io-stats.v5.patch

diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index 3e17875..97e8e51 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -4284,6 +4284,25 @@ COPY postgres_log FROM '/full/path/to/logfile.csv' WITH csv;
       </listitem>
      </varlistentry>
 
+     <varlistentry id="guc-track-iotiming" xreflabel="track_iotiming">
+      <term><varname>track_iotiming</varname> (<type>boolean</type>)</term>
+      <indexterm>
+       <primary><varname>track_iotiming</> configuration parameter</primary>
+      </indexterm>
+      <listitem>
+       <para>
+        Enables timing of database I/O calls.
+        This parameter is off by default, because it may cause significant
+        overhead if the platform doesn't support fast timing information.
+        Only superusers can change this setting.
+       </para>
+       <para>
+        You can use the <xref linkend="pgtesttiming"> tool to find out the
+        overhead of timing on your system.
+       </para>
+      </listitem>
+     </varlistentry>
+
      <varlistentry id="guc-track-functions" xreflabel="track_functions">
       <term><varname>track_functions</varname> (<type>enum</type>)</term>
       <indexterm>
diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 840e54a..ba69382 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -116,7 +116,8 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re
    <productname>PostgreSQL</productname>'s <firstterm>statistics collector</>
    is a subsystem that supports collection and reporting of information about
    server activity.  Presently, the collector can count accesses to tables
-   and indexes in both disk-block and individual-row terms.  It also tracks
+   and indexes in both disk-block and individual-row terms and time disk-block
+   accesses.  It also tracks
    the total number of rows in each table, and information about vacuum and
    analyze actions for each table.  It can also count calls to user-defined
    functions and the total time spent in each one.
@@ -145,6 +146,11 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re
   </para>
 
   <para>
+   The parameter <xref linkend="guc-track-iotiming"> enables timing of I/O
+   requests.
+  </para>
+
+  <para>
    The parameter <xref linkend="guc-track-functions"> enables tracking of
    usage of user-defined functions.
   </para>
@@ -483,7 +489,10 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re
    handles disk I/O, data that is not in the
    <productname>PostgreSQL</> buffer cache might still reside in the
    kernel's I/O cache, and might therefore still be fetched without
-   requiring a physical read. Users interested in obtaining more
+   requiring a physical read.
+   Timing info shows how much user queries were delayed by buffer reads
+   in aggregate.
+   Users interested in obtaining more
    detailed information on <productname>PostgreSQL</> I/O behavior are
    advised to use the <productname>PostgreSQL</> statistics collector
    in combination with operating system utilities that allow insight
@@ -885,6 +894,14 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
      the <function>pg_stat_get_db_blocks_hit</function> function.</entry>
     </row>
     <row>
+     <entry>blks_time</entry>
+     <entry><type>bigint</></entry>
+     <entry>Total time spent waiting for disk blocks to be read
+     in this database, in microseconds.
+     This value can also be returned by directly calling
+     the <function>pg_stat_get_db_blocks_time</function> function.</entry>
+    </row>
+    <row>
      <entry>tup_returned</entry>
      <entry><type>bigint</></entry>
      <entry>The number of rows returned by queries in this database.
@@ -1441,6 +1458,14 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
      the <function>pg_stat_get_blocks_hit</function> function.</entry>
     </row>
     <row>
+     <entry>heap_blks_time</entry>
+     <entry><type>bigint</></entry>
+     <entry>Total time spent waiting for disk blocks to be read
+     from this table, in microseconds.
+     This value can also be returned by directly calling
+     the <function>pg_stat_get_blocks_time</function> function.</entry>
+    </row>
+    <row>
      <entry>idx_blks_read</entry>
      <entry><type>name</></entry>
      <entry>Number of disk blocks read from all indexes on this table</entry>
@@ -1451,6 +1476,12 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
      <entry>Number of buffer hits in all indexes of this table.</entry>
     </row>
     <row>
+     <entry>idx_blks_time</entry>
+     <entry><type>bigint</></entry>
+     <entry>Total time spent waiting for disk blocks to be read
+     from all indexes on this table, in microseconds.</entry>
+    </row>
+    <row>
      <entry>toast_blks_read</entry>
      <entry><type>name</></entry>
      <entry>Number of disk blocks read from this table's TOAST table (if any)</entry>
@@ -1461,6 +1492,12 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
      <entry>Number of buffer hits in this table's TOAST table (if any)</entry>
     </row>
     <row>
+     <entry>toast_blks_time</entry>
+     <entry><type>bigint</></entry>
+     <entry>Total time spent waiting for disk blocks to be read
+     from this table's TOAST table (if any), in microseconds.</entry>
+    </row>
+    <row>
      <entry>tidx_blks_read</entry>
      <entry><type>name</></entry>
      <entry>Number of disk blocks read from this table's TOAST table index (if any)</entry>
@@ -1470,6 +1507,12 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
      <entry><type>name</></entry>
      <entry>Number of buffer hits in this table's TOAST table index (if any)</entry>
     </row>
+    <row>
+     <entry>tidx_blks_time</entry>
+     <entry><type>bigint</></entry>
+     <entry>Total time spent waiting for disk blocks to be read
+     from this table's TOAST table index (if any), in microseconds.</entry>
+    </row>
    </tbody>
    </tgroup>
   </table>
@@ -1536,6 +1579,14 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
      This value can also be returned by directly calling
      the <function>pg_stat_get_blocks_hit</function> function.</entry>
     </row>
+    <row>
+     <entry>idx_blks_time</entry>
+     <entry><type>bigint</></entry>
+     <entry>Total time spent waiting for disk blocks to be read
+     from the index, in microseconds.
+     This value can also be returned by directly calling
+     the <function>pg_stat_get_blocks_time</function> function.</entry>
+    </row>
    </tbody>
    </tgroup>
   </table>
@@ -1703,6 +1754,14 @@ SELECT pg_stat_get_backend_pid(s.backendid) AS pid,
       </entry>
      </row>
 
+     <row>
+      <entry><literal><function>pg_stat_get_xact_blocks_time</function>(<type>oid</type>)</literal></entry>
+      <entry><type>bigint</type></entry>
+      <entry>
+       Total time spent waiting for disk block requests for table or index, in the current transaction,
+       in microseconds.
+      </entry>
+     </row>
 
      <row>
       <entry><literal><function>pg_stat_get_wal_senders()</function></literal></entry>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index ab594eb..430bea1 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -440,15 +440,19 @@ CREATE VIEW pg_statio_all_tables AS
             pg_stat_get_blocks_fetched(C.oid) -
                     pg_stat_get_blocks_hit(C.oid) AS heap_blks_read,
             pg_stat_get_blocks_hit(C.oid) AS heap_blks_hit,
+            pg_stat_get_blocks_time(C.oid) AS heap_blks_time,
             sum(pg_stat_get_blocks_fetched(I.indexrelid) -
                     pg_stat_get_blocks_hit(I.indexrelid))::bigint AS idx_blks_read,
             sum(pg_stat_get_blocks_hit(I.indexrelid))::bigint AS idx_blks_hit,
+            sum(pg_stat_get_blocks_time(I.indexrelid))::bigint AS idx_blks_time,
             pg_stat_get_blocks_fetched(T.oid) -
                     pg_stat_get_blocks_hit(T.oid) AS toast_blks_read,
             pg_stat_get_blocks_hit(T.oid) AS toast_blks_hit,
+            pg_stat_get_blocks_time(T.oid) AS toast_blks_time,
             pg_stat_get_blocks_fetched(X.oid) -
                     pg_stat_get_blocks_hit(X.oid) AS tidx_blks_read,
-            pg_stat_get_blocks_hit(X.oid) AS tidx_blks_hit
+            pg_stat_get_blocks_hit(X.oid) AS tidx_blks_hit,
+            pg_stat_get_blocks_time(X.oid) AS tidx_blks_time
     FROM pg_class C LEFT JOIN
             pg_index I ON C.oid = I.indrelid LEFT JOIN
             pg_class T ON C.reltoastrelid = T.oid LEFT JOIN
@@ -502,7 +506,8 @@ CREATE VIEW pg_statio_all_indexes AS
             I.relname AS indexrelname,
             pg_stat_get_blocks_fetched(I.oid) -
                     pg_stat_get_blocks_hit(I.oid) AS idx_blks_read,
-            pg_stat_get_blocks_hit(I.oid) AS idx_blks_hit
+            pg_stat_get_blocks_hit(I.oid) AS idx_blks_hit,
+            pg_stat_get_blocks_time(I.oid) AS idx_blks_time
     FROM pg_class C JOIN
             pg_index X ON C.oid = X.indrelid JOIN
             pg_class I ON I.oid = X.indexrelid
@@ -526,7 +531,8 @@ CREATE VIEW pg_statio_all_sequences AS
             C.relname AS relname,
             pg_stat_get_blocks_fetched(C.oid) -
                     pg_stat_get_blocks_hit(C.oid) AS blks_read,
-            pg_stat_get_blocks_hit(C.oid) AS blks_hit
+            pg_stat_get_blocks_hit(C.oid) AS blks_hit,
+            pg_stat_get_blocks_time(C.oid) AS blks_time
     FROM pg_class C
             LEFT JOIN pg_namespace N ON (N.oid = C.relnamespace)
     WHERE C.relkind = 'S';
@@ -595,6 +601,7 @@ CREATE VIEW pg_stat_database AS
             pg_stat_get_db_blocks_fetched(D.oid) -
                     pg_stat_get_db_blocks_hit(D.oid) AS blks_read,
             pg_stat_get_db_blocks_hit(D.oid) AS blks_hit,
+            pg_stat_get_db_blocks_time(D.oid) AS blks_time,
             pg_stat_get_db_tuples_returned(D.oid) AS tup_returned,
             pg_stat_get_db_tuples_fetched(D.oid) AS tup_fetched,
             pg_stat_get_db_tuples_inserted(D.oid) AS tup_inserted,
diff --git a/src/backend/commands/explain.c b/src/backend/commands/explain.c
index a14cae1..3707f7e 100644
--- a/src/backend/commands/explain.c
+++ b/src/backend/commands/explain.c
@@ -1236,9 +1236,11 @@ ExplainNode(PlanState *planstate, List *ancestors,
 									 usage->local_blks_written > 0);
 			bool		has_temp = (usage->temp_blks_read > 0 ||
 									usage->temp_blks_written > 0);
+			bool		has_timing = (!INSTR_TIME_IS_ZERO(usage->time_read) ||
+									  !INSTR_TIME_IS_ZERO(usage->time_write));
 
 			/* Show only positive counter values. */
-			if (has_shared || has_local || has_temp)
+			if (has_shared || has_local || has_temp || has_timing)
 			{
 				appendStringInfoSpaces(es->str, es->indent * 2);
 				appendStringInfoString(es->str, "Buffers:");
@@ -1289,6 +1291,18 @@ ExplainNode(PlanState *planstate, List *ancestors,
 						appendStringInfo(es->str, " written=%ld",
 										 usage->temp_blks_written);
 				}
+				if (has_timing)
+				{
+					appendStringInfoString(es->str, " timing");
+					if (!INSTR_TIME_IS_ZERO(usage->time_read)) {
+						appendStringInfo(es->str, " read=%0.2f",
+										 INSTR_TIME_GET_MILLISEC(usage->time_read));
+					}
+					if (!INSTR_TIME_IS_ZERO(usage->time_write)) {
+						appendStringInfo(es->str, " write=%0.2f",
+										 INSTR_TIME_GET_MILLISEC(usage->time_write));
+					}
+				}
 				appendStringInfoChar(es->str, '\n');
 			}
 		}
@@ -1304,6 +1318,8 @@ ExplainNode(PlanState *planstate, List *ancestors,
 			ExplainPropertyLong("Local Written Blocks", usage->local_blks_written, es);
 			ExplainPropertyLong("Temp Read Blocks", usage->temp_blks_read, es);
 			ExplainPropertyLong("Temp Written Blocks", usage->temp_blks_written, es);
+			ExplainPropertyFloat("Read Waits", INSTR_TIME_GET_MILLISEC(usage->time_read), 3, es);
+			ExplainPropertyFloat("Write Waits", INSTR_TIME_GET_MILLISEC(usage->time_write), 3, es);
 		}
 	}
 
diff --git a/src/backend/executor/instrument.c b/src/backend/executor/instrument.c
index 6e9f450..92e56d1 100644
--- a/src/backend/executor/instrument.c
+++ b/src/backend/executor/instrument.c
@@ -145,4 +145,6 @@ BufferUsageAccumDiff(BufferUsage *dst,
 	dst->local_blks_written += add->local_blks_written - sub->local_blks_written;
 	dst->temp_blks_read += add->temp_blks_read - sub->temp_blks_read;
 	dst->temp_blks_written += add->temp_blks_written - sub->temp_blks_written;
+	INSTR_TIME_ACCUM_DIFF(dst->time_read, add->time_read, sub->time_read);
+	INSTR_TIME_ACCUM_DIFF(dst->time_write, add->time_write, sub->time_write);
 }
diff --git a/src/backend/postmaster/pgstat.c b/src/backend/postmaster/pgstat.c
index a53fc52..e9f6a3b 100644
--- a/src/backend/postmaster/pgstat.c
+++ b/src/backend/postmaster/pgstat.c
@@ -3338,6 +3338,7 @@ pgstat_get_db_entry(Oid databaseid, bool create)
 		result->n_xact_rollback = 0;
 		result->n_blocks_fetched = 0;
 		result->n_blocks_hit = 0;
+		INSTR_TIME_SET_ZERO(result->blocks_time);
 		result->n_tuples_returned = 0;
 		result->n_tuples_fetched = 0;
 		result->n_tuples_inserted = 0;
@@ -3412,6 +3413,7 @@ pgstat_get_tab_entry(PgStat_StatDBEntry *dbentry, Oid tableoid, bool create)
 		result->changes_since_analyze = 0;
 		result->blocks_fetched = 0;
 		result->blocks_hit = 0;
+		INSTR_TIME_SET_ZERO(result->blocks_time);
 		result->vacuum_timestamp = 0;
 		result->vacuum_count = 0;
 		result->autovac_vacuum_timestamp = 0;
@@ -4100,6 +4102,7 @@ pgstat_recv_tabstat(PgStat_MsgTabstat *msg, int len)
 			tabentry->changes_since_analyze = tabmsg->t_counts.t_changed_tuples;
 			tabentry->blocks_fetched = tabmsg->t_counts.t_blocks_fetched;
 			tabentry->blocks_hit = tabmsg->t_counts.t_blocks_hit;
+			tabentry->blocks_time = tabmsg->t_counts.t_blocks_time;
 
 			tabentry->vacuum_timestamp = 0;
 			tabentry->vacuum_count = 0;
@@ -4127,6 +4130,7 @@ pgstat_recv_tabstat(PgStat_MsgTabstat *msg, int len)
 			tabentry->changes_since_analyze += tabmsg->t_counts.t_changed_tuples;
 			tabentry->blocks_fetched += tabmsg->t_counts.t_blocks_fetched;
 			tabentry->blocks_hit += tabmsg->t_counts.t_blocks_hit;
+			INSTR_TIME_ADD(tabentry->blocks_time, tabmsg->t_counts.t_blocks_time);
 		}
 
 		/* Clamp n_live_tuples in case of negative delta_live_tuples */
@@ -4144,6 +4148,7 @@ pgstat_recv_tabstat(PgStat_MsgTabstat *msg, int len)
 		dbentry->n_tuples_deleted += tabmsg->t_counts.t_tuples_deleted;
 		dbentry->n_blocks_fetched += tabmsg->t_counts.t_blocks_fetched;
 		dbentry->n_blocks_hit += tabmsg->t_counts.t_blocks_hit;
+		INSTR_TIME_ADD(dbentry->blocks_time, tabmsg->t_counts.t_blocks_time);
 	}
 }
 
@@ -4257,6 +4262,7 @@ pgstat_recv_resetcounter(PgStat_MsgResetcounter *msg, int len)
 	dbentry->n_xact_rollback = 0;
 	dbentry->n_blocks_fetched = 0;
 	dbentry->n_blocks_hit = 0;
+	INSTR_TIME_SET_ZERO(dbentry->blocks_time);
 	dbentry->n_tuples_returned = 0;
 	dbentry->n_tuples_fetched = 0;
 	dbentry->n_tuples_inserted = 0;
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index 3924a51..a28d8e3 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -67,6 +67,7 @@
 bool		zero_damaged_pages = false;
 int			bgwriter_lru_maxpages = 100;
 double		bgwriter_lru_multiplier = 2.0;
+bool		track_iotiming = false;
 
 /*
  * How many buffers PrefetchBuffer callers should try to stay ahead of their
@@ -86,7 +87,7 @@ static volatile BufferDesc *PinCountWaitBuf = NULL;
 static Buffer ReadBuffer_common(SMgrRelation reln, char relpersistence,
 				  ForkNumber forkNum, BlockNumber blockNum,
 				  ReadBufferMode mode, BufferAccessStrategy strategy,
-				  bool *hit);
+				  bool *hit, instr_time *io_time);
 static bool PinBuffer(volatile BufferDesc *buf, BufferAccessStrategy strategy);
 static void PinBuffer_Locked(volatile BufferDesc *buf);
 static void UnpinBuffer(volatile BufferDesc *buf, bool fixOwner);
@@ -224,6 +225,7 @@ ReadBufferExtended(Relation reln, ForkNumber forkNum, BlockNumber blockNum,
 				   ReadBufferMode mode, BufferAccessStrategy strategy)
 {
 	bool		hit;
+	instr_time	io_time;
 	Buffer		buf;
 
 	/* Open it at the smgr level if not already done */
@@ -245,9 +247,11 @@ ReadBufferExtended(Relation reln, ForkNumber forkNum, BlockNumber blockNum,
 	 */
 	pgstat_count_buffer_read(reln);
 	buf = ReadBuffer_common(reln->rd_smgr, reln->rd_rel->relpersistence,
-							forkNum, blockNum, mode, strategy, &hit);
+							forkNum, blockNum, mode, strategy, &hit, &io_time);
 	if (hit)
 		pgstat_count_buffer_hit(reln);
+	else
+		pgstat_count_buffer_time(reln, io_time);
 	return buf;
 }
 
@@ -267,11 +271,12 @@ ReadBufferWithoutRelcache(RelFileNode rnode, ForkNumber forkNum,
 						  BufferAccessStrategy strategy)
 {
 	bool		hit;
+	instr_time	io_time;
 
 	SMgrRelation smgr = smgropen(rnode, InvalidBackendId);
 
 	return ReadBuffer_common(smgr, RELPERSISTENCE_PERMANENT, forkNum, blockNum,
-							 mode, strategy, &hit);
+							 mode, strategy, &hit, &io_time);
 }
 
 
@@ -279,19 +284,22 @@ ReadBufferWithoutRelcache(RelFileNode rnode, ForkNumber forkNum,
  * ReadBuffer_common -- common logic for all ReadBuffer variants
  *
  * *hit is set to true if the request was satisfied from shared buffer cache.
+ * If track_iotiming is enabled, *io_time is set to the time the read took.
  */
 static Buffer
 ReadBuffer_common(SMgrRelation smgr, char relpersistence, ForkNumber forkNum,
 				  BlockNumber blockNum, ReadBufferMode mode,
-				  BufferAccessStrategy strategy, bool *hit)
+				  BufferAccessStrategy strategy, bool *hit, instr_time *io_time)
 {
 	volatile BufferDesc *bufHdr;
 	Block		bufBlock;
 	bool		found;
 	bool		isExtend;
 	bool		isLocalBuf = SmgrIsTemp(smgr);
+	instr_time io_start, io_end;
 
 	*hit = false;
+	INSTR_TIME_SET_ZERO(*io_time);
 
 	/* Make sure we will have room to remember the buffer pin */
 	ResourceOwnerEnlargeBuffers(CurrentResourceOwner);
@@ -437,8 +445,18 @@ ReadBuffer_common(SMgrRelation smgr, char relpersistence, ForkNumber forkNum,
 			MemSet((char *) bufBlock, 0, BLCKSZ);
 		else
 		{
+			if (track_iotiming)
+				INSTR_TIME_SET_CURRENT(io_start);
+
 			smgrread(smgr, forkNum, blockNum, (char *) bufBlock);
 
+			if (track_iotiming)
+			{
+				INSTR_TIME_SET_CURRENT(io_end);
+				INSTR_TIME_ACCUM_DIFF(*io_time, io_end, io_start);
+				INSTR_TIME_ADD(pgBufferUsage.time_read, *io_time);
+			}
+
 			/* check for garbage data */
 			if (!PageHeaderIsValid((PageHeader) bufBlock))
 			{
@@ -1874,6 +1892,7 @@ FlushBuffer(volatile BufferDesc *buf, SMgrRelation reln)
 {
 	XLogRecPtr	recptr;
 	ErrorContextCallback errcontext;
+	instr_time io_start, io_end;
 
 	/*
 	 * Acquire the buffer's io_in_progress lock.  If StartBufferIO returns
@@ -1921,12 +1940,21 @@ FlushBuffer(volatile BufferDesc *buf, SMgrRelation reln)
 	buf->flags &= ~BM_JUST_DIRTIED;
 	UnlockBufHdr(buf);
 
+	if (track_iotiming)
+		INSTR_TIME_SET_CURRENT(io_start);
+
 	smgrwrite(reln,
 			  buf->tag.forkNum,
 			  buf->tag.blockNum,
 			  (char *) BufHdrGetBlock(buf),
 			  false);
 
+	if (track_iotiming)
+	{
+		INSTR_TIME_SET_CURRENT(io_end);
+		INSTR_TIME_ACCUM_DIFF(pgBufferUsage.time_write, io_end, io_start);
+	}
+
 	pgBufferUsage.shared_blks_written++;
 
 	/*
@@ -2670,6 +2698,7 @@ WaitIO(volatile BufferDesc *buf)
 static bool
 StartBufferIO(volatile BufferDesc *buf, bool forInput)
 {
+	instr_time wait_start, wait_end;
 	Assert(!InProgressBuf);
 
 	for (;;)
@@ -2678,7 +2707,28 @@ StartBufferIO(volatile BufferDesc *buf, bool forInput)
 		 * Grab the io_in_progress lock so that other processes can wait for
 		 * me to finish the I/O.
 		 */
-		LWLockAcquire(buf->io_in_progress_lock, LW_EXCLUSIVE);
+		if (forInput && track_iotiming) {
+			/*
+			 * We need to time the lock wait to account for I/O waits where
+			 * someone else is doing the work for us. Conditional acquire
+			 * avoids double timing overhead when we do the I/O ourselves.
+			 */
+			if (!LWLockConditionalAcquire(buf->io_in_progress_lock, LW_EXCLUSIVE)) {
+				INSTR_TIME_SET_CURRENT(wait_start);
+
+				LWLockAcquire(buf->io_in_progress_lock, LW_EXCLUSIVE);
+
+				/*
+				 * Only do backend local accounting, stats collector will get the
+				 * wait from the backend doing the I/O.
+				 */
+				INSTR_TIME_SET_CURRENT(wait_end);
+				INSTR_TIME_ACCUM_DIFF(pgBufferUsage.time_read, wait_end, wait_start);
+			}
+		} else {
+			LWLockAcquire(buf->io_in_progress_lock, LW_EXCLUSIVE);
+		}
+
 
 		LockBufHdr(buf);
 
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 68b2527..d0b6c57 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -22,6 +22,7 @@
 #include "utils/builtins.h"
 #include "utils/inet.h"
 #include "utils/timestamp.h"
+#include "portability/instr_time.h"
 
 /* bogus ... these externs should be in a header file */
 extern Datum pg_stat_get_numscans(PG_FUNCTION_ARGS);
@@ -35,6 +36,7 @@ extern Datum pg_stat_get_live_tuples(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_dead_tuples(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_blocks_fetched(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_blocks_hit(PG_FUNCTION_ARGS);
+extern Datum pg_stat_get_blocks_time(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_last_vacuum_time(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_last_autovacuum_time(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_last_analyze_time(PG_FUNCTION_ARGS);
@@ -67,6 +69,7 @@ extern Datum pg_stat_get_db_xact_commit(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_db_xact_rollback(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_db_blocks_fetched(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_db_blocks_hit(PG_FUNCTION_ARGS);
+extern Datum pg_stat_get_db_blocks_time(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_db_tuples_returned(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_db_tuples_fetched(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_db_tuples_inserted(PG_FUNCTION_ARGS);
@@ -102,6 +105,7 @@ extern Datum pg_stat_get_xact_tuples_deleted(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_xact_tuples_hot_updated(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_xact_blocks_fetched(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_xact_blocks_hit(PG_FUNCTION_ARGS);
+extern Datum pg_stat_get_xact_blocks_time(PG_FUNCTION_ARGS);
 
 extern Datum pg_stat_get_xact_function_calls(PG_FUNCTION_ARGS);
 extern Datum pg_stat_get_xact_function_time(PG_FUNCTION_ARGS);
@@ -292,6 +296,22 @@ pg_stat_get_blocks_hit(PG_FUNCTION_ARGS)
 }
 
 Datum
+pg_stat_get_blocks_time(PG_FUNCTION_ARGS)
+{
+	Oid			relid = PG_GETARG_OID(0);
+	double		result;
+	PgStat_StatTabEntry *tabentry;
+
+	if ((tabentry = pgstat_fetch_stat_tabentry(relid)) == NULL)
+		result = 0;
+	else
+		/* Cast overflows in about 300'000 years of io time */
+		result = (int64) INSTR_TIME_GET_MICROSEC(tabentry->blocks_time);
+
+	PG_RETURN_INT64(result);
+}
+
+Datum
 pg_stat_get_last_vacuum_time(PG_FUNCTION_ARGS)
 {
 	Oid			relid = PG_GETARG_OID(0);
@@ -1120,6 +1140,22 @@ pg_stat_get_db_blocks_hit(PG_FUNCTION_ARGS)
 
 
 Datum
+pg_stat_get_db_blocks_time(PG_FUNCTION_ARGS)
+{
+	Oid			dbid = PG_GETARG_OID(0);
+	int64		result;
+	PgStat_StatDBEntry *dbentry;
+
+	if ((dbentry = pgstat_fetch_stat_dbentry(dbid)) == NULL)
+		result = 0;
+	else
+		result = (int64) INSTR_TIME_GET_MICROSEC(dbentry->blocks_time);
+
+	PG_RETURN_INT64(result);
+}
+
+
+Datum
 pg_stat_get_db_tuples_returned(PG_FUNCTION_ARGS)
 {
 	Oid			dbid = PG_GETARG_OID(0);
@@ -1565,6 +1601,21 @@ pg_stat_get_xact_blocks_hit(PG_FUNCTION_ARGS)
 }
 
 Datum
+pg_stat_get_xact_blocks_time(PG_FUNCTION_ARGS)
+{
+	Oid			relid = PG_GETARG_OID(0);
+	int64		result;
+	PgStat_TableStatus *tabentry;
+
+	if ((tabentry = find_tabstat_entry(relid)) == NULL)
+		result = 0;
+	else
+		result = (int64) INSTR_TIME_GET_MICROSEC(tabentry->t_counts.t_blocks_time);
+
+	PG_RETURN_INT64(result);
+}
+
+Datum
 pg_stat_get_xact_function_calls(PG_FUNCTION_ARGS)
 {
 	Oid			funcid = PG_GETARG_OID(0);
diff --git a/src/backend/utils/misc/guc.c b/src/backend/utils/misc/guc.c
index 3d2fe3e..8ea391a 100644
--- a/src/backend/utils/misc/guc.c
+++ b/src/backend/utils/misc/guc.c
@@ -1018,6 +1018,15 @@ static struct config_bool ConfigureNamesBool[] =
 		true,
 		NULL, NULL, NULL
 	},
+	{
+		{"track_iotiming", PGC_SUSET, STATS_COLLECTOR,
+			gettext_noop("Collects timing information for database IO activity."),
+			NULL
+		},
+		&track_iotiming,
+		false,
+		NULL, NULL, NULL
+	},
 
 	{
 		{"update_process_title", PGC_SUSET, STATS_COLLECTOR,
diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample
index 96da086..cbae3bf 100644
--- a/src/backend/utils/misc/postgresql.conf.sample
+++ b/src/backend/utils/misc/postgresql.conf.sample
@@ -424,6 +424,7 @@
 
 #track_activities = on
 #track_counts = on
+#track_iotiming = off
 #track_functions = none			# none, pl, all
 #track_activity_query_size = 1024 	# (change requires restart)
 #update_process_title = on
diff --git a/src/include/catalog/pg_proc.h b/src/include/catalog/pg_proc.h
index 2db8489..c4ac58f 100644
--- a/src/include/catalog/pg_proc.h
+++ b/src/include/catalog/pg_proc.h
@@ -2680,6 +2680,12 @@ DATA(insert OID = 3063 ( pg_stat_get_buf_fsync_backend PGNSP PGUID 12 1 0 0 0 f
 DESCR("statistics: number of backend buffer writes that did their own fsync");
 DATA(insert OID = 2859 ( pg_stat_get_buf_alloc			PGNSP PGUID 12 1 0 0 0 f f f f t f s 0 0 20 "" _null_ _null_ _null_ _null_ pg_stat_get_buf_alloc _null_ _null_ _null_ ));
 DESCR("statistics: number of buffer allocations");
+DATA(insert OID = 3160 (  pg_stat_get_blocks_time		PGNSP PGUID 12 1 0 0 0 f f f f t f s 1 0 20 "26" _null_ _null_ _null_ _null_ pg_stat_get_blocks_time _null_ _null_ _null_ ));
+DESCR("statistics: duration of block reads");
+DATA(insert OID = 3161 (  pg_stat_get_db_blocks_time	PGNSP PGUID 12 1 0 0 0 f f f f t f s 1 0 20 "26" _null_ _null_ _null_ _null_ pg_stat_get_db_blocks_time _null_ _null_ _null_ ));
+DESCR("statistics: duration of block reads for database");
+DATA(insert OID = 3166 (  pg_stat_get_xact_blocks_time	PGNSP PGUID 12 1 0 0 0 f f f f t f v 1 0 20 "26" _null_ _null_ _null_ _null_ pg_stat_get_xact_blocks_time _null_ _null_ _null_ ));
+DESCR("statistics: duration of block reads in current transaction");
 
 DATA(insert OID = 2978 (  pg_stat_get_function_calls		PGNSP PGUID 12 1 0 0 0 f f f f t f s 1 0 20 "26" _null_ _null_ _null_ _null_ pg_stat_get_function_calls _null_ _null_ _null_ ));
 DESCR("statistics: number of function calls");
diff --git a/src/include/executor/instrument.h b/src/include/executor/instrument.h
index 066f684..ad13914 100644
--- a/src/include/executor/instrument.h
+++ b/src/include/executor/instrument.h
@@ -28,6 +28,8 @@ typedef struct BufferUsage
 	long		local_blks_written;		/* # of local disk blocks written */
 	long		temp_blks_read; /* # of temp blocks read */
 	long		temp_blks_written;		/* # of temp blocks written */
+	instr_time	time_read;	/* time spent reading */
+	instr_time	time_write;	/* time spent writing */
 } BufferUsage;
 
 /* Flag bits included in InstrAlloc's instrument_options bitmask */
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 1281bd8..175b5df 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -97,6 +97,7 @@ typedef struct PgStat_TableCounts
 
 	PgStat_Counter t_blocks_fetched;
 	PgStat_Counter t_blocks_hit;
+	instr_time     t_blocks_time;
 } PgStat_TableCounts;
 
 /* Possible targets for resetting cluster-wide shared values */
@@ -522,6 +523,7 @@ typedef struct PgStat_StatDBEntry
 	PgStat_Counter n_xact_rollback;
 	PgStat_Counter n_blocks_fetched;
 	PgStat_Counter n_blocks_hit;
+	instr_time	blocks_time;
 	PgStat_Counter n_tuples_returned;
 	PgStat_Counter n_tuples_fetched;
 	PgStat_Counter n_tuples_inserted;
@@ -573,6 +575,7 @@ typedef struct PgStat_StatTabEntry
 
 	PgStat_Counter blocks_fetched;
 	PgStat_Counter blocks_hit;
+	instr_time	blocks_time;
 
 	TimestampTz vacuum_timestamp;		/* user initiated vacuum */
 	PgStat_Counter vacuum_count;
@@ -816,6 +819,13 @@ extern void pgstat_initstats(Relation rel);
 		if ((rel)->pgstat_info != NULL)								\
 			(rel)->pgstat_info->t_counts.t_blocks_hit++;			\
 	} while (0)
+#define pgstat_count_buffer_time(rel, io)							\
+	do {															\
+		if ((rel)->pgstat_info != NULL)								\
+			INSTR_TIME_ADD(											\
+				(rel)->pgstat_info->t_counts.t_blocks_time,			\
+				(io));												\
+	} while (0)
 
 extern void pgstat_count_heap_insert(Relation rel, int n);
 extern void pgstat_count_heap_update(Relation rel, bool hot);
diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index de1bbd0..d72bad9 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -48,6 +48,7 @@ extern PGDLLIMPORT int NBuffers;
 extern bool zero_damaged_pages;
 extern int	bgwriter_lru_maxpages;
 extern double bgwriter_lru_multiplier;
+extern bool track_iotiming;
 extern int	target_prefetch_pages;
 
 /* in buf_init.c */
diff --git a/src/test/regress/expected/rules.out b/src/test/regress/expected/rules.out
index aaf0cca..ce6228c 100644
--- a/src/test/regress/expected/rules.out
+++ b/src/test/regress/expected/rules.out
@@ -1296,7 +1296,7 @@ SELECT viewname, definition FROM pg_views WHERE schemaname <> 'information_schem
  pg_stat_all_indexes             | SELECT c.oid AS relid, i.oid AS indexrelid, n.nspname AS schemaname, c.relname, i.relname AS indexrelname, pg_stat_get_numscans(i.oid) AS idx_scan, pg_stat_get_tuples_returned(i.oid) AS idx_tup_read, pg_stat_get_tuples_fetched(i.oid) AS idx_tup_fetch FROM (((pg_class c JOIN pg_index x ON ((c.oid = x.indrelid))) JOIN pg_class i ON ((i.oid = x.indexrelid))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = ANY (ARRAY['r'::"char", 't'::"char"]));
  pg_stat_all_tables              | SELECT c.oid AS relid, n.nspname AS schemaname, c.relname, pg_stat_get_numscans(c.oid) AS seq_scan, pg_stat_get_tuples_returned(c.oid) AS seq_tup_read, (sum(pg_stat_get_numscans(i.indexrelid)))::bigint AS idx_scan, ((sum(pg_stat_get_tuples_fetched(i.indexrelid)))::bigint + pg_stat_get_tuples_fetched(c.oid)) AS idx_tup_fetch, pg_stat_get_tuples_inserted(c.oid) AS n_tup_ins, pg_stat_get_tuples_updated(c.oid) AS n_tup_upd, pg_stat_get_tuples_deleted(c.oid) AS n_tup_del, pg_stat_get_tuples_hot_updated(c.oid) AS n_tup_hot_upd, pg_stat_get_live_tuples(c.oid) AS n_live_tup, pg_stat_get_dead_tuples(c.oid) AS n_dead_tup, pg_stat_get_last_vacuum_time(c.oid) AS last_vacuum, pg_stat_get_last_autovacuum_time(c.oid) AS last_autovacuum, pg_stat_get_last_analyze_time(c.oid) AS last_analyze, pg_stat_get_last_autoanalyze_time(c.oid) AS last_autoanalyze, pg_stat_get_vacuum_count(c.oid) AS vacuum_count, pg_stat_get_autovacuum_count(c.oid) AS autovacuum_count, pg_stat_get_analyze_count(c.oid) AS analyze_count, pg_stat_get_autoanalyze_count(c.oid) AS autoanalyze_count FROM ((pg_class c LEFT JOIN pg_index i ON ((c.oid = i.indrelid))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = ANY (ARRAY['r'::"char", 't'::"char"])) GROUP BY c.oid, n.nspname, c.relname;
  pg_stat_bgwriter                | SELECT pg_stat_get_bgwriter_timed_checkpoints() AS checkpoints_timed, pg_stat_get_bgwriter_requested_checkpoints() AS checkpoints_req, pg_stat_get_bgwriter_buf_written_checkpoints() AS buffers_checkpoint, pg_stat_get_bgwriter_buf_written_clean() AS buffers_clean, pg_stat_get_bgwriter_maxwritten_clean() AS maxwritten_clean, pg_stat_get_buf_written_backend() AS buffers_backend, pg_stat_get_buf_fsync_backend() AS buffers_backend_fsync, pg_stat_get_buf_alloc() AS buffers_alloc, pg_stat_get_bgwriter_stat_reset_time() AS stats_reset;
- pg_stat_database                | SELECT d.oid AS datid, d.datname, pg_stat_get_db_numbackends(d.oid) AS numbackends, pg_stat_get_db_xact_commit(d.oid) AS xact_commit, pg_stat_get_db_xact_rollback(d.oid) AS xact_rollback, (pg_stat_get_db_blocks_fetched(d.oid) - pg_stat_get_db_blocks_hit(d.oid)) AS blks_read, pg_stat_get_db_blocks_hit(d.oid) AS blks_hit, pg_stat_get_db_tuples_returned(d.oid) AS tup_returned, pg_stat_get_db_tuples_fetched(d.oid) AS tup_fetched, pg_stat_get_db_tuples_inserted(d.oid) AS tup_inserted, pg_stat_get_db_tuples_updated(d.oid) AS tup_updated, pg_stat_get_db_tuples_deleted(d.oid) AS tup_deleted, pg_stat_get_db_conflict_all(d.oid) AS conflicts, pg_stat_get_db_temp_files(d.oid) AS temp_files, pg_stat_get_db_temp_bytes(d.oid) AS temp_bytes, pg_stat_get_db_deadlocks(d.oid) AS deadlocks, pg_stat_get_db_stat_reset_time(d.oid) AS stats_reset FROM pg_database d;
+ pg_stat_database                | SELECT d.oid AS datid, d.datname, pg_stat_get_db_numbackends(d.oid) AS numbackends, pg_stat_get_db_xact_commit(d.oid) AS xact_commit, pg_stat_get_db_xact_rollback(d.oid) AS xact_rollback, (pg_stat_get_db_blocks_fetched(d.oid) - pg_stat_get_db_blocks_hit(d.oid)) AS blks_read, pg_stat_get_db_blocks_hit(d.oid) AS blks_hit, pg_stat_get_db_blocks_time(d.oid) AS blks_time, pg_stat_get_db_tuples_returned(d.oid) AS tup_returned, pg_stat_get_db_tuples_fetched(d.oid) AS tup_fetched, pg_stat_get_db_tuples_inserted(d.oid) AS tup_inserted, pg_stat_get_db_tuples_updated(d.oid) AS tup_updated, pg_stat_get_db_tuples_deleted(d.oid) AS tup_deleted, pg_stat_get_db_conflict_all(d.oid) AS conflicts, pg_stat_get_db_temp_files(d.oid) AS temp_files, pg_stat_get_db_temp_bytes(d.oid) AS temp_bytes, pg_stat_get_db_deadlocks(d.oid) AS deadlocks, pg_stat_get_db_stat_reset_time(d.oid) AS stats_reset FROM pg_database d;
  pg_stat_database_conflicts      | SELECT d.oid AS datid, d.datname, pg_stat_get_db_conflict_tablespace(d.oid) AS confl_tablespace, pg_stat_get_db_conflict_lock(d.oid) AS confl_lock, pg_stat_get_db_conflict_snapshot(d.oid) AS confl_snapshot, pg_stat_get_db_conflict_bufferpin(d.oid) AS confl_bufferpin, pg_stat_get_db_conflict_startup_deadlock(d.oid) AS confl_deadlock FROM pg_database d;
  pg_stat_replication             | SELECT s.pid, s.usesysid, u.rolname AS usename, s.application_name, s.client_addr, s.client_hostname, s.client_port, s.backend_start, w.state, w.sent_location, w.write_location, w.flush_location, w.replay_location, w.sync_priority, w.sync_state FROM pg_stat_get_activity(NULL::integer) s(datid, pid, usesysid, application_name, state, query, waiting, xact_start, query_start, backend_start, state_change, client_addr, client_hostname, client_port), pg_authid u, pg_stat_get_wal_senders() w(pid, state, sent_location, write_location, flush_location, replay_location, sync_priority, sync_state) WHERE ((s.usesysid = u.oid) AND (s.pid = w.pid));
  pg_stat_sys_indexes             | SELECT pg_stat_all_indexes.relid, pg_stat_all_indexes.indexrelid, pg_stat_all_indexes.schemaname, pg_stat_all_indexes.relname, pg_stat_all_indexes.indexrelname, pg_stat_all_indexes.idx_scan, pg_stat_all_indexes.idx_tup_read, pg_stat_all_indexes.idx_tup_fetch FROM pg_stat_all_indexes WHERE ((pg_stat_all_indexes.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_stat_all_indexes.schemaname ~ '^pg_toast'::text));
@@ -1308,15 +1308,15 @@ SELECT viewname, definition FROM pg_views WHERE schemaname <> 'information_schem
  pg_stat_xact_sys_tables         | SELECT pg_stat_xact_all_tables.relid, pg_stat_xact_all_tables.schemaname, pg_stat_xact_all_tables.relname, pg_stat_xact_all_tables.seq_scan, pg_stat_xact_all_tables.seq_tup_read, pg_stat_xact_all_tables.idx_scan, pg_stat_xact_all_tables.idx_tup_fetch, pg_stat_xact_all_tables.n_tup_ins, pg_stat_xact_all_tables.n_tup_upd, pg_stat_xact_all_tables.n_tup_del, pg_stat_xact_all_tables.n_tup_hot_upd FROM pg_stat_xact_all_tables WHERE ((pg_stat_xact_all_tables.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_stat_xact_all_tables.schemaname ~ '^pg_toast'::text));
  pg_stat_xact_user_functions     | SELECT p.oid AS funcid, n.nspname AS schemaname, p.proname AS funcname, pg_stat_get_xact_function_calls(p.oid) AS calls, (pg_stat_get_xact_function_time(p.oid) / 1000) AS total_time, (pg_stat_get_xact_function_self_time(p.oid) / 1000) AS self_time FROM (pg_proc p LEFT JOIN pg_namespace n ON ((n.oid = p.pronamespace))) WHERE ((p.prolang <> (12)::oid) AND (pg_stat_get_xact_function_calls(p.oid) IS NOT NULL));
  pg_stat_xact_user_tables        | SELECT pg_stat_xact_all_tables.relid, pg_stat_xact_all_tables.schemaname, pg_stat_xact_all_tables.relname, pg_stat_xact_all_tables.seq_scan, pg_stat_xact_all_tables.seq_tup_read, pg_stat_xact_all_tables.idx_scan, pg_stat_xact_all_tables.idx_tup_fetch, pg_stat_xact_all_tables.n_tup_ins, pg_stat_xact_all_tables.n_tup_upd, pg_stat_xact_all_tables.n_tup_del, pg_stat_xact_all_tables.n_tup_hot_upd FROM pg_stat_xact_all_tables WHERE ((pg_stat_xact_all_tables.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_stat_xact_all_tables.schemaname !~ '^pg_toast'::text));
- pg_statio_all_indexes           | SELECT c.oid AS relid, i.oid AS indexrelid, n.nspname AS schemaname, c.relname, i.relname AS indexrelname, (pg_stat_get_blocks_fetched(i.oid) - pg_stat_get_blocks_hit(i.oid)) AS idx_blks_read, pg_stat_get_blocks_hit(i.oid) AS idx_blks_hit FROM (((pg_class c JOIN pg_index x ON ((c.oid = x.indrelid))) JOIN pg_class i ON ((i.oid = x.indexrelid))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = ANY (ARRAY['r'::"char", 't'::"char"]));
- pg_statio_all_sequences         | SELECT c.oid AS relid, n.nspname AS schemaname, c.relname, (pg_stat_get_blocks_fetched(c.oid) - pg_stat_get_blocks_hit(c.oid)) AS blks_read, pg_stat_get_blocks_hit(c.oid) AS blks_hit FROM (pg_class c LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = 'S'::"char");
- pg_statio_all_tables            | SELECT c.oid AS relid, n.nspname AS schemaname, c.relname, (pg_stat_get_blocks_fetched(c.oid) - pg_stat_get_blocks_hit(c.oid)) AS heap_blks_read, pg_stat_get_blocks_hit(c.oid) AS heap_blks_hit, (sum((pg_stat_get_blocks_fetched(i.indexrelid) - pg_stat_get_blocks_hit(i.indexrelid))))::bigint AS idx_blks_read, (sum(pg_stat_get_blocks_hit(i.indexrelid)))::bigint AS idx_blks_hit, (pg_stat_get_blocks_fetched(t.oid) - pg_stat_get_blocks_hit(t.oid)) AS toast_blks_read, pg_stat_get_blocks_hit(t.oid) AS toast_blks_hit, (pg_stat_get_blocks_fetched(x.oid) - pg_stat_get_blocks_hit(x.oid)) AS tidx_blks_read, pg_stat_get_blocks_hit(x.oid) AS tidx_blks_hit FROM ((((pg_class c LEFT JOIN pg_index i ON ((c.oid = i.indrelid))) LEFT JOIN pg_class t ON ((c.reltoastrelid = t.oid))) LEFT JOIN pg_class x ON ((t.reltoastidxid = x.oid))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = ANY (ARRAY['r'::"char", 't'::"char"])) GROUP BY c.oid, n.nspname, c.relname, t.oid, x.oid;
- pg_statio_sys_indexes           | SELECT pg_statio_all_indexes.relid, pg_statio_all_indexes.indexrelid, pg_statio_all_indexes.schemaname, pg_statio_all_indexes.relname, pg_statio_all_indexes.indexrelname, pg_statio_all_indexes.idx_blks_read, pg_statio_all_indexes.idx_blks_hit FROM pg_statio_all_indexes WHERE ((pg_statio_all_indexes.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_statio_all_indexes.schemaname ~ '^pg_toast'::text));
- pg_statio_sys_sequences         | SELECT pg_statio_all_sequences.relid, pg_statio_all_sequences.schemaname, pg_statio_all_sequences.relname, pg_statio_all_sequences.blks_read, pg_statio_all_sequences.blks_hit FROM pg_statio_all_sequences WHERE ((pg_statio_all_sequences.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_statio_all_sequences.schemaname ~ '^pg_toast'::text));
- pg_statio_sys_tables            | SELECT pg_statio_all_tables.relid, pg_statio_all_tables.schemaname, pg_statio_all_tables.relname, pg_statio_all_tables.heap_blks_read, pg_statio_all_tables.heap_blks_hit, pg_statio_all_tables.idx_blks_read, pg_statio_all_tables.idx_blks_hit, pg_statio_all_tables.toast_blks_read, pg_statio_all_tables.toast_blks_hit, pg_statio_all_tables.tidx_blks_read, pg_statio_all_tables.tidx_blks_hit FROM pg_statio_all_tables WHERE ((pg_statio_all_tables.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_statio_all_tables.schemaname ~ '^pg_toast'::text));
- pg_statio_user_indexes          | SELECT pg_statio_all_indexes.relid, pg_statio_all_indexes.indexrelid, pg_statio_all_indexes.schemaname, pg_statio_all_indexes.relname, pg_statio_all_indexes.indexrelname, pg_statio_all_indexes.idx_blks_read, pg_statio_all_indexes.idx_blks_hit FROM pg_statio_all_indexes WHERE ((pg_statio_all_indexes.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_statio_all_indexes.schemaname !~ '^pg_toast'::text));
- pg_statio_user_sequences        | SELECT pg_statio_all_sequences.relid, pg_statio_all_sequences.schemaname, pg_statio_all_sequences.relname, pg_statio_all_sequences.blks_read, pg_statio_all_sequences.blks_hit FROM pg_statio_all_sequences WHERE ((pg_statio_all_sequences.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_statio_all_sequences.schemaname !~ '^pg_toast'::text));
- pg_statio_user_tables           | SELECT pg_statio_all_tables.relid, pg_statio_all_tables.schemaname, pg_statio_all_tables.relname, pg_statio_all_tables.heap_blks_read, pg_statio_all_tables.heap_blks_hit, pg_statio_all_tables.idx_blks_read, pg_statio_all_tables.idx_blks_hit, pg_statio_all_tables.toast_blks_read, pg_statio_all_tables.toast_blks_hit, pg_statio_all_tables.tidx_blks_read, pg_statio_all_tables.tidx_blks_hit FROM pg_statio_all_tables WHERE ((pg_statio_all_tables.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_statio_all_tables.schemaname !~ '^pg_toast'::text));
+ pg_statio_all_indexes           | SELECT c.oid AS relid, i.oid AS indexrelid, n.nspname AS schemaname, c.relname, i.relname AS indexrelname, (pg_stat_get_blocks_fetched(i.oid) - pg_stat_get_blocks_hit(i.oid)) AS idx_blks_read, pg_stat_get_blocks_hit(i.oid) AS idx_blks_hit, pg_stat_get_blocks_time(i.oid) AS idx_blks_time FROM (((pg_class c JOIN pg_index x ON ((c.oid = x.indrelid))) JOIN pg_class i ON ((i.oid = x.indexrelid))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = ANY (ARRAY['r'::"char", 't'::"char"]));
+ pg_statio_all_sequences         | SELECT c.oid AS relid, n.nspname AS schemaname, c.relname, (pg_stat_get_blocks_fetched(c.oid) - pg_stat_get_blocks_hit(c.oid)) AS blks_read, pg_stat_get_blocks_hit(c.oid) AS blks_hit, pg_stat_get_blocks_time(c.oid) AS blks_time FROM (pg_class c LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = 'S'::"char");
+ pg_statio_all_tables            | SELECT c.oid AS relid, n.nspname AS schemaname, c.relname, (pg_stat_get_blocks_fetched(c.oid) - pg_stat_get_blocks_hit(c.oid)) AS heap_blks_read, pg_stat_get_blocks_hit(c.oid) AS heap_blks_hit, pg_stat_get_blocks_time(c.oid) AS heap_blks_time, (sum((pg_stat_get_blocks_fetched(i.indexrelid) - pg_stat_get_blocks_hit(i.indexrelid))))::bigint AS idx_blks_read, (sum(pg_stat_get_blocks_hit(i.indexrelid)))::bigint AS idx_blks_hit, (sum(pg_stat_get_blocks_time(i.indexrelid)))::bigint AS idx_blks_time, (pg_stat_get_blocks_fetched(t.oid) - pg_stat_get_blocks_hit(t.oid)) AS toast_blks_read, pg_stat_get_blocks_hit(t.oid) AS toast_blks_hit, pg_stat_get_blocks_time(t.oid) AS toast_blks_time, (pg_stat_get_blocks_fetched(x.oid) - pg_stat_get_blocks_hit(x.oid)) AS tidx_blks_read, pg_stat_get_blocks_hit(x.oid) AS tidx_blks_hit, pg_stat_get_blocks_time(x.oid) AS tidx_blks_time FROM ((((pg_class c LEFT JOIN pg_index i ON ((c.oid = i.indrelid))) LEFT JOIN pg_class t ON ((c.reltoastrelid = t.oid))) LEFT JOIN pg_class x ON ((t.reltoastidxid = x.oid))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE (c.relkind = ANY (ARRAY['r'::"char", 't'::"char"])) GROUP BY c.oid, n.nspname, c.relname, t.oid, x.oid;
+ pg_statio_sys_indexes           | SELECT pg_statio_all_indexes.relid, pg_statio_all_indexes.indexrelid, pg_statio_all_indexes.schemaname, pg_statio_all_indexes.relname, pg_statio_all_indexes.indexrelname, pg_statio_all_indexes.idx_blks_read, pg_statio_all_indexes.idx_blks_hit, pg_statio_all_indexes.idx_blks_time FROM pg_statio_all_indexes WHERE ((pg_statio_all_indexes.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_statio_all_indexes.schemaname ~ '^pg_toast'::text));
+ pg_statio_sys_sequences         | SELECT pg_statio_all_sequences.relid, pg_statio_all_sequences.schemaname, pg_statio_all_sequences.relname, pg_statio_all_sequences.blks_read, pg_statio_all_sequences.blks_hit, pg_statio_all_sequences.blks_time FROM pg_statio_all_sequences WHERE ((pg_statio_all_sequences.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_statio_all_sequences.schemaname ~ '^pg_toast'::text));
+ pg_statio_sys_tables            | SELECT pg_statio_all_tables.relid, pg_statio_all_tables.schemaname, pg_statio_all_tables.relname, pg_statio_all_tables.heap_blks_read, pg_statio_all_tables.heap_blks_hit, pg_statio_all_tables.heap_blks_time, pg_statio_all_tables.idx_blks_read, pg_statio_all_tables.idx_blks_hit, pg_statio_all_tables.idx_blks_time, pg_statio_all_tables.toast_blks_read, pg_statio_all_tables.toast_blks_hit, pg_statio_all_tables.toast_blks_time, pg_statio_all_tables.tidx_blks_read, pg_statio_all_tables.tidx_blks_hit, pg_statio_all_tables.tidx_blks_time FROM pg_statio_all_tables WHERE ((pg_statio_all_tables.schemaname = ANY (ARRAY['pg_catalog'::name, 'information_schema'::name])) OR (pg_statio_all_tables.schemaname ~ '^pg_toast'::text));
+ pg_statio_user_indexes          | SELECT pg_statio_all_indexes.relid, pg_statio_all_indexes.indexrelid, pg_statio_all_indexes.schemaname, pg_statio_all_indexes.relname, pg_statio_all_indexes.indexrelname, pg_statio_all_indexes.idx_blks_read, pg_statio_all_indexes.idx_blks_hit, pg_statio_all_indexes.idx_blks_time FROM pg_statio_all_indexes WHERE ((pg_statio_all_indexes.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_statio_all_indexes.schemaname !~ '^pg_toast'::text));
+ pg_statio_user_sequences        | SELECT pg_statio_all_sequences.relid, pg_statio_all_sequences.schemaname, pg_statio_all_sequences.relname, pg_statio_all_sequences.blks_read, pg_statio_all_sequences.blks_hit, pg_statio_all_sequences.blks_time FROM pg_statio_all_sequences WHERE ((pg_statio_all_sequences.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_statio_all_sequences.schemaname !~ '^pg_toast'::text));
+ pg_statio_user_tables           | SELECT pg_statio_all_tables.relid, pg_statio_all_tables.schemaname, pg_statio_all_tables.relname, pg_statio_all_tables.heap_blks_read, pg_statio_all_tables.heap_blks_hit, pg_statio_all_tables.heap_blks_time, pg_statio_all_tables.idx_blks_read, pg_statio_all_tables.idx_blks_hit, pg_statio_all_tables.idx_blks_time, pg_statio_all_tables.toast_blks_read, pg_statio_all_tables.toast_blks_hit, pg_statio_all_tables.toast_blks_time, pg_statio_all_tables.tidx_blks_read, pg_statio_all_tables.tidx_blks_hit, pg_statio_all_tables.tidx_blks_time FROM pg_statio_all_tables WHERE ((pg_statio_all_tables.schemaname <> ALL (ARRAY['pg_catalog'::name, 'information_schema'::name])) AND (pg_statio_all_tables.schemaname !~ '^pg_toast'::text));
  pg_stats                        | SELECT n.nspname AS schemaname, c.relname AS tablename, a.attname, s.stainherit AS inherited, s.stanullfrac AS null_frac, s.stawidth AS avg_width, s.stadistinct AS n_distinct, CASE WHEN (s.stakind1 = 1) THEN s.stavalues1 WHEN (s.stakind2 = 1) THEN s.stavalues2 WHEN (s.stakind3 = 1) THEN s.stavalues3 WHEN (s.stakind4 = 1) THEN s.stavalues4 WHEN (s.stakind5 = 1) THEN s.stavalues5 ELSE NULL::anyarray END AS most_common_vals, CASE WHEN (s.stakind1 = 1) THEN s.stanumbers1 WHEN (s.stakind2 = 1) THEN s.stanumbers2 WHEN (s.stakind3 = 1) THEN s.stanumbers3 WHEN (s.stakind4 = 1) THEN s.stanumbers4 WHEN (s.stakind5 = 1) THEN s.stanumbers5 ELSE NULL::real[] END AS most_common_freqs, CASE WHEN (s.stakind1 = 2) THEN s.stavalues1 WHEN (s.stakind2 = 2) THEN s.stavalues2 WHEN (s.stakind3 = 2) THEN s.stavalues3 WHEN (s.stakind4 = 2) THEN s.stavalues4 WHEN (s.stakind5 = 2) THEN s.stavalues5 ELSE NULL::anyarray END AS histogram_bounds, CASE WHEN (s.stakind1 = 3) THEN s.stanumbers1[1] WHEN (s.stakind2 = 3) THEN s.stanumbers2[1] WHEN (s.stakind3 = 3) THEN s.stanumbers3[1] WHEN (s.stakind4 = 3) THEN s.stanumbers4[1] WHEN (s.stakind5 = 3) THEN s.stanumbers5[1] ELSE NULL::real END AS correlation, CASE WHEN (s.stakind1 = 4) THEN s.stavalues1 WHEN (s.stakind2 = 4) THEN s.stavalues2 WHEN (s.stakind3 = 4) THEN s.stavalues3 WHEN (s.stakind4 = 4) THEN s.stavalues4 WHEN (s.stakind5 = 4) THEN s.stavalues5 ELSE NULL::anyarray END AS most_common_elems, CASE WHEN (s.stakind1 = 4) THEN s.stanumbers1 WHEN (s.stakind2 = 4) THEN s.stanumbers2 WHEN (s.stakind3 = 4) THEN s.stanumbers3 WHEN (s.stakind4 = 4) THEN s.stanumbers4 WHEN (s.stakind5 = 4) THEN s.stanumbers5 ELSE NULL::real[] END AS most_common_elem_freqs, CASE WHEN (s.stakind1 = 5) THEN s.stanumbers1 WHEN (s.stakind2 = 5) THEN s.stanumbers2 WHEN (s.stakind3 = 5) THEN s.stanumbers3 WHEN (s.stakind4 = 5) THEN s.stanumbers4 WHEN (s.stakind5 = 5) THEN s.stanumbers5 ELSE NULL::real[] END AS elem_count_histogram FROM (((pg_statistic s JOIN pg_class c ON ((c.oid = s.starelid))) JOIN pg_attribute a ON (((c.oid = a.attrelid) AND (a.attnum = s.staattnum)))) LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) WHERE ((NOT a.attisdropped) AND has_column_privilege(c.oid, a.attnum, 'select'::text));
  pg_tables                       | SELECT n.nspname AS schemaname, c.relname AS tablename, pg_get_userbyid(c.relowner) AS tableowner, t.spcname AS tablespace, c.relhasindex AS hasindexes, c.relhasrules AS hasrules, c.relhastriggers AS hastriggers FROM ((pg_class c LEFT JOIN pg_namespace n ON ((n.oid = c.relnamespace))) LEFT JOIN pg_tablespace t ON ((t.oid = c.reltablespace))) WHERE (c.relkind = 'r'::"char");
  pg_timezone_abbrevs             | SELECT pg_timezone_abbrevs.abbrev, pg_timezone_abbrevs.utc_offset, pg_timezone_abbrevs.is_dst FROM pg_timezone_abbrevs() pg_timezone_abbrevs(abbrev, utc_offset, is_dst);