Feedback-ID: ie3de48e3:Fastmail
Date: Fri, 7 Nov 2025 14:37:32 +0100
From: =?utf-8?Q?=C3=81lvaro?= Herrera <alvherre@kurilemu.de>
To: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Cc: Peter Eisentraut <peter@eisentraut.org>,
	Michael Paquier <michael@paquier.xyz>, Heikki Linnakangas <hlinnaka@iki.fi>,
	Quan Zongliang <quanzongliang@yeah.net>, pgsql-hackers@lists.postgresql.org
Subject: Re: Consistently use the XLogRecPtrIsInvalid() macro
Message-ID: <202511071335.iw4fygwd4rxo@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <aQ2O99S+VjEZSf3/@ip-10-97-1-34.eu-west-3.compute.internal>
Archived-At: 
 <https://www.postgresql.org/message-id/202511071335.iw4fygwd4rxo%40alvherre.pgsql>
Precedence: bulk

On 2025-Nov-07, Bertrand Drouvot wrote:

> Agree, will modify the .cocci scripts that way.

I just noticed that we missed this ... maybe you want to include it also?

diff --git a/src/backend/replication/syncrep.c b/src/backend/replication/syncrep.c
index a0c79958fd5..1f11c8646f5 100644
--- a/src/backend/replication/syncrep.c
+++ b/src/backend/replication/syncrep.c
@@ -355,7 +355,7 @@ SyncRepWaitForLSN(XLogRecPtr lsn, bool commit)
 	pg_read_barrier();
 	Assert(dlist_node_is_detached(&MyProc->syncRepLinks));
 	MyProc->syncRepState = SYNC_REP_NOT_WAITING;
-	MyProc->waitLSN = 0;
+	MyProc->waitLSN = InvalidXLogRecPtr;
 
 	/* reset ps display to remove the suffix */
 	if (update_process_title)
@@ -1028,7 +1028,7 @@ SyncRepQueueIsOrderedByLSN(int mode)
 
 	Assert(mode >= 0 && mode < NUM_SYNC_REP_WAIT_MODE);
 
-	lastLSN = 0;
+	lastLSN = InvalidXLogRecPtr;
 
 	dlist_foreach(iter, &WalSndCtl->SyncRepQueue[mode])
 	{
diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c
index 1504fafe6d8..ce0d6a7539c 100644
--- a/src/backend/storage/lmgr/proc.c
+++ b/src/backend/storage/lmgr/proc.c
@@ -509,7 +509,7 @@ InitProcess(void)
 	MyProc->recoveryConflictPending = false;
 
 	/* Initialize fields for sync rep */
-	MyProc->waitLSN = 0;
+	MyProc->waitLSN = InvalidXLogRecPtr;
 	MyProc->syncRepState = SYNC_REP_NOT_WAITING;
 	dlist_node_init(&MyProc->syncRepLinks);
 

> Now that XLogRecPtrIsValid() is available in back branches, I agree that we
> can be less conservative and not wait until v24. v21 looks like good timing to
> me.

Cool,  please resubmit.

-- 
Álvaro Herrera        Breisgau, Deutschland  —  https://www.EnterpriseDB.com/
"They proved that being American is not just for some people"
                                               (George Takei)