agora inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH v3 1/3] Add CREATE OR REPLACE MATERIALIZED VIEW
208+ messages / 4 participants
[nested] [flat]

* [PATCH v3 1/3] Add CREATE OR REPLACE MATERIALIZED VIEW
@ 2024-05-21 16:35  Erik Wienhold <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Erik Wienhold @ 2024-05-21 16:35 UTC (permalink / raw)

---
 .../sgml/ref/create_materialized_view.sgml    |  15 +-
 src/backend/commands/createas.c               | 207 ++++++++++++++----
 src/backend/commands/tablecmds.c              |   8 +-
 src/backend/commands/view.c                   | 106 ++++++---
 src/backend/parser/gram.y                     |  15 ++
 src/bin/psql/tab-complete.c                   |  15 +-
 src/include/commands/view.h                   |   3 +
 src/include/nodes/parsenodes.h                |   2 +-
 src/include/nodes/primnodes.h                 |   1 +
 src/test/regress/expected/matview.out         | 191 ++++++++++++++++
 src/test/regress/sql/matview.sql              | 108 +++++++++
 11 files changed, 582 insertions(+), 89 deletions(-)

diff --git a/doc/src/sgml/ref/create_materialized_view.sgml b/doc/src/sgml/ref/create_materialized_view.sgml
index 0d2fea2b97..b5a8e3441a 100644
--- a/doc/src/sgml/ref/create_materialized_view.sgml
+++ b/doc/src/sgml/ref/create_materialized_view.sgml
@@ -21,7 +21,7 @@ PostgreSQL documentation
 
  <refsynopsisdiv>
 <synopsis>
-CREATE MATERIALIZED VIEW [ IF NOT EXISTS ] <replaceable>table_name</replaceable>
+CREATE [ OR REPLACE ] MATERIALIZED VIEW [ IF NOT EXISTS ] <replaceable>table_name</replaceable>
     [ (<replaceable>column_name</replaceable> [, ...] ) ]
     [ USING <replaceable class="parameter">method</replaceable> ]
     [ WITH ( <replaceable class="parameter">storage_parameter</replaceable> [= <replaceable class="parameter">value</replaceable>] [, ... ] ) ]
@@ -60,6 +60,17 @@ CREATE MATERIALIZED VIEW [ IF NOT EXISTS ] <replaceable>table_name</replaceable>
   <title>Parameters</title>
 
   <variablelist>
+   <varlistentry>
+    <term><literal>OR REPLACE</literal></term>
+    <listitem>
+     <para>
+      Replaces a materialized view if it already exists.
+      Specifying <literal>OR REPLACE</literal> together with
+      <literal>IF NOT EXISTS</literal> is an error.
+     </para>
+    </listitem>
+   </varlistentry>
+
    <varlistentry>
     <term><literal>IF NOT EXISTS</literal></term>
     <listitem>
@@ -67,7 +78,7 @@ CREATE MATERIALIZED VIEW [ IF NOT EXISTS ] <replaceable>table_name</replaceable>
       Do not throw an error if a materialized view with the same name already
       exists. A notice is issued in this case.  Note that there is no guarantee
       that the existing materialized view is anything like the one that would
-      have been created.
+      have been created, unless you use <literal>OR REPLACE</literal> instead.
      </para>
     </listitem>
    </varlistentry>
diff --git a/src/backend/commands/createas.c b/src/backend/commands/createas.c
index 0b629b1f79..e4ed3748f9 100644
--- a/src/backend/commands/createas.c
+++ b/src/backend/commands/createas.c
@@ -79,55 +79,151 @@ static void intorel_destroy(DestReceiver *self);
 static ObjectAddress
 create_ctas_internal(List *attrList, IntoClause *into)
 {
-	CreateStmt *create = makeNode(CreateStmt);
-	bool		is_matview;
+	bool		is_matview,
+				replace = false;
 	char		relkind;
-	Datum		toast_options;
-	const char *const validnsps[] = HEAP_RELOPT_NAMESPACES;
+	Oid			matviewOid = InvalidOid;
 	ObjectAddress intoRelationAddr;
 
 	/* This code supports both CREATE TABLE AS and CREATE MATERIALIZED VIEW */
 	is_matview = (into->viewQuery != NULL);
 	relkind = is_matview ? RELKIND_MATVIEW : RELKIND_RELATION;
 
-	/*
-	 * Create the target relation by faking up a CREATE TABLE parsetree and
-	 * passing it to DefineRelation.
-	 */
-	create->relation = into->rel;
-	create->tableElts = attrList;
-	create->inhRelations = NIL;
-	create->ofTypename = NULL;
-	create->constraints = NIL;
-	create->options = into->options;
-	create->oncommit = into->onCommit;
-	create->tablespacename = into->tableSpaceName;
-	create->if_not_exists = false;
-	create->accessMethod = into->accessMethod;
+	/* Check if an existing materialized view needs to be replaced. */
+	if (is_matview)
+	{
+		LOCKMODE	lockmode;
 
-	/*
-	 * Create the relation.  (This will error out if there's an existing view,
-	 * so we don't need more code to complain if "replace" is false.)
-	 */
-	intoRelationAddr = DefineRelation(create, relkind, InvalidOid, NULL, NULL);
+		lockmode = into->replace ? AccessExclusiveLock : NoLock;
+		(void) RangeVarGetAndCheckCreationNamespace(into->rel, lockmode,
+													&matviewOid);
+		replace = OidIsValid(matviewOid) && into->replace;
+	}
 
-	/*
-	 * If necessary, create a TOAST table for the target table.  Note that
-	 * NewRelationCreateToastTable ends with CommandCounterIncrement(), so
-	 * that the TOAST table will be visible for insertion.
-	 */
-	CommandCounterIncrement();
+	if (is_matview && replace)
+	{
+		Relation	rel;
+		List	   *atcmds = NIL;
+		AlterTableCmd *atcmd;
+		TupleDesc	descriptor;
+
+		rel = relation_open(matviewOid, NoLock);
+
+		if (rel->rd_rel->relkind != RELKIND_MATVIEW)
+			ereport(ERROR,
+					errcode(ERRCODE_WRONG_OBJECT_TYPE),
+					errmsg("\"%s\" is not a materialized view",
+						   RelationGetRelationName(rel)));
+
+		CheckTableNotInUse(rel, "CREATE OR REPLACE MATERIALIZED VIEW");
+
+		descriptor = BuildDescForRelation(attrList);
+		checkViewColumns(descriptor, rel->rd_att, true);
+
+		/* Add new attributes via ALTER TABLE. */
+		if (list_length(attrList) > rel->rd_att->natts)
+		{
+			ListCell   *c;
+			int			skip = rel->rd_att->natts;
+
+			foreach(c, attrList)
+			{
+				if (skip > 0)
+				{
+					skip--;
+					continue;
+				}
+				atcmd = makeNode(AlterTableCmd);
+				atcmd->subtype = AT_AddColumnToView;
+				atcmd->def = (Node *) lfirst(c);
+				atcmds = lappend(atcmds, atcmd);
+			}
+		}
+
+		/* Set access method via ALTER TABLE. */
+		if (into->accessMethod != NULL)
+		{
+			atcmd = makeNode(AlterTableCmd);
+			atcmd->subtype = AT_SetAccessMethod;
+			atcmd->name = into->accessMethod;
+			atcmds = lappend(atcmds, atcmd);
+		}
+
+		/* Set tablespace via ALTER TABLE. */
+		if (into->tableSpaceName != NULL)
+		{
+			atcmd = makeNode(AlterTableCmd);
+			atcmd->subtype = AT_SetTableSpace;
+			atcmd->name = into->tableSpaceName;
+			atcmds = lappend(atcmds, atcmd);
+		}
+
+		/* Set storage parameters via ALTER TABLE. */
+		if (into->options != NIL)
+		{
+			atcmd = makeNode(AlterTableCmd);
+			atcmd->subtype = AT_ReplaceRelOptions;
+			atcmd->def = (Node *) into->options;
+			atcmds = lappend(atcmds, atcmd);
+		}
+
+		if (atcmds != NIL)
+		{
+			AlterTableInternal(matviewOid, atcmds, true);
+			CommandCounterIncrement();
+		}
+
+		relation_close(rel, NoLock);
+
+		ObjectAddressSet(intoRelationAddr, RelationRelationId, matviewOid);
+	}
+	else
+	{
+		CreateStmt *create = makeNode(CreateStmt);
+		Datum		toast_options;
+		const static char *validnsps[] = HEAP_RELOPT_NAMESPACES;
+
+		/*
+		 * Create the target relation by faking up a CREATE TABLE parsetree
+		 * and passing it to DefineRelation.
+		 */
+		create->relation = into->rel;
+		create->tableElts = attrList;
+		create->inhRelations = NIL;
+		create->ofTypename = NULL;
+		create->constraints = NIL;
+		create->options = into->options;
+		create->oncommit = into->onCommit;
+		create->tablespacename = into->tableSpaceName;
+		create->if_not_exists = false;
+		create->accessMethod = into->accessMethod;
+
+		/*
+		 * Create the relation.  (This will error out if there's an existing
+		 * view, so we don't need more code to complain if "replace" is
+		 * false.)
+		 */
+		intoRelationAddr = DefineRelation(create, relkind, InvalidOid, NULL,
+										  NULL);
 
-	/* parse and validate reloptions for the toast table */
-	toast_options = transformRelOptions((Datum) 0,
-										create->options,
-										"toast",
-										validnsps,
-										true, false);
+		/*
+		 * If necessary, create a TOAST table for the target table.  Note that
+		 * NewRelationCreateToastTable ends with CommandCounterIncrement(), so
+		 * that the TOAST table will be visible for insertion.
+		 */
+		CommandCounterIncrement();
+
+		/* parse and validate reloptions for the toast table */
+		toast_options = transformRelOptions((Datum) 0,
+											create->options,
+											"toast",
+											validnsps,
+											true, false);
 
-	(void) heap_reloptions(RELKIND_TOASTVALUE, toast_options, true);
+		(void) heap_reloptions(RELKIND_TOASTVALUE, toast_options, true);
 
-	NewRelationCreateToastTable(intoRelationAddr.objectId, toast_options);
+		NewRelationCreateToastTable(intoRelationAddr.objectId, toast_options);
+	}
 
 	/* Create the "view" part of a materialized view. */
 	if (is_matview)
@@ -135,7 +231,7 @@ create_ctas_internal(List *attrList, IntoClause *into)
 		/* StoreViewQuery scribbles on tree, so make a copy */
 		Query	   *query = (Query *) copyObject(into->viewQuery);
 
-		StoreViewQuery(intoRelationAddr.objectId, query, false);
+		StoreViewQuery(intoRelationAddr.objectId, query, replace);
 		CommandCounterIncrement();
 	}
 
@@ -231,7 +327,26 @@ ExecCreateTableAs(ParseState *pstate, CreateTableAsStmt *stmt,
 
 	/* Check if the relation exists or not */
 	if (CreateTableAsRelExists(stmt))
+	{
+		/* An existing materialized view can be replaced. */
+		if (is_matview && into->replace)
+		{
+			RefreshMatViewStmt *refresh;
+
+			/* Change the relation to match the new query and other options. */
+			(void) create_ctas_nodata(query->targetList, into);
+
+			/* Refresh the materialized view with a fake statement. */
+			refresh = makeNode(RefreshMatViewStmt);
+			refresh->relation = into->rel;
+			refresh->skipData = into->skipData;
+			refresh->concurrent = false;
+
+			return ExecRefreshMatView(refresh, NULL, NULL);
+		}
+
 		return InvalidObjectAddress;
+	}
 
 	/*
 	 * Create the tuple receiver object and insert info it will need
@@ -392,14 +507,15 @@ CreateTableAsRelExists(CreateTableAsStmt *ctas)
 	oldrelid = get_relname_relid(into->rel->relname, nspid);
 	if (OidIsValid(oldrelid))
 	{
-		if (!ctas->if_not_exists)
+		if (!ctas->if_not_exists && !into->replace)
 			ereport(ERROR,
 					(errcode(ERRCODE_DUPLICATE_TABLE),
 					 errmsg("relation \"%s\" already exists",
 							into->rel->relname)));
 
 		/*
-		 * The relation exists and IF NOT EXISTS has been specified.
+		 * The relation exists and IF NOT EXISTS or OR REPLACE has been
+		 * specified.
 		 *
 		 * If we are in an extension script, insist that the pre-existing
 		 * object be a member of the extension, to avoid security risks.
@@ -407,11 +523,12 @@ CreateTableAsRelExists(CreateTableAsStmt *ctas)
 		ObjectAddressSet(address, RelationRelationId, oldrelid);
 		checkMembershipInCurrentExtension(&address);
 
-		/* OK to skip */
-		ereport(NOTICE,
-				(errcode(ERRCODE_DUPLICATE_TABLE),
-				 errmsg("relation \"%s\" already exists, skipping",
-						into->rel->relname)));
+		if (ctas->if_not_exists)
+			/* OK to skip */
+			ereport(NOTICE,
+					(errcode(ERRCODE_DUPLICATE_TABLE),
+					 errmsg("relation \"%s\" already exists, skipping",
+							into->rel->relname)));
 		return true;
 	}
 
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index b3cc6f8f69..bcc08f9420 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -4480,7 +4480,7 @@ AlterTableGetLockLevel(List *cmds)
 				 * Subcommands that may be visible to concurrent SELECTs
 				 */
 			case AT_DropColumn: /* change visible to SELECT */
-			case AT_AddColumnToView:	/* CREATE VIEW */
+			case AT_AddColumnToView:	/* via CREATE OR REPLACE [MATERIALIZED] VIEW */
 			case AT_DropOids:	/* used to equiv to DropColumn */
 			case AT_EnableAlwaysRule:	/* may change SELECT rules */
 			case AT_EnableReplicaRule:	/* may change SELECT rules */
@@ -4783,8 +4783,8 @@ ATPrepCmd(List **wqueue, Relation rel, AlterTableCmd *cmd,
 			/* Recursion occurs during execution phase */
 			pass = AT_PASS_ADD_COL;
 			break;
-		case AT_AddColumnToView:	/* add column via CREATE OR REPLACE VIEW */
-			ATSimplePermissions(cmd->subtype, rel, ATT_VIEW);
+		case AT_AddColumnToView:	/* via CREATE OR REPLACE [MATERIALIZED] VIEW */
+			ATSimplePermissions(cmd->subtype, rel, ATT_VIEW | ATT_MATVIEW);
 			ATPrepAddColumn(wqueue, rel, recurse, recursing, true, cmd,
 							lockmode, context);
 			/* Recursion occurs during execution phase */
@@ -5178,7 +5178,7 @@ ATExecCmd(List **wqueue, AlteredTableInfo *tab,
 	switch (cmd->subtype)
 	{
 		case AT_AddColumn:		/* ADD COLUMN */
-		case AT_AddColumnToView:	/* add column via CREATE OR REPLACE VIEW */
+		case AT_AddColumnToView:	/* via CREATE OR REPLACE [MATERIALIZED] VIEW */
 			address = ATExecAddColumn(wqueue, tab, rel, &cmd,
 									  cmd->recurse, false,
 									  lockmode, cur_pass, context);
diff --git a/src/backend/commands/view.c b/src/backend/commands/view.c
index fdad833832..76532aa35d 100644
--- a/src/backend/commands/view.c
+++ b/src/backend/commands/view.c
@@ -30,8 +30,6 @@
 #include "utils/lsyscache.h"
 #include "utils/rel.h"
 
-static void checkViewColumns(TupleDesc newdesc, TupleDesc olddesc);
-
 /*---------------------------------------------------------------------
  * DefineVirtualRelation
  *
@@ -130,7 +128,7 @@ DefineVirtualRelation(RangeVar *relation, List *tlist, bool replace,
 		 * column list.
 		 */
 		descriptor = BuildDescForRelation(attrList);
-		checkViewColumns(descriptor, rel->rd_att);
+		checkViewColumns(descriptor, rel->rd_att, false);
 
 		/*
 		 * If new attributes have been added, we must add pg_attribute entries
@@ -263,15 +261,22 @@ DefineVirtualRelation(RangeVar *relation, List *tlist, bool replace,
  * added to generate specific complaints.  Also, we allow the new view to have
  * more columns than the old.
  */
-static void
-checkViewColumns(TupleDesc newdesc, TupleDesc olddesc)
+void
+checkViewColumns(TupleDesc newdesc, TupleDesc olddesc, bool is_matview)
 {
 	int			i;
 
 	if (newdesc->natts < olddesc->natts)
-		ereport(ERROR,
-				(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
-				 errmsg("cannot drop columns from view")));
+	{
+		if (is_matview)
+			ereport(ERROR,
+					errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+					errmsg("cannot drop columns from materialized view"));
+		else
+			ereport(ERROR,
+					(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+					 errmsg("cannot drop columns from view")));
+	}
 
 	for (i = 0; i < olddesc->natts; i++)
 	{
@@ -280,17 +285,34 @@ checkViewColumns(TupleDesc newdesc, TupleDesc olddesc)
 
 		/* XXX msg not right, but we don't support DROP COL on view anyway */
 		if (newattr->attisdropped != oldattr->attisdropped)
-			ereport(ERROR,
-					(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
-					 errmsg("cannot drop columns from view")));
+		{
+			if (is_matview)
+				ereport(ERROR,
+						errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						errmsg("cannot drop columns from materialized view"));
+			else
+				ereport(ERROR,
+						(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						 errmsg("cannot drop columns from view")));
+		}
 
 		if (strcmp(NameStr(newattr->attname), NameStr(oldattr->attname)) != 0)
-			ereport(ERROR,
-					(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
-					 errmsg("cannot change name of view column \"%s\" to \"%s\"",
-							NameStr(oldattr->attname),
-							NameStr(newattr->attname)),
-					 errhint("Use ALTER VIEW ... RENAME COLUMN ... to change name of view column instead.")));
+		{
+			if (is_matview)
+				ereport(ERROR,
+						errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						errmsg("cannot change name of materialized view column \"%s\" to \"%s\"",
+							   NameStr(oldattr->attname),
+							   NameStr(newattr->attname)),
+						errhint("Use ALTER MATERIALIZED VIEW ... RENAME COLUMN ... to change name of materialized view column instead."));
+			else
+				ereport(ERROR,
+						(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						 errmsg("cannot change name of view column \"%s\" to \"%s\"",
+								NameStr(oldattr->attname),
+								NameStr(newattr->attname)),
+						 errhint("Use ALTER VIEW ... RENAME COLUMN ... to change name of view column instead.")));
+		}
 
 		/*
 		 * We cannot allow type, typmod, or collation to change, since these
@@ -299,26 +321,48 @@ checkViewColumns(TupleDesc newdesc, TupleDesc olddesc)
 		 */
 		if (newattr->atttypid != oldattr->atttypid ||
 			newattr->atttypmod != oldattr->atttypmod)
-			ereport(ERROR,
-					(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
-					 errmsg("cannot change data type of view column \"%s\" from %s to %s",
-							NameStr(oldattr->attname),
-							format_type_with_typemod(oldattr->atttypid,
-													 oldattr->atttypmod),
-							format_type_with_typemod(newattr->atttypid,
-													 newattr->atttypmod))));
+		{
+			if (is_matview)
+				ereport(ERROR,
+						errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						errmsg("cannot change data type of materialized view column \"%s\" from %s to %s",
+							   NameStr(oldattr->attname),
+							   format_type_with_typemod(oldattr->atttypid,
+														oldattr->atttypmod),
+							   format_type_with_typemod(newattr->atttypid,
+														newattr->atttypmod)));
+			else
+				ereport(ERROR,
+						(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						 errmsg("cannot change data type of view column \"%s\" from %s to %s",
+								NameStr(oldattr->attname),
+								format_type_with_typemod(oldattr->atttypid,
+														 oldattr->atttypmod),
+								format_type_with_typemod(newattr->atttypid,
+														 newattr->atttypmod))));
+		}
 
 		/*
 		 * At this point, attcollations should be both valid or both invalid,
 		 * so applying get_collation_name unconditionally should be fine.
 		 */
 		if (newattr->attcollation != oldattr->attcollation)
-			ereport(ERROR,
-					(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
-					 errmsg("cannot change collation of view column \"%s\" from \"%s\" to \"%s\"",
-							NameStr(oldattr->attname),
-							get_collation_name(oldattr->attcollation),
-							get_collation_name(newattr->attcollation))));
+		{
+			if (is_matview)
+				ereport(ERROR,
+						errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						errmsg("cannot change collation of materialized view column \"%s\" from \"%s\" to \"%s\"",
+							   NameStr(oldattr->attname),
+							   get_collation_name(oldattr->attcollation),
+							   get_collation_name(newattr->attcollation)));
+			else
+				ereport(ERROR,
+						(errcode(ERRCODE_INVALID_TABLE_DEFINITION),
+						 errmsg("cannot change collation of view column \"%s\" from \"%s\" to \"%s\"",
+								NameStr(oldattr->attname),
+								get_collation_name(oldattr->attcollation),
+								get_collation_name(newattr->attcollation))));
+		}
 	}
 
 	/*
diff --git a/src/backend/parser/gram.y b/src/backend/parser/gram.y
index 84cef57a70..b02b2f42d2 100644
--- a/src/backend/parser/gram.y
+++ b/src/backend/parser/gram.y
@@ -4779,6 +4779,21 @@ CreateMatViewStmt:
 					$8->skipData = !($11);
 					$$ = (Node *) ctas;
 				}
+		| CREATE OR REPLACE OptNoLog MATERIALIZED VIEW create_mv_target AS SelectStmt opt_with_data
+				{
+					CreateTableAsStmt *ctas = makeNode(CreateTableAsStmt);
+
+					ctas->query = $9;
+					ctas->into = $7;
+					ctas->objtype = OBJECT_MATVIEW;
+					ctas->is_select_into = false;
+					ctas->if_not_exists = false;
+					/* cram additional flags into the IntoClause */
+					$7->rel->relpersistence = $4;
+					$7->skipData = !($10);
+					$7->replace = true;
+					$$ = (Node *) ctas;
+				}
 		;
 
 create_mv_target:
diff --git a/src/bin/psql/tab-complete.c b/src/bin/psql/tab-complete.c
index a7ccde6d7d..c50a3781db 100644
--- a/src/bin/psql/tab-complete.c
+++ b/src/bin/psql/tab-complete.c
@@ -1811,7 +1811,7 @@ psql_completion(const char *text, int start, int end)
 	/* complete with something you can create or replace */
 	else if (TailMatches("CREATE", "OR", "REPLACE"))
 		COMPLETE_WITH("FUNCTION", "PROCEDURE", "LANGUAGE", "RULE", "VIEW",
-					  "AGGREGATE", "TRANSFORM", "TRIGGER");
+					  "AGGREGATE", "TRANSFORM", "TRIGGER", "MATERIALIZED VIEW");
 
 /* DROP, but not DROP embedded in other commands */
 	/* complete with something you can drop */
@@ -3599,13 +3599,16 @@ psql_completion(const char *text, int start, int end)
 		COMPLETE_WITH("SELECT");
 
 /* CREATE MATERIALIZED VIEW */
-	else if (Matches("CREATE", "MATERIALIZED"))
+	else if (Matches("CREATE", "MATERIALIZED") ||
+			 Matches("CREATE", "OR", "REPLACE", "MATERIALIZED"))
 		COMPLETE_WITH("VIEW");
-	/* Complete CREATE MATERIALIZED VIEW <name> with AS */
-	else if (Matches("CREATE", "MATERIALIZED", "VIEW", MatchAny))
+	/* Complete CREATE [ OR REPLACE ] MATERIALIZED VIEW <name> with AS */
+	else if (Matches("CREATE", "MATERIALIZED", "VIEW", MatchAny) ||
+			 Matches("CREATE", "OR", "REPLACE", "MATERIALIZED", "VIEW", MatchAny))
 		COMPLETE_WITH("AS");
-	/* Complete "CREATE MATERIALIZED VIEW <sth> AS with "SELECT" */
-	else if (Matches("CREATE", "MATERIALIZED", "VIEW", MatchAny, "AS"))
+	/* Complete "CREATE [ OR REPLACE ] MATERIALIZED VIEW <sth> AS with "SELECT" */
+	else if (Matches("CREATE", "MATERIALIZED", "VIEW", MatchAny, "AS") ||
+			 Matches("CREATE", "OR", "REPLACE", "MATERIALIZED", "VIEW", MatchAny, "AS"))
 		COMPLETE_WITH("SELECT");
 
 /* CREATE EVENT TRIGGER */
diff --git a/src/include/commands/view.h b/src/include/commands/view.h
index d2d8588989..7eacdaaceb 100644
--- a/src/include/commands/view.h
+++ b/src/include/commands/view.h
@@ -22,4 +22,7 @@ extern ObjectAddress DefineView(ViewStmt *stmt, const char *queryString,
 
 extern void StoreViewQuery(Oid viewOid, Query *viewParse, bool replace);
 
+extern void checkViewColumns(TupleDesc newdesc, TupleDesc olddesc,
+							 bool is_matview);
+
 #endif							/* VIEW_H */
diff --git a/src/include/nodes/parsenodes.h b/src/include/nodes/parsenodes.h
index 124d853e49..3ba536eee3 100644
--- a/src/include/nodes/parsenodes.h
+++ b/src/include/nodes/parsenodes.h
@@ -2338,7 +2338,7 @@ typedef struct AlterTableStmt
 typedef enum AlterTableType
 {
 	AT_AddColumn,				/* add column */
-	AT_AddColumnToView,			/* implicitly via CREATE OR REPLACE VIEW */
+	AT_AddColumnToView,			/* implicitly via CREATE OR REPLACE [MATERIALIZED] VIEW */
 	AT_ColumnDefault,			/* alter column default */
 	AT_CookedColumnDefault,		/* add a pre-cooked column default */
 	AT_DropNotNull,				/* alter column drop not null */
diff --git a/src/include/nodes/primnodes.h b/src/include/nodes/primnodes.h
index ea47652adb..4b0ee5d10d 100644
--- a/src/include/nodes/primnodes.h
+++ b/src/include/nodes/primnodes.h
@@ -168,6 +168,7 @@ typedef struct IntoClause
 	/* materialized view's SELECT query */
 	Node	   *viewQuery pg_node_attr(query_jumble_ignore);
 	bool		skipData;		/* true for WITH NO DATA */
+	bool		replace;		/* replace existing matview? */
 } IntoClause;
 
 
diff --git a/src/test/regress/expected/matview.out b/src/test/regress/expected/matview.out
index 038ab73517..e2e2a13396 100644
--- a/src/test/regress/expected/matview.out
+++ b/src/test/regress/expected/matview.out
@@ -694,3 +694,194 @@ NOTICE:  relation "matview_ine_tab" already exists, skipping
 (0 rows)
 
 DROP MATERIALIZED VIEW matview_ine_tab;
+--
+-- test CREATE OR REPLACE MATERIALIZED VIEW
+--
+-- matview does not already exist
+DROP MATERIALIZED VIEW IF EXISTS mvtest_replace;
+NOTICE:  materialized view "mvtest_replace" does not exist, skipping
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 1 AS a;
+SELECT * FROM mvtest_replace;
+ a 
+---
+ 1
+(1 row)
+
+-- replace query with data
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 2 AS a;
+SELECT * FROM mvtest_replace;
+ a 
+---
+ 2
+(1 row)
+
+-- replace query without data
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 3 AS a
+  WITH NO DATA;
+SELECT * FROM mvtest_replace; -- error: not populated
+ERROR:  materialized view "mvtest_replace" has not been populated
+HINT:  Use the REFRESH MATERIALIZED VIEW command.
+REFRESH MATERIALIZED VIEW mvtest_replace;
+SELECT * FROM mvtest_replace;
+ a 
+---
+ 3
+(1 row)
+
+-- add column
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 4 AS a, 1 b;
+SELECT * FROM mvtest_replace;
+ a | b 
+---+---
+ 4 | 1
+(1 row)
+
+-- replace table options
+SELECT m.*, c.relname, c.reloptions, s.spcname, a.amname
+  FROM mvtest_replace m
+    CROSS JOIN pg_class c
+    LEFT JOIN pg_tablespace s ON s.oid = c.reltablespace
+    LEFT JOIN pg_am a ON a.oid = c.relam
+  WHERE c.relname = 'mvtest_replace';
+ a | b |    relname     | reloptions | spcname | amname 
+---+---+----------------+------------+---------+--------
+ 4 | 1 | mvtest_replace |            |         | heap
+(1 row)
+
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace
+  USING heap2
+  WITH (fillfactor = 50)
+  TABLESPACE regress_tblspace
+  AS SELECT 5 AS a, 1 AS b;
+SELECT m.*, c.relname, c.reloptions, s.spcname, a.amname
+  FROM mvtest_replace m
+    CROSS JOIN pg_class c
+    LEFT JOIN pg_tablespace s ON s.oid = c.reltablespace
+    LEFT JOIN pg_am a ON a.oid = c.relam
+  WHERE c.relname = 'mvtest_replace';
+ a | b |    relname     |   reloptions    |     spcname      | amname 
+---+---+----------------+-----------------+------------------+--------
+ 5 | 1 | mvtest_replace | {fillfactor=50} | regress_tblspace | heap2
+(1 row)
+
+-- can replace matview that has a dependent view
+CREATE VIEW mvtest_replace_v AS
+  SELECT * FROM mvtest_replace;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 6 AS a, 1 AS b;
+SELECT * FROM mvtest_replace, mvtest_replace_v;
+ a | b | a | b 
+---+---+---+---
+ 6 | 1 | 6 | 1
+(1 row)
+
+DROP VIEW mvtest_replace_v;
+-- index gets rebuilt when replacing with data
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 7 AS a, 1 AS b;
+CREATE UNIQUE INDEX ON mvtest_replace (b);
+SELECT * FROM mvtest_replace;
+ a | b 
+---+---
+ 7 | 1
+(1 row)
+
+SET enable_seqscan = off; -- force index scan
+EXPLAIN (COSTS OFF) SELECT * FROM mvtest_replace WHERE b = 1;
+                       QUERY PLAN                        
+---------------------------------------------------------
+ Index Scan using mvtest_replace_b_idx on mvtest_replace
+   Index Cond: (b = 1)
+(2 rows)
+
+SELECT * FROM mvtest_replace WHERE b = 1;
+ a | b 
+---+---
+ 7 | 1
+(1 row)
+
+RESET enable_seqscan;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 8 AS a, 1 AS b;
+SET enable_seqscan = off; -- force index scan
+EXPLAIN (COSTS OFF) SELECT * FROM mvtest_replace WHERE b = 1;
+                       QUERY PLAN                        
+---------------------------------------------------------
+ Index Scan using mvtest_replace_b_idx on mvtest_replace
+   Index Cond: (b = 1)
+(2 rows)
+
+SELECT * FROM mvtest_replace WHERE b = 1;
+ a | b 
+---+---
+ 8 | 1
+(1 row)
+
+RESET enable_seqscan;
+-- cannot change column data type
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 9 AS a, 'x' AS b; -- error
+ERROR:  cannot change data type of materialized view column "b" from integer to text
+SELECT * FROM mvtest_replace;
+ a | b 
+---+---
+ 8 | 1
+(1 row)
+
+-- cannot rename column
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 10 AS a, 1 AS b2; -- error
+ERROR:  cannot change name of materialized view column "b" to "b2"
+HINT:  Use ALTER MATERIALIZED VIEW ... RENAME COLUMN ... to change name of materialized view column instead.
+SELECT * FROM mvtest_replace;
+ a | b 
+---+---
+ 8 | 1
+(1 row)
+
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 11 AS a, 1 AS b, 'y' COLLATE "C" AS c;
+SELECT * FROM mvtest_replace;
+ a  | b | c 
+----+---+---
+ 11 | 1 | y
+(1 row)
+
+-- cannot change column collation
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 12 AS a, 1 AS b, 'x' COLLATE "POSIX" AS c; -- error
+ERROR:  cannot change collation of materialized view column "c" from "C" to "POSIX"
+SELECT * FROM mvtest_replace;
+ a  | b | c 
+----+---+---
+ 11 | 1 | y
+(1 row)
+
+-- cannot drop column
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 13 AS a, 1 AS b; -- error
+ERROR:  cannot drop columns from materialized view
+SELECT * FROM mvtest_replace;
+ a  | b | c 
+----+---+---
+ 11 | 1 | y
+(1 row)
+
+-- must target a matview
+CREATE VIEW mvtest_not_mv AS
+  SELECT 1 AS a;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_not_mv AS
+  SELECT 1 AS a; -- error
+ERROR:  "mvtest_not_mv" is not a materialized view
+DROP VIEW mvtest_not_mv;
+-- cannot use OR REPLACE with IF NOT EXISTS
+CREATE OR REPLACE MATERIALIZED VIEW IF NOT EXISTS mvtest_replace AS
+  SELECT 1 AS a;
+ERROR:  syntax error at or near "NOT"
+LINE 1: CREATE OR REPLACE MATERIALIZED VIEW IF NOT EXISTS mvtest_rep...
+                                               ^
+DROP MATERIALIZED VIEW mvtest_replace;
diff --git a/src/test/regress/sql/matview.sql b/src/test/regress/sql/matview.sql
index b74ee305e0..c12f0243c9 100644
--- a/src/test/regress/sql/matview.sql
+++ b/src/test/regress/sql/matview.sql
@@ -314,3 +314,111 @@ EXPLAIN (ANALYZE, COSTS OFF, SUMMARY OFF, TIMING OFF)
   CREATE MATERIALIZED VIEW IF NOT EXISTS matview_ine_tab AS
     SELECT 1 / 0 WITH NO DATA; -- ok
 DROP MATERIALIZED VIEW matview_ine_tab;
+
+--
+-- test CREATE OR REPLACE MATERIALIZED VIEW
+--
+
+-- matview does not already exist
+DROP MATERIALIZED VIEW IF EXISTS mvtest_replace;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 1 AS a;
+SELECT * FROM mvtest_replace;
+
+-- replace query with data
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 2 AS a;
+SELECT * FROM mvtest_replace;
+
+-- replace query without data
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 3 AS a
+  WITH NO DATA;
+SELECT * FROM mvtest_replace; -- error: not populated
+REFRESH MATERIALIZED VIEW mvtest_replace;
+SELECT * FROM mvtest_replace;
+
+-- add column
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 4 AS a, 1 b;
+SELECT * FROM mvtest_replace;
+
+-- replace table options
+SELECT m.*, c.relname, c.reloptions, s.spcname, a.amname
+  FROM mvtest_replace m
+    CROSS JOIN pg_class c
+    LEFT JOIN pg_tablespace s ON s.oid = c.reltablespace
+    LEFT JOIN pg_am a ON a.oid = c.relam
+  WHERE c.relname = 'mvtest_replace';
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace
+  USING heap2
+  WITH (fillfactor = 50)
+  TABLESPACE regress_tblspace
+  AS SELECT 5 AS a, 1 AS b;
+SELECT m.*, c.relname, c.reloptions, s.spcname, a.amname
+  FROM mvtest_replace m
+    CROSS JOIN pg_class c
+    LEFT JOIN pg_tablespace s ON s.oid = c.reltablespace
+    LEFT JOIN pg_am a ON a.oid = c.relam
+  WHERE c.relname = 'mvtest_replace';
+
+-- can replace matview that has a dependent view
+CREATE VIEW mvtest_replace_v AS
+  SELECT * FROM mvtest_replace;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 6 AS a, 1 AS b;
+SELECT * FROM mvtest_replace, mvtest_replace_v;
+DROP VIEW mvtest_replace_v;
+
+-- index gets rebuilt when replacing with data
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 7 AS a, 1 AS b;
+CREATE UNIQUE INDEX ON mvtest_replace (b);
+SELECT * FROM mvtest_replace;
+SET enable_seqscan = off; -- force index scan
+EXPLAIN (COSTS OFF) SELECT * FROM mvtest_replace WHERE b = 1;
+SELECT * FROM mvtest_replace WHERE b = 1;
+RESET enable_seqscan;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 8 AS a, 1 AS b;
+SET enable_seqscan = off; -- force index scan
+EXPLAIN (COSTS OFF) SELECT * FROM mvtest_replace WHERE b = 1;
+SELECT * FROM mvtest_replace WHERE b = 1;
+RESET enable_seqscan;
+
+-- cannot change column data type
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 9 AS a, 'x' AS b; -- error
+SELECT * FROM mvtest_replace;
+
+-- cannot rename column
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 10 AS a, 1 AS b2; -- error
+SELECT * FROM mvtest_replace;
+
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 11 AS a, 1 AS b, 'y' COLLATE "C" AS c;
+SELECT * FROM mvtest_replace;
+
+-- cannot change column collation
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 12 AS a, 1 AS b, 'x' COLLATE "POSIX" AS c; -- error
+SELECT * FROM mvtest_replace;
+
+-- cannot drop column
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_replace AS
+  SELECT 13 AS a, 1 AS b; -- error
+SELECT * FROM mvtest_replace;
+
+-- must target a matview
+CREATE VIEW mvtest_not_mv AS
+  SELECT 1 AS a;
+CREATE OR REPLACE MATERIALIZED VIEW mvtest_not_mv AS
+  SELECT 1 AS a; -- error
+DROP VIEW mvtest_not_mv;
+
+-- cannot use OR REPLACE with IF NOT EXISTS
+CREATE OR REPLACE MATERIALIZED VIEW IF NOT EXISTS mvtest_replace AS
+  SELECT 1 AS a;
+
+DROP MATERIALIZED VIEW mvtest_replace;
-- 
2.46.0


--muysnh7l3kienazz
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v3-0002-Deprecate-CREATE-MATERIALIZED-VIEW-IF-NOT-EXISTS.patch"



^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread

* [PATCH v1] Add per-backend AIO statistics
@ 2026-06-11 09:45  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 208+ messages in thread

From: Bertrand Drouvot @ 2026-06-11 09:45 UTC (permalink / raw)

This commit adds per-backend AIO statistics, providing per-backend AIO behavior
details.

This data can be retrieved with a new system function called
pg_stat_get_backend_aio(), that returns the following counters based on the PID
provided in input:

- started: total number of AIO operations initiated
- executed_sync: IOs that were executed synchronously (fallback path)
- executed_async: IOs that were submitted asynchronously to the IO method
- completed_self: IO completions processed by the issuing backend itself
- completed_other: IO completions processed on behalf of another backend
- handle_waits: times the backend had to wait for a free AIO handle
- submitted: number of submitted calls to the IO method

These counters are useful for understanding and tuning AIO behavior:

- executed_async / started. A ratio near zero means the backend is falling back
to synchronous execution (TOAST chunk fetches, temp buffers, ...).

- a non-zero handle_waits means the backend exhausted all its AIO handles. That
could mean that io_max_concurrency is too low.

- completed_self vs completed_other reveals cross-backend completion
patterns. That helps see how IO completion work is distributed and could
help interpret per backend IO statistics values.

- executed_async / submitted gives the average batch size per submit call.

This commit is straight-forward, relying on the infrastructure provided
by 9aea73fc61d4 (backend-level pgstats).

XXX: Bump catalog version. No need to touch PGSTAT_FILE_FORMAT_ID as backend
statistics are never written to disk.

Author: Bertrand Drouvot <[email protected]>
Reviewed-by:
Discussion:
---
 doc/src/sgml/monitoring.sgml                |  70 +++++++++++++
 src/backend/storage/aio/aio.c               |  13 +++
 src/backend/utils/activity/pgstat_backend.c | 109 ++++++++++++++++++++
 src/backend/utils/adt/pgstatfuncs.c         |  60 +++++++++++
 src/include/catalog/pg_proc.dat             |   7 ++
 src/include/pgstat.h                        |  25 +++++
 src/include/utils/pgstat_internal.h         |   3 +-
 src/test/modules/test_aio/t/001_aio.pl      |  32 ++++++
 src/tools/pgindent/typedefs.list            |   1 +
 9 files changed, 319 insertions(+), 1 deletion(-)
  24.4% doc/src/sgml/
   3.2% src/backend/storage/aio/
  25.3% src/backend/utils/activity/
  20.5% src/backend/utils/adt/
   4.9% src/include/catalog/
  10.0% src/include/
  11.1% src/test/modules/test_aio/t/

diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 12b9ee20d4a..1c566cf0e0e 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -5641,6 +5641,76 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        </para></entry>
       </row>
 
+      <row>
+       <entry id="pg-stat-get-backend-aio" role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_stat_get_backend_aio</primary>
+        </indexterm>
+        <function>pg_stat_get_backend_aio</function> ( <type>integer</type> )
+        <returnvalue>record</returnvalue>
+       </para>
+       <para>
+        Returns <acronym>AIO</acronym> (Asynchronous I/O) statistics about the
+        backend with the specified process ID. The returned values are:
+       <itemizedlist>
+        <listitem>
+         <para>
+          <literal>started</literal>: Total IOs initiated.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+           <literal>executed_sync</literal>: IOs executed synchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>executed_async</literal>: IOs submitted asynchronously.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_self</literal>: IO completions processed by this
+          backend.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>completed_other</literal>: IO completions processed on
+          behalf of other backends.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>handle_waits</literal>: Times waited for a free AIO handle.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>submitted</literal>: Number of submit calls to the IO
+          method. Compare with <literal>executed_async</literal> to determine
+          average batch size.
+         </para>
+        </listitem>
+        <listitem>
+         <para>
+          <literal>stats_reset</literal>: Timestamp of last stats reset.
+         </para>
+        </listitem>
+       </itemizedlist>
+       </para>
+       <para>
+        The <literal>completed_other</literal> column is only meaningful
+        when <varname>io_method</varname> is set to <literal>io_uring</literal>;
+        with <literal>worker</literal> mode, IO completions are processed by
+        IO worker processes which do not track these statistics.
+       </para>
+       <para>
+        The function does not return AIO statistics for the checkpointer,
+        the background writer, the startup process and the autovacuum launcher.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/storage/aio/aio.c b/src/backend/storage/aio/aio.c
index 8f7e26607b9..507f1727d41 100644
--- a/src/backend/storage/aio/aio.c
+++ b/src/backend/storage/aio/aio.c
@@ -40,6 +40,7 @@
 
 #include "lib/ilist.h"
 #include "miscadmin.h"
+#include "pgstat.h"
 #include "port/atomics.h"
 #include "storage/aio.h"
 #include "storage/aio_internal.h"
@@ -458,6 +459,8 @@ pgaio_io_stage(PgAioHandle *ioh, PgAioOp op)
 				   "staged (synchronous: %d, in_batch: %d)",
 				   needs_synchronous, pgaio_my_backend->in_batchmode);
 
+	pgstat_count_backend_aio_start(needs_synchronous);
+
 	if (!needs_synchronous)
 	{
 		pgaio_my_backend->staged_ios[pgaio_my_backend->num_staged_ios++] = ioh;
@@ -544,6 +547,12 @@ pgaio_io_process_completion(PgAioHandle *ioh, int result)
 	/* condition variable broadcast ensures state is visible before wakeup */
 	ConditionVariableBroadcast(&ioh->cv);
 
+	/* Track AIO completion stats */
+	if (ioh->owner_procno == MyProcNumber)
+		pgstat_count_backend_aio_complete_self();
+	else
+		pgstat_count_backend_aio_complete_other();
+
 	/* contains call to pgaio_io_call_complete_local() */
 	if (ioh->owner_procno == MyProcNumber)
 		pgaio_io_reclaim(ioh);
@@ -762,6 +771,8 @@ pgaio_io_wait_for_free(void)
 {
 	int			reclaimed = 0;
 
+	pgstat_count_backend_aio_handle_wait();
+
 	pgaio_debug(DEBUG2, "waiting for free IO with %d pending, %u in-flight, %u idle IOs",
 				pgaio_my_backend->num_staged_ios,
 				dclist_count(&pgaio_my_backend->in_flight_ios),
@@ -1150,6 +1161,8 @@ pgaio_submit_staged(void)
 
 	Assert(total_submitted == did_submit);
 
+	pgstat_count_backend_aio_submitted();
+
 	pgaio_my_backend->num_staged_ios = 0;
 
 	pgaio_debug(DEBUG4,
diff --git a/src/backend/utils/activity/pgstat_backend.c b/src/backend/utils/activity/pgstat_backend.c
index b736b2ccc6f..4ad755f7f18 100644
--- a/src/backend/utils/activity/pgstat_backend.c
+++ b/src/backend/utils/activity/pgstat_backend.c
@@ -40,6 +40,7 @@
 static PgStat_BackendPending PendingBackendStats;
 static bool backend_has_iostats = false;
 static bool backend_has_lockstats = false;
+static bool backend_has_aiostats = false;
 
 /*
  * WAL usage counters saved from pgWalUsage at the previous call to
@@ -120,6 +121,74 @@ pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type)
 	pgstat_report_fixed = true;
 }
 
+/*
+ * Utility routines to report AIO stats for backends, kept here to avoid
+ * exposing PendingBackendStats to the outside world.
+ */
+void
+pgstat_count_backend_aio_start(bool synchronous)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.started++;
+	if (synchronous)
+		PendingBackendStats.aio_counters.executed_sync++;
+	else
+		PendingBackendStats.aio_counters.executed_async++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_self(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_self++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_complete_other(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.completed_other++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_handle_wait(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.handle_waits++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
+void
+pgstat_count_backend_aio_submitted(void)
+{
+	if (!pgstat_tracks_backend_bktype(MyBackendType))
+		return;
+
+	PendingBackendStats.aio_counters.submitted++;
+
+	backend_has_aiostats = true;
+	pgstat_report_fixed = true;
+}
+
 /*
  * Returns statistics of a backend by proc number.
  */
@@ -326,6 +395,38 @@ pgstat_flush_backend_entry_lock(PgStat_EntryRef *entry_ref)
 	backend_has_lockstats = false;
 }
 
+/*
+ * Flush out locally pending backend AIO statistics.  Locking is managed
+ * by the caller.
+ */
+static void
+pgstat_flush_backend_entry_aio(PgStat_EntryRef *entry_ref)
+{
+	PgStatShared_Backend *shbackendent;
+	PgStat_AioCounters *bktype_shstats;
+
+	if (!backend_has_aiostats)
+		return;
+
+	shbackendent = (PgStatShared_Backend *) entry_ref->shared_stats;
+	bktype_shstats = &shbackendent->stats.aio_counters;
+
+#define AIOSTAT_ACC(fld) \
+	(bktype_shstats->fld += PendingBackendStats.aio_counters.fld)
+	AIOSTAT_ACC(started);
+	AIOSTAT_ACC(executed_sync);
+	AIOSTAT_ACC(executed_async);
+	AIOSTAT_ACC(completed_self);
+	AIOSTAT_ACC(completed_other);
+	AIOSTAT_ACC(handle_waits);
+	AIOSTAT_ACC(submitted);
+#undef AIOSTAT_ACC
+
+	MemSet(&PendingBackendStats.aio_counters, 0, sizeof(PgStat_AioCounters));
+
+	backend_has_aiostats = false;
+}
+
 /*
  * Flush out locally pending backend statistics
  *
@@ -354,6 +455,10 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if ((flags & PGSTAT_BACKEND_FLUSH_LOCK) && backend_has_lockstats)
 		has_pending_data = true;
 
+	/* Some AIO data pending? */
+	if ((flags & PGSTAT_BACKEND_FLUSH_AIO) && backend_has_aiostats)
+		has_pending_data = true;
+
 	if (!has_pending_data)
 		return false;
 
@@ -372,6 +477,9 @@ pgstat_flush_backend(bool nowait, uint32 flags)
 	if (flags & PGSTAT_BACKEND_FLUSH_LOCK)
 		pgstat_flush_backend_entry_lock(entry_ref);
 
+	if (flags & PGSTAT_BACKEND_FLUSH_AIO)
+		pgstat_flush_backend_entry_aio(entry_ref);
+
 	pgstat_unlock_entry(entry_ref);
 
 	return false;
@@ -411,6 +519,7 @@ pgstat_create_backend(ProcNumber procnum)
 	MemSet(&PendingBackendStats, 0, sizeof(PgStat_BackendPending));
 	backend_has_iostats = false;
 	backend_has_lockstats = false;
+	backend_has_aiostats = false;
 
 	/*
 	 * Initialize prevBackendWalUsage with pgWalUsage so that
diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c
index 565d0e70768..fb62a56f3fe 100644
--- a/src/backend/utils/adt/pgstatfuncs.c
+++ b/src/backend/utils/adt/pgstatfuncs.c
@@ -1722,6 +1722,66 @@ pg_stat_get_backend_wal(PG_FUNCTION_ARGS)
 	return (pg_stat_wal_build_tuple(bktype_stats, backend_stats->stat_reset_timestamp));
 }
 
+/*
+ * Returns AIO statistics for a backend with given PID.
+ */
+Datum
+pg_stat_get_backend_aio(PG_FUNCTION_ARGS)
+{
+#define PG_STAT_BACKEND_AIO_COLS	8
+	TupleDesc	tupdesc;
+	Datum		values[PG_STAT_BACKEND_AIO_COLS] = {0};
+	bool		nulls[PG_STAT_BACKEND_AIO_COLS] = {0};
+	int			pid;
+	PgStat_Backend *backend_stats;
+	PgStat_AioCounters aio_counters;
+
+	pid = PG_GETARG_INT32(0);
+	backend_stats = pgstat_fetch_stat_backend_by_pid(pid, NULL);
+
+	if (!backend_stats)
+		PG_RETURN_NULL();
+
+	aio_counters = backend_stats->aio_counters;
+
+	/* Initialise attributes information in the tuple descriptor */
+	tupdesc = CreateTemplateTupleDesc(PG_STAT_BACKEND_AIO_COLS);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "started",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 2, "executed_sync",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 3, "executed_async",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 4, "completed_self",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 5, "completed_other",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 6, "handle_waits",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 7, "submitted",
+					   INT8OID, -1, 0);
+	TupleDescInitEntry(tupdesc, (AttrNumber) 8, "stats_reset",
+					   TIMESTAMPTZOID, -1, 0);
+	TupleDescFinalize(tupdesc);
+	BlessTupleDesc(tupdesc);
+
+	/* Fill values */
+	values[0] = Int64GetDatum(aio_counters.started);
+	values[1] = Int64GetDatum(aio_counters.executed_sync);
+	values[2] = Int64GetDatum(aio_counters.executed_async);
+	values[3] = Int64GetDatum(aio_counters.completed_self);
+	values[4] = Int64GetDatum(aio_counters.completed_other);
+	values[5] = Int64GetDatum(aio_counters.handle_waits);
+	values[6] = Int64GetDatum(aio_counters.submitted);
+
+	if (backend_stats->stat_reset_timestamp != 0)
+		values[7] = TimestampTzGetDatum(backend_stats->stat_reset_timestamp);
+	else
+		nulls[7] = true;
+
+	PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls)));
+}
+
 /*
  * Returns statistics of WAL activity
  */
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 3cb84359adf..a089860ada4 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -6108,6 +6108,13 @@
   proargmodes => '{i,o,o,o,o,o}',
   proargnames => '{backend_pid,locktype,waits,wait_time,fastpath_exceeded,stats_reset}',
   prosrc => 'pg_stat_get_backend_lock' },
+{ oid => '9082', descr => 'statistics: backend AIO activity',
+  proname => 'pg_stat_get_backend_aio', provolatile => 'v', proparallel => 'r',
+  prorettype => 'record', proargtypes => 'int4',
+  proallargtypes => '{int4,int8,int8,int8,int8,int8,int8,int8,timestamptz}',
+  proargmodes => '{i,o,o,o,o,o,o,o,o}',
+  proargnames => '{backend_pid,started,executed_sync,executed_async,completed_self,completed_other,handle_waits,submitted,stats_reset}',
+  prosrc => 'pg_stat_get_backend_aio' },
 { oid => '6248', descr => 'statistics: information about WAL prefetching',
   proname => 'pg_stat_get_recovery_prefetch', prorows => '1', proretset => 't',
   provolatile => 'v', prorettype => 'record', proargtypes => '',
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 58a44857f13..64afb2bc082 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -514,6 +514,21 @@ typedef struct PgStat_WalStats
 	TimestampTz stat_reset_timestamp;
 } PgStat_WalStats;
 
+/* -------
+ * PgStat_AioCounters	AIO activity counters
+ * -------
+ */
+typedef struct PgStat_AioCounters
+{
+	PgStat_Counter started;
+	PgStat_Counter executed_sync;
+	PgStat_Counter executed_async;
+	PgStat_Counter completed_self;
+	PgStat_Counter completed_other;
+	PgStat_Counter handle_waits;
+	PgStat_Counter submitted;
+} PgStat_AioCounters;
+
 /* -------
  * PgStat_Backend		Backend statistics
  * -------
@@ -524,6 +539,7 @@ typedef struct PgStat_Backend
 	PgStat_BktypeIO io_stats;
 	PgStat_WalCounters wal_counters;
 	PgStat_PendingLock lock_stats;
+	PgStat_AioCounters aio_counters;
 } PgStat_Backend;
 
 /* ---------
@@ -542,6 +558,8 @@ typedef struct PgStat_BackendPending
 	 * PGSTAT_KIND_LOCK.
 	 */
 	PgStat_PendingLock pending_lock;
+	/* Store the AIO statistics counters */
+	PgStat_AioCounters aio_counters;
 } PgStat_BackendPending;
 
 /*
@@ -598,6 +616,13 @@ extern void pgstat_count_backend_io_op(IOObject io_object,
 extern void pgstat_count_backend_lock_waits(uint8 locktag_type, PgStat_Counter usecs);
 extern void pgstat_count_backend_lock_fastpath_exceeded(uint8 locktag_type);
 
+/* used by aio.c for AIO stats tracked in backends */
+extern void pgstat_count_backend_aio_start(bool synchronous);
+extern void pgstat_count_backend_aio_complete_self(void);
+extern void pgstat_count_backend_aio_complete_other(void);
+extern void pgstat_count_backend_aio_handle_wait(void);
+extern void pgstat_count_backend_aio_submitted(void);
+
 extern PgStat_Backend *pgstat_fetch_stat_backend(ProcNumber procNumber);
 extern PgStat_Backend *pgstat_fetch_stat_backend_by_pid(int pid,
 														BackendType *bktype);
diff --git a/src/include/utils/pgstat_internal.h b/src/include/utils/pgstat_internal.h
index b3dc3ff7d8b..b2092fb42b9 100644
--- a/src/include/utils/pgstat_internal.h
+++ b/src/include/utils/pgstat_internal.h
@@ -706,7 +706,8 @@ extern void pgstat_archiver_snapshot_cb(void);
 #define PGSTAT_BACKEND_FLUSH_IO		(1 << 0)	/* Flush I/O statistics */
 #define PGSTAT_BACKEND_FLUSH_WAL   (1 << 1) /* Flush WAL statistics */
 #define PGSTAT_BACKEND_FLUSH_LOCK  (1 << 2) /* Flush lock statistics */
-#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK)
+#define PGSTAT_BACKEND_FLUSH_AIO   (1 << 3) /* Flush AIO statistics */
+#define PGSTAT_BACKEND_FLUSH_ALL   (PGSTAT_BACKEND_FLUSH_IO | PGSTAT_BACKEND_FLUSH_WAL | PGSTAT_BACKEND_FLUSH_LOCK | PGSTAT_BACKEND_FLUSH_AIO)
 
 extern bool pgstat_flush_backend(bool nowait, uint32 flags);
 extern bool pgstat_backend_flush_cb(bool nowait);
diff --git a/src/test/modules/test_aio/t/001_aio.pl b/src/test/modules/test_aio/t/001_aio.pl
index 63cadd64c15..8ecf3d3ad91 100644
--- a/src/test/modules/test_aio/t/001_aio.pl
+++ b/src/test/modules/test_aio/t/001_aio.pl
@@ -1842,6 +1842,37 @@ read_buffers('$table', 0, 4)|,
 	$psql_c->quit();
 }
 
+# Test per-backend AIO statistics counters
+sub test_aio_stats
+{
+	my $io_method = shift;
+	my $node = shift;
+
+	my $psql = $node->background_psql('postgres', on_error_stop => 0);
+
+	# Reset backend stats, evict relation, then read it back to force
+	# physical IO through the AIO layer.
+	$psql->query_safe(qq(SELECT pg_stat_reset_backend_stats(pg_backend_pid())));
+	$psql->query_safe(qq(SELECT evict_rel('tbl_ok')));
+	$psql->query_safe(qq(SELECT count(*) FROM tbl_ok));
+	$psql->query_safe(qq(SELECT pg_stat_force_next_flush()));
+
+	# started must be > 0 after physical IO
+	my $started = $psql->query_safe(
+		qq(SELECT started FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	cmp_ok($started, '>', 0,
+		"$io_method: AIO stats: started > 0 after physical IO");
+
+	# invariant: started = executed_sync + executed_async
+	my $consistent = $psql->query_safe(
+		qq(SELECT started = executed_sync + executed_async
+		   FROM pg_stat_get_backend_aio(pg_backend_pid())));
+	is($consistent, 't',
+		"$io_method: AIO stats: started = executed_sync + executed_async");
+
+	$psql->quit();
+}
+
 # Run all tests that for the specified node / io_method
 sub test_io_method
 {
@@ -1878,6 +1909,7 @@ CHECKPOINT;
 	test_ignore_checksum($io_method, $node);
 	test_checksum_createdb($io_method, $node);
 	test_read_buffers($io_method, $node);
+	test_aio_stats($io_method, $node);
 
 	# generic injection tests
   SKIP:
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 117e7379f10..6ce3077fa5b 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -2329,6 +2329,7 @@ PgStatShared_ReplSlot
 PgStatShared_SLRU
 PgStatShared_Subscription
 PgStatShared_Wal
+PgStat_AioCounters
 PgStat_ArchiverStats
 PgStat_Backend
 PgStat_BackendPending
-- 
2.34.1


--Pi0/Nu3LVhi0Ij+H--





^ permalink  raw  reply  [nested|flat] 208+ messages in thread