Reduce relcache access in WAL sender streaming logical changes

get_rel_sync_entry(), which is called each time a change needs to be
logically replicated, is a rather hot code path in the WAL sender
sending logical changes.  This code path was doing a relcache access on
relkind and relpartition for each logical change, but we only need to
know this information when building or re-building the cached
information for a relation.

Some measurements prove that this is noticeable in perf profiles,
particularly when attempting to replicate changes from relations that
are not published as these cause less overhead in the WAL sender,
delaying further the replication of changes for relations that are
published.

Issue introduced in 83fd453.

Author: Hou Zhijie
Reviewed-by: Kyotaro Horiguchi, Euler Taveira
Discussion: https://postgr.es/m/OS0PR01MB5716E863AA9E591C1F010F7A947D9@OS0PR01MB5716.jpnprd01.prod.outlook.com
Backpatch-through: 13

diff --git a/src/backend/replication/pgoutput/pgoutput.c b/src/backend/replication/pgoutput/pgoutput.c
index 6f6a203dea77c990532e8c4f3ad5b60bb371a268..a08da859b44253b29dbcc7c04c5ae4e52de07e2e 100644 (file)
--- a/src/backend/replication/pgoutput/pgoutput.c
+++ b/src/backend/replication/pgoutput/pgoutput.c
@@ -1119,8 +1119,6 @@ static RelationSyncEntry *
 get_rel_sync_entry(PGOutputData *data, Oid relid)
 {
    RelationSyncEntry *entry;
-   bool        am_partition = get_rel_relispartition(relid);
-   char        relkind = get_rel_relkind(relid);
    bool        found;
    MemoryContext oldctx;
 
@@ -1160,6 +1158,8 @@ get_rel_sync_entry(PGOutputData *data, Oid relid)
        List       *schemaPubids = GetSchemaPublications(schemaId);
        ListCell   *lc;
        Oid         publish_as_relid = relid;
+       bool        am_partition = get_rel_relispartition(relid);
+       char        relkind = get_rel_relkind(relid);
 
        /* Reload publications if needed before use. */
        if (!publications_valid)