Wake up for latches in CheckpointWriteDelay().

The checkpointer shouldn't ignore its latch.  Other backends may be
waiting for it to drain the request queue.  Hopefully real systems don't
have a full queue often, but the condition is reached easily when
shared_buffers is small.

This involves defining a new wait event, which will appear in the
pg_stat_activity view often due to spread checkpoints.

Back-patch only to 14.  Even though the problem exists in earlier
branches too, it's hard to hit there.  In 14 we stopped using signal
handlers for latches on Linux, *BSD and macOS, which were previously
hiding this problem by interrupting the sleep (though not reliably, as
the signal could arrive before the sleep begins; precisely the problem
latches address).

Reported-by: Andres Freund <[email protected]>
Reviewed-by: Andres Freund <[email protected]>
Discussion: https://postgr.es/m/20220226213942.nb7uvb2pamyu26dj%40alap3.anarazel.de

4 files changed, 15 insertions, 1 deletions

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 9fb62fec8e..8620aaddc7 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -2236,6 +2236,10 @@ postgres   27093  0.0  0.0  30096  2752 ?        Ss   11:34   0:00 postgres: ser
       <entry>Waiting during base backup when throttling activity.</entry>
      </row>
      <row>
+      <entry><literal>CheckpointerWriteDelay</literal></entry>
+      <entry>Waiting between writes while performing a checkpoint.</entry>
+     </row>
+     <row>
       <entry><literal>PgSleep</literal></entry>
       <entry>Waiting due to a call to <function>pg_sleep</function> or
        a sibling function.</entry>
diff --git a/src/backend/postmaster/checkpointer.c b/src/backend/postmaster/checkpointer.c
index 4488e3a443..a59c3cf020 100644
--- a/src/backend/postmaster/checkpointer.c
+++ b/src/backend/postmaster/checkpointer.c
@@ -484,6 +484,9 @@ CheckpointerMain(void)
 			}
 
 			ckpt_active = false;
+
+			/* We may have received an interrupt during the checkpoint. */
+			HandleCheckpointerInterrupts();
 		}
 
 		/* Check for archive_timeout and switch xlog files if necessary. */
@@ -726,7 +729,10 @@ CheckpointWriteDelay(int flags, double progress)
 		 * Checkpointer and bgwriter are no longer related so take the Big
 		 * Sleep.
 		 */
-		pg_usleep(100000L);
+		WaitLatch(MyLatch, WL_LATCH_SET | WL_EXIT_ON_PM_DEATH | WL_TIMEOUT,
+				  100,
+				  WAIT_EVENT_CHECKPOINT_WRITE_DELAY);
+		ResetLatch(MyLatch);
 	}
 	else if (--absorb_counter <= 0)
 	{
diff --git a/src/backend/utils/activity/wait_event.c b/src/backend/utils/activity/wait_event.c
index 60972c3a75..0706e922b5 100644
--- a/src/backend/utils/activity/wait_event.c
+++ b/src/backend/utils/activity/wait_event.c
@@ -485,6 +485,9 @@ pgstat_get_wait_timeout(WaitEventTimeout w)
 		case WAIT_EVENT_BASE_BACKUP_THROTTLE:
 			event_name = "BaseBackupThrottle";
 			break;
+		case WAIT_EVENT_CHECKPOINT_WRITE_DELAY:
+			event_name = "CheckpointWriteDelay";
+			break;
 		case WAIT_EVENT_PG_SLEEP:
 			event_name = "PgSleep";
 			break;
diff --git a/src/include/utils/wait_event.h b/src/include/utils/wait_event.h
index 395d325c5f..d0345c6b49 100644
--- a/src/include/utils/wait_event.h
+++ b/src/include/utils/wait_event.h
@@ -141,6 +141,7 @@ typedef enum
 typedef enum
 {
 	WAIT_EVENT_BASE_BACKUP_THROTTLE = PG_WAIT_TIMEOUT,
+	WAIT_EVENT_CHECKPOINT_WRITE_DELAY,
 	WAIT_EVENT_PG_SLEEP,
 	WAIT_EVENT_RECOVERY_APPLY_DELAY,
 	WAIT_EVENT_RECOVERY_RETRIEVE_RETRY_INTERVAL,