agora inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH 2/3] one more stress test for repack concurrently
338+ messages / 2 participants
[nested] [flat]

* [PATCH 2/3] one more stress test for repack concurrently
@ 2025-12-13 17:46 Mikhail Nikalayeu <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Mikhail Nikalayeu @ 2025-12-13 17:46 UTC (permalink / raw)

---
 contrib/amcheck/meson.build       |   1 +
 contrib/amcheck/t/008_repack_2.pl | 111 ++++++++++++++++++++++++++++++
 2 files changed, 112 insertions(+)
 create mode 100644 contrib/amcheck/t/008_repack_2.pl

diff --git a/contrib/amcheck/meson.build b/contrib/amcheck/meson.build
index cb4bc32e98a..6799df214f9 100644
--- a/contrib/amcheck/meson.build
+++ b/contrib/amcheck/meson.build
@@ -51,6 +51,7 @@ tests += {
       't/005_pitr.pl',
       't/006_verify_gin.pl',
       't/007_repack_1.pl',
+      't/008_repack_2.pl',
     ],
   },
 }
diff --git a/contrib/amcheck/t/008_repack_2.pl b/contrib/amcheck/t/008_repack_2.pl
new file mode 100644
index 00000000000..a287ee2bb71
--- /dev/null
+++ b/contrib/amcheck/t/008_repack_2.pl
@@ -0,0 +1,111 @@
+# Copyright (c) 2021-2026, PostgreSQL Global Development Group
+
+# Test REPACK CONCURRENTLY with concurrent modifications
+use strict;
+use warnings FATAL => 'all';
+
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+
+use Test::More;
+
+# PG_TEST_EXTRA carries a numerical stress value; 0 disables the test,
+# 1 means the test takes about 6 seconds, and the increase should be
+# roughly linear.
+my $matched = ($ENV{PG_TEST_EXTRA} // '') =~ /\bstress_concurrently(?:=(?<stressval>[0-9]*))?\b/;
+my $stressval = $+{stressval} // 1;
+if (!$matched or $stressval == 0)
+{
+	plan skip_all => 'skipping disabled REPACK CONCURRENTLY stress test';
+}
+
+my $node;
+
+#
+# Test set-up
+#
+$node = PostgreSQL::Test::Cluster->new('CIC_test');
+$node->init;
+$node->append_conf('postgresql.conf',
+	'lock_timeout = ' . (1000 * $PostgreSQL::Test::Utils::timeout_default));
+$node->append_conf(
+	'postgresql.conf', qq(
+wal_level = logical
+));
+
+my $duration = 6 * $stressval;
+my $no_hot = int(rand(2));
+
+$node->start;
+$node->safe_psql('postgres', q(CREATE TABLE tbl(id SERIAL PRIMARY KEY, val int)));
+if ($no_hot)
+{
+	$node->safe_psql('postgres', q(CREATE INDEX test_idx ON tbl(val);));
+}
+else
+{
+	$node->safe_psql('postgres', q(CREATE INDEX test_idx ON tbl(id);));
+}
+
+# Load amcheck
+$node->safe_psql('postgres', q(CREATE EXTENSION amcheck));
+
+my $sum = $node->safe_psql('postgres', q(
+	SELECT SUM(val) AS sum FROM tbl
+));
+
+$node->safe_psql('postgres', q(CREATE UNLOGGED SEQUENCE last_j START 1 INCREMENT 1;));
+
+
+$node->pgbench(
+"--no-vacuum --client=30 --jobs=4 --exit-on-abort -T $duration",
+0,
+[qr{actually processed}],
+[qr{^$}],
+'concurrent operations with REINDEX/CREATE INDEX CONCURRENTLY',
+{
+	'concurrent_ops' => qq(
+		SELECT pg_try_advisory_lock(42)::integer AS gotlock \\gset
+		\\if :gotlock
+			REPACK (CONCURRENTLY) tbl USING INDEX tbl_pkey;
+			SELECT bt_index_parent_check('tbl_pkey', heapallindexed => true);
+			SELECT bt_index_parent_check('test_idx', heapallindexed => true);
+			\\sleep 10 ms
+
+			REPACK (CONCURRENTLY) tbl USING INDEX test_idx;
+			SELECT bt_index_parent_check('tbl_pkey', heapallindexed => true);
+			SELECT bt_index_parent_check('test_idx', heapallindexed => true);
+			\\sleep 10 ms
+
+			REPACK (CONCURRENTLY) tbl;
+			SELECT bt_index_parent_check('tbl_pkey', heapallindexed => true);
+			SELECT bt_index_parent_check('test_idx', heapallindexed => true);
+			\\sleep 10 ms
+
+			SELECT pg_advisory_unlock(42);
+		\\else
+			SELECT pg_advisory_lock(43);
+				BEGIN;
+				INSERT INTO tbl(val) VALUES (nextval('last_j')) RETURNING val \\gset p_
+				COMMIT;
+			SELECT pg_advisory_unlock(43);
+			\\sleep 1 ms
+
+			BEGIN
+			--TRANSACTION ISOLATION LEVEL REPEATABLE READ
+			;
+			SELECT 1;
+			\\sleep 1 ms
+			SELECT COUNT(*) AS count FROM tbl WHERE val <= :p_j \\gset p_
+			\\if :p_count != :p_j
+				COMMIT;
+				SELECT (:p_count) / 0;
+			\\endif
+
+			COMMIT;
+		\\endif
+	)
+});
+
+$node->stop;
+done_testing();
-- 
2.47.3


--tbdehtstrqwksfdh
Content-Type: text/x-diff; charset=utf-8
Content-Disposition: attachment;
	filename="0003-Distinguish-properly-when-database-specific-transact.patch"



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread

* [PATCH 7/8] Decouple updating of freezing information from swap_relation_files().
@ 2026-06-16 11:54 Antonin Houska <[email protected]>
  0 siblings, 0 replies; 338+ messages in thread

From: Antonin Houska @ 2026-06-16 11:54 UTC (permalink / raw)

So far, the relfrozenxid and relminmxid attributes of pg_class used to be
updated by swap_relation_files(), however, not every call of the function
should do that. On the contrary, new feature of REPACK (CONCURRENTLY) will
need to update this information in one case that would be tricky to handle
using swap_relation_files().
---
 src/backend/commands/repack.c | 114 +++++++++++++++++-----------------
 1 file changed, 56 insertions(+), 58 deletions(-)

diff --git a/src/backend/commands/repack.c b/src/backend/commands/repack.c
index 6f3e4ecff46..92c1b678e6f 100644
--- a/src/backend/commands/repack.c
+++ b/src/backend/commands/repack.c
@@ -186,6 +186,8 @@ static void copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldInde
 							TransactionId *pFreezeXid,
 							MultiXactId *pCutoffMulti,
 							ChangeContext *chgcxt);
+static void update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+									MultiXactId cutoffMulti);
 static List *get_tables_to_repack(RepackCommand cmd, bool usingindex,
 								  MemoryContext permcxt);
 static List *get_tables_to_repack_partitioned(RepackCommand cmd,
@@ -2014,13 +2016,6 @@ copy_table_data(Relation NewHeap, Relation OldHeap, Relation OldIndex,
  * links) while the latter is the only way to handle cases in which a toast
  * table is added or removed altogether.
  *
- * Additionally, the first relation is marked with relfrozenxid set to
- * frozenXid.  It seems a bit ugly to have this here, but the caller would
- * have to do it anyway, so having it here saves a heap_update.  Note: in
- * the swap-toast-links case, we assume we don't need to change the toast
- * table's relfrozenxid: the new version of the toast table should already
- * have relfrozenxid set to RecentXmin, which is good enough.
- *
  * Lastly, if r2 and its toast table and toast index (if any) are mapped,
  * their OIDs are emitted into mapped_tables[].  This is hacky but beats
  * having to look the information up again later in finish_heap_swap.
@@ -2029,9 +2024,7 @@ static void
 swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 					bool swap_toast_by_content,
 					bool is_internal,
-					TransactionId frozenXid,
-					MultiXactId cutoffMulti,
-					Oid *mapped_tables)
+					Oid *mapped_tables, Oid *r1_toastid)
 {
 	Relation	relRelation;
 	HeapTuple	reltup1,
@@ -2063,6 +2056,9 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	relam1 = relform1->relam;
 	relam2 = relform2->relam;
 
+	if (r1_toastid)
+		*r1_toastid = relform1->reltoastrelid;
+
 	if (RelFileNumberIsValid(relfilenumber1) &&
 		RelFileNumberIsValid(relfilenumber2))
 	{
@@ -2178,15 +2174,6 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	 * and then fail to commit the pg_class update.
 	 */
 
-	/* set rel1's frozen Xid and minimum MultiXid */
-	if (relform1->relkind != RELKIND_INDEX)
-	{
-		Assert(!TransactionIdIsValid(frozenXid) ||
-			   TransactionIdIsNormal(frozenXid));
-		relform1->relfrozenxid = frozenXid;
-		relform1->relminmxid = cutoffMulti;
-	}
-
 	/* swap size statistics too, since new rel has freshly-updated stats */
 	{
 		int32		swap_pages;
@@ -2288,9 +2275,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 									target_is_pg_class,
 									swap_toast_by_content,
 									is_internal,
-									frozenXid,
-									cutoffMulti,
-									mapped_tables);
+									mapped_tables,
+									NULL);
 			}
 			else
 			{
@@ -2391,9 +2377,8 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 							target_is_pg_class,
 							swap_toast_by_content,
 							is_internal,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 	}
 
 	/* Clean up. */
@@ -2403,6 +2388,36 @@ swap_relation_files(Oid r1, Oid r2, bool target_is_pg_class,
 	table_close(relRelation, RowExclusiveLock);
 }
 
+/*
+ * Update relfrozenxid and relminmxid attributes of pg_class entry, specified
+ * by OID.
+ */
+static void
+update_relation_cutoffs(Oid relid, TransactionId frozenXid,
+						MultiXactId cutoffMulti)
+{
+	Relation	relRelation;
+	HeapTuple	reltup;
+	Form_pg_class relform;
+	CatalogIndexState indstate;
+
+	relRelation = table_open(RelationRelationId, RowExclusiveLock);
+	reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(relid));
+	if (!HeapTupleIsValid(reltup))
+		elog(ERROR, "cache lookup failed for relation %u", relid);
+	relform = (Form_pg_class) GETSTRUCT(reltup);
+	relform->relfrozenxid = frozenXid;
+	relform->relminmxid = cutoffMulti;
+
+	indstate = CatalogOpenIndexes(relRelation);
+	CatalogTupleUpdateWithInfo(relRelation, &reltup->t_self, reltup,
+							   indstate);
+	CatalogCloseIndexes(indstate);
+
+	heap_freetuple(reltup);
+	table_close(relRelation, RowExclusiveLock);
+}
+
 /*
  * Remove the transient table that was built by make_new_heap, and finish
  * cleaning up (including rebuilding all indexes on the old heap).
@@ -2420,6 +2435,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 {
 	ObjectAddress object;
 	Oid			mapped_tables[4];
+	Oid			oid_old_toastid;
 	int			i;
 
 	/* Report that we are now swapping relation files */
@@ -2436,7 +2452,7 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 	swap_relation_files(OIDOldHeap, OIDNewHeap,
 						(OIDOldHeap == RelationRelationId),
 						swap_toast_by_content, is_internal,
-						frozenXid, cutoffMulti, mapped_tables);
+						mapped_tables, &oid_old_toastid);
 
 	/*
 	 * If it's a system catalog, queue a sinval message to flush all catcaches
@@ -2492,37 +2508,20 @@ finish_heap_swap(Oid OIDOldHeap, Oid OIDNewHeap,
 								 PROGRESS_REPACK_PHASE_FINAL_CLEANUP);
 
 	/*
-	 * If the relation being rebuilt is pg_class, swap_relation_files()
-	 * couldn't update pg_class's own pg_class entry (check comments in
-	 * swap_relation_files()), thus relfrozenxid was not updated. That's
-	 * annoying because a potential reason for doing a VACUUM FULL is a
-	 * imminent or actual anti-wraparound shutdown.  So, now that we can
-	 * access the new relation using its indices, update relfrozenxid.
-	 * pg_class doesn't have a toast relation, so we don't need to update the
-	 * corresponding toast relation. Not that there's little point moving all
-	 * relfrozenxid updates here since swap_relation_files() needs to write to
-	 * pg_class for non-mapped relations anyway.
+	 * Update relfrozenxid and relminmxid. CCI is needed to see the most
+	 * recent pg_class tuple, created by swap_relation_files(). However that
+	 * CCI makes us use the new relation file. Therefore we cannot do this
+	 * before indexes have been rebuilt too.
 	 */
-	if (OIDOldHeap == RelationRelationId)
-	{
-		Relation	relRelation;
-		HeapTuple	reltup;
-		Form_pg_class relform;
-
-		relRelation = table_open(RelationRelationId, RowExclusiveLock);
-
-		reltup = SearchSysCacheCopy1(RELOID, ObjectIdGetDatum(OIDOldHeap));
-		if (!HeapTupleIsValid(reltup))
-			elog(ERROR, "cache lookup failed for relation %u", OIDOldHeap);
-		relform = (Form_pg_class) GETSTRUCT(reltup);
-
-		relform->relfrozenxid = frozenXid;
-		relform->relminmxid = cutoffMulti;
-
-		CatalogTupleUpdate(relRelation, &reltup->t_self, reltup);
+	CommandCounterIncrement();
+	update_relation_cutoffs(OIDOldHeap, frozenXid, cutoffMulti);
 
-		table_close(relRelation, RowExclusiveLock);
-	}
+	/*
+	 * The same for TOAST, if needed. In the swap-toast-links case, the new
+	 * the toast table should already have relfrozenxid set to RecentXmin.
+	 */
+	if (OidIsValid(oid_old_toastid) && swap_toast_by_content)
+		update_relation_cutoffs(oid_old_toastid, frozenXid, cutoffMulti);
 
 	/* Destroy new heap with old filenumber */
 	object.classId = RelationRelationId;
@@ -4072,9 +4071,8 @@ rebuild_relation_finish_concurrent(Relation NewHeap, Relation OldHeap,
 							(old_table_oid == RelationRelationId),
 							false,	/* swap_toast_by_content */
 							true,
-							InvalidTransactionId,
-							InvalidMultiXactId,
-							mapped_tables);
+							mapped_tables,
+							NULL);
 
 #ifdef USE_ASSERT_CHECKING
 
-- 
2.52.0


--=-=-=
Content-Type: text/plain
Content-Disposition: attachment;
 filename=v01-0008-Make-REPACK-CONCURRENTLY-MVCC-safe.patch



^ permalink  raw  reply  [nested|flat] 338+ messages in thread