Fix full-page writes of internal GIN pages.

Insertion to a non-leaf GIN page didn't make a full-page image of the page,
which is wrong. The code used to do it correctly, but was changed (commit
853d1c3103fa961ae6219f0281885b345593d101) because the redo-routine didn't
track incomplete splits correctly when the page was restored from a full
page image. Of course, that was not right way to fix it, the redo routine
should've been fixed instead. The redo-routine was surreptitiously fixed
in 2010 (commit 4016bdef8aded77b4903c457050622a5a1815c16), so all we need
to do now is revert the code that creates the record to its original form.

This doesn't change the format of the WAL record.

Backpatch to all supported versions.

diff --git a/src/backend/access/gin/gindatapage.c b/src/backend/access/gin/gindatapage.c
index 13ab44869232d1ea73726dcc1ee21f499dc27c61..27faef5124c7dd333af1ebe516281f3ae6c8683b 100644 (file)
--- a/src/backend/access/gin/gindatapage.c
+++ b/src/backend/access/gin/gindatapage.c
@@ -381,7 +381,6 @@ dataPlaceToPage(GinBtree btree, Buffer buf, OffsetNumber off, XLogRecData **prda
 {
        Page            page = BufferGetPage(buf);
        int                     sizeofitem = GinSizeOfDataPageItem(page);
-       int                     cnt = 0;
 
        /* these must be static so they can be returned to caller */
        static XLogRecData rdata[3];
@@ -401,32 +400,25 @@ dataPlaceToPage(GinBtree btree, Buffer buf, OffsetNumber off, XLogRecData **prda
        data.isLeaf = GinPageIsLeaf(page) ? TRUE : FALSE;
 
        /*
-        * Prevent full page write if child's split occurs. That is needed to
-        * remove incomplete splits while replaying WAL
-        *
-        * data.updateBlkno contains new block number (of newly created right
-        * page) for recently splited page.
+        * For incomplete-split tracking, we need updateBlkno information and the
+        * inserted item even when we make a full page image of the page, so put
+        * the buffer reference in a separate XLogRecData entry.
         */
-       if (data.updateBlkno == InvalidBlockNumber)
-       {
-               rdata[0].buffer = buf;
-               rdata[0].buffer_std = FALSE;
-               rdata[0].data = NULL;
-               rdata[0].len = 0;
-               rdata[0].next = &rdata[1];
-               cnt++;
-       }
+       rdata[0].buffer = buf;
+       rdata[0].buffer_std = FALSE;
+       rdata[0].data = NULL;
+       rdata[0].len = 0;
+       rdata[0].next = &rdata[1];
 
-       rdata[cnt].buffer = InvalidBuffer;
-       rdata[cnt].data = (char *) &data;
-       rdata[cnt].len = sizeof(ginxlogInsert);
-       rdata[cnt].next = &rdata[cnt + 1];
-       cnt++;
+       rdata[1].buffer = InvalidBuffer;
+       rdata[1].data = (char *) &data;
+       rdata[1].len = sizeof(ginxlogInsert);
+       rdata[1].next = &rdata[2];
 
-       rdata[cnt].buffer = InvalidBuffer;
-       rdata[cnt].data = (GinPageIsLeaf(page)) ? ((char *) (btree->items + btree->curitem)) : ((char *) &(btree->pitem));
-       rdata[cnt].len = sizeofitem;
-       rdata[cnt].next = NULL;
+       rdata[2].buffer = InvalidBuffer;
+       rdata[2].data = (GinPageIsLeaf(page)) ? ((char *) (btree->items + btree->curitem)) : ((char *) &(btree->pitem));
+       rdata[2].len = sizeofitem;
+       rdata[2].next = NULL;
 
        if (GinPageIsLeaf(page))
        {
@@ -442,7 +434,7 @@ dataPlaceToPage(GinBtree btree, Buffer buf, OffsetNumber off, XLogRecData **prda
                                btree->curitem++;
                        }
                        data.nitem = btree->curitem - savedPos;
-                       rdata[cnt].len = sizeofitem * data.nitem;
+                       rdata[2].len = sizeofitem * data.nitem;
                }
                else
                {

diff --git a/src/backend/access/gin/ginentrypage.c b/src/backend/access/gin/ginentrypage.c
index 8ead38f234ab36e4920650775d2591e3df0ab4be..8cb5148c810b7f660cf44695842e7a49a36c11a0 100644 (file)
--- a/src/backend/access/gin/ginentrypage.c
+++ b/src/backend/access/gin/ginentrypage.c
@@ -486,7 +486,6 @@ entryPlaceToPage(GinBtree btree, Buffer buf, OffsetNumber off, XLogRecData **prd
 {
        Page            page = BufferGetPage(buf);
        OffsetNumber placed;
-       int                     cnt = 0;
 
        /* these must be static so they can be returned to caller */
        static XLogRecData rdata[3];
@@ -509,32 +508,25 @@ entryPlaceToPage(GinBtree btree, Buffer buf, OffsetNumber off, XLogRecData **prd
        data.isLeaf = GinPageIsLeaf(page) ? TRUE : FALSE;
 
        /*
-        * Prevent full page write if child's split occurs. That is needed to
-        * remove incomplete splits while replaying WAL
-        *
-        * data.updateBlkno contains new block number (of newly created right
-        * page) for recently splited page.
+        * For incomplete-split tracking, we need updateBlkno information and the
+        * inserted item even when we make a full page image of the page, so put
+        * the buffer reference in a separate XLogRecData entry.
         */
-       if (data.updateBlkno == InvalidBlockNumber)
-       {
-               rdata[0].buffer = buf;
-               rdata[0].buffer_std = TRUE;
-               rdata[0].data = NULL;
-               rdata[0].len = 0;
-               rdata[0].next = &rdata[1];
-               cnt++;
-       }
-
-       rdata[cnt].buffer = InvalidBuffer;
-       rdata[cnt].data = (char *) &data;
-       rdata[cnt].len = sizeof(ginxlogInsert);
-       rdata[cnt].next = &rdata[cnt + 1];
-       cnt++;
+       rdata[0].buffer = buf;
+       rdata[0].buffer_std = TRUE;
+       rdata[0].data = NULL;
+       rdata[0].len = 0;
+       rdata[0].next = &rdata[1];
 
-       rdata[cnt].buffer = InvalidBuffer;
-       rdata[cnt].data = (char *) btree->entry;
-       rdata[cnt].len = IndexTupleSize(btree->entry);
-       rdata[cnt].next = NULL;
+       rdata[1].buffer = InvalidBuffer;
+       rdata[1].data = (char *) &data;
+       rdata[1].len = sizeof(ginxlogInsert);
+       rdata[1].next = &rdata[2];
+
+       rdata[2].buffer = InvalidBuffer;
+       rdata[2].data = (char *) btree->entry;
+       rdata[2].len = IndexTupleSize(btree->entry);
+       rdata[2].next = NULL;
 
        btree->entry = NULL;
 }