1 files changed, 161 insertions, 39 deletions

diff --git a/fs/bcachefs/data_update.c b/fs/bcachefs/data_update.c
index 899ff46de8e0..0022b51ce3c0 100644
--- a/fs/bcachefs/data_update.c
+++ b/fs/bcachefs/data_update.c
@@ -13,6 +13,8 @@
 #include "keylist.h"
 #include "move.h"
 #include "nocow_locking.h"
+#include "rebalance.h"
+#include "snapshot.h"
 #include "subvolume.h"
 #include "trace.h"
 
@@ -161,11 +163,7 @@ static int __bch2_data_update_index_update(struct btree_trans *trans,
 			if (((1U << i) & m->data_opts.rewrite_ptrs) &&
 			    (ptr = bch2_extent_has_ptr(old, p, bkey_i_to_s(insert))) &&
 			    !ptr->cached) {
-				bch2_bkey_drop_ptr_noerror(bkey_i_to_s(insert), ptr);
-				/*
-				 * See comment below:
 				bch2_extent_ptr_set_cached(bkey_i_to_s(insert), ptr);
-				*/
 				rewrites_found |= 1U << i;
 			}
 			i++;
@@ -211,14 +209,8 @@ restart_drop_extra_replicas:
 			if (!p.ptr.cached &&
 			    durability - ptr_durability >= m->op.opts.data_replicas) {
 				durability -= ptr_durability;
-				bch2_bkey_drop_ptr_noerror(bkey_i_to_s(insert), &entry->ptr);
-				/*
-				 * Currently, we're dropping unneeded replicas
-				 * instead of marking them as cached, since
-				 * cached data in stripe buckets prevents them
-				 * from being reused:
+
 				bch2_extent_ptr_set_cached(bkey_i_to_s(insert), &entry->ptr);
-				*/
 				goto restart_drop_extra_replicas;
 			}
 		}
@@ -248,19 +240,59 @@ restart_drop_extra_replicas:
 
 		next_pos = insert->k.p;
 
+		/*
+		 * Check for nonce offset inconsistency:
+		 * This is debug code - we've been seeing this bug rarely, and
+		 * it's been hard to reproduce, so this should give us some more
+		 * information when it does occur:
+		 */
+		struct printbuf err = PRINTBUF;
+		int invalid = bch2_bkey_invalid(c, bkey_i_to_s_c(insert), __btree_node_type(0, m->btree_id), 0, &err);
+		printbuf_exit(&err);
+
+		if (invalid) {
+			struct printbuf buf = PRINTBUF;
+
+			prt_str(&buf, "about to insert invalid key in data update path");
+			prt_str(&buf, "\nold: ");
+			bch2_bkey_val_to_text(&buf, c, old);
+			prt_str(&buf, "\nk:   ");
+			bch2_bkey_val_to_text(&buf, c, k);
+			prt_str(&buf, "\nnew: ");
+			bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(insert));
+
+			bch2_print_string_as_lines(KERN_ERR, buf.buf);
+			printbuf_exit(&buf);
+
+			bch2_fatal_error(c);
+			goto out;
+		}
+
+		if (trace_data_update_enabled()) {
+			struct printbuf buf = PRINTBUF;
+
+			prt_str(&buf, "\nold: ");
+			bch2_bkey_val_to_text(&buf, c, old);
+			prt_str(&buf, "\nk:   ");
+			bch2_bkey_val_to_text(&buf, c, k);
+			prt_str(&buf, "\nnew: ");
+			bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(insert));
+
+			trace_data_update(c, buf.buf);
+			printbuf_exit(&buf);
+		}
+
 		ret =   bch2_insert_snapshot_whiteouts(trans, m->btree_id,
 						k.k->p, bkey_start_pos(&insert->k)) ?:
 			bch2_insert_snapshot_whiteouts(trans, m->btree_id,
-						k.k->p, insert->k.p);
-		if (ret)
-			goto err;
-
-		ret   = bch2_trans_update(trans, &iter, insert,
+						k.k->p, insert->k.p) ?:
+			bch2_bkey_set_needs_rebalance(c, insert, &op->opts) ?:
+			bch2_trans_update(trans, &iter, insert,
 				BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?:
 			bch2_trans_commit(trans, &op->res,
 				NULL,
-				BTREE_INSERT_NOCHECK_RW|
-				BTREE_INSERT_NOFAIL|
+				BCH_TRANS_COMMIT_no_check_rw|
+				BCH_TRANS_COMMIT_no_enospc|
 				m->data_opts.btree_insert_flags);
 		if (!ret) {
 			bch2_btree_iter_set_pos(&iter, next_pos);
@@ -281,14 +313,14 @@ next:
 		}
 		continue;
 nowork:
-		if (m->ctxt && m->ctxt->stats) {
+		if (m->stats) {
 			BUG_ON(k.k->p.offset <= iter.pos.offset);
-			atomic64_inc(&m->ctxt->stats->keys_raced);
+			atomic64_inc(&m->stats->keys_raced);
 			atomic64_add(k.k->p.offset - iter.pos.offset,
-				     &m->ctxt->stats->sectors_raced);
+				     &m->stats->sectors_raced);
 		}
 
-		this_cpu_inc(c->counters[BCH_COUNTER_move_extent_fail]);
+		count_event(c, move_extent_fail);
 
 		bch2_btree_iter_advance(&iter);
 		goto next;
@@ -323,7 +355,6 @@ void bch2_data_update_exit(struct data_update *update)
 	struct bch_fs *c = update->op.c;
 	struct bkey_ptrs_c ptrs =
 		bch2_bkey_ptrs_c(bkey_i_to_s_c(update->k.k));
-	const struct bch_extent_ptr *ptr;
 
 	bkey_for_each_ptr(ptrs, ptr) {
 		if (c->opts.nocow_enabled)
@@ -337,14 +368,13 @@ void bch2_data_update_exit(struct data_update *update)
 	bch2_bio_free_pages_pool(c, &update->op.wbio.bio);
 }
 
-void bch2_update_unwritten_extent(struct btree_trans *trans,
+static void bch2_update_unwritten_extent(struct btree_trans *trans,
 				  struct data_update *update)
 {
 	struct bch_fs *c = update->op.c;
 	struct bio *bio = &update->op.wbio.bio;
 	struct bkey_i_extent *e;
 	struct write_point *wp;
-	struct bch_extent_ptr *ptr;
 	struct closure cl;
 	struct btree_iter iter;
 	struct bkey_s_c k;
@@ -385,6 +415,8 @@ void bch2_update_unwritten_extent(struct btree_trans *trans,
 			continue;
 		}
 
+		bch_err_fn_ratelimited(c, ret);
+
 		if (ret)
 			return;
 
@@ -417,7 +449,51 @@ void bch2_update_unwritten_extent(struct btree_trans *trans,
 	}
 }
 
+int bch2_extent_drop_ptrs(struct btree_trans *trans,
+			  struct btree_iter *iter,
+			  struct bkey_s_c k,
+			  struct data_update_opts data_opts)
+{
+	struct bch_fs *c = trans->c;
+	struct bkey_i *n;
+	int ret;
+
+	n = bch2_bkey_make_mut_noupdate(trans, k);
+	ret = PTR_ERR_OR_ZERO(n);
+	if (ret)
+		return ret;
+
+	while (data_opts.kill_ptrs) {
+		unsigned i = 0, drop = __fls(data_opts.kill_ptrs);
+		struct bch_extent_ptr *ptr;
+
+		bch2_bkey_drop_ptrs(bkey_i_to_s(n), ptr, i++ == drop);
+		data_opts.kill_ptrs ^= 1U << drop;
+	}
+
+	/*
+	 * If the new extent no longer has any pointers, bch2_extent_normalize()
+	 * will do the appropriate thing with it (turning it into a
+	 * KEY_TYPE_error key, or just a discard if it was a cached extent)
+	 */
+	bch2_extent_normalize(c, bkey_i_to_s(n));
+
+	/*
+	 * Since we're not inserting through an extent iterator
+	 * (BTREE_ITER_ALL_SNAPSHOTS iterators aren't extent iterators),
+	 * we aren't using the extent overwrite path to delete, we're
+	 * just using the normal key deletion path:
+	 */
+	if (bkey_deleted(&n->k) && !(iter->flags & BTREE_ITER_IS_EXTENTS))
+		n->k.size = 0;
+
+	return bch2_trans_relock(trans) ?:
+		bch2_trans_update(trans, iter, n, BTREE_UPDATE_INTERNAL_SNAPSHOT_NODE) ?:
+		bch2_trans_commit(trans, NULL, NULL, BCH_TRANS_COMMIT_no_enospc);
+}
+
 int bch2_data_update_init(struct btree_trans *trans,
+			  struct btree_iter *iter,
 			  struct moving_context *ctxt,
 			  struct data_update *m,
 			  struct write_point_specifier wp,
@@ -430,15 +506,24 @@ int bch2_data_update_init(struct btree_trans *trans,
 	struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
 	const union bch_extent_entry *entry;
 	struct extent_ptr_decoded p;
-	const struct bch_extent_ptr *ptr;
 	unsigned i, reserve_sectors = k.k->size * data_opts.extra_replicas;
 	unsigned ptrs_locked = 0;
-	int ret;
+	int ret = 0;
+
+	/*
+	 * fs is corrupt  we have a key for a snapshot node that doesn't exist,
+	 * and we have to check for this because we go rw before repairing the
+	 * snapshots table - just skip it, we can move it later.
+	 */
+	if (unlikely(k.k->p.snapshot && !bch2_snapshot_equiv(c, k.k->p.snapshot)))
+		return -BCH_ERR_data_update_done;
 
 	bch2_bkey_buf_init(&m->k);
 	bch2_bkey_buf_reassemble(&m->k, c, k);
 	m->btree_id	= btree_id;
 	m->data_opts	= data_opts;
+	m->ctxt		= ctxt;
+	m->stats	= ctxt ? ctxt->stats : NULL;
 
 	bch2_write_op_init(&m->op, c, io_opts);
 	m->op.pos	= bkey_start_pos(k.k);
@@ -451,12 +536,14 @@ int bch2_data_update_init(struct btree_trans *trans,
 		BCH_WRITE_DATA_ENCODED|
 		BCH_WRITE_MOVE|
 		m->data_opts.write_flags;
-	m->op.compression_opt	= io_opts.background_compression ?: io_opts.compression;
+	m->op.compression_opt	= background_compression(io_opts);
 	m->op.watermark		= m->data_opts.btree_insert_flags & BCH_WATERMARK_MASK;
 
 	bkey_for_each_ptr(ptrs, ptr)
 		percpu_ref_get(&bch_dev_bkey_exists(c, ptr->dev)->ref);
 
+	unsigned durability_have = 0, durability_removing = 0;
+
 	i = 0;
 	bkey_for_each_ptr_decode(k.k, ptrs, p, entry) {
 		bool locked;
@@ -468,8 +555,11 @@ int bch2_data_update_init(struct btree_trans *trans,
 				reserve_sectors += k.k->size;
 
 			m->op.nr_replicas += bch2_extent_ptr_desired_durability(c, &p);
-		} else if (!p.ptr.cached) {
+			durability_removing += bch2_extent_ptr_desired_durability(c, &p);
+		} else if (!p.ptr.cached &&
+			   !((1U << i) & m->data_opts.kill_ptrs)) {
 			bch2_dev_list_add_dev(&m->op.devs_have, p.ptr.dev);
+			durability_have += bch2_extent_ptr_durability(c, &p);
 		}
 
 		/*
@@ -487,10 +577,10 @@ int bch2_data_update_init(struct btree_trans *trans,
 
 		if (c->opts.nocow_enabled) {
 			if (ctxt) {
-				move_ctxt_wait_event(ctxt, trans,
+				move_ctxt_wait_event(ctxt,
 						(locked = bch2_bucket_nocow_trylock(&c->nocow_locks,
 									  PTR_BUCKET_POS(c, &p.ptr), 0)) ||
-						!atomic_read(&ctxt->read_sectors));
+						list_empty(&ctxt->ios));
 
 				if (!locked)
 					bch2_bucket_nocow_lock(&c->nocow_locks,
@@ -508,6 +598,39 @@ int bch2_data_update_init(struct btree_trans *trans,
 		i++;
 	}
 
+	unsigned durability_required = max(0, (int) (io_opts.data_replicas - durability_have));
+
+	/*
+	 * If current extent durability is less than io_opts.data_replicas,
+	 * we're not trying to rereplicate the extent up to data_replicas here -
+	 * unless extra_replicas was specified
+	 *
+	 * Increasing replication is an explicit operation triggered by
+	 * rereplicate, currently, so that users don't get an unexpected -ENOSPC
+	 */
+	if (!(m->data_opts.write_flags & BCH_WRITE_CACHED) &&
+	    !durability_required) {
+		m->data_opts.kill_ptrs |= m->data_opts.rewrite_ptrs;
+		m->data_opts.rewrite_ptrs = 0;
+		/* if iter == NULL, it's just a promote */
+		if (iter)
+			ret = bch2_extent_drop_ptrs(trans, iter, k, m->data_opts);
+		goto done;
+	}
+
+	m->op.nr_replicas = min(durability_removing, durability_required) +
+		m->data_opts.extra_replicas;
+
+	/*
+	 * If device(s) were set to durability=0 after data was written to them
+	 * we can end up with a duribilty=0 extent, and the normal algorithm
+	 * that tries not to increase durability doesn't work:
+	 */
+	if (!(durability_have + durability_removing))
+		m->op.nr_replicas = max((unsigned) m->op.nr_replicas, 1);
+
+	m->op.nr_replicas_required = m->op.nr_replicas;
+
 	if (reserve_sectors) {
 		ret = bch2_disk_reservation_add(c, &m->op.res, reserve_sectors,
 				m->data_opts.extra_replicas
@@ -517,14 +640,11 @@ int bch2_data_update_init(struct btree_trans *trans,
 			goto err;
 	}
 
-	m->op.nr_replicas += m->data_opts.extra_replicas;
-	m->op.nr_replicas_required = m->op.nr_replicas;
-
-	BUG_ON(!m->op.nr_replicas);
+	if (bkey_extent_is_unwritten(k)) {
+		bch2_update_unwritten_extent(trans, m);
+		goto done;
+	}
 
-	/* Special handling required: */
-	if (bkey_extent_is_unwritten(k))
-		return -BCH_ERR_unwritten_extent_update;
 	return 0;
 err:
 	i = 0;
@@ -539,12 +659,14 @@ err:
 	bch2_bkey_buf_exit(&m->k, c);
 	bch2_bio_free_pages_pool(c, &m->op.wbio.bio);
 	return ret;
+done:
+	bch2_data_update_exit(m);
+	return ret ?: -BCH_ERR_data_update_done;
 }
 
 void bch2_data_update_opts_normalize(struct bkey_s_c k, struct data_update_opts *opts)
 {
 	struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
-	const struct bch_extent_ptr *ptr;
 	unsigned i = 0;
 
 	bkey_for_each_ptr(ptrs, ptr) {