Fix "invalid spinlock number: 0" error in pg_stat_wal_receiver.

Commit 2c8dd05 added the atomic variable writtenUpto into
walreceiver's shared memory information. It's initialized only
when walreceiver started up but could be read via pg_stat_wal_receiver
view anytime, i.e., even before it's initialized. In the server built
with --disable-atomics and --disable-spinlocks, this uninitialized
atomic variable read could cause "invalid spinlock number: 0" error.

This commit changed writtenUpto so that it's initialized at
the postmaster startup, to avoid the uninitialized variable read
via pg_stat_wal_receiver and fix the error.

Also this commit moved the read of writtenUpto after the release
of spinlock protecting walreceiver's shared variables. This is
necessary to prevent new spinlock from being taken by atomic
variable read while holding another spinlock, and to shorten
the spinlock duration. This change leads writtenUpto not to be
consistent with the other walreceiver's shared variables protected
by a spinlock. But this is OK because writtenUpto should not be
used for data integrity checks.

Back-patch to v13 where commit 2c8dd05 introduced the bug.

Author: Fujii Masao
Reviewed-by: Michael Paquier, Thomas Munro, Andres Freund
Discussion: https://postgr.es/m/7ef8708c-5b6b-edd3-2cf2-7783f1c7c175@oss.nttdata.com

Author: MasaoFujii

src/backend/replication/walreceiver.c

@@ -249,7 +249,7 @@ WalReceiverMain(void)
 
 	SpinLockRelease(&walrcv->mutex);
 
-	pg_atomic_init_u64(&WalRcv->writtenUpto, 0);
+	pg_atomic_write_u64(&WalRcv->writtenUpto, 0);
 
 	/* Arrange to clean up at walreceiver exit */
 	on_shmem_exit(WalRcvDie, 0);
@@ -1325,7 +1325,6 @@ pg_stat_get_wal_receiver(PG_FUNCTION_ARGS)
 	state = WalRcv->walRcvState;
 	receive_start_lsn = WalRcv->receiveStart;
 	receive_start_tli = WalRcv->receiveStartTLI;
-	written_lsn = pg_atomic_read_u64(&WalRcv->writtenUpto);
 	flushed_lsn = WalRcv->flushedUpto;
 	received_tli = WalRcv->receivedTLI;
 	last_send_time = WalRcv->lastMsgSendTime;
@@ -1345,6 +1344,14 @@ pg_stat_get_wal_receiver(PG_FUNCTION_ARGS)
 	if (pid == 0 || !ready_to_display)
 		PG_RETURN_NULL();
 
+	/*
+	 * Read "writtenUpto" without holding a spinlock.  Note that it may not be
+	 * consistent with the other shared variables of the WAL receiver
+	 * protected by a spinlock, but this should not be used for data integrity
+	 * checks.
+	 */
+	written_lsn = pg_atomic_read_u64(&WalRcv->writtenUpto);
+
 	/* determine result type */
 	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
 		elog(ERROR, "return type must be a row type");

src/backend/replication/walreceiverfuncs.c

@@ -63,6 +63,7 @@ WalRcvShmemInit(void)
 		MemSet(WalRcv, 0, WalRcvShmemSize());
 		WalRcv->walRcvState = WALRCV_STOPPED;
 		SpinLockInit(&WalRcv->mutex);
+		pg_atomic_init_u64(&WalRcv->writtenUpto, 0);
 		WalRcv->latch = NULL;
 	}
 }

src/test/regress/expected/sysviews.out

@@ -83,6 +83,13 @@ select count(*) = 1 as ok from pg_stat_wal;
  t
 (1 row)
 
+-- We expect no walreceiver running in this test
+select count(*) = 0 as ok from pg_stat_wal_receiver;
+ ok 
+----
+ t
+(1 row)
+
 -- This is to record the prevailing planner enable_foo settings during
 -- a regression test run.
 select name, setting from pg_settings where name like 'enable%';

src/test/regress/sql/sysviews.sql

@@ -40,6 +40,9 @@ select count(*) >= 0 as ok from pg_prepared_xacts;
 -- There must be only one record
 select count(*) = 1 as ok from pg_stat_wal;
 
+-- We expect no walreceiver running in this test
+select count(*) = 0 as ok from pg_stat_wal_receiver;
+
 -- This is to record the prevailing planner enable_foo settings during
 -- a regression test run.
 select name, setting from pg_settings where name like 'enable%';