Update bcachefs sources to 6d44812757dd bcachefs: BCH_IOCTL_FSCK_ONLINE

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>

28 files changed, 787 insertions, 404 deletions

diff --git a/libbcachefs/bcachefs.h b/libbcachefs/bcachefs.h
index bb2a0cc4..66de8c0c 100644
--- a/libbcachefs/bcachefs.h
+++ b/libbcachefs/bcachefs.h
@@ -264,36 +264,54 @@ do {									\
 
 #define bch2_fmt(_c, fmt)		bch2_log_msg(_c, fmt "\n")
 
+void __bch2_print(struct bch_fs *c, const char *fmt, ...);
+
+#define maybe_dev_to_fs(_c)	_Generic((_c),				\
+	struct bch_dev *:	((struct bch_dev *) (_c))->fs,		\
+	struct bch_fs *:	(_c))
+
+#define bch2_print(_c, ...) __bch2_print(maybe_dev_to_fs(_c), __VA_ARGS__)
+
+#define bch2_print_ratelimited(_c, ...)					\
+do {									\
+	static DEFINE_RATELIMIT_STATE(_rs,				\
+				      DEFAULT_RATELIMIT_INTERVAL,	\
+				      DEFAULT_RATELIMIT_BURST);		\
+									\
+	if (__ratelimit(&_rs))						\
+		bch2_print(_c, __VA_ARGS__);				\
+} while (0)
+
 #define bch_info(c, fmt, ...) \
-	printk(KERN_INFO bch2_fmt(c, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_INFO bch2_fmt(c, fmt), ##__VA_ARGS__)
 #define bch_notice(c, fmt, ...) \
-	printk(KERN_NOTICE bch2_fmt(c, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_NOTICE bch2_fmt(c, fmt), ##__VA_ARGS__)
 #define bch_warn(c, fmt, ...) \
-	printk(KERN_WARNING bch2_fmt(c, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_WARNING bch2_fmt(c, fmt), ##__VA_ARGS__)
 #define bch_warn_ratelimited(c, fmt, ...) \
-	printk_ratelimited(KERN_WARNING bch2_fmt(c, fmt), ##__VA_ARGS__)
+	bch2_print_ratelimited(c, KERN_WARNING bch2_fmt(c, fmt), ##__VA_ARGS__)
 
 #define bch_err(c, fmt, ...) \
-	printk(KERN_ERR bch2_fmt(c, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_ERR bch2_fmt(c, fmt), ##__VA_ARGS__)
 #define bch_err_dev(ca, fmt, ...) \
-	printk(KERN_ERR bch2_fmt_dev(ca, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_ERR bch2_fmt_dev(ca, fmt), ##__VA_ARGS__)
 #define bch_err_dev_offset(ca, _offset, fmt, ...) \
-	printk(KERN_ERR bch2_fmt_dev_offset(ca, _offset, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_ERR bch2_fmt_dev_offset(ca, _offset, fmt), ##__VA_ARGS__)
 #define bch_err_inum(c, _inum, fmt, ...) \
-	printk(KERN_ERR bch2_fmt_inum(c, _inum, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_ERR bch2_fmt_inum(c, _inum, fmt), ##__VA_ARGS__)
 #define bch_err_inum_offset(c, _inum, _offset, fmt, ...) \
-	printk(KERN_ERR bch2_fmt_inum_offset(c, _inum, _offset, fmt), ##__VA_ARGS__)
+	bch2_print(c, KERN_ERR bch2_fmt_inum_offset(c, _inum, _offset, fmt), ##__VA_ARGS__)
 
 #define bch_err_ratelimited(c, fmt, ...) \
-	printk_ratelimited(KERN_ERR bch2_fmt(c, fmt), ##__VA_ARGS__)
+	bch2_print_ratelimited(c, KERN_ERR bch2_fmt(c, fmt), ##__VA_ARGS__)
 #define bch_err_dev_ratelimited(ca, fmt, ...) \
-	printk_ratelimited(KERN_ERR bch2_fmt_dev(ca, fmt), ##__VA_ARGS__)
+	bch2_print_ratelimited(ca, KERN_ERR bch2_fmt_dev(ca, fmt), ##__VA_ARGS__)
 #define bch_err_dev_offset_ratelimited(ca, _offset, fmt, ...) \
-	printk_ratelimited(KERN_ERR bch2_fmt_dev_offset(ca, _offset, fmt), ##__VA_ARGS__)
+	bch2_print_ratelimited(ca, KERN_ERR bch2_fmt_dev_offset(ca, _offset, fmt), ##__VA_ARGS__)
 #define bch_err_inum_ratelimited(c, _inum, fmt, ...) \
-	printk_ratelimited(KERN_ERR bch2_fmt_inum(c, _inum, fmt), ##__VA_ARGS__)
+	bch2_print_ratelimited(c, KERN_ERR bch2_fmt_inum(c, _inum, fmt), ##__VA_ARGS__)
 #define bch_err_inum_offset_ratelimited(c, _inum, _offset, fmt, ...) \
-	printk_ratelimited(KERN_ERR bch2_fmt_inum_offset(c, _inum, _offset, fmt), ##__VA_ARGS__)
+	bch2_print_ratelimited(c, KERN_ERR bch2_fmt_inum_offset(c, _inum, _offset, fmt), ##__VA_ARGS__)
 
 #define bch_err_fn(_c, _ret)						\
 do {									\
@@ -446,6 +464,12 @@ enum bch_time_stats {
 
 struct btree;
 
+struct log_output {
+	spinlock_t		lock;
+	wait_queue_head_t	wait;
+	struct printbuf		buf;
+};
+
 enum gc_phase {
 	GC_PHASE_NOT_RUNNING,
 	GC_PHASE_START,
@@ -700,6 +724,7 @@ struct bch_fs {
 	struct super_block	*vfs_sb;
 	dev_t			dev;
 	char			name[40];
+	struct log_output	*output;
 
 	/* ro/rw, add/remove/resize devices: */
 	struct rw_semaphore	state_lock;
diff --git a/libbcachefs/bcachefs_ioctl.h b/libbcachefs/bcachefs_ioctl.h
index 43822c17..2ac6272c 100644
--- a/libbcachefs/bcachefs_ioctl.h
+++ b/libbcachefs/bcachefs_ioctl.h
@@ -83,6 +83,10 @@ struct bch_ioctl_incremental {
 
 #define BCH_IOCTL_DEV_USAGE_V2	_IOWR(0xbc,	18, struct bch_ioctl_dev_usage_v2)
 
+#define BCH_IOCTL_FSCK_OFFLINE		_IOW(0xbc,	19,  struct bch_ioctl_fsck_offline)
+
+#define BCH_IOCTL_FSCK_ONLINE		_IOW(0xbc,	20,  struct bch_ioctl_fsck_online)
+
 /* ioctl below act on a particular file, not the filesystem as a whole: */
 
 #define BCHFS_IOC_REINHERIT_ATTRS	_IOR(0xbc, 64, const char __user *)
@@ -386,4 +390,24 @@ struct bch_ioctl_subvolume {
 #define BCH_SUBVOL_SNAPSHOT_CREATE	(1U << 0)
 #define BCH_SUBVOL_SNAPSHOT_RO		(1U << 1)
 
+/*
+ * BCH_IOCTL_FSCK_OFFLINE: run fsck from the 'bcachefs fsck' userspace command,
+ * but with the kernel's implementation of fsck:
+ */
+struct bch_ioctl_fsck_offline {
+	__u64			flags;
+	__u64			opts;		/* string */
+	__u64			nr_devs;
+	__u64			devs[0];
+};
+
+/*
+ * BCH_IOCTL_FSCK_ONLINE: run fsck from the 'bcachefs fsck' userspace command,
+ * but with the kernel's implementation of fsck:
+ */
+struct bch_ioctl_fsck_online {
+	__u64			flags;
+	__u64			opts;		/* string */
+};
+
 #endif /* _BCACHEFS_IOCTL_H */
diff --git a/libbcachefs/btree_cache.c b/libbcachefs/btree_cache.c
index 47e7770d..9574c8c4 100644
--- a/libbcachefs/btree_cache.c
+++ b/libbcachefs/btree_cache.c
@@ -9,6 +9,7 @@
 #include "debug.h"
 #include "errcode.h"
 #include "error.h"
+#include "journal.h"
 #include "trace.h"
 
 #include <linux/prefetch.h>
@@ -424,14 +425,11 @@ void bch2_fs_btree_cache_exit(struct bch_fs *c)
 		BUG_ON(btree_node_read_in_flight(b) ||
 		       btree_node_write_in_flight(b));
 
-		if (btree_node_dirty(b))
-			bch2_btree_complete_write(c, b, btree_current_write(b));
-		clear_btree_node_dirty_acct(c, b);
-
 		btree_node_data_free(c, b);
 	}
 
-	BUG_ON(atomic_read(&c->btree_cache.dirty));
+	BUG_ON(!bch2_journal_error(&c->journal) &&
+	       atomic_read(&c->btree_cache.dirty));
 
 	list_splice(&bc->freed_pcpu, &bc->freed_nonpcpu);
 
@@ -502,19 +500,21 @@ void bch2_fs_btree_cache_init_early(struct btree_cache *bc)
  * cannibalize_bucket() will take. This means every time we unlock the root of
  * the btree, we need to release this lock if we have it held.
  */
-void bch2_btree_cache_cannibalize_unlock(struct bch_fs *c)
+void bch2_btree_cache_cannibalize_unlock(struct btree_trans *trans)
 {
+	struct bch_fs *c = trans->c;
 	struct btree_cache *bc = &c->btree_cache;
 
 	if (bc->alloc_lock == current) {
-		trace_and_count(c, btree_cache_cannibalize_unlock, c);
+		trace_and_count(c, btree_cache_cannibalize_unlock, trans);
 		bc->alloc_lock = NULL;
 		closure_wake_up(&bc->alloc_wait);
 	}
 }
 
-int bch2_btree_cache_cannibalize_lock(struct bch_fs *c, struct closure *cl)
+int bch2_btree_cache_cannibalize_lock(struct btree_trans *trans, struct closure *cl)
 {
+	struct bch_fs *c = trans->c;
 	struct btree_cache *bc = &c->btree_cache;
 	struct task_struct *old;
 
@@ -523,7 +523,7 @@ int bch2_btree_cache_cannibalize_lock(struct bch_fs *c, struct closure *cl)
 		goto success;
 
 	if (!cl) {
-		trace_and_count(c, btree_cache_cannibalize_lock_fail, c);
+		trace_and_count(c, btree_cache_cannibalize_lock_fail, trans);
 		return -BCH_ERR_ENOMEM_btree_cache_cannibalize_lock;
 	}
 
@@ -537,11 +537,11 @@ int bch2_btree_cache_cannibalize_lock(struct bch_fs *c, struct closure *cl)
 		goto success;
 	}
 
-	trace_and_count(c, btree_cache_cannibalize_lock_fail, c);
+	trace_and_count(c, btree_cache_cannibalize_lock_fail, trans);
 	return -BCH_ERR_btree_cache_cannibalize_lock_blocked;
 
 success:
-	trace_and_count(c, btree_cache_cannibalize_lock, c);
+	trace_and_count(c, btree_cache_cannibalize_lock, trans);
 	return 0;
 }
 
@@ -675,7 +675,7 @@ err:
 
 		mutex_unlock(&bc->lock);
 
-		trace_and_count(c, btree_cache_cannibalize, c);
+		trace_and_count(c, btree_cache_cannibalize, trans);
 		goto out;
 	}
 
@@ -751,7 +751,7 @@ static noinline struct btree *bch2_btree_node_fill(struct btree_trans *trans,
 	if (path && sync)
 		bch2_trans_unlock_noassert(trans);
 
-	bch2_btree_node_read(c, b, sync);
+	bch2_btree_node_read(trans, b, sync);
 
 	if (!sync)
 		return NULL;
@@ -1041,7 +1041,7 @@ retry:
 			goto retry;
 
 		if (IS_ERR(b) &&
-		    !bch2_btree_cache_cannibalize_lock(c, NULL))
+		    !bch2_btree_cache_cannibalize_lock(trans, NULL))
 			goto retry;
 
 		if (IS_ERR(b))
@@ -1089,7 +1089,7 @@ lock_node:
 	EBUG_ON(BTREE_NODE_LEVEL(b->data) != level);
 	btree_check_header(c, b);
 out:
-	bch2_btree_cache_cannibalize_unlock(c);
+	bch2_btree_cache_cannibalize_unlock(trans);
 	return b;
 }
 
diff --git a/libbcachefs/btree_cache.h b/libbcachefs/btree_cache.h
index cfb80b20..4e1af588 100644
--- a/libbcachefs/btree_cache.h
+++ b/libbcachefs/btree_cache.h
@@ -17,8 +17,8 @@ int __bch2_btree_node_hash_insert(struct btree_cache *, struct btree *);
 int bch2_btree_node_hash_insert(struct btree_cache *, struct btree *,
 				unsigned, enum btree_id);
 
-void bch2_btree_cache_cannibalize_unlock(struct bch_fs *);
-int bch2_btree_cache_cannibalize_lock(struct bch_fs *, struct closure *);
+void bch2_btree_cache_cannibalize_unlock(struct btree_trans *);
+int bch2_btree_cache_cannibalize_lock(struct btree_trans *, struct closure *);
 
 struct btree *__bch2_btree_node_mem_alloc(struct bch_fs *);
 struct btree *bch2_btree_node_mem_alloc(struct btree_trans *, bool);
diff --git a/libbcachefs/btree_io.c b/libbcachefs/btree_io.c
index 3c663c59..a6ac68fe 100644
--- a/libbcachefs/btree_io.c
+++ b/libbcachefs/btree_io.c
@@ -1575,16 +1575,17 @@ static int btree_node_read_all_replicas(struct bch_fs *c, struct btree *b, bool
 	return 0;
 }
 
-void bch2_btree_node_read(struct bch_fs *c, struct btree *b,
+void bch2_btree_node_read(struct btree_trans *trans, struct btree *b,
 			  bool sync)
 {
+	struct bch_fs *c = trans->c;
 	struct extent_ptr_decoded pick;
 	struct btree_read_bio *rb;
 	struct bch_dev *ca;
 	struct bio *bio;
 	int ret;
 
-	trace_and_count(c, btree_node_read, c, b);
+	trace_and_count(c, btree_node_read, trans, b);
 
 	if (bch2_verify_all_btree_replicas &&
 	    !btree_node_read_all_replicas(c, b, sync))
@@ -1663,12 +1664,12 @@ static int __bch2_btree_root_read(struct btree_trans *trans, enum btree_id id,
 	closure_init_stack(&cl);
 
 	do {
-		ret = bch2_btree_cache_cannibalize_lock(c, &cl);
+		ret = bch2_btree_cache_cannibalize_lock(trans, &cl);
 		closure_sync(&cl);
 	} while (ret);
 
 	b = bch2_btree_node_mem_alloc(trans, level != 0);
-	bch2_btree_cache_cannibalize_unlock(c);
+	bch2_btree_cache_cannibalize_unlock(trans);
 
 	BUG_ON(IS_ERR(b));
 
@@ -1677,7 +1678,7 @@ static int __bch2_btree_root_read(struct btree_trans *trans, enum btree_id id,
 
 	set_btree_node_read_in_flight(b);
 
-	bch2_btree_node_read(c, b, true);
+	bch2_btree_node_read(trans, b, true);
 
 	if (btree_node_read_error(b)) {
 		bch2_btree_node_hash_remove(&c->btree_cache, b);
@@ -1704,8 +1705,8 @@ 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));
 }
 
-void bch2_btree_complete_write(struct bch_fs *c, struct btree *b,
-			      struct btree_write *w)
+static void bch2_btree_complete_write(struct bch_fs *c, struct btree *b,
+				      struct btree_write *w)
 {
 	unsigned long old, new, v = READ_ONCE(b->will_make_reachable);
 
diff --git a/libbcachefs/btree_io.h b/libbcachefs/btree_io.h
index 7e03dd76..e251cb6b 100644
--- a/libbcachefs/btree_io.h
+++ b/libbcachefs/btree_io.h
@@ -130,13 +130,10 @@ void bch2_btree_init_next(struct btree_trans *, struct btree *);
 
 int bch2_btree_node_read_done(struct bch_fs *, struct bch_dev *,
 			      struct btree *, bool, bool *);
-void bch2_btree_node_read(struct bch_fs *, struct btree *, bool);
+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);
 
-void bch2_btree_complete_write(struct bch_fs *, struct btree *,
-			      struct btree_write *);
-
 bool bch2_btree_post_write_cleanup(struct bch_fs *, struct btree *);
 
 enum btree_write_flags {
diff --git a/libbcachefs/btree_iter.c b/libbcachefs/btree_iter.c
index 4d673d47..929f33df 100644
--- a/libbcachefs/btree_iter.c
+++ b/libbcachefs/btree_iter.c
@@ -977,7 +977,7 @@ retry_all:
 		closure_init_stack(&cl);
 
 		do {
-			ret = bch2_btree_cache_cannibalize_lock(c, &cl);
+			ret = bch2_btree_cache_cannibalize_lock(trans, &cl);
 			closure_sync(&cl);
 		} while (ret);
 	}
@@ -1013,7 +1013,7 @@ retry_all:
 	 * then failed to relock a path - that's fine.
 	 */
 err:
-	bch2_btree_cache_cannibalize_unlock(c);
+	bch2_btree_cache_cannibalize_unlock(trans);
 
 	trans->in_traverse_all = false;
 
@@ -1298,7 +1298,7 @@ static inline void __bch2_path_free(struct btree_trans *trans, struct btree_path
 {
 	__bch2_btree_path_unlock(trans, path);
 	btree_path_list_remove(trans, path);
-	trans->paths_allocated &= ~(1ULL << path->idx);
+	__clear_bit(path->idx, trans->paths_allocated);
 }
 
 void bch2_path_put(struct btree_trans *trans, struct btree_path *path, bool intent)
@@ -1471,6 +1471,7 @@ static void bch2_trans_update_max_paths(struct btree_trans *trans)
 {
 	struct btree_transaction_stats *s = btree_trans_stats(trans);
 	struct printbuf buf = PRINTBUF;
+	size_t nr = bitmap_weight(trans->paths_allocated, BTREE_ITER_MAX);
 
 	if (!s)
 		return;
@@ -1479,9 +1480,8 @@ static void bch2_trans_update_max_paths(struct btree_trans *trans)
 
 	if (!buf.allocation_failure) {
 		mutex_lock(&s->lock);
-		if (s->nr_max_paths < hweight64(trans->paths_allocated)) {
-			s->nr_max_paths = trans->nr_max_paths =
-				hweight64(trans->paths_allocated);
+		if (nr > s->nr_max_paths) {
+			s->nr_max_paths = nr;
 			swap(s->max_paths_text, buf.buf);
 		}
 		mutex_unlock(&s->lock);
@@ -1489,7 +1489,7 @@ static void bch2_trans_update_max_paths(struct btree_trans *trans)
 
 	printbuf_exit(&buf);
 
-	trans->nr_max_paths = hweight64(trans->paths_allocated);
+	trans->nr_max_paths = nr;
 }
 
 noinline __cold
@@ -1518,13 +1518,12 @@ static inline struct btree_path *btree_path_alloc(struct btree_trans *trans,
 						  struct btree_path *pos)
 {
 	struct btree_path *path;
-	unsigned idx;
+	size_t idx = find_first_zero_bit(trans->paths_allocated, BTREE_ITER_MAX);
 
-	if (unlikely(trans->paths_allocated ==
-		     ~((~0ULL << 1) << (BTREE_ITER_MAX - 1))))
+	if (unlikely(idx == BTREE_ITER_MAX))
 		btree_path_overflow(trans);
 
-	idx = __ffs64(~trans->paths_allocated);
+	BUG_ON(idx > BTREE_ITER_MAX);
 
 	/*
 	 * Do this before marking the new path as allocated, since it won't be
@@ -1533,7 +1532,7 @@ static inline struct btree_path *btree_path_alloc(struct btree_trans *trans,
 	if (unlikely(idx > trans->nr_max_paths))
 		bch2_trans_update_max_paths(trans);
 
-	trans->paths_allocated |= 1ULL << idx;
+	__set_bit(idx, trans->paths_allocated);
 
 	path = &trans->paths[idx];
 	path->idx		= idx;
@@ -2503,7 +2502,7 @@ static void btree_trans_verify_sorted_refs(struct btree_trans *trans)
 	struct btree_path *path;
 	unsigned i;
 
-	BUG_ON(trans->nr_sorted != hweight64(trans->paths_allocated));
+	BUG_ON(trans->nr_sorted != bitmap_weight(trans->paths_allocated, BTREE_ITER_MAX));
 
 	trans_for_each_path(trans, path) {
 		BUG_ON(path->sorted_idx >= trans->nr_sorted);
@@ -2513,7 +2512,7 @@ static void btree_trans_verify_sorted_refs(struct btree_trans *trans)
 	for (i = 0; i < trans->nr_sorted; i++) {
 		unsigned idx = trans->sorted[i];
 
-		EBUG_ON(!(trans->paths_allocated & (1ULL << idx)));
+		BUG_ON(!test_bit(idx, trans->paths_allocated));
 		BUG_ON(trans->paths[idx].sorted_idx != i);
 	}
 }
diff --git a/libbcachefs/btree_iter.h b/libbcachefs/btree_iter.h
index 75beb183..ea4fc8a2 100644
--- a/libbcachefs/btree_iter.h
+++ b/libbcachefs/btree_iter.h
@@ -66,17 +66,10 @@ static inline void btree_trans_sort_paths(struct btree_trans *trans)
 static inline struct btree_path *
 __trans_next_path(struct btree_trans *trans, unsigned idx)
 {
-	u64 l;
-
+	idx = find_next_bit(trans->paths_allocated, BTREE_ITER_MAX, idx);
 	if (idx == BTREE_ITER_MAX)
 		return NULL;
-
-	l = trans->paths_allocated >> idx;
-	if (!l)
-		return NULL;
-
-	idx += __ffs64(l);
-	EBUG_ON(idx >= BTREE_ITER_MAX);
+	EBUG_ON(idx > BTREE_ITER_MAX);
 	EBUG_ON(trans->paths[idx].idx != idx);
 	return &trans->paths[idx];
 }
@@ -92,17 +85,11 @@ __trans_next_path(struct btree_trans *trans, unsigned idx)
 static inline struct btree_path *
 __trans_next_path_safe(struct btree_trans *trans, unsigned *idx)
 {
-	u64 l;
-
+	*idx = find_next_bit(trans->paths_allocated, BTREE_ITER_MAX, *idx);
 	if (*idx == BTREE_ITER_MAX)
 		return NULL;
 
-	l = trans->paths_allocated >> *idx;
-	if (!l)
-		return NULL;
-
-	*idx += __ffs64(l);
-	EBUG_ON(*idx >= BTREE_ITER_MAX);
+	EBUG_ON(*idx > BTREE_ITER_MAX);
 	return &trans->paths[*idx];
 }
 
@@ -631,7 +618,7 @@ int __bch2_btree_trans_too_many_iters(struct btree_trans *);
 
 static inline int btree_trans_too_many_iters(struct btree_trans *trans)
 {
-	if (hweight64(trans->paths_allocated) > BTREE_ITER_MAX - 8)
+	if (bitmap_weight(trans->paths_allocated, BTREE_ITER_MAX) > BTREE_ITER_MAX - 8)
 		return __bch2_btree_trans_too_many_iters(trans);
 
 	return 0;
diff --git a/libbcachefs/btree_key_cache.c b/libbcachefs/btree_key_cache.c
index c5e8a461..b39b28b4 100644
--- a/libbcachefs/btree_key_cache.c
+++ b/libbcachefs/btree_key_cache.c
@@ -997,8 +997,6 @@ void bch2_fs_btree_key_cache_exit(struct btree_key_cache *bc)
 	list_for_each_entry_safe(ck, n, &items, list) {
 		cond_resched();
 
-		bch2_journal_pin_drop(&c->journal, &ck->journal);
-
 		list_del(&ck->list);
 		kfree(ck->k);
 		six_lock_exit(&ck->c.lock);
diff --git a/libbcachefs/btree_locking.h b/libbcachefs/btree_locking.h
index 11b0a2c8..a49f1dd1 100644
--- a/libbcachefs/btree_locking.h
+++ b/libbcachefs/btree_locking.h
@@ -263,7 +263,7 @@ static inline int btree_node_lock(struct btree_trans *trans,
 	int ret = 0;
 
 	EBUG_ON(level >= BTREE_MAX_DEPTH);
-	EBUG_ON(!(trans->paths_allocated & (1ULL << path->idx)));
+	EBUG_ON(!test_bit(path->idx, trans->paths_allocated));
 
 	if (likely(six_trylock_type(&b->lock, type)) ||
 	    btree_node_lock_increment(trans, b, level, (enum btree_node_locked_type) type) ||
diff --git a/libbcachefs/btree_types.h b/libbcachefs/btree_types.h
index ca752660..78d9f585 100644
--- a/libbcachefs/btree_types.h
+++ b/libbcachefs/btree_types.h
@@ -414,7 +414,7 @@ struct btree_trans {
 	unsigned		extra_journal_res;
 	unsigned		nr_max_paths;
 
-	u64			paths_allocated;
+	unsigned long		paths_allocated[BITS_TO_LONGS(BTREE_ITER_MAX)];
 
 	unsigned		mem_top;
 	unsigned		mem_max;
diff --git a/libbcachefs/btree_update.c b/libbcachefs/btree_update.c
index ba42f578..254794c1 100644
--- a/libbcachefs/btree_update.c
+++ b/libbcachefs/btree_update.c
@@ -531,6 +531,19 @@ int __must_check bch2_trans_update(struct btree_trans *trans, struct btree_iter
 	return bch2_trans_update_by_path(trans, path, k, flags, _RET_IP_);
 }
 
+static noinline int bch2_btree_insert_clone_trans(struct btree_trans *trans,
+						  enum btree_id btree,
+						  struct bkey_i *k)
+{
+	struct bkey_i *n = bch2_trans_kmalloc(trans, bkey_bytes(&k->k));
+	int ret = PTR_ERR_OR_ZERO(n);
+	if (ret)
+		return ret;
+
+	bkey_copy(n, k);
+	return bch2_btree_insert_trans(trans, btree, n, 0);
+}
+
 int __must_check bch2_trans_update_buffered(struct btree_trans *trans,
 					    enum btree_id btree,
 					    struct bkey_i *k)
@@ -541,6 +554,9 @@ int __must_check bch2_trans_update_buffered(struct btree_trans *trans,
 	EBUG_ON(trans->nr_wb_updates > trans->wb_updates_size);
 	EBUG_ON(k->k.u64s > BTREE_WRITE_BUFERED_U64s_MAX);
 
+	if (unlikely(trans->journal_replay_not_finished))
+		return bch2_btree_insert_clone_trans(trans, btree, k);
+
 	trans_for_each_wb_update(trans, i) {
 		if (i->btree == btree && bpos_eq(i->k.k.p, k->k.p)) {
 			bkey_copy(&i->k, k);
diff --git a/libbcachefs/btree_update_interior.c b/libbcachefs/btree_update_interior.c
index c9f07ca4..970faec1 100644
--- a/libbcachefs/btree_update_interior.c
+++ b/libbcachefs/btree_update_interior.c
@@ -163,9 +163,11 @@ bool bch2_btree_node_format_fits(struct bch_fs *c, struct btree *b,
 
 /* Btree node freeing/allocation: */
 
-static void __btree_node_free(struct bch_fs *c, struct btree *b)
+static void __btree_node_free(struct btree_trans *trans, struct btree *b)
 {
-	trace_and_count(c, btree_node_free, c, b);
+	struct bch_fs *c = trans->c;
+
+	trace_and_count(c, btree_node_free, trans, b);
 
 	BUG_ON(btree_node_write_blocked(b));
 	BUG_ON(btree_node_dirty(b));
@@ -191,7 +193,7 @@ static void bch2_btree_node_free_inmem(struct btree_trans *trans,
 
 	bch2_btree_node_lock_write_nofail(trans, path, &b->c);
 	bch2_btree_node_hash_remove(&c->btree_cache, b);
-	__btree_node_free(c, b);
+	__btree_node_free(trans, b);
 	six_unlock_write(&b->c.lock);
 	mark_btree_node_locked_noreset(path, level, BTREE_NODE_INTENT_LOCKED);
 
@@ -362,7 +364,7 @@ static struct btree *bch2_btree_node_alloc(struct btree_update *as,
 	ret = bch2_btree_node_hash_insert(&c->btree_cache, b, level, as->btree_id);
 	BUG_ON(ret);
 
-	trace_and_count(c, btree_node_alloc, c, b);
+	trace_and_count(c, btree_node_alloc, trans, b);
 	bch2_increment_clock(c, btree_sectors(c), WRITE);
 	return b;
 }
@@ -452,7 +454,7 @@ static void bch2_btree_reserve_put(struct btree_update *as, struct btree_trans *
 
 			btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_intent);
 			btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_write);
-			__btree_node_free(c, b);
+			__btree_node_free(trans, b);
 			six_unlock_write(&b->c.lock);
 			six_unlock_intent(&b->c.lock);
 		}
@@ -465,7 +467,6 @@ static int bch2_btree_reserve_get(struct btree_trans *trans,
 				  unsigned flags,
 				  struct closure *cl)
 {
-	struct bch_fs *c = as->c;
 	struct btree *b;
 	unsigned interior;
 	int ret = 0;
@@ -476,7 +477,7 @@ static int bch2_btree_reserve_get(struct btree_trans *trans,
 	 * Protects reaping from the btree node cache and using the btree node
 	 * open bucket reserve:
 	 */
-	ret = bch2_btree_cache_cannibalize_lock(c, cl);
+	ret = bch2_btree_cache_cannibalize_lock(trans, cl);
 	if (ret)
 		return ret;
 
@@ -495,7 +496,7 @@ static int bch2_btree_reserve_get(struct btree_trans *trans,
 		}
 	}
 err:
-	bch2_btree_cache_cannibalize_unlock(c);
+	bch2_btree_cache_cannibalize_unlock(trans);
 	return ret;
 }
 
@@ -1067,6 +1068,17 @@ bch2_btree_update_start(struct btree_trans *trans, struct btree_path *path,
 	flags &= ~BCH_WATERMARK_MASK;
 	flags |= watermark;
 
+	if (!(flags & BCH_TRANS_COMMIT_journal_reclaim) &&
+	    watermark < c->journal.watermark) {
+		struct journal_res res = { 0 };
+
+		ret = drop_locks_do(trans,
+			bch2_journal_res_get(&c->journal, &res, 1,
+					     watermark|JOURNAL_RES_GET_CHECK));
+		if (ret)
+			return ERR_PTR(ret);
+	}
+
 	while (1) {
 		nr_nodes[!!update_level] += 1 + split;
 		update_level++;
@@ -1211,7 +1223,7 @@ static void bch2_btree_set_root(struct btree_update *as,
 	struct bch_fs *c = as->c;
 	struct btree *old;
 
-	trace_and_count(c, btree_node_set_root, c, b);
+	trace_and_count(c, btree_node_set_root, trans, b);
 
 	old = btree_node_root(c, b);
 
@@ -1465,7 +1477,7 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans,
 	if (b->nr.live_u64s > BTREE_SPLIT_THRESHOLD(c)) {
 		struct btree *n[2];
 
-		trace_and_count(c, btree_node_split, c, b);
+		trace_and_count(c, btree_node_split, trans, b);
 
 		n[0] = n1 = bch2_btree_node_alloc(as, trans, b->c.level);
 		n[1] = n2 = bch2_btree_node_alloc(as, trans, b->c.level);
@@ -1523,7 +1535,7 @@ static int btree_split(struct btree_update *as, struct btree_trans *trans,
 			btree_split_insert_keys(as, trans, path, n3, &as->parent_keys);
 		}
 	} else {
-		trace_and_count(c, btree_node_compact, c, b);
+		trace_and_count(c, btree_node_compact, trans, b);
 
 		n1 = bch2_btree_node_alloc_replacement(as, trans, b);
 
@@ -1843,7 +1855,7 @@ int __bch2_foreground_maybe_merge(struct btree_trans *trans,
 	if (ret)
 		goto err;
 
-	trace_and_count(c, btree_node_merge, c, b);
+	trace_and_count(c, btree_node_merge, trans, b);
 
 	bch2_btree_interior_update_will_free_node(as, b);
 	bch2_btree_interior_update_will_free_node(as, m);
@@ -1946,7 +1958,7 @@ int bch2_btree_node_rewrite(struct btree_trans *trans,
 	mark_btree_node_locked(trans, new_path, n->c.level, BTREE_NODE_INTENT_LOCKED);
 	bch2_btree_path_level_init(trans, new_path, n);
 
-	trace_and_count(c, btree_node_rewrite, c, b);
+	trace_and_count(c, btree_node_rewrite, trans, b);
 
 	if (parent) {
 		bch2_keylist_add(&as->parent_keys, &n->key);
@@ -2228,7 +2240,7 @@ int bch2_btree_node_update_key(struct btree_trans *trans, struct btree_iter *ite
 	 * btree_iter_traverse():
 	 */
 	if (btree_ptr_hash_val(new_key) != b->hash_val) {
-		ret = bch2_btree_cache_cannibalize_lock(c, &cl);
+		ret = bch2_btree_cache_cannibalize_lock(trans, &cl);
 		if (ret) {
 			ret = drop_locks_do(trans, (closure_sync(&cl), 0));
 			if (ret)
@@ -2252,7 +2264,7 @@ int bch2_btree_node_update_key(struct btree_trans *trans, struct btree_iter *ite
 		six_unlock_intent(&new_hash->c.lock);
 	}
 	closure_sync(&cl);
-	bch2_btree_cache_cannibalize_unlock(c);
+	bch2_btree_cache_cannibalize_unlock(trans);
 	return ret;
 }
 
@@ -2313,12 +2325,12 @@ static int __bch2_btree_root_alloc(struct btree_trans *trans, enum btree_id id)
 	closure_init_stack(&cl);
 
 	do {
-		ret = bch2_btree_cache_cannibalize_lock(c, &cl);
+		ret = bch2_btree_cache_cannibalize_lock(trans, &cl);
 		closure_sync(&cl);
 	} while (ret);
 
 	b = bch2_btree_node_mem_alloc(trans, false);
-	bch2_btree_cache_cannibalize_unlock(c);
+	bch2_btree_cache_cannibalize_unlock(trans);
 
 	set_btree_node_fake(b);
 	set_btree_node_need_rewrite(b);
diff --git a/libbcachefs/buckets.c b/libbcachefs/buckets.c
index 312bd0c8..27c74388 100644
--- a/libbcachefs/buckets.c
+++ b/libbcachefs/buckets.c
@@ -1164,107 +1164,6 @@ int bch2_mark_reservation(struct btree_trans *trans,
 	return mem_trigger_run_overwrite_then_insert(__mark_reservation, trans, btree_id, level, old, new, flags);
 }
 
-static s64 __bch2_mark_reflink_p(struct btree_trans *trans,
-				 struct bkey_s_c_reflink_p p,
-				 u64 start, u64 end,
-				 u64 *idx, unsigned flags, size_t r_idx)
-{
-	struct bch_fs *c = trans->c;
-	struct reflink_gc *r;
-	int add = !(flags & BTREE_TRIGGER_OVERWRITE) ? 1 : -1;
-	u64 next_idx = end;
-	s64 ret = 0;
-	struct printbuf buf = PRINTBUF;
-
-	if (r_idx >= c->reflink_gc_nr)
-		goto not_found;
-
-	r = genradix_ptr(&c->reflink_gc_table, r_idx);
-	next_idx = min(next_idx, r->offset - r->size);
-	if (*idx < next_idx)
-		goto not_found;
-
-	BUG_ON((s64) r->refcount + add < 0);
-
-	r->refcount += add;
-	*idx = r->offset;
-	return 0;
-not_found:
-	if (fsck_err(c, reflink_p_to_missing_reflink_v,
-		     "pointer to missing indirect extent\n"
-		     "  %s\n"
-		     "  missing range %llu-%llu",
-		     (bch2_bkey_val_to_text(&buf, c, p.s_c), buf.buf),
-		     *idx, next_idx)) {
-		struct bkey_i_error *new;
-
-		new = bch2_trans_kmalloc(trans, sizeof(*new));
-		ret = PTR_ERR_OR_ZERO(new);
-		if (ret)
-			goto err;
-
-		bkey_init(&new->k);
-		new->k.type	= KEY_TYPE_error;
-		new->k.p		= bkey_start_pos(p.k);
-		new->k.p.offset += *idx - start;
-		bch2_key_resize(&new->k, next_idx - *idx);
-		ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, &new->k_i,
-					  BTREE_TRIGGER_NORUN);
-	}
-
-	*idx = next_idx;
-err:
-fsck_err:
-	printbuf_exit(&buf);
-	return ret;
-}
-
-static int __mark_reflink_p(struct btree_trans *trans,
-			    enum btree_id btree_id, unsigned level,
-			    struct bkey_s_c k, unsigned flags)
-{
-	struct bch_fs *c = trans->c;
-	struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k);
-	struct reflink_gc *ref;
-	size_t l, r, m;
-	u64 idx = le64_to_cpu(p.v->idx), start = idx;
-	u64 end = le64_to_cpu(p.v->idx) + p.k->size;
-	int ret = 0;
-
-	BUG_ON(!(flags & BTREE_TRIGGER_GC));
-
-	if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_reflink_p_fix) {
-		idx -= le32_to_cpu(p.v->front_pad);
-		end += le32_to_cpu(p.v->back_pad);
-	}
-
-	l = 0;
-	r = c->reflink_gc_nr;
-	while (l < r) {
-		m = l + (r - l) / 2;
-
-		ref = genradix_ptr(&c->reflink_gc_table, m);
-		if (ref->offset <= idx)
-			l = m + 1;
-		else
-			r = m;
-	}
-
-	while (idx < end && !ret)
-		ret = __bch2_mark_reflink_p(trans, p, start, end,
-					    &idx, flags, l++);
-
-	return ret;
-}
-
-int bch2_mark_reflink_p(struct btree_trans *trans,
-			enum btree_id btree_id, unsigned level,
-			struct bkey_s_c old, struct bkey_s_c new,
-			unsigned flags)
-{
-	return mem_trigger_run_overwrite_then_insert(__mark_reflink_p, trans, btree_id, level, old, new, flags);
-}
-
 void bch2_trans_fs_usage_revert(struct btree_trans *trans,
 				struct replicas_delta_list *deltas)
 {
@@ -1732,105 +1631,6 @@ int bch2_trans_mark_reservation(struct btree_trans *trans,
 	return trigger_run_overwrite_then_insert(__trans_mark_reservation, trans, btree_id, level, old, new, flags);
 }
 
-static int trans_mark_reflink_p_segment(struct btree_trans *trans,
-			struct bkey_s_c_reflink_p p,
-			u64 *idx, unsigned flags)
-{
-	struct bch_fs *c = trans->c;
-	struct btree_iter iter;
-	struct bkey_i *k;
-	__le64 *refcount;
-	int add = !(flags & BTREE_TRIGGER_OVERWRITE) ? 1 : -1;
-	struct printbuf buf = PRINTBUF;
-	int ret;
-
-	k = bch2_bkey_get_mut_noupdate(trans, &iter,
-			BTREE_ID_reflink, POS(0, *idx),
-			BTREE_ITER_WITH_UPDATES);
-	ret = PTR_ERR_OR_ZERO(k);
-	if (ret)
-		goto err;
-
-	refcount = bkey_refcount(k);
-	if (!refcount) {
-		bch2_bkey_val_to_text(&buf, c, p.s_c);
-		bch2_trans_inconsistent(trans,
-			"nonexistent indirect extent at %llu while marking\n  %s",
-			*idx, buf.buf);
-		ret = -EIO;
-		goto err;
-	}
-
-	if (!*refcount && (flags & BTREE_TRIGGER_OVERWRITE)) {
-		bch2_bkey_val_to_text(&buf, c, p.s_c);
-		bch2_trans_inconsistent(trans,
-			"indirect extent refcount underflow at %llu while marking\n  %s",
-			*idx, buf.buf);
-		ret = -EIO;
-		goto err;
-	}
-
-	if (flags & BTREE_TRIGGER_INSERT) {
-		struct bch_reflink_p *v = (struct bch_reflink_p *) p.v;
-		u64 pad;
-
-		pad = max_t(s64, le32_to_cpu(v->front_pad),
-			    le64_to_cpu(v->idx) - bkey_start_offset(&k->k));
-		BUG_ON(pad > U32_MAX);
-		v->front_pad = cpu_to_le32(pad);
-
-		pad = max_t(s64, le32_to_cpu(v->back_pad),
-			    k->k.p.offset - p.k->size - le64_to_cpu(v->idx));
-		BUG_ON(pad > U32_MAX);
-		v->back_pad = cpu_to_le32(pad);
-	}
-
-	le64_add_cpu(refcount, add);
-
-	bch2_btree_iter_set_pos_to_extent_start(&iter);
-	ret = bch2_trans_update(trans, &iter, k, 0);
-	if (ret)
-		goto err;
-
-	*idx = k->k.p.offset;
-err:
-	bch2_trans_iter_exit(trans, &iter);
-	printbuf_exit(&buf);
-	return ret;
-}
-
-static int __trans_mark_reflink_p(struct btree_trans *trans,
-				enum btree_id btree_id, unsigned level,
-				struct bkey_s_c k, unsigned flags)
-{
-	struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k);
-	u64 idx, end_idx;
-	int ret = 0;
-
-	idx	= le64_to_cpu(p.v->idx) - le32_to_cpu(p.v->front_pad);
-	end_idx = le64_to_cpu(p.v->idx) + p.k->size +
-		le32_to_cpu(p.v->back_pad);
-
-	while (idx < end_idx && !ret)
-		ret = trans_mark_reflink_p_segment(trans, p, &idx, flags);
-	return ret;
-}
-
-int bch2_trans_mark_reflink_p(struct btree_trans *trans,
-			      enum btree_id btree_id, unsigned level,
-			      struct bkey_s_c old,
-			      struct bkey_i *new,
-			      unsigned flags)
-{
-	if (flags & BTREE_TRIGGER_INSERT) {
-		struct bch_reflink_p *v = &bkey_i_to_reflink_p(new)->v;
-
-		v->front_pad = v->back_pad = 0;
-	}
-
-	return trigger_run_overwrite_then_insert(__trans_mark_reflink_p, trans, btree_id, level, old, new, flags);
-}
-
 static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans,
 				    struct bch_dev *ca, size_t b,
 				    enum bch_data_type type,
diff --git a/libbcachefs/buckets.h b/libbcachefs/buckets.h
index bc088673..379101d7 100644
--- a/libbcachefs/buckets.h
+++ b/libbcachefs/buckets.h
@@ -335,14 +335,10 @@ int bch2_mark_stripe(struct btree_trans *, enum btree_id, unsigned,
 		     struct bkey_s_c, struct bkey_s_c, unsigned);
 int bch2_mark_reservation(struct btree_trans *, enum btree_id, unsigned,
 			  struct bkey_s_c, struct bkey_s_c, unsigned);
-int bch2_mark_reflink_p(struct btree_trans *, enum btree_id, unsigned,
-			struct bkey_s_c, struct bkey_s_c, unsigned);
 
 int bch2_trans_mark_extent(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_i *, unsigned);
 int bch2_trans_mark_stripe(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_i *, unsigned);
 int bch2_trans_mark_reservation(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_i *, unsigned);
-int bch2_trans_mark_reflink_p(struct btree_trans *, enum btree_id, unsigned, struct bkey_s_c, struct bkey_i *, unsigned);
-
 #define mem_trigger_run_overwrite_then_insert(_fn, _trans, _btree_id, _level, _old, _new, _flags)\
 ({												\
 	int ret = 0;										\
diff --git a/libbcachefs/chardev.c b/libbcachefs/chardev.c
index ba0436ae..a042e07c 100644
--- a/libbcachefs/chardev.c
+++ b/libbcachefs/chardev.c
@@ -29,6 +29,63 @@ static int copy_to_user_errcode(void __user *to, const void *from, unsigned long
 	return copy_to_user(to, from, n) ? -EFAULT : 0;
 }
 
+struct thread_with_file {
+	struct task_struct	*task;
+	int			ret;
+};
+
+static void thread_with_file_exit(struct thread_with_file *thr)
+{
+	if (thr->task) {
+		kthread_stop(thr->task);
+		put_task_struct(thr->task);
+	}
+}
+
+static int run_thread_with_file(struct thread_with_file *thr,
+				const struct file_operations *fops,
+				int (*fn)(void *), const char *fmt, ...)
+{
+	va_list args;
+	struct file *file = NULL;
+	int ret, fd = -1;
+	struct printbuf name = PRINTBUF;
+	unsigned fd_flags = O_RDONLY|O_CLOEXEC|O_NONBLOCK;
+
+	va_start(args, fmt);
+	prt_vprintf(&name, fmt, args);
+	va_end(args);
+
+	thr->ret = 0;
+	thr->task = kthread_create(fn, thr, name.buf);
+	ret = PTR_ERR_OR_ZERO(thr->task);
+	if (ret)
+		goto err;
+
+	ret = get_unused_fd_flags(fd_flags);
+	if (ret < 0)
+		goto err_stop_task;
+	fd = ret;
+
+	file = anon_inode_getfile(name.buf, fops, thr, fd_flags);
+	ret = PTR_ERR_OR_ZERO(file);
+	if (ret)
+		goto err_put_fd;
+
+	fd_install(fd, file);
+	get_task_struct(thr->task);
+	wake_up_process(thr->task);
+	printbuf_exit(&name);
+	return fd;
+err_put_fd:
+	put_unused_fd(fd);
+err_stop_task:
+	kthread_stop(thr->task);
+err:
+	printbuf_exit(&name);
+	return ret;
+}
+
 /* returns with ref on ca->ref */
 static struct bch_dev *bch2_device_lookup(struct bch_fs *c, u64 dev,
 					  unsigned flags)
@@ -138,8 +195,177 @@ static long bch2_ioctl_incremental(struct bch_ioctl_incremental __user *user_arg
 }
 #endif
 
+struct fsck_thread {
+	struct thread_with_file	thr;
+	struct printbuf		buf;
+	struct bch_fs		*c;
+	char			**devs;
+	size_t			nr_devs;
+	struct bch_opts		opts;
+
+	struct log_output	output;
+	DARRAY(char)		output2;
+};
+
+static void bch2_fsck_thread_free(struct fsck_thread *thr)
+{
+	thread_with_file_exit(&thr->thr);
+	if (thr->devs)
+		for (size_t i = 0; i < thr->nr_devs; i++)
+			kfree(thr->devs[i]);
+	darray_exit(&thr->output2);
+	printbuf_exit(&thr->output.buf);
+	kfree(thr->devs);
+	kfree(thr);
+}
+
+static int bch2_fsck_thread_release(struct inode *inode, struct file *file)
+{
+	struct fsck_thread *thr = container_of(file->private_data, struct fsck_thread, thr);
+
+	bch2_fsck_thread_free(thr);
+	return 0;
+}
+
+static ssize_t bch2_fsck_thread_read(struct file *file, char __user *buf,
+				     size_t len, loff_t *ppos)
+{
+	struct fsck_thread *thr = container_of(file->private_data, struct fsck_thread, thr);
+	size_t copied = 0, b;
+	int ret = 0;
+
+	ret = wait_event_interruptible(thr->output.wait,
+			thr->output.buf.pos || thr->output2.nr);
+	if (ret)
+		return ret;
+
+	while (len) {
+		ret = darray_make_room(&thr->output2, thr->output.buf.pos);
+		if (ret)
+			break;
+
+		spin_lock_irq(&thr->output.lock);
+		b = min_t(size_t, darray_room(thr->output2), thr->output.buf.pos);
+
+		memcpy(&darray_top(thr->output2), thr->output.buf.buf, b);
+		memmove(thr->output.buf.buf,
+			thr->output.buf.buf + b,
+			thr->output.buf.pos - b);
+
+		thr->output2.nr += b;
+		thr->output.buf.pos -= b;
+		spin_unlock_irq(&thr->output.lock);
+
+		b = min(len, thr->output2.nr);
+		if (!b)
+			break;
+
+		b -= copy_to_user(buf, thr->output2.data, b);
+		if (!b) {
+			ret = -EFAULT;
+			break;
+		}
+
+		copied	+= b;
+		buf	+= b;
+		len	-= b;
+
+		memmove(thr->output2.data,
+			thr->output2.data + b,
+			thr->output2.nr - b);
+		thr->output2.nr -= b;
+	}
+
+	return copied ?: ret;
+}
+
+static const struct file_operations fsck_thread_ops = {
+	.release	= bch2_fsck_thread_release,
+	.read		= bch2_fsck_thread_read,
+	.llseek		= no_llseek,
+};
+
+static int bch2_fsck_offline_thread_fn(void *arg)
+{
+	struct fsck_thread *thr = container_of(arg, struct fsck_thread, thr);
+	struct bch_fs *c = bch2_fs_open(thr->devs, thr->nr_devs, thr->opts);
+
+	thr->thr.ret = PTR_ERR_OR_ZERO(c);
+	if (!thr->thr.ret)
+		bch2_fs_stop(c);
+	return 0;
+}
+
+static long bch2_ioctl_fsck_offline(struct bch_ioctl_fsck_offline __user *user_arg)
+{
+	struct bch_ioctl_fsck_offline arg;
+	struct fsck_thread *thr = NULL;
+	u64 *devs = NULL;
+	long ret = 0;
+
+	if (copy_from_user(&arg, user_arg, sizeof(arg)))
+		return -EFAULT;
+
+	if (arg.flags)
+		return -EINVAL;
+
+	if (!(devs = kcalloc(arg.nr_devs, sizeof(*devs), GFP_KERNEL)) ||
+	    !(thr = kzalloc(sizeof(*thr), GFP_KERNEL)) ||
+	    !(thr->devs = kcalloc(arg.nr_devs, sizeof(*thr->devs), GFP_KERNEL))) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	thr->nr_devs = arg.nr_devs;
+	thr->output.buf	= PRINTBUF;
+	thr->output.buf.atomic++;
+	spin_lock_init(&thr->output.lock);
+	init_waitqueue_head(&thr->output.wait);
+	darray_init(&thr->output2);
+
+	if (copy_from_user(devs, &user_arg->devs[0], sizeof(user_arg->devs[0]) * arg.nr_devs)) {
+		ret = -EINVAL;
+		goto err;
+	}
+
+	for (size_t i = 0; i < arg.nr_devs; i++) {
+		thr->devs[i] = strndup_user((char __user *)(unsigned long) devs[i], PATH_MAX);
+		ret = PTR_ERR_OR_ZERO(thr->devs[i]);
+		if (ret)
+			goto err;
+	}
+
+	if (arg.opts) {
+		char *optstr = strndup_user((char __user *)(unsigned long) arg.opts, 1 << 16);
+
+		ret =   PTR_ERR_OR_ZERO(optstr) ?:
+			bch2_parse_mount_opts(NULL, &thr->opts, optstr);
+		kfree(optstr);
+
+		if (ret)
+			goto err;
+	}
+
+	opt_set(thr->opts, log_output, (u64)(unsigned long)&thr->output);
+
+	ret = run_thread_with_file(&thr->thr,
+				   &fsck_thread_ops,
+				   bch2_fsck_offline_thread_fn,
+				   "bch-fsck");
+err:
+	if (ret < 0) {
+		if (thr)
+			bch2_fsck_thread_free(thr);
+		pr_err("ret %s", bch2_err_str(ret));
+	}
+	kfree(devs);
+	return ret;
+}
+
 static long bch2_global_ioctl(unsigned cmd, void __user *arg)
 {
+	long ret;
+
 	switch (cmd) {
 #if 0
 	case BCH_IOCTL_ASSEMBLE:
@@ -147,9 +373,18 @@ static long bch2_global_ioctl(unsigned cmd, void __user *arg)
 	case BCH_IOCTL_INCREMENTAL:
 		return bch2_ioctl_incremental(arg);
 #endif
+	case BCH_IOCTL_FSCK_OFFLINE: {
+		ret = bch2_ioctl_fsck_offline(arg);
+		break;
+	}
 	default:
-		return -ENOTTY;
+		ret = -ENOTTY;
+		break;
 	}
+
+	if (ret < 0)
+		ret = bch2_err_class(ret);
+	return ret;
 }
 
 static long bch2_ioctl_query_uuid(struct bch_fs *c,
@@ -299,31 +534,27 @@ static long bch2_ioctl_disk_set_state(struct bch_fs *c,
 }
 
 struct bch_data_ctx {
+	struct thread_with_file		thr;
+
 	struct bch_fs			*c;
 	struct bch_ioctl_data		arg;
 	struct bch_move_stats		stats;
-
-	int				ret;
-
-	struct task_struct		*thread;
 };
 
 static int bch2_data_thread(void *arg)
 {
-	struct bch_data_ctx *ctx = arg;
-
-	ctx->ret = bch2_data_job(ctx->c, &ctx->stats, ctx->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);
 	ctx->stats.data_type = U8_MAX;
 	return 0;
 }
 
 static int bch2_data_job_release(struct inode *inode, struct file *file)
 {
-	struct bch_data_ctx *ctx = file->private_data;
+	struct bch_data_ctx *ctx = container_of(file->private_data, struct bch_data_ctx, thr);
 
-	kthread_stop(ctx->thread);
-	put_task_struct(ctx->thread);
+	thread_with_file_exit(&ctx->thr);
 	kfree(ctx);
 	return 0;
 }
@@ -331,7 +562,7 @@ static int bch2_data_job_release(struct inode *inode, struct file *file)
 static ssize_t bch2_data_job_read(struct file *file, char __user *buf,
 				  size_t len, loff_t *ppos)
 {
-	struct bch_data_ctx *ctx = file->private_data;
+	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,
@@ -357,10 +588,8 @@ static const struct file_operations bcachefs_data_ops = {
 static long bch2_ioctl_data(struct bch_fs *c,
 			    struct bch_ioctl_data arg)
 {
-	struct bch_data_ctx *ctx = NULL;
-	struct file *file = NULL;
-	unsigned flags = O_RDONLY|O_CLOEXEC|O_NONBLOCK;
-	int ret, fd = -1;
+	struct bch_data_ctx *ctx;
+	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -375,36 +604,12 @@ static long bch2_ioctl_data(struct bch_fs *c,
 	ctx->c = c;
 	ctx->arg = arg;
 
-	ctx->thread = kthread_create(bch2_data_thread, ctx,
-				     "bch-data/%s", c->name);
-	if (IS_ERR(ctx->thread)) {
-		ret = PTR_ERR(ctx->thread);
-		goto err;
-	}
-
-	ret = get_unused_fd_flags(flags);
+	ret = run_thread_with_file(&ctx->thr,
+				   &bcachefs_data_ops,
+				   bch2_data_thread,
+				   "bch-data/%s", c->name);
 	if (ret < 0)
-		goto err;
-	fd = ret;
-
-	file = anon_inode_getfile("[bcachefs]", &bcachefs_data_ops, ctx, flags);
-	if (IS_ERR(file)) {
-		ret = PTR_ERR(file);
-		goto err;
-	}
-
-	fd_install(fd, file);
-
-	get_task_struct(ctx->thread);
-	wake_up_process(ctx->thread);
-
-	return fd;
-err:
-	if (fd >= 0)
-		put_unused_fd(fd);
-	if (!IS_ERR_OR_NULL(ctx->thread))
-		kthread_stop(ctx->thread);
-	kfree(ctx);
+		kfree(ctx);
 	return ret;
 }
 
@@ -690,6 +895,50 @@ static long bch2_ioctl_disk_resize_journal(struct bch_fs *c,
 	return ret;
 }
 
+static int bch2_fsck_online_thread_fn(void *arg)
+{
+	struct fsck_thread *thr = container_of(arg, struct fsck_thread, thr);
+	struct bch_fs *c = thr->c;
+#if 0
+	struct bch_fs *c = bch2_fs_open(thr->devs, thr->nr_devs, thr->opts);
+
+	thr->thr.ret = PTR_ERR_OR_ZERO(c);
+	if (!thr->thr.ret)
+		bch2_fs_stop(c);
+#endif
+	return 0;
+}
+
+static long bch2_ioctl_fsck_online(struct bch_fs *c,
+				   struct bch_ioctl_fsck_online arg)
+{
+	struct fsck_thread *thr = NULL;
+	long ret = 0;
+
+	if (arg.flags)
+		return -EINVAL;
+
+	thr = kzalloc(sizeof(*thr), GFP_KERNEL);
+	if (!thr)
+		return -ENOMEM;
+
+	thr->c = c;
+	thr->output.buf	= PRINTBUF;
+	thr->output.buf.atomic++;
+	spin_lock_init(&thr->output.lock);
+	init_waitqueue_head(&thr->output.wait);
+	darray_init(&thr->output2);
+
+	ret = run_thread_with_file(&thr->thr,
+				   &fsck_thread_ops,
+				   bch2_fsck_online_thread_fn,
+				   "bch-fsck");
+	bch_err_fn(c, ret);
+	if (ret < 0)
+		bch2_fsck_thread_free(thr);
+	return ret;
+}
+
 #define BCH_IOCTL(_name, _argtype)					\
 do {									\
 	_argtype i;							\
@@ -745,7 +994,8 @@ long bch2_fs_ioctl(struct bch_fs *c, unsigned cmd, void __user *arg)
 		BCH_IOCTL(disk_resize, struct bch_ioctl_disk_resize);
 	case BCH_IOCTL_DISK_RESIZE_JOURNAL:
 		BCH_IOCTL(disk_resize_journal, struct bch_ioctl_disk_resize_journal);
-
+	case BCH_IOCTL_FSCK_ONLINE:
+		BCH_IOCTL(fsck_online, struct bch_ioctl_fsck_online);
 	default:
 		return -ENOTTY;
 	}
diff --git a/libbcachefs/compress.c b/libbcachefs/compress.c
index 51af8ea2..33df8cf8 100644
--- a/libbcachefs/compress.c
+++ b/libbcachefs/compress.c
@@ -572,10 +572,6 @@ static int __bch2_fs_compress_init(struct bch_fs *c, u64 features)
 	ZSTD_parameters params = zstd_get_params(zstd_max_clevel(),
 						 c->opts.encoded_extent_max);
 
-	/*
-	 * ZSTD is lying: if we allocate the size of the workspace it says it
-	 * requires, it returns memory allocation errors
-	 */
 	c->zstd_workspace_size = zstd_cctx_workspace_bound(&params.cParams);
 
 	struct {
diff --git a/libbcachefs/data_update.c b/libbcachefs/data_update.c
index 0d58a872..22d4bb78 100644
--- a/libbcachefs/data_update.c
+++ b/libbcachefs/data_update.c
@@ -485,7 +485,7 @@ int bch2_extent_drop_ptrs(struct btree_trans *trans,
 	 * we aren't using the extent overwrite path to delete, we're
 	 * just using the normal key deletion path:
 	 */
-	if (bkey_deleted(&n->k))
+	if (bkey_deleted(&n->k) && !(iter->flags & BTREE_ITER_IS_EXTENTS))
 		n->k.size = 0;
 
 	return bch2_trans_relock(trans) ?:
@@ -605,7 +605,7 @@ int bch2_data_update_init(struct btree_trans *trans,
 		m->data_opts.rewrite_ptrs = 0;
 		/* if iter == NULL, it's just a promote */
 		if (iter)
-			ret = bch2_extent_drop_ptrs(trans, iter, k, data_opts);
+			ret = bch2_extent_drop_ptrs(trans, iter, k, m->data_opts);
 		goto done;
 	}
 
diff --git a/libbcachefs/errcode.h b/libbcachefs/errcode.h
index e3e2be79..87c13f13 100644
--- a/libbcachefs/errcode.h
+++ b/libbcachefs/errcode.h
@@ -174,6 +174,7 @@
 	x(EINVAL,			insufficient_devices_to_start)		\
 	x(EINVAL,			invalid)				\
 	x(EINVAL,			internal_fsck_err)			\
+	x(EINVAL,			opt_parse_error)			\
 	x(EROFS,			erofs_trans_commit)			\
 	x(EROFS,			erofs_no_writes)			\
 	x(EROFS,			erofs_journal_err)			\
diff --git a/libbcachefs/fs-ioctl.c b/libbcachefs/fs-ioctl.c
index 5a39bcb5..561fc1da 100644
--- a/libbcachefs/fs-ioctl.c
+++ b/libbcachefs/fs-ioctl.c
@@ -287,34 +287,26 @@ static int bch2_ioc_goingdown(struct bch_fs *c, u32 __user *arg)
 
 	bch_notice(c, "shutdown by ioctl type %u", flags);
 
-	down_write(&c->vfs_sb->s_umount);
-
 	switch (flags) {
 	case FSOP_GOING_FLAGS_DEFAULT:
 		ret = freeze_bdev(c->vfs_sb->s_bdev);
 		if (ret)
-			goto err;
-
+			break;
 		bch2_journal_flush(&c->journal);
-		c->vfs_sb->s_flags |= SB_RDONLY;
 		bch2_fs_emergency_read_only(c);
 		thaw_bdev(c->vfs_sb->s_bdev);
 		break;
-
 	case FSOP_GOING_FLAGS_LOGFLUSH:
 		bch2_journal_flush(&c->journal);
 		fallthrough;
-
 	case FSOP_GOING_FLAGS_NOLOGFLUSH:
-		c->vfs_sb->s_flags |= SB_RDONLY;
 		bch2_fs_emergency_read_only(c);
 		break;
 	default:
 		ret = -EINVAL;
 		break;
 	}
-err:
-	up_write(&c->vfs_sb->s_umount);
+
 	return ret;
 }
 
@@ -341,6 +333,10 @@ static long __bch2_ioctl_subvolume_create(struct bch_fs *c, struct file *filp,
 	     (arg.flags & BCH_SUBVOL_SNAPSHOT_RO)))
 		return -EINVAL;
 
+	if ((arg.flags & BCH_SUBVOL_SNAPSHOT_CREATE) &&
+	    !arg.src_ptr)
+		return -EOPNOTSUPP;
+
 	if (arg.flags & BCH_SUBVOL_SNAPSHOT_CREATE)
 		create_flags |= BCH_CREATE_SNAPSHOT;
 
diff --git a/libbcachefs/opts.c b/libbcachefs/opts.c
index 8dd4046c..8e6f230e 100644
--- a/libbcachefs/opts.c
+++ b/libbcachefs/opts.c
@@ -279,14 +279,14 @@ int bch2_opt_validate(const struct bch_option *opt, u64 v, struct printbuf *err)
 		if (err)
 			prt_printf(err, "%s: not a multiple of 512",
 			       opt->attr.name);
-		return -EINVAL;
+		return -BCH_ERR_opt_parse_error;
 	}
 
 	if ((opt->flags & OPT_MUST_BE_POW_2) && !is_power_of_2(v)) {
 		if (err)
 			prt_printf(err, "%s: must be a power of two",
 			       opt->attr.name);
-		return -EINVAL;
+		return -BCH_ERR_opt_parse_error;
 	}
 
 	if (opt->fn.validate)
diff --git a/libbcachefs/opts.h b/libbcachefs/opts.h
index 8526f177..91026dfb 100644
--- a/libbcachefs/opts.h
+++ b/libbcachefs/opts.h
@@ -419,6 +419,11 @@ enum fsck_err_opts {
 	  OPT_BOOL(),							\
 	  BCH2_NO_SB_OPT,		false,				\
 	  NULL,		"Allocate the buckets_nouse bitmap")		\
+	x(log_output,			u64,				\
+	  0,								\
+	  OPT_UINT(0, S64_MAX),						\
+	  BCH2_NO_SB_OPT,		false,				\
+	  NULL,		"Allocate the buckets_nouse bitmap")		\
 	x(project,			u8,				\
 	  OPT_INODE,							\
 	  OPT_BOOL(),							\
diff --git a/libbcachefs/recovery.c b/libbcachefs/recovery.c
index 98f1454c..69b49845 100644
--- a/libbcachefs/recovery.c
+++ b/libbcachefs/recovery.c
@@ -171,10 +171,12 @@ static int bch2_journal_replay(struct bch_fs *c)
 
 		struct journal_key *k = keys->d + i;
 
-		ret = commit_do(trans, NULL, NULL,
-				BCH_TRANS_COMMIT_no_enospc|
-				BCH_TRANS_COMMIT_journal_reclaim|
-				(!k->allocated ? BCH_TRANS_COMMIT_no_journal_res : 0),
+		/* Skip fastpath if we're low on space in the journal */
+		ret = c->journal.watermark ? -1 :
+			commit_do(trans, NULL, NULL,
+				  BCH_TRANS_COMMIT_no_enospc|
+				  BCH_TRANS_COMMIT_journal_reclaim|
+				  (!k->allocated ? BCH_TRANS_COMMIT_no_journal_res : 0),
 			     bch2_journal_replay_key(trans, k));
 		BUG_ON(!ret && !k->overwritten);
 		if (ret) {
@@ -657,13 +659,13 @@ static int bch2_run_recovery_pass(struct bch_fs *c, enum bch_recovery_pass pass)
 		struct recovery_pass_fn *p = recovery_pass_fns + pass;
 
 		if (!(p->when & PASS_SILENT))
-			printk(KERN_INFO bch2_log_msg(c, "%s..."),
-			       bch2_recovery_passes[pass]);
+			bch2_print(c, KERN_INFO bch2_log_msg(c, "%s..."),
+				   bch2_recovery_passes[pass]);
 		ret = p->fn(c);
 		if (ret)
 			return ret;
 		if (!(p->when & PASS_SILENT))
-			printk(KERN_CONT " done\n");
+			bch2_print(c, KERN_CONT " done\n");
 
 		c->recovery_passes_complete |= BIT_ULL(pass);
 	}
diff --git a/libbcachefs/reflink.c b/libbcachefs/reflink.c
index 07ddf3e8..1d56470e 100644
--- a/libbcachefs/reflink.c
+++ b/libbcachefs/reflink.c
@@ -3,6 +3,7 @@
 #include "bkey_buf.h"
 #include "btree_update.h"
 #include "buckets.h"
+#include "error.h"
 #include "extents.h"
 #include "inode.h"
 #include "io_misc.h"
@@ -73,6 +74,206 @@ bool bch2_reflink_p_merge(struct bch_fs *c, struct bkey_s _l, struct bkey_s_c _r
 	return true;
 }
 
+static int trans_mark_reflink_p_segment(struct btree_trans *trans,
+			struct bkey_s_c_reflink_p p,
+			u64 *idx, unsigned flags)
+{
+	struct bch_fs *c = trans->c;
+	struct btree_iter iter;
+	struct bkey_i *k;
+	__le64 *refcount;
+	int add = !(flags & BTREE_TRIGGER_OVERWRITE) ? 1 : -1;
+	struct printbuf buf = PRINTBUF;
+	int ret;
+
+	k = bch2_bkey_get_mut_noupdate(trans, &iter,
+			BTREE_ID_reflink, POS(0, *idx),
+			BTREE_ITER_WITH_UPDATES);
+	ret = PTR_ERR_OR_ZERO(k);
+	if (ret)
+		goto err;
+
+	refcount = bkey_refcount(k);
+	if (!refcount) {
+		bch2_bkey_val_to_text(&buf, c, p.s_c);
+		bch2_trans_inconsistent(trans,
+			"nonexistent indirect extent at %llu while marking\n  %s",
+			*idx, buf.buf);
+		ret = -EIO;
+		goto err;
+	}
+
+	if (!*refcount && (flags & BTREE_TRIGGER_OVERWRITE)) {
+		bch2_bkey_val_to_text(&buf, c, p.s_c);
+		bch2_trans_inconsistent(trans,
+			"indirect extent refcount underflow at %llu while marking\n  %s",
+			*idx, buf.buf);
+		ret = -EIO;
+		goto err;
+	}
+
+	if (flags & BTREE_TRIGGER_INSERT) {
+		struct bch_reflink_p *v = (struct bch_reflink_p *) p.v;
+		u64 pad;
+
+		pad = max_t(s64, le32_to_cpu(v->front_pad),
+			    le64_to_cpu(v->idx) - bkey_start_offset(&k->k));
+		BUG_ON(pad > U32_MAX);
+		v->front_pad = cpu_to_le32(pad);
+
+		pad = max_t(s64, le32_to_cpu(v->back_pad),
+			    k->k.p.offset - p.k->size - le64_to_cpu(v->idx));
+		BUG_ON(pad > U32_MAX);
+		v->back_pad = cpu_to_le32(pad);
+	}
+
+	le64_add_cpu(refcount, add);
+
+	bch2_btree_iter_set_pos_to_extent_start(&iter);
+	ret = bch2_trans_update(trans, &iter, k, 0);
+	if (ret)
+		goto err;
+
+	*idx = k->k.p.offset;
+err:
+	bch2_trans_iter_exit(trans, &iter);
+	printbuf_exit(&buf);
+	return ret;
+}
+
+static int __trans_mark_reflink_p(struct btree_trans *trans,
+				enum btree_id btree_id, unsigned level,
+				struct bkey_s_c k, unsigned flags)
+{
+	struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k);
+	u64 idx, end_idx;
+	int ret = 0;
+
+	idx	= le64_to_cpu(p.v->idx) - le32_to_cpu(p.v->front_pad);
+	end_idx = le64_to_cpu(p.v->idx) + p.k->size +
+		le32_to_cpu(p.v->back_pad);
+
+	while (idx < end_idx && !ret)
+		ret = trans_mark_reflink_p_segment(trans, p, &idx, flags);
+	return ret;
+}
+
+int bch2_trans_mark_reflink_p(struct btree_trans *trans,
+			      enum btree_id btree_id, unsigned level,
+			      struct bkey_s_c old,
+			      struct bkey_i *new,
+			      unsigned flags)
+{
+	if (flags & BTREE_TRIGGER_INSERT) {
+		struct bch_reflink_p *v = &bkey_i_to_reflink_p(new)->v;
+
+		v->front_pad = v->back_pad = 0;
+	}
+
+	return trigger_run_overwrite_then_insert(__trans_mark_reflink_p, trans, btree_id, level, old, new, flags);
+}
+
+static s64 __bch2_mark_reflink_p(struct btree_trans *trans,
+				 struct bkey_s_c_reflink_p p,
+				 u64 start, u64 end,
+				 u64 *idx, unsigned flags, size_t r_idx)
+{
+	struct bch_fs *c = trans->c;
+	struct reflink_gc *r;
+	int add = !(flags & BTREE_TRIGGER_OVERWRITE) ? 1 : -1;
+	u64 next_idx = end;
+	s64 ret = 0;
+	struct printbuf buf = PRINTBUF;
+
+	if (r_idx >= c->reflink_gc_nr)
+		goto not_found;
+
+	r = genradix_ptr(&c->reflink_gc_table, r_idx);
+	next_idx = min(next_idx, r->offset - r->size);
+	if (*idx < next_idx)
+		goto not_found;
+
+	BUG_ON((s64) r->refcount + add < 0);
+
+	r->refcount += add;
+	*idx = r->offset;
+	return 0;
+not_found:
+	if (fsck_err(c, reflink_p_to_missing_reflink_v,
+		     "pointer to missing indirect extent\n"
+		     "  %s\n"
+		     "  missing range %llu-%llu",
+		     (bch2_bkey_val_to_text(&buf, c, p.s_c), buf.buf),
+		     *idx, next_idx)) {
+		struct bkey_i_error *new;
+
+		new = bch2_trans_kmalloc(trans, sizeof(*new));
+		ret = PTR_ERR_OR_ZERO(new);
+		if (ret)
+			goto err;
+
+		bkey_init(&new->k);
+		new->k.type	= KEY_TYPE_error;
+		new->k.p		= bkey_start_pos(p.k);
+		new->k.p.offset += *idx - start;
+		bch2_key_resize(&new->k, next_idx - *idx);
+		ret = bch2_btree_insert_trans(trans, BTREE_ID_extents, &new->k_i,
+					  BTREE_TRIGGER_NORUN);
+	}
+
+	*idx = next_idx;
+err:
+fsck_err:
+	printbuf_exit(&buf);
+	return ret;
+}
+
+static int __mark_reflink_p(struct btree_trans *trans,
+			    enum btree_id btree_id, unsigned level,
+			    struct bkey_s_c k, unsigned flags)
+{
+	struct bch_fs *c = trans->c;
+	struct bkey_s_c_reflink_p p = bkey_s_c_to_reflink_p(k);
+	struct reflink_gc *ref;
+	size_t l, r, m;
+	u64 idx = le64_to_cpu(p.v->idx), start = idx;
+	u64 end = le64_to_cpu(p.v->idx) + p.k->size;
+	int ret = 0;
+
+	BUG_ON(!(flags & BTREE_TRIGGER_GC));
+
+	if (c->sb.version_upgrade_complete >= bcachefs_metadata_version_reflink_p_fix) {
+		idx -= le32_to_cpu(p.v->front_pad);
+		end += le32_to_cpu(p.v->back_pad);
+	}
+
+	l = 0;
+	r = c->reflink_gc_nr;
+	while (l < r) {
+		m = l + (r - l) / 2;
+
+		ref = genradix_ptr(&c->reflink_gc_table, m);
+		if (ref->offset <= idx)
+			l = m + 1;
+		else
+			r = m;
+	}
+
+	while (idx < end && !ret)
+		ret = __bch2_mark_reflink_p(trans, p, start, end,
+					    &idx, flags, l++);
+
+	return ret;
+}
+
+int bch2_mark_reflink_p(struct btree_trans *trans,
+			enum btree_id btree_id, unsigned level,
+			struct bkey_s_c old, struct bkey_s_c new,
+			unsigned flags)
+{
+	return mem_trigger_run_overwrite_then_insert(__mark_reflink_p, trans, btree_id, level, old, new, flags);
+}
+
 /* indirect extents */
 
 int bch2_reflink_v_invalid(struct bch_fs *c, struct bkey_s_c k,
@@ -121,6 +322,14 @@ int bch2_trans_mark_reflink_v(struct btree_trans *trans,
 {
 	check_indirect_extent_deleting(new, &flags);
 
+	if (old.k->type == KEY_TYPE_reflink_v &&
+	    new->k.type == KEY_TYPE_reflink_v &&
+	    old.k->u64s == new->k.u64s &&
+	    !memcmp(bkey_s_c_to_reflink_v(old).v->start,
+		    bkey_i_to_reflink_v(new)->v.start,
+		    bkey_val_bytes(&new->k) - 8))
+		return 0;
+
 	return bch2_trans_mark_extent(trans, btree_id, level, old, new, flags);
 }
 
diff --git a/libbcachefs/reflink.h b/libbcachefs/reflink.h
index 8ccf3f9c..6cc9c4a7 100644
--- a/libbcachefs/reflink.h
+++ b/libbcachefs/reflink.h
@@ -9,6 +9,10 @@ int bch2_reflink_p_invalid(struct bch_fs *, struct bkey_s_c,
 void bch2_reflink_p_to_text(struct printbuf *, struct bch_fs *,
 			    struct bkey_s_c);
 bool bch2_reflink_p_merge(struct bch_fs *, struct bkey_s, struct bkey_s_c);
+int bch2_trans_mark_reflink_p(struct btree_trans *, enum btree_id, unsigned,
+			      struct bkey_s_c, struct bkey_i *, unsigned);
+int bch2_mark_reflink_p(struct btree_trans *, enum btree_id, unsigned,
+			struct bkey_s_c, struct bkey_s_c, unsigned);
 
 #define bch2_bkey_ops_reflink_p ((struct bkey_ops) {		\
 	.key_invalid	= bch2_reflink_p_invalid,		\
diff --git a/libbcachefs/super.c b/libbcachefs/super.c
index e7f186b4..3abccdbf 100644
--- a/libbcachefs/super.c
+++ b/libbcachefs/super.c
@@ -80,6 +80,25 @@ const char * const bch2_fs_flag_strs[] = {
 	NULL
 };
 
+void __bch2_print(struct bch_fs *c, const char *fmt, ...)
+{
+	va_list args;
+
+	va_start(args, fmt);
+	if (likely(!c->output)) {
+		vprintk(fmt, args);
+	} else {
+		unsigned long flags;
+
+		spin_lock_irqsave(&c->output->lock, flags);
+		prt_vprintf(&c->output->buf, fmt, args);
+		spin_unlock_irqrestore(&c->output->lock, flags);
+
+		wake_up(&c->output->wait);
+	}
+	va_end(args);
+}
+
 #define KTYPE(type)							\
 static const struct attribute_group type ## _group = {			\
 	.attrs = type ## _files						\
@@ -703,6 +722,8 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 		goto out;
 	}
 
+	c->output = (void *)(unsigned long) opts.log_output;
+
 	__module_get(THIS_MODULE);
 
 	closure_init(&c->cl, NULL);
diff --git a/libbcachefs/sysfs.c b/libbcachefs/sysfs.c
index 4a7c93bc..1b82a3a9 100644
--- a/libbcachefs/sysfs.c
+++ b/libbcachefs/sysfs.c
@@ -278,8 +278,8 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c
 		if (!btree_type_has_ptrs(id))
 			continue;
 
-		for_each_btree_key(trans, iter, id, POS_MIN,
-				   BTREE_ITER_ALL_SNAPSHOTS, k, ret) {
+		ret = for_each_btree_key2(trans, iter, id, POS_MIN,
+					  BTREE_ITER_ALL_SNAPSHOTS, k, ({
 			struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
 			struct bch_extent_crc_unpacked crc;
 			const union bch_extent_entry *entry;
@@ -305,8 +305,8 @@ static int bch2_compression_stats_to_text(struct printbuf *out, struct bch_fs *c
 				s[t].sectors_compressed += k.k->size;
 				s[t].sectors_uncompressed += k.k->size;
 			}
-		}
-		bch2_trans_iter_exit(trans, &iter);
+			0;
+		}));
 	}
 
 	bch2_trans_put(trans);
diff --git a/libbcachefs/trace.h b/libbcachefs/trace.h
index 6e2ad6f3..cfa7ee78 100644
--- a/libbcachefs/trace.h
+++ b/libbcachefs/trace.h
@@ -72,7 +72,7 @@ DECLARE_EVENT_CLASS(trans_str,
 		  __entry->trans_fn, (void *) __entry->caller_ip, __get_str(str))
 );
 
-DECLARE_EVENT_CLASS(btree_node,
+DECLARE_EVENT_CLASS(btree_node_nofs,
 	TP_PROTO(struct bch_fs *c, struct btree *b),
 	TP_ARGS(c, b),
 
@@ -97,6 +97,33 @@ DECLARE_EVENT_CLASS(btree_node,
 		  __entry->pos_inode, __entry->pos_offset, __entry->pos_snapshot)
 );
 
+DECLARE_EVENT_CLASS(btree_node,
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b),
+
+	TP_STRUCT__entry(
+		__field(dev_t,		dev			)
+		__array(char,		trans_fn, 32		)
+		__field(u8,		level			)
+		__field(u8,		btree_id		)
+		TRACE_BPOS_entries(pos)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= trans->c->dev;
+		strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn));
+		__entry->level		= b->c.level;
+		__entry->btree_id	= b->c.btree_id;
+		TRACE_BPOS_assign(pos, b->key.k.p);
+	),
+
+	TP_printk("%d,%d %s %u %s %llu:%llu:%u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev), __entry->trans_fn,
+		  __entry->level,
+		  bch2_btree_id_str(__entry->btree_id),
+		  __entry->pos_inode, __entry->pos_offset, __entry->pos_snapshot)
+);
+
 DECLARE_EVENT_CLASS(bch_fs,
 	TP_PROTO(struct bch_fs *c),
 	TP_ARGS(c),
@@ -112,6 +139,23 @@ DECLARE_EVENT_CLASS(bch_fs,
 	TP_printk("%d,%d", MAJOR(__entry->dev), MINOR(__entry->dev))
 );
 
+DECLARE_EVENT_CLASS(btree_trans,
+	TP_PROTO(struct btree_trans *trans),
+	TP_ARGS(trans),
+
+	TP_STRUCT__entry(
+		__field(dev_t,		dev			)
+		__array(char,		trans_fn, 32		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= trans->c->dev;
+		strscpy(__entry->trans_fn, trans->fn, sizeof(__entry->trans_fn));
+	),
+
+	TP_printk("%d,%d %s", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->trans_fn)
+);
+
 DECLARE_EVENT_CLASS(bio,
 	TP_PROTO(struct bio *bio),
 	TP_ARGS(bio),
@@ -330,36 +374,36 @@ TRACE_EVENT(btree_cache_scan,
 		  __entry->nr_to_scan, __entry->can_free, __entry->ret)
 );
 
-DEFINE_EVENT(btree_node, btree_cache_reap,
+DEFINE_EVENT(btree_node_nofs, btree_cache_reap,
 	TP_PROTO(struct bch_fs *c, struct btree *b),
 	TP_ARGS(c, b)
 );
 
-DEFINE_EVENT(bch_fs, btree_cache_cannibalize_lock_fail,
-	TP_PROTO(struct bch_fs *c),
-	TP_ARGS(c)
+DEFINE_EVENT(btree_trans, btree_cache_cannibalize_lock_fail,
+	TP_PROTO(struct btree_trans *trans),
+	TP_ARGS(trans)
 );
 
-DEFINE_EVENT(bch_fs, btree_cache_cannibalize_lock,
-	TP_PROTO(struct bch_fs *c),
-	TP_ARGS(c)
+DEFINE_EVENT(btree_trans, btree_cache_cannibalize_lock,
+	TP_PROTO(struct btree_trans *trans),
+	TP_ARGS(trans)
 );
 
-DEFINE_EVENT(bch_fs, btree_cache_cannibalize,
-	TP_PROTO(struct bch_fs *c),
-	TP_ARGS(c)
+DEFINE_EVENT(btree_trans, btree_cache_cannibalize,
+	TP_PROTO(struct btree_trans *trans),
+	TP_ARGS(trans)
 );
 
-DEFINE_EVENT(bch_fs, btree_cache_cannibalize_unlock,
-	TP_PROTO(struct bch_fs *c),
-	TP_ARGS(c)
+DEFINE_EVENT(btree_trans, btree_cache_cannibalize_unlock,
+	TP_PROTO(struct btree_trans *trans),
+	TP_ARGS(trans)
 );
 
 /* Btree */
 
 DEFINE_EVENT(btree_node, btree_node_read,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 TRACE_EVENT(btree_node_write,
@@ -383,13 +427,13 @@ TRACE_EVENT(btree_node_write,
 );
 
 DEFINE_EVENT(btree_node, btree_node_alloc,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 DEFINE_EVENT(btree_node, btree_node_free,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 TRACE_EVENT(btree_reserve_get_fail,
@@ -421,28 +465,28 @@ TRACE_EVENT(btree_reserve_get_fail,
 );
 
 DEFINE_EVENT(btree_node, btree_node_compact,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 DEFINE_EVENT(btree_node, btree_node_merge,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 DEFINE_EVENT(btree_node, btree_node_split,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 DEFINE_EVENT(btree_node, btree_node_rewrite,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 DEFINE_EVENT(btree_node, btree_node_set_root,
-	TP_PROTO(struct bch_fs *c, struct btree *b),
-	TP_ARGS(c, b)
+	TP_PROTO(struct btree_trans *trans, struct btree *b),
+	TP_ARGS(trans, b)
 );
 
 TRACE_EVENT(btree_path_relock_fail,