Implement "distributed" checkpoints in which the checkpoint I/O is spread

over a fairly long period of time, rather than being spat out in a burst.
This happens only for background checkpoints carried out by the bgwriter;
other cases, such as a shutdown checkpoint, are still done at full speed.

Remove the "all buffers" scan in the bgwriter, and associated stats
infrastructure, since this seems no longer very useful when the checkpoint
itself is properly throttled.

Original patch by Itagaki Takahiro, reworked by Heikki Linnakangas,
and some minor API editorialization by me.

7 files changed, 32 insertions, 30 deletions

diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h
index 1c741f38fd0..e4f9747de7d 100644
--- a/src/include/access/xlog.h
+++ b/src/include/access/xlog.h
@@ -6,7 +6,7 @@
  * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/access/xlog.h,v 1.78 2007/05/30 20:12:02 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/access/xlog.h,v 1.79 2007/06/28 00:02:39 tgl Exp $
  */
 #ifndef XLOG_H
 #define XLOG_H
@@ -157,6 +157,14 @@ extern const char XLOG_sync_method_default[];
 extern bool XLOG_DEBUG;
 #endif
 
+/* OR-able flags for RequestCheckpoint, CreateCheckPoint and subsidiaries */
+#define CHECKPOINT_IS_SHUTDOWN	0x0001		/* Checkpoint is for shutdown */
+#define CHECKPOINT_IMMEDIATE	0x0002		/* Do it without delays */
+#define CHECKPOINT_FORCE		0x0004		/* Force even if no activity */
+#define CHECKPOINT_WARNONTIME	0x0008		/* Enable CheckPointWarning */
+#define CHECKPOINT_WAIT			0x0010		/* Wait for completion */
+
+
 extern XLogRecPtr XLogInsert(RmgrId rmid, uint8 info, XLogRecData *rdata);
 extern void XLogFlush(XLogRecPtr RecPtr);
 extern bool XLogNeedsFlush(XLogRecPtr RecPtr);
@@ -171,9 +179,10 @@ extern void BootStrapXLOG(void);
 extern void StartupXLOG(void);
 extern void ShutdownXLOG(int code, Datum arg);
 extern void InitXLOGAccess(void);
-extern void CreateCheckPoint(bool shutdown, bool force);
+extern void CreateCheckPoint(int flags);
 extern void XLogPutNextOid(Oid nextOid);
 extern XLogRecPtr GetRedoRecPtr(void);
+extern XLogRecPtr GetInsertRecPtr(void);
 extern void GetNextXidAndEpoch(TransactionId *xid, uint32 *epoch);
 
 #endif   /* XLOG_H */
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index 415af0e6cb2..2975721590a 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -37,7 +37,7 @@
  * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/catalog/catversion.h,v 1.412 2007/06/15 20:56:51 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/catalog/catversion.h,v 1.413 2007/06/28 00:02:39 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -53,6 +53,6 @@
  */
 
 /*							yyyymmddN */
-#define CATALOG_VERSION_NO	200706151
+#define CATALOG_VERSION_NO	200706271
 
 #endif
diff --git a/src/include/catalog/pg_proc.h b/src/include/catalog/pg_proc.h
index 11e13b36aae..ecd1f118c65 100644
--- a/src/include/catalog/pg_proc.h
+++ b/src/include/catalog/pg_proc.h
@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/catalog/pg_proc.h,v 1.460 2007/06/15 20:56:51 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/catalog/pg_proc.h,v 1.461 2007/06/28 00:02:39 tgl Exp $
  *
  * NOTES
  *	  The script catalog/genbki.sh reads this file and generates .bki
@@ -2933,14 +2933,10 @@ DATA(insert OID = 2770 ( pg_stat_get_bgwriter_requested_checkpoints PGNSP PGUID
 DESCR("Statistics: Number of backend requested checkpoints started by the bgwriter");
 DATA(insert OID = 2771 ( pg_stat_get_bgwriter_buf_written_checkpoints PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_buf_written_checkpoints - _null_ ));
 DESCR("Statistics: Number of buffers written by the bgwriter during checkpoints");
-DATA(insert OID = 2772 ( pg_stat_get_bgwriter_buf_written_lru PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_buf_written_lru - _null_ ));
-DESCR("Statistics: Number of buffers written by the bgwriter during LRU scans");
-DATA(insert OID = 2773 ( pg_stat_get_bgwriter_buf_written_all PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_buf_written_all - _null_ ));
-DESCR("Statistics: Number of buffers written by the bgwriter during all-buffer scans");
-DATA(insert OID = 2774 ( pg_stat_get_bgwriter_maxwritten_lru PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_maxwritten_lru - _null_ ));
-DESCR("Statistics: Number of times the bgwriter stopped processing when it had written too many buffers during LRU scans");
-DATA(insert OID = 2775 ( pg_stat_get_bgwriter_maxwritten_all PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_maxwritten_all - _null_ ));
-DESCR("Statistics: Number of times the bgwriter stopped processing when it had written too many buffers during all-buffer scans");
+DATA(insert OID = 2772 ( pg_stat_get_bgwriter_buf_written_clean PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_buf_written_clean - _null_ ));
+DESCR("Statistics: Number of buffers written by the bgwriter for cleaning dirty buffers");
+DATA(insert OID = 2773 ( pg_stat_get_bgwriter_maxwritten_clean PGNSP PGUID 12 1 0 f f t f s 0 20 "" _null_ _null_ _null_ pg_stat_get_bgwriter_maxwritten_clean - _null_ ));
+DESCR("Statistics: Number of times the bgwriter stopped processing when it had written too many buffers while cleaning");
 DATA(insert OID = 2230 (  pg_stat_clear_snapshot		PGNSP PGUID 12 1 0 f f f f v 0 2278  "" _null_ _null_ _null_	pg_stat_clear_snapshot - _null_ ));
 DESCR("Statistics: Discard current transaction's statistics snapshot");
 DATA(insert OID = 2274 (  pg_stat_reset					PGNSP PGUID 12 1 0 f f f f v 0 2278  "" _null_ _null_ _null_	pg_stat_reset - _null_ ));
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index da910333531..acdd039b3e4 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -5,7 +5,7 @@
  *
  *	Copyright (c) 2001-2007, PostgreSQL Global Development Group
  *
- *	$PostgreSQL: pgsql/src/include/pgstat.h,v 1.62 2007/06/09 18:49:55 tgl Exp $
+ *	$PostgreSQL: pgsql/src/include/pgstat.h,v 1.63 2007/06/28 00:02:40 tgl Exp $
  * ----------
  */
 #ifndef PGSTAT_H
@@ -291,10 +291,8 @@ typedef struct PgStat_MsgBgWriter
 	PgStat_Counter  m_timed_checkpoints;
 	PgStat_Counter	m_requested_checkpoints;
 	PgStat_Counter	m_buf_written_checkpoints;
-	PgStat_Counter	m_buf_written_lru;
-	PgStat_Counter	m_buf_written_all;
-	PgStat_Counter	m_maxwritten_lru;
-	PgStat_Counter	m_maxwritten_all;
+	PgStat_Counter	m_buf_written_clean;
+	PgStat_Counter	m_maxwritten_clean;
 } PgStat_MsgBgWriter;
 
 
@@ -392,10 +390,8 @@ typedef struct PgStat_GlobalStats
 	PgStat_Counter  timed_checkpoints;
 	PgStat_Counter  requested_checkpoints;
 	PgStat_Counter  buf_written_checkpoints;
-	PgStat_Counter  buf_written_lru;
-	PgStat_Counter  buf_written_all;
-	PgStat_Counter  maxwritten_lru;
-	PgStat_Counter  maxwritten_all;
+	PgStat_Counter  buf_written_clean;
+	PgStat_Counter  maxwritten_clean;
 } PgStat_GlobalStats;
 
 
diff --git a/src/include/postmaster/bgwriter.h b/src/include/postmaster/bgwriter.h
index 8698687733d..7b62cb02eca 100644
--- a/src/include/postmaster/bgwriter.h
+++ b/src/include/postmaster/bgwriter.h
@@ -5,7 +5,7 @@
  *
  * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
  *
- * $PostgreSQL: pgsql/src/include/postmaster/bgwriter.h,v 1.9 2007/01/05 22:19:57 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/postmaster/bgwriter.h,v 1.10 2007/06/28 00:02:40 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -20,10 +20,12 @@
 extern int	BgWriterDelay;
 extern int	CheckPointTimeout;
 extern int	CheckPointWarning;
+extern double CheckPointCompletionTarget;
 
 extern void BackgroundWriterMain(void);
 
-extern void RequestCheckpoint(bool waitforit, bool warnontime);
+extern void RequestCheckpoint(int flags);
+extern void CheckpointWriteDelay(int flags, double progress);
 
 extern bool ForwardFsyncRequest(RelFileNode rnode, BlockNumber segno);
 extern void AbsorbFsyncRequests(void);
diff --git a/src/include/storage/buf_internals.h b/src/include/storage/buf_internals.h
index d5eef8734ff..b6772f90f2c 100644
--- a/src/include/storage/buf_internals.h
+++ b/src/include/storage/buf_internals.h
@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/storage/buf_internals.h,v 1.90 2007/05/30 20:12:03 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/storage/buf_internals.h,v 1.91 2007/06/28 00:02:40 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -35,6 +35,7 @@
 #define BM_IO_ERROR				(1 << 4)		/* previous I/O failed */
 #define BM_JUST_DIRTIED			(1 << 5)		/* dirtied since write started */
 #define BM_PIN_COUNT_WAITER		(1 << 6)		/* have waiter for sole pin */
+#define BM_CHECKPOINT_NEEDED	(1 << 7)		/* must write for checkpoint */
 
 typedef bits16 BufFlags;
 
diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 9ae83b4253e..95c603f961f 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/storage/bufmgr.h,v 1.104 2007/05/30 20:12:03 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/storage/bufmgr.h,v 1.105 2007/06/28 00:02:40 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -33,9 +33,7 @@ extern DLLIMPORT int NBuffers;
 /* in bufmgr.c */
 extern bool zero_damaged_pages;
 extern double bgwriter_lru_percent;
-extern double bgwriter_all_percent;
 extern int	bgwriter_lru_maxpages;
-extern int	bgwriter_all_maxpages;
 
 /* in buf_init.c */
 extern DLLIMPORT char *BufferBlocks;
@@ -136,11 +134,12 @@ extern char *ShowBufferUsage(void);
 extern void ResetBufferUsage(void);
 extern void AtEOXact_Buffers(bool isCommit);
 extern void PrintBufferLeakWarning(Buffer buffer);
-extern void FlushBufferPool(void);
+extern void CheckPointBuffers(int flags);
 extern BlockNumber BufferGetBlockNumber(Buffer buffer);
 extern BlockNumber RelationGetNumberOfBlocks(Relation relation);
 extern void RelationTruncate(Relation rel, BlockNumber nblocks);
 extern void FlushRelationBuffers(Relation rel);
+extern void FlushDatabaseBuffers(Oid dbid);
 extern void DropRelFileNodeBuffers(RelFileNode rnode, bool istemp,
 					   BlockNumber firstDelBlock);
 extern void DropDatabaseBuffers(Oid dbid);
@@ -161,7 +160,6 @@ extern void LockBufferForCleanup(Buffer buffer);
 extern void AbortBufferIO(void);
 
 extern void BufmgrCommit(void);
-extern void BufferSync(void);
 extern void BgBufferSync(void);
 
 extern void AtProcExit_LocalBuffers(void);