Update bcachefs sources to 78c6c8127e21 bcachefs: Fix check_inode_hash_info_matches_root()

37 files changed, 428 insertions, 632 deletions

diff --git a/.bcachefs_revision b/.bcachefs_revision
index ae0826de..29c360d6 100644
--- a/.bcachefs_revision
+++ b/.bcachefs_revision
@@ -1 +1 @@
-f4b9a91fce7373d163343e5ca3a9b483d113706d
+78c6c8127e21fe2c8bf5c1d6a5e6832e28136f8f
diff --git a/libbcachefs/alloc_background.c b/libbcachefs/alloc_background.c
index 94e7bc88..fc2ef33b 100644
--- a/libbcachefs/alloc_background.c
+++ b/libbcachefs/alloc_background.c
@@ -1402,7 +1402,8 @@ int bch2_check_discard_freespace_key(struct btree_trans *trans, struct btree_ite
 
 	struct btree_iter alloc_iter;
 	struct bkey_s_c alloc_k = bch2_bkey_get_iter(trans, &alloc_iter,
-						     BTREE_ID_alloc, bucket, BTREE_ITER_cached);
+						     BTREE_ID_alloc, bucket,
+						     async_repair ? BTREE_ITER_cached : 0);
 	int ret = bkey_err(alloc_k);
 	if (ret)
 		return ret;
diff --git a/libbcachefs/backpointers.c b/libbcachefs/backpointers.c
index 655be233..ebeb6a5f 100644
--- a/libbcachefs/backpointers.c
+++ b/libbcachefs/backpointers.c
@@ -244,31 +244,27 @@ struct bkey_s_c bch2_backpointer_get_key(struct btree_trans *trans,
 	if (unlikely(bp.v->btree_id >= btree_id_nr_alive(c)))
 		return bkey_s_c_null;
 
-	bch2_trans_node_iter_init(trans, iter,
-				  bp.v->btree_id,
-				  bp.v->pos,
-				  0,
-				  bp.v->level,
-				  iter_flags);
-	struct bkey_s_c k = bch2_btree_iter_peek_slot(iter);
-	if (bkey_err(k)) {
-		bch2_trans_iter_exit(trans, iter);
-		return k;
-	}
+	if (likely(!bp.v->level)) {
+		bch2_trans_node_iter_init(trans, iter,
+					  bp.v->btree_id,
+					  bp.v->pos,
+					  0, 0,
+					  iter_flags);
+		struct bkey_s_c k = bch2_btree_iter_peek_slot(iter);
+		if (bkey_err(k)) {
+			bch2_trans_iter_exit(trans, iter);
+			return k;
+		}
 
-	if (k.k &&
-	    extent_matches_bp(c, bp.v->btree_id, bp.v->level, k, bp))
-		return k;
+		if (k.k &&
+		    extent_matches_bp(c, bp.v->btree_id, bp.v->level, k, bp))
+			return k;
 
-	bch2_trans_iter_exit(trans, iter);
-
-	if (!bp.v->level) {
+		bch2_trans_iter_exit(trans, iter);
 		int ret = backpointer_target_not_found(trans, bp, k, last_flushed);
 		return ret ? bkey_s_c_err(ret) : bkey_s_c_null;
 	} else {
 		struct btree *b = bch2_backpointer_get_node(trans, bp, iter, last_flushed);
-		if (b == ERR_PTR(-BCH_ERR_backpointer_to_overwritten_btree_node))
-			return bkey_s_c_null;
 		if (IS_ERR_OR_NULL(b))
 			return ((struct bkey_s_c) { .k = ERR_CAST(b) });
 
diff --git a/libbcachefs/bcachefs.h b/libbcachefs/bcachefs.h
index 13acfbf3..161cf2f0 100644
--- a/libbcachefs/bcachefs.h
+++ b/libbcachefs/bcachefs.h
@@ -687,8 +687,7 @@ struct btree_trans_buf {
 	x(gc_gens)							\
 	x(snapshot_delete_pagecache)					\
 	x(sysfs)							\
-	x(btree_write_buffer)						\
-	x(btree_node_scrub)
+	x(btree_write_buffer)
 
 enum bch_write_ref {
 #define x(n) BCH_WRITE_REF_##n,
diff --git a/libbcachefs/bcachefs_format.h b/libbcachefs/bcachefs_format.h
index 06809305..f70f0108 100644
--- a/libbcachefs/bcachefs_format.h
+++ b/libbcachefs/bcachefs_format.h
@@ -685,7 +685,8 @@ struct bch_sb_field_ext {
 	x(reflink_p_may_update_opts,	BCH_VERSION(1, 16))		\
 	x(inode_depth,			BCH_VERSION(1, 17))		\
 	x(persistent_inode_cursors,	BCH_VERSION(1, 18))		\
-	x(autofix_errors,		BCH_VERSION(1, 19))
+	x(autofix_errors,		BCH_VERSION(1, 19))		\
+	x(directory_size,		BCH_VERSION(1, 20))
 
 enum bcachefs_metadata_version {
 	bcachefs_metadata_version_min = 9,
diff --git a/libbcachefs/bcachefs_ioctl.h b/libbcachefs/bcachefs_ioctl.h
index f176f192..3c23bdf7 100644
--- a/libbcachefs/bcachefs_ioctl.h
+++ b/libbcachefs/bcachefs_ioctl.h
@@ -215,10 +215,6 @@ struct bch_ioctl_data {
 	union {
 	struct {
 		__u32		dev;
-		__u32		data_types;
-	}			scrub;
-	struct {
-		__u32		dev;
 		__u32		pad;
 	}			migrate;
 	struct {
@@ -241,19 +237,11 @@ struct bch_ioctl_data_progress {
 
 	__u64			sectors_done;
 	__u64			sectors_total;
-	__u64			sectors_error_corrected;
-	__u64			sectors_error_uncorrected;
 } __packed __aligned(8);
 
-enum bch_ioctl_data_event_ret {
-	BCH_IOCTL_DATA_EVENT_RET_done		= 1,
-	BCH_IOCTL_DATA_EVENT_RET_device_offline	= 2,
-};
-
 struct bch_ioctl_data_event {
 	__u8			type;
-	__u8			ret;
-	__u8			pad[6];
+	__u8			pad[7];
 	union {
 	struct bch_ioctl_data_progress p;
 	__u64			pad2[15];
diff --git a/libbcachefs/btree_io.c b/libbcachefs/btree_io.c
index 13ab827d..e371e60e 100644
--- a/libbcachefs/btree_io.c
+++ b/libbcachefs/btree_io.c
@@ -1,7 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include "bcachefs.h"
-#include "bkey_buf.h"
 #include "bkey_methods.h"
 #include "bkey_sort.h"
 #include "btree_cache.h"
@@ -1353,7 +1352,7 @@ start:
 
 		can_retry = bch2_bkey_pick_read_device(c,
 				bkey_i_to_s_c(&b->key),
-				&failed, &rb->pick, -1) > 0;
+				&failed, &rb->pick) > 0;
 
 		if (!bio->bi_status &&
 		    !bch2_btree_node_read_done(c, ca, b, can_retry, &saw_error)) {
@@ -1698,7 +1697,7 @@ void bch2_btree_node_read(struct btree_trans *trans, struct btree *b,
 		return;
 
 	ret = bch2_bkey_pick_read_device(c, bkey_i_to_s_c(&b->key),
-					 NULL, &pick, -1);
+					 NULL, &pick);
 
 	if (ret <= 0) {
 		struct printbuf buf = PRINTBUF;
@@ -1812,190 +1811,6 @@ int bch2_btree_root_read(struct bch_fs *c, enum btree_id id,
 	return bch2_trans_run(c, __bch2_btree_root_read(trans, id, k, level));
 }
 
-struct btree_node_scrub {
-	struct bch_fs		*c;
-	struct bch_dev		*ca;
-	void			*buf;
-	bool			used_mempool;
-	unsigned		written;
-
-	enum btree_id		btree;
-	unsigned		level;
-	struct bkey_buf		key;
-	__le64			seq;
-
-	struct work_struct	work;
-	struct bio		bio;
-};
-
-static bool btree_node_scrub_check(struct bch_fs *c, struct btree_node *data, unsigned ptr_written,
-				   struct printbuf *err)
-{
-	unsigned written = 0;
-
-	if (le64_to_cpu(data->magic) != bset_magic(c)) {
-		prt_printf(err, "bad magic: want %llx, got %llx",
-			   bset_magic(c), le64_to_cpu(data->magic));
-		return false;
-	}
-
-	while (written < (ptr_written ?: btree_sectors(c))) {
-		struct btree_node_entry *bne;
-		struct bset *i;
-		bool first = !written;
-
-		if (first) {
-			bne = NULL;
-			i = &data->keys;
-		} else {
-			bne = (void *) data + (written << 9);
-			i = &bne->keys;
-
-			if (!ptr_written && i->seq != data->keys.seq)
-				break;
-		}
-
-		struct nonce nonce = btree_nonce(i, written << 9);
-		bool good_csum_type = bch2_checksum_type_valid(c, BSET_CSUM_TYPE(i));
-
-		if (first) {
-			if (good_csum_type) {
-				struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, data);
-				if (bch2_crc_cmp(data->csum, csum)) {
-					bch2_csum_err_msg(err, BSET_CSUM_TYPE(i), data->csum, csum);
-					return false;
-				}
-			}
-
-			written += vstruct_sectors(data, c->block_bits);
-		} else {
-			if (good_csum_type) {
-				struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne);
-				if (bch2_crc_cmp(bne->csum, csum)) {
-					bch2_csum_err_msg(err, BSET_CSUM_TYPE(i), bne->csum, csum);
-					return false;
-				}
-			}
-
-			written += vstruct_sectors(bne, c->block_bits);
-		}
-	}
-
-	return true;
-}
-
-static void btree_node_scrub_work(struct work_struct *work)
-{
-	struct btree_node_scrub *scrub = container_of(work, struct btree_node_scrub, work);
-	struct bch_fs *c = scrub->c;
-	struct printbuf err = PRINTBUF;
-
-	__bch2_btree_pos_to_text(&err, c, scrub->btree, scrub->level,
-				 bkey_i_to_s_c(scrub->key.k));
-	prt_newline(&err);
-
-	if (!btree_node_scrub_check(c, scrub->buf, scrub->written, &err)) {
-		struct btree_trans *trans = bch2_trans_get(c);
-
-		struct btree_iter iter;
-		bch2_trans_node_iter_init(trans, &iter, scrub->btree,
-					  scrub->key.k->k.p, 0, scrub->level - 1, 0);
-
-		struct btree *b;
-		int ret = lockrestart_do(trans, PTR_ERR_OR_ZERO(b = bch2_btree_iter_peek_node(&iter)));
-		if (ret)
-			goto err;
-
-		if (bkey_i_to_btree_ptr_v2(&b->key)->v.seq == scrub->seq) {
-			bch_err(c, "error validating btree node during scrub on %s at btree %s",
-				scrub->ca->name, err.buf);
-
-			ret = bch2_btree_node_rewrite(trans, &iter, b, 0);
-		}
-err:
-		bch2_trans_iter_exit(trans, &iter);
-		bch2_trans_begin(trans);
-		bch2_trans_put(trans);
-	}
-
-	printbuf_exit(&err);
-	bch2_bkey_buf_exit(&scrub->key, c);;
-	btree_bounce_free(c, c->opts.btree_node_size, scrub->used_mempool, scrub->buf);
-	percpu_ref_put(&scrub->ca->io_ref);
-	kfree(scrub);
-	bch2_write_ref_put(c, BCH_WRITE_REF_btree_node_scrub);
-}
-
-static void btree_node_scrub_endio(struct bio *bio)
-{
-	struct btree_node_scrub *scrub = container_of(bio, struct btree_node_scrub, bio);
-
-	queue_work(scrub->c->btree_read_complete_wq, &scrub->work);
-}
-
-int bch2_btree_node_scrub(struct btree_trans *trans,
-			  enum btree_id btree, unsigned level,
-			  struct bkey_s_c k, unsigned dev)
-{
-	if (k.k->type != KEY_TYPE_btree_ptr_v2)
-		return 0;
-
-	struct bch_fs *c = trans->c;
-
-	if (!bch2_write_ref_tryget(c, BCH_WRITE_REF_btree_node_scrub))
-		return -BCH_ERR_erofs_no_writes;
-
-	struct extent_ptr_decoded pick;
-	int ret = bch2_bkey_pick_read_device(c, k, NULL, &pick, dev);
-	if (ret <= 0)
-		goto err;
-
-	struct bch_dev *ca = bch2_dev_get_ioref(c, pick.ptr.dev, READ);
-	if (!ca) {
-		ret = -BCH_ERR_device_offline;
-		goto err;
-	}
-
-	bool used_mempool = false;
-	void *buf = btree_bounce_alloc(c, c->opts.btree_node_size, &used_mempool);
-
-	unsigned vecs = buf_pages(buf, c->opts.btree_node_size);
-
-	struct btree_node_scrub *scrub =
-		kzalloc(sizeof(*scrub) + sizeof(struct bio_vec) * vecs, GFP_KERNEL);
-	if (!scrub) {
-		ret = -ENOMEM;
-		goto err_free;
-	}
-
-	scrub->c		= c;
-	scrub->ca		= ca;
-	scrub->buf		= buf;
-	scrub->used_mempool	= used_mempool;
-	scrub->written		= btree_ptr_sectors_written(k);
-
-	scrub->btree		= btree;
-	scrub->level		= level;
-	bch2_bkey_buf_init(&scrub->key);
-	bch2_bkey_buf_reassemble(&scrub->key, c, k);
-	scrub->seq		= bkey_s_c_to_btree_ptr_v2(k).v->seq;
-
-	INIT_WORK(&scrub->work, btree_node_scrub_work);
-
-	bio_init(&scrub->bio, ca->disk_sb.bdev, scrub->bio.bi_inline_vecs, vecs, REQ_OP_READ);
-	bch2_bio_map(&scrub->bio, scrub->buf, c->opts.btree_node_size);
-	scrub->bio.bi_iter.bi_sector	= pick.ptr.offset;
-	scrub->bio.bi_end_io		= btree_node_scrub_endio;
-	submit_bio(&scrub->bio);
-	return 0;
-err_free:
-	btree_bounce_free(c, c->opts.btree_node_size, used_mempool, buf);
-	percpu_ref_put(&ca->io_ref);
-err:
-	bch2_write_ref_put(c, BCH_WRITE_REF_btree_node_scrub);
-	return ret;
-}
-
 static void bch2_btree_complete_write(struct bch_fs *c, struct btree *b,
 				      struct btree_write *w)
 {
diff --git a/libbcachefs/btree_io.h b/libbcachefs/btree_io.h
index 75ead381..6f9e4a6d 100644
--- a/libbcachefs/btree_io.h
+++ b/libbcachefs/btree_io.h
@@ -132,9 +132,6 @@ void bch2_btree_node_read(struct btree_trans *, struct btree *, bool);
 int bch2_btree_root_read(struct bch_fs *, enum btree_id,
 			 const struct bkey_i *, unsigned);
 
-int bch2_btree_node_scrub(struct btree_trans *, enum btree_id, unsigned,
-			  struct bkey_s_c, unsigned);
-
 bool bch2_btree_post_write_cleanup(struct bch_fs *, struct btree *);
 
 enum btree_write_flags {
diff --git a/libbcachefs/btree_key_cache.c b/libbcachefs/btree_key_cache.c
index 7636a5e9..3b62296c 100644
--- a/libbcachefs/btree_key_cache.c
+++ b/libbcachefs/btree_key_cache.c
@@ -291,10 +291,8 @@ static noinline int btree_key_cache_fill(struct btree_trans *trans,
 					 struct btree_path *ck_path,
 					 unsigned flags)
 {
-	if (flags & BTREE_ITER_cached_nofill) {
-		ck_path->uptodate = BTREE_ITER_UPTODATE;
+	if (flags & BTREE_ITER_cached_nofill)
 		return 0;
-	}
 
 	struct bch_fs *c = trans->c;
 	struct btree_iter iter;
diff --git a/libbcachefs/btree_locking.c b/libbcachefs/btree_locking.c
index 85039314..10b805a6 100644
--- a/libbcachefs/btree_locking.c
+++ b/libbcachefs/btree_locking.c
@@ -109,6 +109,12 @@ static noinline void lock_graph_pop_all(struct lock_graph *g)
 		lock_graph_up(g);
 }
 
+static noinline void lock_graph_pop_from(struct lock_graph *g, struct trans_waiting_for_lock *i)
+{
+	while (g->g + g->nr > i)
+		lock_graph_up(g);
+}
+
 static void __lock_graph_down(struct lock_graph *g, struct btree_trans *trans)
 {
 	g->g[g->nr++] = (struct trans_waiting_for_lock) {
@@ -124,15 +130,20 @@ static void lock_graph_down(struct lock_graph *g, struct btree_trans *trans)
 	__lock_graph_down(g, trans);
 }
 
-static bool lock_graph_remove_non_waiters(struct lock_graph *g)
+static bool lock_graph_remove_non_waiters(struct lock_graph *g,
+					  struct trans_waiting_for_lock *from)
 {
 	struct trans_waiting_for_lock *i;
 
-	for (i = g->g + 1; i < g->g + g->nr; i++)
+	if (from->trans->locking != from->node_want) {
+		lock_graph_pop_from(g, from);
+		return true;
+	}
+
+	for (i = from + 1; i < g->g + g->nr; i++)
 		if (i->trans->locking != i->node_want ||
 		    i->trans->locking_wait.start_time != i[-1].lock_start_time) {
-			while (g->g + g->nr > i)
-				lock_graph_up(g);
+			lock_graph_pop_from(g, i);
 			return true;
 		}
 
@@ -179,13 +190,14 @@ static int btree_trans_abort_preference(struct btree_trans *trans)
 	return 3;
 }
 
-static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle)
+static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle,
+				struct trans_waiting_for_lock *from)
 {
 	struct trans_waiting_for_lock *i, *abort = NULL;
 	unsigned best = 0, pref;
 	int ret;
 
-	if (lock_graph_remove_non_waiters(g))
+	if (lock_graph_remove_non_waiters(g, from))
 		return 0;
 
 	/* Only checking, for debugfs: */
@@ -195,7 +207,7 @@ static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle)
 		goto out;
 	}
 
-	for (i = g->g; i < g->g + g->nr; i++) {
+	for (i = from; i < g->g + g->nr; i++) {
 		pref = btree_trans_abort_preference(i->trans);
 		if (pref > best) {
 			abort = i;
@@ -229,8 +241,9 @@ static noinline int break_cycle(struct lock_graph *g, struct printbuf *cycle)
 	ret = abort_lock(g, abort);
 out:
 	if (ret)
-		while (g->nr)
-			lock_graph_up(g);
+		lock_graph_pop_all(g);
+	else
+		lock_graph_pop_from(g, abort);
 	return ret;
 }
 
@@ -243,7 +256,7 @@ static int lock_graph_descend(struct lock_graph *g, struct btree_trans *trans,
 	for (i = g->g; i < g->g + g->nr; i++)
 		if (i->trans == trans) {
 			closure_put(&trans->ref);
-			return break_cycle(g, cycle);
+			return break_cycle(g, cycle, i);
 		}
 
 	if (g->nr == ARRAY_SIZE(g->g)) {
@@ -252,8 +265,7 @@ static int lock_graph_descend(struct lock_graph *g, struct btree_trans *trans,
 		if (orig_trans->lock_may_not_fail)
 			return 0;
 
-		while (g->nr)
-			lock_graph_up(g);
+		lock_graph_pop_all(g);
 
 		if (cycle)
 			return 0;
@@ -281,7 +293,7 @@ int bch2_check_for_deadlock(struct btree_trans *trans, struct printbuf *cycle)
 
 	g.nr = 0;
 
-	if (trans->lock_must_abort) {
+	if (trans->lock_must_abort && !trans->lock_may_not_fail) {
 		if (cycle)
 			return -1;
 
@@ -336,7 +348,7 @@ next:
 				 * structures - which means it can't be blocked
 				 * waiting on a lock:
 				 */
-				if (!lock_graph_remove_non_waiters(&g)) {
+				if (!lock_graph_remove_non_waiters(&g, g.g)) {
 					/*
 					 * If lock_graph_remove_non_waiters()
 					 * didn't do anything, it must be
@@ -512,7 +524,6 @@ bool bch2_btree_node_upgrade(struct btree_trans *trans,
 			     struct btree_path *path, unsigned level)
 {
 	struct btree *b = path->l[level].b;
-	struct six_lock_count count = bch2_btree_node_lock_counts(trans, path, &b->c, level);
 
 	if (!is_btree_node(path, level))
 		return false;
@@ -536,24 +547,11 @@ bool bch2_btree_node_upgrade(struct btree_trans *trans,
 	if (race_fault())
 		return false;
 
-	if (btree_node_locked(path, level)) {
-		bool ret;
-
-		six_lock_readers_add(&b->c.lock, -count.n[SIX_LOCK_read]);
-		ret = six_lock_tryupgrade(&b->c.lock);
-		six_lock_readers_add(&b->c.lock, count.n[SIX_LOCK_read]);
-
-		if (ret)
-			goto success;
-	} else {
-		if (six_relock_type(&b->c.lock, SIX_LOCK_intent, path->l[level].lock_seq))
-			goto success;
-	}
+	if (btree_node_locked(path, level)
+	    ? six_lock_tryupgrade(&b->c.lock)
+	    : six_relock_type(&b->c.lock, SIX_LOCK_intent, path->l[level].lock_seq))
+		goto success;
 
-	/*
-	 * Do we already have an intent lock via another path? If so, just bump
-	 * lock count:
-	 */
 	if (btree_node_lock_seq_matches(path, b, level) &&
 	    btree_node_lock_increment(trans, &b->c, level, BTREE_NODE_INTENT_LOCKED)) {
 		btree_node_unlock(trans, path, level);
diff --git a/libbcachefs/btree_update_interior.c b/libbcachefs/btree_update_interior.c
index ab111fec..f4aeadbe 100644
--- a/libbcachefs/btree_update_interior.c
+++ b/libbcachefs/btree_update_interior.c
@@ -2189,26 +2189,6 @@ err:
 	goto out;
 }
 
-int bch2_btree_node_rewrite_key(struct btree_trans *trans,
-				enum btree_id btree, unsigned level,
-				struct bpos pos, unsigned flags)
-{
-	BUG_ON(!level);
-
-	/* Traverse one depth lower to get a pointer to the node itself: */
-	struct btree_iter iter;
-	bch2_trans_node_iter_init(trans, &iter, btree, pos, 0, level - 1, 0);
-	struct btree *b = bch2_btree_iter_peek_node(&iter);
-	int ret = PTR_ERR_OR_ZERO(b);
-	if (ret)
-		goto err;
-
-	ret = bch2_btree_node_rewrite(trans, &iter, b, flags);
-err:
-	bch2_trans_iter_exit(trans, &iter);
-	return ret;
-}
-
 struct async_btree_rewrite {
 	struct bch_fs		*c;
 	struct work_struct	work;
diff --git a/libbcachefs/btree_update_interior.h b/libbcachefs/btree_update_interior.h
index fa5a88f9..7930ffea 100644
--- a/libbcachefs/btree_update_interior.h
+++ b/libbcachefs/btree_update_interior.h
@@ -169,11 +169,7 @@ static inline int bch2_foreground_maybe_merge(struct btree_trans *trans,
 
 int bch2_btree_node_rewrite(struct btree_trans *, struct btree_iter *,
 			    struct btree *, unsigned);
-int bch2_btree_node_rewrite_key(struct btree_trans *,
-				enum btree_id, unsigned,
-				struct bpos, unsigned);
 void bch2_btree_node_rewrite_async(struct bch_fs *, struct btree *);
-
 int bch2_btree_node_update_key(struct btree_trans *, struct btree_iter *,
 			       struct btree *, struct bkey_i *,
 			       unsigned, bool);
diff --git a/libbcachefs/chardev.c b/libbcachefs/chardev.c
index 0eb32074..46e9e321 100644
--- a/libbcachefs/chardev.c
+++ b/libbcachefs/chardev.c
@@ -312,10 +312,7 @@ static int bch2_data_thread(void *arg)
 	struct bch_data_ctx *ctx = container_of(arg, struct bch_data_ctx, thr);
 
 	ctx->thr.ret = bch2_data_job(ctx->c, &ctx->stats, ctx->arg);
-	if (ctx->thr.ret == -BCH_ERR_device_offline)
-		ctx->stats.ret = BCH_IOCTL_DATA_EVENT_RET_device_offline;
-	else
-		ctx->stats.ret = BCH_IOCTL_DATA_EVENT_RET_done;
+	ctx->stats.data_type = U8_MAX;
 	return 0;
 }
 
@@ -334,30 +331,14 @@ static ssize_t bch2_data_job_read(struct file *file, char __user *buf,
 	struct bch_data_ctx *ctx = container_of(file->private_data, struct bch_data_ctx, thr);
 	struct bch_fs *c = ctx->c;
 	struct bch_ioctl_data_event e = {
-		.type				= BCH_DATA_EVENT_PROGRESS,
-		.ret				= ctx->stats.ret,
-		.p.data_type			= ctx->stats.data_type,
-		.p.btree_id			= ctx->stats.pos.btree,
-		.p.pos				= ctx->stats.pos.pos,
-		.p.sectors_done			= atomic64_read(&ctx->stats.sectors_seen),
-		.p.sectors_error_corrected	= atomic64_read(&ctx->stats.sectors_error_corrected),
-		.p.sectors_error_uncorrected	= atomic64_read(&ctx->stats.sectors_error_uncorrected),
+		.type			= BCH_DATA_EVENT_PROGRESS,
+		.p.data_type		= ctx->stats.data_type,
+		.p.btree_id		= ctx->stats.pos.btree,
+		.p.pos			= ctx->stats.pos.pos,
+		.p.sectors_done		= atomic64_read(&ctx->stats.sectors_seen),
+		.p.sectors_total	= bch2_fs_usage_read_short(c).used,
 	};
 
-	if (ctx->arg.op == BCH_DATA_OP_scrub) {
-		struct bch_dev *ca = bch2_dev_tryget(c, ctx->arg.scrub.dev);
-		if (ca) {
-			struct bch_dev_usage u;
-			bch2_dev_usage_read_fast(ca, &u);
-			for (unsigned i = BCH_DATA_btree; i < ARRAY_SIZE(u.d); i++)
-				if (ctx->arg.scrub.data_types & BIT(i))
-					e.p.sectors_total += u.d[i].sectors;
-			bch2_dev_put(ca);
-		}
-	} else {
-		e.p.sectors_total	= bch2_fs_usage_read_short(c).used;
-	}
-
 	if (len < sizeof(e))
 		return -EINVAL;
 
diff --git a/libbcachefs/data_update.c b/libbcachefs/data_update.c
index 58521493..19ee424c 100644
--- a/libbcachefs/data_update.c
+++ b/libbcachefs/data_update.c
@@ -524,35 +524,42 @@ void bch2_data_update_opts_to_text(struct printbuf *out, struct bch_fs *c,
 				   struct bch_io_opts *io_opts,
 				   struct data_update_opts *data_opts)
 {
-	printbuf_tabstop_push(out, 20);
-	prt_str(out, "rewrite ptrs:\t");
+	if (!out->nr_tabstops)
+		printbuf_tabstop_push(out, 20);
+
+	prt_printf(out, "rewrite ptrs:\t");
 	bch2_prt_u64_base2(out, data_opts->rewrite_ptrs);
 	prt_newline(out);
 
-	prt_str(out, "kill ptrs:\t");
+	prt_printf(out, "kill ptrs:\t");
 	bch2_prt_u64_base2(out, data_opts->kill_ptrs);
 	prt_newline(out);
 
-	prt_str(out, "target:\t");
+	prt_printf(out, "target:\t");
 	bch2_target_to_text(out, c, data_opts->target);
 	prt_newline(out);
 
-	prt_str(out, "compression:\t");
+	prt_printf(out, "compression:\t");
 	bch2_compression_opt_to_text(out, io_opts->background_compression);
 	prt_newline(out);
 
-	prt_str(out, "opts.replicas:\t");
+	prt_printf(out, "opts.replicas:\t");
 	prt_u64(out, io_opts->data_replicas);
+	prt_newline(out);
 
-	prt_str(out, "extra replicas:\t");
+	prt_printf(out, "extra replicas:\t");
 	prt_u64(out, data_opts->extra_replicas);
+	prt_newline(out);
 }
 
 void bch2_data_update_to_text(struct printbuf *out, struct data_update *m)
 {
 	bch2_bkey_val_to_text(out, m->op.c, bkey_i_to_s_c(m->k.k));
 	prt_newline(out);
+	printbuf_indent_add(out, 2);
 	bch2_data_update_opts_to_text(out, m->op.c, &m->op.opts, &m->data_opts);
+	bch2_write_op_to_text(out, &m->op);
+	printbuf_indent_sub(out, 2);
 }
 
 int bch2_extent_drop_ptrs(struct btree_trans *trans,
diff --git a/libbcachefs/data_update.h b/libbcachefs/data_update.h
index 144b935c..e4b50723 100644
--- a/libbcachefs/data_update.h
+++ b/libbcachefs/data_update.h
@@ -15,9 +15,6 @@ struct data_update_opts {
 	u8		extra_replicas;
 	unsigned	btree_insert_flags;
 	unsigned	write_flags;
-
-	int		read_dev;
-	bool		scrub;
 };
 
 void bch2_data_update_opts_to_text(struct printbuf *, struct bch_fs *,
diff --git a/libbcachefs/debug.c b/libbcachefs/debug.c
index 3722c2ac..b5de52a5 100644
--- a/libbcachefs/debug.c
+++ b/libbcachefs/debug.c
@@ -189,7 +189,7 @@ void bch2_btree_node_ondisk_to_text(struct printbuf *out, struct bch_fs *c,
 	unsigned offset = 0;
 	int ret;
 
-	if (bch2_bkey_pick_read_device(c, bkey_i_to_s_c(&b->key), NULL, &pick, -1) <= 0) {
+	if (bch2_bkey_pick_read_device(c, bkey_i_to_s_c(&b->key), NULL, &pick) <= 0) {
 		prt_printf(out, "error getting device to read from: invalid device\n");
 		return;
 	}
diff --git a/libbcachefs/dirent.h b/libbcachefs/dirent.h
index 362b3b2f..a633f83c 100644
--- a/libbcachefs/dirent.h
+++ b/libbcachefs/dirent.h
@@ -31,6 +31,11 @@ static inline unsigned dirent_val_u64s(unsigned len)
 			    sizeof(u64));
 }
 
+static inline unsigned int dirent_occupied_size(const struct qstr *name)
+{
+	return (BKEY_U64s + dirent_val_u64s(name->len)) * sizeof(u64);
+}
+
 int bch2_dirent_read_target(struct btree_trans *, subvol_inum,
 			    struct bkey_s_c_dirent, subvol_inum *);
 
diff --git a/libbcachefs/ec_format.h b/libbcachefs/ec_format.h
index 64ef52e0..b9770f24 100644
--- a/libbcachefs/ec_format.h
+++ b/libbcachefs/ec_format.h
@@ -20,6 +20,23 @@ struct bch_stripe {
 	 */
 	__u8			disk_label;
 
+	/*
+	 * Variable length sections:
+	 * - Pointers
+	 * - Checksums
+	 *   2D array of [stripe block/device][csum block], with checksum block
+	 *   size given by csum_granularity_bits
+	 * - Block sector counts: per-block array of u16s
+	 *
+	 * XXX:
+	 * Either checksums should have come last, or we should have included a
+	 * checksum_size field (the size in bytes of the checksum itself, not
+	 * the blocksize the checksum covers).
+	 *
+	 * Currently we aren't able to access the block sector counts if the
+	 * checksum type is unknown.
+	 */
+
 	struct bch_extent_ptr	ptrs[];
 } __packed __aligned(8);
 
diff --git a/libbcachefs/errcode.h b/libbcachefs/errcode.h
index cae4fc3e..4590cd0c 100644
--- a/libbcachefs/errcode.h
+++ b/libbcachefs/errcode.h
@@ -269,7 +269,6 @@
 	x(EIO,				invalidate_stripe_to_dev)		\
 	x(EIO,				no_encryption_key)			\
 	x(EIO,				insufficient_journal_devices)		\
-	x(EIO,				device_offline)				\
 	x(BCH_ERR_btree_node_read_err,	btree_node_read_err_fixable)		\
 	x(BCH_ERR_btree_node_read_err,	btree_node_read_err_want_retry)		\
 	x(BCH_ERR_btree_node_read_err,	btree_node_read_err_must_retry)		\
diff --git a/libbcachefs/extents.c b/libbcachefs/extents.c
index 78a51d96..05d5f71a 100644
--- a/libbcachefs/extents.c
+++ b/libbcachefs/extents.c
@@ -114,9 +114,8 @@ static inline bool ptr_better(struct bch_fs *c,
  * other devices, it will still pick a pointer from avoid.
  */
 int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k,
-			      struct bch_io_failures *failed,
-			      struct extent_ptr_decoded *pick,
-			      int dev)
+			       struct bch_io_failures *failed,
+			       struct extent_ptr_decoded *pick)
 {
 	struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
 	const union bch_extent_entry *entry;
@@ -138,10 +137,6 @@ int bch2_bkey_pick_read_device(struct bch_fs *c, struct bkey_s_c k,
 			break;
 		}
 
-		/* Are we being asked to read from a specific device? */
-		if (dev >= 0 && p.ptr.dev != dev)
-			continue;
-
 		/*
 		 * If there are any dirty pointers it's an error if we can't
 		 * read:
diff --git a/libbcachefs/extents.h b/libbcachefs/extents.h
index 8fae6b23..620b284a 100644
--- a/libbcachefs/extents.h
+++ b/libbcachefs/extents.h
@@ -404,7 +404,7 @@ void bch2_mark_io_failure(struct bch_io_failures *,
 			  struct extent_ptr_decoded *);
 int bch2_bkey_pick_read_device(struct bch_fs *, struct bkey_s_c,
 			       struct bch_io_failures *,
-			       struct extent_ptr_decoded *, int);
+			       struct extent_ptr_decoded *);
 
 /* KEY_TYPE_btree_ptr: */
 
diff --git a/libbcachefs/fs-common.c b/libbcachefs/fs-common.c
index 2c3d46ac..d70d9f63 100644
--- a/libbcachefs/fs-common.c
+++ b/libbcachefs/fs-common.c
@@ -152,6 +152,7 @@ int bch2_create_trans(struct btree_trans *trans,
 		if (is_subdir_for_nlink(new_inode))
 			dir_u->bi_nlink++;
 		dir_u->bi_mtime = dir_u->bi_ctime = now;
+		dir_u->bi_size += dirent_occupied_size(name);
 
 		ret = bch2_inode_write(trans, &dir_iter, dir_u);
 		if (ret)
@@ -220,6 +221,7 @@ int bch2_link_trans(struct btree_trans *trans,
 	}
 
 	dir_u->bi_mtime = dir_u->bi_ctime = now;
+	dir_u->bi_size += dirent_occupied_size(name);
 
 	dir_hash = bch2_hash_info_init(c, dir_u);
 
@@ -322,6 +324,7 @@ int bch2_unlink_trans(struct btree_trans *trans,
 
 	dir_u->bi_mtime = dir_u->bi_ctime = inode_u->bi_ctime = now;
 	dir_u->bi_nlink -= is_subdir_for_nlink(inode_u);
+	dir_u->bi_size	-= dirent_occupied_size(name);
 
 	ret =   bch2_hash_delete_at(trans, bch2_dirent_hash_desc,
 				    &dir_hash, &dirent_iter,
@@ -460,6 +463,14 @@ int bch2_rename_trans(struct btree_trans *trans,
 		goto err;
 	}
 
+	if (mode == BCH_RENAME) {
+		src_dir_u->bi_size -= dirent_occupied_size(src_name);
+		dst_dir_u->bi_size += dirent_occupied_size(dst_name);
+	}
+
+	if (mode == BCH_RENAME_OVERWRITE)
+		src_dir_u->bi_size -= dirent_occupied_size(src_name);
+
 	if (src_inode_u->bi_parent_subvol)
 		src_inode_u->bi_parent_subvol = dst_dir.subvol;
 
diff --git a/libbcachefs/fs.c b/libbcachefs/fs.c
index 3f83f131..90ade8f6 100644
--- a/libbcachefs/fs.c
+++ b/libbcachefs/fs.c
@@ -67,6 +67,9 @@ void bch2_inode_update_after_write(struct btree_trans *trans,
 	i_gid_write(&inode->v, bi->bi_gid);
 	inode->v.i_mode	= bi->bi_mode;
 
+	if (fields & ATTR_SIZE)
+		i_size_write(&inode->v, bi->bi_size);
+
 	if (fields & ATTR_ATIME)
 		inode_set_atime_to_ts(&inode->v, bch2_time_to_timespec(c, bi->bi_atime));
 	if (fields & ATTR_MTIME)
@@ -582,7 +585,7 @@ err_before_quota:
 
 	if (!(flags & BCH_CREATE_TMPFILE)) {
 		bch2_inode_update_after_write(trans, dir, &dir_u,
-					      ATTR_MTIME|ATTR_CTIME);
+					      ATTR_MTIME|ATTR_CTIME|ATTR_SIZE);
 		mutex_unlock(&dir->ei_update_lock);
 	}
 
@@ -739,7 +742,7 @@ static int __bch2_link(struct bch_fs *c,
 
 	if (likely(!ret)) {
 		bch2_inode_update_after_write(trans, dir, &dir_u,
-					      ATTR_MTIME|ATTR_CTIME);
+					      ATTR_MTIME|ATTR_CTIME|ATTR_SIZE);
 		bch2_inode_update_after_write(trans, inode, &inode_u, ATTR_CTIME);
 	}
 
@@ -792,7 +795,7 @@ int __bch2_unlink(struct inode *vdir, struct dentry *dentry,
 		goto err;
 
 	bch2_inode_update_after_write(trans, dir, &dir_u,
-				      ATTR_MTIME|ATTR_CTIME);
+				      ATTR_MTIME|ATTR_CTIME|ATTR_SIZE);
 	bch2_inode_update_after_write(trans, inode, &inode_u,
 				      ATTR_MTIME);
 
@@ -970,11 +973,11 @@ err_tx_restart:
 	       dst_inode->v.i_ino != dst_inode_u.bi_inum);
 
 	bch2_inode_update_after_write(trans, src_dir, &src_dir_u,
-				      ATTR_MTIME|ATTR_CTIME);
+				      ATTR_MTIME|ATTR_CTIME|ATTR_SIZE);
 
 	if (src_dir != dst_dir)
 		bch2_inode_update_after_write(trans, dst_dir, &dst_dir_u,
-					      ATTR_MTIME|ATTR_CTIME);
+					      ATTR_MTIME|ATTR_CTIME|ATTR_SIZE);
 
 	bch2_inode_update_after_write(trans, src_inode, &src_inode_u,
 				      ATTR_CTIME);
diff --git a/libbcachefs/fsck.c b/libbcachefs/fsck.c
index 206fc046..8fcf7c8e 100644
--- a/libbcachefs/fsck.c
+++ b/libbcachefs/fsck.c
@@ -832,11 +832,13 @@ struct inode_walker {
 	struct bpos			last_pos;
 
 	DARRAY(struct inode_walker_entry) inodes;
+	snapshot_id_list		deletes;
 };
 
 static void inode_walker_exit(struct inode_walker *w)
 {
 	darray_exit(&w->inodes);
+	darray_exit(&w->deletes);
 }
 
 static struct inode_walker inode_walker_init(void)
@@ -960,8 +962,9 @@ static int get_visible_inodes(struct btree_trans *trans,
 	int ret;
 
 	w->inodes.nr = 0;
+	w->deletes.nr = 0;
 
-	for_each_btree_key_norestart(trans, iter, BTREE_ID_inodes, POS(0, inum),
+	for_each_btree_key_reverse_norestart(trans, iter, BTREE_ID_inodes, SPOS(0, inum, s->pos.snapshot),
 			   BTREE_ITER_all_snapshots, k, ret) {
 		if (k.k->p.offset != inum)
 			break;
@@ -969,10 +972,13 @@ static int get_visible_inodes(struct btree_trans *trans,
 		if (!ref_visible(c, s, s->pos.snapshot, k.k->p.snapshot))
 			continue;
 
-		if (bkey_is_inode(k.k))
-			add_inode(c, w, k);
+		if (snapshot_list_has_ancestor(c, &w->deletes, k.k->p.snapshot))
+			continue;
 
-		if (k.k->p.snapshot >= s->pos.snapshot)
+		ret = bkey_is_inode(k.k)
+			? add_inode(c, w, k)
+			: snapshot_list_add(c, &w->deletes, k.k->p.snapshot);
+		if (ret)
 			break;
 	}
 	bch2_trans_iter_exit(trans, &iter);
@@ -1110,6 +1116,37 @@ err:
 	return ret;
 }
 
+static int check_directory_size(struct btree_trans *trans,
+				struct bch_inode_unpacked *inode_u,
+				struct bkey_s_c inode_k, bool *write_inode)
+{
+	struct btree_iter iter;
+	struct bkey_s_c k;
+	u64 new_size = 0;
+	int ret;
+
+	for_each_btree_key_max_norestart(trans, iter, BTREE_ID_dirents,
+			SPOS(inode_k.k->p.offset, 0, inode_k.k->p.snapshot),
+			POS(inode_k.k->p.offset, U64_MAX),
+			0, k, ret) {
+		if (k.k->type != KEY_TYPE_dirent)
+			continue;
+
+		struct bkey_s_c_dirent dirent = bkey_s_c_to_dirent(k);
+		struct qstr name = bch2_dirent_get_name(dirent);
+
+		new_size += dirent_occupied_size(&name);
+	}
+	bch2_trans_iter_exit(trans, &iter);
+
+	if (!ret && inode_u->bi_size != new_size) {
+		inode_u->bi_size = new_size;
+		*write_inode = true;
+	}
+
+	return ret;
+}
+
 static int check_inode(struct btree_trans *trans,
 		       struct btree_iter *iter,
 		       struct bkey_s_c k,
@@ -1298,6 +1335,16 @@ static int check_inode(struct btree_trans *trans,
 		u.bi_journal_seq = journal_cur_seq(&c->journal);
 		do_update = true;
 	}
+
+	if (S_ISDIR(u.bi_mode)) {
+		ret = check_directory_size(trans, &u, k, &do_update);
+
+		fsck_err_on(ret,
+			    trans, directory_size_mismatch,
+			    "directory inode %llu:%u with the mismatch directory size",
+			    u.bi_inum, k.k->p.snapshot);
+		ret = 0;
+	}
 do_update:
 	if (do_update) {
 		ret = __bch2_fsck_write_inode(trans, &u);
@@ -2380,6 +2427,30 @@ static int check_dirent(struct btree_trans *trans, struct btree_iter *iter,
 			if (ret)
 				goto err;
 		}
+
+		darray_for_each(target->deletes, i)
+			if (fsck_err_on(!snapshot_list_has_id(&s->ids, *i),
+					trans, dirent_to_overwritten_inode,
+					"dirent points to inode overwritten in snapshot %u:\n%s",
+					*i,
+					(printbuf_reset(&buf),
+					 bch2_bkey_val_to_text(&buf, c, k),
+					 buf.buf))) {
+				struct btree_iter delete_iter;
+				bch2_trans_iter_init(trans, &delete_iter,
+						     BTREE_ID_dirents,
+						     SPOS(k.k->p.inode, k.k->p.offset, *i),
+						     BTREE_ITER_intent);
+				ret =   bch2_btree_iter_traverse(&delete_iter) ?:
+					bch2_hash_delete_at(trans, bch2_dirent_hash_desc,
+							  hash_info,
+							  &delete_iter,
+							  BTREE_UPDATE_internal_snapshot_node);
+				bch2_trans_iter_exit(trans, &delete_iter);
+				if (ret)
+					goto err;
+
+			}
 	}
 
 	ret = bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc);
diff --git a/libbcachefs/io_read.c b/libbcachefs/io_read.c
index 87fe5997..8c7b2d3d 100644
--- a/libbcachefs/io_read.c
+++ b/libbcachefs/io_read.c
@@ -91,13 +91,18 @@ static const struct rhashtable_params bch_promote_params = {
 	.automatic_shrinking	= true,
 };
 
+static inline bool have_io_error(struct bch_io_failures *failed)
+{
+	return failed && failed->nr;
+}
+
 static inline int should_promote(struct bch_fs *c, struct bkey_s_c k,
 				  struct bpos pos,
 				  struct bch_io_opts opts,
 				  unsigned flags,
 				  struct bch_io_failures *failed)
 {
-	if (!failed) {
+	if (!have_io_error(failed)) {
 		BUG_ON(!opts.promote_target);
 
 		if (!(flags & BCH_READ_MAY_PROMOTE))
@@ -224,7 +229,7 @@ static struct promote_op *__promote_alloc(struct btree_trans *trans,
 
 	struct data_update_opts update_opts = {};
 
-	if (!failed) {
+	if (!have_io_error(failed)) {
 		update_opts.target = opts.promote_target;
 		update_opts.extra_replicas = 1;
 		update_opts.write_flags = BCH_WRITE_ALLOC_NOWAIT|BCH_WRITE_CACHED;
@@ -286,7 +291,7 @@ static struct promote_op *promote_alloc(struct btree_trans *trans,
 	 * if failed != NULL we're not actually doing a promote, we're
 	 * recovering from an io/checksum error
 	 */
-	bool promote_full = (failed ||
+	bool promote_full = (have_io_error(failed) ||
 			     *read_full ||
 			     READ_ONCE(c->opts.promote_whole_extents));
 	/* data might have to be decompressed in the write path: */
@@ -444,7 +449,7 @@ retry:
 	ret = __bch2_read_extent(trans, rbio, bvec_iter,
 				 rbio->read_pos,
 				 rbio->data_btree,
-				 k, 0, failed, flags, -1);
+				 k, 0, failed, flags);
 	if (ret == READ_RETRY)
 		goto retry;
 	if (ret)
@@ -499,7 +504,6 @@ static void bch2_rbio_error(struct bch_read_bio *rbio, int retry,
 			    blk_status_t error)
 {
 	rbio->retry = retry;
-	rbio->saw_error = true;
 
 	if (rbio->flags & BCH_READ_IN_RETRY)
 		return;
@@ -741,7 +745,7 @@ static void __bch2_read_endio(struct work_struct *work)
 			bio_copy_data_iter(dst, &dst_iter, src, &src_iter);
 		}
 	}
-nodecode:
+
 	if (rbio->promote) {
 		/*
 		 * Re encrypt data we decrypted, so it's consistent with
@@ -754,7 +758,7 @@ nodecode:
 		promote_start(rbio->promote, rbio);
 		rbio->promote = NULL;
 	}
-
+nodecode:
 	if (likely(!(rbio->flags & BCH_READ_IN_RETRY))) {
 		rbio = bch2_rbio_free(rbio);
 		bch2_rbio_done(rbio);
@@ -879,7 +883,7 @@ int __bch2_read_extent(struct btree_trans *trans, struct bch_read_bio *orig,
 		       struct bvec_iter iter, struct bpos read_pos,
 		       enum btree_id data_btree, struct bkey_s_c k,
 		       unsigned offset_into_extent,
-		       struct bch_io_failures *failed, unsigned flags, int dev)
+		       struct bch_io_failures *failed, unsigned flags)
 {
 	struct bch_fs *c = trans->c;
 	struct extent_ptr_decoded pick;
@@ -889,8 +893,6 @@ int __bch2_read_extent(struct btree_trans *trans, struct bch_read_bio *orig,
 	struct bpos data_pos = bkey_start_pos(k.k);
 	int pick_ret;
 
-	//BUG_ON(failed && failed->nr);
-
 	if (bkey_extent_is_inline_data(k.k)) {
 		unsigned bytes = min_t(unsigned, iter.bi_size,
 				       bkey_inline_data_bytes(k.k));
@@ -903,7 +905,7 @@ int __bch2_read_extent(struct btree_trans *trans, struct bch_read_bio *orig,
 		goto out_read_done;
 	}
 retry_pick:
-	pick_ret = bch2_bkey_pick_read_device(c, k, failed, &pick, dev);
+	pick_ret = bch2_bkey_pick_read_device(c, k, failed, &pick);
 
 	/* hole or reservation - just zero fill: */
 	if (!pick_ret)
@@ -955,30 +957,7 @@ retry_pick:
 	 */
 	bch2_trans_unlock(trans);
 
-	if (!(flags & BCH_READ_NODECODE)) {
-		if (!(flags & BCH_READ_LAST_FRAGMENT) ||
-		    bio_flagged(&orig->bio, BIO_CHAIN))
-			flags |= BCH_READ_MUST_CLONE;
-
-		narrow_crcs = !(flags & BCH_READ_IN_RETRY) &&
-			bch2_can_narrow_extent_crcs(k, pick.crc);
-
-		if (narrow_crcs && (flags & BCH_READ_USER_MAPPED))
-			flags |= BCH_READ_MUST_BOUNCE;
-
-		EBUG_ON(offset_into_extent + bvec_iter_sectors(iter) > k.k->size);
-
-		if (crc_is_compressed(pick.crc) ||
-		    (pick.crc.csum_type != BCH_CSUM_none &&
-		     (bvec_iter_sectors(iter) != pick.crc.uncompressed_size ||
-		      (bch2_csum_type_is_encryption(pick.crc.csum_type) &&
-		       (flags & BCH_READ_USER_MAPPED)) ||
-		      (flags & BCH_READ_MUST_BOUNCE)))) {
-			read_full = true;
-			bounce = true;
-		}
-	} else {
-		read_full = true;
+	if (flags & BCH_READ_NODECODE) {
 		/*
 		 * can happen if we retry, and the extent we were going to read
 		 * has been merged in the meantime:
@@ -990,10 +969,32 @@ retry_pick:
 		}
 
 		iter.bi_size	= pick.crc.compressed_size << 9;
+		goto get_bio;
+	}
+
+	if (!(flags & BCH_READ_LAST_FRAGMENT) ||
+	    bio_flagged(&orig->bio, BIO_CHAIN))
+		flags |= BCH_READ_MUST_CLONE;
+
+	narrow_crcs = !(flags & BCH_READ_IN_RETRY) &&
+		bch2_can_narrow_extent_crcs(k, pick.crc);
+
+	if (narrow_crcs && (flags & BCH_READ_USER_MAPPED))
+		flags |= BCH_READ_MUST_BOUNCE;
+
+	EBUG_ON(offset_into_extent + bvec_iter_sectors(iter) > k.k->size);
+
+	if (crc_is_compressed(pick.crc) ||
+	    (pick.crc.csum_type != BCH_CSUM_none &&
+	     (bvec_iter_sectors(iter) != pick.crc.uncompressed_size ||
+	      (bch2_csum_type_is_encryption(pick.crc.csum_type) &&
+	       (flags & BCH_READ_USER_MAPPED)) ||
+	      (flags & BCH_READ_MUST_BOUNCE)))) {
+		read_full = true;
+		bounce = true;
 	}
 
-	if ((orig->opts.promote_target && !(flags & BCH_READ_NODECODE)) ||
-	    (failed && failed->nr))
+	if (orig->opts.promote_target || have_io_error(failed))
 		promote = promote_alloc(trans, iter, k, &pick, orig->opts, flags,
 					&rbio, &bounce, &read_full, failed);
 
@@ -1014,7 +1015,7 @@ retry_pick:
 		pick.crc.offset			= 0;
 		pick.crc.live_size		= bvec_iter_sectors(iter);
 	}
-
+get_bio:
 	if (rbio) {
 		/*
 		 * promote already allocated bounce rbio:
@@ -1265,7 +1266,7 @@ void __bch2_read(struct bch_fs *c, struct bch_read_bio *rbio,
 
 		ret = __bch2_read_extent(trans, rbio, bvec_iter, iter.pos,
 					 data_btree, k,
-					 offset_into_extent, failed, flags, -1);
+					 offset_into_extent, failed, flags);
 		if (ret)
 			goto err;
 
diff --git a/libbcachefs/io_read.h b/libbcachefs/io_read.h
index b62fcee7..a82e8a94 100644
--- a/libbcachefs/io_read.h
+++ b/libbcachefs/io_read.h
@@ -41,7 +41,6 @@ struct bch_read_bio {
 				have_ioref:1,
 				narrow_crcs:1,
 				hole:1,
-				saw_error:1,
 				retry:2,
 				context:2;
 	};
@@ -125,7 +124,7 @@ enum bch_read_flags {
 int __bch2_read_extent(struct btree_trans *, struct bch_read_bio *,
 		       struct bvec_iter, struct bpos, enum btree_id,
 		       struct bkey_s_c, unsigned,
-		       struct bch_io_failures *, unsigned, int);
+		       struct bch_io_failures *, unsigned);
 
 static inline void bch2_read_extent(struct btree_trans *trans,
 			struct bch_read_bio *rbio, struct bpos read_pos,
@@ -133,7 +132,7 @@ static inline void bch2_read_extent(struct btree_trans *trans,
 			unsigned offset_into_extent, unsigned flags)
 {
 	__bch2_read_extent(trans, rbio, rbio->bio.bi_iter, read_pos,
-			   data_btree, k, offset_into_extent, NULL, flags, -1);
+			   data_btree, k, offset_into_extent, NULL, flags);
 }
 
 void __bch2_read(struct bch_fs *, struct bch_read_bio *, struct bvec_iter,
diff --git a/libbcachefs/journal_io.c b/libbcachefs/journal_io.c
index e1773ac2..7f2efe85 100644
--- a/libbcachefs/journal_io.c
+++ b/libbcachefs/journal_io.c
@@ -1114,10 +1114,8 @@ reread:
 				       (printbuf_reset(&err),
 					prt_str(&err, "journal "),
 					bch2_csum_err_msg(&err, csum_type, j->csum, csum),
-					err.buf))) {
+					err.buf)))
 			saw_bad = true;
-			bch2_fatal_error(c);
-		}
 
 		ret = bch2_encrypt(c, JSET_CSUM_TYPE(j), journal_nonce(j),
 			     j->encrypted_start,
diff --git a/libbcachefs/move.c b/libbcachefs/move.c
index cc852492..6c6ece52 100644
--- a/libbcachefs/move.c
+++ b/libbcachefs/move.c
@@ -88,13 +88,12 @@ static void move_free(struct moving_io *io)
 	if (io->b)
 		atomic_dec(&io->b->count);
 
-	bch2_data_update_exit(&io->write);
-
 	mutex_lock(&ctxt->lock);
 	list_del(&io->io_list);
 	wake_up(&ctxt->wait);
 	mutex_unlock(&ctxt->lock);
 
+	bch2_data_update_exit(&io->write);
 	kfree(io);
 }
 
@@ -114,20 +113,7 @@ static void move_write_done(struct bch_write_op *op)
 
 static void move_write(struct moving_io *io)
 {
-	struct moving_context *ctxt = io->write.ctxt;
-
-	if (ctxt->stats) {
-		if (io->rbio.bio.bi_status)
-			atomic64_add(io->rbio.bvec_iter.bi_size >> 9,
-				     &ctxt->stats->sectors_error_uncorrected);
-		else if (io->rbio.saw_error)
-			atomic64_add(io->rbio.bvec_iter.bi_size >> 9,
-				     &ctxt->stats->sectors_error_corrected);
-	}
-
-	if (unlikely(io->rbio.bio.bi_status ||
-		     io->rbio.hole ||
-		     io->write.data_opts.scrub)) {
+	if (unlikely(io->rbio.bio.bi_status || io->rbio.hole)) {
 		move_free(io);
 		return;
 	}
@@ -286,8 +272,7 @@ int bch2_move_extent(struct moving_context *ctxt,
 	bch2_data_update_opts_normalize(k, &data_opts);
 
 	if (!data_opts.rewrite_ptrs &&
-	    !data_opts.extra_replicas &&
-	    !data_opts.scrub) {
+	    !data_opts.extra_replicas) {
 		if (data_opts.kill_ptrs)
 			return bch2_extent_drop_ptrs(trans, iter, k, &io_opts, &data_opts);
 		return 0;
@@ -333,18 +318,12 @@ int bch2_move_extent(struct moving_context *ctxt,
 	io->rbio.bio.bi_iter.bi_sector	= bkey_start_offset(k.k);
 	io->rbio.bio.bi_end_io		= move_read_endio;
 
-	if (!data_opts.scrub) {
-		ret = bch2_data_update_init(trans, iter, ctxt, &io->write, ctxt->wp,
-					    io_opts, data_opts, iter->btree_id, k);
-		if (ret)
-			goto err_free_pages;
+	ret = bch2_data_update_init(trans, iter, ctxt, &io->write, ctxt->wp,
+				    io_opts, data_opts, iter->btree_id, k);
+	if (ret)
+		goto err_free_pages;
 
-		io->write.op.end_io	= move_write_done;
-	} else {
-		bch2_bkey_buf_init(&io->write.k);
-		io->write.op.c		= c;
-		io->write.data_opts	= data_opts;
-	}
+	io->write.op.end_io = move_write_done;
 
 	if (ctxt->rate)
 		bch2_ratelimit_increment(ctxt->rate, k.k->size);
@@ -376,14 +355,11 @@ int bch2_move_extent(struct moving_context *ctxt,
 	 * ctxt when doing wakeup
 	 */
 	closure_get(&ctxt->cl);
-	__bch2_read_extent(trans, &io->rbio,
-			   io->rbio.bio.bi_iter,
-			   bkey_start_pos(k.k),
-			   iter->btree_id, k, 0,
-			   NULL,
-			   BCH_READ_NODECODE|
-			   BCH_READ_LAST_FRAGMENT,
-			   data_opts.scrub ?  data_opts.read_dev : -1);
+	bch2_read_extent(trans, &io->rbio,
+			 bkey_start_pos(k.k),
+			 iter->btree_id, k, 0,
+			 BCH_READ_NODECODE|
+			 BCH_READ_LAST_FRAGMENT);
 	return 0;
 err_free_pages:
 	bio_free_pages(&io->write.op.wbio.bio);
@@ -712,22 +688,21 @@ int bch2_move_data(struct bch_fs *c,
 		   bool wait_on_copygc,
 		   move_pred_fn pred, void *arg)
 {
+
 	struct moving_context ctxt;
+	int ret;
 
 	bch2_moving_ctxt_init(&ctxt, c, rate, stats, wp, wait_on_copygc);
-	int ret = __bch2_move_data(&ctxt, start, end, pred, arg);
+	ret = __bch2_move_data(&ctxt, start, end, pred, arg);
 	bch2_moving_ctxt_exit(&ctxt);
 
 	return ret;
 }
 
-static int __bch2_move_data_phys(struct moving_context *ctxt,
-			struct move_bucket_in_flight *bucket_in_flight,
-			unsigned dev,
-			u64 bucket_start,
-			u64 bucket_end,
-			unsigned data_types,
-			move_pred_fn pred, void *arg)
+int bch2_evacuate_bucket(struct moving_context *ctxt,
+			   struct move_bucket_in_flight *bucket_in_flight,
+			   struct bpos bucket, int gen,
+			   struct data_update_opts _data_opts)
 {
 	struct btree_trans *trans = ctxt->trans;
 	struct bch_fs *c = trans->c;
@@ -736,22 +711,16 @@ static int __bch2_move_data_phys(struct moving_context *ctxt,
 	struct btree_iter iter = {}, bp_iter = {};
 	struct bkey_buf sk;
 	struct bkey_s_c k;
+	struct data_update_opts data_opts;
 	unsigned sectors_moved = 0;
 	struct bkey_buf last_flushed;
 	int ret = 0;
 
-	struct bch_dev *ca = bch2_dev_tryget(c, dev);
+	struct bch_dev *ca = bch2_dev_tryget(c, bucket.inode);
 	if (!ca)
 		return 0;
 
-	bucket_end = min(bucket_end, ca->mi.nbuckets);
-
-	struct bpos bp_start	= bucket_pos_to_bp_start(ca, POS(dev, bucket_start));
-	struct bpos bp_end	= bucket_pos_to_bp_end(ca, POS(dev, bucket_end));
-	bch2_dev_put(ca);
-	ca = NULL;
-
-	//trace_bucket_evacuate(c, &bucket);
+	trace_bucket_evacuate(c, &bucket);
 
 	bch2_bkey_buf_init(&last_flushed);
 	bkey_init(&last_flushed.k->k);
@@ -762,7 +731,8 @@ static int __bch2_move_data_phys(struct moving_context *ctxt,
 	 */
 	bch2_trans_begin(trans);
 
-	bch2_trans_iter_init(trans, &bp_iter, BTREE_ID_backpointers, bp_start, 0);
+	bch2_trans_iter_init(trans, &bp_iter, BTREE_ID_backpointers,
+			     bucket_pos_to_bp_start(ca, bucket), 0);
 
 	bch_err_msg(c, ret, "looking up alloc key");
 	if (ret)
@@ -786,7 +756,7 @@ static int __bch2_move_data_phys(struct moving_context *ctxt,
 		if (ret)
 			goto err;
 
-		if (!k.k || bkey_gt(k.k->p, bp_end))
+		if (!k.k || bkey_gt(k.k->p, bucket_pos_to_bp_end(ca, bucket)))
 			break;
 
 		if (k.k->type != KEY_TYPE_backpointer)
@@ -794,145 +764,107 @@ static int __bch2_move_data_phys(struct moving_context *ctxt,
 
 		struct bkey_s_c_backpointer bp = bkey_s_c_to_backpointer(k);
 
-		if (ctxt->stats)
-			ctxt->stats->offset = bp.k->p.offset >> MAX_EXTENT_COMPRESS_RATIO_SHIFT;
+		if (!bp.v->level) {
+			k = bch2_backpointer_get_key(trans, bp, &iter, 0, &last_flushed);
+			ret = bkey_err(k);
+			if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
+				continue;
+			if (ret)
+				goto err;
+			if (!k.k)
+				goto next;
 
-		if (!(data_types & BIT(bp.v->data_type)))
-			goto next;
+			bch2_bkey_buf_reassemble(&sk, c, k);
+			k = bkey_i_to_s_c(sk.k);
 
-		k = bch2_backpointer_get_key(trans, bp, &iter, 0, &last_flushed);
-		ret = bkey_err(k);
-		if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
-			continue;
-		if (ret)
-			goto err;
-		if (!k.k)
-			goto next;
-
-		if (!bp.v->level) {
 			ret = bch2_move_get_io_opts_one(trans, &io_opts, &iter, k);
 			if (ret) {
 				bch2_trans_iter_exit(trans, &iter);
 				continue;
 			}
-		}
 
-		struct data_update_opts data_opts = {};
-		if (!pred(c, arg, k, &io_opts, &data_opts)) {
-			bch2_trans_iter_exit(trans, &iter);
-			goto next;
-		}
+			data_opts = _data_opts;
+			data_opts.target	= io_opts.background_target;
+			data_opts.rewrite_ptrs = 0;
+
+			unsigned sectors = bp.v->bucket_len; /* move_extent will drop locks */
+			unsigned i = 0;
+			const union bch_extent_entry *entry;
+			struct extent_ptr_decoded p;
+			bkey_for_each_ptr_decode(k.k, bch2_bkey_ptrs_c(k), p, entry) {
+				if (p.ptr.dev == bucket.inode) {
+					if (p.ptr.cached) {
+						bch2_trans_iter_exit(trans, &iter);
+						goto next;
+					}
+					data_opts.rewrite_ptrs |= 1U << i;
+					break;
+				}
+				i++;
+			}
 
-		if (data_opts.scrub &&
-		    !bch2_dev_idx_is_online(c, data_opts.read_dev)) {
+			ret = bch2_move_extent(ctxt, bucket_in_flight,
+					       &iter, k, io_opts, data_opts);
 			bch2_trans_iter_exit(trans, &iter);
-			ret = -BCH_ERR_device_offline;
-			break;
-		}
 
-		bch2_bkey_buf_reassemble(&sk, c, k);
-		k = bkey_i_to_s_c(sk.k);
+			if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
+				continue;
+			if (ret == -ENOMEM) {
+				/* memory allocation failure, wait for some IO to finish */
+				bch2_move_ctxt_wait_for_io(ctxt);
+				continue;
+			}
+			if (ret)
+				goto err;
 
-		/* move_extent will drop locks */
-		unsigned sectors = bp.v->bucket_len;
+			if (ctxt->stats)
+				atomic64_add(sectors, &ctxt->stats->sectors_seen);
+			sectors_moved += sectors;
+		} else {
+			struct btree *b;
 
-		if (!bp.v->level)
-			ret = bch2_move_extent(ctxt, bucket_in_flight, &iter, k, io_opts, data_opts);
-		else if (!data_opts.scrub)
-			ret = bch2_btree_node_rewrite_key(trans, bp.v->btree_id, bp.v->level, k.k->p, 0);
-		else
-			ret = bch2_btree_node_scrub(trans, bp.v->btree_id, bp.v->level, k, data_opts.read_dev);
+			b = bch2_backpointer_get_node(trans, bp, &iter, &last_flushed);
+			ret = PTR_ERR_OR_ZERO(b);
+			if (ret == -BCH_ERR_backpointer_to_overwritten_btree_node)
+				goto next;
+			if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
+				continue;
+			if (ret)
+				goto err;
+			if (!b)
+				goto next;
 
-		bch2_trans_iter_exit(trans, &iter);
+			unsigned sectors = btree_ptr_sectors_written(bkey_i_to_s_c(&b->key));
 
-		if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
-			continue;
-		if (ret == -ENOMEM) {
-			/* memory allocation failure, wait for some IO to finish */
-			bch2_move_ctxt_wait_for_io(ctxt);
-			continue;
-		}
-		if (ret)
-			goto err;
+			ret = bch2_btree_node_rewrite(trans, &iter, b, 0);
+			bch2_trans_iter_exit(trans, &iter);
 
-		if (ctxt->stats)
-			atomic64_add(sectors, &ctxt->stats->sectors_seen);
-		sectors_moved += sectors;
+			if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
+				continue;
+			if (ret)
+				goto err;
+
+			if (ctxt->rate)
+				bch2_ratelimit_increment(ctxt->rate, sectors);
+			if (ctxt->stats) {
+				atomic64_add(sectors, &ctxt->stats->sectors_seen);
+				atomic64_add(sectors, &ctxt->stats->sectors_moved);
+			}
+			sectors_moved += btree_sectors(c);
+		}
 next:
 		bch2_btree_iter_advance(&bp_iter);
 	}
 
-	//trace_evacuate_bucket(c, &bucket, sectors_moved, ca->mi.bucket_size, ret);
+	trace_evacuate_bucket(c, &bucket, sectors_moved, ca->mi.bucket_size, ret);
 err:
 	bch2_trans_iter_exit(trans, &bp_iter);
+	bch2_dev_put(ca);
 	bch2_bkey_buf_exit(&sk, c);
 	bch2_bkey_buf_exit(&last_flushed, c);
 	return ret;
 }
 
-static int bch2_move_data_phys(struct bch_fs *c,
-			       unsigned dev,
-			       u64 start,
-			       u64 end,
-			       unsigned data_types,
-			       struct bch_ratelimit *rate,
-			       struct bch_move_stats *stats,
-			       struct write_point_specifier wp,
-			       bool wait_on_copygc,
-			       move_pred_fn pred, void *arg)
-{
-	struct moving_context ctxt;
-
-	bch2_trans_run(c, bch2_btree_write_buffer_flush_sync(trans));
-
-	bch2_moving_ctxt_init(&ctxt, c, rate, stats, wp, wait_on_copygc);
-	ctxt.stats->phys = true;
-
-	int ret = __bch2_move_data_phys(&ctxt, NULL, dev, start, end, data_types, pred, arg);
-	bch2_moving_ctxt_exit(&ctxt);
-
-	return ret;
-}
-
-struct evacuate_bucket_arg {
-	struct bpos		bucket;
-	int			gen;
-	struct data_update_opts	data_opts;
-};
-
-static bool evacuate_bucket_pred(struct bch_fs *c, void *_arg, struct bkey_s_c k,
-				 struct bch_io_opts *io_opts,
-				 struct data_update_opts *data_opts)
-{
-	struct evacuate_bucket_arg *arg = _arg;
-
-	*data_opts = arg->data_opts;
-
-	unsigned i = 0;
-	bkey_for_each_ptr(bch2_bkey_ptrs_c(k), ptr) {
-		if (ptr->dev == arg->bucket.inode && !ptr->cached)
-			data_opts->rewrite_ptrs |= BIT(i);
-		i++;
-	}
-
-	return data_opts->rewrite_ptrs != 0;
-}
-
-int bch2_evacuate_bucket(struct moving_context *ctxt,
-			   struct move_bucket_in_flight *bucket_in_flight,
-			   struct bpos bucket, int gen,
-			   struct data_update_opts data_opts)
-{
-	struct evacuate_bucket_arg arg = { bucket, gen, data_opts, };
-
-	return __bch2_move_data_phys(ctxt, bucket_in_flight,
-				   bucket.inode,
-				   bucket.offset,
-				   bucket.offset + 1,
-				   ~0,
-				   evacuate_bucket_pred, &arg);
-}
-
 typedef bool (*move_btree_pred)(struct bch_fs *, void *,
 				struct btree *, struct bch_io_opts *,
 				struct data_update_opts *);
@@ -1171,30 +1103,6 @@ static bool drop_extra_replicas_btree_pred(struct bch_fs *c, void *arg,
 	return drop_extra_replicas_pred(c, arg, bkey_i_to_s_c(&b->key), io_opts, data_opts);
 }
 
-static bool scrub_pred(struct bch_fs *c, void *_arg,
-		       struct bkey_s_c k,
-		       struct bch_io_opts *io_opts,
-		       struct data_update_opts *data_opts)
-{
-	struct bch_ioctl_data *arg = _arg;
-
-	if (k.k->type != KEY_TYPE_btree_ptr_v2) {
-		struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
-		const union bch_extent_entry *entry;
-		struct extent_ptr_decoded p;
-		bkey_for_each_ptr_decode(k.k, ptrs, p, entry)
-			if (p.ptr.dev == arg->migrate.dev) {
-				if (!p.crc.csum_type)
-					return false;
-				break;
-			}
-	}
-
-	data_opts->scrub	= true;
-	data_opts->read_dev	= arg->migrate.dev;
-	return true;
-}
-
 int bch2_data_job(struct bch_fs *c,
 		  struct bch_move_stats *stats,
 		  struct bch_ioctl_data op)
@@ -1209,16 +1117,6 @@ int bch2_data_job(struct bch_fs *c,
 	bch2_move_stats_init(stats, bch2_data_ops_strs[op.op]);
 
 	switch (op.op) {
-	case BCH_DATA_OP_scrub:
-		ret = bch2_move_data_phys(c, op.scrub.dev, 0, U64_MAX,
-					  op.scrub.data_types,
-					  NULL,
-					  stats,
-					  writepoint_hashed((unsigned long) current),
-					  false,
-					  scrub_pred, &op) ?: ret;
-		break;
-
 	case BCH_DATA_OP_rereplicate:
 		stats->data_type = BCH_DATA_journal;
 		ret = bch2_journal_flush_device_pins(&c->journal, -1);
@@ -1317,7 +1215,7 @@ static void bch2_moving_ctxt_to_text(struct printbuf *out, struct bch_fs *c, str
 
 	mutex_lock(&ctxt->lock);
 	list_for_each_entry(io, &ctxt->ios, io_list)
-		bch2_write_op_to_text(out, &io->write.op);
+		bch2_data_update_to_text(out, &io->write);
 	mutex_unlock(&ctxt->lock);
 
 	printbuf_indent_sub(out, 4);
diff --git a/libbcachefs/move_types.h b/libbcachefs/move_types.h
index 82e473ed..e22841ef 100644
--- a/libbcachefs/move_types.h
+++ b/libbcachefs/move_types.h
@@ -3,31 +3,17 @@
 #define _BCACHEFS_MOVE_TYPES_H
 
 #include "bbpos_types.h"
-#include "bcachefs_ioctl.h"
 
 struct bch_move_stats {
+	enum bch_data_type	data_type;
+	struct bbpos		pos;
 	char			name[32];
-	bool			phys;
-	enum bch_ioctl_data_event_ret	ret;
-
-	union {
-	struct {
-		enum bch_data_type	data_type;
-		struct bbpos		pos;
-	};
-	struct {
-		unsigned		dev;
-		u64			offset;
-	};
-	};
 
 	atomic64_t		keys_moved;
 	atomic64_t		keys_raced;
 	atomic64_t		sectors_seen;
 	atomic64_t		sectors_moved;
 	atomic64_t		sectors_raced;
-	atomic64_t		sectors_error_corrected;
-	atomic64_t		sectors_error_uncorrected;
 };
 
 struct move_bucket_key {
diff --git a/libbcachefs/recovery_passes_types.h b/libbcachefs/recovery_passes_types.h
index 71baad41..41855796 100644
--- a/libbcachefs/recovery_passes_types.h
+++ b/libbcachefs/recovery_passes_types.h
@@ -53,7 +53,7 @@
 	x(check_dirents,			27, PASS_FSCK)				\
 	x(check_xattrs,				28, PASS_FSCK)				\
 	x(check_root,				29, PASS_ONLINE|PASS_FSCK)		\
-	x(check_unreachable_inodes,		40, PASS_ONLINE|PASS_FSCK)		\
+	x(check_unreachable_inodes,		40, PASS_FSCK)				\
 	x(check_subvolume_structure,		36, PASS_ONLINE|PASS_FSCK)		\
 	x(check_directory_structure,		30, PASS_ONLINE|PASS_FSCK)		\
 	x(check_nlinks,				31, PASS_FSCK)				\
diff --git a/libbcachefs/sb-downgrade.c b/libbcachefs/sb-downgrade.c
index 051214fd..14f6b6a5 100644
--- a/libbcachefs/sb-downgrade.c
+++ b/libbcachefs/sb-downgrade.c
@@ -90,7 +90,10 @@
 	  BIT_ULL(BCH_RECOVERY_PASS_check_allocations),		\
 	  BCH_FSCK_ERR_accounting_mismatch,			\
 	  BCH_FSCK_ERR_accounting_key_replicas_nr_devs_0,	\
-	  BCH_FSCK_ERR_accounting_key_junk_at_end)
+	  BCH_FSCK_ERR_accounting_key_junk_at_end)		\
+	x(directory_size,					\
+	  BIT_ULL(BCH_RECOVERY_PASS_check_inodes),		\
+	  BCH_FSCK_ERR_directory_size_mismatch)			\
 
 #define DOWNGRADE_TABLE()					\
 	x(bucket_stripe_sectors,				\
diff --git a/libbcachefs/sb-errors_format.h b/libbcachefs/sb-errors_format.h
index e26317c3..0b4fe899 100644
--- a/libbcachefs/sb-errors_format.h
+++ b/libbcachefs/sb-errors_format.h
@@ -256,6 +256,7 @@ enum bch_fsck_flags {
 	x(dirent_in_missing_dir_inode,				227,	0)		\
 	x(dirent_in_non_dir_inode,				228,	0)		\
 	x(dirent_to_missing_inode,				229,	0)		\
+	x(dirent_to_overwritten_inode,				302,	0)		\
 	x(dirent_to_missing_subvol,				230,	0)		\
 	x(dirent_to_itself,					231,	0)		\
 	x(quota_type_invalid,					232,	0)		\
@@ -312,7 +313,8 @@ enum bch_fsck_flags {
 	x(logged_op_but_clean,					283,	FSCK_AUTOFIX)	\
 	x(compression_opt_not_marked_in_sb,			295,	FSCK_AUTOFIX)	\
 	x(compression_type_not_marked_in_sb,			296,	FSCK_AUTOFIX)	\
-	x(MAX,							302,	0)
+	x(directory_size_mismatch,				303,	FSCK_AUTOFIX)	\
+	x(MAX,							304,	0)
 
 enum bch_sb_error_id {
 #define x(t, n, ...) BCH_FSCK_ERR_##t = n,
diff --git a/libbcachefs/sb-members.h b/libbcachefs/sb-members.h
index b29b6c6c..762083b5 100644
--- a/libbcachefs/sb-members.h
+++ b/libbcachefs/sb-members.h
@@ -23,18 +23,6 @@ static inline bool bch2_dev_is_online(struct bch_dev *ca)
 	return !percpu_ref_is_zero(&ca->io_ref);
 }
 
-static inline struct bch_dev *bch2_dev_rcu(struct bch_fs *, unsigned);
-
-static inline bool bch2_dev_idx_is_online(struct bch_fs *c, unsigned dev)
-{
-	rcu_read_lock();
-	struct bch_dev *ca = bch2_dev_rcu(c, dev);
-	bool ret = ca && bch2_dev_is_online(ca);
-	rcu_read_unlock();
-
-	return ret;
-}
-
 static inline bool bch2_dev_is_readable(struct bch_dev *ca)
 {
 	return bch2_dev_is_online(ca) &&
diff --git a/libbcachefs/six.c b/libbcachefs/six.c
index 537bf049..7e7c66a1 100644
--- a/libbcachefs/six.c
+++ b/libbcachefs/six.c
@@ -491,8 +491,12 @@ static int six_lock_slowpath(struct six_lock *lock, enum six_lock_type type,
 				list_del(&wait->list);
 			raw_spin_unlock(&lock->wait_lock);
 
-			if (unlikely(acquired))
+			if (unlikely(acquired)) {
 				do_six_unlock_type(lock, type);
+			} else if (type == SIX_LOCK_write) {
+				six_clear_bitmask(lock, SIX_LOCK_HELD_write);
+				six_lock_wakeup(lock, atomic_read(&lock->state), SIX_LOCK_read);
+			}
 			break;
 		}
 
@@ -501,10 +505,6 @@ static int six_lock_slowpath(struct six_lock *lock, enum six_lock_type type,
 
 	__set_current_state(TASK_RUNNING);
 out:
-	if (ret && type == SIX_LOCK_write) {
-		six_clear_bitmask(lock, SIX_LOCK_HELD_write);
-		six_lock_wakeup(lock, atomic_read(&lock->state), SIX_LOCK_read);
-	}
 	trace_contention_end(lock, 0);
 
 	return ret;
diff --git a/libbcachefs/snapshot.c b/libbcachefs/snapshot.c
index cf6b3256..c54091a2 100644
--- a/libbcachefs/snapshot.c
+++ b/libbcachefs/snapshot.c
@@ -1563,7 +1563,8 @@ int bch2_delete_dead_snapshots(struct bch_fs *c)
 	 */
 	ret = for_each_btree_key(trans, iter, BTREE_ID_snapshots, POS_MIN, 0, k,
 		check_should_delete_snapshot(trans, k, &delete_leaves, &delete_interior));
-	bch_err_msg(c, ret, "walking snapshots");
+	if (!bch2_err_matches(ret, EROFS))
+		bch_err_msg(c, ret, "walking snapshots");
 	if (ret)
 		goto err;
 
@@ -1602,7 +1603,8 @@ int bch2_delete_dead_snapshots(struct bch_fs *c)
 
 		bch2_disk_reservation_put(c, &res);
 
-		bch_err_msg(c, ret, "deleting keys from dying snapshots");
+		if (!bch2_err_matches(ret, EROFS))
+			bch_err_msg(c, ret, "deleting keys from dying snapshots");
 		if (ret)
 			goto err;
 	}
@@ -1610,7 +1612,8 @@ int bch2_delete_dead_snapshots(struct bch_fs *c)
 	darray_for_each(delete_leaves, i) {
 		ret = commit_do(trans, NULL, NULL, 0,
 			bch2_snapshot_node_delete(trans, *i));
-		bch_err_msg(c, ret, "deleting snapshot %u", *i);
+		if (!bch2_err_matches(ret, EROFS))
+			bch_err_msg(c, ret, "deleting snapshot %u", *i);
 		if (ret)
 			goto err;
 	}
@@ -1630,7 +1633,8 @@ int bch2_delete_dead_snapshots(struct bch_fs *c)
 	darray_for_each(delete_interior, i) {
 		ret = commit_do(trans, NULL, NULL, 0,
 			bch2_snapshot_node_delete(trans, i->id));
-		bch_err_msg(c, ret, "deleting snapshot %u", i->id);
+		if (!bch2_err_matches(ret, EROFS))
+			bch_err_msg(c, ret, "deleting snapshot %u", i->id);
 		if (ret)
 			goto err;
 	}
@@ -1638,7 +1642,8 @@ err:
 	darray_exit(&delete_interior);
 	darray_exit(&delete_leaves);
 	bch2_trans_put(trans);
-	bch_err_fn(c, ret);
+	if (!bch2_err_matches(ret, EROFS))
+		bch_err_fn(c, ret);
 	return ret;
 }
 
diff --git a/libbcachefs/str_hash.c b/libbcachefs/str_hash.c
index f5977c5c..8c2c5539 100644
--- a/libbcachefs/str_hash.c
+++ b/libbcachefs/str_hash.c
@@ -167,10 +167,19 @@ found:;
 		goto err;
 
 	struct bch_hash_info hash2 = bch2_hash_info_init(c, &inode);
-	if (memcmp(hash_info, &hash2, sizeof(hash2))) {
+	if (hash_info->type != hash2.type ||
+	    memcmp(&hash_info->siphash_key, &hash2.siphash_key, sizeof(hash2.siphash_key))) {
 		ret = repair_inode_hash_info(trans, &inode);
 		if (!ret) {
-			bch_err(c, "inode hash info mismatch with root, but mismatch not found");
+			bch_err(c, "inode hash info mismatch with root, but mismatch not found\n"
+				"%u %llx %llx\n"
+				"%u %llx %llx",
+				hash_info->type,
+				hash_info->siphash_key.k0,
+				hash_info->siphash_key.k1,
+				hash2.type,
+				hash2.siphash_key.k0,
+				hash2.siphash_key.k1);
 			ret = -BCH_ERR_fsck_repair_unimplemented;
 		}
 	}
diff --git a/libbcachefs/trace.h b/libbcachefs/trace.h
index e4c7dc0e..9d40b7d4 100644
--- a/libbcachefs/trace.h
+++ b/libbcachefs/trace.h
@@ -785,6 +785,27 @@ TRACE_EVENT(bucket_invalidate,
 
 /* Moving IO */
 
+TRACE_EVENT(bucket_evacuate,
+	TP_PROTO(struct bch_fs *c, struct bpos *bucket),
+	TP_ARGS(c, bucket),
+
+	TP_STRUCT__entry(
+		__field(dev_t,		dev			)
+		__field(u32,		dev_idx			)
+		__field(u64,		bucket			)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= c->dev;
+		__entry->dev_idx	= bucket->inode;
+		__entry->bucket		= bucket->offset;
+	),
+
+	TP_printk("%d:%d %u:%llu",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->dev_idx, __entry->bucket)
+);
+
 DEFINE_EVENT(fs_str, move_extent,
 	TP_PROTO(struct bch_fs *c, const char *str),
 	TP_ARGS(c, str)
@@ -848,6 +869,37 @@ TRACE_EVENT(move_data,
 		  __entry->sectors_raced)
 );
 
+TRACE_EVENT(evacuate_bucket,
+	TP_PROTO(struct bch_fs *c, struct bpos *bucket,
+		 unsigned sectors, unsigned bucket_size,
+		 int ret),
+	TP_ARGS(c, bucket, sectors, bucket_size, ret),
+
+	TP_STRUCT__entry(
+		__field(dev_t,		dev		)
+		__field(u64,		member		)
+		__field(u64,		bucket		)
+		__field(u32,		sectors		)
+		__field(u32,		bucket_size	)
+		__field(int,		ret		)
+	),
+
+	TP_fast_assign(
+		__entry->dev			= c->dev;
+		__entry->member			= bucket->inode;
+		__entry->bucket			= bucket->offset;
+		__entry->sectors		= sectors;
+		__entry->bucket_size		= bucket_size;
+		__entry->ret			= ret;
+	),
+
+	TP_printk("%d,%d %llu:%llu sectors %u/%u ret %i",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->member, __entry->bucket,
+		  __entry->sectors, __entry->bucket_size,
+		  __entry->ret)
+);
+
 TRACE_EVENT(copygc,
 	TP_PROTO(struct bch_fs *c,
 		 u64 sectors_moved, u64 sectors_not_moved,