bcachefs: bch2_btree_iter_relock_intent()

This adds a new helper for btree_cache.c that does what we want where
the iterator is still being traverse - and also eliminates some
unnecessary transaction restarts.

Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>

4 files changed, 42 insertions, 18 deletions

diff --git a/fs/bcachefs/btree_cache.c b/fs/bcachefs/btree_cache.c
index e2c02ae98f83..6d5cf2a5a159 100644
--- a/fs/bcachefs/btree_cache.c
+++ b/fs/bcachefs/btree_cache.c
@@ -693,14 +693,9 @@ static noinline struct btree *bch2_btree_node_fill(struct bch_fs *c,
 	if (!sync)
 		return NULL;
 
-	/*
-	 * XXX: this will probably always fail because btree_iter_relock()
-	 * currently fails for iterators that aren't pointed at a valid btree
-	 * node
-	 */
 	if (iter &&
 	    (!bch2_trans_relock(iter->trans) ||
-	     !bch2_btree_iter_relock(iter, _THIS_IP_)))
+	     !bch2_btree_iter_relock_intent(iter)))
 		return ERR_PTR(-EINTR);
 
 	if (!six_relock_type(&b->c.lock, lock_type, seq))
@@ -760,11 +755,12 @@ static inline void btree_check_header(struct bch_fs *c, struct btree *b)
  * The btree node will have either a read or a write lock held, depending on
  * the @write parameter.
  */
-struct btree *bch2_btree_node_get(struct bch_fs *c, struct btree_iter *iter,
+struct btree *bch2_btree_node_get(struct btree_trans *trans, struct btree_iter *iter,
 				  const struct bkey_i *k, unsigned level,
 				  enum six_lock_type lock_type,
 				  unsigned long trace_ip)
 {
+	struct bch_fs *c = trans->c;
 	struct btree_cache *bc = &c->btree_cache;
 	struct btree *b;
 	struct bset_tree *t;
@@ -838,7 +834,7 @@ lock_node:
 			if (bch2_btree_node_relock(iter, level + 1))
 				goto retry;
 
-			trace_trans_restart_btree_node_reused(iter->trans->ip,
+			trace_trans_restart_btree_node_reused(trans->ip,
 							      trace_ip,
 							      iter->btree_id,
 							      &iter->real_pos);
@@ -850,18 +846,17 @@ lock_node:
 		u32 seq = b->c.lock.state.seq;
 
 		six_unlock_type(&b->c.lock, lock_type);
-		bch2_trans_unlock(iter->trans);
+		bch2_trans_unlock(trans);
 
 		bch2_btree_node_wait_on_read(b);
 
 		/*
-		 * XXX: check if this always fails - btree_iter_relock()
-		 * currently fails for iterators that aren't pointed at a valid
-		 * btree node
+		 * should_be_locked is not set on this iterator yet, so we need
+		 * to relock it specifically:
 		 */
 		if (iter &&
-		    (!bch2_trans_relock(iter->trans) ||
-		     !bch2_btree_iter_relock(iter, _THIS_IP_)))
+		    (!bch2_trans_relock(trans) ||
+		     !bch2_btree_iter_relock_intent(iter)))
 			return ERR_PTR(-EINTR);
 
 		if (!six_relock_type(&b->c.lock, lock_type, seq))
diff --git a/fs/bcachefs/btree_cache.h b/fs/bcachefs/btree_cache.h
index 40dd263a7caa..6e9f08597d94 100644
--- a/fs/bcachefs/btree_cache.h
+++ b/fs/bcachefs/btree_cache.h
@@ -20,7 +20,7 @@ int bch2_btree_cache_cannibalize_lock(struct bch_fs *, struct closure *);
 struct btree *__bch2_btree_node_mem_alloc(struct bch_fs *);
 struct btree *bch2_btree_node_mem_alloc(struct bch_fs *);
 
-struct btree *bch2_btree_node_get(struct bch_fs *, struct btree_iter *,
+struct btree *bch2_btree_node_get(struct btree_trans *, struct btree_iter *,
 				  const struct bkey_i *, unsigned,
 				  enum six_lock_type, unsigned long);
 
diff --git a/fs/bcachefs/btree_iter.c b/fs/bcachefs/btree_iter.c
index 745bf48241fd..5c3404699136 100644
--- a/fs/bcachefs/btree_iter.c
+++ b/fs/bcachefs/btree_iter.c
@@ -205,7 +205,6 @@ static inline bool btree_iter_get_locks(struct btree_iter *iter, bool upgrade,
 					is_btree_node(iter, l)
 					? iter->l[l].b->c.lock.state.seq
 					: 0);
-
 			fail_idx = l;
 			btree_iter_set_dirty(iter, BTREE_ITER_NEED_TRAVERSE);
 		}
@@ -382,6 +381,34 @@ void bch2_btree_trans_verify_locks(struct btree_trans *trans)
 static inline void bch2_btree_iter_verify_locks(struct btree_iter *iter) {}
 #endif
 
+/*
+ * Only for btree_cache.c - only relocks intent locks
+ */
+bool bch2_btree_iter_relock_intent(struct btree_iter *iter)
+{
+	unsigned l;
+
+	for (l = iter->level;
+	     l < iter->locks_want && btree_iter_node(iter, l);
+	     l++) {
+		if (!bch2_btree_node_relock(iter, l)) {
+			trace_node_relock_fail(iter->trans->ip, _RET_IP_,
+					iter->btree_id, &iter->real_pos,
+					l, iter->l[l].lock_seq,
+					is_btree_node(iter, l)
+					? 0
+					: (unsigned long) iter->l[l].b,
+					is_btree_node(iter, l)
+					? iter->l[l].b->c.lock.state.seq
+					: 0);
+			btree_iter_set_dirty(iter, BTREE_ITER_NEED_TRAVERSE);
+			return false;
+		}
+	}
+
+	return true;
+}
+
 __flatten
 bool bch2_btree_iter_relock(struct btree_iter *iter, unsigned long trace_ip)
 {
@@ -1172,7 +1199,8 @@ static noinline void btree_node_mem_ptr_set(struct btree_iter *iter,
 static __always_inline int btree_iter_down(struct btree_iter *iter,
 					   unsigned long trace_ip)
 {
-	struct bch_fs *c = iter->trans->c;
+	struct btree_trans *trans  = iter->trans;
+	struct bch_fs *c = trans->c;
 	struct btree_iter_level *l = &iter->l[iter->level];
 	struct btree *b;
 	unsigned level = iter->level - 1;
@@ -1186,7 +1214,7 @@ static __always_inline int btree_iter_down(struct btree_iter *iter,
 	bch2_bkey_buf_unpack(&tmp, c, l->b,
 			 bch2_btree_node_iter_peek(&l->iter, l->b));
 
-	b = bch2_btree_node_get(c, iter, tmp.k, level, lock_type, trace_ip);
+	b = bch2_btree_node_get(trans, iter, tmp.k, level, lock_type, trace_ip);
 	ret = PTR_ERR_OR_ZERO(b);
 	if (unlikely(ret))
 		goto err;
diff --git a/fs/bcachefs/btree_iter.h b/fs/bcachefs/btree_iter.h
index 7385cca43f8b..3889683e16f8 100644
--- a/fs/bcachefs/btree_iter.h
+++ b/fs/bcachefs/btree_iter.h
@@ -111,6 +111,7 @@ void bch2_btree_node_iter_fix(struct btree_iter *, struct btree *,
 			      struct btree_node_iter *, struct bkey_packed *,
 			      unsigned, unsigned);
 
+bool bch2_btree_iter_relock_intent(struct btree_iter *);
 bool bch2_btree_iter_relock(struct btree_iter *, unsigned long);
 
 bool bch2_trans_relock(struct btree_trans *);