2336dfb23f36..4662c7e2d259 100644
--- a/fs/smb/client/smb2proto.h
+++ b/fs/smb/client/smb2proto.h
@@ -283,7 +283,7 @@ extern enum securityEnum smb2_select_sectype(struct TCP_Server_Info *,
 					enum securityEnum);
 int smb2_parse_contexts(struct TCP_Server_Info *server,
 			struct kvec *rsp_iov,
-			unsigned int *epoch,
+			__u16 *epoch,
 			char *lease_key, __u8 *oplock,
 			struct smb2_file_all_info *buf,
 			struct create_posix_rsp *posix);
diff --git a/fs/stat.c b/fs/stat.c
index 2c0e111a098a..f13308bfdc98 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -281,6 +281,8 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat,
 			  u32 request_mask)
 {
 	int error = vfs_getattr(path, stat, request_mask, flags);
+	if (error)
+		return error;
 
 	if (request_mask & STATX_MNT_ID_UNIQUE) {
 		stat->mnt_id = real_mount(path->mnt)->mnt_id_unique;
@@ -302,7 +304,7 @@ static int vfs_statx_path(struct path *path, int flags, struct kstat *stat,
 	if (S_ISBLK(stat->mode))
 		bdev_statx(path, stat, request_mask);
 
-	return error;
+	return 0;
 }
 
 static int vfs_statx_fd(int fd, int flags, struct kstat *stat,
diff --git a/fs/vboxsf/super.c b/fs/vboxsf/super.c
index e95b8a48d8a0..1d94bb784108 100644
--- a/fs/vboxsf/super.c
+++ b/fs/vboxsf/super.c
@@ -21,7 +21,8 @@
 
 #define VBOXSF_SUPER_MAGIC 0x786f4256 /* 'VBox' little endian */
 
-static const unsigned char VBSF_MOUNT_SIGNATURE[4] = "\000\377\376\375";
+static const unsigned char VBSF_MOUNT_SIGNATURE[4] = { '\000', '\377', '\376',
+						       '\375' };
 
 static int follow_symlinks;
 module_param(follow_symlinks, int, 0444);
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 40ad22fb808b..0ef19f1469ec 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3563,12 +3563,12 @@ xfs_bmap_btalloc_at_eof(
 	int			error;
 
 	/*
-	 * If there are already extents in the file, try an exact EOF block
-	 * allocation to extend the file as a contiguous extent. If that fails,
-	 * or it's the first allocation in a file, just try for a stripe aligned
-	 * allocation.
+	 * If there are already extents in the file, and xfs_bmap_adjacent() has
+	 * given a better blkno, try an exact EOF block allocation to extend the
+	 * file as a contiguous extent. If that fails, or it's the first
+	 * allocation in a file, just try for a stripe aligned allocation.
 	 */
-	if (ap->offset) {
+	if (ap->eof) {
 		xfs_extlen_t	nextminlen = 0;
 
 		/*
@@ -3736,7 +3736,8 @@ xfs_bmap_btalloc_best_length(
 	int			error;
 
 	ap->blkno = XFS_INO_TO_FSB(args->mp, ap->ip->i_ino);
-	xfs_bmap_adjacent(ap);
+	if (!xfs_bmap_adjacent(ap))
+		ap->eof = false;
 
 	/*
 	 * Search for an allocation group with a single extent large enough for
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index d1d4a0a22e13..15bb790359f8 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -41,8 +41,7 @@ struct kmem_cache *xfs_buf_cache;
  *
  * xfs_buf_rele:
  *	b_lock
- *	  pag_buf_lock
- *	    lru_lock
+ *	  lru_lock
  *
  * xfs_buftarg_drain_rele
  *	lru_lock
@@ -220,23 +219,25 @@ _xfs_buf_alloc(
 	 */
 	flags &= ~(XBF_UNMAPPED | XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD);
 
-	spin_lock_init(&bp->b_lock);
+	/*
+	 * A new buffer is held and locked by the owner.  This ensures that the
+	 * buffer is owned by the caller and racing RCU lookups right after
+	 * inserting into the hash table are safe (and will have to wait for
+	 * the unlock to do anything non-trivial).
+	 */
 	bp->b_hold = 1;
+	sema_init(&bp->b_sema, 0); /* held, no waiters */
+
+	spin_lock_init(&bp->b_lock);
 	atomic_set(&bp->b_lru_ref, 1);
 	init_completion(&bp->b_iowait);
 	INIT_LIST_HEAD(&bp->b_lru);
 	INIT_LIST_HEAD(&bp->b_list);
 	INIT_LIST_HEAD(&bp->b_li_list);
-	sema_init(&bp->b_sema, 0); /* held, no waiters */
 	bp->b_target = target;
 	bp->b_mount = target->bt_mount;
 	bp->b_flags = flags;
 
-	/*
-	 * Set length and io_length to the same value initially.
-	 * I/O routines should use io_length, which will be the same in
-	 * most cases but may be reset (e.g. XFS recovery).
-	 */
 	error = xfs_buf_get_maps(bp, nmaps);
 	if (error)  {
 		kmem_cache_free(xfs_buf_cache, bp);
@@ -502,7 +503,6 @@ int
 xfs_buf_cache_init(
 	struct xfs_buf_cache	*bch)
 {
-	spin_lock_init(&bch->bc_lock);
 	return rhashtable_init(&bch->bc_hash, &xfs_buf_hash_params);
 }
 
@@ -652,17 +652,20 @@ xfs_buf_find_insert(
 	if (error)
 		goto out_free_buf;
 
-	spin_lock(&bch->bc_lock);
+	/* The new buffer keeps the perag reference until it is freed. */
+	new_bp->b_pag = pag;
+
+	rcu_read_lock();
 	bp = rhashtable_lookup_get_insert_fast(&bch->bc_hash,
 			&new_bp->b_rhash_head, xfs_buf_hash_params);
 	if (IS_ERR(bp)) {
+		rcu_read_unlock();
 		error = PTR_ERR(bp);
-		spin_unlock(&bch->bc_lock);
 		goto out_free_buf;
 	}
 	if (bp && xfs_buf_try_hold(bp)) {
 		/* found an existing buffer */
-		spin_unlock(&bch->bc_lock);
+		rcu_read_unlock();
 		error = xfs_buf_find_lock(bp, flags);
 		if (error)
 			xfs_buf_rele(bp);
@@ -670,10 +673,8 @@ xfs_buf_find_insert(
 			*bpp = bp;
 		goto out_free_buf;
 	}
+	rcu_read_unlock();
 
-	/* The new buffer keeps the perag reference until it is freed. */
-	new_bp->b_pag = pag;
-	spin_unlock(&bch->bc_lock);
 	*bpp = new_bp;
 	return 0;
 
@@ -1090,7 +1091,6 @@ xfs_buf_rele_cached(
 	}
 
 	/* we are asked to drop the last reference */
-	spin_lock(&bch->bc_lock);
 	__xfs_buf_ioacct_dec(bp);
 	if (!(bp->b_flags & XBF_STALE) && atomic_read(&bp->b_lru_ref)) {
 		/*
@@ -1102,7 +1102,6 @@ xfs_buf_rele_cached(
 			bp->b_state &= ~XFS_BSTATE_DISPOSE;
 		else
 			bp->b_hold--;
-		spin_unlock(&bch->bc_lock);
 	} else {
 		bp->b_hold--;
 		/*
@@ -1120,7 +1119,6 @@ xfs_buf_rele_cached(
 		ASSERT(!(bp->b_flags & _XBF_DELWRI_Q));
 		rhashtable_remove_fast(&bch->bc_hash, &bp->b_rhash_head,
 				xfs_buf_hash_params);
-		spin_unlock(&bch->bc_lock);
 		if (pag)
 			xfs_perag_put(pag);
 		freebuf = true;
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 7e73663c5d4a..3b4ed42e11c0 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -80,7 +80,6 @@ typedef unsigned int xfs_buf_flags_t;
 #define XFS_BSTATE_IN_FLIGHT	 (1 << 1)	/* I/O in flight */
 
 struct xfs_buf_cache {
-	spinlock_t		bc_lock;
 	struct rhashtable	bc_hash;
 };
 
diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c
index f340a2015c4c..0b41bdfecdfb 100644
--- a/fs/xfs/xfs_exchrange.c
+++ b/fs/xfs/xfs_exchrange.c
@@ -329,22 +329,6 @@ out_trans_cancel:
  * successfully but before locks are dropped.
  */
 
-/* Verify that we have security clearance to perform this operation. */
-static int
-xfs_exchange_range_verify_area(
-	struct xfs_exchrange	*fxr)
-{
-	int			ret;
-
-	ret = remap_verify_area(fxr->file1, fxr->file1_offset, fxr->length,
-			true);
-	if (ret)
-		return ret;
-
-	return remap_verify_area(fxr->file2, fxr->file2_offset, fxr->length,
-			true);
-}
-
 /*
  * Performs necessary checks before doing a range exchange, having stabilized
  * mutable inode attributes via i_rwsem.
@@ -355,11 +339,13 @@ xfs_exchange_range_checks(
 	unsigned int		alloc_unit)
 {
 	struct inode		*inode1 = file_inode(fxr->file1);
+	loff_t			size1 = i_size_read(inode1);
 	struct inode		*inode2 = file_inode(fxr->file2);
+	loff_t			size2 = i_size_read(inode2);
 	uint64_t		allocmask = alloc_unit - 1;
 	int64_t			test_len;
 	uint64_t		blen;
-	loff_t			size1, size2, tmp;
+	loff_t			tmp;
 	int			error;
 
 	/* Don't touch certain kinds of inodes */
@@ -368,24 +354,25 @@ xfs_exchange_range_checks(
 	if (IS_SWAPFILE(inode1) || IS_SWAPFILE(inode2))
 		return -ETXTBSY;
 
-	size1 = i_size_read(inode1);
-	size2 = i_size_read(inode2);
-
 	/* Ranges cannot start after EOF. */
 	if (fxr->file1_offset > size1 || fxr->file2_offset > size2)
 		return -EINVAL;
 
-	/*
-	 * If the caller said to exchange to EOF, we set the length of the
-	 * request large enough to cover everything to the end of both files.
-	 */
 	if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) {
+		/*
+		 * If the caller said to exchange to EOF, we set the length of
+		 * the request large enough to cover everything to the end of
+		 * both files.
+		 */
 		fxr->length = max_t(int64_t, size1 - fxr->file1_offset,
 					     size2 - fxr->file2_offset);
-
-		error = xfs_exchange_range_verify_area(fxr);
-		if (error)
-			return error;
+	} else {
+		/*
+		 * Otherwise we require both ranges to end within EOF.
+		 */
+		if (fxr->file1_offset + fxr->length > size1 ||
+		    fxr->file2_offset + fxr->length > size2)
+			return -EINVAL;
 	}
 
 	/*
@@ -402,15 +389,6 @@ xfs_exchange_range_checks(
 		return -EINVAL;
 
 	/*
-	 * We require both ranges to end within EOF, unless we're exchanging
-	 * to EOF.
-	 */
-	if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) &&
-	    (fxr->file1_offset + fxr->length > size1 ||
-	     fxr->file2_offset + fxr->length > size2))
-		return -EINVAL;
-
-	/*
 	 * Make sure we don't hit any file size limits.  If we hit any size
 	 * limits such that test_length was adjusted, we abort the whole
 	 * operation.
@@ -747,6 +725,7 @@ xfs_exchange_range(
 {
 	struct inode		*inode1 = file_inode(fxr->file1);
 	struct inode		*inode2 = file_inode(fxr->file2);
+	loff_t			check_len = fxr->length;
 	int			ret;
 
 	BUILD_BUG_ON(XFS_EXCHANGE_RANGE_ALL_FLAGS &
@@ -779,14 +758,18 @@ xfs_exchange_range(
 		return -EBADF;
 
 	/*
-	 * If we're not exchanging to EOF, we can check the areas before
-	 * stabilizing both files' i_size.
+	 * If we're exchanging to EOF we can't calculate the length until taking
+	 * the iolock.  Pass a 0 length to remap_verify_area similar to the
+	 * FICLONE and FICLONERANGE ioctls that support cloning to EOF as well.
 	 */
-	if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)) {
-		ret = xfs_exchange_range_verify_area(fxr);
-		if (ret)
-			return ret;
-	}
+	if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)
+		check_len = 0;
+	ret = remap_verify_area(fxr->file1, fxr->file1_offset, check_len, true);
+	if (ret)
+		return ret;
+	ret = remap_verify_area(fxr->file2, fxr->file2_offset, check_len, true);
+	if (ret)
+		return ret;
 
 	/* Update cmtime if the fd/inode don't forbid it. */
 	if (!(fxr->file1->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode1))
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index c95fe1b1de4e..b1f9f156ec88 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1404,8 +1404,11 @@ xfs_inactive(
 		goto out;
 
 	/* Try to clean out the cow blocks if there are any. */
-	if (xfs_inode_has_cow_data(ip))
-		xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true);
+	if (xfs_inode_has_cow_data(ip)) {
+		error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true);
+		if (error)
+			goto out;
+	}
 
 	if (VFS_I(ip)->i_nlink != 0) {
 		/*
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 50fa3ef89f6c..d61460309a78 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -976,10 +976,8 @@ xfs_dax_write_iomap_end(
 	if (!xfs_is_cow_inode(ip))
 		return 0;
 
-	if (!written) {
-		xfs_reflink_cancel_cow_range(ip, pos, length, true);
-		return 0;
-	}
+	if (!written)
+		return xfs_reflink_cancel_cow_range(ip, pos, length, true);
 
 	return xfs_reflink_end_cow(ip, pos, written);
 }