v8-0001-Use-barrier-semantics-when-reading-writing-writte.patch

From 73ec91e1ad442ec052c8f1f964c10c810def6047 Mon Sep 17 00:00:00 2001
From: alterego655 <824662526@qq.com>
Date: Fri, 10 Apr 2026 10:45:52 +0800
Subject: [PATCH v8 1/7] Use barrier semantics when reading/writing writtenUpto

The walreceiver publishes its write position lock-free via writtenUpto.
On weakly-ordered architectures (ARM, PowerPC), both sides of this
handshake need explicit barriers so that the lock-less reader sees a
consistent state.

Use pg_atomic_write_membarrier_u64() at both write sites and
pg_atomic_read_membarrier_u64() in GetWalRcvWriteRecPtr().  This matches
the barrier semantics that GetWalRcvFlushRecPtr() and other LSN-position
functions get implicitly from their spinlock acquire/release, and
protects from bugs caused by expectations of similar barrier guarantees
from different LSN-position functions.

Reported-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/zqbppucpmkeqecfy4s5kscnru4tbk6khp3ozqz6ad2zijz354k%40w4bdf4z3wqoz
---
 src/backend/replication/walreceiver.c      |  2 +-
 src/backend/replication/walreceiverfuncs.c | 10 +++++++---
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/src/backend/replication/walreceiver.c b/src/backend/replication/walreceiver.c
index 8185412a810..c0ac75b54d7 100644
--- a/src/backend/replication/walreceiver.c
+++ b/src/backend/replication/walreceiver.c
@@ -975,7 +975,7 @@ XLogWalRcvWrite(char *buf, Size nbytes, XLogRecPtr recptr, TimeLineID tli)
 	}
 
 	/* Update shared-memory status */
-	pg_atomic_write_u64(&WalRcv->writtenUpto, LogstreamResult.Write);
+	pg_atomic_write_membarrier_u64(&WalRcv->writtenUpto, LogstreamResult.Write);
 
 	/*
 	 * If we wrote an LSN that someone was waiting for, notify the waiters.
diff --git a/src/backend/replication/walreceiverfuncs.c b/src/backend/replication/walreceiverfuncs.c
index bd5d47be964..9447913c0e7 100644
--- a/src/backend/replication/walreceiverfuncs.c
+++ b/src/backend/replication/walreceiverfuncs.c
@@ -321,7 +321,8 @@ RequestXLogStreaming(TimeLineID tli, XLogRecPtr recptr, const char *conninfo,
 		walrcv->flushedUpto = recptr;
 		walrcv->receivedTLI = tli;
 		walrcv->latestChunkStart = recptr;
-		pg_atomic_write_u64(&walrcv->writtenUpto, recptr);
+		/* Pairs with pg_atomic_read_membarrier_u64() in GetWalRcvWriteRecPtr(). */
+		pg_atomic_write_membarrier_u64(&walrcv->writtenUpto, recptr);
 	}
 	walrcv->receiveStart = recptr;
 	walrcv->receiveStartTLI = tli;
@@ -363,14 +364,17 @@ GetWalRcvFlushRecPtr(XLogRecPtr *latestChunkStart, TimeLineID *receiveTLI)
 
 /*
  * Returns the last+1 byte position that walreceiver has written.
- * This returns a recently written value without taking a lock.
+ *
+ * Use pg_atomic_read_membarrier_u64() to ensure that callers see up-to-date
+ * shared memory state, matching the barrier semantics provided by the
+ * spinlock in GetWalRcvFlushRecPtr() and other LSN-position functions.
  */
 XLogRecPtr
 GetWalRcvWriteRecPtr(void)
 {
 	WalRcvData *walrcv = WalRcv;
 
-	return pg_atomic_read_u64(&walrcv->writtenUpto);
+	return pg_atomic_read_membarrier_u64(&walrcv->writtenUpto);
 }
 
 /*
-- 
2.51.0