agora inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH 1/2] REPACK: do not require the user to have REPLICATION
560+ messages / 2 participants
[nested] [flat]

* [PATCH 1/2] REPACK: do not require the user to have REPLICATION
@ 2026-04-20 09:38 Álvaro Herrera <[email protected]>
  0 siblings, 0 replies; 560+ messages in thread

From: Álvaro Herrera @ 2026-04-20 09:38 UTC (permalink / raw)

Although REPACK (CONCURRENTLY) uses replication slots, there is no
concern that the slot will leak data of other users, because the
MAINTAIN privilege on the table is required anyway; requiring
REPLICATION is user-unfriendly without providing any actual protection.

A related aspect is that the REPLICATION attribute is not needed to
prevent REPACK from stealing slots from logical replication, since
commit e76d8c749c31 made REPACK use a separate pool of replication
slots.

Because there are now successful concurrent repack runs in the
regression tests, we're forced to run test_plan_advice under
wal_level=replica.

Author: Antonin Houska <[email protected]>
Reported-by: Justin Pryzby <[email protected]>
Reviewed-by: Chao Li <[email protected]>
Discussion: https://postgr.es/m/aeJHPNmL4vVy3oPw@pryzbyj2023
---
 src/backend/commands/repack_worker.c          |  1 -
 .../test_plan_advice/t/001_replan_regress.pl  |  1 +
 src/test/regress/expected/cluster.out         | 20 +++++++++++++++++--
 src/test/regress/sql/cluster.sql              | 11 ++++++++--
 4 files changed, 28 insertions(+), 5 deletions(-)

diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index b17edd771e2..e4a4860805b 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -214,7 +214,6 @@ repack_setup_logical_decoding(Oid relid)
 	/*
 	 * Make sure we can use logical decoding.
 	 */
-	CheckSlotPermissions();
 	CheckLogicalDecodingRequirements(true);
 
 	/*
diff --git a/src/test/modules/test_plan_advice/t/001_replan_regress.pl b/src/test/modules/test_plan_advice/t/001_replan_regress.pl
index 38ffa4d11ae..452b179a665 100644
--- a/src/test/modules/test_plan_advice/t/001_replan_regress.pl
+++ b/src/test/modules/test_plan_advice/t/001_replan_regress.pl
@@ -18,6 +18,7 @@ $node->init();
 # Set up our desired configuration.
 $node->append_conf('postgresql.conf', <<EOM);
 shared_preload_libraries='test_plan_advice'
+wal_level=replica
 pg_plan_advice.always_explain_supplied_advice=false
 pg_plan_advice.feedback_warnings=true
 EOM
diff --git a/src/test/regress/expected/cluster.out b/src/test/regress/expected/cluster.out
index 6127b215a86..e17bc91fae1 100644
--- a/src/test/regress/expected/cluster.out
+++ b/src/test/regress/expected/cluster.out
@@ -543,15 +543,17 @@ ERROR:  REPACK (CONCURRENTLY) is not supported for partitioned tables
 HINT:  Consider running the command on individual partitions.
 DROP TABLE clstrpart;
 -- Ownership of partitions is checked
-CREATE TABLE ptnowner(i int unique) PARTITION BY LIST (i);
+CREATE TABLE ptnowner(i int unique not null) PARTITION BY LIST (i);
 CREATE INDEX ptnowner_i_idx ON ptnowner(i);
 CREATE TABLE ptnowner1 PARTITION OF ptnowner FOR VALUES IN (1);
-CREATE ROLE regress_ptnowner;
+CREATE ROLE regress_ptnowner LOGIN;
 CREATE TABLE ptnowner2 PARTITION OF ptnowner FOR VALUES IN (2);
 ALTER TABLE ptnowner1 OWNER TO regress_ptnowner;
 SET SESSION AUTHORIZATION regress_ptnowner;
 CLUSTER ptnowner USING ptnowner_i_idx;
 ERROR:  permission denied for table ptnowner
+ALTER TABLE ptnowner1 REPLICA IDENTITY USING INDEX ptnowner1_i_key;
+REPACK (CONCURRENTLY) ptnowner1;
 RESET SESSION AUTHORIZATION;
 ALTER TABLE ptnowner OWNER TO regress_ptnowner;
 CREATE TEMP TABLE ptnowner_oldnodes AS
@@ -560,6 +562,11 @@ CREATE TEMP TABLE ptnowner_oldnodes AS
 SET SESSION AUTHORIZATION regress_ptnowner;
 CLUSTER ptnowner USING ptnowner_i_idx;
 WARNING:  permission denied to execute CLUSTER on "ptnowner2", skipping it
+-- still can't repack without a replica identity
+ALTER TABLE ptnowner1 REPLICA IDENTITY DEFAULT;
+REPACK (CONCURRENTLY) ptnowner1;
+ERROR:  cannot process relation "ptnowner1"
+HINT:  Relation "ptnowner1" has no identity index.
 RESET SESSION AUTHORIZATION;
 SELECT a.relname, a.relfilenode=b.relfilenode FROM pg_class a
   JOIN ptnowner_oldnodes b USING (oid) ORDER BY a.relname COLLATE "C";
@@ -570,6 +577,15 @@ SELECT a.relname, a.relfilenode=b.relfilenode FROM pg_class a
  ptnowner2 | t
 (3 rows)
 
+SELECT a.relname, a.relfilenode=b.relfilenode FROM pg_class a
+  JOIN ptnowner_oldnodes b USING (oid) ORDER BY a.relname COLLATE "C";
+  relname  | ?column? 
+-----------+----------
+ ptnowner  | t
+ ptnowner1 | f
+ ptnowner2 | t
+(3 rows)
+
 DROP TABLE ptnowner;
 DROP ROLE regress_ptnowner;
 -- Test CLUSTER with external tuplesorting
diff --git a/src/test/regress/sql/cluster.sql b/src/test/regress/sql/cluster.sql
index d14063a9683..1f471a8821a 100644
--- a/src/test/regress/sql/cluster.sql
+++ b/src/test/regress/sql/cluster.sql
@@ -254,14 +254,16 @@ REPACK (CONCURRENTLY) clstrpart;
 DROP TABLE clstrpart;
 
 -- Ownership of partitions is checked
-CREATE TABLE ptnowner(i int unique) PARTITION BY LIST (i);
+CREATE TABLE ptnowner(i int unique not null) PARTITION BY LIST (i);
 CREATE INDEX ptnowner_i_idx ON ptnowner(i);
 CREATE TABLE ptnowner1 PARTITION OF ptnowner FOR VALUES IN (1);
-CREATE ROLE regress_ptnowner;
+CREATE ROLE regress_ptnowner LOGIN;
 CREATE TABLE ptnowner2 PARTITION OF ptnowner FOR VALUES IN (2);
 ALTER TABLE ptnowner1 OWNER TO regress_ptnowner;
 SET SESSION AUTHORIZATION regress_ptnowner;
 CLUSTER ptnowner USING ptnowner_i_idx;
+ALTER TABLE ptnowner1 REPLICA IDENTITY USING INDEX ptnowner1_i_key;
+REPACK (CONCURRENTLY) ptnowner1;
 RESET SESSION AUTHORIZATION;
 ALTER TABLE ptnowner OWNER TO regress_ptnowner;
 CREATE TEMP TABLE ptnowner_oldnodes AS
@@ -269,7 +271,12 @@ CREATE TEMP TABLE ptnowner_oldnodes AS
   JOIN pg_class AS c ON c.oid=tree.relid;
 SET SESSION AUTHORIZATION regress_ptnowner;
 CLUSTER ptnowner USING ptnowner_i_idx;
+-- still can't repack without a replica identity
+ALTER TABLE ptnowner1 REPLICA IDENTITY DEFAULT;
+REPACK (CONCURRENTLY) ptnowner1;
 RESET SESSION AUTHORIZATION;
+SELECT a.relname, a.relfilenode=b.relfilenode FROM pg_class a
+  JOIN ptnowner_oldnodes b USING (oid) ORDER BY a.relname COLLATE "C";
 SELECT a.relname, a.relfilenode=b.relfilenode FROM pg_class a
   JOIN ptnowner_oldnodes b USING (oid) ORDER BY a.relname COLLATE "C";
 DROP TABLE ptnowner;
-- 
2.47.3


--m6zy3l65ushq557m
Content-Type: text/x-diff; charset=utf-8
Content-Disposition: attachment;
	filename="0002-REPACK-do-not-require-LOGIN-privileges.patch"



^ permalink  raw  reply  [nested|flat] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ 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; 560+ 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] 560+ messages in thread