Fix WAL format incompatibility introduced by backpatching of 52ac6cd2d0

52ac6cd2d0 added new field to ginxlogDeletePage and was backpatched to 9.4.
That led to problems when patched postgres instance applies WAL records
generated by non-patched one.  WAL records generated by non-patched instance
don't contain new field, which patched one is expecting to see.

Thankfully, we can distinguish patched and non-patched WAL records by their data
size.  If we see that WAL record is generated by non-patched instance, we skip
processing of new field.  This commit comes with some assertions.  In
particular, if it appears that on some platform struct data size didn't change
then static assertion will trigger.

Reported-by: Simon Riggs
Discussion: https://postgr.es/m/CANP8%2Bj%2BK4whxf7ET7%2BgO%2BG-baC3-WxqqH%3DnV4X2CgfEPA3Yu3g%40mail.gmail.com
Author: Alexander Korotkov
Reviewed-by: Simon Riggs, Alvaro Herrera
Backpatch-through: 9.4

1 files changed, 18 insertions, 1 deletions

diff --git a/src/backend/access/gin/ginxlog.c b/src/backend/access/gin/ginxlog.c
index 291f22ef77f..b86990f3ac7 100644
--- a/src/backend/access/gin/ginxlog.c
+++ b/src/backend/access/gin/ginxlog.c
@@ -531,7 +531,24 @@ ginRedoDeletePage(XLogReaderState *record)
 		page = BufferGetPage(dbuffer);
 		Assert(GinPageIsData(page));
 		GinPageGetOpaque(page)->flags = GIN_DELETED;
-		GinPageSetDeleteXid(page, data->deleteXid);
+
+		/*
+		 * deleteXid field of ginxlogDeletePage was added during backpatching.
+		 * But, non-backpatched instances will continue generate WAL without
+		 * this field.  We should be able to correctly apply that.  We can
+		 * distinguish new WAL records by size their data, because
+		 * ginxlogDeletePage changes its size on both 32-bit and 64-bit
+		 * platforms.
+		 */
+		StaticAssertStmt(sizeof(ginxlogDeletePage) !=
+						 sizeof(ginxlogDeletePageOld),
+						 "ginxlogDeletePage size should be changed "
+						 "with addition of deleteXid field");
+		Assert(XLogRecGetDataLen(record) == sizeof(ginxlogDeletePage) ||
+			   XLogRecGetDataLen(record) == sizeof(ginxlogDeletePageOld));
+		if (XLogRecGetDataLen(record) == sizeof(ginxlogDeletePage))
+			GinPageSetDeleteXid(page, data->deleteXid);
+
 		PageSetLSN(page, lsn);
 		MarkBufferDirty(dbuffer);
 	}