Change WAL-logging scheme for multixacts to be more like regular

transaction IDs, rather than like subtrans; in particular, the information
now survives a database restart.  Per previous discussion, this is
essential for PITR log shipping and for 2PC.

doc/src/sgml/ref/pg_resetxlog.sgml

 <!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/pg_resetxlog.sgml,v 1.10 2005/04/28 21:47:10 tgl Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/pg_resetxlog.sgml,v 1.11 2005/06/08 15:50:21 tgl Exp $
 PostgreSQL documentation
 -->
 
@@ -23,6 +23,7 @@ PostgreSQL documentation
    <arg> -o <replaceable class="parameter">oid</replaceable> </arg>
    <arg> -x <replaceable class="parameter">xid</replaceable> </arg>
    <arg> -m <replaceable class="parameter">mxid</replaceable> </arg>
+   <arg> -O <replaceable class="parameter">mxoff</replaceable> </arg>
    <arg> -l <replaceable class="parameter">timelineid</replaceable>,<replaceable class="parameter">fileid</replaceable>,<replaceable class="parameter">seg</replaceable> </arg>
    <arg choice="plain"><replaceable>datadir</replaceable></arg>
   </cmdsynopsis>
@@ -32,8 +33,8 @@ PostgreSQL documentation
   <title>Description</title>
   <para>
    <command>pg_resetxlog</command> clears the write-ahead log (WAL) and
-   optionally resets some other control information (stored in the
-   <filename>pg_control</> file).  This function is sometimes needed
+   optionally resets some other control information stored in the
+   <filename>pg_control</> file.  This function is sometimes needed
    if these files have become corrupted.  It should be used only as a
    last resort, when the server will not start due to such corruption.
   </para>
@@ -60,8 +61,9 @@ PostgreSQL documentation
    by specifying the <literal>-f</> (force) switch.  In this case plausible
    values will be substituted for the missing data.  Most of the fields can be
    expected to match, but manual assistance may be needed for the next OID,
-   next transaction ID, WAL starting address, and database locale fields.
-   The first three of these can be set using the switches discussed below.
+   next transaction ID, next multi-transaction ID and offset,
+   WAL starting address, and database locale fields.
+   The first five of these can be set using the switches discussed below.
    <command>pg_resetxlog</command>'s own environment is the source for its
    guess at the locale fields; take care that <envar>LANG</> and so forth
    match the environment that <command>initdb</> was run in.
@@ -74,9 +76,10 @@ PostgreSQL documentation
   </para>
 
   <para>
-   The <literal>-o</>, <literal>-x</>, <literal>-m</>, and <literal>-l</>
+   The <literal>-o</>, <literal>-x</>, <literal>-m</>, <literal>-O</>,
+   and <literal>-l</>
    switches allow the next OID, next transaction ID, next multi-transaction
-   ID, and WAL starting address values to
+   ID, next multi-transaction offset, and WAL starting address values to
    be set manually.  These are only needed when
    <command>pg_resetxlog</command> is unable to determine appropriate values
    by reading <filename>pg_control</>.  Safe values may be determined as
@@ -108,6 +111,17 @@ PostgreSQL documentation
      </para>
     </listitem>
 
+    <listitem>
+     <para>
+      A safe value for the next multi-transaction offset (<literal>-O</>)
+      may be determined by looking for the numerically largest
+      file name in the directory <filename>pg_multixact/members</> under the
+      data directory, adding one, and then multiplying by 65536.  As above,
+      the file names are in hexadecimal, so the easiest way to do this is to
+      specify the switch value in hexadecimal and add four zeroes.
+     </para>
+    </listitem>
+
     <listitem>
      <para>
       The WAL starting address (<literal>-l</>) should be

src/backend/access/heap/heapam.c

@@ -8,7 +8,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/access/heap/heapam.c,v 1.193 2005/06/06 20:22:56 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/access/heap/heapam.c,v 1.194 2005/06/08 15:50:21 tgl Exp $
  *
  *
  * INTERFACE ROUTINES
@@ -2219,6 +2219,8 @@ l3:
 	 * Else the same IDs might be re-used after a crash, which would be
 	 * disastrous if this page made it to disk before the crash.  Essentially
 	 * we have to enforce the WAL log-before-data rule even in this case.
+	 * (Also, in a PITR log-shipping or 2PC environment, we have to have XLOG
+	 * entries for everything anyway.)
 	 */
 	if (!relation->rd_istemp)
 	{
@@ -2228,6 +2230,8 @@ l3:
 
 		xlrec.target.node = relation->rd_node;
 		xlrec.target.tid = tuple->t_self;
+		xlrec.locking_xid = xid;
+		xlrec.xid_is_mxact = ((new_infomask & HEAP_XMAX_IS_MULTI) != 0);
 		xlrec.shared_lock = (mode == LockTupleShared);
 		rdata[0].data = (char *) &xlrec;
 		rdata[0].len = SizeOfHeapLock;
@@ -2900,17 +2904,18 @@ heap_xlog_lock(XLogRecPtr lsn, XLogRecord *record)
 
 	htup = (HeapTupleHeader) PageGetItem(page, lp);
 
-	/*
-	 * Presently, we don't bother to restore the locked state, but
-	 * just set the XMAX_INVALID bit.
-	 */
 	htup->t_infomask &= ~(HEAP_XMAX_COMMITTED |
 						  HEAP_XMAX_INVALID |
 						  HEAP_XMAX_IS_MULTI |
 						  HEAP_IS_LOCKED |
 						  HEAP_MOVED);
-	htup->t_infomask |= HEAP_XMAX_INVALID;
-	HeapTupleHeaderSetXmax(htup, record->xl_xid);
+	if (xlrec->xid_is_mxact)
+		htup->t_infomask |= HEAP_XMAX_IS_MULTI;
+	if (xlrec->shared_lock)
+		htup->t_infomask |= HEAP_XMAX_SHARED_LOCK;
+	else
+		htup->t_infomask |= HEAP_XMAX_EXCL_LOCK;
+	HeapTupleHeaderSetXmax(htup, xlrec->locking_xid);
 	HeapTupleHeaderSetCmax(htup, FirstCommandId);
 	/* Make sure there is no forward chain link in t_ctid */
 	htup->t_ctid = xlrec->target.tid;
@@ -3010,6 +3015,11 @@ heap_desc(char *buf, uint8 xl_info, char *rec)
 			strcat(buf, "shared_lock: ");
 		else
 			strcat(buf, "exclusive_lock: ");
+		if (xlrec->xid_is_mxact)
+			strcat(buf, "mxid ");
+		else
+			strcat(buf, "xid ");
+		sprintf(buf + strlen(buf), "%u ", xlrec->locking_xid);
 		out_target(buf, &(xlrec->target));
 	}
 	else

src/backend/access/transam/rmgr.c

@@ -3,7 +3,7 @@
  *
  * Resource managers definition
  *
- * $PostgreSQL: pgsql/src/backend/access/transam/rmgr.c,v 1.18 2005/06/06 17:01:22 tgl Exp $
+ * $PostgreSQL: pgsql/src/backend/access/transam/rmgr.c,v 1.19 2005/06/08 15:50:26 tgl Exp $
  */
 #include "postgres.h"
 
@@ -11,6 +11,7 @@
 #include "access/gist_private.h"
 #include "access/hash.h"
 #include "access/heapam.h"
+#include "access/multixact.h"
 #include "access/nbtree.h"
 #include "access/rtree.h"
 #include "access/xact.h"
@@ -28,7 +29,7 @@ const RmgrData RmgrTable[RM_MAX_ID + 1] = {
 	{"CLOG", clog_redo, clog_desc, NULL, NULL},
 	{"Database", dbase_redo, dbase_desc, NULL, NULL},
 	{"Tablespace", tblspc_redo, tblspc_desc, NULL, NULL},
-	{"Reserved 6", NULL, NULL, NULL, NULL},
+	{"MultiXact", multixact_redo, multixact_desc, NULL, NULL},
 	{"Reserved 7", NULL, NULL, NULL, NULL},
 	{"Reserved 8", NULL, NULL, NULL, NULL},
 	{"Reserved 9", NULL, NULL, NULL, NULL},

src/backend/access/transam/xlog.c

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/backend/access/transam/xlog.c,v 1.197 2005/06/06 20:22:57 tgl Exp $
+ * $PostgreSQL: pgsql/src/backend/access/transam/xlog.c,v 1.198 2005/06/08 15:50:26 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -3688,12 +3688,13 @@ BootStrapXLOG(void)
 	checkPoint.nextXid = FirstNormalTransactionId;
 	checkPoint.nextOid = FirstBootstrapObjectId;
 	checkPoint.nextMulti = FirstMultiXactId;
+	checkPoint.nextMultiOffset = 0;
 	checkPoint.time = time(NULL);
 
 	ShmemVariableCache->nextXid = checkPoint.nextXid;
 	ShmemVariableCache->nextOid = checkPoint.nextOid;
 	ShmemVariableCache->oidCount = 0;
-	MultiXactSetNextMXact(checkPoint.nextMulti);
+	MultiXactSetNextMXact(checkPoint.nextMulti, checkPoint.nextMultiOffset);
 
 	/* Set up the XLOG page header */
 	page->xlp_magic = XLOG_PAGE_MAGIC;
@@ -4344,8 +4345,11 @@ StartupXLOG(void)
 					checkPoint.undo.xlogid, checkPoint.undo.xrecoff,
 					wasShutdown ? "TRUE" : "FALSE")));
 	ereport(LOG,
-			(errmsg("next transaction ID: %u; next OID: %u; next MultiXactId: %u",
-					checkPoint.nextXid, checkPoint.nextOid, checkPoint.nextMulti)));
+			(errmsg("next transaction ID: %u; next OID: %u",
+					checkPoint.nextXid, checkPoint.nextOid)));
+	ereport(LOG,
+			(errmsg("next MultiXactId: %u; next MultiXactOffset: %u",
+					checkPoint.nextMulti, checkPoint.nextMultiOffset)));
 	if (!TransactionIdIsNormal(checkPoint.nextXid))
 		ereport(PANIC,
 				(errmsg("invalid next transaction ID")));
@@ -4353,7 +4357,7 @@ StartupXLOG(void)
 	ShmemVariableCache->nextXid = checkPoint.nextXid;
 	ShmemVariableCache->nextOid = checkPoint.nextOid;
 	ShmemVariableCache->oidCount = 0;
-	MultiXactSetNextMXact(checkPoint.nextMulti);
+	MultiXactSetNextMXact(checkPoint.nextMulti, checkPoint.nextMultiOffset);
 
 	/*
 	 * We must replay WAL entries using the same TimeLineID they were
@@ -5080,7 +5084,9 @@ CreateCheckPoint(bool shutdown, bool force)
 		checkPoint.nextOid += ShmemVariableCache->oidCount;
 	LWLockRelease(OidGenLock);
 
-	checkPoint.nextMulti = MultiXactGetCheckptMulti(shutdown);
+	MultiXactGetCheckptMulti(shutdown,
+							 &checkPoint.nextMulti,
+							 &checkPoint.nextMultiOffset);
 
 	/*
 	 * Having constructed the checkpoint record, ensure all shmem disk
@@ -5228,25 +5234,6 @@ XLogPutNextOid(Oid nextOid)
 	 */
 }
 
-/*
- * Write a NEXT_MULTIXACT log record
- */
-void
-XLogPutNextMultiXactId(MultiXactId nextMulti)
-{
-	XLogRecData rdata;
-
-	rdata.data = (char *) (&nextMulti);
-	rdata.len = sizeof(MultiXactId);
-	rdata.buffer = InvalidBuffer;
-	rdata.next = NULL;
-	(void) XLogInsert(RM_XLOG_ID, XLOG_NEXTMULTI, &rdata);
-	/*
-	 * We do not flush here either; this assumes that heap_lock_tuple() will
-	 * always generate a WAL record.  See notes therein.
-	 */
-}
-
 /*
  * XLOG resource manager's routines
  */
@@ -5266,14 +5253,6 @@ xlog_redo(XLogRecPtr lsn, XLogRecord *record)
 			ShmemVariableCache->oidCount = 0;
 		}
 	}
-	else if (info == XLOG_NEXTMULTI)
-	{
-		MultiXactId	nextMulti;
-
-		memcpy(&nextMulti, XLogRecGetData(record), sizeof(MultiXactId));
-
-		MultiXactAdvanceNextMXact(nextMulti);
-	}
 	else if (info == XLOG_CHECKPOINT_SHUTDOWN)
 	{
 		CheckPoint	checkPoint;
@@ -5283,7 +5262,8 @@ xlog_redo(XLogRecPtr lsn, XLogRecord *record)
 		ShmemVariableCache->nextXid = checkPoint.nextXid;
 		ShmemVariableCache->nextOid = checkPoint.nextOid;
 		ShmemVariableCache->oidCount = 0;
-		MultiXactSetNextMXact(checkPoint.nextMulti);
+		MultiXactSetNextMXact(checkPoint.nextMulti,
+							  checkPoint.nextMultiOffset);
 
 		/*
 		 * TLI may change in a shutdown checkpoint, but it shouldn't
@@ -5315,7 +5295,8 @@ xlog_redo(XLogRecPtr lsn, XLogRecord *record)
 			ShmemVariableCache->nextOid = checkPoint.nextOid;
 			ShmemVariableCache->oidCount = 0;
 		}
-		MultiXactAdvanceNextMXact(checkPoint.nextMulti);
+		MultiXactAdvanceNextMXact(checkPoint.nextMulti,
+								  checkPoint.nextMultiOffset);
 		/* TLI should not change in an on-line checkpoint */
 		if (checkPoint.ThisTimeLineID != ThisTimeLineID)
 			ereport(PANIC,
@@ -5335,12 +5316,13 @@ xlog_desc(char *buf, uint8 xl_info, char *rec)
 		CheckPoint *checkpoint = (CheckPoint *) rec;
 
 		sprintf(buf + strlen(buf), "checkpoint: redo %X/%X; undo %X/%X; "
-				"tli %u; xid %u; oid %u; multi %u; %s",
+				"tli %u; xid %u; oid %u; multi %u; offset %u; %s",
 				checkpoint->redo.xlogid, checkpoint->redo.xrecoff,
 				checkpoint->undo.xlogid, checkpoint->undo.xrecoff,
 				checkpoint->ThisTimeLineID, checkpoint->nextXid,
 				checkpoint->nextOid,
 				checkpoint->nextMulti,
+				checkpoint->nextMultiOffset,
 			 (info == XLOG_CHECKPOINT_SHUTDOWN) ? "shutdown" : "online");
 	}
 	else if (info == XLOG_NEXTOID)
@@ -5350,13 +5332,6 @@ xlog_desc(char *buf, uint8 xl_info, char *rec)
 		memcpy(&nextOid, rec, sizeof(Oid));
 		sprintf(buf + strlen(buf), "nextOid: %u", nextOid);
 	}
-	else if (info == XLOG_NEXTMULTI)
-	{
-		MultiXactId	multi;
-
-		memcpy(&multi, rec, sizeof(MultiXactId));
-		sprintf(buf + strlen(buf), "nextMultiXact: %u", multi);
-	}
 	else
 		strcat(buf, "UNKNOWN");
 }

src/bin/pg_controldata/pg_controldata.c

@@ -6,7 +6,7 @@
  * copyright (c) Oliver Elphick <olly@lfix.co.uk>, 2001;
  * licence: BSD
  *
- * $PostgreSQL: pgsql/src/bin/pg_controldata/pg_controldata.c,v 1.24 2005/06/02 05:55:29 tgl Exp $
+ * $PostgreSQL: pgsql/src/bin/pg_controldata/pg_controldata.c,v 1.25 2005/06/08 15:50:27 tgl Exp $
  */
 #include "postgres.h"
 
@@ -166,6 +166,7 @@ main(int argc, char *argv[])
 	printf(_("Latest checkpoint's NextXID:          %u\n"), ControlFile.checkPointCopy.nextXid);
 	printf(_("Latest checkpoint's NextOID:          %u\n"), ControlFile.checkPointCopy.nextOid);
 	printf(_("Latest checkpoint's NextMultiXactId:  %u\n"), ControlFile.checkPointCopy.nextMulti);
+	printf(_("Latest checkpoint's NextMultiOffset:  %u\n"), ControlFile.checkPointCopy.nextMultiOffset);
 	printf(_("Time of latest checkpoint:            %s\n"), ckpttime_str);
 	printf(_("Database block size:                  %u\n"), ControlFile.blcksz);
 	printf(_("Blocks per segment of large relation: %u\n"), ControlFile.relseg_size);

src/bin/pg_resetxlog/pg_resetxlog.c

@@ -23,7 +23,7 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/bin/pg_resetxlog/pg_resetxlog.c,v 1.33 2005/06/02 05:55:29 tgl Exp $
+ * $PostgreSQL: pgsql/src/bin/pg_resetxlog/pg_resetxlog.c,v 1.34 2005/06/08 15:50:27 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -77,6 +77,7 @@ main(int argc, char *argv[])
 	TransactionId set_xid = 0;
 	Oid			set_oid = 0;
 	MultiXactId	set_mxid = 0;
+	MultiXactOffset set_mxoff = -1;
 	uint32		minXlogTli = 0,
 				minXlogId = 0,
 				minXlogSeg = 0;
@@ -106,7 +107,7 @@ main(int argc, char *argv[])
 	}
 
 
-	while ((c = getopt(argc, argv, "fl:m:no:x:")) != -1)
+	while ((c = getopt(argc, argv, "fl:m:no:O:x:")) != -1)
 	{
 		switch (c)
 		{
@@ -163,6 +164,21 @@ main(int argc, char *argv[])
 				}
 				break;
 
+			case 'O':
+				set_mxoff = strtoul(optarg, &endptr, 0);
+				if (endptr == optarg || *endptr != '\0')
+				{
+					fprintf(stderr, _("%s: invalid argument for option -O\n"), progname);
+					fprintf(stderr, _("Try \"%s --help\" for more information.\n"), progname);
+					exit(1);
+				}
+				if (set_mxoff == -1)
+				{
+					fprintf(stderr, _("%s: multi transaction offset (-O) must not be -1\n"), progname);
+					exit(1);
+				}
+				break;
+
 			case 'l':
 				minXlogTli = strtoul(optarg, &endptr, 0);
 				if (endptr == optarg || *endptr != ',')
@@ -265,6 +281,9 @@ main(int argc, char *argv[])
 	if (set_mxid != 0)
 		ControlFile.checkPointCopy.nextMulti = set_mxid;
 
+	if (set_mxoff != -1)
+		ControlFile.checkPointCopy.nextMultiOffset = set_mxoff;
+
 	if (minXlogTli > ControlFile.checkPointCopy.ThisTimeLineID)
 		ControlFile.checkPointCopy.ThisTimeLineID = minXlogTli;
 
@@ -426,6 +445,7 @@ GuessControlValues(void)
 	ControlFile.checkPointCopy.nextXid = (TransactionId) 514;	/* XXX */
 	ControlFile.checkPointCopy.nextOid = FirstBootstrapObjectId;
 	ControlFile.checkPointCopy.nextMulti = FirstMultiXactId;
+	ControlFile.checkPointCopy.nextMultiOffset = 0;
 	ControlFile.checkPointCopy.time = time(NULL);
 
 	ControlFile.state = DB_SHUTDOWNED;
@@ -463,7 +483,7 @@ GuessControlValues(void)
 
 	/*
 	 * XXX eventually, should try to grovel through old XLOG to develop
-	 * more accurate values for TimeLineID, nextXID, and nextOID.
+	 * more accurate values for TimeLineID, nextXID, etc.
 	 */
 }
 
@@ -500,6 +520,7 @@ PrintControlValues(bool guessed)
 	printf(_("Latest checkpoint's NextXID:          %u\n"), ControlFile.checkPointCopy.nextXid);
 	printf(_("Latest checkpoint's NextOID:          %u\n"), ControlFile.checkPointCopy.nextOid);
 	printf(_("Latest checkpoint's NextMultiXactId:  %u\n"), ControlFile.checkPointCopy.nextMulti);
+	printf(_("Latest checkpoint's NextMultiOffset:  %u\n"), ControlFile.checkPointCopy.nextMultiOffset);
 	printf(_("Database block size:                  %u\n"), ControlFile.blcksz);
 	printf(_("Blocks per segment of large relation: %u\n"), ControlFile.relseg_size);
 	printf(_("Maximum length of identifiers:        %u\n"), ControlFile.nameDataLen);
@@ -777,6 +798,7 @@ usage(void)
 	printf(_("  -o OID          set next OID\n"));
 	printf(_("  -x XID          set next transaction ID\n"));
 	printf(_("  -m multiXID     set next multi transaction ID\n"));
+	printf(_("  -O multiOffset  set next multi transaction offset\n"));
 	printf(_("  --help          show this help, then exit\n"));
 	printf(_("  --version       output version information, then exit\n"));
 	printf(_("\nReport bugs to <pgsql-bugs@postgresql.org>.\n"));

src/include/access/htup.h

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/access/htup.h,v 1.74 2005/04/28 21:47:17 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/access/htup.h,v 1.75 2005/06/08 15:50:27 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -505,6 +505,8 @@ typedef struct xl_heap_newpage
 typedef struct xl_heap_lock
 {
 	xl_heaptid	target;			/* locked tuple id */
+	TransactionId locking_xid;	/* might be a MultiXactId not xid */
+	bool		xid_is_mxact;	/* is it? */
 	bool		shared_lock;	/* shared or exclusive row lock? */
 } xl_heap_lock;
 

src/include/access/multixact.h

@@ -6,16 +6,38 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/access/multixact.h,v 1.2 2005/05/03 19:42:41 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/access/multixact.h,v 1.3 2005/06/08 15:50:28 tgl Exp $
  */
 #ifndef MULTIXACT_H
 #define MULTIXACT_H
 
+#include "access/xlog.h"
+
 #define InvalidMultiXactId	((MultiXactId) 0)
 #define FirstMultiXactId	((MultiXactId) 1)
 
 #define MultiXactIdIsValid(multi) ((multi) != InvalidMultiXactId)
 
+/* ----------------
+ *		multixact-related XLOG entries
+ * ----------------
+ */
+
+#define XLOG_MULTIXACT_ZERO_OFF_PAGE	0x00
+#define XLOG_MULTIXACT_ZERO_MEM_PAGE	0x10
+#define XLOG_MULTIXACT_CREATE_ID		0x20
+
+typedef struct xl_multixact_create
+{
+	MultiXactId		mid;		/* new MultiXact's ID */
+	MultiXactOffset	moff;		/* its starting offset in members file */
+	int32			nxids;		/* number of member XIDs */
+	TransactionId	xids[1];	/* VARIABLE LENGTH ARRAY */
+} xl_multixact_create;
+
+#define MinSizeOfMultiXactCreate offsetof(xl_multixact_create, xids)
+
+
 extern MultiXactId MultiXactIdCreate(TransactionId xid1, TransactionId xid2);
 extern MultiXactId MultiXactIdExpand(MultiXactId multi, TransactionId xid);
 extern bool MultiXactIdIsRunning(MultiXactId multi);
@@ -29,9 +51,16 @@ extern void MultiXactShmemInit(void);
 extern void BootStrapMultiXact(void);
 extern void StartupMultiXact(void);
 extern void ShutdownMultiXact(void);
-extern MultiXactId MultiXactGetCheckptMulti(bool is_shutdown);
+extern void MultiXactGetCheckptMulti(bool is_shutdown,
+									 MultiXactId *nextMulti,
+									 MultiXactOffset *nextMultiOffset);
 extern void CheckPointMultiXact(void);
-extern void MultiXactSetNextMXact(MultiXactId nextMulti);
-extern void MultiXactAdvanceNextMXact(MultiXactId minMulti);
+extern void MultiXactSetNextMXact(MultiXactId nextMulti,
+								  MultiXactOffset nextMultiOffset);
+extern void MultiXactAdvanceNextMXact(MultiXactId minMulti,
+									  MultiXactOffset minMultiOffset);
+
+extern void multixact_redo(XLogRecPtr lsn, XLogRecord *record);
+extern void multixact_desc(char *buf, uint8 xl_info, char *rec);
 
 #endif   /* MULTIXACT_H */

src/include/access/xlog.h

@@ -6,7 +6,7 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/access/xlog.h,v 1.64 2005/06/06 20:22:58 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/access/xlog.h,v 1.65 2005/06/08 15:50:28 tgl Exp $
  */
 #ifndef XLOG_H
 #define XLOG_H
@@ -165,7 +165,6 @@ extern void ShutdownXLOG(int code, Datum arg);
 extern void InitXLOGAccess(void);
 extern void CreateCheckPoint(bool shutdown, bool force);
 extern void XLogPutNextOid(Oid nextOid);
-extern void XLogPutNextMultiXactId(MultiXactId multi);
 extern XLogRecPtr GetRedoRecPtr(void);
 
 #endif   /* XLOG_H */

src/include/c.h

@@ -12,7 +12,7 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/c.h,v 1.184 2005/05/25 21:40:41 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/c.h,v 1.185 2005/06/08 15:50:28 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -388,6 +388,8 @@ typedef uint32 SubTransactionId;
 /* MultiXactId must be equivalent to TransactionId, to fit in t_xmax */
 typedef TransactionId MultiXactId;
 
+typedef uint32 MultiXactOffset;
+
 typedef uint32 CommandId;
 
 #define FirstCommandId	((CommandId) 0)

src/include/catalog/pg_control.h

@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1996-2005, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/catalog/pg_control.h,v 1.22 2005/06/02 05:55:29 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/catalog/pg_control.h,v 1.23 2005/06/08 15:50:28 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -22,7 +22,7 @@
 
 
 /* Version identifier for this pg_control format */
-#define PG_CONTROL_VERSION	810
+#define PG_CONTROL_VERSION	811
 
 /*
  * Body of CheckPoint XLOG records.  This is declared here because we keep
@@ -40,13 +40,13 @@ typedef struct CheckPoint
 	TransactionId nextXid;		/* next free XID */
 	Oid			nextOid;		/* next free OID */
 	MultiXactId	nextMulti;		/* next free MultiXactId */
+	MultiXactOffset nextMultiOffset;	/* next free MultiXact offset */
 	time_t		time;			/* time stamp of checkpoint */
 } CheckPoint;
 
 /* XLOG info values for XLOG rmgr */
 #define XLOG_CHECKPOINT_SHUTDOWN		0x00
 #define XLOG_CHECKPOINT_ONLINE			0x10
-#define XLOG_NEXTMULTI					0x20
 #define XLOG_NEXTOID					0x30