Support unlogged tables.

The contents of an unlogged table are WAL-logged; thus, they are not
available on standby servers and are truncated whenever the database
system enters recovery.  Indexes on unlogged tables are also unlogged.
Unlogged GiST indexes are not currently supported.

Author: robertmhaas

doc/src/sgml/catalogs.sgml

@@ -1644,7 +1644,8 @@
       <entry><type>bool</type></entry>
       <entry></entry>
       <entry>
-       <literal>p</> = permanent table, <literal>t</> = temporary table
+       <literal>p</> = permanent table, <literal>u</> = unlogged table,
+       <literal>t</> = temporary table
       </entry>
      </row>
 

doc/src/sgml/indexam.sgml

@@ -167,6 +167,17 @@ ambuild (Relation heapRelation,
 
   <para>
 <programlisting>
+void
+ambuildempty (Relation indexRelation);
+</programlisting>
+   Build an empty index, and write it to the initialization fork (INIT_FORKNUM)
+   of the given relation.  This method is called only for unlogged tables; the
+   empty index written to the initialization fork will be copied over the main
+   relation fork on each server restart.
+  </para>
+
+  <para>
+<programlisting>
 bool
 aminsert (Relation indexRelation,
           Datum *values,

doc/src/sgml/ref/create_table.sgml

@@ -21,7 +21,7 @@ PostgreSQL documentation
 
  <refsynopsisdiv>
 <synopsis>
-CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } ] TABLE [ IF NOT EXISTS ] <replaceable class="PARAMETER">table_name</replaceable> ( [
+CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } | UNLOGGED ] TABLE [ IF NOT EXISTS ] <replaceable class="PARAMETER">table_name</replaceable> ( [
   { <replaceable class="PARAMETER">column_name</replaceable> <replaceable class="PARAMETER">data_type</replaceable> [ <replaceable class="PARAMETER">column_constraint</replaceable> [ ... ] ]
     | <replaceable>table_constraint</replaceable>
     | LIKE <replaceable>parent_table</replaceable> [ <replaceable>like_option</replaceable> ... ] }
@@ -32,7 +32,7 @@ CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } ] TABLE [ IF NOT EXISTS ] <repl
 [ ON COMMIT { PRESERVE ROWS | DELETE ROWS | DROP } ]
 [ TABLESPACE <replaceable class="PARAMETER">tablespace</replaceable> ]
 
-CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } ] TABLE [ IF NOT EXISTS ] <replaceable class="PARAMETER">table_name</replaceable>
+CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } | UNLOGGED ] TABLE [ IF NOT EXISTS ] <replaceable class="PARAMETER">table_name</replaceable>
     OF <replaceable class="PARAMETER">type_name</replaceable> [ (
   { <replaceable class="PARAMETER">column_name</replaceable> WITH OPTIONS [ <replaceable class="PARAMETER">column_constraint</replaceable> [ ... ] ]
     | <replaceable>table_constraint</replaceable> }
@@ -164,6 +164,23 @@ CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } ] TABLE [ IF NOT EXISTS ] <repl
     </listitem>
    </varlistentry>
 
+   <varlistentry>
+    <term><literal>UNLOGGED</></term>
+    <listitem>
+     <para>
+      If specified, the table is created as an unlogged table.  Data written
+      to unlogged tables is not written to the write-ahead log (see <xref
+      linkend="wal">), which makes them considerably faster than ordinary
+      tables.  However, they are not crash-safe: an unlogged table is
+      automatically truncated after a crash or unclean shutdown.  The contents
+      of an unlogged table are also not replicated to standby servers.
+      Any indexes created on an unlogged table are automatically unlogged as
+      well; however, unlogged <link linkend="GiST">GiST indexes</link> are
+      currently not supported and cannot be created on an unlogged table.
+     </para>
+    </listitem>
+   </varlistentry>
+
    <varlistentry>
     <term><literal>IF NOT EXISTS</></term>
     <listitem>

doc/src/sgml/ref/create_table_as.sgml

@@ -21,7 +21,7 @@ PostgreSQL documentation
 
  <refsynopsisdiv>
 <synopsis>
-CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } ] TABLE <replaceable>table_name</replaceable>
+CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } | UNLOGGED ] TABLE <replaceable>table_name</replaceable>
     [ (<replaceable>column_name</replaceable> [, ...] ) ]
     [ WITH ( <replaceable class="PARAMETER">storage_parameter</replaceable> [= <replaceable class="PARAMETER">value</replaceable>] [, ... ] ) | WITH OIDS | WITHOUT OIDS ]
     [ ON COMMIT { PRESERVE ROWS | DELETE ROWS | DROP } ]
@@ -81,6 +81,16 @@ CREATE [ [ GLOBAL | LOCAL ] { TEMPORARY | TEMP } ] TABLE <replaceable>table_name
     </listitem>
    </varlistentry>
 
+   <varlistentry>
+    <term><literal>UNLOGGED</></term>
+    <listitem>
+     <para>
+      If specified, the table is created as an unlogged table.
+      Refer to <xref linkend="sql-createtable"> for details.
+     </para>
+    </listitem>
+   </varlistentry>
+
    <varlistentry>
     <term><replaceable>table_name</replaceable></term>
     <listitem>

doc/src/sgml/ref/pg_dump.sgml

@@ -669,6 +669,17 @@ PostgreSQL documentation
       </listitem>
      </varlistentry>
 
+     <varlistentry>
+      <term><option>--no-unlogged-table-data</option></term>
+      <listitem>
+       <para>
+        Do not dump the contents of unlogged tables.  This option has no
+        effect on whether or not the table definitions (schema) are dumped;
+        it only suppresses dumping the table data.
+       </para>
+      </listitem>
+     </varlistentry>
+
      <varlistentry>
       <term><option>--quote-all-identifiers</></term>
       <listitem>

doc/src/sgml/ref/pg_dumpall.sgml

@@ -201,6 +201,17 @@ PostgreSQL documentation
       </listitem>
      </varlistentry>
 
+     <varlistentry>
+      <term><option>--no-unlogged-table-data</option></term>
+      <listitem>
+       <para>
+        Do not dump the contents of unlogged tables.  This option has no
+        effect on whether or not the table definitions (schema) are dumped;
+        it only suppresses dumping the table data.
+       </para>
+      </listitem>
+     </varlistentry>
+
      <varlistentry>
       <term><option>-r</option></term>
       <term><option>--roles-only</option></term>

doc/src/sgml/storage.sgml

@@ -147,7 +147,9 @@ the relation.  The free space map is stored in a file named with the filenode
 number plus the suffix <literal>_fsm</>.  Tables also have a
 <firstterm>visibility map</>, stored in a fork with the suffix <literal>_vm</>,
 to track which pages are known to have no dead tuples.  The visibility map is
-described further in <xref linkend="storage-vm">.
+described further in <xref linkend="storage-vm">.  Unlogged tables and indexes
+have a third fork, known as the initialization fork, which is stored in a fork
+with the suffix <literal>_init</literal> (see <xref linkend="storage-init">).
 </para>
 
 <caution>
@@ -485,6 +487,24 @@ a bit is not set, it might or might not be true.
 
 </sect1>
 
+<sect1 id="storage-init">
+
+<title>The Initialization Fork</title>
+
+<indexterm>
+ <primary>Initialization Fork</primary>
+</indexterm>
+
+<para>
+Each unlogged table, and each index on an unlogged table, has an initialization
+fork.  The initialization fork is an empty table or index of the appropriate
+type.  When an unlogged table must be reset to empty due to a crash, the
+initialization fork is copied over the main fork, and any other forks are
+erased (they will be recreated automatically as needed).
+</para>
+
+</sect1>
+
 <sect1 id="storage-page-layout">
 
 <title>Database Page Layout</title>

src/backend/access/gin/gininsert.c

@@ -19,6 +19,7 @@
 #include "catalog/index.h"
 #include "miscadmin.h"
 #include "storage/bufmgr.h"
+#include "storage/smgr.h"
 #include "storage/indexfsm.h"
 #include "utils/memutils.h"
 
@@ -411,6 +412,47 @@ ginbuild(PG_FUNCTION_ARGS)
 	PG_RETURN_POINTER(result);
 }
 
+/*
+ *	ginbuildempty() -- build an empty gin index in the initialization fork
+ */
+Datum
+ginbuildempty(PG_FUNCTION_ARGS)
+{
+	Relation	index = (Relation) PG_GETARG_POINTER(0);
+	Buffer		RootBuffer,
+				MetaBuffer;
+
+	/* An empty GIN index has two pages. */
+	MetaBuffer =
+		ReadBufferExtended(index, INIT_FORKNUM, P_NEW, RBM_NORMAL, NULL);
+	LockBuffer(MetaBuffer, BUFFER_LOCK_EXCLUSIVE);
+	RootBuffer =
+		ReadBufferExtended(index, INIT_FORKNUM, P_NEW, RBM_NORMAL, NULL);
+	LockBuffer(RootBuffer, BUFFER_LOCK_EXCLUSIVE);
+
+	/* Initialize both pages, mark them dirty, unlock and release buffer. */
+	START_CRIT_SECTION();
+	GinInitMetabuffer(MetaBuffer);
+	MarkBufferDirty(MetaBuffer);
+	GinInitBuffer(RootBuffer, GIN_LEAF);
+	MarkBufferDirty(RootBuffer);
+
+	/* XLOG the new pages */
+	log_newpage(&index->rd_smgr->smgr_rnode.node, INIT_FORKNUM,
+				BufferGetBlockNumber(MetaBuffer),
+				BufferGetPage(MetaBuffer));
+	log_newpage(&index->rd_smgr->smgr_rnode.node, INIT_FORKNUM,
+				BufferGetBlockNumber(RootBuffer),
+				BufferGetPage(RootBuffer));
+	END_CRIT_SECTION();
+
+	/* Unlock and release the buffers. */
+	UnlockReleaseBuffer(MetaBuffer);
+	UnlockReleaseBuffer(RootBuffer);
+
+	PG_RETURN_VOID();
+}
+
 /*
  * Inserts value during normal insertion
  */

src/backend/access/gist/gist.c

@@ -218,6 +218,19 @@ gistbuildCallback(Relation index,
 	MemoryContextReset(buildstate->tmpCtx);
 }
 
+/*
+ *	gistbuildempty() -- build an empty gist index in the initialization fork
+ */
+Datum
+gistbuildempty(PG_FUNCTION_ARGS)
+{
+	ereport(ERROR,
+			(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+			 errmsg("unlogged GIST indexes are not supported")));
+
+	PG_RETURN_VOID();
+}
+
 /*
  *	gistinsert -- wrapper for GiST tuple insertion.
  *

src/backend/access/hash/hash.c

@@ -69,7 +69,7 @@ hashbuild(PG_FUNCTION_ARGS)
 	estimate_rel_size(heap, NULL, &relpages, &reltuples);
 
 	/* Initialize the hash index metadata page and initial buckets */
-	num_buckets = _hash_metapinit(index, reltuples);
+	num_buckets = _hash_metapinit(index, reltuples, MAIN_FORKNUM);
 
 	/*
 	 * If we just insert the tuples into the index in scan order, then
@@ -113,6 +113,19 @@ hashbuild(PG_FUNCTION_ARGS)
 	PG_RETURN_POINTER(result);
 }
 
+/*
+ *	hashbuildempty() -- build an empty hash index in the initialization fork
+ */
+Datum
+hashbuildempty(PG_FUNCTION_ARGS)
+{
+	Relation	index = (Relation) PG_GETARG_POINTER(0);
+
+	_hash_metapinit(index, 0, INIT_FORKNUM);
+
+	PG_RETURN_VOID();
+}
+
 /*
  * Per-tuple callback from IndexBuildHeapScan
  */

src/backend/access/hash/hashovfl.c

@@ -259,7 +259,7 @@ _hash_getovflpage(Relation rel, Buffer metabuf)
 		 * convenient to pre-mark them as "in use" too.
 		 */
 		bit = metap->hashm_spares[splitnum];
-		_hash_initbitmap(rel, metap, bitno_to_blkno(metap, bit));
+		_hash_initbitmap(rel, metap, bitno_to_blkno(metap, bit), MAIN_FORKNUM);
 		metap->hashm_spares[splitnum]++;
 	}
 	else
@@ -280,7 +280,7 @@ _hash_getovflpage(Relation rel, Buffer metabuf)
 	 * with metapage write lock held; would be better to use a lock that
 	 * doesn't block incoming searches.
 	 */
-	newbuf = _hash_getnewbuf(rel, blkno);
+	newbuf = _hash_getnewbuf(rel, blkno, MAIN_FORKNUM);
 
 	metap->hashm_spares[splitnum]++;
 
@@ -503,7 +503,8 @@ _hash_freeovflpage(Relation rel, Buffer ovflbuf,
  * All bits in the new bitmap page are set to "1", indicating "in use".
  */
 void
-_hash_initbitmap(Relation rel, HashMetaPage metap, BlockNumber blkno)
+_hash_initbitmap(Relation rel, HashMetaPage metap, BlockNumber blkno,
+				 ForkNumber forkNum)
 {
 	Buffer		buf;
 	Page		pg;
@@ -520,7 +521,7 @@ _hash_initbitmap(Relation rel, HashMetaPage metap, BlockNumber blkno)
 	 * page while holding the metapage lock, but this path is taken so seldom
 	 * that it's not worth worrying about.
 	 */
-	buf = _hash_getnewbuf(rel, blkno);
+	buf = _hash_getnewbuf(rel, blkno, forkNum);
 	pg = BufferGetPage(buf);
 
 	/* initialize the page's special space */

src/backend/access/hash/hashpage.c

@@ -183,9 +183,9 @@ _hash_getinitbuf(Relation rel, BlockNumber blkno)
  *		extend the index at a time.
  */
 Buffer
-_hash_getnewbuf(Relation rel, BlockNumber blkno)
+_hash_getnewbuf(Relation rel, BlockNumber blkno, ForkNumber forkNum)
 {
-	BlockNumber nblocks = RelationGetNumberOfBlocks(rel);
+	BlockNumber nblocks = RelationGetNumberOfBlocksInFork(rel, forkNum);
 	Buffer		buf;
 
 	if (blkno == P_NEW)
@@ -197,13 +197,13 @@ _hash_getnewbuf(Relation rel, BlockNumber blkno)
 	/* smgr insists we use P_NEW to extend the relation */
 	if (blkno == nblocks)
 	{
-		buf = ReadBuffer(rel, P_NEW);
+		buf = ReadBufferExtended(rel, forkNum, P_NEW, RBM_NORMAL, NULL);
 		if (BufferGetBlockNumber(buf) != blkno)
 			elog(ERROR, "unexpected hash relation size: %u, should be %u",
 				 BufferGetBlockNumber(buf), blkno);
 	}
 	else
-		buf = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_ZERO, NULL);
+		buf = ReadBufferExtended(rel, forkNum, blkno, RBM_ZERO, NULL);
 
 	LockBuffer(buf, HASH_WRITE);
 
@@ -324,7 +324,7 @@ _hash_chgbufaccess(Relation rel,
  * multiple buffer locks is ignored.
  */
 uint32
-_hash_metapinit(Relation rel, double num_tuples)
+_hash_metapinit(Relation rel, double num_tuples, ForkNumber forkNum)
 {
 	HashMetaPage metap;
 	HashPageOpaque pageopaque;
@@ -340,7 +340,7 @@ _hash_metapinit(Relation rel, double num_tuples)
 	uint32		i;
 
 	/* safety check */
-	if (RelationGetNumberOfBlocks(rel) != 0)
+	if (RelationGetNumberOfBlocksInFork(rel, forkNum) != 0)
 		elog(ERROR, "cannot initialize non-empty hash index \"%s\"",
 			 RelationGetRelationName(rel));
 
@@ -383,7 +383,7 @@ _hash_metapinit(Relation rel, double num_tuples)
 	 * calls to occur.	This ensures that the smgr level has the right idea of
 	 * the physical index length.
 	 */
-	metabuf = _hash_getnewbuf(rel, HASH_METAPAGE);
+	metabuf = _hash_getnewbuf(rel, HASH_METAPAGE, forkNum);
 	pg = BufferGetPage(metabuf);
 
 	pageopaque = (HashPageOpaque) PageGetSpecialPointer(pg);
@@ -451,7 +451,7 @@ _hash_metapinit(Relation rel, double num_tuples)
 		/* Allow interrupts, in case N is huge */
 		CHECK_FOR_INTERRUPTS();
 
-		buf = _hash_getnewbuf(rel, BUCKET_TO_BLKNO(metap, i));
+		buf = _hash_getnewbuf(rel, BUCKET_TO_BLKNO(metap, i), forkNum);
 		pg = BufferGetPage(buf);
 		pageopaque = (HashPageOpaque) PageGetSpecialPointer(pg);
 		pageopaque->hasho_prevblkno = InvalidBlockNumber;
@@ -468,7 +468,7 @@ _hash_metapinit(Relation rel, double num_tuples)
 	/*
 	 * Initialize first bitmap page
 	 */
-	_hash_initbitmap(rel, metap, num_buckets + 1);
+	_hash_initbitmap(rel, metap, num_buckets + 1, forkNum);
 
 	/* all done */
 	_hash_wrtbuf(rel, metabuf);
@@ -785,7 +785,7 @@ _hash_splitbucket(Relation rel,
 	oopaque = (HashPageOpaque) PageGetSpecialPointer(opage);
 
 	nblkno = start_nblkno;
-	nbuf = _hash_getnewbuf(rel, nblkno);
+	nbuf = _hash_getnewbuf(rel, nblkno, MAIN_FORKNUM);
 	npage = BufferGetPage(nbuf);
 
 	/* initialize the new bucket's primary page */