Revamp the WAL record format.

Each WAL record now carries information about the modified relation and
block(s) in a standardized format. That makes it easier to write tools that
need that information, like pg_rewind, prefetching the blocks to speed up
recovery, etc.

There's a whole new API for building WAL records, replacing the XLogRecData
chains used previously. The new API consists of XLogRegister* functions,
which are called for each buffer and chunk of data that is added to the
record. The new API also gives more control over when a full-page image is
written, by passing flags to the XLogRegisterBuffer function.

This also simplifies the XLogReadBufferForRedo() calls. The function can dig
the relation and block number from the WAL record, so they no longer need to
be passed as arguments.

For the convenience of redo routines, XLogReader now disects each WAL record
after reading it, copying the main data part and the per-block data into
MAXALIGNed buffers. The data chunks are not aligned within the WAL record,
but the redo routines can assume that the pointers returned by XLogRecGet*
functions are. Redo routines are now passed the XLogReaderState, which
contains the record in the already-disected format, instead of the plain
XLogRecord.

The new record format also makes the fixed size XLogRecord header smaller,
by removing the xl_len field. The length of the "main data" portion is now
stored at the end of the WAL record, and there's a separate header after
XLogRecord for it. The alignment padding at the end of XLogRecord is also
removed. This compansates for the fact that the new format would otherwise
be more bulky than the old format.

Reviewed by Andres Freund, Amit Kapila, Michael Paquier, Alvaro Herrera,
Fujii Masao.

1 files changed, 20 insertions, 44 deletions

diff --git a/src/backend/access/gin/ginentrypage.c b/src/backend/access/gin/ginentrypage.c
index 84dc1e228c1..2dae7b95499 100644
--- a/src/backend/access/gin/ginentrypage.c
+++ b/src/backend/access/gin/ginentrypage.c
@@ -22,7 +22,7 @@
 static void entrySplitPage(GinBtree btree, Buffer origbuf,
 			   GinBtreeStack *stack,
 			   void *insertPayload,
-			   BlockNumber updateblkno, XLogRecData **prdata,
+			   BlockNumber updateblkno,
 			   Page *newlpage, Page *newrpage);
 
 /*
@@ -515,33 +515,33 @@ entryPreparePage(GinBtree btree, Page page, OffsetNumber off,
  * On insertion to an internal node, in addition to inserting the given item,
  * the downlink of the existing item at 'off' is updated to point to
  * 'updateblkno'.
+ *
+ * On INSERTED, registers the buffer as buffer ID 0, with data.
+ * On SPLIT, returns rdata that represents the split pages in *prdata.
  */
 static GinPlaceToPageRC
 entryPlaceToPage(GinBtree btree, Buffer buf, GinBtreeStack *stack,
 				 void *insertPayload, BlockNumber updateblkno,
-				 XLogRecData **prdata, Page *newlpage, Page *newrpage)
+				 Page *newlpage, Page *newrpage)
 {
 	GinBtreeEntryInsertData *insertData = insertPayload;
 	Page		page = BufferGetPage(buf);
 	OffsetNumber off = stack->off;
 	OffsetNumber placed;
-	int			cnt = 0;
 
-	/* these must be static so they can be returned to caller */
-	static XLogRecData rdata[3];
+	/* this must be static so it can be returned to caller. */
 	static ginxlogInsertEntry data;
 
 	/* quick exit if it doesn't fit */
 	if (!entryIsEnoughSpace(btree, buf, off, insertData))
 	{
 		entrySplitPage(btree, buf, stack, insertPayload, updateblkno,
-					   prdata, newlpage, newrpage);
+					   newlpage, newrpage);
 		return SPLIT;
 	}
 
 	START_CRIT_SECTION();
 
-	*prdata = rdata;
 	entryPreparePage(btree, page, off, insertData, updateblkno);
 
 	placed = PageAddItem(page,
@@ -552,21 +552,17 @@ entryPlaceToPage(GinBtree btree, Buffer buf, GinBtreeStack *stack,
 		elog(ERROR, "failed to add item to index page in \"%s\"",
 			 RelationGetRelationName(btree->index));
 
-	data.isDelete = insertData->isDelete;
-	data.offset = off;
-
-	rdata[cnt].buffer = buf;
-	rdata[cnt].buffer_std = true;
-	rdata[cnt].data = (char *) &data;
-	rdata[cnt].len = offsetof(ginxlogInsertEntry, tuple);
-	rdata[cnt].next = &rdata[cnt + 1];
-	cnt++;
-
-	rdata[cnt].buffer = buf;
-	rdata[cnt].buffer_std = true;
-	rdata[cnt].data = (char *) insertData->entry;
-	rdata[cnt].len = IndexTupleSize(insertData->entry);
-	rdata[cnt].next = NULL;
+	if (RelationNeedsWAL(btree->index))
+	{
+		data.isDelete = insertData->isDelete;
+		data.offset = off;
+
+		XLogRegisterBuffer(0, buf, REGBUF_STANDARD);
+		XLogRegisterBufData(0, (char *) &data,
+							offsetof(ginxlogInsertEntry, tuple));
+		XLogRegisterBufData(0, (char *) insertData->entry,
+							IndexTupleSize(insertData->entry));
+	}
 
 	return INSERTED;
 }
@@ -581,7 +577,7 @@ static void
 entrySplitPage(GinBtree btree, Buffer origbuf,
 			   GinBtreeStack *stack,
 			   void *insertPayload,
-			   BlockNumber updateblkno, XLogRecData **prdata,
+			   BlockNumber updateblkno,
 			   Page *newlpage, Page *newrpage)
 {
 	GinBtreeEntryInsertData *insertData = insertPayload;
@@ -590,7 +586,6 @@ entrySplitPage(GinBtree btree, Buffer origbuf,
 				maxoff,
 				separator = InvalidOffsetNumber;
 	Size		totalsize = 0;
-	Size		tupstoresize;
 	Size		lsize = 0,
 				size;
 	char	   *ptr;
@@ -599,13 +594,8 @@ entrySplitPage(GinBtree btree, Buffer origbuf,
 	Page		lpage = PageGetTempPageCopy(BufferGetPage(origbuf));
 	Page		rpage = PageGetTempPageCopy(BufferGetPage(origbuf));
 	Size		pageSize = PageGetPageSize(lpage);
+	char		tupstore[2 * BLCKSZ];
 
-	/* these must be static so they can be returned to caller */
-	static XLogRecData rdata[2];
-	static ginxlogSplitEntry data;
-	static char tupstore[2 * BLCKSZ];
-
-	*prdata = rdata;
 	entryPreparePage(btree, lpage, off, insertData, updateblkno);
 
 	/*
@@ -638,7 +628,6 @@ entrySplitPage(GinBtree btree, Buffer origbuf,
 		ptr += size;
 		totalsize += size + sizeof(ItemIdData);
 	}
-	tupstoresize = ptr - tupstore;
 
 	/*
 	 * Initialize the left and right pages, and copy all the tuples back to
@@ -673,19 +662,6 @@ entrySplitPage(GinBtree btree, Buffer origbuf,
 		ptr += MAXALIGN(IndexTupleSize(itup));
 	}
 
-	data.separator = separator;
-	data.nitem = maxoff;
-
-	rdata[0].buffer = InvalidBuffer;
-	rdata[0].data = (char *) &data;
-	rdata[0].len = sizeof(ginxlogSplitEntry);
-	rdata[0].next = &rdata[1];
-
-	rdata[1].buffer = InvalidBuffer;
-	rdata[1].data = tupstore;
-	rdata[1].len = tupstoresize;
-	rdata[1].next = NULL;
-
 	*newlpage = lpage;
 	*newrpage = rpage;
 }