Fix buffer pin leak in heap update redo routine.

In a heap update, if the old and new tuple were on different pages, and the
new page no longer existed (because it was subsequently truncated away by
vacuum), heap_xlog_update forgot to release the pin on the old buffer. This
bug was introduced by the "Fix multiple problems in WAL replay" patch,
commit 3bbf668de9 (on master branch).

With full_page_writes=off, this triggered an "incorrect local pin count"
error later in replay, if the old page was vacuumed.

This fixes bug #7969, reported by Yunong Xiao. Backpatch to 9.0, like the
commit that introduced this bug.

src/backend/access/heap/heapam.c

@@ -5367,7 +5367,11 @@ newt:;
 								 ItemPointerGetBlockNumber(&(xlrec->newtid)),
 								 false);
 		if (!BufferIsValid(nbuffer))
+		{
+			if (BufferIsValid(obuffer))
+				UnlockReleaseBuffer(obuffer);
 			return;
+		}
 		page = (Page) BufferGetPage(nbuffer);
 
 		if (XLByteLE(lsn, PageGetLSN(page)))	/* changes are applied */