agora inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH v15 1/8] Row pattern recognition patch for raw parser.
201+ messages / 2 participants
[nested] [flat]

* [PATCH v15 1/8] Row pattern recognition patch for raw parser.
@ 2024-03-28 10:30  Tatsuo Ishii <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Tatsuo Ishii @ 2024-03-28 10:30 UTC (permalink / raw)

---
 src/backend/parser/gram.y       | 220 ++++++++++++++++++++++++++++++--
 src/include/nodes/parsenodes.h  |  67 ++++++++++
 src/include/parser/kwlist.h     |   8 ++
 src/include/parser/parse_node.h |   1 +
 4 files changed, 285 insertions(+), 11 deletions(-)

diff --git a/src/backend/parser/gram.y b/src/backend/parser/gram.y
index c1b0cff1c9..45817b296e 100644
--- a/src/backend/parser/gram.y
+++ b/src/backend/parser/gram.y
@@ -671,6 +671,21 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 				json_object_constructor_null_clause_opt
 				json_array_constructor_null_clause_opt
 
+%type <target>	row_pattern_measure_item
+				row_pattern_definition
+%type <node>	opt_row_pattern_common_syntax
+				opt_row_pattern_skip_to
+				row_pattern_subset_item
+				row_pattern_term
+%type <list>	opt_row_pattern_measures
+				row_pattern_measure_list
+				row_pattern_definition_list
+				opt_row_pattern_subset_clause
+				row_pattern_subset_list
+				row_pattern_subset_rhs
+				row_pattern
+%type <boolean>	opt_row_pattern_initial_or_seek
+				first_or_last
 
 /*
  * Non-keyword token types.  These are hard-wired into the "flex" lexer.
@@ -714,7 +729,7 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 	CURRENT_TIME CURRENT_TIMESTAMP CURRENT_USER CURSOR CYCLE
 
 	DATA_P DATABASE DAY_P DEALLOCATE DEC DECIMAL_P DECLARE DEFAULT DEFAULTS
-	DEFERRABLE DEFERRED DEFINER DELETE_P DELIMITER DELIMITERS DEPENDS DEPTH DESC
+	DEFERRABLE DEFERRED DEFINE DEFINER DELETE_P DELIMITER DELIMITERS DEPENDS DEPTH DESC
 	DETACH DICTIONARY DISABLE_P DISCARD DISTINCT DO DOCUMENT_P DOMAIN_P
 	DOUBLE_P DROP
 
@@ -730,7 +745,7 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 	HANDLER HAVING HEADER_P HOLD HOUR_P
 
 	IDENTITY_P IF_P ILIKE IMMEDIATE IMMUTABLE IMPLICIT_P IMPORT_P IN_P INCLUDE
-	INCLUDING INCREMENT INDENT INDEX INDEXES INHERIT INHERITS INITIALLY INLINE_P
+	INCLUDING INCREMENT INDENT INDEX INDEXES INHERIT INHERITS INITIAL INITIALLY INLINE_P
 	INNER_P INOUT INPUT_P INSENSITIVE INSERT INSTEAD INT_P INTEGER
 	INTERSECT INTERVAL INTO INVOKER IS ISNULL ISOLATION
 
@@ -743,7 +758,7 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 	LEADING LEAKPROOF LEAST LEFT LEVEL LIKE LIMIT LISTEN LOAD LOCAL
 	LOCALTIME LOCALTIMESTAMP LOCATION LOCK_P LOCKED LOGGED
 
-	MAPPING MATCH MATCHED MATERIALIZED MAXVALUE MERGE MERGE_ACTION METHOD
+	MAPPING MATCH MATCHED MATERIALIZED MAXVALUE MEASURES MERGE MERGE_ACTION METHOD
 	MINUTE_P MINVALUE MODE MONTH_P MOVE
 
 	NAME_P NAMES NATIONAL NATURAL NCHAR NEW NEXT NFC NFD NFKC NFKD NO NONE
@@ -755,8 +770,8 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 	ORDER ORDINALITY OTHERS OUT_P OUTER_P
 	OVER OVERLAPS OVERLAY OVERRIDING OWNED OWNER
 
-	PARALLEL PARAMETER PARSER PARTIAL PARTITION PASSING PASSWORD
-	PERIOD PLACING PLANS POLICY
+	PARALLEL PARAMETER PARSER PARTIAL PARTITION PASSING PASSWORD PAST
+	PATTERN_P PERIOD PERMUTE PLACING PLANS POLICY
 	POSITION PRECEDING PRECISION PRESERVE PREPARE PREPARED PRIMARY
 	PRIOR PRIVILEGES PROCEDURAL PROCEDURE PROCEDURES PROGRAM PUBLICATION
 
@@ -767,12 +782,13 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 	RESET RESTART RESTRICT RETURN RETURNING RETURNS REVOKE RIGHT ROLE ROLLBACK ROLLUP
 	ROUTINE ROUTINES ROW ROWS RULE
 
-	SAVEPOINT SCALAR SCHEMA SCHEMAS SCROLL SEARCH SECOND_P SECURITY SELECT
+	SAVEPOINT SCALAR SCHEMA SCHEMAS SCROLL SEARCH SECOND_P SECURITY SEEK SELECT
 	SEQUENCE SEQUENCES
+
 	SERIALIZABLE SERVER SESSION SESSION_USER SET SETS SETOF SHARE SHOW
 	SIMILAR SIMPLE SKIP SMALLINT SNAPSHOT SOME SQL_P STABLE STANDALONE_P
 	START STATEMENT STATISTICS STDIN STDOUT STORAGE STORED STRICT_P STRING_P STRIP_P
-	SUBSCRIPTION SUBSTRING SUPPORT SYMMETRIC SYSID SYSTEM_P SYSTEM_USER
+	SUBSCRIPTION SUBSET SUBSTRING SUPPORT SYMMETRIC SYSID SYSTEM_P SYSTEM_USER
 
 	TABLE TABLES TABLESAMPLE TABLESPACE TEMP TEMPLATE TEMPORARY TEXT_P THEN
 	TIES TIME TIMESTAMP TO TRAILING TRANSACTION TRANSFORM
@@ -878,6 +894,7 @@ static Node *makeRecursiveViewSelect(char *relname, List *aliases, Node *query);
 %nonassoc	UNBOUNDED		/* ideally would have same precedence as IDENT */
 %nonassoc	IDENT PARTITION RANGE ROWS GROUPS PRECEDING FOLLOWING CUBE ROLLUP
 			SET KEYS OBJECT_P SCALAR VALUE_P WITH WITHOUT
+			MEASURES AFTER INITIAL SEEK PATTERN_P
 %left		Op OPERATOR		/* multi-character ops and user-defined operators */
 %left		'+' '-'
 %left		'*' '/' '%'
@@ -16037,7 +16054,8 @@ over_clause: OVER window_specification
 		;
 
 window_specification: '(' opt_existing_window_name opt_partition_clause
-						opt_sort_clause opt_frame_clause ')'
+						opt_sort_clause opt_row_pattern_measures opt_frame_clause
+						opt_row_pattern_common_syntax ')'
 				{
 					WindowDef  *n = makeNode(WindowDef);
 
@@ -16045,10 +16063,12 @@ window_specification: '(' opt_existing_window_name opt_partition_clause
 					n->refname = $2;
 					n->partitionClause = $3;
 					n->orderClause = $4;
+					n->rowPatternMeasures = $5;
 					/* copy relevant fields of opt_frame_clause */
-					n->frameOptions = $5->frameOptions;
-					n->startOffset = $5->startOffset;
-					n->endOffset = $5->endOffset;
+					n->frameOptions = $6->frameOptions;
+					n->startOffset = $6->startOffset;
+					n->endOffset = $6->endOffset;
+					n->rpCommonSyntax = (RPCommonSyntax *)$7;
 					n->location = @1;
 					$$ = n;
 				}
@@ -16072,6 +16092,31 @@ opt_partition_clause: PARTITION BY expr_list		{ $$ = $3; }
 			| /*EMPTY*/								{ $$ = NIL; }
 		;
 
+/*
+ * ROW PATTERN_P MEASURES
+ */
+opt_row_pattern_measures: MEASURES row_pattern_measure_list	{ $$ = $2; }
+			| /*EMPTY*/								{ $$ = NIL; }
+		;
+
+row_pattern_measure_list:
+			row_pattern_measure_item
+					{ $$ = list_make1($1); }
+			| row_pattern_measure_list ',' row_pattern_measure_item
+					{ $$ = lappend($1, $3); }
+		;
+
+row_pattern_measure_item:
+			a_expr AS ColLabel
+				{
+					$$ = makeNode(ResTarget);
+					$$->name = $3;
+					$$->indirection = NIL;
+					$$->val = (Node *) $1;
+					$$->location = @1;
+				}
+		;
+
 /*
  * For frame clauses, we return a WindowDef, but only some fields are used:
  * frameOptions, startOffset, and endOffset.
@@ -16231,6 +16276,143 @@ opt_window_exclusion_clause:
 			| /*EMPTY*/				{ $$ = 0; }
 		;
 
+opt_row_pattern_common_syntax:
+opt_row_pattern_skip_to opt_row_pattern_initial_or_seek
+				PATTERN_P '(' row_pattern ')'
+				opt_row_pattern_subset_clause
+				DEFINE row_pattern_definition_list
+			{
+				RPCommonSyntax *n = makeNode(RPCommonSyntax);
+				n->rpSkipTo = ((RPCommonSyntax *)$1)->rpSkipTo;
+				n->rpSkipVariable = ((RPCommonSyntax *)$1)->rpSkipVariable;
+				n->initial = $2;
+				n->rpPatterns = $5;
+				n->rpSubsetClause = $7;
+				n->rpDefs = $9;
+				$$ = (Node *) n;
+			}
+			| /*EMPTY*/		{ $$ = NULL; }
+	;
+
+opt_row_pattern_skip_to:
+			AFTER MATCH SKIP TO NEXT ROW
+				{
+					RPCommonSyntax *n = makeNode(RPCommonSyntax);
+					n->rpSkipTo = ST_NEXT_ROW;
+					n->rpSkipVariable = NULL;
+					$$ = (Node *) n;
+			}
+			| AFTER MATCH SKIP PAST LAST_P ROW
+				{
+					RPCommonSyntax *n = makeNode(RPCommonSyntax);
+					n->rpSkipTo = ST_PAST_LAST_ROW;
+					n->rpSkipVariable = NULL;
+					$$ = (Node *) n;
+				}
+			| AFTER MATCH SKIP TO first_or_last ColId
+				{
+					RPCommonSyntax *n = makeNode(RPCommonSyntax);
+					n->rpSkipTo = $5? ST_FIRST_VARIABLE : ST_LAST_VARIABLE;
+					n->rpSkipVariable = $6;
+					$$ = (Node *) n;
+				}
+/*
+			| AFTER MATCH SKIP TO LAST_P ColId		%prec LAST_P
+				{
+					RPCommonSyntax *n = makeNode(RPCommonSyntax);
+					n->rpSkipTo = ST_LAST_VARIABLE;
+					n->rpSkipVariable = $6;
+					$$ = n;
+				}
+			| AFTER MATCH SKIP TO ColId
+				{
+					RPCommonSyntax *n = makeNode(RPCommonSyntax);
+					n->rpSkipTo = ST_VARIABLE;
+					n->rpSkipVariable = $5;
+					$$ = n;
+				}
+*/
+			| /*EMPTY*/
+				{
+					RPCommonSyntax *n = makeNode(RPCommonSyntax);
+					/* temporary set default to ST_NEXT_ROW */
+					n->rpSkipTo = ST_PAST_LAST_ROW;
+					n->rpSkipVariable = NULL;
+					$$ = (Node *) n;
+				}
+	;
+
+first_or_last:
+			FIRST_P		{ $$ = true; }
+			| LAST_P	{ $$ = false; }
+	;
+
+opt_row_pattern_initial_or_seek:
+			INITIAL			{ $$ = true; }
+			| SEEK
+				{
+					ereport(ERROR,
+							(errcode(ERRCODE_SYNTAX_ERROR),
+							 errmsg("SEEK is not supported"),
+							 errhint("Use INITIAL."),
+							 parser_errposition(@1)));
+				}
+			| /*EMPTY*/		{ $$ = true; }
+		;
+
+row_pattern:
+			row_pattern_term							{ $$ = list_make1($1); }
+			| row_pattern row_pattern_term				{ $$ = lappend($1, $2); }
+		;
+
+row_pattern_term:
+			ColId	{ $$ = (Node *) makeSimpleA_Expr(AEXPR_OP, "", (Node *)makeString($1), NULL, @1); }
+			| ColId '*'	{ $$ = (Node *) makeSimpleA_Expr(AEXPR_OP, "*", (Node *)makeString($1), NULL, @1); }
+			| ColId '+'	{ $$ = (Node *) makeSimpleA_Expr(AEXPR_OP, "+", (Node *)makeString($1), NULL, @1); }
+			| ColId '?'	{ $$ = (Node *) makeSimpleA_Expr(AEXPR_OP, "?", (Node *)makeString($1), NULL, @1); }
+		;
+
+opt_row_pattern_subset_clause:
+			SUBSET row_pattern_subset_list	{ $$ = $2; }
+			| /*EMPTY*/												{ $$ = NIL; }
+		;
+
+row_pattern_subset_list:
+			row_pattern_subset_item									{ $$ = list_make1($1); }
+			| row_pattern_subset_list ',' row_pattern_subset_item	{ $$ = lappend($1, $3); }
+			| /*EMPTY*/												{ $$ = NIL; }
+		;
+
+row_pattern_subset_item: ColId '=' '(' row_pattern_subset_rhs ')'
+			{
+				RPSubsetItem *n = makeNode(RPSubsetItem);
+				n->name = $1;
+				n->rhsVariable = $4;
+				$$ = (Node *) n;
+			}
+		;
+
+row_pattern_subset_rhs:
+			ColId								{ $$ = list_make1(makeStringConst($1, @1)); }
+			| row_pattern_subset_rhs ',' ColId	{ $$ = lappend($1, makeStringConst($3, @1)); }
+			| /*EMPTY*/							{ $$ = NIL; }
+		;
+
+row_pattern_definition_list:
+			row_pattern_definition										{ $$ = list_make1($1); }
+			| row_pattern_definition_list ',' row_pattern_definition	{ $$ = lappend($1, $3); }
+		;
+
+row_pattern_definition:
+			ColId AS a_expr
+				{
+					$$ = makeNode(ResTarget);
+					$$->name = $1;
+					$$->indirection = NIL;
+					$$->val = (Node *) $3;
+					$$->location = @1;
+				}
+		;
 
 /*
  * Supporting nonterminals for expressions.
@@ -17441,6 +17623,7 @@ unreserved_keyword:
 			| INDEXES
 			| INHERIT
 			| INHERITS
+			| INITIAL
 			| INLINE_P
 			| INPUT_P
 			| INSENSITIVE
@@ -17469,6 +17652,7 @@ unreserved_keyword:
 			| MATCHED
 			| MATERIALIZED
 			| MAXVALUE
+			| MEASURES
 			| MERGE
 			| METHOD
 			| MINUTE_P
@@ -17512,7 +17696,10 @@ unreserved_keyword:
 			| PARTITION
 			| PASSING
 			| PASSWORD
+			| PAST
+			| PATTERN_P
 			| PERIOD
+			| PERMUTE
 			| PLANS
 			| POLICY
 			| PRECEDING
@@ -17564,6 +17751,7 @@ unreserved_keyword:
 			| SEARCH
 			| SECOND_P
 			| SECURITY
+			| SEEK
 			| SEQUENCE
 			| SEQUENCES
 			| SERIALIZABLE
@@ -17590,6 +17778,7 @@ unreserved_keyword:
 			| STRING_P
 			| STRIP_P
 			| SUBSCRIPTION
+			| SUBSET
 			| SUPPORT
 			| SYSID
 			| SYSTEM_P
@@ -17782,6 +17971,7 @@ reserved_keyword:
 			| CURRENT_USER
 			| DEFAULT
 			| DEFERRABLE
+			| DEFINE
 			| DESC
 			| DISTINCT
 			| DO
@@ -17945,6 +18135,7 @@ bare_label_keyword:
 			| DEFAULTS
 			| DEFERRABLE
 			| DEFERRED
+			| DEFINE
 			| DEFINER
 			| DELETE_P
 			| DELIMITER
@@ -18022,6 +18213,7 @@ bare_label_keyword:
 			| INDEXES
 			| INHERIT
 			| INHERITS
+			| INITIAL
 			| INITIALLY
 			| INLINE_P
 			| INNER_P
@@ -18075,6 +18267,7 @@ bare_label_keyword:
 			| MATCHED
 			| MATERIALIZED
 			| MAXVALUE
+			| MEASURES
 			| MERGE
 			| MERGE_ACTION
 			| METHOD
@@ -18130,7 +18323,10 @@ bare_label_keyword:
 			| PARTITION
 			| PASSING
 			| PASSWORD
+			| PAST
+			| PATTERN_P
 			| PERIOD
+			| PERMUTE
 			| PLACING
 			| PLANS
 			| POLICY
@@ -18188,6 +18384,7 @@ bare_label_keyword:
 			| SCROLL
 			| SEARCH
 			| SECURITY
+			| SEEK
 			| SELECT
 			| SEQUENCE
 			| SEQUENCES
@@ -18220,6 +18417,7 @@ bare_label_keyword:
 			| STRING_P
 			| STRIP_P
 			| SUBSCRIPTION
+			| SUBSET
 			| SUBSTRING
 			| SUPPORT
 			| SYMMETRIC
diff --git a/src/include/nodes/parsenodes.h b/src/include/nodes/parsenodes.h
index b89baef95d..28cf577089 100644
--- a/src/include/nodes/parsenodes.h
+++ b/src/include/nodes/parsenodes.h
@@ -549,6 +549,48 @@ typedef struct SortBy
 	ParseLoc	location;		/* operator location, or -1 if none/unknown */
 } SortBy;
 
+/*
+ * AFTER MATCH row pattern skip to types in row pattern common syntax
+ */
+typedef enum RPSkipTo
+{
+	ST_NONE,					/* AFTER MATCH omitted */
+	ST_NEXT_ROW,				/* SKIP TO NEXT ROW */
+	ST_PAST_LAST_ROW,			/* SKIP TO PAST LAST ROW */
+	ST_FIRST_VARIABLE,			/* SKIP TO FIRST variable name */
+	ST_LAST_VARIABLE,			/* SKIP TO LAST variable name */
+	ST_VARIABLE					/* SKIP TO variable name */
+}			RPSkipTo;
+
+/*
+ * Row Pattern SUBSET clause item
+ */
+typedef struct RPSubsetItem
+{
+	NodeTag		type;
+	char	   *name;			/* Row Pattern SUBSET clause variable name */
+	List	   *rhsVariable;	/* Row Pattern SUBSET rhs variables (list of
+								 * char *string) */
+}			RPSubsetItem;
+
+/*
+ * RowPatternCommonSyntax - raw representation of row pattern common syntax
+ *
+ */
+typedef struct RPCommonSyntax
+{
+	NodeTag		type;
+	RPSkipTo	rpSkipTo;		/* Row Pattern AFTER MATCH SKIP type */
+	char	   *rpSkipVariable; /* Row Pattern Skip To variable name, if any */
+	bool		initial;		/* true if <row pattern initial or seek> is
+								 * initial */
+	List	   *rpPatterns;		/* PATTERN variables (list of A_Expr) */
+	List	   *rpSubsetClause; /* row pattern subset clause (list of
+								 * RPSubsetItem), if any */
+	List	   *rpDefs;			/* row pattern definitions clause (list of
+								 * ResTarget) */
+}			RPCommonSyntax;
+
 /*
  * WindowDef - raw representation of WINDOW and OVER clauses
  *
@@ -564,6 +606,9 @@ typedef struct WindowDef
 	char	   *refname;		/* referenced window name, if any */
 	List	   *partitionClause;	/* PARTITION BY expression list */
 	List	   *orderClause;	/* ORDER BY (list of SortBy) */
+	List	   *rowPatternMeasures; /* row pattern measures (list of
+									 * ResTarget) */
+	RPCommonSyntax *rpCommonSyntax; /* row pattern common syntax */
 	int			frameOptions;	/* frame_clause options, see below */
 	Node	   *startOffset;	/* expression for starting bound, if any */
 	Node	   *endOffset;		/* expression for ending bound, if any */
@@ -1516,6 +1561,11 @@ typedef struct GroupingSet
  * the orderClause might or might not be copied (see copiedOrder); the framing
  * options are never copied, per spec.
  *
+ * "defineClause" is Row Pattern Recognition DEFINE clause (list of
+ * TargetEntry). TargetEntry.resname represents row pattern definition
+ * variable name. "patternVariable" and "patternRegexp" represents PATTERN
+ * clause.
+ *
  * The information relevant for the query jumbling is the partition clause
  * type and its bounds.
  */
@@ -1547,6 +1597,23 @@ typedef struct WindowClause
 	Index		winref;			/* ID referenced by window functions */
 	/* did we copy orderClause from refname? */
 	bool		copiedOrder pg_node_attr(query_jumble_ignore);
+	/* Row Pattern AFTER MACH SKIP clause */
+	RPSkipTo	rpSkipTo;		/* Row Pattern Skip To type */
+	char	   *rpSkipVariable; /* Row Pattern Skip To variable */
+	bool		initial;		/* true if <row pattern initial or seek> is
+								 * initial */
+	/* Row Pattern DEFINE clause (list of TargetEntry) */
+	List	   *defineClause;
+	/* Row Pattern DEFINE variable initial names (list of String) */
+	List	   *defineInitial;
+	/* Row Pattern PATTERN variable name (list of String) */
+	List	   *patternVariable;
+
+	/*
+	 * Row Pattern PATTERN regular expression quantifier ('+' or ''. list of
+	 * String)
+	 */
+	List	   *patternRegexp;
 } WindowClause;
 
 /*
diff --git a/src/include/parser/kwlist.h b/src/include/parser/kwlist.h
index 57514d064b..4a47b13c34 100644
--- a/src/include/parser/kwlist.h
+++ b/src/include/parser/kwlist.h
@@ -129,6 +129,7 @@ PG_KEYWORD("default", DEFAULT, RESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("defaults", DEFAULTS, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("deferrable", DEFERRABLE, RESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("deferred", DEFERRED, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("define", DEFINE, RESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("definer", DEFINER, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("delete", DELETE_P, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("delimiter", DELIMITER, UNRESERVED_KEYWORD, BARE_LABEL)
@@ -215,6 +216,7 @@ PG_KEYWORD("index", INDEX, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("indexes", INDEXES, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("inherit", INHERIT, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("inherits", INHERITS, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("initial", INITIAL, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("initially", INITIALLY, RESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("inline", INLINE_P, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("inner", INNER_P, TYPE_FUNC_NAME_KEYWORD, BARE_LABEL)
@@ -272,6 +274,7 @@ PG_KEYWORD("match", MATCH, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("matched", MATCHED, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("materialized", MATERIALIZED, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("maxvalue", MAXVALUE, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("measures", MEASURES, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("merge", MERGE, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("merge_action", MERGE_ACTION, COL_NAME_KEYWORD, BARE_LABEL)
 PG_KEYWORD("method", METHOD, UNRESERVED_KEYWORD, BARE_LABEL)
@@ -335,7 +338,10 @@ PG_KEYWORD("partial", PARTIAL, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("partition", PARTITION, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("passing", PASSING, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("password", PASSWORD, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("past", PAST, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("pattern", PATTERN_P, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("period", PERIOD, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("permute", PERMUTE, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("placing", PLACING, RESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("plans", PLANS, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("policy", POLICY, UNRESERVED_KEYWORD, BARE_LABEL)
@@ -396,6 +402,7 @@ PG_KEYWORD("scroll", SCROLL, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("search", SEARCH, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("second", SECOND_P, UNRESERVED_KEYWORD, AS_LABEL)
 PG_KEYWORD("security", SECURITY, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("seek", SEEK, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("select", SELECT, RESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("sequence", SEQUENCE, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("sequences", SEQUENCES, UNRESERVED_KEYWORD, BARE_LABEL)
@@ -428,6 +435,7 @@ PG_KEYWORD("strict", STRICT_P, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("string", STRING_P, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("strip", STRIP_P, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("subscription", SUBSCRIPTION, UNRESERVED_KEYWORD, BARE_LABEL)
+PG_KEYWORD("subset", SUBSET, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("substring", SUBSTRING, COL_NAME_KEYWORD, BARE_LABEL)
 PG_KEYWORD("support", SUPPORT, UNRESERVED_KEYWORD, BARE_LABEL)
 PG_KEYWORD("symmetric", SYMMETRIC, RESERVED_KEYWORD, BARE_LABEL)
diff --git a/src/include/parser/parse_node.h b/src/include/parser/parse_node.h
index 5b781d87a9..66935afff8 100644
--- a/src/include/parser/parse_node.h
+++ b/src/include/parser/parse_node.h
@@ -51,6 +51,7 @@ typedef enum ParseExprKind
 	EXPR_KIND_WINDOW_FRAME_RANGE,	/* window frame clause with RANGE */
 	EXPR_KIND_WINDOW_FRAME_ROWS,	/* window frame clause with ROWS */
 	EXPR_KIND_WINDOW_FRAME_GROUPS,	/* window frame clause with GROUPS */
+	EXPR_KIND_RPR_DEFINE,		/* DEFINE */
 	EXPR_KIND_SELECT_TARGET,	/* SELECT target list item */
 	EXPR_KIND_INSERT_TARGET,	/* INSERT target list item */
 	EXPR_KIND_UPDATE_SOURCE,	/* UPDATE assignment source item */
-- 
2.25.1


----Next_Part(Thu_Mar_28_19_59_25_2024_076)--
Content-Type: Text/X-Patch; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="v15-0002-Row-pattern-recognition-patch-parse-analysis.patch"



^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread

* [PATCH v1] bufmgr: aio: Prototype for not waiting for already-in-progress IO
@ 2025-08-15 15:01  Andres Freund <[email protected]>
  0 siblings, 0 replies; 201+ messages in thread

From: Andres Freund @ 2025-08-15 15:01 UTC (permalink / raw)

Author:
Reviewed-by:
Discussion: https://postgr.es/m/
Backpatch:
---
 src/include/storage/bufmgr.h        |   1 +
 src/backend/storage/buffer/bufmgr.c | 133 ++++++++++++++++++++++++++--
 2 files changed, 125 insertions(+), 9 deletions(-)

diff --git a/src/include/storage/bufmgr.h b/src/include/storage/bufmgr.h
index 41fdc1e7693..7ddb867bc99 100644
--- a/src/include/storage/bufmgr.h
+++ b/src/include/storage/bufmgr.h
@@ -137,6 +137,7 @@ struct ReadBuffersOperation
 	int			flags;
 	int16		nblocks;
 	int16		nblocks_done;
+	bool		foreign_io;
 	PgAioWaitRef io_wref;
 	PgAioReturn io_return;
 };
diff --git a/src/backend/storage/buffer/bufmgr.c b/src/backend/storage/buffer/bufmgr.c
index fd7e21d96d3..de755fd53ad 100644
--- a/src/backend/storage/buffer/bufmgr.c
+++ b/src/backend/storage/buffer/bufmgr.c
@@ -1557,6 +1557,41 @@ ReadBuffersCanStartIOOnce(Buffer buffer, bool nowait)
 		return StartBufferIO(GetBufferDescriptor(buffer - 1), true, nowait);
 }
 
+static inline bool
+ReadBuffersIOAlreadyInProgress(ReadBuffersOperation *operation, Buffer buffer)
+{
+	BufferDesc *desc;
+	uint32		buf_state;
+	PgAioWaitRef iow;
+
+	pgaio_wref_clear(&iow);
+
+	if (BufferIsLocal(buffer))
+	{
+		desc = GetLocalBufferDescriptor(-buffer - 1);
+		buf_state = pg_atomic_read_u32(&desc->state);
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+	}
+	else
+	{
+		desc = GetBufferDescriptor(buffer - 1);
+		buf_state = LockBufHdr(desc);
+
+		if (buf_state & BM_IO_IN_PROGRESS)
+			iow = desc->io_wref;
+		UnlockBufHdr(desc, buf_state);
+	}
+
+	if (pgaio_wref_valid(&iow))
+	{
+		operation->io_wref = iow;
+		return true;
+	}
+
+	return false;
+}
+
 /*
  * Helper for AsyncReadBuffers that tries to get the buffer ready for IO.
  */
@@ -1689,7 +1724,7 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 			 *
 			 * we first check if we already know the IO is complete.
 			 */
-			if (aio_ret->result.status == PGAIO_RS_UNKNOWN &&
+			if ((operation->foreign_io || aio_ret->result.status == PGAIO_RS_UNKNOWN) &&
 				!pgaio_wref_check_done(&operation->io_wref))
 			{
 				instr_time	io_start = pgstat_prepare_io_time(track_io_timing);
@@ -1708,11 +1743,38 @@ WaitReadBuffers(ReadBuffersOperation *operation)
 				Assert(pgaio_wref_check_done(&operation->io_wref));
 			}
 
-			/*
-			 * We now are sure the IO completed. Check the results. This
-			 * includes reporting on errors if there were any.
-			 */
-			ProcessReadBuffersResult(operation);
+			if (operation->foreign_io)
+			{
+				Buffer		buffer = operation->buffers[operation->nblocks_done];
+				BufferDesc *desc;
+				uint32		buf_state;
+
+				if (BufferIsLocal(buffer))
+				{
+					desc = GetLocalBufferDescriptor(-buffer - 1);
+					buf_state = pg_atomic_read_u32(&desc->state);
+				}
+				else
+				{
+					desc = GetBufferDescriptor(buffer - 1);
+					buf_state = LockBufHdr(desc);
+					UnlockBufHdr(desc, buf_state);
+				}
+
+				if (buf_state & BM_VALID)
+				{
+					operation->nblocks_done += 1;
+					Assert(operation->nblocks_done <= operation->nblocks);
+				}
+			}
+			else
+			{
+				/*
+				 * We now are sure the IO completed. Check the results. This
+				 * includes reporting on errors if there were any.
+				 */
+				ProcessReadBuffersResult(operation);
+			}
 		}
 
 		/*
@@ -1798,6 +1860,56 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		io_object = IOOBJECT_RELATION;
 	}
 
+	/*
+	 * If AIO is in progress, be it in this backend or another backend, we
+	 * just associate the wait reference with the operation and wait in
+	 * WaitReadBuffers(). This turns out to be important for performance in
+	 * two workloads:
+	 *
+	 * 1) A read stream that has to read the same block multiple times within
+	 *	  the readahead distance. This can happen e.g. for the table accesses
+	 *	  of an index scan.
+	 *
+	 * 2) Concurrent scans by multiple backends on the same relation.
+	 *
+	 * If we were to synchronously wait for the in-progress IO, we'd not be
+	 * able to keep enough I/O in flight.
+	 *
+	 *
+	 * If we do find there is ongoing I/O for the buffer, we set up a 1-block
+	 * ReadBuffersOperation that WaitReadBuffers then can wait on.
+	 */
+	if (1 && ReadBuffersIOAlreadyInProgress(operation, buffers[nblocks_done]))
+	{
+		/* FIXME: probably need to wait if io_method == sync? */
+
+		*nblocks_progress = 1;
+		did_start_io = true;
+		operation->foreign_io = true;
+
+		if (0)
+			elog(LOG, "using foreign IO path");
+
+		/* FIXME: trace point */
+
+		/*
+		 * FIXME: how should this be accounted for in stats? Account as a hit
+		 * for now, quite likely *we* started this IO.
+		 */
+		if (persistence == RELPERSISTENCE_TEMP)
+			pgBufferUsage.local_blks_hit += 1;
+		else
+			pgBufferUsage.shared_blks_hit += 1;
+
+		if (operation->rel)
+			pgstat_count_buffer_hit(operation->rel);
+		pgstat_count_io_op(io_object, io_context, IOOP_HIT, 1, 0);
+		if (VacuumCostActive)
+			VacuumCostBalance += VacuumCostPageHit;
+
+		return true;
+	}
+
 	/*
 	 * If zero_damaged_pages is enabled, add the READ_BUFFERS_ZERO_ON_ERROR
 	 * flag. The reason for that is that, hopefully, zero_damaged_pages isn't
@@ -1855,9 +1967,9 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 	/*
 	 * Check if we can start IO on the first to-be-read buffer.
 	 *
-	 * If an I/O is already in progress in another backend, we want to wait
-	 * for the outcome: either done, or something went wrong and we will
-	 * retry.
+	 * If a synchronous I/O in progress in another backend (it can't be this
+	 * backend), we want to wait for the outcome: either done, or something
+	 * went wrong and we will retry.
 	 */
 	if (!ReadBuffersCanStartIO(buffers[nblocks_done], false))
 	{
@@ -1970,6 +2082,7 @@ AsyncReadBuffers(ReadBuffersOperation *operation, int *nblocks_progress)
 		if (VacuumCostActive)
 			VacuumCostBalance += VacuumCostPageMiss * io_buffers_len;
 
+		operation->foreign_io = false;
 		*nblocks_progress = io_buffers_len;
 		did_start_io = true;
 	}
@@ -5986,6 +6099,8 @@ WaitIO(BufferDesc *buf)
 		 */
 		if (pgaio_wref_valid(&iow))
 		{
+			if (0)
+				elog(LOG, "foreign wait");
 			pgaio_wref_wait(&iow);
 
 			/*
-- 
2.48.1.76.g4e746b1a31.dirty


--j4imdtjxqt5tb6mo--





^ permalink  raw  reply  [nested|flat] 201+ messages in thread