bcachefs: Improved inode create optimization

This shards new inodes into different btree nodes by using the processor
ID for the high bits of the new inode number. Much faster than the
previous inode create optimization - this also helps with sharding in
the other btrees that index by inode number.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>

fs/bcachefs/bcachefs.h

@@ -814,11 +814,8 @@ mempool_t		bio_bounce_pages;
 	struct mutex		verify_lock;
 #endif
 
-	struct mutex		inode_create_lock;
-	unsigned		unused_inodes_nr;
-	u64			unused_inodes[64];
-	u32			unused_inodes_gens[64];
-	u64			unused_inode_hint;
+	u64			*unused_inode_hints;
+	unsigned		inode_shard_bits;
 
 	/*
 	 * A btree node on disk could have too many bsets for an iterator to fit

fs/bcachefs/inode.c

@@ -361,55 +361,6 @@ static inline u32 bkey_generation(struct bkey_s_c k)
 	}
 }
 
-static int scan_free_inums(struct btree_trans *trans)
-{
-	struct bch_fs *c = trans->c;
-	struct btree_iter *iter = NULL;
-	struct bkey_s_c k;
-	u64 min = BLOCKDEV_INODE_MAX;
-	u64 max = c->opts.inodes_32bit
-		? S32_MAX : S64_MAX;
-	u64 start = max(min, READ_ONCE(c->unused_inode_hint));
-	int ret = 0;
-
-	iter = bch2_trans_get_iter(trans, BTREE_ID_INODES, POS(0, start),
-				   BTREE_ITER_SLOTS);
-	if (IS_ERR(iter))
-		return PTR_ERR(iter);
-again:
-	for_each_btree_key_continue(iter, BTREE_ITER_SLOTS, k, ret) {
-		if (bkey_cmp(iter->pos, POS(0, max)) > 0)
-			break;
-
-		/*
-		 * This doesn't check the btree key cache, but we don't care:
-		 * we have to recheck with an intent lock held on the slot we're
-		 * inserting to anyways:
-		 */
-		if (k.k->type != KEY_TYPE_inode) {
-			if (c->unused_inodes_nr < ARRAY_SIZE(c->unused_inodes)) {
-				c->unused_inodes[c->unused_inodes_nr] = k.k->p.offset;
-				c->unused_inodes_gens[c->unused_inodes_nr] = bkey_generation(k);
-				c->unused_inodes_nr++;
-			}
-
-			if (c->unused_inodes_nr == ARRAY_SIZE(c->unused_inodes))
-				goto out;
-		}
-	}
-
-	if (!ret && start != min) {
-		max = start;
-		start = min;
-		bch2_btree_iter_set_pos(iter, POS(0, start));
-		goto again;
-	}
-out:
-	c->unused_inode_hint = iter->pos.offset;
-	bch2_trans_iter_put(trans, iter);
-	return ret;
-}
-
 int bch2_inode_create(struct btree_trans *trans,
 		      struct bch_inode_unpacked *inode_u)
 {
@@ -417,64 +368,68 @@ int bch2_inode_create(struct btree_trans *trans,
 	struct bkey_inode_buf *inode_p;
 	struct btree_iter *iter = NULL;
 	struct bkey_s_c k;
-	u64 inum;
-	u32 generation;
-	int ret = 0;
+	u64 min, max, start, *hint;
+	int ret;
+
+	unsigned cpu = raw_smp_processor_id();
+	unsigned bits = (c->opts.inodes_32bit
+		? 31 : 63) - c->inode_shard_bits;
+
+	min = (cpu << bits);
+	max = (cpu << bits) | ~(ULLONG_MAX << bits);
+
+	min = max_t(u64, min, BLOCKDEV_INODE_MAX);
+	hint = c->unused_inode_hints + cpu;
+
+	start = READ_ONCE(*hint);
+
+	if (start >= max || start < min)
+		start = min;
 
 	inode_p = bch2_trans_kmalloc(trans, sizeof(*inode_p));
 	if (IS_ERR(inode_p))
 		return PTR_ERR(inode_p);
+again:
+	for_each_btree_key(trans, iter, BTREE_ID_INODES, POS(0, start),
+			   BTREE_ITER_SLOTS|BTREE_ITER_INTENT, k, ret) {
+		if (bkey_cmp(iter->pos, POS(0, max)) > 0)
+			break;
 
-	iter = bch2_trans_get_iter(trans, BTREE_ID_INODES, POS_MIN,
-				   BTREE_ITER_CACHED|
-				   BTREE_ITER_INTENT);
-	if (IS_ERR(iter))
-		return PTR_ERR(iter);
-retry:
-	if (!mutex_trylock(&c->inode_create_lock)) {
-		bch2_trans_unlock(trans);
-		mutex_lock(&c->inode_create_lock);
-		if (!bch2_trans_relock(trans)) {
-			mutex_unlock(&c->inode_create_lock);
-			ret = -EINTR;
-			goto err;
-		}
-	}
-
-	if (!c->unused_inodes_nr)
-		ret = scan_free_inums(trans);
-	if (!ret && !c->unused_inodes_nr)
-		ret = -ENOSPC;
-	if (!ret) {
-		--c->unused_inodes_nr;
-		inum		= c->unused_inodes[c->unused_inodes_nr];
-		generation	= c->unused_inodes_gens[c->unused_inodes_nr];
+		/*
+		 * There's a potential cache coherency issue with the btree key
+		 * cache code here - we're iterating over the btree, skipping
+		 * that cache. We should never see an empty slot that isn't
+		 * actually empty due to a pending update in the key cache
+		 * because the update that creates the inode isn't done with a
+		 * cached iterator, but - better safe than sorry, check the
+		 * cache before using a slot:
+		 */
+		if (k.k->type != KEY_TYPE_inode &&
+		    !bch2_btree_key_cache_find(c, BTREE_ID_INODES, iter->pos))
+			goto found_slot;
 	}
 
-	mutex_unlock(&c->inode_create_lock);
-
-	if (ret)
-		goto err;
-
-	bch2_btree_iter_set_pos(iter, POS(0, inum));
+	bch2_trans_iter_put(trans, iter);
 
-	/* Recheck that the slot is free with an intent lock held: */
-	k = bch2_btree_iter_peek_cached(iter);
-	ret = bkey_err(k);
 	if (ret)
-		goto err;
+		return ret;
 
-	if (k.k->type == KEY_TYPE_inode)
-		goto retry;
+	if (start != min) {
+		/* Retry from start */
+		start = min;
+		goto again;
+	}
 
-	inode_u->bi_inum	= inum;
-	inode_u->bi_generation	= generation;
+	return -ENOSPC;
+found_slot:
+	*hint			= k.k->p.offset;
+	inode_u->bi_inum	= k.k->p.offset;
+	inode_u->bi_generation	= bkey_generation(k);
 
 	bch2_inode_pack(inode_p, inode_u);
 	bch2_trans_update(trans, iter, &inode_p->inode.k_i, 0);
-err:
 	bch2_trans_iter_put(trans, iter);
-	return ret;
+	return 0;
 }
 
 int bch2_inode_rm(struct bch_fs *c, u64 inode_nr)

fs/bcachefs/super.c

@@ -493,6 +493,7 @@ static void __bch2_fs_free(struct bch_fs *c)
 	kfree(c->replicas_gc.entries);
 	kfree(rcu_dereference_protected(c->disk_groups, 1));
 	kfree(c->journal_seq_blacklist_table);
+	kfree(c->unused_inode_hints);
 	free_heap(&c->copygc_heap);
 
 	if (c->journal_reclaim_wq)
@@ -703,8 +704,6 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 
 	seqcount_init(&c->usage_lock);
 
-	mutex_init(&c->inode_create_lock);
-
 	c->copy_gc_enabled		= 1;
 	c->rebalance.enabled		= 1;
 	c->promote_whole_extents	= true;
@@ -746,6 +745,8 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 		(btree_blocks(c) + 1) * 2 *
 		sizeof(struct sort_iter_set);
 
+	c->inode_shard_bits = ilog2(roundup_pow_of_two(num_possible_cpus()));
+
 	if (!(c->wq = alloc_workqueue("bcachefs",
 				WQ_FREEZABLE|WQ_MEM_RECLAIM, 1)) ||
 	    !(c->copygc_wq = alloc_workqueue("bcachefs_copygc",
@@ -764,6 +765,8 @@ static struct bch_fs *bch2_fs_alloc(struct bch_sb *sb, struct bch_opts opts)
 	    mempool_init_kvpmalloc_pool(&c->btree_bounce_pool, 1,
 					btree_bytes(c)) ||
 	    mempool_init_kmalloc_pool(&c->large_bkey_pool, 1, 2048) ||
+	    !(c->unused_inode_hints = kcalloc(1U << c->inode_shard_bits,
+					      sizeof(u64), GFP_KERNEL)) ||
 	    bch2_io_clock_init(&c->io_clock[READ]) ||
 	    bch2_io_clock_init(&c->io_clock[WRITE]) ||
 	    bch2_fs_journal_init(&c->journal) ||