0003-patch-2-gin_check_parent_keys_consistency.patch

From a7dfd221a809ccd062fec36cd2b7acf025642bfb Mon Sep 17 00:00:00 2001
From: Arseniy Mukhin <arseniy.mukhin.dev@gmail.com>
Date: Mon, 9 Jun 2025 20:39:13 +0300
Subject: [PATCH 3/5] patch 2: gin_check_parent_keys_consistency

This commit address issues with parent_key checks for the entry tree in gin_index_check():
- parenttup is always NULL. We want to set parenttup to NULL for the rightmost tuple of the level only, in all other cases it should be valid parent_key.
- use GinGetDownlink while retrieving child blkno to avoid triggering Assert (as core GIN code does).
---
 contrib/amcheck/t/006_verify_gin.pl | 80 +++++++++++++++++++++++++++++
 contrib/amcheck/verify_gin.c        |  8 +--
 2 files changed, 84 insertions(+), 4 deletions(-)

diff --git a/contrib/amcheck/t/006_verify_gin.pl b/contrib/amcheck/t/006_verify_gin.pl
index 46f693fbb08..fa6baa51546 100644
--- a/contrib/amcheck/t/006_verify_gin.pl
+++ b/contrib/amcheck/t/006_verify_gin.pl
@@ -31,6 +31,8 @@ $node->safe_psql(
 invalid_entry_order_leaf_page_test();
 invalid_entry_order_inner_page_test();
 invalid_entry_columns_order_test();
+inconsistent_with_parent_key__parent_key_corrupted_test();
+inconsistent_with_parent_key__child_key_corrupted_test();
 
 sub invalid_entry_order_leaf_page_test
 {
@@ -161,6 +163,84 @@ sub invalid_entry_columns_order_test
 	like($stderr, qr/$expected/);
 }
 
+sub inconsistent_with_parent_key__parent_key_corrupted_test
+{
+	my $relname = "test";
+	my $indexname = "test_gin_idx";
+
+	$node->safe_psql(
+		'postgres', qq(
+		DROP TABLE IF EXISTS $relname;
+		CREATE TABLE $relname (a text[]);
+		CREATE INDEX $indexname ON $relname USING gin (a);
+		INSERT INTO $relname (a) VALUES (('{' || 'llllllllll' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'mmmmmmmmmm' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'nnnnnnnnnn' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'xxxxxxxxxx' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'yyyyyyyyyy' || random_string(1870) ||'}')::text[]);
+		SELECT gin_clean_pending_list('$indexname');
+	));
+	my $relpath = relation_filepath($indexname);
+
+	$node->stop;
+
+	my $blkno = 1;  # root
+
+	# we have nnnnnnnnnn... as parent key in the root, so replace it with something smaller then child's keys
+	string_replace_block(
+		$relpath,
+		'nnnnnnnnnn',
+		'aaaaaaaaaa',
+		$blksize,
+		$blkno
+	);
+
+	$node->start;
+
+	my ($result, $stdout, $stderr) = $node->psql('postgres', qq(SELECT gin_index_check('$indexname')));
+	my $expected = "index \"$indexname\" has inconsistent records on page 5 offset 3";
+	like($stderr, qr/$expected/);
+}
+
+sub inconsistent_with_parent_key__child_key_corrupted_test
+{
+	my $relname = "test";
+	my $indexname = "test_gin_idx";
+
+	$node->safe_psql(
+		'postgres', qq(
+		DROP TABLE IF EXISTS $relname;
+		CREATE TABLE $relname (a text[]);
+		CREATE INDEX $indexname ON $relname USING gin (a);
+		INSERT INTO $relname (a) VALUES (('{' || 'llllllllll' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'mmmmmmmmmm' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'nnnnnnnnnn' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'xxxxxxxxxx' || random_string(1870) ||'}')::text[]);
+		INSERT INTO $relname (a) VALUES (('{' || 'yyyyyyyyyy' || random_string(1870) ||'}')::text[]);
+		SELECT gin_clean_pending_list('$indexname');
+	 ));
+	my $relpath = relation_filepath($indexname);
+
+	$node->stop;
+
+	my $blkno = 5;  # leaf
+
+	# we have nnnnnnnnnn... as parent key in the root, so replace child key with something bigger
+	string_replace_block(
+		$relpath,
+		'nnnnnnnnnn',
+		'pppppppppp',
+		$blksize,
+		$blkno
+	);
+
+	$node->start;
+
+	my ($result, $stdout, $stderr) = $node->psql('postgres', qq(SELECT gin_index_check('$indexname')));
+	my $expected = "index \"$indexname\" has inconsistent records on page 5 offset 3";
+	like($stderr, qr/$expected/);
+}
+
 # Returns the filesystem path for the named relation.
 sub relation_filepath
 {
diff --git a/contrib/amcheck/verify_gin.c b/contrib/amcheck/verify_gin.c
index 26b98571b56..8f6a5410cb7 100644
--- a/contrib/amcheck/verify_gin.c
+++ b/contrib/amcheck/verify_gin.c
@@ -609,10 +609,10 @@ gin_check_parent_keys_consistency(Relation rel,
 				ptr = (GinScanItem *) palloc(sizeof(GinScanItem));
 				ptr->depth = stack->depth + 1;
 				/* last tuple in layer has no high key */
-				if (i != maxoff && !GinPageGetOpaque(page)->rightlink)
-					ptr->parenttup = CopyIndexTuple(idxtuple);
-				else
+				if (i == maxoff && rightlink == InvalidBlockNumber)
 					ptr->parenttup = NULL;
+				else
+					ptr->parenttup = CopyIndexTuple(idxtuple);
 				ptr->parentblk = stack->blkno;
 				ptr->blkno = GinGetDownlink(idxtuple);
 				ptr->parentlsn = lsn;
@@ -750,7 +750,7 @@ gin_refind_parent(Relation rel, BlockNumber parentblkno,
 		ItemId		p_iid = PageGetItemIdCareful(rel, parentblkno, parentpage, o);
 		IndexTuple	itup = (IndexTuple) PageGetItem(parentpage, p_iid);
 
-		if (ItemPointerGetBlockNumber(&(itup->t_tid)) == childblkno)
+		if (GinGetDownlink(itup) == childblkno)
 		{
 			/* Found it! Make copy and return it */
 			result = CopyIndexTuple(itup);
-- 
2.43.0