public inbox for [email protected]
help / color / mirror / Atom feed[PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely
29+ messages / 14 participants
[nested] [flat]
* [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely
@ 2019-02-20 11:35 Etsuro Fujita <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Etsuro Fujita @ 2019-02-20 11:35 UTC (permalink / raw)
---
contrib/postgres_fdw/deparse.c | 37 +-
.../postgres_fdw/expected/postgres_fdw.out | 618 ++++++------------
contrib/postgres_fdw/postgres_fdw.c | 203 +++++-
contrib/postgres_fdw/postgres_fdw.h | 3 +-
contrib/postgres_fdw/sql/postgres_fdw.sql | 2 +-
src/backend/optimizer/plan/planner.c | 10 +-
src/include/nodes/pathnodes.h | 17 +
7 files changed, 461 insertions(+), 429 deletions(-)
diff --git a/contrib/postgres_fdw/deparse.c b/contrib/postgres_fdw/deparse.c
index 97dd07bee8..079406f4f3 100644
--- a/contrib/postgres_fdw/deparse.c
+++ b/contrib/postgres_fdw/deparse.c
@@ -169,6 +169,7 @@ static void deparseSelectSql(List *tlist, bool is_subquery, List **retrieved_att
static void deparseLockingClause(deparse_expr_cxt *context);
static void appendOrderByClause(List *pathkeys, bool has_final_sort,
deparse_expr_cxt *context);
+static void appendLimitClause(deparse_expr_cxt *context);
static void appendConditions(List *exprs, deparse_expr_cxt *context);
static void deparseFromExprForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, bool use_alias,
@@ -930,7 +931,7 @@ build_tlist_to_deparse(RelOptInfo *foreignrel)
void
deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
List *tlist, List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit, bool is_subquery,
List **retrieved_attrs, List **params_list)
{
deparse_expr_cxt context;
@@ -988,6 +989,10 @@ deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
if (pathkeys)
appendOrderByClause(pathkeys, has_final_sort, &context);
+ /* Add LIMIT clause if necessary */
+ if (has_limit)
+ appendLimitClause(&context);
+
/* Add any necessary FOR UPDATE/SHARE. */
deparseLockingClause(&context);
}
@@ -1591,7 +1596,8 @@ deparseRangeTblRef(StringInfo buf, PlannerInfo *root, RelOptInfo *foreignrel,
/* Deparse the subquery representing the relation. */
appendStringInfoChar(buf, '(');
deparseSelectStmtForRel(buf, root, foreignrel, NIL,
- fpinfo->remote_conds, NIL, false, true,
+ fpinfo->remote_conds, NIL,
+ false, false, true,
&retrieved_attrs, params_list);
appendStringInfoChar(buf, ')');
@@ -3160,6 +3166,33 @@ appendOrderByClause(List *pathkeys, bool has_final_sort,
reset_transmission_modes(nestlevel);
}
+/*
+ * Deparse LIMIT/OFFSET clause.
+ */
+static void
+appendLimitClause(deparse_expr_cxt *context)
+{
+ PlannerInfo *root = context->root;
+ StringInfo buf = context->buf;
+ int nestlevel;
+
+ /* Make sure any constants in the exprs are printed portably */
+ nestlevel = set_transmission_modes();
+
+ if (root->parse->limitCount)
+ {
+ appendStringInfoString(buf, " LIMIT ");
+ deparseExpr((Expr *) root->parse->limitCount, context);
+ }
+ if (root->parse->limitOffset)
+ {
+ appendStringInfoString(buf, " OFFSET ");
+ deparseExpr((Expr *) root->parse->limitOffset, context);
+ }
+
+ reset_transmission_modes(nestlevel);
+}
+
/*
* appendFunctionName
* Deparses function name from given function oid.
diff --git a/contrib/postgres_fdw/expected/postgres_fdw.out b/contrib/postgres_fdw/expected/postgres_fdw.out
index ee36acc850..6e89a6234c 100644
--- a/contrib/postgres_fdw/expected/postgres_fdw.out
+++ b/contrib/postgres_fdw/expected/postgres_fdw.out
@@ -236,11 +236,10 @@ ALTER FOREIGN TABLE ft2 OPTIONS (use_remote_estimate 'true');
-- ===================================================================
-- single table without alias
EXPLAIN (COSTS OFF) SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------
- Limit
- -> Foreign Scan on ft1
-(2 rows)
+ QUERY PLAN
+---------------------
+ Foreign Scan on ft1
+(1 row)
SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -288,14 +287,12 @@ SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
-- whole-row reference
EXPLAIN (VERBOSE, COSTS OFF) SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: t1.*, c3, c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.*, c3, c1
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(3 rows)
SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
t1
@@ -335,14 +332,12 @@ SELECT * FROM ft1 t1 WHERE t1.c1 = 101 AND t1.c6 = '1' AND t1.c7 >= '1';
-- with FOR UPDATE/SHARE
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -351,14 +346,12 @@ SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
(1 row)
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -1015,15 +1008,13 @@ ANALYZE ft5;
-- join two tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -1043,18 +1034,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3, t1.c3
- -> Sort
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Sort Key: t1.c3, t1.c1
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1))))
-(9 rows)
+ Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1)))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1074,15 +1060,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t
-- left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1102,15 +1086,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.
-- left outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1175,15 +1157,13 @@ SELECT t1.c1, t1.c2, t2.c1, t2.c2 FROM ft4 t1 LEFT JOIN (SELECT * FROM ft5 WHERE
-- right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1203,15 +1183,13 @@ SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2
-- right outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1231,15 +1209,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGH
-- full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 45::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
c1 | c1
@@ -1283,15 +1259,13 @@ SELECT t1.c1, t2.c1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL
EXPLAIN (VERBOSE, COSTS OFF)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: 1
- -> Foreign Scan
- Output: 1
- Relations: (public.ft4) FULL JOIN (public.ft5)
- Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE))
-(6 rows)
+ Relations: (public.ft4) FULL JOIN (public.ft5)
+ Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
?column?
@@ -1406,15 +1380,13 @@ SELECT t1.c1, ss.a, ss.b FROM (SELECT c1 FROM "S 1"."T 3" WHERE c1 = 50) t1 INNE
-- full outer join + inner join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t3.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1, t3.c1
- Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST
-(6 rows)
+ Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST LIMIT 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
c1 | c1 | c1
@@ -1434,15 +1406,13 @@ SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 a
-- full outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1462,15 +1432,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1490,15 +1458,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT
-- right outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1518,15 +1484,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1546,15 +1510,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1574,15 +1536,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- right outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1602,15 +1562,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1661,15 +1619,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) WHERE (t1.c1
-- full outer join + WHERE clause with shippable extensions set
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t1.c3 FROM ft1 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) WHERE postgres_fdw_abs(t1.c1) > 0 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t1.c3
- Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0))
-(6 rows)
+ Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
ALTER SERVER loopback OPTIONS (DROP extensions);
-- full outer join + WHERE clause with shippable extensions not set
@@ -1691,37 +1647,13 @@ ALTER SERVER loopback OPTIONS (ADD extensions 'postgres_fdw');
-- tests whole-row reference for row marks
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
c1 | c1
@@ -1740,37 +1672,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1 FOR UPDATE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1 FOR UPDATE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
c1 | c1
@@ -1790,37 +1698,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join two tables with FOR SHARE clause
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
c1 | c1
@@ -1839,37 +1723,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1 FOR SHARE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1 FOR SHARE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
c1 | c1
@@ -1923,15 +1783,13 @@ WITH t (c1_1, c1_3, c2_1) AS MATERIALIZED (SELECT t1.c1, t1.c3, t2.c1 FROM ft1 t
-- ctid with whole-row reference
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.ctid, t1, t2, t1.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- -> Foreign Scan
- Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
-- SEMI JOIN, not pushed down
EXPLAIN (VERBOSE, COSTS OFF)
@@ -1999,27 +1857,16 @@ SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2
119
(10 rows)
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Sort
- Output: t1.c1, t2.c1
- Sort Key: t1.c1, t2.c1
- -> Nested Loop
- Output: t1.c1, t2.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
- -> Materialize
- Output: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
-(15 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1" FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (TRUE)) ORDER BY r1."C 1" ASC NULLS LAST, r2."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -2341,50 +2188,13 @@ SET enable_hashjoin TO false;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- -> Foreign Scan
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
- Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Merge Cond: (ft1.c2 = ft5.c1)
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*, ft4.c1, ft4.c2, ft4.c3, ft4.*
- Merge Cond: (ft1.c2 = ft4.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Sort Key: ft1.c2
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Merge Cond: (ft1.c1 = ft2.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Sort Key: ft1.c1
- -> Foreign Scan on public.ft1
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) FOR UPDATE
- -> Materialize
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- -> Foreign Scan on public.ft2
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) ORDER BY "C 1" ASC NULLS LAST FOR UPDATE
- -> Sort
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Sort Key: ft4.c1
- -> Foreign Scan on public.ft4
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 3" FOR UPDATE
- -> Sort
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Sort Key: ft5.c1
- -> Foreign Scan on public.ft5
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 4" FOR UPDATE
-(41 rows)
+ Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
+ Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
+(4 rows)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
@@ -2452,15 +2262,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1
ALTER VIEW v4 OWNER TO regress_view_owner;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, ft5.c2, ft5.c1
- -> Foreign Scan
- Output: ft4.c1, ft5.c2, ft5.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5)
- Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5)
+ Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2517,15 +2325,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c
ALTER VIEW v4 OWNER TO CURRENT_USER;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, t2.c2, t2.c1
- -> Foreign Scan
- Output: ft4.c1, t2.c2, t2.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -4021,14 +3827,12 @@ SELECT * FROM ft1 t1 WHERE t1.tableoid = 'ft1'::regclass LIMIT 1;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------
- Limit
- Output: ((tableoid)::regclass), c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(5 rows)
+ QUERY PLAN
+-----------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
tableoid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4053,14 +3857,12 @@ SELECT * FROM ft1 t1 WHERE t1.ctid = '(0,2)';
EXPLAIN (VERBOSE, COSTS OFF)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1"
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
ctid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4265,12 +4067,10 @@ INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
Remote SQL: INSERT INTO "S 1"."T 1"("C 1", c2, c3, c4, c5, c6, c7, c8) VALUES ($1, $2, $3, $4, $5, $6, $7, $8)
-> Subquery Scan on "*SELECT*"
Output: "*SELECT*"."?column?", "*SELECT*"."?column?_1", NULL::integer, "*SELECT*"."?column?_2", NULL::timestamp with time zone, NULL::timestamp without time zone, NULL::character varying, 'ft2 '::character(10), NULL::user_enum
- -> Limit
- Output: ((ft2_1.c1 + 1000)), ((ft2_1.c2 + 100)), ((ft2_1.c3 || ft2_1.c3))
- -> Foreign Scan on public.ft2 ft2_1
- Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
- Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1"
-(9 rows)
+ -> Foreign Scan on public.ft2 ft2_1
+ Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
+ Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1" LIMIT 20::bigint
+(7 rows)
INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
INSERT INTO ft2 (c1,c2,c3)
@@ -6017,14 +5817,12 @@ VACUUM ANALYZE "S 1"."T 1";
-- FIRST behavior here.
-- ORDER BY DESC NULLS LAST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 795::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6043,14 +5841,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
-- ORDER BY DESC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6069,14 +5865,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
-- ORDER BY ASC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
diff --git a/contrib/postgres_fdw/postgres_fdw.c b/contrib/postgres_fdw/postgres_fdw.c
index 6174d2be07..a4d35fb789 100644
--- a/contrib/postgres_fdw/postgres_fdw.c
+++ b/contrib/postgres_fdw/postgres_fdw.c
@@ -251,11 +251,14 @@ typedef struct PgFdwAnalyzeState
* We store:
*
* 1) Boolean flag showing if the remote query has the final sort
+ * 2) Boolean flag showing if the remote query has the LIMIT clause
*/
enum FdwPathPrivateIndex
{
/* has-final-sort flag (as an integer Value node) */
- FdwPathPrivateHasFinalSort
+ FdwPathPrivateHasFinalSort,
+ /* has-limit flag (as an integer Value node) */
+ FdwPathPrivateHasLimit
};
/* Struct for extra information passed to estimate_path_cost_size */
@@ -263,6 +266,10 @@ typedef struct
{
PathTarget *target;
bool has_final_sort;
+ bool has_limit;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
} PgFdwPathExtraData;
/*
@@ -400,6 +407,7 @@ static void adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost);
static bool ec_member_matches_foreign(PlannerInfo *root, RelOptInfo *rel,
@@ -481,6 +489,10 @@ static void add_foreign_grouping_paths(PlannerInfo *root,
static void add_foreign_ordered_paths(PlannerInfo *root,
RelOptInfo *input_rel,
RelOptInfo *ordered_rel);
+static void add_foreign_final_paths(PlannerInfo *root,
+ RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra);
static void apply_server_options(PgFdwRelationInfo *fpinfo);
static void apply_table_options(PgFdwRelationInfo *fpinfo);
static void merge_fdw_options(PgFdwRelationInfo *fpinfo,
@@ -1179,14 +1191,19 @@ postgresGetForeignPlan(PlannerInfo *root,
List *retrieved_attrs;
StringInfoData sql;
bool has_final_sort = false;
+ bool has_limit = false;
ListCell *lc;
/*
* Get private info created by postgresGetForeignUpperPaths, if any.
*/
if (best_path->fdw_private)
+ {
has_final_sort = intVal(list_nth(best_path->fdw_private,
FdwPathPrivateHasFinalSort));
+ has_limit = intVal(list_nth(best_path->fdw_private,
+ FdwPathPrivateHasLimit));
+ }
if (IS_SIMPLE_REL(foreignrel))
{
@@ -1336,7 +1353,7 @@ postgresGetForeignPlan(PlannerInfo *root,
initStringInfo(&sql);
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_exprs, best_path->path.pathkeys,
- has_final_sort, false,
+ has_final_sort, has_limit, false,
&retrieved_attrs, ¶ms_list);
/* Remember remote_exprs for possible use by postgresPlanDirectModify */
@@ -2597,6 +2614,7 @@ estimate_path_cost_size(PlannerInfo *root,
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_conds, pathkeys,
fpextra ? fpextra->has_final_sort : false,
+ fpextra ? fpextra->has_limit : false,
false, &retrieved_attrs, NULL);
/* Get the remote estimate */
@@ -2875,11 +2893,11 @@ estimate_path_cost_size(PlannerInfo *root,
}
/*
- * If this is an UPPERREL_ORDERED step performed on the final
- * scan/join relation, the costs obtained from the cache wouldn't yet
- * contain the eval costs for the final scan/join target, which would
- * have been updated by apply_scanjoin_target_to_paths(); add the eval
- * costs now.
+ * If this is UPPERREL_ORDERED and/or UPPERREL_FINAL steps performed
+ * on the final scan/join relation, the costs obtained from the cache
+ * wouldn't yet contain the eval costs for the final scan/join target,
+ * which would have been updated by apply_scanjoin_target_to_paths();
+ * add the eval costs now.
*/
if (fpextra && !IS_UPPER_REL(foreignrel))
{
@@ -2907,6 +2925,7 @@ estimate_path_cost_size(PlannerInfo *root,
Assert(fpinfo->stage == UPPERREL_GROUP_AGG);
adjust_foreign_grouping_path_cost(root, pathkeys,
retrieved_rows, width,
+ fpextra->limit_tuples,
&startup_cost, &run_cost);
}
else
@@ -2917,6 +2936,14 @@ estimate_path_cost_size(PlannerInfo *root,
}
total_cost = startup_cost + run_cost;
+
+ /* Adjust the cost estimates if we have LIMIT */
+ if (fpextra && fpextra->has_limit)
+ {
+ adjust_limit_rows_costs(&rows, &startup_cost, &total_cost,
+ fpextra->offset_est, fpextra->count_est);
+ retrieved_rows = rows;
+ }
}
/*
@@ -2925,7 +2952,8 @@ estimate_path_cost_size(PlannerInfo *root,
* have different expressions from the underlying relation's reltarget
* (see make_sort_input_target()); adjust tlist eval costs.
*/
- if (fpextra && fpextra->target != foreignrel->reltarget)
+ if (fpextra && fpextra->has_final_sort &&
+ fpextra->target != foreignrel->reltarget)
{
QualCost oldcost = foreignrel->reltarget->cost;
QualCost newcost = fpextra->target->cost;
@@ -2961,6 +2989,24 @@ estimate_path_cost_size(PlannerInfo *root,
total_cost += fpinfo->fdw_tuple_cost * retrieved_rows;
total_cost += cpu_tuple_cost * retrieved_rows;
+ /*
+ * If we have LIMIT, we should perfer applying the restriction remotely
+ * rather than locally, as the former avoids extra row fetches from the
+ * remote that the latter might cause. But since we don't account for
+ * such row fetches when estimating the costs of the local restriction in
+ * create_limit_path(), there would be no difference between their costs;
+ * instead, tweak the costs estimated above to ensure we'll prefer the
+ * remote restriction if the LIMIT is a useful one.
+ */
+ if (fpextra && fpextra->has_limit &&
+ fpextra->limit_tuples > 0 &&
+ fpextra->limit_tuples < fpinfo->rows)
+ {
+ Assert(fpinfo->rows > 0);
+ total_cost -= (total_cost - startup_cost) * 0.05 *
+ (fpinfo->rows - fpextra->limit_tuples) / fpinfo->rows;
+ }
+
/* Return results. */
*p_rows = rows;
*p_width = width;
@@ -3027,6 +3073,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost)
{
@@ -3051,7 +3098,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
width,
0.0,
work_mem,
- -1.0);
+ limit_tuples);
*p_startup_cost = sort_path.startup_cost;
*p_run_cost = sort_path.total_cost - sort_path.startup_cost;
@@ -5587,7 +5634,8 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
/* Ignore stages we don't support; and skip any duplicate calls. */
if ((stage != UPPERREL_GROUP_AGG &&
- stage != UPPERREL_ORDERED) ||
+ stage != UPPERREL_ORDERED &&
+ stage != UPPERREL_FINAL) ||
output_rel->fdw_private)
return;
@@ -5605,6 +5653,10 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
case UPPERREL_ORDERED:
add_foreign_ordered_paths(root, input_rel, output_rel);
break;
+ case UPPERREL_FINAL:
+ add_foreign_final_paths(root, input_rel, output_rel,
+ (FinalPathExtraData *) extra);
+ break;
default:
elog(ERROR, "unexpected upper relation: %d", (int) stage);
break;
@@ -5803,7 +5855,7 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
* Build the fdw_private list that will be used by postgresGetForeignPlan.
* Items in the list must match order in enum FdwPathPrivateIndex.
*/
- fdw_private = list_make1(makeInteger(true));
+ fdw_private = list_make2(makeInteger(true), makeInteger(false));
/* Create foreign ordering ForeignPath */
ordered_path = create_foreign_upper_path(root,
@@ -5820,6 +5872,135 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
add_path(ordered_rel, (Path *) ordered_path);
}
+/*
+ * add_foreign_final_paths
+ * Add foreign paths for performing the final processing step remotely.
+ *
+ * Given input_rel contains the source-data Paths. The paths are added to the
+ * given final_rel.
+ */
+static void
+add_foreign_final_paths(PlannerInfo *root, RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra)
+{
+ Query *parse = root->parse;
+ PgFdwRelationInfo *ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ PgFdwRelationInfo *fpinfo = (PgFdwRelationInfo *) final_rel->fdw_private;
+ bool has_final_sort = false;
+ List *pathkeys = NIL;
+ PgFdwPathExtraData *fpextra;
+ double rows;
+ int width;
+ Cost startup_cost;
+ Cost total_cost;
+ List *fdw_private;
+ ForeignPath *final_path;
+
+ /*
+ * Currently, we only support this for SELECT commands
+ */
+ if (parse->commandType != CMD_SELECT)
+ return;
+
+ /*
+ * No work if there is no FOR UPDATE/SHARE clause and if there is no need
+ * to add a LIMIT node
+ */
+ if (!parse->rowMarks && !extra->limit_needed)
+ return;
+
+ /* Save the input_rel as outerrel in fpinfo */
+ fpinfo->outerrel = input_rel;
+
+ /*
+ * Copy foreign table, foreign server, user mapping, FDW options etc.
+ * details from the input relation's fpinfo.
+ */
+ fpinfo->table = ifpinfo->table;
+ fpinfo->server = ifpinfo->server;
+ fpinfo->user = ifpinfo->user;
+ merge_fdw_options(fpinfo, ifpinfo, NULL);
+
+ /*
+ * For now we don't support cases where there are any SRFs in the tlist
+ */
+ if (parse->hasTargetSRFs)
+ return;
+
+ /*
+ * If the input_rel is the ordered rel, replace it with its underlying
+ * scan, join, or grouping rel
+ */
+ if (IS_UPPER_REL(input_rel) && ifpinfo->stage == UPPERREL_ORDERED)
+ {
+ input_rel = ifpinfo->outerrel;
+ ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ has_final_sort = true;
+ pathkeys = root->sort_pathkeys;
+ }
+
+ /* Assess if it is safe to push down the LIMIT and OFFSET, if any */
+ if (extra->limit_needed)
+ {
+ /*
+ * If the underlying rel has any local conditions, the LIMIT/OFFSET
+ * cannot be pushed down.
+ */
+ if (ifpinfo->local_conds)
+ return;
+
+ /*
+ * Also, the LIMIT/OFFSET cannot be pushed down if their expressions
+ * are not safe to remote.
+ */
+ if (!is_foreign_expr(root, input_rel, (Expr *) parse->limitOffset) ||
+ !is_foreign_expr(root, input_rel, (Expr *) parse->limitCount))
+ return;
+ }
+
+ /* Safe to pushdown */
+ fpinfo->pushdown_safe = true;
+
+ /* Initialize the selectivity and cost of the local_conds */
+ fpinfo->local_conds_sel = 1.0;
+ fpinfo->local_conds_cost.startup = 0.0;
+ fpinfo->local_conds_cost.per_tuple = 0.0;
+
+ fpextra = (PgFdwPathExtraData *) palloc0(sizeof(PgFdwPathExtraData));
+ fpextra->target = root->upper_targets[UPPERREL_FINAL];
+ fpextra->has_final_sort = has_final_sort;
+ fpextra->has_limit = extra->limit_needed;
+ fpextra->limit_tuples = extra->limit_tuples;
+ fpextra->count_est = extra->count_est;
+ fpextra->offset_est = extra->offset_est;
+
+ /* Estimate the cost of performing the final steps remotely */
+ estimate_path_cost_size(root, input_rel, NIL, pathkeys, fpextra,
+ &rows, &width, &startup_cost, &total_cost);
+
+ /*
+ * Build the fdw_private list that will be used by postgresGetForeignPlan.
+ * Items in the list must match order in enum FdwPathPrivateIndex.
+ */
+ fdw_private = list_make2(makeInteger(has_final_sort),
+ makeInteger(extra->limit_needed));
+
+ /* Create foreign final ForeignPath */
+ final_path = create_foreign_upper_path(root,
+ input_rel,
+ root->upper_targets[UPPERREL_FINAL],
+ rows,
+ startup_cost,
+ total_cost,
+ pathkeys,
+ NULL, /* no extra plan */
+ fdw_private);
+
+ /* And add it to the final rel */
+ add_path(final_rel, (Path *) final_path);
+}
+
/*
* Create a tuple from the specified row of the PGresult.
*
diff --git a/contrib/postgres_fdw/postgres_fdw.h b/contrib/postgres_fdw/postgres_fdw.h
index 936e3f498a..aabe1dec68 100644
--- a/contrib/postgres_fdw/postgres_fdw.h
+++ b/contrib/postgres_fdw/postgres_fdw.h
@@ -185,7 +185,8 @@ extern List *build_tlist_to_deparse(RelOptInfo *foreignrel);
extern void deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, List *tlist,
List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit,
+ bool is_subquery,
List **retrieved_attrs, List **params_list);
extern const char *get_jointype_name(JoinType jointype);
diff --git a/contrib/postgres_fdw/sql/postgres_fdw.sql b/contrib/postgres_fdw/sql/postgres_fdw.sql
index eb9d1ad59d..621d4faa1f 100644
--- a/contrib/postgres_fdw/sql/postgres_fdw.sql
+++ b/contrib/postgres_fdw/sql/postgres_fdw.sql
@@ -506,7 +506,7 @@ SELECT t1.c1 FROM ft1 t1 WHERE EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c1)
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
diff --git a/src/backend/optimizer/plan/planner.c b/src/backend/optimizer/plan/planner.c
index 1a58d733fa..fe5272ecf9 100644
--- a/src/backend/optimizer/plan/planner.c
+++ b/src/backend/optimizer/plan/planner.c
@@ -1688,6 +1688,7 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
bool final_target_parallel_safe;
RelOptInfo *current_rel;
RelOptInfo *final_rel;
+ FinalPathExtraData extra;
ListCell *lc;
/* Tweak caller-supplied tuple_fraction if have LIMIT/OFFSET */
@@ -2254,6 +2255,11 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
}
}
+ extra.limit_needed = limit_needed(parse);
+ extra.limit_tuples = limit_tuples;
+ extra.count_est = count_est;
+ extra.offset_est = offset_est;
+
/*
* If there is an FDW that's responsible for all baserels of the query,
* let it consider adding ForeignPaths.
@@ -2262,12 +2268,12 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
final_rel->fdwroutine->GetForeignUpperPaths)
final_rel->fdwroutine->GetForeignUpperPaths(root, UPPERREL_FINAL,
current_rel, final_rel,
- NULL);
+ &extra);
/* Let extensions possibly add some more paths */
if (create_upper_paths_hook)
(*create_upper_paths_hook) (root, UPPERREL_FINAL,
- current_rel, final_rel, NULL);
+ current_rel, final_rel, &extra);
/* Note: currently, we leave it to callers to do set_cheapest() */
}
diff --git a/src/include/nodes/pathnodes.h b/src/include/nodes/pathnodes.h
index a008ae07da..d01c152b9f 100644
--- a/src/include/nodes/pathnodes.h
+++ b/src/include/nodes/pathnodes.h
@@ -2429,6 +2429,23 @@ typedef struct
PartitionwiseAggregateType patype;
} GroupPathExtraData;
+/*
+ * Struct for extra information passed to subroutines of grouping_planner
+ *
+ * limit_needed is true if we actually need a Limit plan node
+ * limit_tuples is the max number of tuples needed according to LIMIT/OFFSET
+ * count_est and offset_est are the values of the LIMIT/OFFSET clauses
+ * estimated by preprocess_limit() in planner.c (see comments for
+ * preprocess_limit() for more information).
+ */
+typedef struct
+{
+ bool limit_needed;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
+} FinalPathExtraData;
+
/*
* For speed reasons, cost estimation for join paths is performed in two
* phases: the first phase tries to quickly derive a lower bound for the
--
2.19.2
--------------080909020109040609010109--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely
@ 2019-03-06 11:37 Etsuro Fujita <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Etsuro Fujita @ 2019-03-06 11:37 UTC (permalink / raw)
---
contrib/postgres_fdw/deparse.c | 37 +-
.../postgres_fdw/expected/postgres_fdw.out | 618 ++++++------------
contrib/postgres_fdw/postgres_fdw.c | 290 +++++++-
contrib/postgres_fdw/postgres_fdw.h | 3 +-
contrib/postgres_fdw/sql/postgres_fdw.sql | 2 +-
src/backend/optimizer/plan/planner.c | 10 +-
src/include/nodes/pathnodes.h | 17 +
7 files changed, 540 insertions(+), 437 deletions(-)
diff --git a/contrib/postgres_fdw/deparse.c b/contrib/postgres_fdw/deparse.c
index 97dd07bee8..079406f4f3 100644
--- a/contrib/postgres_fdw/deparse.c
+++ b/contrib/postgres_fdw/deparse.c
@@ -169,6 +169,7 @@ static void deparseSelectSql(List *tlist, bool is_subquery, List **retrieved_att
static void deparseLockingClause(deparse_expr_cxt *context);
static void appendOrderByClause(List *pathkeys, bool has_final_sort,
deparse_expr_cxt *context);
+static void appendLimitClause(deparse_expr_cxt *context);
static void appendConditions(List *exprs, deparse_expr_cxt *context);
static void deparseFromExprForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, bool use_alias,
@@ -930,7 +931,7 @@ build_tlist_to_deparse(RelOptInfo *foreignrel)
void
deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
List *tlist, List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit, bool is_subquery,
List **retrieved_attrs, List **params_list)
{
deparse_expr_cxt context;
@@ -988,6 +989,10 @@ deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
if (pathkeys)
appendOrderByClause(pathkeys, has_final_sort, &context);
+ /* Add LIMIT clause if necessary */
+ if (has_limit)
+ appendLimitClause(&context);
+
/* Add any necessary FOR UPDATE/SHARE. */
deparseLockingClause(&context);
}
@@ -1591,7 +1596,8 @@ deparseRangeTblRef(StringInfo buf, PlannerInfo *root, RelOptInfo *foreignrel,
/* Deparse the subquery representing the relation. */
appendStringInfoChar(buf, '(');
deparseSelectStmtForRel(buf, root, foreignrel, NIL,
- fpinfo->remote_conds, NIL, false, true,
+ fpinfo->remote_conds, NIL,
+ false, false, true,
&retrieved_attrs, params_list);
appendStringInfoChar(buf, ')');
@@ -3160,6 +3166,33 @@ appendOrderByClause(List *pathkeys, bool has_final_sort,
reset_transmission_modes(nestlevel);
}
+/*
+ * Deparse LIMIT/OFFSET clause.
+ */
+static void
+appendLimitClause(deparse_expr_cxt *context)
+{
+ PlannerInfo *root = context->root;
+ StringInfo buf = context->buf;
+ int nestlevel;
+
+ /* Make sure any constants in the exprs are printed portably */
+ nestlevel = set_transmission_modes();
+
+ if (root->parse->limitCount)
+ {
+ appendStringInfoString(buf, " LIMIT ");
+ deparseExpr((Expr *) root->parse->limitCount, context);
+ }
+ if (root->parse->limitOffset)
+ {
+ appendStringInfoString(buf, " OFFSET ");
+ deparseExpr((Expr *) root->parse->limitOffset, context);
+ }
+
+ reset_transmission_modes(nestlevel);
+}
+
/*
* appendFunctionName
* Deparses function name from given function oid.
diff --git a/contrib/postgres_fdw/expected/postgres_fdw.out b/contrib/postgres_fdw/expected/postgres_fdw.out
index 592dd3e05f..4759f58193 100644
--- a/contrib/postgres_fdw/expected/postgres_fdw.out
+++ b/contrib/postgres_fdw/expected/postgres_fdw.out
@@ -236,11 +236,10 @@ ALTER FOREIGN TABLE ft2 OPTIONS (use_remote_estimate 'true');
-- ===================================================================
-- single table without alias
EXPLAIN (COSTS OFF) SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------
- Limit
- -> Foreign Scan on ft1
-(2 rows)
+ QUERY PLAN
+---------------------
+ Foreign Scan on ft1
+(1 row)
SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -288,14 +287,12 @@ SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
-- whole-row reference
EXPLAIN (VERBOSE, COSTS OFF) SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: t1.*, c3, c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.*, c3, c1
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(3 rows)
SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
t1
@@ -335,14 +332,12 @@ SELECT * FROM ft1 t1 WHERE t1.c1 = 101 AND t1.c6 = '1' AND t1.c7 >= '1';
-- with FOR UPDATE/SHARE
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -351,14 +346,12 @@ SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
(1 row)
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -1015,15 +1008,13 @@ ANALYZE ft5;
-- join two tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -1043,18 +1034,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3, t1.c3
- -> Sort
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Sort Key: t1.c3, t1.c1
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1))))
-(9 rows)
+ Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1)))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1074,15 +1060,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t
-- left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1102,15 +1086,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.
-- left outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1175,15 +1157,13 @@ SELECT t1.c1, t1.c2, t2.c1, t2.c2 FROM ft4 t1 LEFT JOIN (SELECT * FROM ft5 WHERE
-- right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1203,15 +1183,13 @@ SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2
-- right outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1231,15 +1209,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGH
-- full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 45::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
c1 | c1
@@ -1283,15 +1259,13 @@ SELECT t1.c1, t2.c1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL
EXPLAIN (VERBOSE, COSTS OFF)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: 1
- -> Foreign Scan
- Output: 1
- Relations: (public.ft4) FULL JOIN (public.ft5)
- Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE))
-(6 rows)
+ Relations: (public.ft4) FULL JOIN (public.ft5)
+ Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
?column?
@@ -1406,15 +1380,13 @@ SELECT t1.c1, ss.a, ss.b FROM (SELECT c1 FROM "S 1"."T 3" WHERE c1 = 50) t1 INNE
-- full outer join + inner join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t3.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1, t3.c1
- Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST
-(6 rows)
+ Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST LIMIT 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
c1 | c1 | c1
@@ -1434,15 +1406,13 @@ SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 a
-- full outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1462,15 +1432,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1490,15 +1458,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT
-- right outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1518,15 +1484,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1546,15 +1510,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1574,15 +1536,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- right outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1602,15 +1562,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1661,15 +1619,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) WHERE (t1.c1
-- full outer join + WHERE clause with shippable extensions set
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t1.c3 FROM ft1 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) WHERE postgres_fdw_abs(t1.c1) > 0 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t1.c3
- Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0))
-(6 rows)
+ Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
ALTER SERVER loopback OPTIONS (DROP extensions);
-- full outer join + WHERE clause with shippable extensions not set
@@ -1691,37 +1647,13 @@ ALTER SERVER loopback OPTIONS (ADD extensions 'postgres_fdw');
-- tests whole-row reference for row marks
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
c1 | c1
@@ -1740,37 +1672,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1 FOR UPDATE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1 FOR UPDATE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
c1 | c1
@@ -1790,37 +1698,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join two tables with FOR SHARE clause
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
c1 | c1
@@ -1839,37 +1723,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1 FOR SHARE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1 FOR SHARE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
c1 | c1
@@ -1923,15 +1783,13 @@ WITH t (c1_1, c1_3, c2_1) AS MATERIALIZED (SELECT t1.c1, t1.c3, t2.c1 FROM ft1 t
-- ctid with whole-row reference
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.ctid, t1, t2, t1.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- -> Foreign Scan
- Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
-- SEMI JOIN, not pushed down
EXPLAIN (VERBOSE, COSTS OFF)
@@ -1999,27 +1857,16 @@ SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2
119
(10 rows)
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Sort
- Output: t1.c1, t2.c1
- Sort Key: t1.c1, t2.c1
- -> Nested Loop
- Output: t1.c1, t2.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
- -> Materialize
- Output: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
-(15 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1" FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (TRUE)) ORDER BY r1."C 1" ASC NULLS LAST, r2."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -2341,50 +2188,13 @@ SET enable_hashjoin TO false;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- -> Foreign Scan
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
- Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Merge Cond: (ft1.c2 = ft5.c1)
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*, ft4.c1, ft4.c2, ft4.c3, ft4.*
- Merge Cond: (ft1.c2 = ft4.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Sort Key: ft1.c2
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Merge Cond: (ft1.c1 = ft2.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Sort Key: ft1.c1
- -> Foreign Scan on public.ft1
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) FOR UPDATE
- -> Materialize
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- -> Foreign Scan on public.ft2
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) ORDER BY "C 1" ASC NULLS LAST FOR UPDATE
- -> Sort
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Sort Key: ft4.c1
- -> Foreign Scan on public.ft4
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 3" FOR UPDATE
- -> Sort
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Sort Key: ft5.c1
- -> Foreign Scan on public.ft5
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 4" FOR UPDATE
-(41 rows)
+ Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
+ Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
+(4 rows)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
@@ -2452,15 +2262,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1
ALTER VIEW v4 OWNER TO regress_view_owner;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, ft5.c2, ft5.c1
- -> Foreign Scan
- Output: ft4.c1, ft5.c2, ft5.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5)
- Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5)
+ Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2517,15 +2325,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c
ALTER VIEW v4 OWNER TO CURRENT_USER;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, t2.c2, t2.c1
- -> Foreign Scan
- Output: ft4.c1, t2.c2, t2.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -4024,14 +3830,12 @@ SELECT * FROM ft1 t1 WHERE t1.tableoid = 'ft1'::regclass LIMIT 1;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------
- Limit
- Output: ((tableoid)::regclass), c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(5 rows)
+ QUERY PLAN
+-----------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
tableoid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4056,14 +3860,12 @@ SELECT * FROM ft1 t1 WHERE t1.ctid = '(0,2)';
EXPLAIN (VERBOSE, COSTS OFF)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1"
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
ctid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4268,12 +4070,10 @@ INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
Remote SQL: INSERT INTO "S 1"."T 1"("C 1", c2, c3, c4, c5, c6, c7, c8) VALUES ($1, $2, $3, $4, $5, $6, $7, $8)
-> Subquery Scan on "*SELECT*"
Output: "*SELECT*"."?column?", "*SELECT*"."?column?_1", NULL::integer, "*SELECT*"."?column?_2", NULL::timestamp with time zone, NULL::timestamp without time zone, NULL::character varying, 'ft2 '::character(10), NULL::user_enum
- -> Limit
- Output: ((ft2_1.c1 + 1000)), ((ft2_1.c2 + 100)), ((ft2_1.c3 || ft2_1.c3))
- -> Foreign Scan on public.ft2 ft2_1
- Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
- Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1"
-(9 rows)
+ -> Foreign Scan on public.ft2 ft2_1
+ Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
+ Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1" LIMIT 20::bigint
+(7 rows)
INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
INSERT INTO ft2 (c1,c2,c3)
@@ -6020,14 +5820,12 @@ VACUUM ANALYZE "S 1"."T 1";
-- FIRST behavior here.
-- ORDER BY DESC NULLS LAST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 795::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6046,14 +5844,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
-- ORDER BY DESC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6072,14 +5868,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
-- ORDER BY ASC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
diff --git a/contrib/postgres_fdw/postgres_fdw.c b/contrib/postgres_fdw/postgres_fdw.c
index e22472ae3d..405f1f0c92 100644
--- a/contrib/postgres_fdw/postgres_fdw.c
+++ b/contrib/postgres_fdw/postgres_fdw.c
@@ -251,11 +251,14 @@ typedef struct PgFdwAnalyzeState
* We store:
*
* 1) Boolean flag showing if the remote query has the final sort
+ * 2) Boolean flag showing if the remote query has the LIMIT clause
*/
enum FdwPathPrivateIndex
{
/* has-final-sort flag (as an integer Value node) */
- FdwPathPrivateHasFinalSort
+ FdwPathPrivateHasFinalSort,
+ /* has-limit flag (as an integer Value node) */
+ FdwPathPrivateHasLimit
};
/* Struct for extra information passed to estimate_path_cost_size */
@@ -263,6 +266,10 @@ typedef struct
{
PathTarget *target;
bool has_final_sort;
+ bool has_limit;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
} PgFdwPathExtraData;
/*
@@ -400,6 +407,7 @@ static void adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost);
static bool ec_member_matches_foreign(PlannerInfo *root, RelOptInfo *rel,
@@ -481,6 +489,10 @@ static void add_foreign_grouping_paths(PlannerInfo *root,
static void add_foreign_ordered_paths(PlannerInfo *root,
RelOptInfo *input_rel,
RelOptInfo *ordered_rel);
+static void add_foreign_final_paths(PlannerInfo *root,
+ RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra);
static void apply_server_options(PgFdwRelationInfo *fpinfo);
static void apply_table_options(PgFdwRelationInfo *fpinfo);
static void merge_fdw_options(PgFdwRelationInfo *fpinfo,
@@ -1179,14 +1191,19 @@ postgresGetForeignPlan(PlannerInfo *root,
List *retrieved_attrs;
StringInfoData sql;
bool has_final_sort = false;
+ bool has_limit = false;
ListCell *lc;
/*
* Get private info created by postgresGetForeignUpperPaths, if any.
*/
if (best_path->fdw_private)
+ {
has_final_sort = intVal(list_nth(best_path->fdw_private,
FdwPathPrivateHasFinalSort));
+ has_limit = intVal(list_nth(best_path->fdw_private,
+ FdwPathPrivateHasLimit));
+ }
if (IS_SIMPLE_REL(foreignrel))
{
@@ -1336,7 +1353,7 @@ postgresGetForeignPlan(PlannerInfo *root,
initStringInfo(&sql);
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_exprs, best_path->path.pathkeys,
- has_final_sort, false,
+ has_final_sort, has_limit, false,
&retrieved_attrs, ¶ms_list);
/* Remember remote_exprs for possible use by postgresPlanDirectModify */
@@ -2598,6 +2615,7 @@ estimate_path_cost_size(PlannerInfo *root,
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_conds, pathkeys,
fpextra ? fpextra->has_final_sort : false,
+ fpextra ? fpextra->has_limit : false,
false, &retrieved_attrs, NULL);
/* Get the remote estimate */
@@ -2677,10 +2695,11 @@ estimate_path_cost_size(PlannerInfo *root,
/*
* When we're called from postgresGetForeignUpperPaths() with the
- * UPPERREL_ORDERED stage, the costs of scanning a foreign base or
- * join relation obtained from the cache wouldn't yet contain the
- * eval costs for the final scan/join target that has been updated
- * by apply_scanjoin_target_to_paths(); add the eval costs now.
+ * UPPERREL_ORDERED or UPPERREL_FINAL stage, the costs of scanning
+ * a foreign base or join relation obtained from the cache would
+ * not yet contain the eval costs for the final scan/join target
+ * that has been updated by apply_scanjoin_target_to_paths(); add
+ * the eval costs now.
*/
if (fpextra && !IS_UPPER_REL(foreignrel))
{
@@ -2904,6 +2923,7 @@ estimate_path_cost_size(PlannerInfo *root,
Assert(fpinfo->stage == UPPERREL_GROUP_AGG);
adjust_foreign_grouping_path_cost(root, pathkeys,
retrieved_rows, width,
+ fpextra->limit_tuples,
&startup_cost, &run_cost);
}
else
@@ -2914,15 +2934,24 @@ estimate_path_cost_size(PlannerInfo *root,
}
total_cost = startup_cost + run_cost;
+
+ /* Adjust the cost estimates if we have LIMIT */
+ if (fpextra && fpextra->has_limit)
+ {
+ adjust_limit_rows_costs(&rows, &startup_cost, &total_cost,
+ fpextra->offset_est, fpextra->count_est);
+ retrieved_rows = rows;
+ }
}
/*
* If this includes an UPPERREL_ORDERED step, the given target, which will
* be applied to the resulting path, might have different expressions from
- * the relation's reltarget (see make_sort_input_target()); adjust tlist
+ * the foreignrel's reltarget (see make_sort_input_target()); adjust tlist
* eval costs.
*/
- if (fpextra && fpextra->target != foreignrel->reltarget)
+ if (fpextra && fpextra->has_final_sort &&
+ fpextra->target != foreignrel->reltarget)
{
QualCost oldcost = foreignrel->reltarget->cost;
QualCost newcost = fpextra->target->cost;
@@ -2958,6 +2987,26 @@ estimate_path_cost_size(PlannerInfo *root,
total_cost += fpinfo->fdw_tuple_cost * retrieved_rows;
total_cost += cpu_tuple_cost * retrieved_rows;
+ /*
+ * If we have LIMIT, we should perfer performing the restriction remotely
+ * rather than locally, as the former avoids extra row fetches from the
+ * remote that the latter might cause. But since the core code doesn't
+ * account for such fetches when estimating the costs of the local
+ * restriction (see create_limit_path()), there would be no difference
+ * between the costs of the local restriction and the costs of the remote
+ * restriction estimated above; tweak the costs of the remote restriction
+ * to ensure we'll prefer the remote restriction if the LIMIT is a useful
+ * one.
+ */
+ if (fpextra && fpextra->has_limit &&
+ fpextra->limit_tuples > 0 &&
+ fpextra->limit_tuples < fpinfo->rows)
+ {
+ Assert(fpinfo->rows > 0);
+ total_cost -= (total_cost - startup_cost) * 0.05 *
+ (fpinfo->rows - fpextra->limit_tuples) / fpinfo->rows;
+ }
+
/* Return results. */
*p_rows = rows;
*p_width = width;
@@ -3024,6 +3073,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost)
{
@@ -3048,7 +3098,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
width,
0.0,
work_mem,
- -1.0);
+ limit_tuples);
*p_startup_cost = sort_path.startup_cost;
*p_run_cost = sort_path.total_cost - sort_path.startup_cost;
@@ -5588,7 +5638,8 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
/* Ignore stages we don't support; and skip any duplicate calls. */
if ((stage != UPPERREL_GROUP_AGG &&
- stage != UPPERREL_ORDERED) ||
+ stage != UPPERREL_ORDERED &&
+ stage != UPPERREL_FINAL) ||
output_rel->fdw_private)
return;
@@ -5606,6 +5657,10 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
case UPPERREL_ORDERED:
add_foreign_ordered_paths(root, input_rel, output_rel);
break;
+ case UPPERREL_FINAL:
+ add_foreign_final_paths(root, input_rel, output_rel,
+ (FinalPathExtraData *) extra);
+ break;
default:
elog(ERROR, "unexpected upper relation: %d", (int) stage);
break;
@@ -5690,6 +5745,9 @@ add_foreign_grouping_paths(PlannerInfo *root, RelOptInfo *input_rel,
grouped_rel->rows = fpinfo->rows;
+ /* Report estimated grouped_rel size to planner. */
+ /* grouped_rel->rows = fpinfo->rows; */
+
/* Create and add foreign path to the grouping relation. */
grouppath = create_foreign_upper_path(root,
grouped_rel,
@@ -5769,14 +5827,6 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
}
}
- /*
- * No point in further work for the case where the input relation is the
- * final scan/join relation because we would have already considered
- * performing the final sort remotely before we get here.
- */
- if (!IS_UPPER_REL(input_rel))
- return;
-
/*
* We try to create paths below by extending paths for the underlying
* base, join, or grouping relation to perform the final sort remotely,
@@ -5833,7 +5883,7 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
* Build the fdw_private list that will be used by postgresGetForeignPlan.
* Items in the list must match order in enum FdwPathPrivateIndex.
*/
- fdw_private = list_make1(makeInteger(true));
+ fdw_private = list_make2(makeInteger(true), makeInteger(false));
/* Create foreign ordering ForeignPath */
ordered_path = create_foreign_upper_path(root,
@@ -5850,6 +5900,208 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
add_path(ordered_rel, (Path *) ordered_path);
}
+/*
+ * add_foreign_final_paths
+ * Add foreign paths for performing the final processing remotely.
+ *
+ * Given input_rel contains the source-data Paths. The paths are added to the
+ * given final_rel.
+ */
+static void
+add_foreign_final_paths(PlannerInfo *root, RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra)
+{
+ Query *parse = root->parse;
+ PgFdwRelationInfo *ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ PgFdwRelationInfo *fpinfo = (PgFdwRelationInfo *) final_rel->fdw_private;
+ bool has_final_sort = false;
+ List *pathkeys = NIL;
+ PgFdwPathExtraData *fpextra;
+ double rows;
+ int width;
+ Cost startup_cost;
+ Cost total_cost;
+ List *fdw_private;
+ ForeignPath *final_path;
+ bool save_use_remote_estimate = false;
+
+ /*
+ * Currently, we only support this for SELECT commands
+ */
+ if (parse->commandType != CMD_SELECT)
+ return;
+
+ /*
+ * No work if there is no FOR UPDATE/SHARE clause and if there is no need
+ * to add a LIMIT node
+ */
+ if (!parse->rowMarks && !extra->limit_needed)
+ return;
+
+ /* Save the input_rel as outerrel in fpinfo */
+ fpinfo->outerrel = input_rel;
+
+ /*
+ * Copy foreign table, foreign server, user mapping, FDW options etc.
+ * details from the input relation's fpinfo.
+ */
+ fpinfo->table = ifpinfo->table;
+ fpinfo->server = ifpinfo->server;
+ fpinfo->user = ifpinfo->user;
+ merge_fdw_options(fpinfo, ifpinfo, NULL);
+
+ /*
+ * If there is no need to add a LIMIT node, we might already have a path
+ * in the input_rel's pathlist that implements all behavior of the query.
+ * Note: we would have already accounted for the query's FOR UPDATE/SHARE
+ * (if any) before we get here.
+ */
+ if (!extra->limit_needed)
+ {
+ ListCell *lc;
+
+ Assert(parse->rowMarks);
+
+ foreach(lc, input_rel->pathlist)
+ {
+ Path *path = (Path *) lfirst(lc);
+
+ /*
+ * apply_scanjoin_target_to_paths() might put ProjectionPath on
+ * top of ForeignPath; look through ProjectionPath and see if the
+ * path underneath it is ForeignPath.
+ */
+ if (IsA(path, ForeignPath) ||
+ (IsA(path, ProjectionPath) &&
+ IsA(((ProjectionPath *) path)->subpath, ForeignPath)))
+ {
+ /* Create foreign final ForeignPath */
+ final_path = create_foreign_upper_path(root,
+ path->parent,
+ path->pathtarget,
+ path->rows,
+ path->startup_cost,
+ path->total_cost,
+ path->pathkeys,
+ NULL, /* no extra plan */
+ NULL);
+
+ /* And add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+ return;
+ }
+ }
+ }
+
+ Assert(extra->limit_needed);
+
+ /*
+ * We try to create paths below by extending paths for the underlying
+ * base, join, or grouping relation to perform the final sort and the
+ * LIMIT restriction remotely if the input_rel is an ordered relation, or
+ * to perform the LIMIT restriction remotely otherwise, which is stored
+ * into the fdw_private list of the resulting paths. (Note that in the
+ * former case we can estimate the effects of the bounded sort by doing
+ * this.)
+ */
+
+ /*
+ * For now we don't support cases where there are any SRFs in the tlist
+ */
+ if (parse->hasTargetSRFs)
+ return;
+
+ /*
+ * If the input_rel is an ordered relation, replace the input_rel with
+ * its underlying base, join, or grouping relation
+ */
+ if (IS_UPPER_REL(input_rel) && ifpinfo->stage == UPPERREL_ORDERED)
+ {
+ input_rel = ifpinfo->outerrel;
+ ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ has_final_sort = true;
+ pathkeys = root->sort_pathkeys;
+ }
+
+ /* Assess if it is safe to push down the LIMIT and OFFSET */
+
+ /*
+ * If the input relation has any local conditions, the LIMIT/OFFSET cannot
+ * be pushed down.
+ */
+ if (ifpinfo->local_conds)
+ return;
+
+ /*
+ * Also, the LIMIT/OFFSET cannot be pushed down, if their expressions are
+ * not safe to remote.
+ */
+ if (!is_foreign_expr(root, input_rel, (Expr *) parse->limitOffset) ||
+ !is_foreign_expr(root, input_rel, (Expr *) parse->limitCount))
+ return;
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ /* Initialize the selectivity and cost of the local_conds */
+ fpinfo->local_conds_sel = 1.0;
+ fpinfo->local_conds_cost.startup = 0.0;
+ fpinfo->local_conds_cost.per_tuple = 0.0;
+
+ fpextra = (PgFdwPathExtraData *) palloc0(sizeof(PgFdwPathExtraData));
+ fpextra->target = root->upper_targets[UPPERREL_FINAL];
+ fpextra->has_final_sort = has_final_sort;
+ fpextra->has_limit = extra->limit_needed;
+ fpextra->limit_tuples = extra->limit_tuples;
+ fpextra->count_est = extra->count_est;
+ fpextra->offset_est = extra->offset_est;
+
+ /*
+ * Estimate the costs of performing the final sort (if any) and the LIMIT
+ * restriction remotely. If there is no final sort, we don't need to
+ * execute EXPLAIN anymore, since the costs of the remote restriction can
+ * be estimated using the cost estimates we already have for the
+ * underlying base, join, or grouping relation, in the same way as we
+ * estimate the costs when use_remote_estimate is false. Since it is
+ * pretty expensive to execute EXPLAIN, force use_remote_estimate to false
+ * in that case.
+ */
+ if (!fpextra->has_final_sort)
+ {
+ save_use_remote_estimate = ifpinfo->use_remote_estimate;
+ ifpinfo->use_remote_estimate = false;
+ }
+ estimate_path_cost_size(root, input_rel, NIL, pathkeys, fpextra,
+ &rows, &width, &startup_cost, &total_cost);
+ if (!fpextra->has_final_sort)
+ ifpinfo->use_remote_estimate = save_use_remote_estimate;
+
+ /*
+ * Build the fdw_private list that will be used by postgresGetForeignPlan.
+ * Items in the list must match order in enum FdwPathPrivateIndex.
+ */
+ fdw_private = list_make2(makeInteger(has_final_sort),
+ makeInteger(extra->limit_needed));
+
+ /* Create foreign final ForeignPath */
+ final_path = create_foreign_upper_path(root,
+ input_rel,
+ root->upper_targets[UPPERREL_FINAL],
+ rows,
+ startup_cost,
+ total_cost,
+ pathkeys,
+ NULL, /* no extra plan */
+ fdw_private);
+
+ /* And add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+}
+
/*
* Create a tuple from the specified row of the PGresult.
*
diff --git a/contrib/postgres_fdw/postgres_fdw.h b/contrib/postgres_fdw/postgres_fdw.h
index 936e3f498a..aabe1dec68 100644
--- a/contrib/postgres_fdw/postgres_fdw.h
+++ b/contrib/postgres_fdw/postgres_fdw.h
@@ -185,7 +185,8 @@ extern List *build_tlist_to_deparse(RelOptInfo *foreignrel);
extern void deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, List *tlist,
List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit,
+ bool is_subquery,
List **retrieved_attrs, List **params_list);
extern const char *get_jointype_name(JoinType jointype);
diff --git a/contrib/postgres_fdw/sql/postgres_fdw.sql b/contrib/postgres_fdw/sql/postgres_fdw.sql
index eb9d1ad59d..621d4faa1f 100644
--- a/contrib/postgres_fdw/sql/postgres_fdw.sql
+++ b/contrib/postgres_fdw/sql/postgres_fdw.sql
@@ -506,7 +506,7 @@ SELECT t1.c1 FROM ft1 t1 WHERE EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c1)
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
diff --git a/src/backend/optimizer/plan/planner.c b/src/backend/optimizer/plan/planner.c
index bc81535905..8deef7dc08 100644
--- a/src/backend/optimizer/plan/planner.c
+++ b/src/backend/optimizer/plan/planner.c
@@ -1709,6 +1709,7 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
bool final_target_parallel_safe;
RelOptInfo *current_rel;
RelOptInfo *final_rel;
+ FinalPathExtraData extra;
ListCell *lc;
/* Tweak caller-supplied tuple_fraction if have LIMIT/OFFSET */
@@ -2275,6 +2276,11 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
}
}
+ extra.limit_needed = limit_needed(parse);
+ extra.limit_tuples = limit_tuples;
+ extra.count_est = count_est;
+ extra.offset_est = offset_est;
+
/*
* If there is an FDW that's responsible for all baserels of the query,
* let it consider adding ForeignPaths.
@@ -2283,12 +2289,12 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
final_rel->fdwroutine->GetForeignUpperPaths)
final_rel->fdwroutine->GetForeignUpperPaths(root, UPPERREL_FINAL,
current_rel, final_rel,
- NULL);
+ &extra);
/* Let extensions possibly add some more paths */
if (create_upper_paths_hook)
(*create_upper_paths_hook) (root, UPPERREL_FINAL,
- current_rel, final_rel, NULL);
+ current_rel, final_rel, &extra);
/* Note: currently, we leave it to callers to do set_cheapest() */
}
diff --git a/src/include/nodes/pathnodes.h b/src/include/nodes/pathnodes.h
index a008ae07da..d01c152b9f 100644
--- a/src/include/nodes/pathnodes.h
+++ b/src/include/nodes/pathnodes.h
@@ -2429,6 +2429,23 @@ typedef struct
PartitionwiseAggregateType patype;
} GroupPathExtraData;
+/*
+ * Struct for extra information passed to subroutines of grouping_planner
+ *
+ * limit_needed is true if we actually need a Limit plan node
+ * limit_tuples is the max number of tuples needed according to LIMIT/OFFSET
+ * count_est and offset_est are the values of the LIMIT/OFFSET clauses
+ * estimated by preprocess_limit() in planner.c (see comments for
+ * preprocess_limit() for more information).
+ */
+typedef struct
+{
+ bool limit_needed;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
+} FinalPathExtraData;
+
/*
* For speed reasons, cost estimation for join paths is performed in two
* phases: the first phase tries to quickly derive a lower bound for the
--
2.19.2
--------------060600050702040203040200--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely
@ 2019-03-07 09:49 Etsuro Fujita <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Etsuro Fujita @ 2019-03-07 09:49 UTC (permalink / raw)
---
contrib/postgres_fdw/deparse.c | 37 +-
.../postgres_fdw/expected/postgres_fdw.out | 618 ++++++------------
contrib/postgres_fdw/postgres_fdw.c | 277 +++++++-
contrib/postgres_fdw/postgres_fdw.h | 3 +-
contrib/postgres_fdw/sql/postgres_fdw.sql | 2 +-
src/backend/optimizer/plan/planner.c | 10 +-
src/include/nodes/pathnodes.h | 17 +
7 files changed, 536 insertions(+), 428 deletions(-)
diff --git a/contrib/postgres_fdw/deparse.c b/contrib/postgres_fdw/deparse.c
index 97dd07bee8..079406f4f3 100644
--- a/contrib/postgres_fdw/deparse.c
+++ b/contrib/postgres_fdw/deparse.c
@@ -169,6 +169,7 @@ static void deparseSelectSql(List *tlist, bool is_subquery, List **retrieved_att
static void deparseLockingClause(deparse_expr_cxt *context);
static void appendOrderByClause(List *pathkeys, bool has_final_sort,
deparse_expr_cxt *context);
+static void appendLimitClause(deparse_expr_cxt *context);
static void appendConditions(List *exprs, deparse_expr_cxt *context);
static void deparseFromExprForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, bool use_alias,
@@ -930,7 +931,7 @@ build_tlist_to_deparse(RelOptInfo *foreignrel)
void
deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
List *tlist, List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit, bool is_subquery,
List **retrieved_attrs, List **params_list)
{
deparse_expr_cxt context;
@@ -988,6 +989,10 @@ deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
if (pathkeys)
appendOrderByClause(pathkeys, has_final_sort, &context);
+ /* Add LIMIT clause if necessary */
+ if (has_limit)
+ appendLimitClause(&context);
+
/* Add any necessary FOR UPDATE/SHARE. */
deparseLockingClause(&context);
}
@@ -1591,7 +1596,8 @@ deparseRangeTblRef(StringInfo buf, PlannerInfo *root, RelOptInfo *foreignrel,
/* Deparse the subquery representing the relation. */
appendStringInfoChar(buf, '(');
deparseSelectStmtForRel(buf, root, foreignrel, NIL,
- fpinfo->remote_conds, NIL, false, true,
+ fpinfo->remote_conds, NIL,
+ false, false, true,
&retrieved_attrs, params_list);
appendStringInfoChar(buf, ')');
@@ -3160,6 +3166,33 @@ appendOrderByClause(List *pathkeys, bool has_final_sort,
reset_transmission_modes(nestlevel);
}
+/*
+ * Deparse LIMIT/OFFSET clause.
+ */
+static void
+appendLimitClause(deparse_expr_cxt *context)
+{
+ PlannerInfo *root = context->root;
+ StringInfo buf = context->buf;
+ int nestlevel;
+
+ /* Make sure any constants in the exprs are printed portably */
+ nestlevel = set_transmission_modes();
+
+ if (root->parse->limitCount)
+ {
+ appendStringInfoString(buf, " LIMIT ");
+ deparseExpr((Expr *) root->parse->limitCount, context);
+ }
+ if (root->parse->limitOffset)
+ {
+ appendStringInfoString(buf, " OFFSET ");
+ deparseExpr((Expr *) root->parse->limitOffset, context);
+ }
+
+ reset_transmission_modes(nestlevel);
+}
+
/*
* appendFunctionName
* Deparses function name from given function oid.
diff --git a/contrib/postgres_fdw/expected/postgres_fdw.out b/contrib/postgres_fdw/expected/postgres_fdw.out
index 592dd3e05f..4759f58193 100644
--- a/contrib/postgres_fdw/expected/postgres_fdw.out
+++ b/contrib/postgres_fdw/expected/postgres_fdw.out
@@ -236,11 +236,10 @@ ALTER FOREIGN TABLE ft2 OPTIONS (use_remote_estimate 'true');
-- ===================================================================
-- single table without alias
EXPLAIN (COSTS OFF) SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------
- Limit
- -> Foreign Scan on ft1
-(2 rows)
+ QUERY PLAN
+---------------------
+ Foreign Scan on ft1
+(1 row)
SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -288,14 +287,12 @@ SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
-- whole-row reference
EXPLAIN (VERBOSE, COSTS OFF) SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: t1.*, c3, c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.*, c3, c1
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(3 rows)
SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
t1
@@ -335,14 +332,12 @@ SELECT * FROM ft1 t1 WHERE t1.c1 = 101 AND t1.c6 = '1' AND t1.c7 >= '1';
-- with FOR UPDATE/SHARE
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -351,14 +346,12 @@ SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
(1 row)
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -1015,15 +1008,13 @@ ANALYZE ft5;
-- join two tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -1043,18 +1034,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3, t1.c3
- -> Sort
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Sort Key: t1.c3, t1.c1
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1))))
-(9 rows)
+ Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1)))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1074,15 +1060,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t
-- left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1102,15 +1086,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.
-- left outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1175,15 +1157,13 @@ SELECT t1.c1, t1.c2, t2.c1, t2.c2 FROM ft4 t1 LEFT JOIN (SELECT * FROM ft5 WHERE
-- right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1203,15 +1183,13 @@ SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2
-- right outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1231,15 +1209,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGH
-- full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 45::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
c1 | c1
@@ -1283,15 +1259,13 @@ SELECT t1.c1, t2.c1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL
EXPLAIN (VERBOSE, COSTS OFF)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: 1
- -> Foreign Scan
- Output: 1
- Relations: (public.ft4) FULL JOIN (public.ft5)
- Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE))
-(6 rows)
+ Relations: (public.ft4) FULL JOIN (public.ft5)
+ Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
?column?
@@ -1406,15 +1380,13 @@ SELECT t1.c1, ss.a, ss.b FROM (SELECT c1 FROM "S 1"."T 3" WHERE c1 = 50) t1 INNE
-- full outer join + inner join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t3.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1, t3.c1
- Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST
-(6 rows)
+ Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST LIMIT 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
c1 | c1 | c1
@@ -1434,15 +1406,13 @@ SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 a
-- full outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1462,15 +1432,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1490,15 +1458,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT
-- right outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1518,15 +1484,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1546,15 +1510,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1574,15 +1536,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- right outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1602,15 +1562,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1661,15 +1619,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) WHERE (t1.c1
-- full outer join + WHERE clause with shippable extensions set
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t1.c3 FROM ft1 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) WHERE postgres_fdw_abs(t1.c1) > 0 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t1.c3
- Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0))
-(6 rows)
+ Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
ALTER SERVER loopback OPTIONS (DROP extensions);
-- full outer join + WHERE clause with shippable extensions not set
@@ -1691,37 +1647,13 @@ ALTER SERVER loopback OPTIONS (ADD extensions 'postgres_fdw');
-- tests whole-row reference for row marks
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
c1 | c1
@@ -1740,37 +1672,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1 FOR UPDATE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1 FOR UPDATE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
c1 | c1
@@ -1790,37 +1698,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join two tables with FOR SHARE clause
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
c1 | c1
@@ -1839,37 +1723,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1 FOR SHARE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1 FOR SHARE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
c1 | c1
@@ -1923,15 +1783,13 @@ WITH t (c1_1, c1_3, c2_1) AS MATERIALIZED (SELECT t1.c1, t1.c3, t2.c1 FROM ft1 t
-- ctid with whole-row reference
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.ctid, t1, t2, t1.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- -> Foreign Scan
- Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
-- SEMI JOIN, not pushed down
EXPLAIN (VERBOSE, COSTS OFF)
@@ -1999,27 +1857,16 @@ SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2
119
(10 rows)
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Sort
- Output: t1.c1, t2.c1
- Sort Key: t1.c1, t2.c1
- -> Nested Loop
- Output: t1.c1, t2.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
- -> Materialize
- Output: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
-(15 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1" FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (TRUE)) ORDER BY r1."C 1" ASC NULLS LAST, r2."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -2341,50 +2188,13 @@ SET enable_hashjoin TO false;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- -> Foreign Scan
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
- Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Merge Cond: (ft1.c2 = ft5.c1)
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*, ft4.c1, ft4.c2, ft4.c3, ft4.*
- Merge Cond: (ft1.c2 = ft4.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Sort Key: ft1.c2
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Merge Cond: (ft1.c1 = ft2.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Sort Key: ft1.c1
- -> Foreign Scan on public.ft1
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) FOR UPDATE
- -> Materialize
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- -> Foreign Scan on public.ft2
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) ORDER BY "C 1" ASC NULLS LAST FOR UPDATE
- -> Sort
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Sort Key: ft4.c1
- -> Foreign Scan on public.ft4
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 3" FOR UPDATE
- -> Sort
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Sort Key: ft5.c1
- -> Foreign Scan on public.ft5
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 4" FOR UPDATE
-(41 rows)
+ Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
+ Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
+(4 rows)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
@@ -2452,15 +2262,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1
ALTER VIEW v4 OWNER TO regress_view_owner;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, ft5.c2, ft5.c1
- -> Foreign Scan
- Output: ft4.c1, ft5.c2, ft5.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5)
- Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5)
+ Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2517,15 +2325,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c
ALTER VIEW v4 OWNER TO CURRENT_USER;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, t2.c2, t2.c1
- -> Foreign Scan
- Output: ft4.c1, t2.c2, t2.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -4024,14 +3830,12 @@ SELECT * FROM ft1 t1 WHERE t1.tableoid = 'ft1'::regclass LIMIT 1;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------
- Limit
- Output: ((tableoid)::regclass), c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(5 rows)
+ QUERY PLAN
+-----------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
tableoid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4056,14 +3860,12 @@ SELECT * FROM ft1 t1 WHERE t1.ctid = '(0,2)';
EXPLAIN (VERBOSE, COSTS OFF)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1"
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
ctid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4268,12 +4070,10 @@ INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
Remote SQL: INSERT INTO "S 1"."T 1"("C 1", c2, c3, c4, c5, c6, c7, c8) VALUES ($1, $2, $3, $4, $5, $6, $7, $8)
-> Subquery Scan on "*SELECT*"
Output: "*SELECT*"."?column?", "*SELECT*"."?column?_1", NULL::integer, "*SELECT*"."?column?_2", NULL::timestamp with time zone, NULL::timestamp without time zone, NULL::character varying, 'ft2 '::character(10), NULL::user_enum
- -> Limit
- Output: ((ft2_1.c1 + 1000)), ((ft2_1.c2 + 100)), ((ft2_1.c3 || ft2_1.c3))
- -> Foreign Scan on public.ft2 ft2_1
- Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
- Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1"
-(9 rows)
+ -> Foreign Scan on public.ft2 ft2_1
+ Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
+ Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1" LIMIT 20::bigint
+(7 rows)
INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
INSERT INTO ft2 (c1,c2,c3)
@@ -6020,14 +5820,12 @@ VACUUM ANALYZE "S 1"."T 1";
-- FIRST behavior here.
-- ORDER BY DESC NULLS LAST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 795::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6046,14 +5844,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
-- ORDER BY DESC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6072,14 +5868,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
-- ORDER BY ASC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
diff --git a/contrib/postgres_fdw/postgres_fdw.c b/contrib/postgres_fdw/postgres_fdw.c
index c342bd319c..3240ffbb11 100644
--- a/contrib/postgres_fdw/postgres_fdw.c
+++ b/contrib/postgres_fdw/postgres_fdw.c
@@ -251,11 +251,14 @@ typedef struct PgFdwAnalyzeState
* We store:
*
* 1) Boolean flag showing if the remote query has the final sort
+ * 2) Boolean flag showing if the remote query has the LIMIT clause
*/
enum FdwPathPrivateIndex
{
/* has-final-sort flag (as an integer Value node) */
- FdwPathPrivateHasFinalSort
+ FdwPathPrivateHasFinalSort,
+ /* has-limit flag (as an integer Value node) */
+ FdwPathPrivateHasLimit
};
/* Struct for extra information passed to estimate_path_cost_size */
@@ -263,6 +266,10 @@ typedef struct
{
PathTarget *target;
bool has_final_sort;
+ bool has_limit;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
} PgFdwPathExtraData;
/*
@@ -400,6 +407,7 @@ static void adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost);
static bool ec_member_matches_foreign(PlannerInfo *root, RelOptInfo *rel,
@@ -481,6 +489,10 @@ static void add_foreign_grouping_paths(PlannerInfo *root,
static void add_foreign_ordered_paths(PlannerInfo *root,
RelOptInfo *input_rel,
RelOptInfo *ordered_rel);
+static void add_foreign_final_paths(PlannerInfo *root,
+ RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra);
static void apply_server_options(PgFdwRelationInfo *fpinfo);
static void apply_table_options(PgFdwRelationInfo *fpinfo);
static void merge_fdw_options(PgFdwRelationInfo *fpinfo,
@@ -1179,14 +1191,19 @@ postgresGetForeignPlan(PlannerInfo *root,
List *retrieved_attrs;
StringInfoData sql;
bool has_final_sort = false;
+ bool has_limit = false;
ListCell *lc;
/*
* Get private info created by postgresGetForeignUpperPaths, if any.
*/
if (best_path->fdw_private)
+ {
has_final_sort = intVal(list_nth(best_path->fdw_private,
FdwPathPrivateHasFinalSort));
+ has_limit = intVal(list_nth(best_path->fdw_private,
+ FdwPathPrivateHasLimit));
+ }
if (IS_SIMPLE_REL(foreignrel))
{
@@ -1336,7 +1353,7 @@ postgresGetForeignPlan(PlannerInfo *root,
initStringInfo(&sql);
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_exprs, best_path->path.pathkeys,
- has_final_sort, false,
+ has_final_sort, has_limit, false,
&retrieved_attrs, ¶ms_list);
/* Remember remote_exprs for possible use by postgresPlanDirectModify */
@@ -2598,6 +2615,7 @@ estimate_path_cost_size(PlannerInfo *root,
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_conds, pathkeys,
fpextra ? fpextra->has_final_sort : false,
+ fpextra ? fpextra->has_limit : false,
false, &retrieved_attrs, NULL);
/* Get the remote estimate */
@@ -2677,10 +2695,11 @@ estimate_path_cost_size(PlannerInfo *root,
/*
* When we're called from postgresGetForeignUpperPaths() with the
- * UPPERREL_ORDERED stage, the costs of scanning a foreign base or
- * join relation obtained from the cache would not yet contain the
- * eval costs for the final scan/join target that has been updated
- * by apply_scanjoin_target_to_paths(); add the eval costs now.
+ * UPPERREL_ORDERED or UPPERREL_FINAL stage, the costs of scanning
+ * a foreign base or join relation obtained from the cache would
+ * not yet contain the eval costs for the final scan/join target
+ * that has been updated by apply_scanjoin_target_to_paths(); add
+ * the eval costs now.
*/
if (fpextra && !IS_UPPER_REL(foreignrel))
{
@@ -2904,6 +2923,7 @@ estimate_path_cost_size(PlannerInfo *root,
Assert(fpinfo->stage == UPPERREL_GROUP_AGG);
adjust_foreign_grouping_path_cost(root, pathkeys,
retrieved_rows, width,
+ fpextra->limit_tuples,
&startup_cost, &run_cost);
}
else
@@ -2914,6 +2934,14 @@ estimate_path_cost_size(PlannerInfo *root,
}
total_cost = startup_cost + run_cost;
+
+ /* Adjust the cost estimates if we have LIMIT */
+ if (fpextra && fpextra->has_limit)
+ {
+ adjust_limit_rows_costs(&rows, &startup_cost, &total_cost,
+ fpextra->offset_est, fpextra->count_est);
+ retrieved_rows = rows;
+ }
}
/*
@@ -2922,7 +2950,8 @@ estimate_path_cost_size(PlannerInfo *root,
* the foreignrel's reltarget (see make_sort_input_target()); adjust tlist
* eval costs.
*/
- if (fpextra && fpextra->target != foreignrel->reltarget)
+ if (fpextra && fpextra->has_final_sort &&
+ fpextra->target != foreignrel->reltarget)
{
QualCost oldcost = foreignrel->reltarget->cost;
QualCost newcost = fpextra->target->cost;
@@ -2958,6 +2987,26 @@ estimate_path_cost_size(PlannerInfo *root,
total_cost += fpinfo->fdw_tuple_cost * retrieved_rows;
total_cost += cpu_tuple_cost * retrieved_rows;
+ /*
+ * If we have LIMIT, we should perfer performing the restriction remotely
+ * rather than locally, as the former avoids extra row fetches from the
+ * remote that the latter might cause. But since the core code doesn't
+ * account for such fetches when estimating the costs of the local
+ * restriction (see create_limit_path()), there would be no difference
+ * between the costs of the local restriction and the costs of the remote
+ * restriction estimated above; tweak the costs of the remote restriction
+ * to ensure we'll prefer the remote restriction if the LIMIT is a useful
+ * one.
+ */
+ if (fpextra && fpextra->has_limit &&
+ fpextra->limit_tuples > 0 &&
+ fpextra->limit_tuples < fpinfo->rows)
+ {
+ Assert(fpinfo->rows > 0);
+ total_cost -= (total_cost - startup_cost) * 0.05 *
+ (fpinfo->rows - fpextra->limit_tuples) / fpinfo->rows;
+ }
+
/* Return results. */
*p_rows = rows;
*p_width = width;
@@ -3024,6 +3073,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost)
{
@@ -3048,7 +3098,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
width,
0.0,
work_mem,
- -1.0);
+ limit_tuples);
*p_startup_cost = sort_path.startup_cost;
*p_run_cost = sort_path.total_cost - sort_path.startup_cost;
@@ -5588,7 +5638,8 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
/* Ignore stages we don't support; and skip any duplicate calls. */
if ((stage != UPPERREL_GROUP_AGG &&
- stage != UPPERREL_ORDERED) ||
+ stage != UPPERREL_ORDERED &&
+ stage != UPPERREL_FINAL) ||
output_rel->fdw_private)
return;
@@ -5606,6 +5657,10 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
case UPPERREL_ORDERED:
add_foreign_ordered_paths(root, input_rel, output_rel);
break;
+ case UPPERREL_FINAL:
+ add_foreign_final_paths(root, input_rel, output_rel,
+ (FinalPathExtraData *) extra);
+ break;
default:
elog(ERROR, "unexpected upper relation: %d", (int) stage);
break;
@@ -5826,7 +5881,7 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
* Build the fdw_private list that will be used by postgresGetForeignPlan.
* Items in the list must match order in enum FdwPathPrivateIndex.
*/
- fdw_private = list_make1(makeInteger(true));
+ fdw_private = list_make2(makeInteger(true), makeInteger(false));
/* Create foreign ordering ForeignPath */
ordered_path = create_foreign_upper_path(root,
@@ -5843,6 +5898,208 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
add_path(ordered_rel, (Path *) ordered_path);
}
+/*
+ * add_foreign_final_paths
+ * Add foreign paths for performing the final processing remotely.
+ *
+ * Given input_rel contains the source-data Paths. The paths are added to the
+ * given final_rel.
+ */
+static void
+add_foreign_final_paths(PlannerInfo *root, RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra)
+{
+ Query *parse = root->parse;
+ PgFdwRelationInfo *ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ PgFdwRelationInfo *fpinfo = (PgFdwRelationInfo *) final_rel->fdw_private;
+ bool has_final_sort = false;
+ List *pathkeys = NIL;
+ PgFdwPathExtraData *fpextra;
+ double rows;
+ int width;
+ Cost startup_cost;
+ Cost total_cost;
+ List *fdw_private;
+ ForeignPath *final_path;
+ bool save_use_remote_estimate = false;
+
+ /*
+ * Currently, we only support this for SELECT commands
+ */
+ if (parse->commandType != CMD_SELECT)
+ return;
+
+ /*
+ * For now we don't support cases where there are any SRFs in the tlist
+ */
+ if (parse->hasTargetSRFs)
+ return;
+
+ /*
+ * No work if there is no FOR UPDATE/SHARE clause and if there is no need
+ * to add a LIMIT node
+ */
+ if (!parse->rowMarks && !extra->limit_needed)
+ return;
+
+ /* Save the input_rel as outerrel in fpinfo */
+ fpinfo->outerrel = input_rel;
+
+ /*
+ * Copy foreign table, foreign server, user mapping, FDW options etc.
+ * details from the input relation's fpinfo.
+ */
+ fpinfo->table = ifpinfo->table;
+ fpinfo->server = ifpinfo->server;
+ fpinfo->user = ifpinfo->user;
+ merge_fdw_options(fpinfo, ifpinfo, NULL);
+
+ /*
+ * If there is no need to add a LIMIT node, there might be a ForeignPath
+ * in the input_rel's pathlist that implements all behavior of the query.
+ * Note: we would have already accounted for the query's FOR UPDATE/SHARE
+ * (if any) before we get here.
+ */
+ if (!extra->limit_needed)
+ {
+ ListCell *lc;
+
+ Assert(parse->rowMarks);
+
+ foreach(lc, input_rel->pathlist)
+ {
+ Path *path = (Path *) lfirst(lc);
+
+ /*
+ * apply_scanjoin_target_to_paths() might put a ProjectionPath on
+ * top of each of its scan/join paths; look through ProjectionPath
+ * and see if the path underneath it is ForeignPath.
+ */
+ if (IsA(path, ForeignPath) ||
+ (IsA(path, ProjectionPath) &&
+ IsA(((ProjectionPath *) path)->subpath, ForeignPath)))
+ {
+ /* Create foreign final ForeignPath */
+ final_path = create_foreign_upper_path(root,
+ path->parent,
+ path->pathtarget,
+ path->rows,
+ path->startup_cost,
+ path->total_cost,
+ path->pathkeys,
+ NULL, /* no extra plan */
+ NULL); /* no fdw_private */
+
+ /* And add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ return;
+ }
+ }
+ }
+
+ /*
+ * We try to create paths below by extending paths for the underlying
+ * base, join, or grouping relation to perform the final sort and the
+ * LIMIT restriction remotely if the input_rel is an ordered relation, or
+ * to perform the LIMIT restriction remotely otherwise, which is stored
+ * into the fdw_private list of the resulting paths. Note: in the former
+ * case we can estimate the effects of the bounded sort by doing this.
+ */
+
+ /*
+ * If the input_rel is an ordered relation, replace the input_rel with
+ * its underlying base, join, or grouping relation
+ */
+ if (IS_UPPER_REL(input_rel) && ifpinfo->stage == UPPERREL_ORDERED)
+ {
+ input_rel = ifpinfo->outerrel;
+ ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ has_final_sort = true;
+ pathkeys = root->sort_pathkeys;
+ }
+
+ /* Assess if it is safe to push down the LIMIT and OFFSET, if any */
+ if (extra->limit_needed)
+ {
+ /*
+ * If the input relation has any local conditions, the LIMIT/OFFSET
+ * cannot be pushed down.
+ */
+ if (ifpinfo->local_conds)
+ return;
+
+ /*
+ * Also, the LIMIT/OFFSET cannot be pushed down, if their expressions
+ * are not safe to remote.
+ */
+ if (!is_foreign_expr(root, input_rel, (Expr *) parse->limitOffset) ||
+ !is_foreign_expr(root, input_rel, (Expr *) parse->limitCount))
+ return;
+ }
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ /* Initialize the selectivity and cost of the local_conds */
+ fpinfo->local_conds_sel = 1.0;
+ fpinfo->local_conds_cost.startup = 0.0;
+ fpinfo->local_conds_cost.per_tuple = 0.0;
+
+ fpextra = (PgFdwPathExtraData *) palloc0(sizeof(PgFdwPathExtraData));
+ fpextra->target = root->upper_targets[UPPERREL_FINAL];
+ fpextra->has_final_sort = has_final_sort;
+ fpextra->has_limit = extra->limit_needed;
+ fpextra->limit_tuples = extra->limit_tuples;
+ fpextra->count_est = extra->count_est;
+ fpextra->offset_est = extra->offset_est;
+
+ /*
+ * Estimate the costs of performing the final sort (if any) and the LIMIT
+ * restriction remotely. If there is no final sort, we wouldn't really
+ * need to execute EXPLAIN anymore to estimate the costs, since they can
+ * be roughly estimated by applying to adjust_limit_rows_costs() the cost
+ * estimates we already have for the underlying base, join, or grouping
+ * relation, in the same way as when use_remote_estimate is false. Since
+ * it's pretty expensive to execute EXPLAIN, force use_remote_estimate to
+ * false in that case.
+ */
+ if (!fpextra->has_final_sort)
+ {
+ save_use_remote_estimate = ifpinfo->use_remote_estimate;
+ ifpinfo->use_remote_estimate = false;
+ }
+ estimate_path_cost_size(root, input_rel, NIL, pathkeys, fpextra,
+ &rows, &width, &startup_cost, &total_cost);
+ if (!fpextra->has_final_sort)
+ ifpinfo->use_remote_estimate = save_use_remote_estimate;
+
+ /*
+ * Build the fdw_private list that will be used by postgresGetForeignPlan.
+ * Items in the list must match order in enum FdwPathPrivateIndex.
+ */
+ fdw_private = list_make2(makeInteger(has_final_sort),
+ makeInteger(extra->limit_needed));
+
+ /* Create foreign final ForeignPath */
+ final_path = create_foreign_upper_path(root,
+ input_rel,
+ root->upper_targets[UPPERREL_FINAL],
+ rows,
+ startup_cost,
+ total_cost,
+ pathkeys,
+ NULL, /* no extra plan */
+ fdw_private);
+
+ /* And add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+}
+
/*
* Create a tuple from the specified row of the PGresult.
*
diff --git a/contrib/postgres_fdw/postgres_fdw.h b/contrib/postgres_fdw/postgres_fdw.h
index 936e3f498a..aabe1dec68 100644
--- a/contrib/postgres_fdw/postgres_fdw.h
+++ b/contrib/postgres_fdw/postgres_fdw.h
@@ -185,7 +185,8 @@ extern List *build_tlist_to_deparse(RelOptInfo *foreignrel);
extern void deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, List *tlist,
List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit,
+ bool is_subquery,
List **retrieved_attrs, List **params_list);
extern const char *get_jointype_name(JoinType jointype);
diff --git a/contrib/postgres_fdw/sql/postgres_fdw.sql b/contrib/postgres_fdw/sql/postgres_fdw.sql
index eb9d1ad59d..621d4faa1f 100644
--- a/contrib/postgres_fdw/sql/postgres_fdw.sql
+++ b/contrib/postgres_fdw/sql/postgres_fdw.sql
@@ -506,7 +506,7 @@ SELECT t1.c1 FROM ft1 t1 WHERE EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c1)
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
diff --git a/src/backend/optimizer/plan/planner.c b/src/backend/optimizer/plan/planner.c
index bc81535905..8deef7dc08 100644
--- a/src/backend/optimizer/plan/planner.c
+++ b/src/backend/optimizer/plan/planner.c
@@ -1709,6 +1709,7 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
bool final_target_parallel_safe;
RelOptInfo *current_rel;
RelOptInfo *final_rel;
+ FinalPathExtraData extra;
ListCell *lc;
/* Tweak caller-supplied tuple_fraction if have LIMIT/OFFSET */
@@ -2275,6 +2276,11 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
}
}
+ extra.limit_needed = limit_needed(parse);
+ extra.limit_tuples = limit_tuples;
+ extra.count_est = count_est;
+ extra.offset_est = offset_est;
+
/*
* If there is an FDW that's responsible for all baserels of the query,
* let it consider adding ForeignPaths.
@@ -2283,12 +2289,12 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
final_rel->fdwroutine->GetForeignUpperPaths)
final_rel->fdwroutine->GetForeignUpperPaths(root, UPPERREL_FINAL,
current_rel, final_rel,
- NULL);
+ &extra);
/* Let extensions possibly add some more paths */
if (create_upper_paths_hook)
(*create_upper_paths_hook) (root, UPPERREL_FINAL,
- current_rel, final_rel, NULL);
+ current_rel, final_rel, &extra);
/* Note: currently, we leave it to callers to do set_cheapest() */
}
diff --git a/src/include/nodes/pathnodes.h b/src/include/nodes/pathnodes.h
index a008ae07da..d01c152b9f 100644
--- a/src/include/nodes/pathnodes.h
+++ b/src/include/nodes/pathnodes.h
@@ -2429,6 +2429,23 @@ typedef struct
PartitionwiseAggregateType patype;
} GroupPathExtraData;
+/*
+ * Struct for extra information passed to subroutines of grouping_planner
+ *
+ * limit_needed is true if we actually need a Limit plan node
+ * limit_tuples is the max number of tuples needed according to LIMIT/OFFSET
+ * count_est and offset_est are the values of the LIMIT/OFFSET clauses
+ * estimated by preprocess_limit() in planner.c (see comments for
+ * preprocess_limit() for more information).
+ */
+typedef struct
+{
+ bool limit_needed;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
+} FinalPathExtraData;
+
/*
* For speed reasons, cost estimation for join paths is performed in two
* phases: the first phase tries to quickly derive a lower bound for the
--
2.19.2
--------------050108040401020008040409--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely
@ 2019-03-20 11:34 Etsuro Fujita <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Etsuro Fujita @ 2019-03-20 11:34 UTC (permalink / raw)
---
contrib/postgres_fdw/deparse.c | 37 +-
.../postgres_fdw/expected/postgres_fdw.out | 669 +++++++-----------
contrib/postgres_fdw/postgres_fdw.c | 327 ++++++++-
contrib/postgres_fdw/postgres_fdw.h | 3 +-
contrib/postgres_fdw/sql/postgres_fdw.sql | 16 +-
src/backend/optimizer/plan/planner.c | 10 +-
src/include/nodes/pathnodes.h | 17 +
7 files changed, 650 insertions(+), 429 deletions(-)
diff --git a/contrib/postgres_fdw/deparse.c b/contrib/postgres_fdw/deparse.c
index 97dd07bee8..079406f4f3 100644
--- a/contrib/postgres_fdw/deparse.c
+++ b/contrib/postgres_fdw/deparse.c
@@ -169,6 +169,7 @@ static void deparseSelectSql(List *tlist, bool is_subquery, List **retrieved_att
static void deparseLockingClause(deparse_expr_cxt *context);
static void appendOrderByClause(List *pathkeys, bool has_final_sort,
deparse_expr_cxt *context);
+static void appendLimitClause(deparse_expr_cxt *context);
static void appendConditions(List *exprs, deparse_expr_cxt *context);
static void deparseFromExprForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, bool use_alias,
@@ -930,7 +931,7 @@ build_tlist_to_deparse(RelOptInfo *foreignrel)
void
deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
List *tlist, List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit, bool is_subquery,
List **retrieved_attrs, List **params_list)
{
deparse_expr_cxt context;
@@ -988,6 +989,10 @@ deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
if (pathkeys)
appendOrderByClause(pathkeys, has_final_sort, &context);
+ /* Add LIMIT clause if necessary */
+ if (has_limit)
+ appendLimitClause(&context);
+
/* Add any necessary FOR UPDATE/SHARE. */
deparseLockingClause(&context);
}
@@ -1591,7 +1596,8 @@ deparseRangeTblRef(StringInfo buf, PlannerInfo *root, RelOptInfo *foreignrel,
/* Deparse the subquery representing the relation. */
appendStringInfoChar(buf, '(');
deparseSelectStmtForRel(buf, root, foreignrel, NIL,
- fpinfo->remote_conds, NIL, false, true,
+ fpinfo->remote_conds, NIL,
+ false, false, true,
&retrieved_attrs, params_list);
appendStringInfoChar(buf, ')');
@@ -3160,6 +3166,33 @@ appendOrderByClause(List *pathkeys, bool has_final_sort,
reset_transmission_modes(nestlevel);
}
+/*
+ * Deparse LIMIT/OFFSET clause.
+ */
+static void
+appendLimitClause(deparse_expr_cxt *context)
+{
+ PlannerInfo *root = context->root;
+ StringInfo buf = context->buf;
+ int nestlevel;
+
+ /* Make sure any constants in the exprs are printed portably */
+ nestlevel = set_transmission_modes();
+
+ if (root->parse->limitCount)
+ {
+ appendStringInfoString(buf, " LIMIT ");
+ deparseExpr((Expr *) root->parse->limitCount, context);
+ }
+ if (root->parse->limitOffset)
+ {
+ appendStringInfoString(buf, " OFFSET ");
+ deparseExpr((Expr *) root->parse->limitOffset, context);
+ }
+
+ reset_transmission_modes(nestlevel);
+}
+
/*
* appendFunctionName
* Deparses function name from given function oid.
diff --git a/contrib/postgres_fdw/expected/postgres_fdw.out b/contrib/postgres_fdw/expected/postgres_fdw.out
index 592dd3e05f..936cd214e2 100644
--- a/contrib/postgres_fdw/expected/postgres_fdw.out
+++ b/contrib/postgres_fdw/expected/postgres_fdw.out
@@ -236,11 +236,10 @@ ALTER FOREIGN TABLE ft2 OPTIONS (use_remote_estimate 'true');
-- ===================================================================
-- single table without alias
EXPLAIN (COSTS OFF) SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------
- Limit
- -> Foreign Scan on ft1
-(2 rows)
+ QUERY PLAN
+---------------------
+ Foreign Scan on ft1
+(1 row)
SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -288,14 +287,12 @@ SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
-- whole-row reference
EXPLAIN (VERBOSE, COSTS OFF) SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: t1.*, c3, c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.*, c3, c1
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(3 rows)
SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
t1
@@ -335,14 +332,12 @@ SELECT * FROM ft1 t1 WHERE t1.c1 = 101 AND t1.c6 = '1' AND t1.c7 >= '1';
-- with FOR UPDATE/SHARE
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -351,14 +346,12 @@ SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
(1 row)
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -968,6 +961,25 @@ SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
9
(1 row)
+-- ORDER BY can be shipped, though
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Limit
+ Output: c1, c2, c3, c4, c5, c6, c7, c8
+ -> Foreign Scan on public.ft1 t1
+ Output: c1, c2, c3, c4, c5, c6, c7, c8
+ Filter: (t1.c1 === t1.c2)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c2 ASC NULLS LAST
+(6 rows)
+
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
+----+----+-------+------------------------------+--------------------------+----+------------+-----
+ 1 | 1 | 00001 | Fri Jan 02 00:00:00 1970 PST | Fri Jan 02 00:00:00 1970 | 1 | 1 | foo
+(1 row)
+
-- but let's put them in an extension ...
ALTER EXTENSION postgres_fdw ADD FUNCTION postgres_fdw_abs(int);
ALTER EXTENSION postgres_fdw ADD OPERATOR === (int, int);
@@ -1005,6 +1017,22 @@ SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
9
(1 row)
+-- and both ORDER BY and LIMIT can be shipped
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" OPERATOR(public.===) c2)) ORDER BY c2 ASC NULLS LAST LIMIT 1::bigint
+(3 rows)
+
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
+----+----+-------+------------------------------+--------------------------+----+------------+-----
+ 1 | 1 | 00001 | Fri Jan 02 00:00:00 1970 PST | Fri Jan 02 00:00:00 1970 | 1 | 1 | foo
+(1 row)
+
-- ===================================================================
-- JOIN queries
-- ===================================================================
@@ -1015,15 +1043,13 @@ ANALYZE ft5;
-- join two tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -1043,18 +1069,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3, t1.c3
- -> Sort
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Sort Key: t1.c3, t1.c1
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1))))
-(9 rows)
+ Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1)))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1074,15 +1095,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t
-- left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1102,15 +1121,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.
-- left outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1175,15 +1192,13 @@ SELECT t1.c1, t1.c2, t2.c1, t2.c2 FROM ft4 t1 LEFT JOIN (SELECT * FROM ft5 WHERE
-- right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1203,15 +1218,13 @@ SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2
-- right outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1231,15 +1244,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGH
-- full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 45::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
c1 | c1
@@ -1283,15 +1294,13 @@ SELECT t1.c1, t2.c1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL
EXPLAIN (VERBOSE, COSTS OFF)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: 1
- -> Foreign Scan
- Output: 1
- Relations: (public.ft4) FULL JOIN (public.ft5)
- Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE))
-(6 rows)
+ Relations: (public.ft4) FULL JOIN (public.ft5)
+ Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
?column?
@@ -1406,15 +1415,13 @@ SELECT t1.c1, ss.a, ss.b FROM (SELECT c1 FROM "S 1"."T 3" WHERE c1 = 50) t1 INNE
-- full outer join + inner join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t3.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1, t3.c1
- Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST
-(6 rows)
+ Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST LIMIT 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
c1 | c1 | c1
@@ -1434,15 +1441,13 @@ SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 a
-- full outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1462,15 +1467,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1490,15 +1493,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT
-- right outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1518,15 +1519,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1546,15 +1545,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1574,15 +1571,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- right outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1602,15 +1597,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1661,15 +1654,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) WHERE (t1.c1
-- full outer join + WHERE clause with shippable extensions set
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t1.c3 FROM ft1 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) WHERE postgres_fdw_abs(t1.c1) > 0 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t1.c3
- Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0))
-(6 rows)
+ Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
ALTER SERVER loopback OPTIONS (DROP extensions);
-- full outer join + WHERE clause with shippable extensions not set
@@ -1691,37 +1682,13 @@ ALTER SERVER loopback OPTIONS (ADD extensions 'postgres_fdw');
-- tests whole-row reference for row marks
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
c1 | c1
@@ -1740,37 +1707,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1 FOR UPDATE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1 FOR UPDATE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
c1 | c1
@@ -1790,37 +1733,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join two tables with FOR SHARE clause
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
c1 | c1
@@ -1839,37 +1758,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1 FOR SHARE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1 FOR SHARE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
c1 | c1
@@ -1923,15 +1818,13 @@ WITH t (c1_1, c1_3, c2_1) AS MATERIALIZED (SELECT t1.c1, t1.c3, t2.c1 FROM ft1 t
-- ctid with whole-row reference
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.ctid, t1, t2, t1.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- -> Foreign Scan
- Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
-- SEMI JOIN, not pushed down
EXPLAIN (VERBOSE, COSTS OFF)
@@ -1999,27 +1892,16 @@ SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2
119
(10 rows)
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Sort
- Output: t1.c1, t2.c1
- Sort Key: t1.c1, t2.c1
- -> Nested Loop
- Output: t1.c1, t2.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
- -> Materialize
- Output: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
-(15 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1" FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (TRUE)) ORDER BY r1."C 1" ASC NULLS LAST, r2."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -2341,50 +2223,13 @@ SET enable_hashjoin TO false;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- -> Foreign Scan
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
- Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft4.c1, ft4.c2, ft4.c3, ft5.c1, ft5.c2, ft5.c3, ft1.*, ft2.*, ft4.*, ft5.*
- Merge Cond: (ft1.c2 = ft5.c1)
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*, ft4.c1, ft4.c2, ft4.c3, ft4.*
- Merge Cond: (ft1.c2 = ft4.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Sort Key: ft1.c2
- -> Merge Join
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*, ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Merge Cond: (ft1.c1 = ft2.c1)
- -> Sort
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Sort Key: ft1.c1
- -> Foreign Scan on public.ft1
- Output: ft1.c1, ft1.c2, ft1.c3, ft1.c4, ft1.c5, ft1.c6, ft1.c7, ft1.c8, ft1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) FOR UPDATE
- -> Materialize
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- -> Foreign Scan on public.ft2
- Output: ft2.c1, ft2.c2, ft2.c3, ft2.c4, ft2.c5, ft2.c6, ft2.c7, ft2.c8, ft2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" < 100)) ORDER BY "C 1" ASC NULLS LAST FOR UPDATE
- -> Sort
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Sort Key: ft4.c1
- -> Foreign Scan on public.ft4
- Output: ft4.c1, ft4.c2, ft4.c3, ft4.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 3" FOR UPDATE
- -> Sort
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Sort Key: ft5.c1
- -> Foreign Scan on public.ft5
- Output: ft5.c1, ft5.c2, ft5.c3, ft5.*
- Remote SQL: SELECT c1, c2, c3 FROM "S 1"."T 4" FOR UPDATE
-(41 rows)
+ Relations: (((public.ft1) INNER JOIN (public.ft2)) INNER JOIN (public.ft4)) INNER JOIN (public.ft5)
+ Remote SQL: SELECT r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8, r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8, r3.c1, r3.c2, r3.c3, r4.c1, r4.c2, r4.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, CASE WHEN (r3.*)::text IS NOT NULL THEN ROW(r3.c1, r3.c2, r3.c3) END, CASE WHEN (r4.*)::text IS NOT NULL THEN ROW(r4.c1, r4.c2, r4.c3) END FROM ((("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")) AND ((r2."C 1" < 100)) AND ((r1."C 1" < 100)))) INNER JOIN "S 1"."T 3" r3 ON (((r1.c2 = r3.c1)))) INNER JOIN "S 1"."T 4" r4 ON (((r1.c2 = r4.c1)))) FOR UPDATE OF r1 FOR UPDATE OF r2 FOR UPDATE OF r3 FOR UPDATE OF r4
+(4 rows)
SELECT * FROM ft1, ft2, ft4, ft5 WHERE ft1.c1 = ft2.c1 AND ft1.c2 = ft4.c1
AND ft1.c2 = ft5.c1 AND ft1.c1 < 100 AND ft2.c1 < 100 FOR UPDATE;
@@ -2452,15 +2297,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1
ALTER VIEW v4 OWNER TO regress_view_owner;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, ft5.c2, ft5.c1
- -> Foreign Scan
- Output: ft4.c1, ft5.c2, ft5.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5)
- Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5)
+ Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2517,15 +2360,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c
ALTER VIEW v4 OWNER TO CURRENT_USER;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, t2.c2, t2.c1
- -> Foreign Scan
- Output: ft4.c1, t2.c2, t2.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2570,6 +2411,22 @@ select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (ran
100 | 50500 | 505.0000000000000000 | 0 | 1000 | 0 | 50500
(5 rows)
+explain (verbose, costs off)
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
+ Output: (count(c6)), (sum(c1)), (avg(c1)), (min(c2)), (max(c1)), (stddev(c2)), ((sum(c1)) * ((random() <= '1'::double precision))::integer), c2
+ Relations: Aggregate on (public.ft1)
+ Remote SQL: SELECT count(c6), sum("C 1"), avg("C 1"), min(c2), max("C 1"), stddev(c2), c2 FROM "S 1"."T 1" WHERE ((c2 < 5)) GROUP BY 7 ORDER BY count(c6) ASC NULLS LAST, sum("C 1") ASC NULLS LAST LIMIT 1::bigint
+(4 rows)
+
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+ count | sum | avg | min | max | stddev | sum2
+-------+-------+----------------------+-----+-----+--------+-------
+ 100 | 49600 | 496.0000000000000000 | 1 | 991 | 0 | 49600
+(1 row)
+
-- Aggregate is not pushed down as aggregation contains random()
explain (verbose, costs off)
select sum(c1 * (random() <= 1)::int) as sum, avg(c1) from ft1;
@@ -4024,14 +3881,12 @@ SELECT * FROM ft1 t1 WHERE t1.tableoid = 'ft1'::regclass LIMIT 1;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------
- Limit
- Output: ((tableoid)::regclass), c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(5 rows)
+ QUERY PLAN
+-----------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
tableoid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4056,14 +3911,12 @@ SELECT * FROM ft1 t1 WHERE t1.ctid = '(0,2)';
EXPLAIN (VERBOSE, COSTS OFF)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1"
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
ctid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4268,12 +4121,10 @@ INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
Remote SQL: INSERT INTO "S 1"."T 1"("C 1", c2, c3, c4, c5, c6, c7, c8) VALUES ($1, $2, $3, $4, $5, $6, $7, $8)
-> Subquery Scan on "*SELECT*"
Output: "*SELECT*"."?column?", "*SELECT*"."?column?_1", NULL::integer, "*SELECT*"."?column?_2", NULL::timestamp with time zone, NULL::timestamp without time zone, NULL::character varying, 'ft2 '::character(10), NULL::user_enum
- -> Limit
- Output: ((ft2_1.c1 + 1000)), ((ft2_1.c2 + 100)), ((ft2_1.c3 || ft2_1.c3))
- -> Foreign Scan on public.ft2 ft2_1
- Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
- Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1"
-(9 rows)
+ -> Foreign Scan on public.ft2 ft2_1
+ Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
+ Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1" LIMIT 20::bigint
+(7 rows)
INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
INSERT INTO ft2 (c1,c2,c3)
@@ -6020,14 +5871,12 @@ VACUUM ANALYZE "S 1"."T 1";
-- FIRST behavior here.
-- ORDER BY DESC NULLS LAST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 795::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6046,14 +5895,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
-- ORDER BY DESC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6072,14 +5919,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
-- ORDER BY ASC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
diff --git a/contrib/postgres_fdw/postgres_fdw.c b/contrib/postgres_fdw/postgres_fdw.c
index 4208bf6baa..54cf169e7e 100644
--- a/contrib/postgres_fdw/postgres_fdw.c
+++ b/contrib/postgres_fdw/postgres_fdw.c
@@ -251,11 +251,14 @@ typedef struct PgFdwAnalyzeState
* We store:
*
* 1) Boolean flag showing if the remote query has the final sort
+ * 2) Boolean flag showing if the remote query has the LIMIT clause
*/
enum FdwPathPrivateIndex
{
/* has-final-sort flag (as an integer Value node) */
- FdwPathPrivateHasFinalSort
+ FdwPathPrivateHasFinalSort,
+ /* has-limit flag (as an integer Value node) */
+ FdwPathPrivateHasLimit
};
/* Struct for extra information passed to estimate_path_cost_size() */
@@ -263,6 +266,10 @@ typedef struct
{
PathTarget *target;
bool has_final_sort;
+ bool has_limit;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
} PgFdwPathExtraData;
/*
@@ -400,6 +407,7 @@ static void adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost);
static bool ec_member_matches_foreign(PlannerInfo *root, RelOptInfo *rel,
@@ -481,6 +489,10 @@ static void add_foreign_grouping_paths(PlannerInfo *root,
static void add_foreign_ordered_paths(PlannerInfo *root,
RelOptInfo *input_rel,
RelOptInfo *ordered_rel);
+static void add_foreign_final_paths(PlannerInfo *root,
+ RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra);
static void apply_server_options(PgFdwRelationInfo *fpinfo);
static void apply_table_options(PgFdwRelationInfo *fpinfo);
static void merge_fdw_options(PgFdwRelationInfo *fpinfo,
@@ -1183,14 +1195,19 @@ postgresGetForeignPlan(PlannerInfo *root,
List *retrieved_attrs;
StringInfoData sql;
bool has_final_sort = false;
+ bool has_limit = false;
ListCell *lc;
/*
* Get FDW private data created by postgresGetForeignUpperPaths(), if any.
*/
if (best_path->fdw_private)
+ {
has_final_sort = intVal(list_nth(best_path->fdw_private,
FdwPathPrivateHasFinalSort));
+ has_limit = intVal(list_nth(best_path->fdw_private,
+ FdwPathPrivateHasLimit));
+ }
if (IS_SIMPLE_REL(foreignrel))
{
@@ -1340,7 +1357,7 @@ postgresGetForeignPlan(PlannerInfo *root,
initStringInfo(&sql);
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_exprs, best_path->path.pathkeys,
- has_final_sort, false,
+ has_final_sort, has_limit, false,
&retrieved_attrs, ¶ms_list);
/* Remember remote_exprs for possible use by postgresPlanDirectModify */
@@ -2525,7 +2542,7 @@ postgresExplainDirectModify(ForeignScanState *node, ExplainState *es)
* param_join_conds are the parameterization clauses with outer relations.
* pathkeys specify the expected sort order if any for given path being costed.
* fpextra specifies additional post-scan/join-processing steps such as the
- * final sort.
+ * final sort and the LIMIT restriction.
*
* The function returns the cost and size estimates in p_row, p_width,
* p_startup_cost and p_total_cost variables.
@@ -2602,6 +2619,7 @@ estimate_path_cost_size(PlannerInfo *root,
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_conds, pathkeys,
fpextra ? fpextra->has_final_sort : false,
+ fpextra ? fpextra->has_limit : false,
false, &retrieved_attrs, NULL);
/* Get the remote estimate */
@@ -2669,15 +2687,30 @@ estimate_path_cost_size(PlannerInfo *root,
retrieved_rows = clamp_row_est(rows / fpinfo->local_conds_sel);
/*
- * We will come here again and again with different set of pathkeys
- * that caller wants to cost. We don't need to calculate the costs of
- * the underlying scan, join, or grouping each time. Instead, use the
- * costs if we have cached them already.
+ * We will come here again and again with different set of pathkeys or
+ * additional post-scan/join-processing steps that caller wants to
+ * cost. We don't need to calculate the costs of the underlying scan,
+ * join, or grouping each time. Instead, use the costs if we have
+ * cached them already.
*/
if (fpinfo->rel_startup_cost >= 0 && fpinfo->rel_total_cost >= 0)
{
startup_cost = fpinfo->rel_startup_cost;
run_cost = fpinfo->rel_total_cost - fpinfo->rel_startup_cost;
+
+ /*
+ * If we estimate the costs of a foreign scan or a foreign join
+ * with additional post-scan/join-processing steps, the scan or
+ * join costs obtained from the cache wouldn't yet contain the
+ * eval costs for the final scan/join target, which would've been
+ * updated by apply_scanjoin_target_to_paths(); add the eval costs
+ * now.
+ */
+ if (fpextra && !IS_UPPER_REL(foreignrel))
+ {
+ startup_cost += foreignrel->reltarget->cost.startup;
+ run_cost += foreignrel->reltarget->cost.per_tuple * rows;
+ }
}
else if (IS_JOIN_REL(foreignrel))
{
@@ -2895,6 +2928,7 @@ estimate_path_cost_size(PlannerInfo *root,
Assert(fpinfo->stage == UPPERREL_GROUP_AGG);
adjust_foreign_grouping_path_cost(root, pathkeys,
retrieved_rows, width,
+ fpextra->limit_tuples,
&startup_cost, &run_cost);
}
else
@@ -2905,6 +2939,14 @@ estimate_path_cost_size(PlannerInfo *root,
}
total_cost = startup_cost + run_cost;
+
+ /* Adjust the cost estimates if we have LIMIT */
+ if (fpextra && fpextra->has_limit)
+ {
+ adjust_limit_rows_costs(&rows, &startup_cost, &total_cost,
+ fpextra->offset_est, fpextra->count_est);
+ retrieved_rows = rows;
+ }
}
/*
@@ -2913,7 +2955,8 @@ estimate_path_cost_size(PlannerInfo *root,
* the foreignrel's reltarget (see make_sort_input_target()); adjust tlist
* eval costs.
*/
- if (fpextra && fpextra->target != foreignrel->reltarget)
+ if (fpextra && fpextra->has_final_sort &&
+ fpextra->target != foreignrel->reltarget)
{
QualCost oldcost = foreignrel->reltarget->cost;
QualCost newcost = fpextra->target->cost;
@@ -2951,6 +2994,29 @@ estimate_path_cost_size(PlannerInfo *root,
total_cost += fpinfo->fdw_tuple_cost * retrieved_rows;
total_cost += cpu_tuple_cost * retrieved_rows;
+ /*
+ * If we have LIMIT, we should perfer performing the restriction remotely
+ * rather than locally, as the former avoids extra row fetches from the
+ * remote that the latter might cause. But since the core code doesn't
+ * account for such fetches when estimating the costs of the local
+ * restriction (see create_limit_path()), there would be no difference
+ * between the costs of the local restriction and the costs of the remote
+ * restriction estimated above if we don't use remote estimates (except
+ * for the case where the foreignrel is a grouping relation and we
+ * estimate the effects of a bounded sort for that relation above). Tweak
+ * the costs of the remote restriction to ensure we'll prefer it if the
+ * LIMIT is a useful one.
+ */
+ if (!fpinfo->use_remote_estimate &&
+ fpextra && fpextra->has_limit &&
+ fpextra->limit_tuples > 0 &&
+ fpextra->limit_tuples < fpinfo->rows)
+ {
+ Assert(fpinfo->rows > 0);
+ total_cost -= (total_cost - startup_cost) * 0.05 *
+ (fpinfo->rows - fpextra->limit_tuples) / fpinfo->rows;
+ }
+
/* Return results. */
*p_rows = rows;
*p_width = width;
@@ -3017,6 +3083,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost)
{
@@ -3041,7 +3108,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
width,
0.0,
work_mem,
- -1.0);
+ limit_tuples);
*p_startup_cost = sort_path.startup_cost;
*p_run_cost = sort_path.total_cost - sort_path.startup_cost;
@@ -5581,7 +5648,8 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
/* Ignore stages we don't support; and skip any duplicate calls. */
if ((stage != UPPERREL_GROUP_AGG &&
- stage != UPPERREL_ORDERED) ||
+ stage != UPPERREL_ORDERED &&
+ stage != UPPERREL_FINAL) ||
output_rel->fdw_private)
return;
@@ -5599,6 +5667,10 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
case UPPERREL_ORDERED:
add_foreign_ordered_paths(root, input_rel, output_rel);
break;
+ case UPPERREL_FINAL:
+ add_foreign_final_paths(root, input_rel, output_rel,
+ (FinalPathExtraData *) extra);
+ break;
default:
elog(ERROR, "unexpected upper relation: %d", (int) stage);
break;
@@ -5812,7 +5884,7 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
* Build the fdw_private list that will be used by postgresGetForeignPlan.
* Items in the list must match order in enum FdwPathPrivateIndex.
*/
- fdw_private = list_make1(makeInteger(true));
+ fdw_private = list_make2(makeInteger(true), makeInteger(false));
/* Create foreign ordering ForeignPath */
ordered_path = create_foreign_upper_path(root,
@@ -5829,6 +5901,239 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
add_path(ordered_rel, (Path *) ordered_path);
}
+/*
+ * add_foreign_final_paths
+ * Add foreign paths for performing the final processing remotely.
+ *
+ * Given input_rel contains the source-data Paths. The paths are added to the
+ * given final_rel.
+ */
+static void
+add_foreign_final_paths(PlannerInfo *root, RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra)
+{
+ Query *parse = root->parse;
+ PgFdwRelationInfo *ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ PgFdwRelationInfo *fpinfo = (PgFdwRelationInfo *) final_rel->fdw_private;
+ bool has_final_sort = false;
+ List *pathkeys = NIL;
+ PgFdwPathExtraData *fpextra;
+ bool save_use_remote_estimate = false;
+ double rows;
+ int width;
+ Cost startup_cost;
+ Cost total_cost;
+ List *fdw_private;
+ ForeignPath *final_path;
+
+ /*
+ * Currently, we only support this for SELECT commands
+ */
+ if (parse->commandType != CMD_SELECT)
+ return;
+
+ /*
+ * No work if there is no FOR UPDATE/SHARE clause and if there is no need
+ * to add a LIMIT node
+ */
+ if (!parse->rowMarks && !extra->limit_needed)
+ return;
+
+ /* We don't support cases where there are any SRFs in the targetlist. */
+ if (parse->hasTargetSRFs)
+ return;
+
+ /* Save the input_rel as outerrel in fpinfo */
+ fpinfo->outerrel = input_rel;
+
+ /*
+ * Copy foreign table, foreign server, user mapping, FDW options etc.
+ * details from the input relation's fpinfo.
+ */
+ fpinfo->table = ifpinfo->table;
+ fpinfo->server = ifpinfo->server;
+ fpinfo->user = ifpinfo->user;
+ merge_fdw_options(fpinfo, ifpinfo, NULL);
+
+ /*
+ * If there is no need to add a LIMIT node, there might be a ForeignPath
+ * in the input_rel's pathlist that implements all behavior of the query.
+ * Note: we would already have accounted for the query's FOR UPDATE/SHARE
+ * (if any) before we get here.
+ */
+ if (!extra->limit_needed)
+ {
+ ListCell *lc;
+
+ Assert(parse->rowMarks);
+
+ /*
+ * The input_rel should be a base, join, or ordered relation; and if
+ * it's an ordered relation, its input relation should be a base or
+ * join relation.
+ */
+ Assert(input_rel->reloptkind == RELOPT_BASEREL ||
+ input_rel->reloptkind == RELOPT_JOINREL ||
+ (input_rel->reloptkind == RELOPT_UPPER_REL &&
+ ifpinfo->stage == UPPERREL_ORDERED &&
+ (ifpinfo->outerrel->reloptkind == RELOPT_BASEREL ||
+ ifpinfo->outerrel->reloptkind == RELOPT_JOINREL)));
+
+ foreach(lc, input_rel->pathlist)
+ {
+ Path *path = (Path *) lfirst(lc);
+
+ /*
+ * The core code might have put a ProjectionPath on top of each
+ * path for the input_rel (see apply_scanjoin_target_to_paths());
+ * look through ProjectionPath and see if the path underneath it
+ * is ForeignPath.
+ */
+ if (IsA(path, ForeignPath) ||
+ (IsA(path, ProjectionPath) &&
+ IsA(((ProjectionPath *) path)->subpath, ForeignPath)))
+ {
+ /*
+ * Create foreign final ForeignPath; this removes the outer
+ * plan (if any), which makes the EXPLAIN output look cleaner
+ */
+ final_path = create_foreign_upper_path(root,
+ path->parent,
+ path->pathtarget,
+ path->rows,
+ path->startup_cost,
+ path->total_cost,
+ path->pathkeys,
+ NULL, /* no extra plan */
+ NULL); /* no fdw_private */
+
+ /* and add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ return;
+ }
+ }
+
+ /*
+ * If we get here it means no ForeignPaths; since we would already
+ * have considered pushing down all operations of the query to the
+ * remote server, give up on it.
+ */
+ return;
+ }
+
+ Assert(extra->limit_needed);
+
+ /*
+ * If the input_rel is an ordered relation, replace the input_rel with its
+ * underlying relation
+ */
+ if (input_rel->reloptkind == RELOPT_UPPER_REL &&
+ ifpinfo->stage == UPPERREL_ORDERED)
+ {
+ input_rel = ifpinfo->outerrel;
+ ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ has_final_sort = true;
+ pathkeys = root->sort_pathkeys;
+ }
+
+ /* The input_rel should be a base, join, or grouping relation */
+ Assert(input_rel->reloptkind == RELOPT_BASEREL ||
+ input_rel->reloptkind == RELOPT_JOINREL ||
+ (input_rel->reloptkind == RELOPT_UPPER_REL &&
+ ifpinfo->stage == UPPERREL_GROUP_AGG));
+
+ /*
+ * We try to create a path below by extending a simple foreign path for
+ * the underlying base, join, or grouping relation to perform the final
+ * sort (if has_final_sort) and the LIMIT restriction remotely, which is
+ * stored into the fdw_private list of the resulting path. (We
+ * re-estimate the costs of sorting the underlying relation, if
+ * has_final_sort.)
+ */
+
+ /*
+ * Assess if it is safe to push down the LIMIT and OFFSET to the remote
+ * server
+ */
+
+ /*
+ * If the underlying relation has any local conditions, the LIMIT/OFFSET
+ * cannot be pushed down.
+ */
+ if (ifpinfo->local_conds)
+ return;
+
+ /*
+ * Also, the LIMIT/OFFSET cannot be pushed down, if their expressions are
+ * not safe to remote.
+ */
+ if (!is_foreign_expr(root, input_rel, (Expr *) parse->limitOffset) ||
+ !is_foreign_expr(root, input_rel, (Expr *) parse->limitCount))
+ return;
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ /* Initialize the selectivity and cost of the local_conds */
+ fpinfo->local_conds_sel = 1.0;
+ fpinfo->local_conds_cost.startup = 0.0;
+ fpinfo->local_conds_cost.per_tuple = 0.0;
+
+ /* Construct PgFdwPathExtraData */
+ fpextra = (PgFdwPathExtraData *) palloc0(sizeof(PgFdwPathExtraData));
+ fpextra->target = root->upper_targets[UPPERREL_FINAL];
+ fpextra->has_final_sort = has_final_sort;
+ fpextra->has_limit = extra->limit_needed;
+ fpextra->limit_tuples = extra->limit_tuples;
+ fpextra->count_est = extra->count_est;
+ fpextra->offset_est = extra->offset_est;
+
+ /*
+ * Estimate the costs of performing the final sort and the LIMIT
+ * restriction remotely. If has_final_sort is false, we wouldn't need to
+ * execute EXPLAIN anymore if use_remote_estimate, since the costs can be
+ * roughly estimated using the costs we already have for the underlying
+ * relation, in the same way as when use_remote_estimate is false. Since
+ * it's pretty expensive to execute EXPLAIN, force use_remote_estimate to
+ * false in that case.
+ */
+ if (!fpextra->has_final_sort)
+ {
+ save_use_remote_estimate = ifpinfo->use_remote_estimate;
+ ifpinfo->use_remote_estimate = false;
+ }
+ estimate_path_cost_size(root, input_rel, NIL, pathkeys, fpextra,
+ &rows, &width, &startup_cost, &total_cost);
+ if (!fpextra->has_final_sort)
+ ifpinfo->use_remote_estimate = save_use_remote_estimate;
+
+ /*
+ * Build the fdw_private list that will be used by postgresGetForeignPlan.
+ * Items in the list must match order in enum FdwPathPrivateIndex.
+ */
+ fdw_private = list_make2(makeInteger(has_final_sort),
+ makeInteger(extra->limit_needed));
+
+ /* Create foreign final ForeignPath */
+ final_path = create_foreign_upper_path(root,
+ input_rel,
+ root->upper_targets[UPPERREL_FINAL],
+ rows,
+ startup_cost,
+ total_cost,
+ pathkeys,
+ NULL, /* no extra plan */
+ fdw_private);
+
+ /* and add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+}
+
/*
* Create a tuple from the specified row of the PGresult.
*
diff --git a/contrib/postgres_fdw/postgres_fdw.h b/contrib/postgres_fdw/postgres_fdw.h
index e9cbbe6234..b382945067 100644
--- a/contrib/postgres_fdw/postgres_fdw.h
+++ b/contrib/postgres_fdw/postgres_fdw.h
@@ -188,7 +188,8 @@ extern List *build_tlist_to_deparse(RelOptInfo *foreignrel);
extern void deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, List *tlist,
List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit,
+ bool is_subquery,
List **retrieved_attrs, List **params_list);
extern const char *get_jointype_name(JoinType jointype);
diff --git a/contrib/postgres_fdw/sql/postgres_fdw.sql b/contrib/postgres_fdw/sql/postgres_fdw.sql
index eb9d1ad59d..b7b8c00aa0 100644
--- a/contrib/postgres_fdw/sql/postgres_fdw.sql
+++ b/contrib/postgres_fdw/sql/postgres_fdw.sql
@@ -349,6 +349,11 @@ EXPLAIN (VERBOSE, COSTS OFF)
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
+-- ORDER BY can be shipped, though
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+
-- but let's put them in an extension ...
ALTER EXTENSION postgres_fdw ADD FUNCTION postgres_fdw_abs(int);
ALTER EXTENSION postgres_fdw ADD OPERATOR === (int, int);
@@ -362,6 +367,11 @@ EXPLAIN (VERBOSE, COSTS OFF)
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
+-- and both ORDER BY and LIMIT can be shipped
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+
-- ===================================================================
-- JOIN queries
-- ===================================================================
@@ -506,7 +516,7 @@ SELECT t1.c1 FROM ft1 t1 WHERE EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c1)
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
@@ -609,6 +619,10 @@ explain (verbose, costs off)
select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2;
select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2;
+explain (verbose, costs off)
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+
-- Aggregate is not pushed down as aggregation contains random()
explain (verbose, costs off)
select sum(c1 * (random() <= 1)::int) as sum, avg(c1) from ft1;
diff --git a/src/backend/optimizer/plan/planner.c b/src/backend/optimizer/plan/planner.c
index e408e77d6f..8b61b8aad4 100644
--- a/src/backend/optimizer/plan/planner.c
+++ b/src/backend/optimizer/plan/planner.c
@@ -1704,6 +1704,7 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
bool final_target_parallel_safe;
RelOptInfo *current_rel;
RelOptInfo *final_rel;
+ FinalPathExtraData extra;
ListCell *lc;
/* Tweak caller-supplied tuple_fraction if have LIMIT/OFFSET */
@@ -2270,6 +2271,11 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
}
}
+ extra.limit_needed = limit_needed(parse);
+ extra.limit_tuples = limit_tuples;
+ extra.count_est = count_est;
+ extra.offset_est = offset_est;
+
/*
* If there is an FDW that's responsible for all baserels of the query,
* let it consider adding ForeignPaths.
@@ -2278,12 +2284,12 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
final_rel->fdwroutine->GetForeignUpperPaths)
final_rel->fdwroutine->GetForeignUpperPaths(root, UPPERREL_FINAL,
current_rel, final_rel,
- NULL);
+ &extra);
/* Let extensions possibly add some more paths */
if (create_upper_paths_hook)
(*create_upper_paths_hook) (root, UPPERREL_FINAL,
- current_rel, final_rel, NULL);
+ current_rel, final_rel, &extra);
/* Note: currently, we leave it to callers to do set_cheapest() */
}
diff --git a/src/include/nodes/pathnodes.h b/src/include/nodes/pathnodes.h
index 253e0b7e48..d7889ca768 100644
--- a/src/include/nodes/pathnodes.h
+++ b/src/include/nodes/pathnodes.h
@@ -2434,6 +2434,23 @@ typedef struct
PartitionwiseAggregateType patype;
} GroupPathExtraData;
+/*
+ * Struct for extra information passed to subroutines of grouping_planner
+ *
+ * limit_needed is true if we actually need a Limit plan node
+ * limit_tuples is the max number of tuples needed according to LIMIT/OFFSET
+ * count_est and offset_est are the values of the LIMIT/OFFSET clauses
+ * estimated by preprocess_limit() in planner.c (see comments for
+ * preprocess_limit() for more information).
+ */
+typedef struct
+{
+ bool limit_needed;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
+} FinalPathExtraData;
+
/*
* For speed reasons, cost estimation for join paths is performed in two
* phases: the first phase tries to quickly derive a lower bound for the
--
2.19.2
--------------080806040701020705050304--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH 4/4] postgres_fdw: Perform UPPERREL_FINAL step remotely
@ 2019-03-27 07:43 Etsuro Fujita <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Etsuro Fujita @ 2019-03-27 07:43 UTC (permalink / raw)
---
contrib/postgres_fdw/deparse.c | 37 +-
.../postgres_fdw/expected/postgres_fdw.out | 620 +++++++-----------
contrib/postgres_fdw/postgres_fdw.c | 344 +++++++++-
contrib/postgres_fdw/postgres_fdw.h | 3 +-
contrib/postgres_fdw/sql/postgres_fdw.sql | 16 +-
src/backend/optimizer/plan/planner.c | 10 +-
src/include/nodes/pathnodes.h | 17 +
7 files changed, 657 insertions(+), 390 deletions(-)
diff --git a/contrib/postgres_fdw/deparse.c b/contrib/postgres_fdw/deparse.c
index 97dd07bee8..079406f4f3 100644
--- a/contrib/postgres_fdw/deparse.c
+++ b/contrib/postgres_fdw/deparse.c
@@ -169,6 +169,7 @@ static void deparseSelectSql(List *tlist, bool is_subquery, List **retrieved_att
static void deparseLockingClause(deparse_expr_cxt *context);
static void appendOrderByClause(List *pathkeys, bool has_final_sort,
deparse_expr_cxt *context);
+static void appendLimitClause(deparse_expr_cxt *context);
static void appendConditions(List *exprs, deparse_expr_cxt *context);
static void deparseFromExprForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, bool use_alias,
@@ -930,7 +931,7 @@ build_tlist_to_deparse(RelOptInfo *foreignrel)
void
deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
List *tlist, List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit, bool is_subquery,
List **retrieved_attrs, List **params_list)
{
deparse_expr_cxt context;
@@ -988,6 +989,10 @@ deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root, RelOptInfo *rel,
if (pathkeys)
appendOrderByClause(pathkeys, has_final_sort, &context);
+ /* Add LIMIT clause if necessary */
+ if (has_limit)
+ appendLimitClause(&context);
+
/* Add any necessary FOR UPDATE/SHARE. */
deparseLockingClause(&context);
}
@@ -1591,7 +1596,8 @@ deparseRangeTblRef(StringInfo buf, PlannerInfo *root, RelOptInfo *foreignrel,
/* Deparse the subquery representing the relation. */
appendStringInfoChar(buf, '(');
deparseSelectStmtForRel(buf, root, foreignrel, NIL,
- fpinfo->remote_conds, NIL, false, true,
+ fpinfo->remote_conds, NIL,
+ false, false, true,
&retrieved_attrs, params_list);
appendStringInfoChar(buf, ')');
@@ -3160,6 +3166,33 @@ appendOrderByClause(List *pathkeys, bool has_final_sort,
reset_transmission_modes(nestlevel);
}
+/*
+ * Deparse LIMIT/OFFSET clause.
+ */
+static void
+appendLimitClause(deparse_expr_cxt *context)
+{
+ PlannerInfo *root = context->root;
+ StringInfo buf = context->buf;
+ int nestlevel;
+
+ /* Make sure any constants in the exprs are printed portably */
+ nestlevel = set_transmission_modes();
+
+ if (root->parse->limitCount)
+ {
+ appendStringInfoString(buf, " LIMIT ");
+ deparseExpr((Expr *) root->parse->limitCount, context);
+ }
+ if (root->parse->limitOffset)
+ {
+ appendStringInfoString(buf, " OFFSET ");
+ deparseExpr((Expr *) root->parse->limitOffset, context);
+ }
+
+ reset_transmission_modes(nestlevel);
+}
+
/*
* appendFunctionName
* Deparses function name from given function oid.
diff --git a/contrib/postgres_fdw/expected/postgres_fdw.out b/contrib/postgres_fdw/expected/postgres_fdw.out
index edf9cd780b..7bcec516db 100644
--- a/contrib/postgres_fdw/expected/postgres_fdw.out
+++ b/contrib/postgres_fdw/expected/postgres_fdw.out
@@ -236,11 +236,10 @@ ALTER FOREIGN TABLE ft2 OPTIONS (use_remote_estimate 'true');
-- ===================================================================
-- single table without alias
EXPLAIN (COSTS OFF) SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------
- Limit
- -> Foreign Scan on ft1
-(2 rows)
+ QUERY PLAN
+---------------------
+ Foreign Scan on ft1
+(1 row)
SELECT * FROM ft1 ORDER BY c3, c1 OFFSET 100 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -288,14 +287,12 @@ SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
-- whole-row reference
EXPLAIN (VERBOSE, COSTS OFF) SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: t1.*, c3, c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.*, c3, c1
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c3 ASC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(3 rows)
SELECT t1 FROM ft1 t1 ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
t1
@@ -335,14 +332,12 @@ SELECT * FROM ft1 t1 WHERE t1.c1 = 101 AND t1.c6 = '1' AND t1.c7 >= '1';
-- with FOR UPDATE/SHARE
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 101)) FOR UPDATE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -351,14 +346,12 @@ SELECT * FROM ft1 t1 WHERE c1 = 101 FOR UPDATE;
(1 row)
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------
- LockRows
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- -> Foreign Scan on public.ft1 t1
- Output: c1, c2, c3, c4, c5, c6, c7, c8, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" = 102)) FOR SHARE
+(3 rows)
SELECT * FROM ft1 t1 WHERE c1 = 102 FOR SHARE;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -968,6 +961,25 @@ SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
9
(1 row)
+-- ORDER BY can be shipped, though
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------
+ Limit
+ Output: c1, c2, c3, c4, c5, c6, c7, c8
+ -> Foreign Scan on public.ft1 t1
+ Output: c1, c2, c3, c4, c5, c6, c7, c8
+ Filter: (t1.c1 === t1.c2)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c2 ASC NULLS LAST
+(6 rows)
+
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
+----+----+-------+------------------------------+--------------------------+----+------------+-----
+ 1 | 1 | 00001 | Fri Jan 02 00:00:00 1970 PST | Fri Jan 02 00:00:00 1970 | 1 | 1 | foo
+(1 row)
+
-- but let's put them in an extension ...
ALTER EXTENSION postgres_fdw ADD FUNCTION postgres_fdw_abs(int);
ALTER EXTENSION postgres_fdw ADD OPERATOR === (int, int);
@@ -1005,6 +1017,22 @@ SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
9
(1 row)
+-- and both ORDER BY and LIMIT can be shipped
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" WHERE (("C 1" OPERATOR(public.===) c2)) ORDER BY c2 ASC NULLS LAST LIMIT 1::bigint
+(3 rows)
+
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+ c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
+----+----+-------+------------------------------+--------------------------+----+------------+-----
+ 1 | 1 | 00001 | Fri Jan 02 00:00:00 1970 PST | Fri Jan 02 00:00:00 1970 | 1 | 1 | foo
+(1 row)
+
-- ===================================================================
-- JOIN queries
-- ===================================================================
@@ -1015,15 +1043,13 @@ ANALYZE ft5;
-- join two tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -1043,18 +1069,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3, t1.c3
- -> Sort
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Sort Key: t1.c3, t1.c1
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3, t1.c3
- Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1))))
-(9 rows)
+ Relations: ((public.ft1 t1) INNER JOIN (public.ft2 t2)) INNER JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3, r1.c3 FROM (("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) INNER JOIN "S 1"."T 3" r4 ON (((r1."C 1" = r4.c1)))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t3 ON (t3.c1 = t1.c1) ORDER BY t1.c3, t1.c1 OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1074,15 +1095,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) JOIN ft4 t
-- left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 LEFT JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1102,15 +1121,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.
-- left outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1175,15 +1192,13 @@ SELECT t1.c1, t1.c2, t2.c1, t2.c2 FROM ft4 t1 LEFT JOIN (SELECT * FROM ft5 WHERE
-- right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t2) LEFT JOIN (public.ft5 t1)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r2 LEFT JOIN "S 1"."T 4" r1 ON (((r1.c1 = r2.c1)))) ORDER BY r2.c1 ASC NULLS LAST, r1.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2.c1, t1.c1 OFFSET 10 LIMIT 10;
c1 | c1
@@ -1203,15 +1218,13 @@ SELECT t1.c1, t2.c1 FROM ft5 t1 RIGHT JOIN ft4 t2 ON (t1.c1 = t2.c1) ORDER BY t2
-- right outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1231,15 +1244,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGH
-- full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1
- Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
- Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4 t1) FULL JOIN (public.ft5 t2)
+ Remote SQL: SELECT r1.c1, r2.c1 FROM ("S 1"."T 3" r1 FULL JOIN "S 1"."T 4" r2 ON (((r1.c1 = r2.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 45::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 45 LIMIT 10;
c1 | c1
@@ -1283,15 +1294,13 @@ SELECT t1.c1, t2.c1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL
EXPLAIN (VERBOSE, COSTS OFF)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: 1
- -> Foreign Scan
- Output: 1
- Relations: (public.ft4) FULL JOIN (public.ft5)
- Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE))
-(6 rows)
+ Relations: (public.ft4) FULL JOIN (public.ft5)
+ Remote SQL: SELECT NULL FROM ((SELECT NULL FROM "S 1"."T 3" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s4 FULL JOIN (SELECT NULL FROM "S 1"."T 4" WHERE ((c1 >= 50)) AND ((c1 <= 60))) s5 ON (TRUE)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT 1 FROM (SELECT c1 FROM ft4 WHERE c1 between 50 and 60) t1 FULL JOIN (SELECT c1 FROM ft5 WHERE c1 between 50 and 60) t2 ON (TRUE) OFFSET 10 LIMIT 10;
?column?
@@ -1406,15 +1415,13 @@ SELECT t1.c1, ss.a, ss.b FROM (SELECT c1 FROM "S 1"."T 3" WHERE c1 = 50) t1 INNE
-- full outer join + inner join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t3.c1
- -> Foreign Scan
- Output: t1.c1, t2.c1, t3.c1
- Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST
-(6 rows)
+ Relations: ((public.ft4 t1) INNER JOIN (public.ft5 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1.c1, r2.c1, r4.c1 FROM (("S 1"."T 3" r1 INNER JOIN "S 1"."T 4" r2 ON (((r1.c1 = (r2.c1 + 1))) AND ((r1.c1 >= 50)) AND ((r1.c1 <= 60)))) FULL JOIN "S 1"."T 3" r4 ON (((r2.c1 = r4.c1)))) ORDER BY r1.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST, r4.c1 ASC NULLS LAST LIMIT 10::bigint
+(4 rows)
SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 and t1.c1 between 50 and 60) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) ORDER BY t1.c1, t2.c1, t3.c1 LIMIT 10;
c1 | c1 | c1
@@ -1434,15 +1441,13 @@ SELECT t1.c1, t2.c1, t3.c1 FROM ft4 t1 INNER JOIN ft5 t2 ON (t1.c1 = t2.c1 + 1 a
-- full outer join three tables
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1462,15 +1467,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1"))))
-(6 rows)
+ Relations: ((public.ft4 t3) LEFT JOIN (public.ft2 t2)) LEFT JOIN (public.ft2 t1)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 3" r4 LEFT JOIN "S 1"."T 1" r2 ON (((r2."C 1" = r4.c1)))) LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1490,15 +1493,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT
-- right outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1518,15 +1519,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- full outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) FULL JOIN (public.ft2 t2)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1546,15 +1545,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + full outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t1) LEFT JOIN (public.ft2 t2)) FULL JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r1 LEFT JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) FULL JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1574,15 +1571,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) FULL
-- right outer join + left outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: ((public.ft2 t2) LEFT JOIN (public.ft2 t1)) LEFT JOIN (public.ft4 t3)
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM (("S 1"."T 1" r2 LEFT JOIN "S 1"."T 1" r1 ON (((r1."C 1" = r2."C 1")))) LEFT JOIN "S 1"."T 3" r4 ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1602,15 +1597,13 @@ SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 RIGHT JOIN ft2 t2 ON (t1.c1 = t2.c1) LEFT
-- left outer join + right outer join
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t3.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t3.c3
- Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
- Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1))))
-(6 rows)
+ Relations: (public.ft4 t3) LEFT JOIN ((public.ft2 t1) INNER JOIN (public.ft2 t2))
+ Remote SQL: SELECT r1."C 1", r2.c2, r4.c3 FROM ("S 1"."T 3" r4 LEFT JOIN ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ON (((r2."C 1" = r4.c1)))) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2, t3.c3 FROM ft2 t1 LEFT JOIN ft2 t2 ON (t1.c1 = t2.c1) RIGHT JOIN ft4 t3 ON (t2.c1 = t3.c1) OFFSET 10 LIMIT 10;
c1 | c2 | c3
@@ -1661,15 +1654,13 @@ SELECT t1.c1, t2.c1 FROM ft4 t1 FULL JOIN ft5 t2 ON (t1.c1 = t2.c1) WHERE (t1.c1
-- full outer join + WHERE clause with shippable extensions set
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2, t1.c3 FROM ft1 t1 FULL JOIN ft2 t2 ON (t1.c1 = t2.c1) WHERE postgres_fdw_abs(t1.c1) > 0 OFFSET 10 LIMIT 10;
- QUERY PLAN
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c2, t1.c3
- -> Foreign Scan
- Output: t1.c1, t2.c2, t1.c3
- Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0))
-(6 rows)
+ Relations: (public.ft1 t1) FULL JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2.c2, r1.c3 FROM ("S 1"."T 1" r1 FULL JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) WHERE ((public.postgres_fdw_abs(r1."C 1") > 0)) LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
ALTER SERVER loopback OPTIONS (DROP extensions);
-- full outer join + WHERE clause with shippable extensions not set
@@ -1691,37 +1682,13 @@ ALTER SERVER loopback OPTIONS (ADD extensions 'postgres_fdw');
-- tests whole-row reference for row marks
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE OF t1;
c1 | c1
@@ -1740,37 +1707,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR UPDATE OF r1 FOR UPDATE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR UPDATE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR UPDATE OF r1 FOR UPDATE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR UPDATE;
c1 | c1
@@ -1790,37 +1733,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
-- join two tables with FOR SHARE clause
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE OF t1;
c1 | c1
@@ -1839,37 +1758,13 @@ SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> LockRows
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Foreign Scan
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST FOR SHARE OF r1 FOR SHARE OF r2
- -> Result
- Output: t1.c1, t2.c1, t1.c3, t1.*, t2.*
- -> Sort
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Sort Key: t1.c3, t1.c1
- -> Merge Join
- Output: t1.c1, t1.c3, t1.*, t2.c1, t2.*
- Merge Cond: (t1.c1 = t2.c1)
- -> Sort
- Output: t1.c1, t1.c3, t1.*
- Sort Key: t1.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1, t1.c3, t1.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
- -> Sort
- Output: t2.c1, t2.*
- Sort Key: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1, t2.*
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" FOR SHARE
-(28 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1", r1.c3, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint FOR SHARE OF r1 FOR SHARE OF r2
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10 FOR SHARE;
c1 | c1
@@ -1923,15 +1818,13 @@ WITH t (c1_1, c1_3, c2_1) AS MATERIALIZED (SELECT t1.c1, t1.c3, t2.c1 FROM ft1 t
-- ctid with whole-row reference
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.ctid, t1, t2, t1.c1 FROM ft1 t1 JOIN ft2 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c3, t1.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- -> Foreign Scan
- Output: t1.ctid, t1.*, t2.*, t1.c1, t1.c3
- Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
- Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1.ctid, CASE WHEN (r1.*)::text IS NOT NULL THEN ROW(r1."C 1", r1.c2, r1.c3, r1.c4, r1.c5, r1.c6, r1.c7, r1.c8) END, CASE WHEN (r2.*)::text IS NOT NULL THEN ROW(r2."C 1", r2.c2, r2.c3, r2.c4, r2.c5, r2.c6, r2.c7, r2.c8) END, r1."C 1", r1.c3 FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (((r1."C 1" = r2."C 1")))) ORDER BY r1.c3 ASC NULLS LAST, r1."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
-- SEMI JOIN, not pushed down
EXPLAIN (VERBOSE, COSTS OFF)
@@ -1999,27 +1892,16 @@ SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2
119
(10 rows)
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: t1.c1, t2.c1
- -> Sort
- Output: t1.c1, t2.c1
- Sort Key: t1.c1, t2.c1
- -> Nested Loop
- Output: t1.c1, t2.c1
- -> Foreign Scan on public.ft1 t1
- Output: t1.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
- -> Materialize
- Output: t2.c1
- -> Foreign Scan on public.ft2 t2
- Output: t2.c1
- Remote SQL: SELECT "C 1" FROM "S 1"."T 1"
-(15 rows)
+ Relations: (public.ft1 t1) INNER JOIN (public.ft2 t2)
+ Remote SQL: SELECT r1."C 1", r2."C 1" FROM ("S 1"."T 1" r1 INNER JOIN "S 1"."T 1" r2 ON (TRUE)) ORDER BY r1."C 1" ASC NULLS LAST, r2."C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 100::bigint
+(4 rows)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
c1 | c1
@@ -2462,15 +2344,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1
ALTER VIEW v4 OWNER TO regress_view_owner;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, ft5.c2, ft5.c1
- -> Foreign Scan
- Output: ft4.c1, ft5.c2, ft5.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5)
- Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5)
+ Remote SQL: SELECT r6.c1, r9.c2, r9.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r9 ON (((r6.c1 = r9.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r9.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN v5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2527,15 +2407,13 @@ SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c
ALTER VIEW v4 OWNER TO CURRENT_USER;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10; -- can be pushed down
- QUERY PLAN
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
Output: ft4.c1, t2.c2, t2.c1
- -> Foreign Scan
- Output: ft4.c1, t2.c2, t2.c1
- Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
- Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST
-(6 rows)
+ Relations: (public.ft4) LEFT JOIN (public.ft5 t2)
+ Remote SQL: SELECT r6.c1, r2.c2, r2.c1 FROM ("S 1"."T 3" r6 LEFT JOIN "S 1"."T 4" r2 ON (((r6.c1 = r2.c1)))) ORDER BY r6.c1 ASC NULLS LAST, r2.c1 ASC NULLS LAST LIMIT 10::bigint OFFSET 10::bigint
+(4 rows)
SELECT t1.c1, t2.c2 FROM v4 t1 LEFT JOIN ft5 t2 ON (t1.c1 = t2.c1) ORDER BY t1.c1, t2.c1 OFFSET 10 LIMIT 10;
c1 | c2
@@ -2580,6 +2458,22 @@ select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (ran
100 | 50500 | 505.0000000000000000 | 0 | 1000 | 0 | 50500
(5 rows)
+explain (verbose, costs off)
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan
+ Output: (count(c6)), (sum(c1)), (avg(c1)), (min(c2)), (max(c1)), (stddev(c2)), ((sum(c1)) * ((random() <= '1'::double precision))::integer), c2
+ Relations: Aggregate on (public.ft1)
+ Remote SQL: SELECT count(c6), sum("C 1"), avg("C 1"), min(c2), max("C 1"), stddev(c2), c2 FROM "S 1"."T 1" WHERE ((c2 < 5)) GROUP BY 7 ORDER BY count(c6) ASC NULLS LAST, sum("C 1") ASC NULLS LAST LIMIT 1::bigint
+(4 rows)
+
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+ count | sum | avg | min | max | stddev | sum2
+-------+-------+----------------------+-----+-----+--------+-------
+ 100 | 49600 | 496.0000000000000000 | 1 | 991 | 0 | 49600
+(1 row)
+
-- Aggregate is not pushed down as aggregation contains random()
explain (verbose, costs off)
select sum(c1 * (random() <= 1)::int) as sum, avg(c1) from ft1;
@@ -4034,14 +3928,12 @@ SELECT * FROM ft1 t1 WHERE t1.tableoid = 'ft1'::regclass LIMIT 1;
EXPLAIN (VERBOSE, COSTS OFF)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------
- Limit
- Output: ((tableoid)::regclass), c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1"
-(5 rows)
+ QUERY PLAN
+-----------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
+ Output: (tableoid)::regclass, c1, c2, c3, c4, c5, c6, c7, c8
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT tableoid::regclass, * FROM ft1 t1 LIMIT 1;
tableoid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4066,14 +3958,12 @@ SELECT * FROM ft1 t1 WHERE t1.ctid = '(0,2)';
EXPLAIN (VERBOSE, COSTS OFF)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
- QUERY PLAN
--------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+-----------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1 t1
Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1 t1
- Output: ctid, c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1"
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8, ctid FROM "S 1"."T 1" LIMIT 1::bigint
+(3 rows)
SELECT ctid, * FROM ft1 t1 LIMIT 1;
ctid | c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -4278,12 +4168,10 @@ INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
Remote SQL: INSERT INTO "S 1"."T 1"("C 1", c2, c3, c4, c5, c6, c7, c8) VALUES ($1, $2, $3, $4, $5, $6, $7, $8)
-> Subquery Scan on "*SELECT*"
Output: "*SELECT*"."?column?", "*SELECT*"."?column?_1", NULL::integer, "*SELECT*"."?column?_2", NULL::timestamp with time zone, NULL::timestamp without time zone, NULL::character varying, 'ft2 '::character(10), NULL::user_enum
- -> Limit
- Output: ((ft2_1.c1 + 1000)), ((ft2_1.c2 + 100)), ((ft2_1.c3 || ft2_1.c3))
- -> Foreign Scan on public.ft2 ft2_1
- Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
- Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1"
-(9 rows)
+ -> Foreign Scan on public.ft2 ft2_1
+ Output: (ft2_1.c1 + 1000), (ft2_1.c2 + 100), (ft2_1.c3 || ft2_1.c3)
+ Remote SQL: SELECT "C 1", c2, c3 FROM "S 1"."T 1" LIMIT 20::bigint
+(7 rows)
INSERT INTO ft2 (c1,c2,c3) SELECT c1+1000,c2+100, c3 || c3 FROM ft2 LIMIT 20;
INSERT INTO ft2 (c1,c2,c3)
@@ -6030,14 +5918,12 @@ VACUUM ANALYZE "S 1"."T 1";
-- FIRST behavior here.
-- ORDER BY DESC NULLS LAST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS LAST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 795::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6056,14 +5942,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS LAST, c1 OFFSET 795 LIMIT 10;
-- ORDER BY DESC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
-----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+---------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 DESC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
@@ -6082,14 +5966,12 @@ SELECT * FROM ft1 ORDER BY c6 DESC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
-- ORDER BY ASC NULLS FIRST options
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
- QUERY PLAN
----------------------------------------------------------------------------------------------------------------------------------
- Limit
+ QUERY PLAN
+--------------------------------------------------------------------------------------------------------------------------------------------------------------
+ Foreign Scan on public.ft1
Output: c1, c2, c3, c4, c5, c6, c7, c8
- -> Foreign Scan on public.ft1
- Output: c1, c2, c3, c4, c5, c6, c7, c8
- Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST
-(5 rows)
+ Remote SQL: SELECT "C 1", c2, c3, c4, c5, c6, c7, c8 FROM "S 1"."T 1" ORDER BY c6 ASC NULLS FIRST, "C 1" ASC NULLS LAST LIMIT 10::bigint OFFSET 15::bigint
+(3 rows)
SELECT * FROM ft1 ORDER BY c6 ASC NULLS FIRST, c1 OFFSET 15 LIMIT 10;
c1 | c2 | c3 | c4 | c5 | c6 | c7 | c8
diff --git a/contrib/postgres_fdw/postgres_fdw.c b/contrib/postgres_fdw/postgres_fdw.c
index 7bd30333f0..10b02cbb2e 100644
--- a/contrib/postgres_fdw/postgres_fdw.c
+++ b/contrib/postgres_fdw/postgres_fdw.c
@@ -251,11 +251,14 @@ typedef struct PgFdwAnalyzeState
* We store:
*
* 1) Boolean flag showing if the remote query has the final sort
+ * 2) Boolean flag showing if the remote query has the LIMIT clause
*/
enum FdwPathPrivateIndex
{
/* has-final-sort flag (as an integer Value node) */
- FdwPathPrivateHasFinalSort
+ FdwPathPrivateHasFinalSort,
+ /* has-limit flag (as an integer Value node) */
+ FdwPathPrivateHasLimit
};
/* Struct for extra information passed to estimate_path_cost_size() */
@@ -263,6 +266,10 @@ typedef struct
{
PathTarget *target;
bool has_final_sort;
+ bool has_limit;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
} PgFdwPathExtraData;
/*
@@ -400,6 +407,7 @@ static void adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost);
static bool ec_member_matches_foreign(PlannerInfo *root, RelOptInfo *rel,
@@ -481,6 +489,10 @@ static void add_foreign_grouping_paths(PlannerInfo *root,
static void add_foreign_ordered_paths(PlannerInfo *root,
RelOptInfo *input_rel,
RelOptInfo *ordered_rel);
+static void add_foreign_final_paths(PlannerInfo *root,
+ RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra);
static void apply_server_options(PgFdwRelationInfo *fpinfo);
static void apply_table_options(PgFdwRelationInfo *fpinfo);
static void merge_fdw_options(PgFdwRelationInfo *fpinfo,
@@ -1183,14 +1195,19 @@ postgresGetForeignPlan(PlannerInfo *root,
List *retrieved_attrs;
StringInfoData sql;
bool has_final_sort = false;
+ bool has_limit = false;
ListCell *lc;
/*
* Get FDW private data created by postgresGetForeignUpperPaths(), if any.
*/
if (best_path->fdw_private)
+ {
has_final_sort = intVal(list_nth(best_path->fdw_private,
FdwPathPrivateHasFinalSort));
+ has_limit = intVal(list_nth(best_path->fdw_private,
+ FdwPathPrivateHasLimit));
+ }
if (IS_SIMPLE_REL(foreignrel))
{
@@ -1340,7 +1357,7 @@ postgresGetForeignPlan(PlannerInfo *root,
initStringInfo(&sql);
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_exprs, best_path->path.pathkeys,
- has_final_sort, false,
+ has_final_sort, has_limit, false,
&retrieved_attrs, ¶ms_list);
/* Remember remote_exprs for possible use by postgresPlanDirectModify */
@@ -2525,7 +2542,7 @@ postgresExplainDirectModify(ForeignScanState *node, ExplainState *es)
* param_join_conds are the parameterization clauses with outer relations.
* pathkeys specify the expected sort order if any for given path being costed.
* fpextra specifies additional post-scan/join-processing steps such as the
- * final sort.
+ * final sort and the LIMIT restriction.
*
* The function returns the cost and size estimates in p_row, p_width,
* p_startup_cost and p_total_cost variables.
@@ -2602,6 +2619,7 @@ estimate_path_cost_size(PlannerInfo *root,
deparseSelectStmtForRel(&sql, root, foreignrel, fdw_scan_tlist,
remote_conds, pathkeys,
fpextra ? fpextra->has_final_sort : false,
+ fpextra ? fpextra->has_limit : false,
false, &retrieved_attrs, NULL);
/* Get the remote estimate */
@@ -2669,15 +2687,32 @@ estimate_path_cost_size(PlannerInfo *root,
retrieved_rows = clamp_row_est(rows / fpinfo->local_conds_sel);
/*
- * We will come here again and again with different set of pathkeys
- * that caller wants to cost. We don't need to calculate the costs of
- * the underlying scan, join, or grouping each time. Instead, use the
- * costs if we have cached them already.
+ * We will come here again and again with different set of pathkeys or
+ * additional post-scan/join-processing steps that caller wants to
+ * cost. We don't need to calculate the costs of the underlying scan,
+ * join, or grouping each time. Instead, use the costs if we have
+ * cached them already.
*/
if (fpinfo->rel_startup_cost >= 0 && fpinfo->rel_total_cost >= 0)
{
startup_cost = fpinfo->rel_startup_cost;
run_cost = fpinfo->rel_total_cost - fpinfo->rel_startup_cost;
+
+ /*
+ * If we estimate the costs of a foreign scan or a foreign join
+ * with additional post-scan/join-processing steps, the scan or
+ * join costs obtained from the cache wouldn't yet contain the
+ * eval costs for the final scan/join target, which would've been
+ * updated by apply_scanjoin_target_to_paths(); add the eval costs
+ * now.
+ */
+ if (fpextra && !IS_UPPER_REL(foreignrel))
+ {
+ Assert(foreignrel->reloptkind == RELOPT_BASEREL ||
+ foreignrel->reloptkind == RELOPT_JOINREL);
+ startup_cost += foreignrel->reltarget->cost.startup;
+ run_cost += foreignrel->reltarget->cost.per_tuple * rows;
+ }
}
else if (IS_JOIN_REL(foreignrel))
{
@@ -2896,6 +2931,7 @@ estimate_path_cost_size(PlannerInfo *root,
fpinfo->stage == UPPERREL_GROUP_AGG);
adjust_foreign_grouping_path_cost(root, pathkeys,
retrieved_rows, width,
+ fpextra->limit_tuples,
&startup_cost, &run_cost);
}
else
@@ -2906,6 +2942,14 @@ estimate_path_cost_size(PlannerInfo *root,
}
total_cost = startup_cost + run_cost;
+
+ /* Adjust the cost estimates if we have LIMIT */
+ if (fpextra && fpextra->has_limit)
+ {
+ adjust_limit_rows_costs(&rows, &startup_cost, &total_cost,
+ fpextra->offset_est, fpextra->count_est);
+ retrieved_rows = rows;
+ }
}
/*
@@ -2914,7 +2958,8 @@ estimate_path_cost_size(PlannerInfo *root,
* the foreignrel's reltarget (see make_sort_input_target()); adjust tlist
* eval costs.
*/
- if (fpextra && fpextra->target != foreignrel->reltarget)
+ if (fpextra && fpextra->has_final_sort &&
+ fpextra->target != foreignrel->reltarget)
{
QualCost oldcost = foreignrel->reltarget->cost;
QualCost newcost = fpextra->target->cost;
@@ -2930,10 +2975,10 @@ estimate_path_cost_size(PlannerInfo *root,
* steps, before adding the costs for transferring data from the foreign
* server. These costs are useful for costing remote joins involving this
* relation or costing other remote operations for this relation such as
- * remote sorts, when the costs can not be obtained from the foreign
- * server. This function will be called at least once for every foreign
- * relation without any parameterization, pathkeys, or additional
- * post-scan/join-processing steps.
+ * remote sorts and remote LIMIT restrictions, when the costs can not be
+ * obtained from the foreign server. This function will be called at
+ * least once for every foreign relation without any parameterization,
+ * pathkeys, or additional post-scan/join-processing steps.
*/
if (pathkeys == NIL && param_join_conds == NIL && fpextra == NULL)
{
@@ -2952,6 +2997,30 @@ estimate_path_cost_size(PlannerInfo *root,
total_cost += fpinfo->fdw_tuple_cost * retrieved_rows;
total_cost += cpu_tuple_cost * retrieved_rows;
+ /*
+ * If we have LIMIT, we should perfer performing the restriction remotely
+ * rather than locally, as the former avoids extra row fetches from the
+ * remote that the latter might cause. But since the core code doesn't
+ * account for such fetches when estimating the costs of the local
+ * restriction (see create_limit_path()), there would be no difference
+ * between the costs of the local restriction and the costs of the remote
+ * restriction estimated above if we don't use remote estimates (except
+ * for the case where the foreignrel is a grouping relation, the given
+ * pathkeys is not NIL, and the effects of a bounded sort for that
+ * relation is accounted for in costing the remote restriction). Instead
+ * tweak the costs of the remote restriction to ensure we'll prefer it if
+ * the LIMIT is a useful one.
+ */
+ if (!fpinfo->use_remote_estimate &&
+ fpextra && fpextra->has_limit &&
+ fpextra->limit_tuples > 0 &&
+ fpextra->limit_tuples < fpinfo->rows)
+ {
+ Assert(fpinfo->rows > 0);
+ total_cost -= (total_cost - startup_cost) * 0.05 *
+ (fpinfo->rows - fpextra->limit_tuples) / fpinfo->rows;
+ }
+
/* Return results. */
*p_rows = rows;
*p_width = width;
@@ -3018,6 +3087,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
List *pathkeys,
double retrieved_rows,
double width,
+ double limit_tuples,
Cost *p_startup_cost,
Cost *p_run_cost)
{
@@ -3042,7 +3112,7 @@ adjust_foreign_grouping_path_cost(PlannerInfo *root,
width,
0.0,
work_mem,
- -1.0);
+ limit_tuples);
*p_startup_cost = sort_path.startup_cost;
*p_run_cost = sort_path.total_cost - sort_path.startup_cost;
@@ -5582,7 +5652,8 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
/* Ignore stages we don't support; and skip any duplicate calls. */
if ((stage != UPPERREL_GROUP_AGG &&
- stage != UPPERREL_ORDERED) ||
+ stage != UPPERREL_ORDERED &&
+ stage != UPPERREL_FINAL) ||
output_rel->fdw_private)
return;
@@ -5600,6 +5671,10 @@ postgresGetForeignUpperPaths(PlannerInfo *root, UpperRelationKind stage,
case UPPERREL_ORDERED:
add_foreign_ordered_paths(root, input_rel, output_rel);
break;
+ case UPPERREL_FINAL:
+ add_foreign_final_paths(root, input_rel, output_rel,
+ (FinalPathExtraData *) extra);
+ break;
default:
elog(ERROR, "unexpected upper relation: %d", (int) stage);
break;
@@ -5814,7 +5889,7 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
* Build the fdw_private list that will be used by postgresGetForeignPlan.
* Items in the list must match order in enum FdwPathPrivateIndex.
*/
- fdw_private = list_make1(makeInteger(true));
+ fdw_private = list_make2(makeInteger(true), makeInteger(false));
/* Create foreign ordering ForeignPath */
ordered_path = create_foreign_upper_path(root,
@@ -5831,6 +5906,245 @@ add_foreign_ordered_paths(PlannerInfo *root, RelOptInfo *input_rel,
add_path(ordered_rel, (Path *) ordered_path);
}
+/*
+ * add_foreign_final_paths
+ * Add foreign paths for performing the final processing remotely.
+ *
+ * Given input_rel contains the source-data Paths. The paths are added to the
+ * given final_rel.
+ */
+static void
+add_foreign_final_paths(PlannerInfo *root, RelOptInfo *input_rel,
+ RelOptInfo *final_rel,
+ FinalPathExtraData *extra)
+{
+ Query *parse = root->parse;
+ PgFdwRelationInfo *ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ PgFdwRelationInfo *fpinfo = (PgFdwRelationInfo *) final_rel->fdw_private;
+ bool has_final_sort = false;
+ List *pathkeys = NIL;
+ PgFdwPathExtraData *fpextra;
+ bool save_use_remote_estimate = false;
+ double rows;
+ int width;
+ Cost startup_cost;
+ Cost total_cost;
+ List *fdw_private;
+ ForeignPath *final_path;
+
+ /*
+ * Currently, we only support this for SELECT commands
+ */
+ if (parse->commandType != CMD_SELECT)
+ return;
+
+ /*
+ * No work if there is no FOR UPDATE/SHARE clause and if there is no need
+ * to add a LIMIT node
+ */
+ if (!parse->rowMarks && !extra->limit_needed)
+ return;
+
+ /* We don't support cases where there are any SRFs in the targetlist */
+ if (parse->hasTargetSRFs)
+ return;
+
+ /* Save the input_rel as outerrel in fpinfo */
+ fpinfo->outerrel = input_rel;
+
+ /*
+ * Copy foreign table, foreign server, user mapping, FDW options etc.
+ * details from the input relation's fpinfo.
+ */
+ fpinfo->table = ifpinfo->table;
+ fpinfo->server = ifpinfo->server;
+ fpinfo->user = ifpinfo->user;
+ merge_fdw_options(fpinfo, ifpinfo, NULL);
+
+ /*
+ * If there is no need to add a LIMIT node, there might be a ForeignPath
+ * in the input_rel's pathlist that implements all behavior of the query.
+ * Note: we would already have accounted for the query's FOR UPDATE/SHARE
+ * (if any) before we get here.
+ */
+ if (!extra->limit_needed)
+ {
+ ListCell *lc;
+
+ Assert(parse->rowMarks);
+
+ /*
+ * Grouping and aggregation are not supported with FOR UPDATE/SHARE,
+ * so the input_rel should be a base, join, or ordered relation; and
+ * if it's an ordered relation, its input relation should be a base
+ * or join relation.
+ */
+ Assert(input_rel->reloptkind == RELOPT_BASEREL ||
+ input_rel->reloptkind == RELOPT_JOINREL ||
+ (input_rel->reloptkind == RELOPT_UPPER_REL &&
+ ifpinfo->stage == UPPERREL_ORDERED &&
+ (ifpinfo->outerrel->reloptkind == RELOPT_BASEREL ||
+ ifpinfo->outerrel->reloptkind == RELOPT_JOINREL)));
+
+ foreach(lc, input_rel->pathlist)
+ {
+ Path *path = (Path *) lfirst(lc);
+
+ /*
+ * apply_scanjoin_target_to_paths() uses create_projection_path()
+ * to adjust each of its input paths if needed, although
+ * create_ordered_paths() uses apply_projection_to_path() to do
+ * that. So the former might have put a ProjectionPath on top of
+ * the ForeignPath; look through ProjectionPath and see if the
+ * path underneath it is ForeignPath.
+ */
+ if (IsA(path, ForeignPath) ||
+ (IsA(path, ProjectionPath) &&
+ IsA(((ProjectionPath *) path)->subpath, ForeignPath)))
+ {
+ /*
+ * Create foreign final ForeignPath; this removes the outer
+ * plan (if any), which makes the EXPLAIN output look cleaner
+ */
+ final_path = create_foreign_upper_path(root,
+ path->parent,
+ path->pathtarget,
+ path->rows,
+ path->startup_cost,
+ path->total_cost,
+ path->pathkeys,
+ NULL, /* no extra plan */
+ NULL); /* no fdw_private */
+
+ /* and add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ return;
+ }
+ }
+
+ /*
+ * If we get here it means no ForeignPaths; since we would already
+ * have considered pushing down all operations for the query to the
+ * remote server, give up on it.
+ */
+ return;
+ }
+
+ Assert(extra->limit_needed);
+
+ /*
+ * If the input_rel is an ordered relation, replace the input_rel with its
+ * input relation
+ */
+ if (input_rel->reloptkind == RELOPT_UPPER_REL &&
+ ifpinfo->stage == UPPERREL_ORDERED)
+ {
+ input_rel = ifpinfo->outerrel;
+ ifpinfo = (PgFdwRelationInfo *) input_rel->fdw_private;
+ has_final_sort = true;
+ pathkeys = root->sort_pathkeys;
+ }
+
+ /* The input_rel should be a base, join, or grouping relation */
+ Assert(input_rel->reloptkind == RELOPT_BASEREL ||
+ input_rel->reloptkind == RELOPT_JOINREL ||
+ (input_rel->reloptkind == RELOPT_UPPER_REL &&
+ ifpinfo->stage == UPPERREL_GROUP_AGG));
+
+ /*
+ * We try to create a path below by extending a simple foreign path for
+ * the underlying base, join, or grouping relation to perform the final
+ * sort (if has_final_sort) and the LIMIT restriction remotely, which is
+ * stored into the fdw_private list of the resulting path. (We
+ * re-estimate the costs of sorting the underlying relation, if
+ * has_final_sort.)
+ */
+
+ /*
+ * Assess if it is safe to push down the LIMIT and OFFSET to the remote
+ * server
+ */
+
+ /*
+ * If the underlying relation has any local conditions, the LIMIT/OFFSET
+ * cannot be pushed down.
+ */
+ if (ifpinfo->local_conds)
+ return;
+
+ /*
+ * Also, the LIMIT/OFFSET cannot be pushed down, if their expressions are
+ * not safe to remote.
+ */
+ if (!is_foreign_expr(root, input_rel, (Expr *) parse->limitOffset) ||
+ !is_foreign_expr(root, input_rel, (Expr *) parse->limitCount))
+ return;
+
+ /* Safe to push down */
+ fpinfo->pushdown_safe = true;
+
+ /* Initialize the selectivity and cost of the local_conds */
+ fpinfo->local_conds_sel = 1.0;
+ fpinfo->local_conds_cost.startup = 0.0;
+ fpinfo->local_conds_cost.per_tuple = 0.0;
+
+ /* Construct PgFdwPathExtraData */
+ fpextra = (PgFdwPathExtraData *) palloc0(sizeof(PgFdwPathExtraData));
+ fpextra->target = root->upper_targets[UPPERREL_FINAL];
+ fpextra->has_final_sort = has_final_sort;
+ fpextra->has_limit = extra->limit_needed;
+ fpextra->limit_tuples = extra->limit_tuples;
+ fpextra->count_est = extra->count_est;
+ fpextra->offset_est = extra->offset_est;
+
+ /*
+ * Estimate the costs of performing the final sort and the LIMIT
+ * restriction remotely. If has_final_sort is false, we wouldn't need to
+ * execute EXPLAIN anymore if use_remote_estimate, since the costs can be
+ * roughly estimated using the costs we already have for the underlying
+ * relation, in the same way as when use_remote_estimate is false. Since
+ * it's pretty expensive to execute EXPLAIN, force use_remote_estimate to
+ * false in that case.
+ */
+ if (!fpextra->has_final_sort)
+ {
+ save_use_remote_estimate = ifpinfo->use_remote_estimate;
+ ifpinfo->use_remote_estimate = false;
+ }
+ estimate_path_cost_size(root, input_rel, NIL, pathkeys, fpextra,
+ &rows, &width, &startup_cost, &total_cost);
+ if (!fpextra->has_final_sort)
+ ifpinfo->use_remote_estimate = save_use_remote_estimate;
+
+ /*
+ * Build the fdw_private list that will be used by postgresGetForeignPlan.
+ * Items in the list must match order in enum FdwPathPrivateIndex.
+ */
+ fdw_private = list_make2(makeInteger(has_final_sort),
+ makeInteger(extra->limit_needed));
+
+ /*
+ * Create foreign final ForeignPath; this removes the outer plan (if any),
+ * which makes the EXPLAIN output look cleaner
+ */
+ final_path = create_foreign_upper_path(root,
+ input_rel,
+ root->upper_targets[UPPERREL_FINAL],
+ rows,
+ startup_cost,
+ total_cost,
+ pathkeys,
+ NULL, /* no extra plan */
+ fdw_private);
+
+ /* and add it to the final_rel */
+ add_path(final_rel, (Path *) final_path);
+}
+
/*
* Create a tuple from the specified row of the PGresult.
*
diff --git a/contrib/postgres_fdw/postgres_fdw.h b/contrib/postgres_fdw/postgres_fdw.h
index e9cbbe6234..b382945067 100644
--- a/contrib/postgres_fdw/postgres_fdw.h
+++ b/contrib/postgres_fdw/postgres_fdw.h
@@ -188,7 +188,8 @@ extern List *build_tlist_to_deparse(RelOptInfo *foreignrel);
extern void deparseSelectStmtForRel(StringInfo buf, PlannerInfo *root,
RelOptInfo *foreignrel, List *tlist,
List *remote_conds, List *pathkeys,
- bool has_final_sort, bool is_subquery,
+ bool has_final_sort, bool has_limit,
+ bool is_subquery,
List **retrieved_attrs, List **params_list);
extern const char *get_jointype_name(JoinType jointype);
diff --git a/contrib/postgres_fdw/sql/postgres_fdw.sql b/contrib/postgres_fdw/sql/postgres_fdw.sql
index 460077fd9a..17a1c69bd8 100644
--- a/contrib/postgres_fdw/sql/postgres_fdw.sql
+++ b/contrib/postgres_fdw/sql/postgres_fdw.sql
@@ -349,6 +349,11 @@ EXPLAIN (VERBOSE, COSTS OFF)
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
+-- ORDER BY can be shipped, though
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+
-- but let's put them in an extension ...
ALTER EXTENSION postgres_fdw ADD FUNCTION postgres_fdw_abs(int);
ALTER EXTENSION postgres_fdw ADD OPERATOR === (int, int);
@@ -362,6 +367,11 @@ EXPLAIN (VERBOSE, COSTS OFF)
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
SELECT count(c3) FROM ft1 t1 WHERE t1.c1 === t1.c2;
+-- and both ORDER BY and LIMIT can be shipped
+EXPLAIN (VERBOSE, COSTS OFF)
+ SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+SELECT * FROM ft1 t1 WHERE t1.c1 === t1.c2 order by t1.c2 limit 1;
+
-- ===================================================================
-- JOIN queries
-- ===================================================================
@@ -506,7 +516,7 @@ SELECT t1.c1 FROM ft1 t1 WHERE EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c1)
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1 FROM ft1 t1 WHERE NOT EXISTS (SELECT 1 FROM ft2 t2 WHERE t1.c1 = t2.c2) ORDER BY t1.c1 OFFSET 100 LIMIT 10;
--- CROSS JOIN, not pushed down
+-- CROSS JOIN can be pushed down
EXPLAIN (VERBOSE, COSTS OFF)
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
SELECT t1.c1, t2.c1 FROM ft1 t1 CROSS JOIN ft2 t2 ORDER BY t1.c1, t2.c1 OFFSET 100 LIMIT 10;
@@ -613,6 +623,10 @@ explain (verbose, costs off)
select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2;
select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2;
+explain (verbose, costs off)
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+select count(c6), sum(c1), avg(c1), min(c2), max(c1), stddev(c2), sum(c1) * (random() <= 1)::int as sum2 from ft1 where c2 < 5 group by c2 order by 1, 2 limit 1;
+
-- Aggregate is not pushed down as aggregation contains random()
explain (verbose, costs off)
select sum(c1 * (random() <= 1)::int) as sum, avg(c1) from ft1;
diff --git a/src/backend/optimizer/plan/planner.c b/src/backend/optimizer/plan/planner.c
index e408e77d6f..8b61b8aad4 100644
--- a/src/backend/optimizer/plan/planner.c
+++ b/src/backend/optimizer/plan/planner.c
@@ -1704,6 +1704,7 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
bool final_target_parallel_safe;
RelOptInfo *current_rel;
RelOptInfo *final_rel;
+ FinalPathExtraData extra;
ListCell *lc;
/* Tweak caller-supplied tuple_fraction if have LIMIT/OFFSET */
@@ -2270,6 +2271,11 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
}
}
+ extra.limit_needed = limit_needed(parse);
+ extra.limit_tuples = limit_tuples;
+ extra.count_est = count_est;
+ extra.offset_est = offset_est;
+
/*
* If there is an FDW that's responsible for all baserels of the query,
* let it consider adding ForeignPaths.
@@ -2278,12 +2284,12 @@ grouping_planner(PlannerInfo *root, bool inheritance_update,
final_rel->fdwroutine->GetForeignUpperPaths)
final_rel->fdwroutine->GetForeignUpperPaths(root, UPPERREL_FINAL,
current_rel, final_rel,
- NULL);
+ &extra);
/* Let extensions possibly add some more paths */
if (create_upper_paths_hook)
(*create_upper_paths_hook) (root, UPPERREL_FINAL,
- current_rel, final_rel, NULL);
+ current_rel, final_rel, &extra);
/* Note: currently, we leave it to callers to do set_cheapest() */
}
diff --git a/src/include/nodes/pathnodes.h b/src/include/nodes/pathnodes.h
index 253e0b7e48..d7889ca768 100644
--- a/src/include/nodes/pathnodes.h
+++ b/src/include/nodes/pathnodes.h
@@ -2434,6 +2434,23 @@ typedef struct
PartitionwiseAggregateType patype;
} GroupPathExtraData;
+/*
+ * Struct for extra information passed to subroutines of grouping_planner
+ *
+ * limit_needed is true if we actually need a Limit plan node
+ * limit_tuples is the max number of tuples needed according to LIMIT/OFFSET
+ * count_est and offset_est are the values of the LIMIT/OFFSET clauses
+ * estimated by preprocess_limit() in planner.c (see comments for
+ * preprocess_limit() for more information).
+ */
+typedef struct
+{
+ bool limit_needed;
+ double limit_tuples;
+ int64 count_est;
+ int64 offset_est;
+} FinalPathExtraData;
+
/*
* For speed reasons, cost estimation for join paths is performed in two
* phases: the first phase tries to quickly derive a lower bound for the
--
2.19.2
--------------060101040300010105040500--
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
@ 2023-12-07 10:19 Andrey M. Borodin <[email protected]>
2023-12-07 14:16 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
0 siblings, 2 replies; 29+ messages in thread
From: Andrey M. Borodin @ 2023-12-07 10:19 UTC (permalink / raw)
To: Li, Yong <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]>; Shyrabokau, Anton <[email protected]>; Debnath, Shawn <[email protected]>
Hi Yong!
+1 to the idea to protect SLRUs from corruption. I'm slightly leaning towards the idea of separating checksums from data pages, but anyway this checksums are better than no checksums.
> On 7 Dec 2023, at 10:06, Li, Yong <[email protected]> wrote:
>
> I am still working on patching the pg_upgrade. Just love to hear your thoughts on the idea and the current patch.
FWIW you can take upgrade code from this patch [0] doing all the same stuff :)
Best regards, Andrey Borodin.
[0] https://www.postgresql.org/message-id/[email protected]
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
@ 2023-12-07 14:16 ` Aleksander Alekseev <[email protected]>
1 sibling, 0 replies; 29+ messages in thread
From: Aleksander Alekseev @ 2023-12-07 14:16 UTC (permalink / raw)
To: PostgreSQL Hackers <[email protected]>; +Cc: Li, Yong <[email protected]>; Andrey M. Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>; Debnath, Shawn <[email protected]>
Hi,
> +1 to the idea to protect SLRUs from corruption. I'm slightly leaning towards the idea of separating checksums from data pages, but anyway this checksums are better than no checksums.
>
> On 7 Dec 2023, at 10:06, Li, Yong <[email protected]> wrote:
>
> I am still working on patching the pg_upgrade. Just love to hear your thoughts on the idea and the current patch.
>
> FWIW you can take upgrade code from this patch [0] doing all the same stuff :)
Sounds like a half-measure to me. If we really want to go down this
rabbit hole IMO SLRU should be moved to shared buffers as proposed
elsewhere [1].
[1]: https://www.postgresql.org/message-id/[email protected]
--
Best regards,
Aleksander Alekseev
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
@ 2023-12-07 17:32 ` Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
1 sibling, 1 reply; 29+ messages in thread
From: Andrey Borodin @ 2023-12-07 17:32 UTC (permalink / raw)
To: PostgreSQL Hackers <[email protected]>; Aleksander Alekseev <[email protected]>; +Cc: Li, Yong <[email protected]>; Shyrabokau, Anton <[email protected]>; Debnath, Shawn <[email protected]>
<br /><br /><br />07.12.2023, 19:17, "Aleksander Alekseev" <[email protected]>:<br /><blockquote><p>Hi,<br /><br /></p><blockquote class="210e7a848e8fcb45wmi-quote"> +1 to the idea to protect SLRUs from corruption. I'm slightly leaning towards the idea of separating checksums from data pages, but anyway this checksums are better than no checksums.<br /><br /> On 7 Dec 2023, at 10:06, Li, Yong <<a href="mailto:[email protected]">[email protected]</a>> wrote:<br /><br /> I am still working on patching the pg_upgrade. Just love to hear your thoughts on the idea and the current patch.<br /><br /> FWIW you can take upgrade code from this patch [0] doing all the same stuff :)<br /></blockquote><p><br />Sounds like a half-measure to me. If we really want to go down this<br />rabbit hole IMO SLRU should be moved to shared buffers as proposed<br />elsewhere [1].<br /></p></blockquote>Thread that I cited stopped in 2018 for this exact reason. 5 years ago. Is this argument still valid?<div>Meanwhile checksums of buffer pages also reside on a page :)</div><div><br /></div><div>Best regards, Andrey Borodin.</div>
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
@ 2023-12-07 18:27 ` Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Debnath, Shawn @ 2023-12-07 18:27 UTC (permalink / raw)
To: Andrey Borodin <[email protected]>; PostgreSQL Hackers <[email protected]>; Aleksander Alekseev <[email protected]>; +Cc: Li, Yong <[email protected]>; Shyrabokau, Anton <[email protected]>; Bagga, Rishu <[email protected]>
>> Sounds like a half-measure to me. If we really want to go down this
>> rabbit hole IMO SLRU should be moved to shared buffers as proposed
>> elsewhere [1].
> Thread that I cited stopped in 2018 for this exact reason. 5 years ago. Is this argument still valid?
Meanwhile checksums of buffer pages also reside on a page :)
I would love to have seen more progress on the set of threads that proposed
the page header and integration of SLRU into buffer cache. The changes were
large, and unfortunately as a result, it didn't get the detailed review
that it needed. The complex nature of the feature allowed for more branches
to be split from the main thread with alternative approaches. Athough this is
great to see, it did result in the set of core requirements around LSN and
checksum tracking via page headers to not get into PG 16.
What is being proposed now is the simple and core functionality of introducing
page headers to SLRU pages while continuing to be in the SLRU cache. This
allows the whole project to be iterative and reviewers to better reason about
the smaller set of changes being introduced into the codebase.
Once the set of on-disk changes are in, we can follow up on optimizations.
It may be moving to buffer cache or reviewing Dilip's approach in [1], we
will have the option to be flexible in our approach.
[1] https://www.postgresql.org/message-id/flat/CAFiTN-vzDvNz=ExGXz6gdyjtzGixKSqs0mKHMmaQ8sOSEFZ33A@mail....
Shawn
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
@ 2023-12-07 19:51 ` Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Robert Haas @ 2023-12-07 19:51 UTC (permalink / raw)
To: Debnath, Shawn <[email protected]>; +Cc: Andrey Borodin <[email protected]>; PostgreSQL Hackers <[email protected]>; Aleksander Alekseev <[email protected]>; Li, Yong <[email protected]>; Shyrabokau, Anton <[email protected]>; Bagga, Rishu <[email protected]>
On Thu, Dec 7, 2023 at 1:28 PM Debnath, Shawn <[email protected]> wrote:
> What is being proposed now is the simple and core functionality of introducing
> page headers to SLRU pages while continuing to be in the SLRU cache. This
> allows the whole project to be iterative and reviewers to better reason about
> the smaller set of changes being introduced into the codebase.
>
> Once the set of on-disk changes are in, we can follow up on optimizations.
> It may be moving to buffer cache or reviewing Dilip's approach in [1], we
> will have the option to be flexible in our approach.
I basically agree with this. I don't think we should let the perfect
be the enemy of the good. Shooting down this patch because it doesn't
do everything that we want is a recipe for getting nothing done at
all.
That said, I don't think that the original post on this thread
provides a sufficiently clear and detailed motivation for making this
change. For this to eventually be committed, it's going to need (among
other things) a commit message that articulates a convincing rationale
for whatever changes it makes. Here's what the original email said:
> It adds a checksum to each SLRU page, tracks page LSN as if it is a standard page and eases future page enhancements.
Of those three things, in my opinion, the first is good and the other
two are too vague. I assume that most people who would be likely to
read a commit message would understand the value of pages having
checksums. But I can't immediately think of what the value of tracking
the page LSN as if it were a standard page might be, so that probably
needs more explanation. Similarly, at least one or two of the future
page enhancements that might be eased should be spelled out, and/or
the ways in which they would be made easier should be articulated.
--
Robert Haas
EDB: http://www.enterprisedb.com
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
@ 2023-12-08 09:35 ` Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2023-12-08 09:35 UTC (permalink / raw)
To: Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; +Cc: Andrey Borodin <[email protected]>; PostgreSQL Hackers <[email protected]>; Aleksander Alekseev <[email protected]>; Shyrabokau, Anton <[email protected]>; Bagga, Rishu <[email protected]>
Given so many different approaches were discussed, I have started a wiki to record and collaborate all efforts towards SLRU improvements. The wiki provides a concise overview of all the ideas discussed and can serve as a portal for all historical discussions. Currently, the wiki summarizes four recent threads ranging from identifier format change to page header change, to moving SLRU into the main buffer pool, to reduce lock contention on SLRU latches. We can keep the patch related discussions in this thread and use the wiki as a live document for larger scale collaborations.
The wiki page is here: https://wiki.postgresql.org/wiki/SLRU_improvements
Regarding the benefits of this patch, here is a detailed explanation:
1. Checksum is added to each page, allowing us to verify if a page has been corrupted when read from the disk.
2. The ad-hoc LSN group structure is removed from the SLRU cache control data and is replaced by the page LSN in the page header. This allows us to use the same WAL protocol as used by pages in the main buffer pool: flush all redo logs up to the page LSN before flushing the page itself. If we move SLRU caches into the main buffer pool, this change fits naturally.
3. It leaves further optimizations open. We can continue to pursue the goal of moving SLRU into the main buffer pool, or we can follow the lock partition idea. This change by itself does not conflict with either proposal.
Also, the patch is now complete and is ready for review. All check-world tests including tap tests passed with this patch.
Regards,
Yong
From: Robert Haas <[email protected]>
Date: Friday, December 8, 2023 at 03:51
To: Debnath, Shawn <[email protected]>
Cc: Andrey Borodin <[email protected]>, PostgreSQL Hackers <[email protected]>, Aleksander Alekseev <[email protected]>, Li, Yong <[email protected]>, Shyrabokau, Anton <[email protected]>, Bagga, Rishu <[email protected]>
Subject: Re: Proposal to add page headers to SLRU pages
External Email
On Thu, Dec 7, 2023 at 1:28 PM Debnath, Shawn <[email protected]> wrote:
> What is being proposed now is the simple and core functionality of introducing
> page headers to SLRU pages while continuing to be in the SLRU cache. This
> allows the whole project to be iterative and reviewers to better reason about
> the smaller set of changes being introduced into the codebase.
>
> Once the set of on-disk changes are in, we can follow up on optimizations.
> It may be moving to buffer cache or reviewing Dilip's approach in [1], we
> will have the option to be flexible in our approach.
I basically agree with this. I don't think we should let the perfect
be the enemy of the good. Shooting down this patch because it doesn't
do everything that we want is a recipe for getting nothing done at
all.
That said, I don't think that the original post on this thread
provides a sufficiently clear and detailed motivation for making this
change. For this to eventually be committed, it's going to need (among
other things) a commit message that articulates a convincing rationale
for whatever changes it makes. Here's what the original email said:
> It adds a checksum to each SLRU page, tracks page LSN as if it is a standard page and eases future page enhancements.
Of those three things, in my opinion, the first is good and the other
two are too vague. I assume that most people who would be likely to
read a commit message would understand the value of pages having
checksums. But I can't immediately think of what the value of tracking
the page LSN as if it were a standard page might be, so that probably
needs more explanation. Similarly, at least one or two of the future
page enhancements that might be eased should be spelled out, and/or
the ways in which they would be made easier should be articulated.
--
Robert Haas
EDB: https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.enterprisedb.com%2F&data=05...;
Attachments:
[application/octet-stream] slru_page_header_v1.patch (54.9K, ../../DM4PR07MB9812DA12DA729E8C709DEDD9B98AA@DM4PR07MB9812.namprd07.prod.outlook.com/3-slru_page_header_v1.patch)
download | inline diff:
src/backend/access/transam/clog.c | 52 ++++-----
src/backend/access/transam/commit_ts.c | 26 +++--
src/backend/access/transam/multixact.c | 63 ++++++----
src/backend/access/transam/slru.c | 113 ++++++------------
src/backend/access/transam/subtrans.c | 12 +-
src/backend/commands/async.c | 27 ++---
src/backend/storage/lmgr/predicate.c | 16 ++-
src/backend/storage/page/bufpage.c | 25 ++++
src/bin/pg_checksums/pg_checksums.c | 9 ++
src/bin/pg_resetwal/t/001_basic.pl | 6 +-
src/bin/pg_upgrade/Makefile | 1 +
src/bin/pg_upgrade/meson.build | 1 +
src/bin/pg_upgrade/nls.mk | 1 +
src/bin/pg_upgrade/pg_upgrade.c | 23 +++-
src/bin/pg_upgrade/pg_upgrade.h | 10 ++
src/bin/pg_upgrade/slru.c | 205 +++++++++++++++++++++++++++++++++
src/include/access/slru.h | 15 +--
src/include/catalog/catversion.h | 2 +-
src/include/storage/bufpage.h | 7 ++
src/test/modules/test_slru/test_slru.c | 13 ++-
20 files changed, 435 insertions(+), 192 deletions(-)
diff --git a/src/backend/access/transam/clog.c b/src/backend/access/transam/clog.c
index cc60eab1e2..33a0bcaea9 100644
--- a/src/backend/access/transam/clog.c
+++ b/src/backend/access/transam/clog.c
@@ -41,6 +41,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/proc.h"
#include "storage/sync.h"
@@ -59,7 +60,7 @@
/* We need two bits per xact, so four xacts fit in a byte */
#define CLOG_BITS_PER_XACT 2
#define CLOG_XACTS_PER_BYTE 4
-#define CLOG_XACTS_PER_PAGE (BLCKSZ * CLOG_XACTS_PER_BYTE)
+#define CLOG_XACTS_PER_PAGE (SizeOfPageContents * CLOG_XACTS_PER_BYTE)
#define CLOG_XACT_BITMASK ((1 << CLOG_BITS_PER_XACT) - 1)
@@ -77,13 +78,6 @@ TransactionIdToPage(TransactionId xid)
#define TransactionIdToByte(xid) (TransactionIdToPgIndex(xid) / CLOG_XACTS_PER_BYTE)
#define TransactionIdToBIndex(xid) ((xid) % (TransactionId) CLOG_XACTS_PER_BYTE)
-/* We store the latest async LSN for each group of transactions */
-#define CLOG_XACTS_PER_LSN_GROUP 32 /* keep this a power of 2 */
-#define CLOG_LSNS_PER_PAGE (CLOG_XACTS_PER_PAGE / CLOG_XACTS_PER_LSN_GROUP)
-
-#define GetLSNIndex(slotno, xid) ((slotno) * CLOG_LSNS_PER_PAGE + \
- ((xid) % (TransactionId) CLOG_XACTS_PER_PAGE) / CLOG_XACTS_PER_LSN_GROUP)
-
/*
* The number of subtransactions below which we consider to apply clog group
* update optimization. Testing reveals that the number higher than this can
@@ -101,7 +95,7 @@ static SlruCtlData XactCtlData;
static int ZeroCLOGPage(int64 pageno, bool writeXlog);
static bool CLOGPagePrecedes(int64 page1, int64 page2);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXact,
Oid oldestXactDb);
static void TransactionIdSetPageStatus(TransactionId xid, int nsubxids,
@@ -583,8 +577,9 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
char *byteptr;
char byteval;
char curval;
+ Page page = XactCtl->shared->page_buffer[slotno];
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(page) + byteno;
curval = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
/*
@@ -613,7 +608,7 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*byteptr = byteval;
/*
- * Update the group LSN if the transaction completion LSN is higher.
+ * Update the page LSN if the transaction completion LSN is higher.
*
* Note: lsn will be invalid when supplied during InRecovery processing,
* so we don't need to do anything special to avoid LSN updates during
@@ -622,10 +617,8 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*/
if (!XLogRecPtrIsInvalid(lsn))
{
- int lsnindex = GetLSNIndex(slotno, xid);
-
- if (XactCtl->shared->group_lsn[lsnindex] < lsn)
- XactCtl->shared->group_lsn[lsnindex] = lsn;
+ if (PageGetLSN(page) < lsn)
+ PageSetLSN(page, lsn);
}
}
@@ -651,19 +644,19 @@ TransactionIdGetStatus(TransactionId xid, XLogRecPtr *lsn)
int byteno = TransactionIdToByte(xid);
int bshift = TransactionIdToBIndex(xid) * CLOG_BITS_PER_XACT;
int slotno;
- int lsnindex;
+ Page page;
char *byteptr;
XidStatus status;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(XactCtl, pageno, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
status = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
- lsnindex = GetLSNIndex(slotno, xid);
- *lsn = XactCtl->shared->group_lsn[lsnindex];
+ *lsn = PageGetLSN(page);
LWLockRelease(XactSLRULock);
@@ -698,14 +691,14 @@ CLOGShmemBuffers(void)
Size
CLOGShmemSize(void)
{
- return SimpleLruShmemSize(CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE);
+ return SimpleLruShmemSize(CLOGShmemBuffers());
}
void
CLOGShmemInit(void)
{
XactCtl->PagePrecedes = CLOGPagePrecedes;
- SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE,
+ SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(),
XactSLRULock, "pg_xact", LWTRANCHE_XACT_BUFFER,
SYNC_HANDLER_CLOG, false);
SlruPagePrecedesUnitTests(XactCtl, CLOG_XACTS_PER_PAGE);
@@ -747,11 +740,17 @@ static int
ZeroCLOGPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(XactCtl, pageno);
+ page = XactCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -807,12 +806,12 @@ TrimCLOG(void)
char *byteptr;
slotno = SimpleLruReadPage(XactCtl, pageno, false, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(XactCtl->shared->page_buffer[slotno]) + byteno;
/* Zero so-far-unused positions in the current byte */
*byteptr &= (1 << bshift) - 1;
/* Zero the rest of the page */
- MemSet(byteptr + 1, 0, BLCKSZ - byteno - 1);
+ MemSet(byteptr + 1, 0, SizeOfPageContents - byteno - 1);
XactCtl->shared->page_dirty[slotno] = true;
}
@@ -836,7 +835,6 @@ CheckPointCLOG(void)
TRACE_POSTGRESQL_CLOG_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that CLOG has room for a newly-allocated XID.
*
@@ -958,12 +956,12 @@ CLOGPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
+ return XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/commit_ts.c b/src/backend/access/transam/commit_ts.c
index 7c642f7b59..6b36ee1022 100644
--- a/src/backend/access/transam/commit_ts.c
+++ b/src/backend/access/transam/commit_ts.c
@@ -31,6 +31,7 @@
#include "funcapi.h"
#include "miscadmin.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "storage/shmem.h"
#include "utils/builtins.h"
#include "utils/snapmgr.h"
@@ -63,7 +64,7 @@ typedef struct CommitTimestampEntry
sizeof(RepOriginId))
#define COMMIT_TS_XACTS_PER_PAGE \
- (BLCKSZ / SizeOfCommitTimestampEntry)
+ (SizeOfPageContents / SizeOfCommitTimestampEntry)
/*
@@ -120,7 +121,7 @@ static int ZeroCommitTsPage(int64 pageno, bool writeXlog);
static bool CommitTsPagePrecedes(int64 page1, int64 page2);
static void ActivateCommitTs(void);
static void DeactivateCommitTs(void);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXid);
/*
@@ -254,11 +255,12 @@ TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
+ Assert(xid == slotno * COMMIT_TS_XACTS_PER_PAGE + entryno);
entry.time = ts;
entry.nodeid = nodeid;
- memcpy(CommitTsCtl->shared->page_buffer[slotno] +
+ memcpy(PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
@@ -337,7 +339,7 @@ TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
- CommitTsCtl->shared->page_buffer[slotno] +
+ PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
@@ -515,7 +517,7 @@ CommitTsShmemBuffers(void)
Size
CommitTsShmemSize(void)
{
- return SimpleLruShmemSize(CommitTsShmemBuffers(), 0) +
+ return SimpleLruShmemSize(CommitTsShmemBuffers()) +
sizeof(CommitTimestampShared);
}
@@ -529,7 +531,7 @@ CommitTsShmemInit(void)
bool found;
CommitTsCtl->PagePrecedes = CommitTsPagePrecedes;
- SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(), 0,
+ SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(),
CommitTsSLRULock, "pg_commit_ts",
LWTRANCHE_COMMITTS_BUFFER,
SYNC_HANDLER_COMMIT_TS,
@@ -582,11 +584,17 @@ static int
ZeroCommitTsPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(CommitTsCtl, pageno);
+ page = CommitTsCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -946,12 +954,12 @@ CommitTsPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
+ return XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c
index db3423f12e..0970aa8855 100644
--- a/src/backend/access/transam/multixact.c
+++ b/src/backend/access/transam/multixact.c
@@ -83,6 +83,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "postmaster/autovacuum.h"
+#include "storage/bufpage.h"
#include "storage/lmgr.h"
#include "storage/pmsignal.h"
#include "storage/proc.h"
@@ -106,7 +107,7 @@
*/
/* We need four bytes per offset */
-#define MULTIXACT_OFFSETS_PER_PAGE (BLCKSZ / sizeof(MultiXactOffset))
+#define MULTIXACT_OFFSETS_PER_PAGE (SizeOfPageContents / sizeof(MultiXactOffset))
#define MultiXactIdToOffsetPage(xid) \
((xid) / (MultiXactOffset) MULTIXACT_OFFSETS_PER_PAGE)
@@ -119,8 +120,8 @@
* additional flag bits for each TransactionId. To do this without getting
* into alignment issues, we store four bytes of flags, and then the
* corresponding 4 Xids. Each such 5-word (20-byte) set we call a "group", and
- * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 409 groups
- * per page. This wastes 12 bytes per page, but that's OK -- simplicity (and
+ * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 408 groups
+ * per page. This wastes 8 bytes per page, but that's OK -- simplicity (and
* performance) trumps space efficiency here.
*
* Note that the "offset" macros work with byte offset, not array indexes, so
@@ -138,7 +139,7 @@
/* size in bytes of a complete group */
#define MULTIXACT_MEMBERGROUP_SIZE \
(sizeof(TransactionId) * MULTIXACT_MEMBERS_PER_MEMBERGROUP + MULTIXACT_FLAGBYTES_PER_GROUP)
-#define MULTIXACT_MEMBERGROUPS_PER_PAGE (BLCKSZ / MULTIXACT_MEMBERGROUP_SIZE)
+#define MULTIXACT_MEMBERGROUPS_PER_PAGE (SizeOfPageContents / MULTIXACT_MEMBERGROUP_SIZE)
#define MULTIXACT_MEMBERS_PER_PAGE \
(MULTIXACT_MEMBERGROUPS_PER_PAGE * MULTIXACT_MEMBERS_PER_MEMBERGROUP)
@@ -366,7 +367,7 @@ static bool MultiXactOffsetWouldWrap(MultiXactOffset boundary,
MultiXactOffset start, uint32 distance);
static bool SetOffsetVacuumLimit(bool is_startup);
static bool find_multixact_start(MultiXactId multi, MultiXactOffset *result);
-static void WriteMZeroPageXlogRec(int64 pageno, uint8 info);
+static XLogRecPtr WriteMZeroPageXlogRec(int64 pageno, uint8 info);
static void WriteMTruncateXlogRec(Oid oldestMultiDB,
MultiXactId startTruncOff,
MultiXactId endTruncOff,
@@ -884,7 +885,7 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
* take the trouble to generalize the slru.c error reporting code.
*/
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
*offptr = offset;
@@ -921,12 +922,12 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
}
memberptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
*memberptr = members[i].xid;
flagsptr = (uint32 *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
flagsval = *flagsptr;
flagsval &= ~(((1 << MXACT_MEMBER_BITS_PER_XACT) - 1) << bshift);
@@ -1348,7 +1349,7 @@ retry:
entryno = MultiXactIdToOffsetEntry(multi);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
@@ -1381,7 +1382,7 @@ retry:
if (pageno != prev_pageno)
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, tmpMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
nextMXOffset = *offptr;
@@ -1424,7 +1425,7 @@ retry:
}
xactptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
if (!TransactionIdIsValid(*xactptr))
{
@@ -1435,7 +1436,7 @@ retry:
flagsoff = MXOffsetToFlagsOffset(offset);
bshift = MXOffsetToFlagsBitShift(offset);
- flagsptr = (uint32 *) (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ flagsptr = (uint32 *) (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
ptr[truelength].xid = *xactptr;
ptr[truelength].status = (*flagsptr >> bshift) & MXACT_MEMBER_XACT_BITMASK;
@@ -1834,8 +1835,8 @@ MultiXactShmemSize(void)
mul_size(sizeof(MultiXactId) * 2, MaxOldestSlot))
size = SHARED_MULTIXACT_STATE_SIZE;
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS, 0));
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS));
return size;
}
@@ -1851,14 +1852,14 @@ MultiXactShmemInit(void)
MultiXactMemberCtl->PagePrecedes = MultiXactMemberPagePrecedes;
SimpleLruInit(MultiXactOffsetCtl,
- "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS, 0,
+ "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS,
MultiXactOffsetSLRULock, "pg_multixact/offsets",
LWTRANCHE_MULTIXACTOFFSET_BUFFER,
SYNC_HANDLER_MULTIXACT_OFFSET,
false);
SlruPagePrecedesUnitTests(MultiXactOffsetCtl, MULTIXACT_OFFSETS_PER_PAGE);
SimpleLruInit(MultiXactMemberCtl,
- "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS, 0,
+ "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS,
MultiXactMemberSLRULock, "pg_multixact/members",
LWTRANCHE_MULTIXACTMEMBER_BUFFER,
SYNC_HANDLER_MULTIXACT_MEMBER,
@@ -1933,11 +1934,17 @@ static int
ZeroMultiXactOffsetPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactOffsetCtl, pageno);
+ page = MultiXactOffsetCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -1949,11 +1956,17 @@ static int
ZeroMultiXactMemberPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactMemberCtl, pageno);
+ page = MultiXactMemberCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2071,10 +2084,10 @@ TrimMultiXact(void)
MultiXactOffset *offptr;
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, nextMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
- MemSet(offptr, 0, BLCKSZ - (entryno * sizeof(MultiXactOffset)));
+ MemSet(offptr, 0, SizeOfPageContents - (entryno * sizeof(MultiXactOffset)));
MultiXactOffsetCtl->shared->page_dirty[slotno] = true;
}
@@ -2104,9 +2117,9 @@ TrimMultiXact(void)
memberoff = MXOffsetToMemberOffset(offset);
slotno = SimpleLruReadPage(MultiXactMemberCtl, pageno, true, offset);
xidptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
- MemSet(xidptr, 0, BLCKSZ - memberoff);
+ MemSet(xidptr, 0, SizeOfPageContents - memberoff);
/*
* Note: we don't need to zero out the flag bits in the remaining
@@ -2758,7 +2771,7 @@ find_multixact_start(MultiXactId multi, MultiXactOffset *result)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(MultiXactOffsetCtl, pageno, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
LWLockRelease(MultiXactOffsetSLRULock);
@@ -3192,12 +3205,12 @@ MultiXactOffsetPrecedes(MultiXactOffset offset1, MultiXactOffset offset2)
* Write an xlog record reflecting the zeroing of either a MEMBERs or
* OFFSETs page (info shows which)
*/
-static void
+static XLogRecPtr
WriteMZeroPageXlogRec(int64 pageno, uint8 info)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_MULTIXACT_ID, info);
+ return XLogInsert(RM_MULTIXACT_ID, info);
}
/*
diff --git a/src/backend/access/transam/slru.c b/src/backend/access/transam/slru.c
index 7a371d9034..d7f3deea7d 100644
--- a/src/backend/access/transam/slru.c
+++ b/src/backend/access/transam/slru.c
@@ -57,6 +57,7 @@
#include "access/xlogutils.h"
#include "miscadmin.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/shmem.h"
@@ -154,13 +155,13 @@ typedef enum
SLRU_WRITE_FAILED,
SLRU_FSYNC_FAILED,
SLRU_CLOSE_FAILED,
+ SLRU_DATA_CORRUPTED,
} SlruErrorCause;
static SlruErrorCause slru_errcause;
static int slru_errno;
-static void SimpleLruZeroLSNs(SlruCtl ctl, int slotno);
static void SimpleLruWaitIO(SlruCtl ctl, int slotno);
static void SlruInternalWritePage(SlruCtl ctl, int slotno, SlruWriteAll fdata);
static bool SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno);
@@ -179,7 +180,7 @@ static void SlruInternalDeleteSegment(SlruCtl ctl, int64 segno);
*/
Size
-SimpleLruShmemSize(int nslots, int nlsns)
+SimpleLruShmemSize(int nslots)
{
Size sz;
@@ -192,9 +193,6 @@ SimpleLruShmemSize(int nslots, int nlsns)
sz += MAXALIGN(nslots * sizeof(int)); /* page_lru_count[] */
sz += MAXALIGN(nslots * sizeof(LWLockPadded)); /* buffer_locks[] */
- if (nlsns > 0)
- sz += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr)); /* group_lsn[] */
-
return BUFFERALIGN(sz) + BLCKSZ * nslots;
}
@@ -204,14 +202,13 @@ SimpleLruShmemSize(int nslots, int nlsns)
* ctl: address of local (unshared) control structure.
* name: name of SLRU. (This is user-visible, pick with care!)
* nslots: number of page slots to use.
- * nlsns: number of LSN groups per page (set to zero if not relevant).
* ctllock: LWLock to use to control access to the shared control structure.
* subdir: PGDATA-relative subdirectory that will contain the files.
* tranche_id: LWLock tranche ID to use for the SLRU's per-buffer LWLocks.
* sync_handler: which set of functions to use to handle sync requests
*/
void
-SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler, bool long_segment_names)
{
@@ -219,7 +216,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
bool found;
shared = (SlruShared) ShmemInitStruct(name,
- SimpleLruShmemSize(nslots, nlsns),
+ SimpleLruShmemSize(nslots),
&found);
if (!IsUnderPostmaster)
@@ -236,7 +233,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->ControlLock = ctllock;
shared->num_slots = nslots;
- shared->lsn_groups_per_page = nlsns;
shared->cur_lru_count = 0;
@@ -261,12 +257,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->buffer_locks = (LWLockPadded *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(LWLockPadded));
- if (nlsns > 0)
- {
- shared->group_lsn = (XLogRecPtr *) (ptr + offset);
- offset += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr));
- }
-
ptr += BUFFERALIGN(offset);
for (slotno = 0; slotno < nslots; slotno++)
{
@@ -281,7 +271,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
}
/* Should fit to estimated shmem size */
- Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots, nlsns));
+ Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots));
}
else
Assert(found);
@@ -323,11 +313,8 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
shared->page_dirty[slotno] = true;
SlruRecentlyUsed(shared, slotno);
- /* Set the buffer to zeroes */
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
-
- /* Set the LSNs for this new page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
+ /* Initialize the page. */
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
/* Assume this page is now the latest active page */
shared->latest_page_number = pageno;
@@ -338,26 +325,6 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
return slotno;
}
-/*
- * Zero all the LSNs we store for this slru page.
- *
- * This should be called each time we create a new page, and each time we read
- * in a page from disk into an existing buffer. (Such an old page cannot
- * have any interesting LSNs, since we'd have flushed them before writing
- * the page in the first place.)
- *
- * This assumes that InvalidXLogRecPtr is bitwise-all-0.
- */
-static void
-SimpleLruZeroLSNs(SlruCtl ctl, int slotno)
-{
- SlruShared shared = ctl->shared;
-
- if (shared->lsn_groups_per_page > 0)
- MemSet(&shared->group_lsn[slotno * shared->lsn_groups_per_page], 0,
- shared->lsn_groups_per_page * sizeof(XLogRecPtr));
-}
-
/*
* Wait for any active I/O on a page slot to finish. (This does not
* guarantee that new I/O hasn't been started before we return, though.
@@ -478,9 +445,6 @@ SimpleLruReadPage(SlruCtl ctl, int64 pageno, bool write_ok,
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
- /* Set the LSNs for this newly read-in page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
-
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
@@ -740,7 +704,7 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
ereport(LOG,
(errmsg("file \"%s\" doesn't exist, reading as zeroes",
path)));
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
return true;
}
@@ -763,6 +727,13 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
return false;
}
+ if (!PageIsVerifiedExtended(shared->page_buffer[slotno], pageno, PIV_REPORT_STAT))
+ {
+ slru_errcause = SLRU_DATA_CORRUPTED;
+ slru_errno = 0;
+ return false;
+ }
+
return true;
}
@@ -789,6 +760,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
off_t offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd = -1;
+ Page page = shared->page_buffer[slotno];
+ XLogRecPtr lsn;
/* update the stats counter of written pages */
pgstat_count_slru_page_written(shared->slru_stats_idx);
@@ -798,41 +771,18 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
* write out data before associated WAL records. This is the same action
* performed during FlushBuffer() in the main buffer manager.
*/
- if (shared->group_lsn != NULL)
+ lsn = PageGetLSN(page);
+ if (!XLogRecPtrIsInvalid(lsn))
{
/*
- * We must determine the largest async-commit LSN for the page. This
- * is a bit tedious, but since this entire function is a slow path
- * anyway, it seems better to do this here than to maintain a per-page
- * LSN variable (which'd need an extra comparison in the
- * transaction-commit path).
+ * As noted above, elog(ERROR) is not acceptable here, so if
+ * XLogFlush were to fail, we must PANIC. This isn't much of a
+ * restriction because XLogFlush is just about all critical
+ * section anyway, but let's make sure.
*/
- XLogRecPtr max_lsn;
- int lsnindex,
- lsnoff;
-
- lsnindex = slotno * shared->lsn_groups_per_page;
- max_lsn = shared->group_lsn[lsnindex++];
- for (lsnoff = 1; lsnoff < shared->lsn_groups_per_page; lsnoff++)
- {
- XLogRecPtr this_lsn = shared->group_lsn[lsnindex++];
-
- if (max_lsn < this_lsn)
- max_lsn = this_lsn;
- }
-
- if (!XLogRecPtrIsInvalid(max_lsn))
- {
- /*
- * As noted above, elog(ERROR) is not acceptable here, so if
- * XLogFlush were to fail, we must PANIC. This isn't much of a
- * restriction because XLogFlush is just about all critical
- * section anyway, but let's make sure.
- */
- START_CRIT_SECTION();
- XLogFlush(max_lsn);
- END_CRIT_SECTION();
- }
+ START_CRIT_SECTION();
+ XLogFlush(lsn);
+ END_CRIT_SECTION();
}
/*
@@ -899,6 +849,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
}
}
+ PageSetChecksumInplace(shared->page_buffer[slotno], pageno);
+
errno = 0;
pgstat_report_wait_start(WAIT_EVENT_SLRU_WRITE);
if (pg_pwrite(fd, shared->page_buffer[slotno], BLCKSZ, offset) != BLCKSZ)
@@ -1019,6 +971,13 @@ SlruReportIOError(SlruCtl ctl, int64 pageno, TransactionId xid)
errdetail("Could not close file \"%s\": %m.",
path)));
break;
+ case SLRU_DATA_CORRUPTED:
+ ereport(ERROR,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("could not access status of transaction %u", xid),
+ errdetail("Invalid page from file \"%s\" at offset %d.",
+ path, offset)));
+ break;
default:
/* can't get here, we trust */
elog(ERROR, "unrecognized SimpleLru error cause: %d",
diff --git a/src/backend/access/transam/subtrans.c b/src/backend/access/transam/subtrans.c
index 64673eaef6..c4a912171f 100644
--- a/src/backend/access/transam/subtrans.c
+++ b/src/backend/access/transam/subtrans.c
@@ -32,6 +32,7 @@
#include "access/subtrans.h"
#include "access/transam.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "utils/snapmgr.h"
@@ -49,7 +50,7 @@
*/
/* We need four bytes per xact */
-#define SUBTRANS_XACTS_PER_PAGE (BLCKSZ / sizeof(TransactionId))
+#define SUBTRANS_XACTS_PER_PAGE (SizeOfPageContents / sizeof(TransactionId))
/*
* Although we return an int64 the actual value can't currently exceed
@@ -93,7 +94,7 @@ SubTransSetParent(TransactionId xid, TransactionId parent)
LWLockAcquire(SubtransSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(SubTransCtl, pageno, true, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
/*
@@ -133,7 +134,7 @@ SubTransGetParent(TransactionId xid)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
parent = *ptr;
@@ -193,14 +194,14 @@ SubTransGetTopmostTransaction(TransactionId xid)
Size
SUBTRANSShmemSize(void)
{
- return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS, 0);
+ return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS);
}
void
SUBTRANSShmemInit(void)
{
SubTransCtl->PagePrecedes = SubTransPagePrecedes;
- SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS, 0,
+ SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS,
SubtransSLRULock, "pg_subtrans",
LWTRANCHE_SUBTRANS_BUFFER, SYNC_HANDLER_NONE,
false);
@@ -305,7 +306,6 @@ CheckPointSUBTRANS(void)
TRACE_POSTGRESQL_SUBTRANS_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that SUBTRANS has room for a newly-allocated XID.
*
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c
index 264f25a8f9..85b0d63cdc 100644
--- a/src/backend/commands/async.c
+++ b/src/backend/commands/async.c
@@ -140,6 +140,7 @@
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/proc.h"
@@ -162,7 +163,7 @@
* than that, so changes in that data structure won't affect user-visible
* restrictions.
*/
-#define NOTIFY_PAYLOAD_MAX_LENGTH (BLCKSZ - NAMEDATALEN - 128)
+#define NOTIFY_PAYLOAD_MAX_LENGTH (SizeOfPageContents - NAMEDATALEN - 128)
/*
* Struct representing an entry in the global notify queue
@@ -311,7 +312,7 @@ static SlruCtlData NotifyCtlData;
#define NotifyCtl (&NotifyCtlData)
#define QUEUE_PAGESIZE BLCKSZ
-
+#define QUEUE_PAGE_CAPACITY (QUEUE_PAGESIZE - MAXALIGN(SizeOfPageHeaderData))
#define QUEUE_FULL_WARN_INTERVAL 5000 /* warn at most once every 5s */
/*
@@ -492,7 +493,7 @@ AsyncShmemSize(void)
size = mul_size(MaxBackends + 1, sizeof(QueueBackendStatus));
size = add_size(size, offsetof(AsyncQueueControl, backend));
- size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS));
return size;
}
@@ -541,7 +542,7 @@ AsyncShmemInit(void)
* names are used in order to avoid wraparound.
*/
NotifyCtl->PagePrecedes = asyncQueuePagePrecedes;
- SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS, 0,
+ SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS,
NotifySLRULock, "pg_notify", LWTRANCHE_NOTIFY_BUFFER,
SYNC_HANDLER_NONE, true);
@@ -1301,14 +1302,14 @@ asyncQueueAdvance(volatile QueuePosition *position, int entryLength)
* written or read.
*/
offset += entryLength;
- Assert(offset <= QUEUE_PAGESIZE);
+ Assert(offset <= QUEUE_PAGE_CAPACITY);
/*
* In a second step check if another entry can possibly be written to the
* page. If so, stay here, we have reached the next position. If not, then
* we need to move on to the next page.
*/
- if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGESIZE)
+ if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGE_CAPACITY)
{
pageno++;
offset = 0;
@@ -1408,7 +1409,7 @@ asyncQueueAddEntries(ListCell *nextNotify)
offset = QUEUE_POS_OFFSET(queue_head);
/* Check whether the entry really fits on the current page */
- if (offset + qe.length <= QUEUE_PAGESIZE)
+ if (offset + qe.length <= QUEUE_PAGE_CAPACITY)
{
/* OK, so advance nextNotify past this item */
nextNotify = lnext(pendingNotifies->events, nextNotify);
@@ -1420,14 +1421,14 @@ asyncQueueAddEntries(ListCell *nextNotify)
* only check dboid and since it won't match any reader's database
* OID, they will ignore this entry and move on.
*/
- qe.length = QUEUE_PAGESIZE - offset;
+ qe.length = QUEUE_PAGE_CAPACITY - offset;
qe.dboid = InvalidOid;
qe.data[0] = '\0'; /* empty channel */
qe.data[1] = '\0'; /* empty payload */
}
/* Now copy qe into the shared buffer page */
- memcpy(NotifyCtl->shared->page_buffer[slotno] + offset,
+ memcpy(PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + offset,
&qe,
qe.length);
@@ -1947,10 +1948,10 @@ asyncQueueReadAllNotifications(void)
else
{
/* fetch all the rest of the page */
- copysize = QUEUE_PAGESIZE - curoffset;
+ copysize = QUEUE_PAGE_CAPACITY - curoffset;
}
- memcpy(page_buffer.buf + curoffset,
- NotifyCtl->shared->page_buffer[slotno] + curoffset,
+ memcpy(PageGetContents(page_buffer.buf) + curoffset,
+ PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + curoffset,
copysize);
/* Release lock that we got from SimpleLruReadPage_ReadOnly() */
LWLockRelease(NotifySLRULock);
@@ -2021,7 +2022,7 @@ asyncQueueProcessPageEntries(volatile QueuePosition *current,
if (QUEUE_POS_EQUAL(thisentry, stop))
break;
- qe = (AsyncQueueEntry *) (page_buffer + QUEUE_POS_OFFSET(thisentry));
+ qe = (AsyncQueueEntry *) (PageGetContents(page_buffer) + QUEUE_POS_OFFSET(thisentry));
/*
* Advance *current over this message, possibly to the next page. As
diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index ff8df7c0bc..39129bce98 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -204,6 +204,7 @@
#include "pgstat.h"
#include "port/pg_lfind.h"
#include "storage/bufmgr.h"
+#include "storage/bufpage.h"
#include "storage/predicate.h"
#include "storage/predicate_internals.h"
#include "storage/proc.h"
@@ -322,8 +323,8 @@ static SlruCtlData SerialSlruCtlData;
#define SerialSlruCtl (&SerialSlruCtlData)
#define SERIAL_PAGESIZE BLCKSZ
-#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
-#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE / SERIAL_ENTRYSIZE)
+#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
+#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE - MAXALIGN(SizeOfPageHeaderData) / SERIAL_ENTRYSIZE)
/*
* Set maximum pages based on the number needed to track all transactions.
@@ -333,7 +334,7 @@ static SlruCtlData SerialSlruCtlData;
#define SerialNextPage(page) (((page) >= SERIAL_MAX_PAGE) ? 0 : (page) + 1)
#define SerialValue(slotno, xid) (*((SerCommitSeqNo *) \
- (SerialSlruCtl->shared->page_buffer[slotno] + \
+ (PageGetContents(SerialSlruCtl->shared->page_buffer[slotno]) + \
((((uint32) (xid)) % SERIAL_ENTRIESPERPAGE) * SERIAL_ENTRYSIZE))))
#define SerialPage(xid) (((uint32) (xid)) / SERIAL_ENTRIESPERPAGE)
@@ -785,10 +786,13 @@ SerialPagePrecedesLogicallyUnitTests(void)
* requires burning ~2B XIDs in single-user mode, a negligible
* possibility. Moreover, if it does happen, the consequence would be
* mild, namely a new transaction failing in SimpleLruReadPage().
+ *
+ * NOTE: After adding the page header, the defect affects two pages.
+ * We now assert correct treatment of its second to prior page.
*/
headPage = oldestPage;
targetPage = newestPage;
- Assert(SerialPagePrecedesLogically(headPage, targetPage - 1));
+ Assert(SerialPagePrecedesLogically(headPage, targetPage - 2));
#if 0
Assert(SerialPagePrecedesLogically(headPage, targetPage));
#endif
@@ -808,7 +812,7 @@ SerialInit(void)
*/
SerialSlruCtl->PagePrecedes = SerialPagePrecedesLogically;
SimpleLruInit(SerialSlruCtl, "Serial",
- NUM_SERIAL_BUFFERS, 0, SerialSLRULock, "pg_serial",
+ NUM_SERIAL_BUFFERS, SerialSLRULock, "pg_serial",
LWTRANCHE_SERIAL_BUFFER, SYNC_HANDLER_NONE,
false);
#ifdef USE_ASSERT_CHECKING
@@ -1348,7 +1352,7 @@ PredicateLockShmemSize(void)
/* Shared memory structures for SLRU tracking of old committed xids. */
size = add_size(size, sizeof(SerialControlData));
- size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS));
return size;
}
diff --git a/src/backend/storage/page/bufpage.c b/src/backend/storage/page/bufpage.c
index 9a302ddc30..723a127594 100644
--- a/src/backend/storage/page/bufpage.c
+++ b/src/backend/storage/page/bufpage.c
@@ -59,6 +59,31 @@ PageInit(Page page, Size pageSize, Size specialSize)
/* p->pd_prune_xid = InvalidTransactionId; done by above MemSet */
}
+/*
+ * PageInitSLRU
+ * Initializes the contents of an SLRU page.
+ * Note that we don't calculate an initial checksum here; that's not done
+ * until it's time to write.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
/*
* PageIsVerifiedExtended
diff --git a/src/bin/pg_checksums/pg_checksums.c b/src/bin/pg_checksums/pg_checksums.c
index 6543d9ce08..cfbc239843 100644
--- a/src/bin/pg_checksums/pg_checksums.c
+++ b/src/bin/pg_checksums/pg_checksums.c
@@ -16,6 +16,7 @@
#include <dirent.h>
#include <limits.h>
+#include <stdbool.h>
#include <time.h>
#include <sys/stat.h>
#include <unistd.h>
@@ -589,12 +590,20 @@ main(int argc, char *argv[])
{
total_size = scan_directory(DataDir, "global", true);
total_size += scan_directory(DataDir, "base", true);
+ total_size += scan_directory(DataDir, "pg_commit_ts", true);
+ total_size += scan_directory(DataDir, "pg_multixact", true);
+ total_size += scan_directory(DataDir, "pg_serial", true);
total_size += scan_directory(DataDir, "pg_tblspc", true);
+ total_size += scan_directory(DataDir, "pg_xact", true);
}
(void) scan_directory(DataDir, "global", false);
(void) scan_directory(DataDir, "base", false);
+ (void) scan_directory(DataDir, "pg_commit_ts", false);
+ (void) scan_directory(DataDir, "pg_multixact", false);
+ (void) scan_directory(DataDir, "pg_serial", false);
(void) scan_directory(DataDir, "pg_tblspc", false);
+ (void) scan_directory(DataDir, "pg_xact", false);
if (showprogress)
progress_report(true);
diff --git a/src/bin/pg_resetwal/t/001_basic.pl b/src/bin/pg_resetwal/t/001_basic.pl
index 18d0882cb1..ae74828e44 100644
--- a/src/bin/pg_resetwal/t/001_basic.pl
+++ b/src/bin/pg_resetwal/t/001_basic.pl
@@ -206,7 +206,7 @@ push @cmd,
sprintf("%d,%d", hex($files[0]) == 0 ? 3 : hex($files[0]), hex($files[-1]));
@files = get_slru_files('pg_multixact/offsets');
-$mult = 32 * $blcksz / 4;
+$mult = 32 * ($blcksz - 24) / 4;
# -m argument is "new,old"
push @cmd, '-m',
sprintf("%d,%d",
@@ -214,11 +214,11 @@ push @cmd, '-m',
hex($files[0]) == 0 ? 1 : hex($files[0] * $mult));
@files = get_slru_files('pg_multixact/members');
-$mult = 32 * int($blcksz / 20) * 4;
+$mult = 32 * int(($blcksz - 24) / 20) * 4;
push @cmd, '-O', (hex($files[-1]) + 1) * $mult;
@files = get_slru_files('pg_xact');
-$mult = 32 * $blcksz * 4;
+$mult = 32 * ($blcksz - 24) * 4;
push @cmd,
'-u', (hex($files[0]) == 0 ? 3 : hex($files[0]) * $mult),
'-x', ((hex($files[-1]) + 1) * $mult);
diff --git a/src/bin/pg_upgrade/Makefile b/src/bin/pg_upgrade/Makefile
index bde91e2beb..02438f5884 100644
--- a/src/bin/pg_upgrade/Makefile
+++ b/src/bin/pg_upgrade/Makefile
@@ -24,6 +24,7 @@ OBJS = \
pg_upgrade.o \
relfilenumber.o \
server.o \
+ slru.o \
tablespace.o \
util.o \
version.o
diff --git a/src/bin/pg_upgrade/meson.build b/src/bin/pg_upgrade/meson.build
index 3e8a08e062..d216731864 100644
--- a/src/bin/pg_upgrade/meson.build
+++ b/src/bin/pg_upgrade/meson.build
@@ -13,6 +13,7 @@ pg_upgrade_sources = files(
'pg_upgrade.c',
'relfilenumber.c',
'server.c',
+ 'slru.c',
'tablespace.c',
'util.c',
'version.c',
diff --git a/src/bin/pg_upgrade/nls.mk b/src/bin/pg_upgrade/nls.mk
index 9e2c1386e2..f5ad9d1397 100644
--- a/src/bin/pg_upgrade/nls.mk
+++ b/src/bin/pg_upgrade/nls.mk
@@ -11,6 +11,7 @@ GETTEXT_FILES = check.c \
parallel.c \
pg_upgrade.c \
relfilenumber.c \
+ slru.c \
server.c \
tablespace.c \
util.c \
diff --git a/src/bin/pg_upgrade/pg_upgrade.c b/src/bin/pg_upgrade/pg_upgrade.c
index 3960af4036..2b7d01058e 100644
--- a/src/bin/pg_upgrade/pg_upgrade.c
+++ b/src/bin/pg_upgrade/pg_upgrade.c
@@ -660,14 +660,23 @@ copy_subdir_files(const char *old_subdir, const char *new_subdir)
static void
copy_xact_xlog_xid(void)
{
+ bool slru_header_changed = false;
+
/*
* Copy old commit logs to new data dir. pg_clog has been renamed to
* pg_xact in post-10 clusters.
*/
- copy_subdir_files(GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact",
- GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact");
+ char *xact_old_dir = GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+ char *xact_new_dir = GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+
+ if (new_cluster.controldata.cat_ver >= SLRU_PAGE_HEADER_CAT_VER &&
+ old_cluster.controldata.cat_ver < SLRU_PAGE_HEADER_CAT_VER)
+ slru_header_changed = true;
+
+ if (slru_header_changed)
+ upgrade_xact_cache(xact_old_dir, xact_new_dir);
+ else
+ copy_subdir_files(xact_old_dir, xact_new_dir);
prep_status("Setting oldest XID for new cluster");
exec_prog(UTILITY_LOG_FILE, NULL, true, true,
@@ -702,7 +711,8 @@ copy_xact_xlog_xid(void)
* server doesn't attempt to read multis older than the cutoff value.
*/
if (old_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
- new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
+ !slru_header_changed)
{
copy_subdir_files("pg_multixact/offsets", "pg_multixact/offsets");
copy_subdir_files("pg_multixact/members", "pg_multixact/members");
@@ -722,7 +732,8 @@ copy_xact_xlog_xid(void)
new_cluster.pgdata);
check_ok();
}
- else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER ||
+ slru_header_changed)
{
/*
* Remove offsets/0000 file created by initdb that no longer matches
diff --git a/src/bin/pg_upgrade/pg_upgrade.h b/src/bin/pg_upgrade/pg_upgrade.h
index a710f325de..bb04224dee 100644
--- a/src/bin/pg_upgrade/pg_upgrade.h
+++ b/src/bin/pg_upgrade/pg_upgrade.h
@@ -115,6 +115,11 @@ extern char *output_files[];
*/
#define MULTIXACT_FORMATCHANGE_CAT_VER 201301231
+/*
+ * A page header was added to each SLRU page in 17.0.
+ */
+#define SLRU_PAGE_HEADER_CAT_VER 202312091
+
/*
* large object chunk size added to pg_controldata,
* commit 5f93c37805e7485488480916b4585e098d3cc883
@@ -454,6 +459,11 @@ uint32 get_major_server_version(ClusterInfo *cluster);
void check_pghost_envvar(void);
+/* slru.c */
+
+void upgrade_xact_cache(const char *src_subdir, const char *dst_subdir);
+
+
/* util.c */
char *quote_identifier(const char *s);
diff --git a/src/bin/pg_upgrade/slru.c b/src/bin/pg_upgrade/slru.c
new file mode 100644
index 0000000000..3d23175640
--- /dev/null
+++ b/src/bin/pg_upgrade/slru.c
@@ -0,0 +1,205 @@
+/*
+ * slru.c
+ *
+ * SLRU functions
+ *
+ * Copyright (c) 2010-2023, PostgreSQL Global Development Group
+ * src/bin/pg_upgrade/slru.c
+ */
+
+#include <dirent.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#include "postgres_fe.h"
+#include "storage/bufpage.h"
+#include "storage/checksum.h"
+
+#include "pg_upgrade.h"
+
+
+/*
+ * Copy SLRU_PAGES_PER_SEGMENT from access/slru.h to avoid including it.
+ */
+#define SLRU_PAGES_PER_SEGMENT 32
+
+#define SEGMENT_SIZE (BLCKSZ * SLRU_PAGES_PER_SEGMENT)
+
+/*
+ * Copy PageInitSLRU from storage/bufpage.c to avoid linking to the backend.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
+
+/*
+ * Filter function for scandir(3) to select only segment files.
+ */
+static int
+segment_file_filter(const struct dirent *dirent)
+{
+ return strspn(dirent->d_name, "0123456789ABCDEF") == strlen(dirent->d_name);
+}
+
+/*
+ * Compare function for scandir(3) to sort segment files.
+ */
+static int
+segment_file_compare(const struct dirent **a, const struct dirent **b)
+{
+ long segno_a;
+ long segno_b;
+
+ segno_a = strtol((*a)->d_name, NULL, 16);
+ segno_b = strtol((*b)->d_name, NULL, 16);
+ return segno_a - segno_b;
+}
+
+static void
+upgrade_file(const char *src_dir, const char *src_file, const char *dst_dir)
+{
+ char src[MAXPGPATH];
+ char dst[MAXPGPATH];
+
+ int seg_name_len;
+ int src_segno;
+ int64 src_pageno;
+ int dst_segno;
+ int64 dst_pageno;
+ int dst_offset;
+
+ int src_fd;
+ int dst_fd;
+
+ char *src_buf;
+ ssize_t src_len;
+ ssize_t src_buf_offset;
+ PGAlignedBlock dst_block;
+ Page page = dst_block.data;
+ int len_to_copy;
+
+ seg_name_len = strlen(src_file);
+ src_segno = (int) strtol(src_file, NULL, 16);
+ src_pageno = src_segno * SLRU_PAGES_PER_SEGMENT;
+
+ dst_pageno = src_pageno * BLCKSZ / SizeOfPageContents;
+ dst_offset = src_pageno * BLCKSZ - dst_pageno * SizeOfPageContents;
+ dst_segno = dst_pageno / SLRU_PAGES_PER_SEGMENT;
+
+ snprintf(src, sizeof(src), "%s/%s", src_dir, src_file);
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+
+ src_buf = pg_malloc(SEGMENT_SIZE);
+ if ((src_fd = open(src, O_RDONLY | PG_BINARY, 0)) == -1)
+ pg_fatal("could not open file \"%s\": %s", src, strerror(errno));
+ if ((src_len = read(src_fd, src_buf, SEGMENT_SIZE)) == -1)
+ pg_fatal("could not read file \"%s\": %s", src, strerror(errno));
+
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+
+ /*
+ * Read the destination page at dst_pageno into the buffer. The page may contain
+ * data from the previous source segment. Initialize the page if the page is new.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+ if (read(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not read file \"%s\": %s", dst, strerror(errno));
+ if (PageIsNew(page))
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Rewind the file position, so the first write will overwrite the page.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+
+ src_buf_offset = 0;
+ while (src_buf_offset < src_len)
+ {
+ len_to_copy = Min(src_len - src_buf_offset, SizeOfPageContents - dst_offset);
+ memcpy(PageGetContents(page) + dst_offset, src_buf + src_buf_offset, len_to_copy);
+ src_buf_offset += len_to_copy;
+
+ if (new_cluster.controldata.data_checksum_version > 0)
+ ((PageHeader) page)->pd_checksum = pg_checksum_page(page, dst_pageno);
+ if (write(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not write file \"%s\": %s", dst, strerror(errno));
+
+ dst_pageno++;
+ dst_offset = 0;
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Switch segments if we reached the end of the current segment.
+ */
+ if (dst_pageno % SLRU_PAGES_PER_SEGMENT == 0)
+ {
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ dst_segno++;
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+ }
+ }
+
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ pg_free(src_buf);
+ close(src_fd);
+}
+
+void
+upgrade_xact_cache(const char *src_subdir, const char *dst_subdir)
+{
+ char src_dir[MAXPGPATH];
+ char dst_dir[MAXPGPATH];
+
+ struct dirent **src_segment_files;
+ int num_files;
+ int i;
+
+ snprintf(src_dir, sizeof(src_dir), "%s/%s", old_cluster.pgdata, src_subdir);
+ snprintf(dst_dir, sizeof(dst_dir), "%s/%s", new_cluster.pgdata, dst_subdir);
+
+ num_files = scandir(src_dir, &src_segment_files, segment_file_filter, segment_file_compare);
+ if (num_files == -1)
+ pg_fatal("could not scan directory \"%s\": %s", src_dir, strerror(errno));
+
+ for (i = 0; i < num_files; i++)
+ {
+ upgrade_file(src_dir, src_segment_files[i]->d_name, dst_dir);
+ free(src_segment_files[i]);
+ }
+ free(src_segment_files);
+}
diff --git a/src/include/access/slru.h b/src/include/access/slru.h
index 091e2202c9..24733166b8 100644
--- a/src/include/access/slru.h
+++ b/src/include/access/slru.h
@@ -68,17 +68,6 @@ typedef struct SlruSharedData
int *page_lru_count;
LWLockPadded *buffer_locks;
- /*
- * Optional array of WAL flush LSNs associated with entries in the SLRU
- * pages. If not zero/NULL, we must flush WAL before writing pages (true
- * for pg_xact, false for multixact, pg_subtrans, pg_notify). group_lsn[]
- * has lsn_groups_per_page entries per buffer slot, each containing the
- * highest LSN known for a contiguous group of SLRU entries on that slot's
- * page.
- */
- XLogRecPtr *group_lsn;
- int lsn_groups_per_page;
-
/*----------
* We mark a page "most recently used" by setting
* page_lru_count[slotno] = ++cur_lru_count;
@@ -147,8 +136,8 @@ typedef struct SlruCtlData
typedef SlruCtlData *SlruCtl;
-extern Size SimpleLruShmemSize(int nslots, int nlsns);
-extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+extern Size SimpleLruShmemSize(int nslots);
+extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler,
bool long_segment_names);
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index c4c59bfe6f..ec823e6d1a 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
*/
/* yyyymmddN */
-#define CATALOG_VERSION_NO 202311271
+#define CATALOG_VERSION_NO 202312091
#endif
diff --git a/src/include/storage/bufpage.h b/src/include/storage/bufpage.h
index 424ecba028..251d9523fa 100644
--- a/src/include/storage/bufpage.h
+++ b/src/include/storage/bufpage.h
@@ -201,6 +201,7 @@ typedef PageHeaderData *PageHeader;
* handling pages.
*/
#define PG_PAGE_LAYOUT_VERSION 4
+#define PG_SLRU_PAGE_LAYOUT_VERSION 1
#define PG_DATA_CHECKSUM_VERSION 1
/* ----------------------------------------------------------------
@@ -257,6 +258,11 @@ PageGetContents(Page page)
return (char *) page + MAXALIGN(SizeOfPageHeaderData);
}
+/*
+ * Space available for storing page contents.
+ */
+#define SizeOfPageContents (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
+
/* ----------------
* functions to access page size info
* ----------------
@@ -486,6 +492,7 @@ StaticAssertDecl(BLCKSZ == ((BLCKSZ / sizeof(size_t)) * sizeof(size_t)),
"BLCKSZ has to be a multiple of sizeof(size_t)");
extern void PageInit(Page page, Size pageSize, Size specialSize);
+extern void PageInitSLRU(Page page, Size pageSize, Size specialSize);
extern bool PageIsVerifiedExtended(Page page, BlockNumber blkno, int flags);
extern OffsetNumber PageAddItemExtended(Page page, Item item, Size size,
OffsetNumber offsetNumber, int flags);
diff --git a/src/test/modules/test_slru/test_slru.c b/src/test/modules/test_slru/test_slru.c
index d0fb9444e8..aa8a29d92e 100644
--- a/src/test/modules/test_slru/test_slru.c
+++ b/src/test/modules/test_slru/test_slru.c
@@ -17,6 +17,7 @@
#include "access/slru.h"
#include "access/transam.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/shmem.h"
@@ -76,8 +77,8 @@ test_slru_page_write(PG_FUNCTION_ARGS)
TestSlruCtl->shared->page_status[slotno] = SLRU_PAGE_VALID;
/* write given data to the page, up to the limit of the page */
- strncpy(TestSlruCtl->shared->page_buffer[slotno], data,
- BLCKSZ - 1);
+ strncpy(PageGetContents(TestSlruCtl->shared->page_buffer[slotno]), data,
+ SizeOfPageContents - 1);
SimpleLruWritePage(TestSlruCtl, slotno);
LWLockRelease(TestSLRULock);
@@ -104,7 +105,7 @@ test_slru_page_read(PG_FUNCTION_ARGS)
LWLockAcquire(TestSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(TestSlruCtl, pageno,
write_ok, InvalidTransactionId);
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -122,7 +123,7 @@ test_slru_page_readonly(PG_FUNCTION_ARGS)
pageno,
InvalidTransactionId);
Assert(LWLockHeldByMe(TestSLRULock));
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -202,7 +203,7 @@ test_slru_shmem_request(void)
prev_shmem_request_hook();
/* reserve shared memory for the test SLRU */
- RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS, 0));
+ RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS));
}
static bool
@@ -238,7 +239,7 @@ test_slru_shmem_startup(void)
TestSlruCtl->PagePrecedes = test_slru_page_precedes_logically;
SimpleLruInit(TestSlruCtl, "TestSLRU",
- NUM_TEST_BUFFERS, 0, TestSLRULock, slru_dir_name,
+ NUM_TEST_BUFFERS, TestSLRULock, slru_dir_name,
test_tranche_id, SYNC_HANDLER_NONE, long_segment_names);
}
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2023-12-19 02:23 ` Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Bagga, Rishu @ 2023-12-19 02:23 UTC (permalink / raw)
To: Li, Yong <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; +Cc: Andrey Borodin <[email protected]>; PostgreSQL Hackers <[email protected]>; Aleksander Alekseev <[email protected]>; Shyrabokau, Anton <[email protected]>
On Thu, Dec 8, 2023 at 1:36 AM Li, Yong <[email protected]> wrote:
>Given so many different approaches were discussed, I have started a
>wiki to record and collaborate all efforts towards SLRU
>improvements. The wiki provides a concise overview of all the ideas
>discussed and can serve as a portal for all historical
>discussions. Currently, the wiki summarizes four recent threads
>ranging from identifier format change to page header change, to moving
>SLRU into the main buffer pool, to reduce lock contention on SLRU
>latches. We can keep the patch related discussions in this thread and
>use the wiki as a live document for larger scale collaborations.
>The wiki page is
>here: https://wiki.postgresql.org/wiki/SLRU_improvements
>Regarding the benefits of this patch, here is a detailed explanation:
1. Checksum is added to each page, allowing us to verify if a page has
been corrupted when read from the disk.
2. The ad-hoc LSN group structure is removed from the SLRU cache
control data and is replaced by the page LSN in the page header.
This allows us to use the same WAL protocol as used by pages in the
main buffer pool: flush all redo logs up to the page LSN before
flushing the page itself. If we move SLRU caches into the main
buffer pool, this change fits naturally.
3. It leaves further optimizations open. We can continue to pursue the
goal of moving SLRU into the main buffer pool, or we can follow the
lock partition idea. This change by itself does not conflict with
either proposal.
>Also, the patch is now complete and is ready for review. All check-
>world tests including tap tests passed with this patch.
Hi Yong,
I agree we should break the effort for the SLRU optimization into
smaller chunks after having worked on some of the bigger patches and
facing difficulty in making progress that way.
The patch looks mostly good to me; though one thing that I thought about
differently with the upgrade portion is where we should keep the logic
of re-writing the CLOG files.
There is a precedent introduced back in Postgres v9.6 in making on disk
page format changes across different in visibility map: [1]
code comment:
* In versions of PostgreSQL prior to catversion 201603011, PostgreSQL's
* visibility map included one bit per heap page; it now includes two.
* When upgrading a cluster from before that time to a current PostgreSQL
* version, we could refuse to copy visibility maps from the old cluster
* to the new cluster; the next VACUUM would recreate them, but at the
* price of scanning the entire table. So, instead, we rewrite the old
* visibility maps in the new format.
This work is being done in file.c – it seems to me the proper way to
proceed would be to continue writing on-disk upgrade logic here.
Besides that this looks good to me, would like to hear what others have to say.
Thanks,
Rishu Bagga
Amazon Web Services (AWS)
[1] https://github.com/postgres/postgres/commit/7087166a88fe0c04fc6636d0d6d6bea1737fc1fb
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
@ 2023-12-19 07:28 ` Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2023-12-19 07:28 UTC (permalink / raw)
To: Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; +Cc: Andrey Borodin <[email protected]>; PostgreSQL Hackers <[email protected]>; Aleksander Alekseev <[email protected]>; Shyrabokau, Anton <[email protected]>
> This work is being done in file.c – it seems to me the proper way to
> proceed would be to continue writing on-disk upgrade logic here.
> Besides that this looks good to me, would like to hear what others have to say.
Thank you, Rishu for taking time to review the code. I've updated the patch
and moved the on-disk upgrade logic to pg_upgrade/file.c.
I have also added this thread to the current Commitfest and hope this patch
will be part of the 17 release.
The commitfest link:
https://commitfest.postgresql.org/46/4709/
Regards,
Yong,
Attachments:
[application/octet-stream] slur_page_header_v2.patch (53.6K, ../../[email protected]/2-slur_page_header_v2.patch)
download | inline diff:
src/backend/access/transam/clog.c | 52 +++++----
src/backend/access/transam/commit_ts.c | 26 +++--
src/backend/access/transam/multixact.c | 63 ++++++-----
src/backend/access/transam/slru.c | 113 +++++++-------------
src/backend/access/transam/subtrans.c | 12 +--
src/backend/commands/async.c | 27 ++---
src/backend/storage/lmgr/predicate.c | 16 +--
src/backend/storage/page/bufpage.c | 25 +++++
src/bin/pg_checksums/pg_checksums.c | 9 ++
src/bin/pg_resetwal/t/001_basic.pl | 6 +-
src/bin/pg_upgrade/file.c | 186 +++++++++++++++++++++++++++++++++
src/bin/pg_upgrade/pg_upgrade.c | 23 ++--
src/bin/pg_upgrade/pg_upgrade.h | 6 ++
src/include/access/slru.h | 15 +--
src/include/catalog/catversion.h | 2 +-
src/include/storage/bufpage.h | 7 ++
src/test/modules/test_slru/test_slru.c | 13 +--
17 files changed, 409 insertions(+), 192 deletions(-)
diff --git a/src/backend/access/transam/clog.c b/src/backend/access/transam/clog.c
index 7dca1df61b..b7f690754b 100644
--- a/src/backend/access/transam/clog.c
+++ b/src/backend/access/transam/clog.c
@@ -41,6 +41,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/proc.h"
#include "storage/sync.h"
@@ -59,7 +60,7 @@
/* We need two bits per xact, so four xacts fit in a byte */
#define CLOG_BITS_PER_XACT 2
#define CLOG_XACTS_PER_BYTE 4
-#define CLOG_XACTS_PER_PAGE (BLCKSZ * CLOG_XACTS_PER_BYTE)
+#define CLOG_XACTS_PER_PAGE (SizeOfPageContents * CLOG_XACTS_PER_BYTE)
#define CLOG_XACT_BITMASK ((1 << CLOG_BITS_PER_XACT) - 1)
@@ -77,13 +78,6 @@ TransactionIdToPage(TransactionId xid)
#define TransactionIdToByte(xid) (TransactionIdToPgIndex(xid) / CLOG_XACTS_PER_BYTE)
#define TransactionIdToBIndex(xid) ((xid) % (TransactionId) CLOG_XACTS_PER_BYTE)
-/* We store the latest async LSN for each group of transactions */
-#define CLOG_XACTS_PER_LSN_GROUP 32 /* keep this a power of 2 */
-#define CLOG_LSNS_PER_PAGE (CLOG_XACTS_PER_PAGE / CLOG_XACTS_PER_LSN_GROUP)
-
-#define GetLSNIndex(slotno, xid) ((slotno) * CLOG_LSNS_PER_PAGE + \
- ((xid) % (TransactionId) CLOG_XACTS_PER_PAGE) / CLOG_XACTS_PER_LSN_GROUP)
-
/*
* The number of subtransactions below which we consider to apply clog group
* update optimization. Testing reveals that the number higher than this can
@@ -101,7 +95,7 @@ static SlruCtlData XactCtlData;
static int ZeroCLOGPage(int64 pageno, bool writeXlog);
static bool CLOGPagePrecedes(int64 page1, int64 page2);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXact,
Oid oldestXactDb);
static void TransactionIdSetPageStatus(TransactionId xid, int nsubxids,
@@ -583,8 +577,9 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
char *byteptr;
char byteval;
char curval;
+ Page page = XactCtl->shared->page_buffer[slotno];
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(page) + byteno;
curval = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
/*
@@ -613,7 +608,7 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*byteptr = byteval;
/*
- * Update the group LSN if the transaction completion LSN is higher.
+ * Update the page LSN if the transaction completion LSN is higher.
*
* Note: lsn will be invalid when supplied during InRecovery processing,
* so we don't need to do anything special to avoid LSN updates during
@@ -622,10 +617,8 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*/
if (!XLogRecPtrIsInvalid(lsn))
{
- int lsnindex = GetLSNIndex(slotno, xid);
-
- if (XactCtl->shared->group_lsn[lsnindex] < lsn)
- XactCtl->shared->group_lsn[lsnindex] = lsn;
+ if (PageGetLSN(page) < lsn)
+ PageSetLSN(page, lsn);
}
}
@@ -651,19 +644,19 @@ TransactionIdGetStatus(TransactionId xid, XLogRecPtr *lsn)
int byteno = TransactionIdToByte(xid);
int bshift = TransactionIdToBIndex(xid) * CLOG_BITS_PER_XACT;
int slotno;
- int lsnindex;
+ Page page;
char *byteptr;
XidStatus status;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(XactCtl, pageno, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
status = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
- lsnindex = GetLSNIndex(slotno, xid);
- *lsn = XactCtl->shared->group_lsn[lsnindex];
+ *lsn = PageGetLSN(page);
LWLockRelease(XactSLRULock);
@@ -698,14 +691,14 @@ CLOGShmemBuffers(void)
Size
CLOGShmemSize(void)
{
- return SimpleLruShmemSize(CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE);
+ return SimpleLruShmemSize(CLOGShmemBuffers());
}
void
CLOGShmemInit(void)
{
XactCtl->PagePrecedes = CLOGPagePrecedes;
- SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE,
+ SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(),
XactSLRULock, "pg_xact", LWTRANCHE_XACT_BUFFER,
SYNC_HANDLER_CLOG, false);
SlruPagePrecedesUnitTests(XactCtl, CLOG_XACTS_PER_PAGE);
@@ -747,11 +740,17 @@ static int
ZeroCLOGPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(XactCtl, pageno);
+ page = XactCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -807,12 +806,12 @@ TrimCLOG(void)
char *byteptr;
slotno = SimpleLruReadPage(XactCtl, pageno, false, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(XactCtl->shared->page_buffer[slotno]) + byteno;
/* Zero so-far-unused positions in the current byte */
*byteptr &= (1 << bshift) - 1;
/* Zero the rest of the page */
- MemSet(byteptr + 1, 0, BLCKSZ - byteno - 1);
+ MemSet(byteptr + 1, 0, SizeOfPageContents - byteno - 1);
XactCtl->shared->page_dirty[slotno] = true;
}
@@ -836,7 +835,6 @@ CheckPointCLOG(void)
TRACE_POSTGRESQL_CLOG_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that CLOG has room for a newly-allocated XID.
*
@@ -958,12 +956,12 @@ CLOGPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
+ return XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/commit_ts.c b/src/backend/access/transam/commit_ts.c
index e6fd9b3349..367459ea04 100644
--- a/src/backend/access/transam/commit_ts.c
+++ b/src/backend/access/transam/commit_ts.c
@@ -31,6 +31,7 @@
#include "funcapi.h"
#include "miscadmin.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "storage/shmem.h"
#include "utils/builtins.h"
#include "utils/snapmgr.h"
@@ -63,7 +64,7 @@ typedef struct CommitTimestampEntry
sizeof(RepOriginId))
#define COMMIT_TS_XACTS_PER_PAGE \
- (BLCKSZ / SizeOfCommitTimestampEntry)
+ (SizeOfPageContents / SizeOfCommitTimestampEntry)
/*
@@ -120,7 +121,7 @@ static int ZeroCommitTsPage(int64 pageno, bool writeXlog);
static bool CommitTsPagePrecedes(int64 page1, int64 page2);
static void ActivateCommitTs(void);
static void DeactivateCommitTs(void);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXid);
/*
@@ -254,11 +255,12 @@ TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
+ Assert(xid == slotno * COMMIT_TS_XACTS_PER_PAGE + entryno);
entry.time = ts;
entry.nodeid = nodeid;
- memcpy(CommitTsCtl->shared->page_buffer[slotno] +
+ memcpy(PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
@@ -337,7 +339,7 @@ TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
- CommitTsCtl->shared->page_buffer[slotno] +
+ PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
@@ -515,7 +517,7 @@ CommitTsShmemBuffers(void)
Size
CommitTsShmemSize(void)
{
- return SimpleLruShmemSize(CommitTsShmemBuffers(), 0) +
+ return SimpleLruShmemSize(CommitTsShmemBuffers()) +
sizeof(CommitTimestampShared);
}
@@ -529,7 +531,7 @@ CommitTsShmemInit(void)
bool found;
CommitTsCtl->PagePrecedes = CommitTsPagePrecedes;
- SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(), 0,
+ SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(),
CommitTsSLRULock, "pg_commit_ts",
LWTRANCHE_COMMITTS_BUFFER,
SYNC_HANDLER_COMMIT_TS,
@@ -582,11 +584,17 @@ static int
ZeroCommitTsPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(CommitTsCtl, pageno);
+ page = CommitTsCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -946,12 +954,12 @@ CommitTsPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
+ return XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c
index db3423f12e..0970aa8855 100644
--- a/src/backend/access/transam/multixact.c
+++ b/src/backend/access/transam/multixact.c
@@ -83,6 +83,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "postmaster/autovacuum.h"
+#include "storage/bufpage.h"
#include "storage/lmgr.h"
#include "storage/pmsignal.h"
#include "storage/proc.h"
@@ -106,7 +107,7 @@
*/
/* We need four bytes per offset */
-#define MULTIXACT_OFFSETS_PER_PAGE (BLCKSZ / sizeof(MultiXactOffset))
+#define MULTIXACT_OFFSETS_PER_PAGE (SizeOfPageContents / sizeof(MultiXactOffset))
#define MultiXactIdToOffsetPage(xid) \
((xid) / (MultiXactOffset) MULTIXACT_OFFSETS_PER_PAGE)
@@ -119,8 +120,8 @@
* additional flag bits for each TransactionId. To do this without getting
* into alignment issues, we store four bytes of flags, and then the
* corresponding 4 Xids. Each such 5-word (20-byte) set we call a "group", and
- * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 409 groups
- * per page. This wastes 12 bytes per page, but that's OK -- simplicity (and
+ * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 408 groups
+ * per page. This wastes 8 bytes per page, but that's OK -- simplicity (and
* performance) trumps space efficiency here.
*
* Note that the "offset" macros work with byte offset, not array indexes, so
@@ -138,7 +139,7 @@
/* size in bytes of a complete group */
#define MULTIXACT_MEMBERGROUP_SIZE \
(sizeof(TransactionId) * MULTIXACT_MEMBERS_PER_MEMBERGROUP + MULTIXACT_FLAGBYTES_PER_GROUP)
-#define MULTIXACT_MEMBERGROUPS_PER_PAGE (BLCKSZ / MULTIXACT_MEMBERGROUP_SIZE)
+#define MULTIXACT_MEMBERGROUPS_PER_PAGE (SizeOfPageContents / MULTIXACT_MEMBERGROUP_SIZE)
#define MULTIXACT_MEMBERS_PER_PAGE \
(MULTIXACT_MEMBERGROUPS_PER_PAGE * MULTIXACT_MEMBERS_PER_MEMBERGROUP)
@@ -366,7 +367,7 @@ static bool MultiXactOffsetWouldWrap(MultiXactOffset boundary,
MultiXactOffset start, uint32 distance);
static bool SetOffsetVacuumLimit(bool is_startup);
static bool find_multixact_start(MultiXactId multi, MultiXactOffset *result);
-static void WriteMZeroPageXlogRec(int64 pageno, uint8 info);
+static XLogRecPtr WriteMZeroPageXlogRec(int64 pageno, uint8 info);
static void WriteMTruncateXlogRec(Oid oldestMultiDB,
MultiXactId startTruncOff,
MultiXactId endTruncOff,
@@ -884,7 +885,7 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
* take the trouble to generalize the slru.c error reporting code.
*/
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
*offptr = offset;
@@ -921,12 +922,12 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
}
memberptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
*memberptr = members[i].xid;
flagsptr = (uint32 *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
flagsval = *flagsptr;
flagsval &= ~(((1 << MXACT_MEMBER_BITS_PER_XACT) - 1) << bshift);
@@ -1348,7 +1349,7 @@ retry:
entryno = MultiXactIdToOffsetEntry(multi);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
@@ -1381,7 +1382,7 @@ retry:
if (pageno != prev_pageno)
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, tmpMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
nextMXOffset = *offptr;
@@ -1424,7 +1425,7 @@ retry:
}
xactptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
if (!TransactionIdIsValid(*xactptr))
{
@@ -1435,7 +1436,7 @@ retry:
flagsoff = MXOffsetToFlagsOffset(offset);
bshift = MXOffsetToFlagsBitShift(offset);
- flagsptr = (uint32 *) (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ flagsptr = (uint32 *) (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
ptr[truelength].xid = *xactptr;
ptr[truelength].status = (*flagsptr >> bshift) & MXACT_MEMBER_XACT_BITMASK;
@@ -1834,8 +1835,8 @@ MultiXactShmemSize(void)
mul_size(sizeof(MultiXactId) * 2, MaxOldestSlot))
size = SHARED_MULTIXACT_STATE_SIZE;
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS, 0));
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS));
return size;
}
@@ -1851,14 +1852,14 @@ MultiXactShmemInit(void)
MultiXactMemberCtl->PagePrecedes = MultiXactMemberPagePrecedes;
SimpleLruInit(MultiXactOffsetCtl,
- "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS, 0,
+ "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS,
MultiXactOffsetSLRULock, "pg_multixact/offsets",
LWTRANCHE_MULTIXACTOFFSET_BUFFER,
SYNC_HANDLER_MULTIXACT_OFFSET,
false);
SlruPagePrecedesUnitTests(MultiXactOffsetCtl, MULTIXACT_OFFSETS_PER_PAGE);
SimpleLruInit(MultiXactMemberCtl,
- "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS, 0,
+ "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS,
MultiXactMemberSLRULock, "pg_multixact/members",
LWTRANCHE_MULTIXACTMEMBER_BUFFER,
SYNC_HANDLER_MULTIXACT_MEMBER,
@@ -1933,11 +1934,17 @@ static int
ZeroMultiXactOffsetPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactOffsetCtl, pageno);
+ page = MultiXactOffsetCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -1949,11 +1956,17 @@ static int
ZeroMultiXactMemberPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactMemberCtl, pageno);
+ page = MultiXactMemberCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2071,10 +2084,10 @@ TrimMultiXact(void)
MultiXactOffset *offptr;
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, nextMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
- MemSet(offptr, 0, BLCKSZ - (entryno * sizeof(MultiXactOffset)));
+ MemSet(offptr, 0, SizeOfPageContents - (entryno * sizeof(MultiXactOffset)));
MultiXactOffsetCtl->shared->page_dirty[slotno] = true;
}
@@ -2104,9 +2117,9 @@ TrimMultiXact(void)
memberoff = MXOffsetToMemberOffset(offset);
slotno = SimpleLruReadPage(MultiXactMemberCtl, pageno, true, offset);
xidptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
- MemSet(xidptr, 0, BLCKSZ - memberoff);
+ MemSet(xidptr, 0, SizeOfPageContents - memberoff);
/*
* Note: we don't need to zero out the flag bits in the remaining
@@ -2758,7 +2771,7 @@ find_multixact_start(MultiXactId multi, MultiXactOffset *result)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(MultiXactOffsetCtl, pageno, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
LWLockRelease(MultiXactOffsetSLRULock);
@@ -3192,12 +3205,12 @@ MultiXactOffsetPrecedes(MultiXactOffset offset1, MultiXactOffset offset2)
* Write an xlog record reflecting the zeroing of either a MEMBERs or
* OFFSETs page (info shows which)
*/
-static void
+static XLogRecPtr
WriteMZeroPageXlogRec(int64 pageno, uint8 info)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_MULTIXACT_ID, info);
+ return XLogInsert(RM_MULTIXACT_ID, info);
}
/*
diff --git a/src/backend/access/transam/slru.c b/src/backend/access/transam/slru.c
index 7a371d9034..d7f3deea7d 100644
--- a/src/backend/access/transam/slru.c
+++ b/src/backend/access/transam/slru.c
@@ -57,6 +57,7 @@
#include "access/xlogutils.h"
#include "miscadmin.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/shmem.h"
@@ -154,13 +155,13 @@ typedef enum
SLRU_WRITE_FAILED,
SLRU_FSYNC_FAILED,
SLRU_CLOSE_FAILED,
+ SLRU_DATA_CORRUPTED,
} SlruErrorCause;
static SlruErrorCause slru_errcause;
static int slru_errno;
-static void SimpleLruZeroLSNs(SlruCtl ctl, int slotno);
static void SimpleLruWaitIO(SlruCtl ctl, int slotno);
static void SlruInternalWritePage(SlruCtl ctl, int slotno, SlruWriteAll fdata);
static bool SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno);
@@ -179,7 +180,7 @@ static void SlruInternalDeleteSegment(SlruCtl ctl, int64 segno);
*/
Size
-SimpleLruShmemSize(int nslots, int nlsns)
+SimpleLruShmemSize(int nslots)
{
Size sz;
@@ -192,9 +193,6 @@ SimpleLruShmemSize(int nslots, int nlsns)
sz += MAXALIGN(nslots * sizeof(int)); /* page_lru_count[] */
sz += MAXALIGN(nslots * sizeof(LWLockPadded)); /* buffer_locks[] */
- if (nlsns > 0)
- sz += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr)); /* group_lsn[] */
-
return BUFFERALIGN(sz) + BLCKSZ * nslots;
}
@@ -204,14 +202,13 @@ SimpleLruShmemSize(int nslots, int nlsns)
* ctl: address of local (unshared) control structure.
* name: name of SLRU. (This is user-visible, pick with care!)
* nslots: number of page slots to use.
- * nlsns: number of LSN groups per page (set to zero if not relevant).
* ctllock: LWLock to use to control access to the shared control structure.
* subdir: PGDATA-relative subdirectory that will contain the files.
* tranche_id: LWLock tranche ID to use for the SLRU's per-buffer LWLocks.
* sync_handler: which set of functions to use to handle sync requests
*/
void
-SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler, bool long_segment_names)
{
@@ -219,7 +216,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
bool found;
shared = (SlruShared) ShmemInitStruct(name,
- SimpleLruShmemSize(nslots, nlsns),
+ SimpleLruShmemSize(nslots),
&found);
if (!IsUnderPostmaster)
@@ -236,7 +233,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->ControlLock = ctllock;
shared->num_slots = nslots;
- shared->lsn_groups_per_page = nlsns;
shared->cur_lru_count = 0;
@@ -261,12 +257,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->buffer_locks = (LWLockPadded *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(LWLockPadded));
- if (nlsns > 0)
- {
- shared->group_lsn = (XLogRecPtr *) (ptr + offset);
- offset += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr));
- }
-
ptr += BUFFERALIGN(offset);
for (slotno = 0; slotno < nslots; slotno++)
{
@@ -281,7 +271,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
}
/* Should fit to estimated shmem size */
- Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots, nlsns));
+ Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots));
}
else
Assert(found);
@@ -323,11 +313,8 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
shared->page_dirty[slotno] = true;
SlruRecentlyUsed(shared, slotno);
- /* Set the buffer to zeroes */
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
-
- /* Set the LSNs for this new page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
+ /* Initialize the page. */
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
/* Assume this page is now the latest active page */
shared->latest_page_number = pageno;
@@ -338,26 +325,6 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
return slotno;
}
-/*
- * Zero all the LSNs we store for this slru page.
- *
- * This should be called each time we create a new page, and each time we read
- * in a page from disk into an existing buffer. (Such an old page cannot
- * have any interesting LSNs, since we'd have flushed them before writing
- * the page in the first place.)
- *
- * This assumes that InvalidXLogRecPtr is bitwise-all-0.
- */
-static void
-SimpleLruZeroLSNs(SlruCtl ctl, int slotno)
-{
- SlruShared shared = ctl->shared;
-
- if (shared->lsn_groups_per_page > 0)
- MemSet(&shared->group_lsn[slotno * shared->lsn_groups_per_page], 0,
- shared->lsn_groups_per_page * sizeof(XLogRecPtr));
-}
-
/*
* Wait for any active I/O on a page slot to finish. (This does not
* guarantee that new I/O hasn't been started before we return, though.
@@ -478,9 +445,6 @@ SimpleLruReadPage(SlruCtl ctl, int64 pageno, bool write_ok,
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
- /* Set the LSNs for this newly read-in page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
-
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
@@ -740,7 +704,7 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
ereport(LOG,
(errmsg("file \"%s\" doesn't exist, reading as zeroes",
path)));
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
return true;
}
@@ -763,6 +727,13 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
return false;
}
+ if (!PageIsVerifiedExtended(shared->page_buffer[slotno], pageno, PIV_REPORT_STAT))
+ {
+ slru_errcause = SLRU_DATA_CORRUPTED;
+ slru_errno = 0;
+ return false;
+ }
+
return true;
}
@@ -789,6 +760,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
off_t offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd = -1;
+ Page page = shared->page_buffer[slotno];
+ XLogRecPtr lsn;
/* update the stats counter of written pages */
pgstat_count_slru_page_written(shared->slru_stats_idx);
@@ -798,41 +771,18 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
* write out data before associated WAL records. This is the same action
* performed during FlushBuffer() in the main buffer manager.
*/
- if (shared->group_lsn != NULL)
+ lsn = PageGetLSN(page);
+ if (!XLogRecPtrIsInvalid(lsn))
{
/*
- * We must determine the largest async-commit LSN for the page. This
- * is a bit tedious, but since this entire function is a slow path
- * anyway, it seems better to do this here than to maintain a per-page
- * LSN variable (which'd need an extra comparison in the
- * transaction-commit path).
+ * As noted above, elog(ERROR) is not acceptable here, so if
+ * XLogFlush were to fail, we must PANIC. This isn't much of a
+ * restriction because XLogFlush is just about all critical
+ * section anyway, but let's make sure.
*/
- XLogRecPtr max_lsn;
- int lsnindex,
- lsnoff;
-
- lsnindex = slotno * shared->lsn_groups_per_page;
- max_lsn = shared->group_lsn[lsnindex++];
- for (lsnoff = 1; lsnoff < shared->lsn_groups_per_page; lsnoff++)
- {
- XLogRecPtr this_lsn = shared->group_lsn[lsnindex++];
-
- if (max_lsn < this_lsn)
- max_lsn = this_lsn;
- }
-
- if (!XLogRecPtrIsInvalid(max_lsn))
- {
- /*
- * As noted above, elog(ERROR) is not acceptable here, so if
- * XLogFlush were to fail, we must PANIC. This isn't much of a
- * restriction because XLogFlush is just about all critical
- * section anyway, but let's make sure.
- */
- START_CRIT_SECTION();
- XLogFlush(max_lsn);
- END_CRIT_SECTION();
- }
+ START_CRIT_SECTION();
+ XLogFlush(lsn);
+ END_CRIT_SECTION();
}
/*
@@ -899,6 +849,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
}
}
+ PageSetChecksumInplace(shared->page_buffer[slotno], pageno);
+
errno = 0;
pgstat_report_wait_start(WAIT_EVENT_SLRU_WRITE);
if (pg_pwrite(fd, shared->page_buffer[slotno], BLCKSZ, offset) != BLCKSZ)
@@ -1019,6 +971,13 @@ SlruReportIOError(SlruCtl ctl, int64 pageno, TransactionId xid)
errdetail("Could not close file \"%s\": %m.",
path)));
break;
+ case SLRU_DATA_CORRUPTED:
+ ereport(ERROR,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("could not access status of transaction %u", xid),
+ errdetail("Invalid page from file \"%s\" at offset %d.",
+ path, offset)));
+ break;
default:
/* can't get here, we trust */
elog(ERROR, "unrecognized SimpleLru error cause: %d",
diff --git a/src/backend/access/transam/subtrans.c b/src/backend/access/transam/subtrans.c
index 1b3b3ad720..1f35ee23c6 100644
--- a/src/backend/access/transam/subtrans.c
+++ b/src/backend/access/transam/subtrans.c
@@ -32,6 +32,7 @@
#include "access/subtrans.h"
#include "access/transam.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "utils/snapmgr.h"
@@ -49,7 +50,7 @@
*/
/* We need four bytes per xact */
-#define SUBTRANS_XACTS_PER_PAGE (BLCKSZ / sizeof(TransactionId))
+#define SUBTRANS_XACTS_PER_PAGE (SizeOfPageContents / sizeof(TransactionId))
/*
* Although we return an int64 the actual value can't currently exceed
@@ -93,7 +94,7 @@ SubTransSetParent(TransactionId xid, TransactionId parent)
LWLockAcquire(SubtransSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(SubTransCtl, pageno, true, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
/*
@@ -133,7 +134,7 @@ SubTransGetParent(TransactionId xid)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
parent = *ptr;
@@ -193,14 +194,14 @@ SubTransGetTopmostTransaction(TransactionId xid)
Size
SUBTRANSShmemSize(void)
{
- return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS, 0);
+ return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS);
}
void
SUBTRANSShmemInit(void)
{
SubTransCtl->PagePrecedes = SubTransPagePrecedes;
- SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS, 0,
+ SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS,
SubtransSLRULock, "pg_subtrans",
LWTRANCHE_SUBTRANS_BUFFER, SYNC_HANDLER_NONE,
false);
@@ -305,7 +306,6 @@ CheckPointSUBTRANS(void)
TRACE_POSTGRESQL_SUBTRANS_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that SUBTRANS has room for a newly-allocated XID.
*
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c
index 264f25a8f9..85b0d63cdc 100644
--- a/src/backend/commands/async.c
+++ b/src/backend/commands/async.c
@@ -140,6 +140,7 @@
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/proc.h"
@@ -162,7 +163,7 @@
* than that, so changes in that data structure won't affect user-visible
* restrictions.
*/
-#define NOTIFY_PAYLOAD_MAX_LENGTH (BLCKSZ - NAMEDATALEN - 128)
+#define NOTIFY_PAYLOAD_MAX_LENGTH (SizeOfPageContents - NAMEDATALEN - 128)
/*
* Struct representing an entry in the global notify queue
@@ -311,7 +312,7 @@ static SlruCtlData NotifyCtlData;
#define NotifyCtl (&NotifyCtlData)
#define QUEUE_PAGESIZE BLCKSZ
-
+#define QUEUE_PAGE_CAPACITY (QUEUE_PAGESIZE - MAXALIGN(SizeOfPageHeaderData))
#define QUEUE_FULL_WARN_INTERVAL 5000 /* warn at most once every 5s */
/*
@@ -492,7 +493,7 @@ AsyncShmemSize(void)
size = mul_size(MaxBackends + 1, sizeof(QueueBackendStatus));
size = add_size(size, offsetof(AsyncQueueControl, backend));
- size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS));
return size;
}
@@ -541,7 +542,7 @@ AsyncShmemInit(void)
* names are used in order to avoid wraparound.
*/
NotifyCtl->PagePrecedes = asyncQueuePagePrecedes;
- SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS, 0,
+ SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS,
NotifySLRULock, "pg_notify", LWTRANCHE_NOTIFY_BUFFER,
SYNC_HANDLER_NONE, true);
@@ -1301,14 +1302,14 @@ asyncQueueAdvance(volatile QueuePosition *position, int entryLength)
* written or read.
*/
offset += entryLength;
- Assert(offset <= QUEUE_PAGESIZE);
+ Assert(offset <= QUEUE_PAGE_CAPACITY);
/*
* In a second step check if another entry can possibly be written to the
* page. If so, stay here, we have reached the next position. If not, then
* we need to move on to the next page.
*/
- if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGESIZE)
+ if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGE_CAPACITY)
{
pageno++;
offset = 0;
@@ -1408,7 +1409,7 @@ asyncQueueAddEntries(ListCell *nextNotify)
offset = QUEUE_POS_OFFSET(queue_head);
/* Check whether the entry really fits on the current page */
- if (offset + qe.length <= QUEUE_PAGESIZE)
+ if (offset + qe.length <= QUEUE_PAGE_CAPACITY)
{
/* OK, so advance nextNotify past this item */
nextNotify = lnext(pendingNotifies->events, nextNotify);
@@ -1420,14 +1421,14 @@ asyncQueueAddEntries(ListCell *nextNotify)
* only check dboid and since it won't match any reader's database
* OID, they will ignore this entry and move on.
*/
- qe.length = QUEUE_PAGESIZE - offset;
+ qe.length = QUEUE_PAGE_CAPACITY - offset;
qe.dboid = InvalidOid;
qe.data[0] = '\0'; /* empty channel */
qe.data[1] = '\0'; /* empty payload */
}
/* Now copy qe into the shared buffer page */
- memcpy(NotifyCtl->shared->page_buffer[slotno] + offset,
+ memcpy(PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + offset,
&qe,
qe.length);
@@ -1947,10 +1948,10 @@ asyncQueueReadAllNotifications(void)
else
{
/* fetch all the rest of the page */
- copysize = QUEUE_PAGESIZE - curoffset;
+ copysize = QUEUE_PAGE_CAPACITY - curoffset;
}
- memcpy(page_buffer.buf + curoffset,
- NotifyCtl->shared->page_buffer[slotno] + curoffset,
+ memcpy(PageGetContents(page_buffer.buf) + curoffset,
+ PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + curoffset,
copysize);
/* Release lock that we got from SimpleLruReadPage_ReadOnly() */
LWLockRelease(NotifySLRULock);
@@ -2021,7 +2022,7 @@ asyncQueueProcessPageEntries(volatile QueuePosition *current,
if (QUEUE_POS_EQUAL(thisentry, stop))
break;
- qe = (AsyncQueueEntry *) (page_buffer + QUEUE_POS_OFFSET(thisentry));
+ qe = (AsyncQueueEntry *) (PageGetContents(page_buffer) + QUEUE_POS_OFFSET(thisentry));
/*
* Advance *current over this message, possibly to the next page. As
diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index 1129b8e4f2..9fd945ad38 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -204,6 +204,7 @@
#include "pgstat.h"
#include "port/pg_lfind.h"
#include "storage/bufmgr.h"
+#include "storage/bufpage.h"
#include "storage/predicate.h"
#include "storage/predicate_internals.h"
#include "storage/proc.h"
@@ -322,8 +323,8 @@ static SlruCtlData SerialSlruCtlData;
#define SerialSlruCtl (&SerialSlruCtlData)
#define SERIAL_PAGESIZE BLCKSZ
-#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
-#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE / SERIAL_ENTRYSIZE)
+#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
+#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE - MAXALIGN(SizeOfPageHeaderData) / SERIAL_ENTRYSIZE)
/*
* Set maximum pages based on the number needed to track all transactions.
@@ -333,7 +334,7 @@ static SlruCtlData SerialSlruCtlData;
#define SerialNextPage(page) (((page) >= SERIAL_MAX_PAGE) ? 0 : (page) + 1)
#define SerialValue(slotno, xid) (*((SerCommitSeqNo *) \
- (SerialSlruCtl->shared->page_buffer[slotno] + \
+ (PageGetContents(SerialSlruCtl->shared->page_buffer[slotno]) + \
((((uint32) (xid)) % SERIAL_ENTRIESPERPAGE) * SERIAL_ENTRYSIZE))))
#define SerialPage(xid) (((uint32) (xid)) / SERIAL_ENTRIESPERPAGE)
@@ -785,10 +786,13 @@ SerialPagePrecedesLogicallyUnitTests(void)
* requires burning ~2B XIDs in single-user mode, a negligible
* possibility. Moreover, if it does happen, the consequence would be
* mild, namely a new transaction failing in SimpleLruReadPage().
+ *
+ * NOTE: After adding the page header, the defect affects two pages.
+ * We now assert correct treatment of its second to prior page.
*/
headPage = oldestPage;
targetPage = newestPage;
- Assert(SerialPagePrecedesLogically(headPage, targetPage - 1));
+ Assert(SerialPagePrecedesLogically(headPage, targetPage - 2));
#if 0
Assert(SerialPagePrecedesLogically(headPage, targetPage));
#endif
@@ -808,7 +812,7 @@ SerialInit(void)
*/
SerialSlruCtl->PagePrecedes = SerialPagePrecedesLogically;
SimpleLruInit(SerialSlruCtl, "Serial",
- NUM_SERIAL_BUFFERS, 0, SerialSLRULock, "pg_serial",
+ NUM_SERIAL_BUFFERS, SerialSLRULock, "pg_serial",
LWTRANCHE_SERIAL_BUFFER, SYNC_HANDLER_NONE,
false);
#ifdef USE_ASSERT_CHECKING
@@ -1348,7 +1352,7 @@ PredicateLockShmemSize(void)
/* Shared memory structures for SLRU tracking of old committed xids. */
size = add_size(size, sizeof(SerialControlData));
- size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS));
return size;
}
diff --git a/src/backend/storage/page/bufpage.c b/src/backend/storage/page/bufpage.c
index 9a302ddc30..723a127594 100644
--- a/src/backend/storage/page/bufpage.c
+++ b/src/backend/storage/page/bufpage.c
@@ -59,6 +59,31 @@ PageInit(Page page, Size pageSize, Size specialSize)
/* p->pd_prune_xid = InvalidTransactionId; done by above MemSet */
}
+/*
+ * PageInitSLRU
+ * Initializes the contents of an SLRU page.
+ * Note that we don't calculate an initial checksum here; that's not done
+ * until it's time to write.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
/*
* PageIsVerifiedExtended
diff --git a/src/bin/pg_checksums/pg_checksums.c b/src/bin/pg_checksums/pg_checksums.c
index 19c083be17..0f43533811 100644
--- a/src/bin/pg_checksums/pg_checksums.c
+++ b/src/bin/pg_checksums/pg_checksums.c
@@ -16,6 +16,7 @@
#include <dirent.h>
#include <limits.h>
+#include <stdbool.h>
#include <time.h>
#include <sys/stat.h>
#include <unistd.h>
@@ -589,12 +590,20 @@ main(int argc, char *argv[])
{
total_size = scan_directory(DataDir, "global", true);
total_size += scan_directory(DataDir, "base", true);
+ total_size += scan_directory(DataDir, "pg_commit_ts", true);
+ total_size += scan_directory(DataDir, "pg_multixact", true);
+ total_size += scan_directory(DataDir, "pg_serial", true);
total_size += scan_directory(DataDir, "pg_tblspc", true);
+ total_size += scan_directory(DataDir, "pg_xact", true);
}
(void) scan_directory(DataDir, "global", false);
(void) scan_directory(DataDir, "base", false);
+ (void) scan_directory(DataDir, "pg_commit_ts", false);
+ (void) scan_directory(DataDir, "pg_multixact", false);
+ (void) scan_directory(DataDir, "pg_serial", false);
(void) scan_directory(DataDir, "pg_tblspc", false);
+ (void) scan_directory(DataDir, "pg_xact", false);
if (showprogress)
progress_report(true);
diff --git a/src/bin/pg_resetwal/t/001_basic.pl b/src/bin/pg_resetwal/t/001_basic.pl
index 18d0882cb1..ae74828e44 100644
--- a/src/bin/pg_resetwal/t/001_basic.pl
+++ b/src/bin/pg_resetwal/t/001_basic.pl
@@ -206,7 +206,7 @@ push @cmd,
sprintf("%d,%d", hex($files[0]) == 0 ? 3 : hex($files[0]), hex($files[-1]));
@files = get_slru_files('pg_multixact/offsets');
-$mult = 32 * $blcksz / 4;
+$mult = 32 * ($blcksz - 24) / 4;
# -m argument is "new,old"
push @cmd, '-m',
sprintf("%d,%d",
@@ -214,11 +214,11 @@ push @cmd, '-m',
hex($files[0]) == 0 ? 1 : hex($files[0] * $mult));
@files = get_slru_files('pg_multixact/members');
-$mult = 32 * int($blcksz / 20) * 4;
+$mult = 32 * int(($blcksz - 24) / 20) * 4;
push @cmd, '-O', (hex($files[-1]) + 1) * $mult;
@files = get_slru_files('pg_xact');
-$mult = 32 * $blcksz * 4;
+$mult = 32 * ($blcksz - 24) * 4;
push @cmd,
'-u', (hex($files[0]) == 0 ? 3 : hex($files[0]) * $mult),
'-x', ((hex($files[-1]) + 1) * $mult);
diff --git a/src/bin/pg_upgrade/file.c b/src/bin/pg_upgrade/file.c
index d173602882..322c2b3ce4 100644
--- a/src/bin/pg_upgrade/file.c
+++ b/src/bin/pg_upgrade/file.c
@@ -9,6 +9,7 @@
#include "postgres_fe.h"
+#include <dirent.h>
#include <sys/stat.h>
#include <fcntl.h>
#ifdef HAVE_COPYFILE_H
@@ -375,3 +376,188 @@ check_hard_link(void)
unlink(new_link_file);
}
+
+
+/*
+ * Copy SLRU_PAGES_PER_SEGMENT from access/slru.h to avoid including it.
+ */
+#define SLRU_PAGES_PER_SEGMENT 32
+
+#define SEGMENT_SIZE (BLCKSZ * SLRU_PAGES_PER_SEGMENT)
+
+/*
+ * Copy PageInitSLRU from storage/bufpage.c to avoid linking to the backend.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
+
+/*
+ * Filter function for scandir(3) to select only segment files.
+ */
+static int
+segment_file_filter(const struct dirent *dirent)
+{
+ return strspn(dirent->d_name, "0123456789ABCDEF") == strlen(dirent->d_name);
+}
+
+/*
+ * Compare function for scandir(3) to sort segment files.
+ */
+static int
+segment_file_compare(const struct dirent **a, const struct dirent **b)
+{
+ long segno_a;
+ long segno_b;
+
+ segno_a = strtol((*a)->d_name, NULL, 16);
+ segno_b = strtol((*b)->d_name, NULL, 16);
+ return segno_a - segno_b;
+}
+
+static void
+upgrade_file(const char *src_dir, const char *src_file, const char *dst_dir)
+{
+ char src[MAXPGPATH];
+ char dst[MAXPGPATH];
+
+ int seg_name_len;
+ int src_segno;
+ int64 src_pageno;
+ int dst_segno;
+ int64 dst_pageno;
+ int dst_offset;
+
+ int src_fd;
+ int dst_fd;
+
+ char *src_buf;
+ ssize_t src_len;
+ ssize_t src_buf_offset;
+ PGAlignedBlock dst_block;
+ Page page = dst_block.data;
+ int len_to_copy;
+
+ seg_name_len = strlen(src_file);
+ src_segno = (int) strtol(src_file, NULL, 16);
+ src_pageno = src_segno * SLRU_PAGES_PER_SEGMENT;
+
+ dst_pageno = src_pageno * BLCKSZ / SizeOfPageContents;
+ dst_offset = src_pageno * BLCKSZ - dst_pageno * SizeOfPageContents;
+ dst_segno = dst_pageno / SLRU_PAGES_PER_SEGMENT;
+
+ snprintf(src, sizeof(src), "%s/%s", src_dir, src_file);
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+
+ src_buf = pg_malloc(SEGMENT_SIZE);
+ if ((src_fd = open(src, O_RDONLY | PG_BINARY, 0)) == -1)
+ pg_fatal("could not open file \"%s\": %s", src, strerror(errno));
+ if ((src_len = read(src_fd, src_buf, SEGMENT_SIZE)) == -1)
+ pg_fatal("could not read file \"%s\": %s", src, strerror(errno));
+
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+
+ /*
+ * Read the destination page at dst_pageno into the buffer. The page may contain
+ * data from the previous source segment. Initialize the page if the page is new.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+ if (read(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not read file \"%s\": %s", dst, strerror(errno));
+ if (PageIsNew(page))
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Rewind the file position, so the first write will overwrite the page.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+
+ src_buf_offset = 0;
+ while (src_buf_offset < src_len)
+ {
+ len_to_copy = Min(src_len - src_buf_offset, SizeOfPageContents - dst_offset);
+ memcpy(PageGetContents(page) + dst_offset, src_buf + src_buf_offset, len_to_copy);
+ src_buf_offset += len_to_copy;
+
+ if (new_cluster.controldata.data_checksum_version > 0)
+ ((PageHeader) page)->pd_checksum = pg_checksum_page(page, dst_pageno);
+ if (write(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not write file \"%s\": %s", dst, strerror(errno));
+
+ dst_pageno++;
+ dst_offset = 0;
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Switch segments if we reached the end of the current segment.
+ */
+ if (dst_pageno % SLRU_PAGES_PER_SEGMENT == 0)
+ {
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ dst_segno++;
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+ }
+ }
+
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ pg_free(src_buf);
+ close(src_fd);
+}
+
+void
+upgrade_xact_cache(const char *src_subdir, const char *dst_subdir)
+{
+ char src_dir[MAXPGPATH];
+ char dst_dir[MAXPGPATH];
+
+ struct dirent **src_segment_files;
+ int num_files;
+ int i;
+
+ snprintf(src_dir, sizeof(src_dir), "%s/%s", old_cluster.pgdata, src_subdir);
+ snprintf(dst_dir, sizeof(dst_dir), "%s/%s", new_cluster.pgdata, dst_subdir);
+
+ num_files = scandir(src_dir, &src_segment_files, segment_file_filter, segment_file_compare);
+ if (num_files == -1)
+ pg_fatal("could not scan directory \"%s\": %s", src_dir, strerror(errno));
+
+ for (i = 0; i < num_files; i++)
+ {
+ upgrade_file(src_dir, src_segment_files[i]->d_name, dst_dir);
+ free(src_segment_files[i]);
+ }
+ free(src_segment_files);
+}
diff --git a/src/bin/pg_upgrade/pg_upgrade.c b/src/bin/pg_upgrade/pg_upgrade.c
index 3960af4036..2b7d01058e 100644
--- a/src/bin/pg_upgrade/pg_upgrade.c
+++ b/src/bin/pg_upgrade/pg_upgrade.c
@@ -660,14 +660,23 @@ copy_subdir_files(const char *old_subdir, const char *new_subdir)
static void
copy_xact_xlog_xid(void)
{
+ bool slru_header_changed = false;
+
/*
* Copy old commit logs to new data dir. pg_clog has been renamed to
* pg_xact in post-10 clusters.
*/
- copy_subdir_files(GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact",
- GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact");
+ char *xact_old_dir = GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+ char *xact_new_dir = GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+
+ if (new_cluster.controldata.cat_ver >= SLRU_PAGE_HEADER_CAT_VER &&
+ old_cluster.controldata.cat_ver < SLRU_PAGE_HEADER_CAT_VER)
+ slru_header_changed = true;
+
+ if (slru_header_changed)
+ upgrade_xact_cache(xact_old_dir, xact_new_dir);
+ else
+ copy_subdir_files(xact_old_dir, xact_new_dir);
prep_status("Setting oldest XID for new cluster");
exec_prog(UTILITY_LOG_FILE, NULL, true, true,
@@ -702,7 +711,8 @@ copy_xact_xlog_xid(void)
* server doesn't attempt to read multis older than the cutoff value.
*/
if (old_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
- new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
+ !slru_header_changed)
{
copy_subdir_files("pg_multixact/offsets", "pg_multixact/offsets");
copy_subdir_files("pg_multixact/members", "pg_multixact/members");
@@ -722,7 +732,8 @@ copy_xact_xlog_xid(void)
new_cluster.pgdata);
check_ok();
}
- else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER ||
+ slru_header_changed)
{
/*
* Remove offsets/0000 file created by initdb that no longer matches
diff --git a/src/bin/pg_upgrade/pg_upgrade.h b/src/bin/pg_upgrade/pg_upgrade.h
index a710f325de..22186ea712 100644
--- a/src/bin/pg_upgrade/pg_upgrade.h
+++ b/src/bin/pg_upgrade/pg_upgrade.h
@@ -115,6 +115,11 @@ extern char *output_files[];
*/
#define MULTIXACT_FORMATCHANGE_CAT_VER 201301231
+/*
+ * A page header was added to each SLRU page in 17.0.
+ */
+#define SLRU_PAGE_HEADER_CAT_VER 202312091
+
/*
* large object chunk size added to pg_controldata,
* commit 5f93c37805e7485488480916b4585e098d3cc883
@@ -405,6 +410,7 @@ void rewriteVisibilityMap(const char *fromfile, const char *tofile,
const char *schemaName, const char *relName);
void check_file_clone(void);
void check_hard_link(void);
+void upgrade_xact_cache(const char *src_subdir, const char *dst_subdir);
/* fopen_priv() is no longer different from fopen() */
#define fopen_priv(path, mode) fopen(path, mode)
diff --git a/src/include/access/slru.h b/src/include/access/slru.h
index 091e2202c9..24733166b8 100644
--- a/src/include/access/slru.h
+++ b/src/include/access/slru.h
@@ -68,17 +68,6 @@ typedef struct SlruSharedData
int *page_lru_count;
LWLockPadded *buffer_locks;
- /*
- * Optional array of WAL flush LSNs associated with entries in the SLRU
- * pages. If not zero/NULL, we must flush WAL before writing pages (true
- * for pg_xact, false for multixact, pg_subtrans, pg_notify). group_lsn[]
- * has lsn_groups_per_page entries per buffer slot, each containing the
- * highest LSN known for a contiguous group of SLRU entries on that slot's
- * page.
- */
- XLogRecPtr *group_lsn;
- int lsn_groups_per_page;
-
/*----------
* We mark a page "most recently used" by setting
* page_lru_count[slotno] = ++cur_lru_count;
@@ -147,8 +136,8 @@ typedef struct SlruCtlData
typedef SlruCtlData *SlruCtl;
-extern Size SimpleLruShmemSize(int nslots, int nlsns);
-extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+extern Size SimpleLruShmemSize(int nslots);
+extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler,
bool long_segment_names);
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index a06a8f0b23..af5aaded56 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
*/
/* yyyymmddN */
-#define CATALOG_VERSION_NO 202312071
+#define CATALOG_VERSION_NO 202312191
#endif
diff --git a/src/include/storage/bufpage.h b/src/include/storage/bufpage.h
index 424ecba028..251d9523fa 100644
--- a/src/include/storage/bufpage.h
+++ b/src/include/storage/bufpage.h
@@ -201,6 +201,7 @@ typedef PageHeaderData *PageHeader;
* handling pages.
*/
#define PG_PAGE_LAYOUT_VERSION 4
+#define PG_SLRU_PAGE_LAYOUT_VERSION 1
#define PG_DATA_CHECKSUM_VERSION 1
/* ----------------------------------------------------------------
@@ -257,6 +258,11 @@ PageGetContents(Page page)
return (char *) page + MAXALIGN(SizeOfPageHeaderData);
}
+/*
+ * Space available for storing page contents.
+ */
+#define SizeOfPageContents (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
+
/* ----------------
* functions to access page size info
* ----------------
@@ -486,6 +492,7 @@ StaticAssertDecl(BLCKSZ == ((BLCKSZ / sizeof(size_t)) * sizeof(size_t)),
"BLCKSZ has to be a multiple of sizeof(size_t)");
extern void PageInit(Page page, Size pageSize, Size specialSize);
+extern void PageInitSLRU(Page page, Size pageSize, Size specialSize);
extern bool PageIsVerifiedExtended(Page page, BlockNumber blkno, int flags);
extern OffsetNumber PageAddItemExtended(Page page, Item item, Size size,
OffsetNumber offsetNumber, int flags);
diff --git a/src/test/modules/test_slru/test_slru.c b/src/test/modules/test_slru/test_slru.c
index d0fb9444e8..aa8a29d92e 100644
--- a/src/test/modules/test_slru/test_slru.c
+++ b/src/test/modules/test_slru/test_slru.c
@@ -17,6 +17,7 @@
#include "access/slru.h"
#include "access/transam.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/shmem.h"
@@ -76,8 +77,8 @@ test_slru_page_write(PG_FUNCTION_ARGS)
TestSlruCtl->shared->page_status[slotno] = SLRU_PAGE_VALID;
/* write given data to the page, up to the limit of the page */
- strncpy(TestSlruCtl->shared->page_buffer[slotno], data,
- BLCKSZ - 1);
+ strncpy(PageGetContents(TestSlruCtl->shared->page_buffer[slotno]), data,
+ SizeOfPageContents - 1);
SimpleLruWritePage(TestSlruCtl, slotno);
LWLockRelease(TestSLRULock);
@@ -104,7 +105,7 @@ test_slru_page_read(PG_FUNCTION_ARGS)
LWLockAcquire(TestSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(TestSlruCtl, pageno,
write_ok, InvalidTransactionId);
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -122,7 +123,7 @@ test_slru_page_readonly(PG_FUNCTION_ARGS)
pageno,
InvalidTransactionId);
Assert(LWLockHeldByMe(TestSLRULock));
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -202,7 +203,7 @@ test_slru_shmem_request(void)
prev_shmem_request_hook();
/* reserve shared memory for the test SLRU */
- RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS, 0));
+ RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS));
}
static bool
@@ -238,7 +239,7 @@ test_slru_shmem_startup(void)
TestSlruCtl->PagePrecedes = test_slru_page_precedes_logically;
SimpleLruInit(TestSlruCtl, "TestSLRU",
- NUM_TEST_BUFFERS, 0, TestSLRULock, slru_dir_name,
+ NUM_TEST_BUFFERS, TestSLRULock, slru_dir_name,
test_tranche_id, SYNC_HANDLER_NONE, long_segment_names);
}
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-01-02 11:35 ` Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Aleksander Alekseev @ 2024-01-02 11:35 UTC (permalink / raw)
To: PostgreSQL Hackers <[email protected]>; +Cc: Li, Yong <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
Hi,
> I have also added this thread to the current Commitfest and hope this patch
> will be part of the 17 release.
>
> The commitfest link:
> https://commitfest.postgresql.org/46/4709/
Thanks for the updated patch.
cfbot seems to have some complaints regarding compiler warnings and
also building the patch on Windows:
http://cfbot.cputube.org/
--
Best regards,
Aleksander Alekseev
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
@ 2024-01-04 15:57 ` Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2024-01-04 15:57 UTC (permalink / raw)
To: Aleksander Alekseev <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
> On Jan 2, 2024, at 19:35, Aleksander Alekseev <[email protected]> wrote:
>
> Thanks for the updated patch.
>
> cfbot seems to have some complaints regarding compiler warnings and
> also building the patch on Windows:
>
> http://cfbot.cputube.org/
Thanks for the information. Here is the updated patch.
Regards,
Yong
Attachments:
[application/octet-stream] slru_page_header_v3.patch (55.2K, ../../[email protected]/2-slru_page_header_v3.patch)
download | inline diff:
.gitignore | 3 +
src/backend/access/transam/clog.c | 52 +++++-----
src/backend/access/transam/commit_ts.c | 26 +++--
src/backend/access/transam/multixact.c | 63 +++++++-----
src/backend/access/transam/slru.c | 113 +++++++---------------
src/backend/access/transam/subtrans.c | 12 +--
src/backend/commands/async.c | 27 +++---
src/backend/storage/lmgr/predicate.c | 16 +--
src/backend/storage/page/bufpage.c | 25 +++++
src/bin/pg_checksums/pg_checksums.c | 9 ++
src/bin/pg_resetwal/t/001_basic.pl | 6 +-
src/bin/pg_upgrade/file.c | 172 +++++++++++++++++++++++++++++++++
src/bin/pg_upgrade/pg_upgrade.c | 23 +++--
src/bin/pg_upgrade/pg_upgrade.h | 6 ++
src/include/access/slru.h | 15 +--
src/include/catalog/catversion.h | 2 +-
src/include/storage/bufpage.h | 7 ++
src/test/modules/test_slru/test_slru.c | 13 +--
18 files changed, 398 insertions(+), 192 deletions(-)
diff --git a/.gitignore b/.gitignore
index 4e911395fe..5b8d23ce9f 100644
--- a/.gitignore
+++ b/.gitignore
@@ -31,6 +31,8 @@ win32ver.rc
*.exe
lib*dll.def
lib*.pc
+*.patch
+compile_commands.json
# Local excludes in root directory
/GNUmakefile
@@ -43,3 +45,4 @@ lib*.pc
/Release/
/tmp_install/
/portlock/
+/.cache
\ No newline at end of file
diff --git a/src/backend/access/transam/clog.c b/src/backend/access/transam/clog.c
index 7dca1df61b..b7f690754b 100644
--- a/src/backend/access/transam/clog.c
+++ b/src/backend/access/transam/clog.c
@@ -41,6 +41,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/proc.h"
#include "storage/sync.h"
@@ -59,7 +60,7 @@
/* We need two bits per xact, so four xacts fit in a byte */
#define CLOG_BITS_PER_XACT 2
#define CLOG_XACTS_PER_BYTE 4
-#define CLOG_XACTS_PER_PAGE (BLCKSZ * CLOG_XACTS_PER_BYTE)
+#define CLOG_XACTS_PER_PAGE (SizeOfPageContents * CLOG_XACTS_PER_BYTE)
#define CLOG_XACT_BITMASK ((1 << CLOG_BITS_PER_XACT) - 1)
@@ -77,13 +78,6 @@ TransactionIdToPage(TransactionId xid)
#define TransactionIdToByte(xid) (TransactionIdToPgIndex(xid) / CLOG_XACTS_PER_BYTE)
#define TransactionIdToBIndex(xid) ((xid) % (TransactionId) CLOG_XACTS_PER_BYTE)
-/* We store the latest async LSN for each group of transactions */
-#define CLOG_XACTS_PER_LSN_GROUP 32 /* keep this a power of 2 */
-#define CLOG_LSNS_PER_PAGE (CLOG_XACTS_PER_PAGE / CLOG_XACTS_PER_LSN_GROUP)
-
-#define GetLSNIndex(slotno, xid) ((slotno) * CLOG_LSNS_PER_PAGE + \
- ((xid) % (TransactionId) CLOG_XACTS_PER_PAGE) / CLOG_XACTS_PER_LSN_GROUP)
-
/*
* The number of subtransactions below which we consider to apply clog group
* update optimization. Testing reveals that the number higher than this can
@@ -101,7 +95,7 @@ static SlruCtlData XactCtlData;
static int ZeroCLOGPage(int64 pageno, bool writeXlog);
static bool CLOGPagePrecedes(int64 page1, int64 page2);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXact,
Oid oldestXactDb);
static void TransactionIdSetPageStatus(TransactionId xid, int nsubxids,
@@ -583,8 +577,9 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
char *byteptr;
char byteval;
char curval;
+ Page page = XactCtl->shared->page_buffer[slotno];
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(page) + byteno;
curval = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
/*
@@ -613,7 +608,7 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*byteptr = byteval;
/*
- * Update the group LSN if the transaction completion LSN is higher.
+ * Update the page LSN if the transaction completion LSN is higher.
*
* Note: lsn will be invalid when supplied during InRecovery processing,
* so we don't need to do anything special to avoid LSN updates during
@@ -622,10 +617,8 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*/
if (!XLogRecPtrIsInvalid(lsn))
{
- int lsnindex = GetLSNIndex(slotno, xid);
-
- if (XactCtl->shared->group_lsn[lsnindex] < lsn)
- XactCtl->shared->group_lsn[lsnindex] = lsn;
+ if (PageGetLSN(page) < lsn)
+ PageSetLSN(page, lsn);
}
}
@@ -651,19 +644,19 @@ TransactionIdGetStatus(TransactionId xid, XLogRecPtr *lsn)
int byteno = TransactionIdToByte(xid);
int bshift = TransactionIdToBIndex(xid) * CLOG_BITS_PER_XACT;
int slotno;
- int lsnindex;
+ Page page;
char *byteptr;
XidStatus status;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(XactCtl, pageno, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
status = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
- lsnindex = GetLSNIndex(slotno, xid);
- *lsn = XactCtl->shared->group_lsn[lsnindex];
+ *lsn = PageGetLSN(page);
LWLockRelease(XactSLRULock);
@@ -698,14 +691,14 @@ CLOGShmemBuffers(void)
Size
CLOGShmemSize(void)
{
- return SimpleLruShmemSize(CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE);
+ return SimpleLruShmemSize(CLOGShmemBuffers());
}
void
CLOGShmemInit(void)
{
XactCtl->PagePrecedes = CLOGPagePrecedes;
- SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE,
+ SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(),
XactSLRULock, "pg_xact", LWTRANCHE_XACT_BUFFER,
SYNC_HANDLER_CLOG, false);
SlruPagePrecedesUnitTests(XactCtl, CLOG_XACTS_PER_PAGE);
@@ -747,11 +740,17 @@ static int
ZeroCLOGPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(XactCtl, pageno);
+ page = XactCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -807,12 +806,12 @@ TrimCLOG(void)
char *byteptr;
slotno = SimpleLruReadPage(XactCtl, pageno, false, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(XactCtl->shared->page_buffer[slotno]) + byteno;
/* Zero so-far-unused positions in the current byte */
*byteptr &= (1 << bshift) - 1;
/* Zero the rest of the page */
- MemSet(byteptr + 1, 0, BLCKSZ - byteno - 1);
+ MemSet(byteptr + 1, 0, SizeOfPageContents - byteno - 1);
XactCtl->shared->page_dirty[slotno] = true;
}
@@ -836,7 +835,6 @@ CheckPointCLOG(void)
TRACE_POSTGRESQL_CLOG_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that CLOG has room for a newly-allocated XID.
*
@@ -958,12 +956,12 @@ CLOGPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
+ return XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/commit_ts.c b/src/backend/access/transam/commit_ts.c
index e6fd9b3349..367459ea04 100644
--- a/src/backend/access/transam/commit_ts.c
+++ b/src/backend/access/transam/commit_ts.c
@@ -31,6 +31,7 @@
#include "funcapi.h"
#include "miscadmin.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "storage/shmem.h"
#include "utils/builtins.h"
#include "utils/snapmgr.h"
@@ -63,7 +64,7 @@ typedef struct CommitTimestampEntry
sizeof(RepOriginId))
#define COMMIT_TS_XACTS_PER_PAGE \
- (BLCKSZ / SizeOfCommitTimestampEntry)
+ (SizeOfPageContents / SizeOfCommitTimestampEntry)
/*
@@ -120,7 +121,7 @@ static int ZeroCommitTsPage(int64 pageno, bool writeXlog);
static bool CommitTsPagePrecedes(int64 page1, int64 page2);
static void ActivateCommitTs(void);
static void DeactivateCommitTs(void);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXid);
/*
@@ -254,11 +255,12 @@ TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
+ Assert(xid == slotno * COMMIT_TS_XACTS_PER_PAGE + entryno);
entry.time = ts;
entry.nodeid = nodeid;
- memcpy(CommitTsCtl->shared->page_buffer[slotno] +
+ memcpy(PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
@@ -337,7 +339,7 @@ TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
- CommitTsCtl->shared->page_buffer[slotno] +
+ PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
@@ -515,7 +517,7 @@ CommitTsShmemBuffers(void)
Size
CommitTsShmemSize(void)
{
- return SimpleLruShmemSize(CommitTsShmemBuffers(), 0) +
+ return SimpleLruShmemSize(CommitTsShmemBuffers()) +
sizeof(CommitTimestampShared);
}
@@ -529,7 +531,7 @@ CommitTsShmemInit(void)
bool found;
CommitTsCtl->PagePrecedes = CommitTsPagePrecedes;
- SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(), 0,
+ SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(),
CommitTsSLRULock, "pg_commit_ts",
LWTRANCHE_COMMITTS_BUFFER,
SYNC_HANDLER_COMMIT_TS,
@@ -582,11 +584,17 @@ static int
ZeroCommitTsPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(CommitTsCtl, pageno);
+ page = CommitTsCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -946,12 +954,12 @@ CommitTsPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
+ return XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c
index db3423f12e..0970aa8855 100644
--- a/src/backend/access/transam/multixact.c
+++ b/src/backend/access/transam/multixact.c
@@ -83,6 +83,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "postmaster/autovacuum.h"
+#include "storage/bufpage.h"
#include "storage/lmgr.h"
#include "storage/pmsignal.h"
#include "storage/proc.h"
@@ -106,7 +107,7 @@
*/
/* We need four bytes per offset */
-#define MULTIXACT_OFFSETS_PER_PAGE (BLCKSZ / sizeof(MultiXactOffset))
+#define MULTIXACT_OFFSETS_PER_PAGE (SizeOfPageContents / sizeof(MultiXactOffset))
#define MultiXactIdToOffsetPage(xid) \
((xid) / (MultiXactOffset) MULTIXACT_OFFSETS_PER_PAGE)
@@ -119,8 +120,8 @@
* additional flag bits for each TransactionId. To do this without getting
* into alignment issues, we store four bytes of flags, and then the
* corresponding 4 Xids. Each such 5-word (20-byte) set we call a "group", and
- * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 409 groups
- * per page. This wastes 12 bytes per page, but that's OK -- simplicity (and
+ * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 408 groups
+ * per page. This wastes 8 bytes per page, but that's OK -- simplicity (and
* performance) trumps space efficiency here.
*
* Note that the "offset" macros work with byte offset, not array indexes, so
@@ -138,7 +139,7 @@
/* size in bytes of a complete group */
#define MULTIXACT_MEMBERGROUP_SIZE \
(sizeof(TransactionId) * MULTIXACT_MEMBERS_PER_MEMBERGROUP + MULTIXACT_FLAGBYTES_PER_GROUP)
-#define MULTIXACT_MEMBERGROUPS_PER_PAGE (BLCKSZ / MULTIXACT_MEMBERGROUP_SIZE)
+#define MULTIXACT_MEMBERGROUPS_PER_PAGE (SizeOfPageContents / MULTIXACT_MEMBERGROUP_SIZE)
#define MULTIXACT_MEMBERS_PER_PAGE \
(MULTIXACT_MEMBERGROUPS_PER_PAGE * MULTIXACT_MEMBERS_PER_MEMBERGROUP)
@@ -366,7 +367,7 @@ static bool MultiXactOffsetWouldWrap(MultiXactOffset boundary,
MultiXactOffset start, uint32 distance);
static bool SetOffsetVacuumLimit(bool is_startup);
static bool find_multixact_start(MultiXactId multi, MultiXactOffset *result);
-static void WriteMZeroPageXlogRec(int64 pageno, uint8 info);
+static XLogRecPtr WriteMZeroPageXlogRec(int64 pageno, uint8 info);
static void WriteMTruncateXlogRec(Oid oldestMultiDB,
MultiXactId startTruncOff,
MultiXactId endTruncOff,
@@ -884,7 +885,7 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
* take the trouble to generalize the slru.c error reporting code.
*/
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
*offptr = offset;
@@ -921,12 +922,12 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
}
memberptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
*memberptr = members[i].xid;
flagsptr = (uint32 *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
flagsval = *flagsptr;
flagsval &= ~(((1 << MXACT_MEMBER_BITS_PER_XACT) - 1) << bshift);
@@ -1348,7 +1349,7 @@ retry:
entryno = MultiXactIdToOffsetEntry(multi);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
@@ -1381,7 +1382,7 @@ retry:
if (pageno != prev_pageno)
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, tmpMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
nextMXOffset = *offptr;
@@ -1424,7 +1425,7 @@ retry:
}
xactptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
if (!TransactionIdIsValid(*xactptr))
{
@@ -1435,7 +1436,7 @@ retry:
flagsoff = MXOffsetToFlagsOffset(offset);
bshift = MXOffsetToFlagsBitShift(offset);
- flagsptr = (uint32 *) (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ flagsptr = (uint32 *) (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
ptr[truelength].xid = *xactptr;
ptr[truelength].status = (*flagsptr >> bshift) & MXACT_MEMBER_XACT_BITMASK;
@@ -1834,8 +1835,8 @@ MultiXactShmemSize(void)
mul_size(sizeof(MultiXactId) * 2, MaxOldestSlot))
size = SHARED_MULTIXACT_STATE_SIZE;
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS, 0));
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS));
return size;
}
@@ -1851,14 +1852,14 @@ MultiXactShmemInit(void)
MultiXactMemberCtl->PagePrecedes = MultiXactMemberPagePrecedes;
SimpleLruInit(MultiXactOffsetCtl,
- "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS, 0,
+ "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS,
MultiXactOffsetSLRULock, "pg_multixact/offsets",
LWTRANCHE_MULTIXACTOFFSET_BUFFER,
SYNC_HANDLER_MULTIXACT_OFFSET,
false);
SlruPagePrecedesUnitTests(MultiXactOffsetCtl, MULTIXACT_OFFSETS_PER_PAGE);
SimpleLruInit(MultiXactMemberCtl,
- "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS, 0,
+ "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS,
MultiXactMemberSLRULock, "pg_multixact/members",
LWTRANCHE_MULTIXACTMEMBER_BUFFER,
SYNC_HANDLER_MULTIXACT_MEMBER,
@@ -1933,11 +1934,17 @@ static int
ZeroMultiXactOffsetPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactOffsetCtl, pageno);
+ page = MultiXactOffsetCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -1949,11 +1956,17 @@ static int
ZeroMultiXactMemberPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactMemberCtl, pageno);
+ page = MultiXactMemberCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2071,10 +2084,10 @@ TrimMultiXact(void)
MultiXactOffset *offptr;
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, nextMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
- MemSet(offptr, 0, BLCKSZ - (entryno * sizeof(MultiXactOffset)));
+ MemSet(offptr, 0, SizeOfPageContents - (entryno * sizeof(MultiXactOffset)));
MultiXactOffsetCtl->shared->page_dirty[slotno] = true;
}
@@ -2104,9 +2117,9 @@ TrimMultiXact(void)
memberoff = MXOffsetToMemberOffset(offset);
slotno = SimpleLruReadPage(MultiXactMemberCtl, pageno, true, offset);
xidptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
- MemSet(xidptr, 0, BLCKSZ - memberoff);
+ MemSet(xidptr, 0, SizeOfPageContents - memberoff);
/*
* Note: we don't need to zero out the flag bits in the remaining
@@ -2758,7 +2771,7 @@ find_multixact_start(MultiXactId multi, MultiXactOffset *result)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(MultiXactOffsetCtl, pageno, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
LWLockRelease(MultiXactOffsetSLRULock);
@@ -3192,12 +3205,12 @@ MultiXactOffsetPrecedes(MultiXactOffset offset1, MultiXactOffset offset2)
* Write an xlog record reflecting the zeroing of either a MEMBERs or
* OFFSETs page (info shows which)
*/
-static void
+static XLogRecPtr
WriteMZeroPageXlogRec(int64 pageno, uint8 info)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_MULTIXACT_ID, info);
+ return XLogInsert(RM_MULTIXACT_ID, info);
}
/*
diff --git a/src/backend/access/transam/slru.c b/src/backend/access/transam/slru.c
index 7a371d9034..d7f3deea7d 100644
--- a/src/backend/access/transam/slru.c
+++ b/src/backend/access/transam/slru.c
@@ -57,6 +57,7 @@
#include "access/xlogutils.h"
#include "miscadmin.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/shmem.h"
@@ -154,13 +155,13 @@ typedef enum
SLRU_WRITE_FAILED,
SLRU_FSYNC_FAILED,
SLRU_CLOSE_FAILED,
+ SLRU_DATA_CORRUPTED,
} SlruErrorCause;
static SlruErrorCause slru_errcause;
static int slru_errno;
-static void SimpleLruZeroLSNs(SlruCtl ctl, int slotno);
static void SimpleLruWaitIO(SlruCtl ctl, int slotno);
static void SlruInternalWritePage(SlruCtl ctl, int slotno, SlruWriteAll fdata);
static bool SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno);
@@ -179,7 +180,7 @@ static void SlruInternalDeleteSegment(SlruCtl ctl, int64 segno);
*/
Size
-SimpleLruShmemSize(int nslots, int nlsns)
+SimpleLruShmemSize(int nslots)
{
Size sz;
@@ -192,9 +193,6 @@ SimpleLruShmemSize(int nslots, int nlsns)
sz += MAXALIGN(nslots * sizeof(int)); /* page_lru_count[] */
sz += MAXALIGN(nslots * sizeof(LWLockPadded)); /* buffer_locks[] */
- if (nlsns > 0)
- sz += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr)); /* group_lsn[] */
-
return BUFFERALIGN(sz) + BLCKSZ * nslots;
}
@@ -204,14 +202,13 @@ SimpleLruShmemSize(int nslots, int nlsns)
* ctl: address of local (unshared) control structure.
* name: name of SLRU. (This is user-visible, pick with care!)
* nslots: number of page slots to use.
- * nlsns: number of LSN groups per page (set to zero if not relevant).
* ctllock: LWLock to use to control access to the shared control structure.
* subdir: PGDATA-relative subdirectory that will contain the files.
* tranche_id: LWLock tranche ID to use for the SLRU's per-buffer LWLocks.
* sync_handler: which set of functions to use to handle sync requests
*/
void
-SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler, bool long_segment_names)
{
@@ -219,7 +216,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
bool found;
shared = (SlruShared) ShmemInitStruct(name,
- SimpleLruShmemSize(nslots, nlsns),
+ SimpleLruShmemSize(nslots),
&found);
if (!IsUnderPostmaster)
@@ -236,7 +233,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->ControlLock = ctllock;
shared->num_slots = nslots;
- shared->lsn_groups_per_page = nlsns;
shared->cur_lru_count = 0;
@@ -261,12 +257,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->buffer_locks = (LWLockPadded *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(LWLockPadded));
- if (nlsns > 0)
- {
- shared->group_lsn = (XLogRecPtr *) (ptr + offset);
- offset += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr));
- }
-
ptr += BUFFERALIGN(offset);
for (slotno = 0; slotno < nslots; slotno++)
{
@@ -281,7 +271,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
}
/* Should fit to estimated shmem size */
- Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots, nlsns));
+ Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots));
}
else
Assert(found);
@@ -323,11 +313,8 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
shared->page_dirty[slotno] = true;
SlruRecentlyUsed(shared, slotno);
- /* Set the buffer to zeroes */
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
-
- /* Set the LSNs for this new page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
+ /* Initialize the page. */
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
/* Assume this page is now the latest active page */
shared->latest_page_number = pageno;
@@ -338,26 +325,6 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
return slotno;
}
-/*
- * Zero all the LSNs we store for this slru page.
- *
- * This should be called each time we create a new page, and each time we read
- * in a page from disk into an existing buffer. (Such an old page cannot
- * have any interesting LSNs, since we'd have flushed them before writing
- * the page in the first place.)
- *
- * This assumes that InvalidXLogRecPtr is bitwise-all-0.
- */
-static void
-SimpleLruZeroLSNs(SlruCtl ctl, int slotno)
-{
- SlruShared shared = ctl->shared;
-
- if (shared->lsn_groups_per_page > 0)
- MemSet(&shared->group_lsn[slotno * shared->lsn_groups_per_page], 0,
- shared->lsn_groups_per_page * sizeof(XLogRecPtr));
-}
-
/*
* Wait for any active I/O on a page slot to finish. (This does not
* guarantee that new I/O hasn't been started before we return, though.
@@ -478,9 +445,6 @@ SimpleLruReadPage(SlruCtl ctl, int64 pageno, bool write_ok,
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
- /* Set the LSNs for this newly read-in page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
-
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
@@ -740,7 +704,7 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
ereport(LOG,
(errmsg("file \"%s\" doesn't exist, reading as zeroes",
path)));
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
return true;
}
@@ -763,6 +727,13 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
return false;
}
+ if (!PageIsVerifiedExtended(shared->page_buffer[slotno], pageno, PIV_REPORT_STAT))
+ {
+ slru_errcause = SLRU_DATA_CORRUPTED;
+ slru_errno = 0;
+ return false;
+ }
+
return true;
}
@@ -789,6 +760,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
off_t offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd = -1;
+ Page page = shared->page_buffer[slotno];
+ XLogRecPtr lsn;
/* update the stats counter of written pages */
pgstat_count_slru_page_written(shared->slru_stats_idx);
@@ -798,41 +771,18 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
* write out data before associated WAL records. This is the same action
* performed during FlushBuffer() in the main buffer manager.
*/
- if (shared->group_lsn != NULL)
+ lsn = PageGetLSN(page);
+ if (!XLogRecPtrIsInvalid(lsn))
{
/*
- * We must determine the largest async-commit LSN for the page. This
- * is a bit tedious, but since this entire function is a slow path
- * anyway, it seems better to do this here than to maintain a per-page
- * LSN variable (which'd need an extra comparison in the
- * transaction-commit path).
+ * As noted above, elog(ERROR) is not acceptable here, so if
+ * XLogFlush were to fail, we must PANIC. This isn't much of a
+ * restriction because XLogFlush is just about all critical
+ * section anyway, but let's make sure.
*/
- XLogRecPtr max_lsn;
- int lsnindex,
- lsnoff;
-
- lsnindex = slotno * shared->lsn_groups_per_page;
- max_lsn = shared->group_lsn[lsnindex++];
- for (lsnoff = 1; lsnoff < shared->lsn_groups_per_page; lsnoff++)
- {
- XLogRecPtr this_lsn = shared->group_lsn[lsnindex++];
-
- if (max_lsn < this_lsn)
- max_lsn = this_lsn;
- }
-
- if (!XLogRecPtrIsInvalid(max_lsn))
- {
- /*
- * As noted above, elog(ERROR) is not acceptable here, so if
- * XLogFlush were to fail, we must PANIC. This isn't much of a
- * restriction because XLogFlush is just about all critical
- * section anyway, but let's make sure.
- */
- START_CRIT_SECTION();
- XLogFlush(max_lsn);
- END_CRIT_SECTION();
- }
+ START_CRIT_SECTION();
+ XLogFlush(lsn);
+ END_CRIT_SECTION();
}
/*
@@ -899,6 +849,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
}
}
+ PageSetChecksumInplace(shared->page_buffer[slotno], pageno);
+
errno = 0;
pgstat_report_wait_start(WAIT_EVENT_SLRU_WRITE);
if (pg_pwrite(fd, shared->page_buffer[slotno], BLCKSZ, offset) != BLCKSZ)
@@ -1019,6 +971,13 @@ SlruReportIOError(SlruCtl ctl, int64 pageno, TransactionId xid)
errdetail("Could not close file \"%s\": %m.",
path)));
break;
+ case SLRU_DATA_CORRUPTED:
+ ereport(ERROR,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("could not access status of transaction %u", xid),
+ errdetail("Invalid page from file \"%s\" at offset %d.",
+ path, offset)));
+ break;
default:
/* can't get here, we trust */
elog(ERROR, "unrecognized SimpleLru error cause: %d",
diff --git a/src/backend/access/transam/subtrans.c b/src/backend/access/transam/subtrans.c
index 1b3b3ad720..1f35ee23c6 100644
--- a/src/backend/access/transam/subtrans.c
+++ b/src/backend/access/transam/subtrans.c
@@ -32,6 +32,7 @@
#include "access/subtrans.h"
#include "access/transam.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "utils/snapmgr.h"
@@ -49,7 +50,7 @@
*/
/* We need four bytes per xact */
-#define SUBTRANS_XACTS_PER_PAGE (BLCKSZ / sizeof(TransactionId))
+#define SUBTRANS_XACTS_PER_PAGE (SizeOfPageContents / sizeof(TransactionId))
/*
* Although we return an int64 the actual value can't currently exceed
@@ -93,7 +94,7 @@ SubTransSetParent(TransactionId xid, TransactionId parent)
LWLockAcquire(SubtransSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(SubTransCtl, pageno, true, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
/*
@@ -133,7 +134,7 @@ SubTransGetParent(TransactionId xid)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
parent = *ptr;
@@ -193,14 +194,14 @@ SubTransGetTopmostTransaction(TransactionId xid)
Size
SUBTRANSShmemSize(void)
{
- return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS, 0);
+ return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS);
}
void
SUBTRANSShmemInit(void)
{
SubTransCtl->PagePrecedes = SubTransPagePrecedes;
- SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS, 0,
+ SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS,
SubtransSLRULock, "pg_subtrans",
LWTRANCHE_SUBTRANS_BUFFER, SYNC_HANDLER_NONE,
false);
@@ -305,7 +306,6 @@ CheckPointSUBTRANS(void)
TRACE_POSTGRESQL_SUBTRANS_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that SUBTRANS has room for a newly-allocated XID.
*
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c
index 264f25a8f9..85b0d63cdc 100644
--- a/src/backend/commands/async.c
+++ b/src/backend/commands/async.c
@@ -140,6 +140,7 @@
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/proc.h"
@@ -162,7 +163,7 @@
* than that, so changes in that data structure won't affect user-visible
* restrictions.
*/
-#define NOTIFY_PAYLOAD_MAX_LENGTH (BLCKSZ - NAMEDATALEN - 128)
+#define NOTIFY_PAYLOAD_MAX_LENGTH (SizeOfPageContents - NAMEDATALEN - 128)
/*
* Struct representing an entry in the global notify queue
@@ -311,7 +312,7 @@ static SlruCtlData NotifyCtlData;
#define NotifyCtl (&NotifyCtlData)
#define QUEUE_PAGESIZE BLCKSZ
-
+#define QUEUE_PAGE_CAPACITY (QUEUE_PAGESIZE - MAXALIGN(SizeOfPageHeaderData))
#define QUEUE_FULL_WARN_INTERVAL 5000 /* warn at most once every 5s */
/*
@@ -492,7 +493,7 @@ AsyncShmemSize(void)
size = mul_size(MaxBackends + 1, sizeof(QueueBackendStatus));
size = add_size(size, offsetof(AsyncQueueControl, backend));
- size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS));
return size;
}
@@ -541,7 +542,7 @@ AsyncShmemInit(void)
* names are used in order to avoid wraparound.
*/
NotifyCtl->PagePrecedes = asyncQueuePagePrecedes;
- SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS, 0,
+ SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS,
NotifySLRULock, "pg_notify", LWTRANCHE_NOTIFY_BUFFER,
SYNC_HANDLER_NONE, true);
@@ -1301,14 +1302,14 @@ asyncQueueAdvance(volatile QueuePosition *position, int entryLength)
* written or read.
*/
offset += entryLength;
- Assert(offset <= QUEUE_PAGESIZE);
+ Assert(offset <= QUEUE_PAGE_CAPACITY);
/*
* In a second step check if another entry can possibly be written to the
* page. If so, stay here, we have reached the next position. If not, then
* we need to move on to the next page.
*/
- if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGESIZE)
+ if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGE_CAPACITY)
{
pageno++;
offset = 0;
@@ -1408,7 +1409,7 @@ asyncQueueAddEntries(ListCell *nextNotify)
offset = QUEUE_POS_OFFSET(queue_head);
/* Check whether the entry really fits on the current page */
- if (offset + qe.length <= QUEUE_PAGESIZE)
+ if (offset + qe.length <= QUEUE_PAGE_CAPACITY)
{
/* OK, so advance nextNotify past this item */
nextNotify = lnext(pendingNotifies->events, nextNotify);
@@ -1420,14 +1421,14 @@ asyncQueueAddEntries(ListCell *nextNotify)
* only check dboid and since it won't match any reader's database
* OID, they will ignore this entry and move on.
*/
- qe.length = QUEUE_PAGESIZE - offset;
+ qe.length = QUEUE_PAGE_CAPACITY - offset;
qe.dboid = InvalidOid;
qe.data[0] = '\0'; /* empty channel */
qe.data[1] = '\0'; /* empty payload */
}
/* Now copy qe into the shared buffer page */
- memcpy(NotifyCtl->shared->page_buffer[slotno] + offset,
+ memcpy(PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + offset,
&qe,
qe.length);
@@ -1947,10 +1948,10 @@ asyncQueueReadAllNotifications(void)
else
{
/* fetch all the rest of the page */
- copysize = QUEUE_PAGESIZE - curoffset;
+ copysize = QUEUE_PAGE_CAPACITY - curoffset;
}
- memcpy(page_buffer.buf + curoffset,
- NotifyCtl->shared->page_buffer[slotno] + curoffset,
+ memcpy(PageGetContents(page_buffer.buf) + curoffset,
+ PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + curoffset,
copysize);
/* Release lock that we got from SimpleLruReadPage_ReadOnly() */
LWLockRelease(NotifySLRULock);
@@ -2021,7 +2022,7 @@ asyncQueueProcessPageEntries(volatile QueuePosition *current,
if (QUEUE_POS_EQUAL(thisentry, stop))
break;
- qe = (AsyncQueueEntry *) (page_buffer + QUEUE_POS_OFFSET(thisentry));
+ qe = (AsyncQueueEntry *) (PageGetContents(page_buffer) + QUEUE_POS_OFFSET(thisentry));
/*
* Advance *current over this message, possibly to the next page. As
diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index 1129b8e4f2..9fd945ad38 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -204,6 +204,7 @@
#include "pgstat.h"
#include "port/pg_lfind.h"
#include "storage/bufmgr.h"
+#include "storage/bufpage.h"
#include "storage/predicate.h"
#include "storage/predicate_internals.h"
#include "storage/proc.h"
@@ -322,8 +323,8 @@ static SlruCtlData SerialSlruCtlData;
#define SerialSlruCtl (&SerialSlruCtlData)
#define SERIAL_PAGESIZE BLCKSZ
-#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
-#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE / SERIAL_ENTRYSIZE)
+#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
+#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE - MAXALIGN(SizeOfPageHeaderData) / SERIAL_ENTRYSIZE)
/*
* Set maximum pages based on the number needed to track all transactions.
@@ -333,7 +334,7 @@ static SlruCtlData SerialSlruCtlData;
#define SerialNextPage(page) (((page) >= SERIAL_MAX_PAGE) ? 0 : (page) + 1)
#define SerialValue(slotno, xid) (*((SerCommitSeqNo *) \
- (SerialSlruCtl->shared->page_buffer[slotno] + \
+ (PageGetContents(SerialSlruCtl->shared->page_buffer[slotno]) + \
((((uint32) (xid)) % SERIAL_ENTRIESPERPAGE) * SERIAL_ENTRYSIZE))))
#define SerialPage(xid) (((uint32) (xid)) / SERIAL_ENTRIESPERPAGE)
@@ -785,10 +786,13 @@ SerialPagePrecedesLogicallyUnitTests(void)
* requires burning ~2B XIDs in single-user mode, a negligible
* possibility. Moreover, if it does happen, the consequence would be
* mild, namely a new transaction failing in SimpleLruReadPage().
+ *
+ * NOTE: After adding the page header, the defect affects two pages.
+ * We now assert correct treatment of its second to prior page.
*/
headPage = oldestPage;
targetPage = newestPage;
- Assert(SerialPagePrecedesLogically(headPage, targetPage - 1));
+ Assert(SerialPagePrecedesLogically(headPage, targetPage - 2));
#if 0
Assert(SerialPagePrecedesLogically(headPage, targetPage));
#endif
@@ -808,7 +812,7 @@ SerialInit(void)
*/
SerialSlruCtl->PagePrecedes = SerialPagePrecedesLogically;
SimpleLruInit(SerialSlruCtl, "Serial",
- NUM_SERIAL_BUFFERS, 0, SerialSLRULock, "pg_serial",
+ NUM_SERIAL_BUFFERS, SerialSLRULock, "pg_serial",
LWTRANCHE_SERIAL_BUFFER, SYNC_HANDLER_NONE,
false);
#ifdef USE_ASSERT_CHECKING
@@ -1348,7 +1352,7 @@ PredicateLockShmemSize(void)
/* Shared memory structures for SLRU tracking of old committed xids. */
size = add_size(size, sizeof(SerialControlData));
- size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS));
return size;
}
diff --git a/src/backend/storage/page/bufpage.c b/src/backend/storage/page/bufpage.c
index 9a302ddc30..723a127594 100644
--- a/src/backend/storage/page/bufpage.c
+++ b/src/backend/storage/page/bufpage.c
@@ -59,6 +59,31 @@ PageInit(Page page, Size pageSize, Size specialSize)
/* p->pd_prune_xid = InvalidTransactionId; done by above MemSet */
}
+/*
+ * PageInitSLRU
+ * Initializes the contents of an SLRU page.
+ * Note that we don't calculate an initial checksum here; that's not done
+ * until it's time to write.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
/*
* PageIsVerifiedExtended
diff --git a/src/bin/pg_checksums/pg_checksums.c b/src/bin/pg_checksums/pg_checksums.c
index 19c083be17..0f43533811 100644
--- a/src/bin/pg_checksums/pg_checksums.c
+++ b/src/bin/pg_checksums/pg_checksums.c
@@ -16,6 +16,7 @@
#include <dirent.h>
#include <limits.h>
+#include <stdbool.h>
#include <time.h>
#include <sys/stat.h>
#include <unistd.h>
@@ -589,12 +590,20 @@ main(int argc, char *argv[])
{
total_size = scan_directory(DataDir, "global", true);
total_size += scan_directory(DataDir, "base", true);
+ total_size += scan_directory(DataDir, "pg_commit_ts", true);
+ total_size += scan_directory(DataDir, "pg_multixact", true);
+ total_size += scan_directory(DataDir, "pg_serial", true);
total_size += scan_directory(DataDir, "pg_tblspc", true);
+ total_size += scan_directory(DataDir, "pg_xact", true);
}
(void) scan_directory(DataDir, "global", false);
(void) scan_directory(DataDir, "base", false);
+ (void) scan_directory(DataDir, "pg_commit_ts", false);
+ (void) scan_directory(DataDir, "pg_multixact", false);
+ (void) scan_directory(DataDir, "pg_serial", false);
(void) scan_directory(DataDir, "pg_tblspc", false);
+ (void) scan_directory(DataDir, "pg_xact", false);
if (showprogress)
progress_report(true);
diff --git a/src/bin/pg_resetwal/t/001_basic.pl b/src/bin/pg_resetwal/t/001_basic.pl
index dcb5fa846e..914afbda29 100644
--- a/src/bin/pg_resetwal/t/001_basic.pl
+++ b/src/bin/pg_resetwal/t/001_basic.pl
@@ -206,7 +206,7 @@ push @cmd,
sprintf("%d,%d", hex($files[0]) == 0 ? 3 : hex($files[0]), hex($files[-1]));
@files = get_slru_files('pg_multixact/offsets');
-$mult = 32 * $blcksz / 4;
+$mult = 32 * ($blcksz - 24) / 4;
# -m argument is "new,old"
push @cmd, '-m',
sprintf("%d,%d",
@@ -214,11 +214,11 @@ push @cmd, '-m',
hex($files[0]) == 0 ? 1 : hex($files[0] * $mult));
@files = get_slru_files('pg_multixact/members');
-$mult = 32 * int($blcksz / 20) * 4;
+$mult = 32 * int(($blcksz - 24) / 20) * 4;
push @cmd, '-O', (hex($files[-1]) + 1) * $mult;
@files = get_slru_files('pg_xact');
-$mult = 32 * $blcksz * 4;
+$mult = 32 * ($blcksz - 24) * 4;
push @cmd,
'-u', (hex($files[0]) == 0 ? 3 : hex($files[0]) * $mult),
'-x', ((hex($files[-1]) + 1) * $mult);
diff --git a/src/bin/pg_upgrade/file.c b/src/bin/pg_upgrade/file.c
index d173602882..c6be340736 100644
--- a/src/bin/pg_upgrade/file.c
+++ b/src/bin/pg_upgrade/file.c
@@ -9,6 +9,7 @@
#include "postgres_fe.h"
+#include <dirent.h>
#include <sys/stat.h>
#include <fcntl.h>
#ifdef HAVE_COPYFILE_H
@@ -375,3 +376,174 @@ check_hard_link(void)
unlink(new_link_file);
}
+
+
+/*
+ * Copy SLRU_PAGES_PER_SEGMENT from access/slru.h to avoid including it.
+ */
+#define SLRU_PAGES_PER_SEGMENT 32
+
+#define SEGMENT_SIZE (BLCKSZ * SLRU_PAGES_PER_SEGMENT)
+
+/*
+ * Copy PageInitSLRU from storage/bufpage.c to avoid linking to the backend.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
+
+/*
+ * Filter function for scandir(3) to select only segment files.
+ */
+static int
+segment_file_filter(const struct dirent *dirent)
+{
+ return strspn(dirent->d_name, "0123456789ABCDEF") == strlen(dirent->d_name);
+}
+
+static void
+upgrade_file(const char *src_dir, const char *src_file, const char *dst_dir)
+{
+ char src[MAXPGPATH];
+ char dst[MAXPGPATH];
+
+ int seg_name_len;
+ int src_segno;
+ int64 src_pageno;
+ int dst_segno;
+ int64 dst_pageno;
+ int dst_offset;
+
+ int src_fd;
+ int dst_fd;
+
+ char *src_buf;
+ ssize_t src_len;
+ ssize_t src_buf_offset;
+ PGAlignedBlock dst_block;
+ Page page = dst_block.data;
+ int len_to_copy;
+
+ seg_name_len = strlen(src_file);
+ src_segno = (int) strtol(src_file, NULL, 16);
+ src_pageno = src_segno * SLRU_PAGES_PER_SEGMENT;
+
+ dst_pageno = src_pageno * BLCKSZ / SizeOfPageContents;
+ dst_offset = src_pageno * BLCKSZ - dst_pageno * SizeOfPageContents;
+ dst_segno = dst_pageno / SLRU_PAGES_PER_SEGMENT;
+
+ snprintf(src, sizeof(src), "%s/%s", src_dir, src_file);
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+
+ src_buf = pg_malloc(SEGMENT_SIZE);
+ if ((src_fd = open(src, O_RDONLY | PG_BINARY, 0)) == -1)
+ pg_fatal("could not open file \"%s\": %s", src, strerror(errno));
+ if ((src_len = read(src_fd, src_buf, SEGMENT_SIZE)) == -1)
+ pg_fatal("could not read file \"%s\": %s", src, strerror(errno));
+
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+
+ /*
+ * Read the destination page at dst_pageno into the buffer. The page may contain
+ * data from the previous source segment. Initialize the page if the page is new.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+ if (read(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not read file \"%s\": %s", dst, strerror(errno));
+ if (PageIsNew(page))
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Rewind the file position, so the first write will overwrite the page.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+
+ src_buf_offset = 0;
+ while (src_buf_offset < src_len)
+ {
+ len_to_copy = Min(src_len - src_buf_offset, SizeOfPageContents - dst_offset);
+ memcpy(PageGetContents(page) + dst_offset, src_buf + src_buf_offset, len_to_copy);
+ src_buf_offset += len_to_copy;
+
+ if (new_cluster.controldata.data_checksum_version > 0)
+ ((PageHeader) page)->pd_checksum = pg_checksum_page(page, dst_pageno);
+ if (write(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not write file \"%s\": %s", dst, strerror(errno));
+
+ dst_pageno++;
+ dst_offset = 0;
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Switch segments if we reached the end of the current segment.
+ */
+ if (dst_pageno % SLRU_PAGES_PER_SEGMENT == 0)
+ {
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ dst_segno++;
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+ }
+ }
+
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ pg_free(src_buf);
+ close(src_fd);
+}
+
+void
+upgrade_xact_cache(const char *src_subdir, const char *dst_subdir)
+{
+ char src_dir[MAXPGPATH];
+ char dst_dir[MAXPGPATH];
+
+ DIR *src_dirp;
+ struct dirent *src_dirent;
+
+ snprintf(src_dir, sizeof(src_dir), "%s/%s", old_cluster.pgdata, src_subdir);
+ snprintf(dst_dir, sizeof(dst_dir), "%s/%s", new_cluster.pgdata, dst_subdir);
+
+ if ((src_dirp = opendir(src_dir)) == NULL)
+ pg_fatal("could not open directory \"%s\": %s", src_dir, strerror(errno));
+
+ while (errno = 0, (src_dirent = readdir(src_dirp)) != NULL)
+ {
+ if (segment_file_filter(src_dirent))
+ upgrade_file(src_dir, src_dirent->d_name, dst_dir);
+ }
+
+ if (closedir(src_dirp) != 0)
+ pg_fatal("could not close directory \"%s\": %s", src_dir, strerror(errno));
+}
diff --git a/src/bin/pg_upgrade/pg_upgrade.c b/src/bin/pg_upgrade/pg_upgrade.c
index 3960af4036..2b7d01058e 100644
--- a/src/bin/pg_upgrade/pg_upgrade.c
+++ b/src/bin/pg_upgrade/pg_upgrade.c
@@ -660,14 +660,23 @@ copy_subdir_files(const char *old_subdir, const char *new_subdir)
static void
copy_xact_xlog_xid(void)
{
+ bool slru_header_changed = false;
+
/*
* Copy old commit logs to new data dir. pg_clog has been renamed to
* pg_xact in post-10 clusters.
*/
- copy_subdir_files(GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact",
- GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact");
+ char *xact_old_dir = GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+ char *xact_new_dir = GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+
+ if (new_cluster.controldata.cat_ver >= SLRU_PAGE_HEADER_CAT_VER &&
+ old_cluster.controldata.cat_ver < SLRU_PAGE_HEADER_CAT_VER)
+ slru_header_changed = true;
+
+ if (slru_header_changed)
+ upgrade_xact_cache(xact_old_dir, xact_new_dir);
+ else
+ copy_subdir_files(xact_old_dir, xact_new_dir);
prep_status("Setting oldest XID for new cluster");
exec_prog(UTILITY_LOG_FILE, NULL, true, true,
@@ -702,7 +711,8 @@ copy_xact_xlog_xid(void)
* server doesn't attempt to read multis older than the cutoff value.
*/
if (old_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
- new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
+ !slru_header_changed)
{
copy_subdir_files("pg_multixact/offsets", "pg_multixact/offsets");
copy_subdir_files("pg_multixact/members", "pg_multixact/members");
@@ -722,7 +732,8 @@ copy_xact_xlog_xid(void)
new_cluster.pgdata);
check_ok();
}
- else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER ||
+ slru_header_changed)
{
/*
* Remove offsets/0000 file created by initdb that no longer matches
diff --git a/src/bin/pg_upgrade/pg_upgrade.h b/src/bin/pg_upgrade/pg_upgrade.h
index d63f13fffc..c955e903a5 100644
--- a/src/bin/pg_upgrade/pg_upgrade.h
+++ b/src/bin/pg_upgrade/pg_upgrade.h
@@ -115,6 +115,11 @@ extern char *output_files[];
*/
#define MULTIXACT_FORMATCHANGE_CAT_VER 201301231
+/*
+ * A page header was added to each SLRU page in 17.0.
+ */
+#define SLRU_PAGE_HEADER_CAT_VER 202312091
+
/*
* large object chunk size added to pg_controldata,
* commit 5f93c37805e7485488480916b4585e098d3cc883
@@ -406,6 +411,7 @@ void rewriteVisibilityMap(const char *fromfile, const char *tofile,
const char *schemaName, const char *relName);
void check_file_clone(void);
void check_hard_link(void);
+void upgrade_xact_cache(const char *src_subdir, const char *dst_subdir);
/* fopen_priv() is no longer different from fopen() */
#define fopen_priv(path, mode) fopen(path, mode)
diff --git a/src/include/access/slru.h b/src/include/access/slru.h
index 091e2202c9..24733166b8 100644
--- a/src/include/access/slru.h
+++ b/src/include/access/slru.h
@@ -68,17 +68,6 @@ typedef struct SlruSharedData
int *page_lru_count;
LWLockPadded *buffer_locks;
- /*
- * Optional array of WAL flush LSNs associated with entries in the SLRU
- * pages. If not zero/NULL, we must flush WAL before writing pages (true
- * for pg_xact, false for multixact, pg_subtrans, pg_notify). group_lsn[]
- * has lsn_groups_per_page entries per buffer slot, each containing the
- * highest LSN known for a contiguous group of SLRU entries on that slot's
- * page.
- */
- XLogRecPtr *group_lsn;
- int lsn_groups_per_page;
-
/*----------
* We mark a page "most recently used" by setting
* page_lru_count[slotno] = ++cur_lru_count;
@@ -147,8 +136,8 @@ typedef struct SlruCtlData
typedef SlruCtlData *SlruCtl;
-extern Size SimpleLruShmemSize(int nslots, int nlsns);
-extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+extern Size SimpleLruShmemSize(int nslots);
+extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler,
bool long_segment_names);
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index b3fdbd9066..be4b8e1c7e 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
*/
/* yyyymmddN */
-#define CATALOG_VERSION_NO 202401021
+#define CATALOG_VERSION_NO 202401031
#endif
diff --git a/src/include/storage/bufpage.h b/src/include/storage/bufpage.h
index 424ecba028..251d9523fa 100644
--- a/src/include/storage/bufpage.h
+++ b/src/include/storage/bufpage.h
@@ -201,6 +201,7 @@ typedef PageHeaderData *PageHeader;
* handling pages.
*/
#define PG_PAGE_LAYOUT_VERSION 4
+#define PG_SLRU_PAGE_LAYOUT_VERSION 1
#define PG_DATA_CHECKSUM_VERSION 1
/* ----------------------------------------------------------------
@@ -257,6 +258,11 @@ PageGetContents(Page page)
return (char *) page + MAXALIGN(SizeOfPageHeaderData);
}
+/*
+ * Space available for storing page contents.
+ */
+#define SizeOfPageContents (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
+
/* ----------------
* functions to access page size info
* ----------------
@@ -486,6 +492,7 @@ StaticAssertDecl(BLCKSZ == ((BLCKSZ / sizeof(size_t)) * sizeof(size_t)),
"BLCKSZ has to be a multiple of sizeof(size_t)");
extern void PageInit(Page page, Size pageSize, Size specialSize);
+extern void PageInitSLRU(Page page, Size pageSize, Size specialSize);
extern bool PageIsVerifiedExtended(Page page, BlockNumber blkno, int flags);
extern OffsetNumber PageAddItemExtended(Page page, Item item, Size size,
OffsetNumber offsetNumber, int flags);
diff --git a/src/test/modules/test_slru/test_slru.c b/src/test/modules/test_slru/test_slru.c
index d0fb9444e8..aa8a29d92e 100644
--- a/src/test/modules/test_slru/test_slru.c
+++ b/src/test/modules/test_slru/test_slru.c
@@ -17,6 +17,7 @@
#include "access/slru.h"
#include "access/transam.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/shmem.h"
@@ -76,8 +77,8 @@ test_slru_page_write(PG_FUNCTION_ARGS)
TestSlruCtl->shared->page_status[slotno] = SLRU_PAGE_VALID;
/* write given data to the page, up to the limit of the page */
- strncpy(TestSlruCtl->shared->page_buffer[slotno], data,
- BLCKSZ - 1);
+ strncpy(PageGetContents(TestSlruCtl->shared->page_buffer[slotno]), data,
+ SizeOfPageContents - 1);
SimpleLruWritePage(TestSlruCtl, slotno);
LWLockRelease(TestSLRULock);
@@ -104,7 +105,7 @@ test_slru_page_read(PG_FUNCTION_ARGS)
LWLockAcquire(TestSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(TestSlruCtl, pageno,
write_ok, InvalidTransactionId);
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -122,7 +123,7 @@ test_slru_page_readonly(PG_FUNCTION_ARGS)
pageno,
InvalidTransactionId);
Assert(LWLockHeldByMe(TestSLRULock));
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -202,7 +203,7 @@ test_slru_shmem_request(void)
prev_shmem_request_hook();
/* reserve shared memory for the test SLRU */
- RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS, 0));
+ RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS));
}
static bool
@@ -238,7 +239,7 @@ test_slru_shmem_startup(void)
TestSlruCtl->PagePrecedes = test_slru_page_precedes_logically;
SimpleLruInit(TestSlruCtl, "TestSLRU",
- NUM_TEST_BUFFERS, 0, TestSLRULock, slru_dir_name,
+ NUM_TEST_BUFFERS, TestSLRULock, slru_dir_name,
test_tranche_id, SYNC_HANDLER_NONE, long_segment_names);
}
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-01-16 09:12 ` Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2024-01-16 09:12 UTC (permalink / raw)
To: Aleksander Alekseev <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
Rebase the patch against the latest HEAD.
Regards,
Yong
Attachments:
[application/octet-stream] slru_page_header_v4.patch (54.8K, ../../[email protected]/3-slru_page_header_v4.patch)
download | inline diff:
src/backend/access/transam/clog.c | 52 +++++-----
src/backend/access/transam/commit_ts.c | 26 +++--
src/backend/access/transam/multixact.c | 63 +++++++-----
src/backend/access/transam/slru.c | 113 +++++++---------------
src/backend/access/transam/subtrans.c | 12 +--
src/backend/commands/async.c | 27 +++---
src/backend/storage/lmgr/predicate.c | 16 +--
src/backend/storage/page/bufpage.c | 25 +++++
src/bin/pg_checksums/pg_checksums.c | 9 ++
src/bin/pg_resetwal/t/001_basic.pl | 6 +-
src/bin/pg_upgrade/file.c | 172 +++++++++++++++++++++++++++++++++
src/bin/pg_upgrade/pg_upgrade.c | 23 +++--
src/bin/pg_upgrade/pg_upgrade.h | 6 ++
src/include/access/slru.h | 15 +--
src/include/catalog/catversion.h | 2 +-
src/include/storage/bufpage.h | 7 ++
src/test/modules/test_slru/test_slru.c | 13 +--
17 files changed, 395 insertions(+), 192 deletions(-)
diff --git a/src/backend/access/transam/clog.c b/src/backend/access/transam/clog.c
index f6e7da7ffc..ebfc6e1ef8 100644
--- a/src/backend/access/transam/clog.c
+++ b/src/backend/access/transam/clog.c
@@ -41,6 +41,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/proc.h"
#include "storage/sync.h"
@@ -59,7 +60,7 @@
/* We need two bits per xact, so four xacts fit in a byte */
#define CLOG_BITS_PER_XACT 2
#define CLOG_XACTS_PER_BYTE 4
-#define CLOG_XACTS_PER_PAGE (BLCKSZ * CLOG_XACTS_PER_BYTE)
+#define CLOG_XACTS_PER_PAGE (SizeOfPageContents * CLOG_XACTS_PER_BYTE)
#define CLOG_XACT_BITMASK ((1 << CLOG_BITS_PER_XACT) - 1)
@@ -77,13 +78,6 @@ TransactionIdToPage(TransactionId xid)
#define TransactionIdToByte(xid) (TransactionIdToPgIndex(xid) / CLOG_XACTS_PER_BYTE)
#define TransactionIdToBIndex(xid) ((xid) % (TransactionId) CLOG_XACTS_PER_BYTE)
-/* We store the latest async LSN for each group of transactions */
-#define CLOG_XACTS_PER_LSN_GROUP 32 /* keep this a power of 2 */
-#define CLOG_LSNS_PER_PAGE (CLOG_XACTS_PER_PAGE / CLOG_XACTS_PER_LSN_GROUP)
-
-#define GetLSNIndex(slotno, xid) ((slotno) * CLOG_LSNS_PER_PAGE + \
- ((xid) % (TransactionId) CLOG_XACTS_PER_PAGE) / CLOG_XACTS_PER_LSN_GROUP)
-
/*
* The number of subtransactions below which we consider to apply clog group
* update optimization. Testing reveals that the number higher than this can
@@ -101,7 +95,7 @@ static SlruCtlData XactCtlData;
static int ZeroCLOGPage(int64 pageno, bool writeXlog);
static bool CLOGPagePrecedes(int64 page1, int64 page2);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXact,
Oid oldestXactDb);
static void TransactionIdSetPageStatus(TransactionId xid, int nsubxids,
@@ -583,8 +577,9 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
char *byteptr;
char byteval;
char curval;
+ Page page = XactCtl->shared->page_buffer[slotno];
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(page) + byteno;
curval = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
/*
@@ -613,7 +608,7 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*byteptr = byteval;
/*
- * Update the group LSN if the transaction completion LSN is higher.
+ * Update the page LSN if the transaction completion LSN is higher.
*
* Note: lsn will be invalid when supplied during InRecovery processing,
* so we don't need to do anything special to avoid LSN updates during
@@ -622,10 +617,8 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*/
if (!XLogRecPtrIsInvalid(lsn))
{
- int lsnindex = GetLSNIndex(slotno, xid);
-
- if (XactCtl->shared->group_lsn[lsnindex] < lsn)
- XactCtl->shared->group_lsn[lsnindex] = lsn;
+ if (PageGetLSN(page) < lsn)
+ PageSetLSN(page, lsn);
}
}
@@ -651,19 +644,19 @@ TransactionIdGetStatus(TransactionId xid, XLogRecPtr *lsn)
int byteno = TransactionIdToByte(xid);
int bshift = TransactionIdToBIndex(xid) * CLOG_BITS_PER_XACT;
int slotno;
- int lsnindex;
+ Page page;
char *byteptr;
XidStatus status;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(XactCtl, pageno, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
status = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
- lsnindex = GetLSNIndex(slotno, xid);
- *lsn = XactCtl->shared->group_lsn[lsnindex];
+ *lsn = PageGetLSN(page);
LWLockRelease(XactSLRULock);
@@ -698,14 +691,14 @@ CLOGShmemBuffers(void)
Size
CLOGShmemSize(void)
{
- return SimpleLruShmemSize(CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE);
+ return SimpleLruShmemSize(CLOGShmemBuffers());
}
void
CLOGShmemInit(void)
{
XactCtl->PagePrecedes = CLOGPagePrecedes;
- SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE,
+ SimpleLruInit(XactCtl, "Xact", CLOGShmemBuffers(),
XactSLRULock, "pg_xact", LWTRANCHE_XACT_BUFFER,
SYNC_HANDLER_CLOG, false);
SlruPagePrecedesUnitTests(XactCtl, CLOG_XACTS_PER_PAGE);
@@ -747,11 +740,17 @@ static int
ZeroCLOGPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(XactCtl, pageno);
+ page = XactCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -807,12 +806,12 @@ TrimCLOG(void)
char *byteptr;
slotno = SimpleLruReadPage(XactCtl, pageno, false, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(XactCtl->shared->page_buffer[slotno]) + byteno;
/* Zero so-far-unused positions in the current byte */
*byteptr &= (1 << bshift) - 1;
/* Zero the rest of the page */
- MemSet(byteptr + 1, 0, BLCKSZ - byteno - 1);
+ MemSet(byteptr + 1, 0, SizeOfPageContents - byteno - 1);
XactCtl->shared->page_dirty[slotno] = true;
}
@@ -836,7 +835,6 @@ CheckPointCLOG(void)
TRACE_POSTGRESQL_CLOG_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that CLOG has room for a newly-allocated XID.
*
@@ -958,12 +956,12 @@ CLOGPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
+ return XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/commit_ts.c b/src/backend/access/transam/commit_ts.c
index 61b82385f3..cc9388f5d2 100644
--- a/src/backend/access/transam/commit_ts.c
+++ b/src/backend/access/transam/commit_ts.c
@@ -31,6 +31,7 @@
#include "funcapi.h"
#include "miscadmin.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "storage/shmem.h"
#include "utils/builtins.h"
#include "utils/snapmgr.h"
@@ -63,7 +64,7 @@ typedef struct CommitTimestampEntry
sizeof(RepOriginId))
#define COMMIT_TS_XACTS_PER_PAGE \
- (BLCKSZ / SizeOfCommitTimestampEntry)
+ (SizeOfPageContents / SizeOfCommitTimestampEntry)
/*
@@ -120,7 +121,7 @@ static int ZeroCommitTsPage(int64 pageno, bool writeXlog);
static bool CommitTsPagePrecedes(int64 page1, int64 page2);
static void ActivateCommitTs(void);
static void DeactivateCommitTs(void);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXid);
/*
@@ -254,11 +255,12 @@ TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
+ Assert(xid == slotno * COMMIT_TS_XACTS_PER_PAGE + entryno);
entry.time = ts;
entry.nodeid = nodeid;
- memcpy(CommitTsCtl->shared->page_buffer[slotno] +
+ memcpy(PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
@@ -337,7 +339,7 @@ TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
- CommitTsCtl->shared->page_buffer[slotno] +
+ PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
@@ -515,7 +517,7 @@ CommitTsShmemBuffers(void)
Size
CommitTsShmemSize(void)
{
- return SimpleLruShmemSize(CommitTsShmemBuffers(), 0) +
+ return SimpleLruShmemSize(CommitTsShmemBuffers()) +
sizeof(CommitTimestampShared);
}
@@ -529,7 +531,7 @@ CommitTsShmemInit(void)
bool found;
CommitTsCtl->PagePrecedes = CommitTsPagePrecedes;
- SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(), 0,
+ SimpleLruInit(CommitTsCtl, "CommitTs", CommitTsShmemBuffers(),
CommitTsSLRULock, "pg_commit_ts",
LWTRANCHE_COMMITTS_BUFFER,
SYNC_HANDLER_COMMIT_TS,
@@ -582,11 +584,17 @@ static int
ZeroCommitTsPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(CommitTsCtl, pageno);
+ page = CommitTsCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -946,12 +954,12 @@ CommitTsPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
+ return XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c
index 59523be901..af6563c889 100644
--- a/src/backend/access/transam/multixact.c
+++ b/src/backend/access/transam/multixact.c
@@ -83,6 +83,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "postmaster/autovacuum.h"
+#include "storage/bufpage.h"
#include "storage/lmgr.h"
#include "storage/pmsignal.h"
#include "storage/proc.h"
@@ -106,7 +107,7 @@
*/
/* We need four bytes per offset */
-#define MULTIXACT_OFFSETS_PER_PAGE (BLCKSZ / sizeof(MultiXactOffset))
+#define MULTIXACT_OFFSETS_PER_PAGE (SizeOfPageContents / sizeof(MultiXactOffset))
#define MultiXactIdToOffsetPage(xid) \
((xid) / (MultiXactOffset) MULTIXACT_OFFSETS_PER_PAGE)
@@ -119,8 +120,8 @@
* additional flag bits for each TransactionId. To do this without getting
* into alignment issues, we store four bytes of flags, and then the
* corresponding 4 Xids. Each such 5-word (20-byte) set we call a "group", and
- * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 409 groups
- * per page. This wastes 12 bytes per page, but that's OK -- simplicity (and
+ * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 408 groups
+ * per page. This wastes 8 bytes per page, but that's OK -- simplicity (and
* performance) trumps space efficiency here.
*
* Note that the "offset" macros work with byte offset, not array indexes, so
@@ -138,7 +139,7 @@
/* size in bytes of a complete group */
#define MULTIXACT_MEMBERGROUP_SIZE \
(sizeof(TransactionId) * MULTIXACT_MEMBERS_PER_MEMBERGROUP + MULTIXACT_FLAGBYTES_PER_GROUP)
-#define MULTIXACT_MEMBERGROUPS_PER_PAGE (BLCKSZ / MULTIXACT_MEMBERGROUP_SIZE)
+#define MULTIXACT_MEMBERGROUPS_PER_PAGE (SizeOfPageContents / MULTIXACT_MEMBERGROUP_SIZE)
#define MULTIXACT_MEMBERS_PER_PAGE \
(MULTIXACT_MEMBERGROUPS_PER_PAGE * MULTIXACT_MEMBERS_PER_MEMBERGROUP)
@@ -366,7 +367,7 @@ static bool MultiXactOffsetWouldWrap(MultiXactOffset boundary,
MultiXactOffset start, uint32 distance);
static bool SetOffsetVacuumLimit(bool is_startup);
static bool find_multixact_start(MultiXactId multi, MultiXactOffset *result);
-static void WriteMZeroPageXlogRec(int64 pageno, uint8 info);
+static XLogRecPtr WriteMZeroPageXlogRec(int64 pageno, uint8 info);
static void WriteMTruncateXlogRec(Oid oldestMultiDB,
MultiXactId startTruncOff,
MultiXactId endTruncOff,
@@ -884,7 +885,7 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
* take the trouble to generalize the slru.c error reporting code.
*/
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
*offptr = offset;
@@ -921,12 +922,12 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
}
memberptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
*memberptr = members[i].xid;
flagsptr = (uint32 *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
flagsval = *flagsptr;
flagsval &= ~(((1 << MXACT_MEMBER_BITS_PER_XACT) - 1) << bshift);
@@ -1348,7 +1349,7 @@ retry:
entryno = MultiXactIdToOffsetEntry(multi);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
@@ -1381,7 +1382,7 @@ retry:
if (pageno != prev_pageno)
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, tmpMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
nextMXOffset = *offptr;
@@ -1424,7 +1425,7 @@ retry:
}
xactptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
if (!TransactionIdIsValid(*xactptr))
{
@@ -1435,7 +1436,7 @@ retry:
flagsoff = MXOffsetToFlagsOffset(offset);
bshift = MXOffsetToFlagsBitShift(offset);
- flagsptr = (uint32 *) (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ flagsptr = (uint32 *) (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
ptr[truelength].xid = *xactptr;
ptr[truelength].status = (*flagsptr >> bshift) & MXACT_MEMBER_XACT_BITMASK;
@@ -1834,8 +1835,8 @@ MultiXactShmemSize(void)
mul_size(sizeof(MultiXactId) * 2, MaxOldestSlot))
size = SHARED_MULTIXACT_STATE_SIZE;
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS, 0));
- size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTOFFSET_BUFFERS));
+ size = add_size(size, SimpleLruShmemSize(NUM_MULTIXACTMEMBER_BUFFERS));
return size;
}
@@ -1851,14 +1852,14 @@ MultiXactShmemInit(void)
MultiXactMemberCtl->PagePrecedes = MultiXactMemberPagePrecedes;
SimpleLruInit(MultiXactOffsetCtl,
- "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS, 0,
+ "MultiXactOffset", NUM_MULTIXACTOFFSET_BUFFERS,
MultiXactOffsetSLRULock, "pg_multixact/offsets",
LWTRANCHE_MULTIXACTOFFSET_BUFFER,
SYNC_HANDLER_MULTIXACT_OFFSET,
false);
SlruPagePrecedesUnitTests(MultiXactOffsetCtl, MULTIXACT_OFFSETS_PER_PAGE);
SimpleLruInit(MultiXactMemberCtl,
- "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS, 0,
+ "MultiXactMember", NUM_MULTIXACTMEMBER_BUFFERS,
MultiXactMemberSLRULock, "pg_multixact/members",
LWTRANCHE_MULTIXACTMEMBER_BUFFER,
SYNC_HANDLER_MULTIXACT_MEMBER,
@@ -1933,11 +1934,17 @@ static int
ZeroMultiXactOffsetPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactOffsetCtl, pageno);
+ page = MultiXactOffsetCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -1949,11 +1956,17 @@ static int
ZeroMultiXactMemberPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactMemberCtl, pageno);
+ page = MultiXactMemberCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2071,10 +2084,10 @@ TrimMultiXact(void)
MultiXactOffset *offptr;
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, nextMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
- MemSet(offptr, 0, BLCKSZ - (entryno * sizeof(MultiXactOffset)));
+ MemSet(offptr, 0, SizeOfPageContents - (entryno * sizeof(MultiXactOffset)));
MultiXactOffsetCtl->shared->page_dirty[slotno] = true;
}
@@ -2104,9 +2117,9 @@ TrimMultiXact(void)
memberoff = MXOffsetToMemberOffset(offset);
slotno = SimpleLruReadPage(MultiXactMemberCtl, pageno, true, offset);
xidptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
- MemSet(xidptr, 0, BLCKSZ - memberoff);
+ MemSet(xidptr, 0, SizeOfPageContents - memberoff);
/*
* Note: we don't need to zero out the flag bits in the remaining
@@ -2758,7 +2771,7 @@ find_multixact_start(MultiXactId multi, MultiXactOffset *result)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(MultiXactOffsetCtl, pageno, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
LWLockRelease(MultiXactOffsetSLRULock);
@@ -3192,12 +3205,12 @@ MultiXactOffsetPrecedes(MultiXactOffset offset1, MultiXactOffset offset2)
* Write an xlog record reflecting the zeroing of either a MEMBERs or
* OFFSETs page (info shows which)
*/
-static void
+static XLogRecPtr
WriteMZeroPageXlogRec(int64 pageno, uint8 info)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_MULTIXACT_ID, info);
+ return XLogInsert(RM_MULTIXACT_ID, info);
}
/*
diff --git a/src/backend/access/transam/slru.c b/src/backend/access/transam/slru.c
index 9ac4790f16..c42a44fcaa 100644
--- a/src/backend/access/transam/slru.c
+++ b/src/backend/access/transam/slru.c
@@ -57,6 +57,7 @@
#include "access/xlogutils.h"
#include "miscadmin.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/shmem.h"
@@ -154,13 +155,13 @@ typedef enum
SLRU_WRITE_FAILED,
SLRU_FSYNC_FAILED,
SLRU_CLOSE_FAILED,
+ SLRU_DATA_CORRUPTED,
} SlruErrorCause;
static SlruErrorCause slru_errcause;
static int slru_errno;
-static void SimpleLruZeroLSNs(SlruCtl ctl, int slotno);
static void SimpleLruWaitIO(SlruCtl ctl, int slotno);
static void SlruInternalWritePage(SlruCtl ctl, int slotno, SlruWriteAll fdata);
static bool SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno);
@@ -179,7 +180,7 @@ static void SlruInternalDeleteSegment(SlruCtl ctl, int64 segno);
*/
Size
-SimpleLruShmemSize(int nslots, int nlsns)
+SimpleLruShmemSize(int nslots)
{
Size sz;
@@ -192,9 +193,6 @@ SimpleLruShmemSize(int nslots, int nlsns)
sz += MAXALIGN(nslots * sizeof(int)); /* page_lru_count[] */
sz += MAXALIGN(nslots * sizeof(LWLockPadded)); /* buffer_locks[] */
- if (nlsns > 0)
- sz += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr)); /* group_lsn[] */
-
return BUFFERALIGN(sz) + BLCKSZ * nslots;
}
@@ -204,14 +202,13 @@ SimpleLruShmemSize(int nslots, int nlsns)
* ctl: address of local (unshared) control structure.
* name: name of SLRU. (This is user-visible, pick with care!)
* nslots: number of page slots to use.
- * nlsns: number of LSN groups per page (set to zero if not relevant).
* ctllock: LWLock to use to control access to the shared control structure.
* subdir: PGDATA-relative subdirectory that will contain the files.
* tranche_id: LWLock tranche ID to use for the SLRU's per-buffer LWLocks.
* sync_handler: which set of functions to use to handle sync requests
*/
void
-SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler, bool long_segment_names)
{
@@ -219,7 +216,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
bool found;
shared = (SlruShared) ShmemInitStruct(name,
- SimpleLruShmemSize(nslots, nlsns),
+ SimpleLruShmemSize(nslots),
&found);
if (!IsUnderPostmaster)
@@ -236,7 +233,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->ControlLock = ctllock;
shared->num_slots = nslots;
- shared->lsn_groups_per_page = nlsns;
shared->cur_lru_count = 0;
@@ -261,12 +257,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->buffer_locks = (LWLockPadded *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(LWLockPadded));
- if (nlsns > 0)
- {
- shared->group_lsn = (XLogRecPtr *) (ptr + offset);
- offset += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr));
- }
-
ptr += BUFFERALIGN(offset);
for (slotno = 0; slotno < nslots; slotno++)
{
@@ -281,7 +271,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
}
/* Should fit to estimated shmem size */
- Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots, nlsns));
+ Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots));
}
else
Assert(found);
@@ -323,11 +313,8 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
shared->page_dirty[slotno] = true;
SlruRecentlyUsed(shared, slotno);
- /* Set the buffer to zeroes */
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
-
- /* Set the LSNs for this new page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
+ /* Initialize the page. */
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
/* Assume this page is now the latest active page */
shared->latest_page_number = pageno;
@@ -338,26 +325,6 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
return slotno;
}
-/*
- * Zero all the LSNs we store for this slru page.
- *
- * This should be called each time we create a new page, and each time we read
- * in a page from disk into an existing buffer. (Such an old page cannot
- * have any interesting LSNs, since we'd have flushed them before writing
- * the page in the first place.)
- *
- * This assumes that InvalidXLogRecPtr is bitwise-all-0.
- */
-static void
-SimpleLruZeroLSNs(SlruCtl ctl, int slotno)
-{
- SlruShared shared = ctl->shared;
-
- if (shared->lsn_groups_per_page > 0)
- MemSet(&shared->group_lsn[slotno * shared->lsn_groups_per_page], 0,
- shared->lsn_groups_per_page * sizeof(XLogRecPtr));
-}
-
/*
* Wait for any active I/O on a page slot to finish. (This does not
* guarantee that new I/O hasn't been started before we return, though.
@@ -478,9 +445,6 @@ SimpleLruReadPage(SlruCtl ctl, int64 pageno, bool write_ok,
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
- /* Set the LSNs for this newly read-in page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
-
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
@@ -740,7 +704,7 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
ereport(LOG,
(errmsg("file \"%s\" doesn't exist, reading as zeroes",
path)));
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
return true;
}
@@ -763,6 +727,13 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
return false;
}
+ if (!PageIsVerifiedExtended(shared->page_buffer[slotno], pageno, PIV_REPORT_STAT))
+ {
+ slru_errcause = SLRU_DATA_CORRUPTED;
+ slru_errno = 0;
+ return false;
+ }
+
return true;
}
@@ -789,6 +760,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
off_t offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd = -1;
+ Page page = shared->page_buffer[slotno];
+ XLogRecPtr lsn;
/* update the stats counter of written pages */
pgstat_count_slru_page_written(shared->slru_stats_idx);
@@ -798,41 +771,18 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
* write out data before associated WAL records. This is the same action
* performed during FlushBuffer() in the main buffer manager.
*/
- if (shared->group_lsn != NULL)
+ lsn = PageGetLSN(page);
+ if (!XLogRecPtrIsInvalid(lsn))
{
/*
- * We must determine the largest async-commit LSN for the page. This
- * is a bit tedious, but since this entire function is a slow path
- * anyway, it seems better to do this here than to maintain a per-page
- * LSN variable (which'd need an extra comparison in the
- * transaction-commit path).
+ * As noted above, elog(ERROR) is not acceptable here, so if
+ * XLogFlush were to fail, we must PANIC. This isn't much of a
+ * restriction because XLogFlush is just about all critical
+ * section anyway, but let's make sure.
*/
- XLogRecPtr max_lsn;
- int lsnindex,
- lsnoff;
-
- lsnindex = slotno * shared->lsn_groups_per_page;
- max_lsn = shared->group_lsn[lsnindex++];
- for (lsnoff = 1; lsnoff < shared->lsn_groups_per_page; lsnoff++)
- {
- XLogRecPtr this_lsn = shared->group_lsn[lsnindex++];
-
- if (max_lsn < this_lsn)
- max_lsn = this_lsn;
- }
-
- if (!XLogRecPtrIsInvalid(max_lsn))
- {
- /*
- * As noted above, elog(ERROR) is not acceptable here, so if
- * XLogFlush were to fail, we must PANIC. This isn't much of a
- * restriction because XLogFlush is just about all critical
- * section anyway, but let's make sure.
- */
- START_CRIT_SECTION();
- XLogFlush(max_lsn);
- END_CRIT_SECTION();
- }
+ START_CRIT_SECTION();
+ XLogFlush(lsn);
+ END_CRIT_SECTION();
}
/*
@@ -899,6 +849,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
}
}
+ PageSetChecksumInplace(shared->page_buffer[slotno], pageno);
+
errno = 0;
pgstat_report_wait_start(WAIT_EVENT_SLRU_WRITE);
if (pg_pwrite(fd, shared->page_buffer[slotno], BLCKSZ, offset) != BLCKSZ)
@@ -1019,6 +971,13 @@ SlruReportIOError(SlruCtl ctl, int64 pageno, TransactionId xid)
errdetail("Could not close file \"%s\": %m.",
path)));
break;
+ case SLRU_DATA_CORRUPTED:
+ ereport(ERROR,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("could not access status of transaction %u", xid),
+ errdetail("Invalid page from file \"%s\" at offset %d.",
+ path, offset)));
+ break;
default:
/* can't get here, we trust */
elog(ERROR, "unrecognized SimpleLru error cause: %d",
diff --git a/src/backend/access/transam/subtrans.c b/src/backend/access/transam/subtrans.c
index b2ed82ac56..a18b357fe2 100644
--- a/src/backend/access/transam/subtrans.c
+++ b/src/backend/access/transam/subtrans.c
@@ -32,6 +32,7 @@
#include "access/subtrans.h"
#include "access/transam.h"
#include "pg_trace.h"
+#include "storage/bufpage.h"
#include "utils/snapmgr.h"
@@ -49,7 +50,7 @@
*/
/* We need four bytes per xact */
-#define SUBTRANS_XACTS_PER_PAGE (BLCKSZ / sizeof(TransactionId))
+#define SUBTRANS_XACTS_PER_PAGE (SizeOfPageContents / sizeof(TransactionId))
/*
* Although we return an int64 the actual value can't currently exceed
@@ -93,7 +94,7 @@ SubTransSetParent(TransactionId xid, TransactionId parent)
LWLockAcquire(SubtransSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(SubTransCtl, pageno, true, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
/*
@@ -133,7 +134,7 @@ SubTransGetParent(TransactionId xid)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
parent = *ptr;
@@ -193,14 +194,14 @@ SubTransGetTopmostTransaction(TransactionId xid)
Size
SUBTRANSShmemSize(void)
{
- return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS, 0);
+ return SimpleLruShmemSize(NUM_SUBTRANS_BUFFERS);
}
void
SUBTRANSShmemInit(void)
{
SubTransCtl->PagePrecedes = SubTransPagePrecedes;
- SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS, 0,
+ SimpleLruInit(SubTransCtl, "Subtrans", NUM_SUBTRANS_BUFFERS,
SubtransSLRULock, "pg_subtrans",
LWTRANCHE_SUBTRANS_BUFFER, SYNC_HANDLER_NONE,
false);
@@ -305,7 +306,6 @@ CheckPointSUBTRANS(void)
TRACE_POSTGRESQL_SUBTRANS_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that SUBTRANS has room for a newly-allocated XID.
*
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c
index 8b24b22293..d3b8d54903 100644
--- a/src/backend/commands/async.c
+++ b/src/backend/commands/async.c
@@ -140,6 +140,7 @@
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/proc.h"
@@ -162,7 +163,7 @@
* than that, so changes in that data structure won't affect user-visible
* restrictions.
*/
-#define NOTIFY_PAYLOAD_MAX_LENGTH (BLCKSZ - NAMEDATALEN - 128)
+#define NOTIFY_PAYLOAD_MAX_LENGTH (SizeOfPageContents - NAMEDATALEN - 128)
/*
* Struct representing an entry in the global notify queue
@@ -311,7 +312,7 @@ static SlruCtlData NotifyCtlData;
#define NotifyCtl (&NotifyCtlData)
#define QUEUE_PAGESIZE BLCKSZ
-
+#define QUEUE_PAGE_CAPACITY (QUEUE_PAGESIZE - MAXALIGN(SizeOfPageHeaderData))
#define QUEUE_FULL_WARN_INTERVAL 5000 /* warn at most once every 5s */
/*
@@ -492,7 +493,7 @@ AsyncShmemSize(void)
size = mul_size(MaxBackends + 1, sizeof(QueueBackendStatus));
size = add_size(size, offsetof(AsyncQueueControl, backend));
- size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_NOTIFY_BUFFERS));
return size;
}
@@ -541,7 +542,7 @@ AsyncShmemInit(void)
* names are used in order to avoid wraparound.
*/
NotifyCtl->PagePrecedes = asyncQueuePagePrecedes;
- SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS, 0,
+ SimpleLruInit(NotifyCtl, "Notify", NUM_NOTIFY_BUFFERS,
NotifySLRULock, "pg_notify", LWTRANCHE_NOTIFY_BUFFER,
SYNC_HANDLER_NONE, true);
@@ -1301,14 +1302,14 @@ asyncQueueAdvance(volatile QueuePosition *position, int entryLength)
* written or read.
*/
offset += entryLength;
- Assert(offset <= QUEUE_PAGESIZE);
+ Assert(offset <= QUEUE_PAGE_CAPACITY);
/*
* In a second step check if another entry can possibly be written to the
* page. If so, stay here, we have reached the next position. If not, then
* we need to move on to the next page.
*/
- if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGESIZE)
+ if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGE_CAPACITY)
{
pageno++;
offset = 0;
@@ -1408,7 +1409,7 @@ asyncQueueAddEntries(ListCell *nextNotify)
offset = QUEUE_POS_OFFSET(queue_head);
/* Check whether the entry really fits on the current page */
- if (offset + qe.length <= QUEUE_PAGESIZE)
+ if (offset + qe.length <= QUEUE_PAGE_CAPACITY)
{
/* OK, so advance nextNotify past this item */
nextNotify = lnext(pendingNotifies->events, nextNotify);
@@ -1420,14 +1421,14 @@ asyncQueueAddEntries(ListCell *nextNotify)
* only check dboid and since it won't match any reader's database
* OID, they will ignore this entry and move on.
*/
- qe.length = QUEUE_PAGESIZE - offset;
+ qe.length = QUEUE_PAGE_CAPACITY - offset;
qe.dboid = InvalidOid;
qe.data[0] = '\0'; /* empty channel */
qe.data[1] = '\0'; /* empty payload */
}
/* Now copy qe into the shared buffer page */
- memcpy(NotifyCtl->shared->page_buffer[slotno] + offset,
+ memcpy(PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + offset,
&qe,
qe.length);
@@ -1947,10 +1948,10 @@ asyncQueueReadAllNotifications(void)
else
{
/* fetch all the rest of the page */
- copysize = QUEUE_PAGESIZE - curoffset;
+ copysize = QUEUE_PAGE_CAPACITY - curoffset;
}
- memcpy(page_buffer.buf + curoffset,
- NotifyCtl->shared->page_buffer[slotno] + curoffset,
+ memcpy(PageGetContents(page_buffer.buf) + curoffset,
+ PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + curoffset,
copysize);
/* Release lock that we got from SimpleLruReadPage_ReadOnly() */
LWLockRelease(NotifySLRULock);
@@ -2021,7 +2022,7 @@ asyncQueueProcessPageEntries(volatile QueuePosition *current,
if (QUEUE_POS_EQUAL(thisentry, stop))
break;
- qe = (AsyncQueueEntry *) (page_buffer + QUEUE_POS_OFFSET(thisentry));
+ qe = (AsyncQueueEntry *) (PageGetContents(page_buffer) + QUEUE_POS_OFFSET(thisentry));
/*
* Advance *current over this message, possibly to the next page. As
diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index ee5ea1175c..ba9ecbe524 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -204,6 +204,7 @@
#include "pgstat.h"
#include "port/pg_lfind.h"
#include "storage/bufmgr.h"
+#include "storage/bufpage.h"
#include "storage/predicate.h"
#include "storage/predicate_internals.h"
#include "storage/proc.h"
@@ -322,8 +323,8 @@ static SlruCtlData SerialSlruCtlData;
#define SerialSlruCtl (&SerialSlruCtlData)
#define SERIAL_PAGESIZE BLCKSZ
-#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
-#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE / SERIAL_ENTRYSIZE)
+#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
+#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE - MAXALIGN(SizeOfPageHeaderData) / SERIAL_ENTRYSIZE)
/*
* Set maximum pages based on the number needed to track all transactions.
@@ -333,7 +334,7 @@ static SlruCtlData SerialSlruCtlData;
#define SerialNextPage(page) (((page) >= SERIAL_MAX_PAGE) ? 0 : (page) + 1)
#define SerialValue(slotno, xid) (*((SerCommitSeqNo *) \
- (SerialSlruCtl->shared->page_buffer[slotno] + \
+ (PageGetContents(SerialSlruCtl->shared->page_buffer[slotno]) + \
((((uint32) (xid)) % SERIAL_ENTRIESPERPAGE) * SERIAL_ENTRYSIZE))))
#define SerialPage(xid) (((uint32) (xid)) / SERIAL_ENTRIESPERPAGE)
@@ -785,10 +786,13 @@ SerialPagePrecedesLogicallyUnitTests(void)
* requires burning ~2B XIDs in single-user mode, a negligible
* possibility. Moreover, if it does happen, the consequence would be
* mild, namely a new transaction failing in SimpleLruReadPage().
+ *
+ * NOTE: After adding the page header, the defect affects two pages.
+ * We now assert correct treatment of its second to prior page.
*/
headPage = oldestPage;
targetPage = newestPage;
- Assert(SerialPagePrecedesLogically(headPage, targetPage - 1));
+ Assert(SerialPagePrecedesLogically(headPage, targetPage - 2));
#if 0
Assert(SerialPagePrecedesLogically(headPage, targetPage));
#endif
@@ -808,7 +812,7 @@ SerialInit(void)
*/
SerialSlruCtl->PagePrecedes = SerialPagePrecedesLogically;
SimpleLruInit(SerialSlruCtl, "Serial",
- NUM_SERIAL_BUFFERS, 0, SerialSLRULock, "pg_serial",
+ NUM_SERIAL_BUFFERS, SerialSLRULock, "pg_serial",
LWTRANCHE_SERIAL_BUFFER, SYNC_HANDLER_NONE,
false);
#ifdef USE_ASSERT_CHECKING
@@ -1348,7 +1352,7 @@ PredicateLockShmemSize(void)
/* Shared memory structures for SLRU tracking of old committed xids. */
size = add_size(size, sizeof(SerialControlData));
- size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS, 0));
+ size = add_size(size, SimpleLruShmemSize(NUM_SERIAL_BUFFERS));
return size;
}
diff --git a/src/backend/storage/page/bufpage.c b/src/backend/storage/page/bufpage.c
index be6f1f62d2..e8193d7f56 100644
--- a/src/backend/storage/page/bufpage.c
+++ b/src/backend/storage/page/bufpage.c
@@ -59,6 +59,31 @@ PageInit(Page page, Size pageSize, Size specialSize)
/* p->pd_prune_xid = InvalidTransactionId; done by above MemSet */
}
+/*
+ * PageInitSLRU
+ * Initializes the contents of an SLRU page.
+ * Note that we don't calculate an initial checksum here; that's not done
+ * until it's time to write.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
/*
* PageIsVerifiedExtended
diff --git a/src/bin/pg_checksums/pg_checksums.c b/src/bin/pg_checksums/pg_checksums.c
index 6fc1326418..7e6596b8fd 100644
--- a/src/bin/pg_checksums/pg_checksums.c
+++ b/src/bin/pg_checksums/pg_checksums.c
@@ -16,6 +16,7 @@
#include <dirent.h>
#include <limits.h>
+#include <stdbool.h>
#include <time.h>
#include <sys/stat.h>
#include <unistd.h>
@@ -589,12 +590,20 @@ main(int argc, char *argv[])
{
total_size = scan_directory(DataDir, "global", true);
total_size += scan_directory(DataDir, "base", true);
+ total_size += scan_directory(DataDir, "pg_commit_ts", true);
+ total_size += scan_directory(DataDir, "pg_multixact", true);
+ total_size += scan_directory(DataDir, "pg_serial", true);
total_size += scan_directory(DataDir, "pg_tblspc", true);
+ total_size += scan_directory(DataDir, "pg_xact", true);
}
(void) scan_directory(DataDir, "global", false);
(void) scan_directory(DataDir, "base", false);
+ (void) scan_directory(DataDir, "pg_commit_ts", false);
+ (void) scan_directory(DataDir, "pg_multixact", false);
+ (void) scan_directory(DataDir, "pg_serial", false);
(void) scan_directory(DataDir, "pg_tblspc", false);
+ (void) scan_directory(DataDir, "pg_xact", false);
if (showprogress)
progress_report(true);
diff --git a/src/bin/pg_resetwal/t/001_basic.pl b/src/bin/pg_resetwal/t/001_basic.pl
index 9829e48106..7b9e034e19 100644
--- a/src/bin/pg_resetwal/t/001_basic.pl
+++ b/src/bin/pg_resetwal/t/001_basic.pl
@@ -206,7 +206,7 @@ push @cmd,
sprintf("%d,%d", hex($files[0]) == 0 ? 3 : hex($files[0]), hex($files[-1]));
@files = get_slru_files('pg_multixact/offsets');
-$mult = 32 * $blcksz / 4;
+$mult = 32 * ($blcksz - 24) / 4;
# -m argument is "new,old"
push @cmd, '-m',
sprintf("%d,%d",
@@ -214,11 +214,11 @@ push @cmd, '-m',
hex($files[0]) == 0 ? 1 : hex($files[0] * $mult));
@files = get_slru_files('pg_multixact/members');
-$mult = 32 * int($blcksz / 20) * 4;
+$mult = 32 * int(($blcksz - 24) / 20) * 4;
push @cmd, '-O', (hex($files[-1]) + 1) * $mult;
@files = get_slru_files('pg_xact');
-$mult = 32 * $blcksz * 4;
+$mult = 32 * ($blcksz - 24) * 4;
push @cmd,
'-u', (hex($files[0]) == 0 ? 3 : hex($files[0]) * $mult),
'-x', ((hex($files[-1]) + 1) * $mult);
diff --git a/src/bin/pg_upgrade/file.c b/src/bin/pg_upgrade/file.c
index 4850a682cb..64184cecdb 100644
--- a/src/bin/pg_upgrade/file.c
+++ b/src/bin/pg_upgrade/file.c
@@ -9,6 +9,7 @@
#include "postgres_fe.h"
+#include <dirent.h>
#include <sys/stat.h>
#include <fcntl.h>
#ifdef HAVE_COPYFILE_H
@@ -375,3 +376,174 @@ check_hard_link(void)
unlink(new_link_file);
}
+
+
+/*
+ * Copy SLRU_PAGES_PER_SEGMENT from access/slru.h to avoid including it.
+ */
+#define SLRU_PAGES_PER_SEGMENT 32
+
+#define SEGMENT_SIZE (BLCKSZ * SLRU_PAGES_PER_SEGMENT)
+
+/*
+ * Copy PageInitSLRU from storage/bufpage.c to avoid linking to the backend.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
+
+/*
+ * Filter function for scandir(3) to select only segment files.
+ */
+static int
+segment_file_filter(const struct dirent *dirent)
+{
+ return strspn(dirent->d_name, "0123456789ABCDEF") == strlen(dirent->d_name);
+}
+
+static void
+upgrade_file(const char *src_dir, const char *src_file, const char *dst_dir)
+{
+ char src[MAXPGPATH];
+ char dst[MAXPGPATH];
+
+ int seg_name_len;
+ int src_segno;
+ int64 src_pageno;
+ int dst_segno;
+ int64 dst_pageno;
+ int dst_offset;
+
+ int src_fd;
+ int dst_fd;
+
+ char *src_buf;
+ ssize_t src_len;
+ ssize_t src_buf_offset;
+ PGAlignedBlock dst_block;
+ Page page = dst_block.data;
+ int len_to_copy;
+
+ seg_name_len = strlen(src_file);
+ src_segno = (int) strtol(src_file, NULL, 16);
+ src_pageno = src_segno * SLRU_PAGES_PER_SEGMENT;
+
+ dst_pageno = src_pageno * BLCKSZ / SizeOfPageContents;
+ dst_offset = src_pageno * BLCKSZ - dst_pageno * SizeOfPageContents;
+ dst_segno = dst_pageno / SLRU_PAGES_PER_SEGMENT;
+
+ snprintf(src, sizeof(src), "%s/%s", src_dir, src_file);
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+
+ src_buf = pg_malloc(SEGMENT_SIZE);
+ if ((src_fd = open(src, O_RDONLY | PG_BINARY, 0)) == -1)
+ pg_fatal("could not open file \"%s\": %s", src, strerror(errno));
+ if ((src_len = read(src_fd, src_buf, SEGMENT_SIZE)) == -1)
+ pg_fatal("could not read file \"%s\": %s", src, strerror(errno));
+
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+
+ /*
+ * Read the destination page at dst_pageno into the buffer. The page may contain
+ * data from the previous source segment. Initialize the page if the page is new.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+ if (read(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not read file \"%s\": %s", dst, strerror(errno));
+ if (PageIsNew(page))
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Rewind the file position, so the first write will overwrite the page.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+
+ src_buf_offset = 0;
+ while (src_buf_offset < src_len)
+ {
+ len_to_copy = Min(src_len - src_buf_offset, SizeOfPageContents - dst_offset);
+ memcpy(PageGetContents(page) + dst_offset, src_buf + src_buf_offset, len_to_copy);
+ src_buf_offset += len_to_copy;
+
+ if (new_cluster.controldata.data_checksum_version > 0)
+ ((PageHeader) page)->pd_checksum = pg_checksum_page(page, dst_pageno);
+ if (write(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not write file \"%s\": %s", dst, strerror(errno));
+
+ dst_pageno++;
+ dst_offset = 0;
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Switch segments if we reached the end of the current segment.
+ */
+ if (dst_pageno % SLRU_PAGES_PER_SEGMENT == 0)
+ {
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ dst_segno++;
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+ }
+ }
+
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ pg_free(src_buf);
+ close(src_fd);
+}
+
+void
+upgrade_xact_cache(const char *src_subdir, const char *dst_subdir)
+{
+ char src_dir[MAXPGPATH];
+ char dst_dir[MAXPGPATH];
+
+ DIR *src_dirp;
+ struct dirent *src_dirent;
+
+ snprintf(src_dir, sizeof(src_dir), "%s/%s", old_cluster.pgdata, src_subdir);
+ snprintf(dst_dir, sizeof(dst_dir), "%s/%s", new_cluster.pgdata, dst_subdir);
+
+ if ((src_dirp = opendir(src_dir)) == NULL)
+ pg_fatal("could not open directory \"%s\": %s", src_dir, strerror(errno));
+
+ while (errno = 0, (src_dirent = readdir(src_dirp)) != NULL)
+ {
+ if (segment_file_filter(src_dirent))
+ upgrade_file(src_dir, src_dirent->d_name, dst_dir);
+ }
+
+ if (closedir(src_dirp) != 0)
+ pg_fatal("could not close directory \"%s\": %s", src_dir, strerror(errno));
+}
diff --git a/src/bin/pg_upgrade/pg_upgrade.c b/src/bin/pg_upgrade/pg_upgrade.c
index 14a36f0503..f630c5b04a 100644
--- a/src/bin/pg_upgrade/pg_upgrade.c
+++ b/src/bin/pg_upgrade/pg_upgrade.c
@@ -660,14 +660,23 @@ copy_subdir_files(const char *old_subdir, const char *new_subdir)
static void
copy_xact_xlog_xid(void)
{
+ bool slru_header_changed = false;
+
/*
* Copy old commit logs to new data dir. pg_clog has been renamed to
* pg_xact in post-10 clusters.
*/
- copy_subdir_files(GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact",
- GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact");
+ char *xact_old_dir = GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+ char *xact_new_dir = GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+
+ if (new_cluster.controldata.cat_ver >= SLRU_PAGE_HEADER_CAT_VER &&
+ old_cluster.controldata.cat_ver < SLRU_PAGE_HEADER_CAT_VER)
+ slru_header_changed = true;
+
+ if (slru_header_changed)
+ upgrade_xact_cache(xact_old_dir, xact_new_dir);
+ else
+ copy_subdir_files(xact_old_dir, xact_new_dir);
prep_status("Setting oldest XID for new cluster");
exec_prog(UTILITY_LOG_FILE, NULL, true, true,
@@ -702,7 +711,8 @@ copy_xact_xlog_xid(void)
* server doesn't attempt to read multis older than the cutoff value.
*/
if (old_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
- new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
+ !slru_header_changed)
{
copy_subdir_files("pg_multixact/offsets", "pg_multixact/offsets");
copy_subdir_files("pg_multixact/members", "pg_multixact/members");
@@ -722,7 +732,8 @@ copy_xact_xlog_xid(void)
new_cluster.pgdata);
check_ok();
}
- else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER ||
+ slru_header_changed)
{
/*
* Remove offsets/0000 file created by initdb that no longer matches
diff --git a/src/bin/pg_upgrade/pg_upgrade.h b/src/bin/pg_upgrade/pg_upgrade.h
index a1d08c3dab..edec949378 100644
--- a/src/bin/pg_upgrade/pg_upgrade.h
+++ b/src/bin/pg_upgrade/pg_upgrade.h
@@ -115,6 +115,11 @@ extern char *output_files[];
*/
#define MULTIXACT_FORMATCHANGE_CAT_VER 201301231
+/*
+ * A page header was added to each SLRU page in 17.0.
+ */
+#define SLRU_PAGE_HEADER_CAT_VER 202312091
+
/*
* large object chunk size added to pg_controldata,
* commit 5f93c37805e7485488480916b4585e098d3cc883
@@ -406,6 +411,7 @@ void rewriteVisibilityMap(const char *fromfile, const char *tofile,
const char *schemaName, const char *relName);
void check_file_clone(void);
void check_hard_link(void);
+void upgrade_xact_cache(const char *src_subdir, const char *dst_subdir);
/* fopen_priv() is no longer different from fopen() */
#define fopen_priv(path, mode) fopen(path, mode)
diff --git a/src/include/access/slru.h b/src/include/access/slru.h
index b05f6bc71d..4684389fda 100644
--- a/src/include/access/slru.h
+++ b/src/include/access/slru.h
@@ -68,17 +68,6 @@ typedef struct SlruSharedData
int *page_lru_count;
LWLockPadded *buffer_locks;
- /*
- * Optional array of WAL flush LSNs associated with entries in the SLRU
- * pages. If not zero/NULL, we must flush WAL before writing pages (true
- * for pg_xact, false for multixact, pg_subtrans, pg_notify). group_lsn[]
- * has lsn_groups_per_page entries per buffer slot, each containing the
- * highest LSN known for a contiguous group of SLRU entries on that slot's
- * page.
- */
- XLogRecPtr *group_lsn;
- int lsn_groups_per_page;
-
/*----------
* We mark a page "most recently used" by setting
* page_lru_count[slotno] = ++cur_lru_count;
@@ -147,8 +136,8 @@ typedef struct SlruCtlData
typedef SlruCtlData *SlruCtl;
-extern Size SimpleLruShmemSize(int nslots, int nlsns);
-extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+extern Size SimpleLruShmemSize(int nslots);
+extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
LWLock *ctllock, const char *subdir, int tranche_id,
SyncRequestHandler sync_handler,
bool long_segment_names);
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index 460d80ac97..a06f2f0776 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
*/
/* yyyymmddN */
-#define CATALOG_VERSION_NO 202401131
+#define CATALOG_VERSION_NO 202401161
#endif
diff --git a/src/include/storage/bufpage.h b/src/include/storage/bufpage.h
index d0df02d39c..2dc83451a6 100644
--- a/src/include/storage/bufpage.h
+++ b/src/include/storage/bufpage.h
@@ -201,6 +201,7 @@ typedef PageHeaderData *PageHeader;
* handling pages.
*/
#define PG_PAGE_LAYOUT_VERSION 4
+#define PG_SLRU_PAGE_LAYOUT_VERSION 1
#define PG_DATA_CHECKSUM_VERSION 1
/* ----------------------------------------------------------------
@@ -257,6 +258,11 @@ PageGetContents(Page page)
return (char *) page + MAXALIGN(SizeOfPageHeaderData);
}
+/*
+ * Space available for storing page contents.
+ */
+#define SizeOfPageContents (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
+
/* ----------------
* functions to access page size info
* ----------------
@@ -486,6 +492,7 @@ StaticAssertDecl(BLCKSZ == ((BLCKSZ / sizeof(size_t)) * sizeof(size_t)),
"BLCKSZ has to be a multiple of sizeof(size_t)");
extern void PageInit(Page page, Size pageSize, Size specialSize);
+extern void PageInitSLRU(Page page, Size pageSize, Size specialSize);
extern bool PageIsVerifiedExtended(Page page, BlockNumber blkno, int flags);
extern OffsetNumber PageAddItemExtended(Page page, Item item, Size size,
OffsetNumber offsetNumber, int flags);
diff --git a/src/test/modules/test_slru/test_slru.c b/src/test/modules/test_slru/test_slru.c
index 4b31f331ca..4e6ca07b6a 100644
--- a/src/test/modules/test_slru/test_slru.c
+++ b/src/test/modules/test_slru/test_slru.c
@@ -17,6 +17,7 @@
#include "access/slru.h"
#include "access/transam.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/shmem.h"
@@ -76,8 +77,8 @@ test_slru_page_write(PG_FUNCTION_ARGS)
TestSlruCtl->shared->page_status[slotno] = SLRU_PAGE_VALID;
/* write given data to the page, up to the limit of the page */
- strncpy(TestSlruCtl->shared->page_buffer[slotno], data,
- BLCKSZ - 1);
+ strncpy(PageGetContents(TestSlruCtl->shared->page_buffer[slotno]), data,
+ SizeOfPageContents - 1);
SimpleLruWritePage(TestSlruCtl, slotno);
LWLockRelease(TestSLRULock);
@@ -104,7 +105,7 @@ test_slru_page_read(PG_FUNCTION_ARGS)
LWLockAcquire(TestSLRULock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(TestSlruCtl, pageno,
write_ok, InvalidTransactionId);
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -122,7 +123,7 @@ test_slru_page_readonly(PG_FUNCTION_ARGS)
pageno,
InvalidTransactionId);
Assert(LWLockHeldByMe(TestSLRULock));
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(TestSLRULock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -202,7 +203,7 @@ test_slru_shmem_request(void)
prev_shmem_request_hook();
/* reserve shared memory for the test SLRU */
- RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS, 0));
+ RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS));
}
static bool
@@ -238,7 +239,7 @@ test_slru_shmem_startup(void)
TestSlruCtl->PagePrecedes = test_slru_page_precedes_logically;
SimpleLruInit(TestSlruCtl, "TestSLRU",
- NUM_TEST_BUFFERS, 0, TestSLRULock, slru_dir_name,
+ NUM_TEST_BUFFERS, TestSLRULock, slru_dir_name,
test_tranche_id, SYNC_HANDLER_NONE, long_segment_names);
}
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-03-06 12:01 ` Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2024-03-06 12:01 UTC (permalink / raw)
To: Aleksander Alekseev <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
Rebase the patch against the latest HEAD.
Regards,
Yong
Attachments:
[application/octet-stream] slru_page_header_v5.patch (55.4K, ../../[email protected]/3-slru_page_header_v5.patch)
download | inline diff:
src/backend/access/transam/clog.c | 52 +++++-----
src/backend/access/transam/commit_ts.c | 26 +++--
src/backend/access/transam/multixact.c | 63 +++++++-----
src/backend/access/transam/slru.c | 112 +++++++--------------
src/backend/access/transam/subtrans.c | 12 +--
src/backend/commands/async.c | 27 +++---
src/backend/storage/lmgr/predicate.c | 16 +--
src/backend/storage/page/bufpage.c | 25 +++++
src/bin/pg_checksums/pg_checksums.c | 9 ++
src/bin/pg_resetwal/t/001_basic.pl | 6 +-
src/bin/pg_upgrade/file.c | 172 +++++++++++++++++++++++++++++++++
src/bin/pg_upgrade/pg_upgrade.c | 23 +++--
src/bin/pg_upgrade/pg_upgrade.h | 6 ++
src/include/access/slru.h | 15 +--
src/include/catalog/catversion.h | 2 +-
src/include/storage/bufpage.h | 7 ++
src/test/modules/test_slru/test_slru.c | 13 +--
17 files changed, 395 insertions(+), 191 deletions(-)
diff --git a/src/backend/access/transam/clog.c b/src/backend/access/transam/clog.c
index 44c253246b..4ed4c82f05 100644
--- a/src/backend/access/transam/clog.c
+++ b/src/backend/access/transam/clog.c
@@ -42,6 +42,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/proc.h"
#include "storage/sync.h"
#include "utils/guc_hooks.h"
@@ -61,7 +62,7 @@
/* We need two bits per xact, so four xacts fit in a byte */
#define CLOG_BITS_PER_XACT 2
#define CLOG_XACTS_PER_BYTE 4
-#define CLOG_XACTS_PER_PAGE (BLCKSZ * CLOG_XACTS_PER_BYTE)
+#define CLOG_XACTS_PER_PAGE (SizeOfPageContents * CLOG_XACTS_PER_BYTE)
#define CLOG_XACT_BITMASK ((1 << CLOG_BITS_PER_XACT) - 1)
/*
@@ -88,13 +89,6 @@ TransactionIdToPage(TransactionId xid)
#define TransactionIdToByte(xid) (TransactionIdToPgIndex(xid) / CLOG_XACTS_PER_BYTE)
#define TransactionIdToBIndex(xid) ((xid) % (TransactionId) CLOG_XACTS_PER_BYTE)
-/* We store the latest async LSN for each group of transactions */
-#define CLOG_XACTS_PER_LSN_GROUP 32 /* keep this a power of 2 */
-#define CLOG_LSNS_PER_PAGE (CLOG_XACTS_PER_PAGE / CLOG_XACTS_PER_LSN_GROUP)
-
-#define GetLSNIndex(slotno, xid) ((slotno) * CLOG_LSNS_PER_PAGE + \
- ((xid) % (TransactionId) CLOG_XACTS_PER_PAGE) / CLOG_XACTS_PER_LSN_GROUP)
-
/*
* The number of subtransactions below which we consider to apply clog group
* update optimization. Testing reveals that the number higher than this can
@@ -112,7 +106,7 @@ static SlruCtlData XactCtlData;
static int ZeroCLOGPage(int64 pageno, bool writeXlog);
static bool CLOGPagePrecedes(int64 page1, int64 page2);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXact,
Oid oldestXactDb);
static void TransactionIdSetPageStatus(TransactionId xid, int nsubxids,
@@ -670,8 +664,9 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
Assert(LWLockHeldByMeInMode(SimpleLruGetBankLock(XactCtl,
XactCtl->shared->page_number[slotno]),
LW_EXCLUSIVE));
+ Page page = XactCtl->shared->page_buffer[slotno];
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(page) + byteno;
curval = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
/*
@@ -700,7 +695,7 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*byteptr = byteval;
/*
- * Update the group LSN if the transaction completion LSN is higher.
+ * Update the page LSN if the transaction completion LSN is higher.
*
* Note: lsn will be invalid when supplied during InRecovery processing,
* so we don't need to do anything special to avoid LSN updates during
@@ -709,10 +704,8 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*/
if (!XLogRecPtrIsInvalid(lsn))
{
- int lsnindex = GetLSNIndex(slotno, xid);
-
- if (XactCtl->shared->group_lsn[lsnindex] < lsn)
- XactCtl->shared->group_lsn[lsnindex] = lsn;
+ if (PageGetLSN(page) < lsn)
+ PageSetLSN(page, lsn);
}
}
@@ -738,19 +731,19 @@ TransactionIdGetStatus(TransactionId xid, XLogRecPtr *lsn)
int byteno = TransactionIdToByte(xid);
int bshift = TransactionIdToBIndex(xid) * CLOG_BITS_PER_XACT;
int slotno;
- int lsnindex;
+ Page page;
char *byteptr;
XidStatus status;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(XactCtl, pageno, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
status = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
- lsnindex = GetLSNIndex(slotno, xid);
- *lsn = XactCtl->shared->group_lsn[lsnindex];
+ *lsn = PageGetLSN(page);
LWLockRelease(SimpleLruGetBankLock(XactCtl, pageno));
@@ -780,7 +773,7 @@ CLOGShmemBuffers(void)
Size
CLOGShmemSize(void)
{
- return SimpleLruShmemSize(CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE);
+ return SimpleLruShmemSize(CLOGShmemBuffers());
}
void
@@ -808,7 +801,7 @@ CLOGShmemInit(void)
Assert(transaction_buffers != 0);
XactCtl->PagePrecedes = CLOGPagePrecedes;
- SimpleLruInit(XactCtl, "transaction", CLOGShmemBuffers(), CLOG_LSNS_PER_PAGE,
+ SimpleLruInit(XactCtl, "transaction", CLOGShmemBuffers(),
"pg_xact", LWTRANCHE_XACT_BUFFER,
LWTRANCHE_XACT_SLRU, SYNC_HANDLER_CLOG, false);
SlruPagePrecedesUnitTests(XactCtl, CLOG_XACTS_PER_PAGE);
@@ -860,11 +853,17 @@ static int
ZeroCLOGPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(XactCtl, pageno);
+ page = XactCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -917,12 +916,12 @@ TrimCLOG(void)
char *byteptr;
slotno = SimpleLruReadPage(XactCtl, pageno, false, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(XactCtl->shared->page_buffer[slotno]) + byteno;
/* Zero so-far-unused positions in the current byte */
*byteptr &= (1 << bshift) - 1;
/* Zero the rest of the page */
- MemSet(byteptr + 1, 0, BLCKSZ - byteno - 1);
+ MemSet(byteptr + 1, 0, SizeOfPageContents - byteno - 1);
XactCtl->shared->page_dirty[slotno] = true;
}
@@ -946,7 +945,6 @@ CheckPointCLOG(void)
TRACE_POSTGRESQL_CLOG_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that CLOG has room for a newly-allocated XID.
*
@@ -1070,12 +1068,12 @@ CLOGPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
+ return XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/commit_ts.c b/src/backend/access/transam/commit_ts.c
index f221494687..dce06d6cf6 100644
--- a/src/backend/access/transam/commit_ts.c
+++ b/src/backend/access/transam/commit_ts.c
@@ -29,6 +29,7 @@
#include "access/xlogutils.h"
#include "funcapi.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/shmem.h"
#include "utils/fmgrprotos.h"
#include "utils/guc_hooks.h"
@@ -61,7 +62,7 @@ typedef struct CommitTimestampEntry
sizeof(RepOriginId))
#define COMMIT_TS_XACTS_PER_PAGE \
- (BLCKSZ / SizeOfCommitTimestampEntry)
+ (SizeOfPageContents / SizeOfCommitTimestampEntry)
/*
@@ -118,7 +119,7 @@ static int ZeroCommitTsPage(int64 pageno, bool writeXlog);
static bool CommitTsPagePrecedes(int64 page1, int64 page2);
static void ActivateCommitTs(void);
static void DeactivateCommitTs(void);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXid);
/*
@@ -253,11 +254,12 @@ TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
+ Assert(xid == slotno * COMMIT_TS_XACTS_PER_PAGE + entryno);
entry.time = ts;
entry.nodeid = nodeid;
- memcpy(CommitTsCtl->shared->page_buffer[slotno] +
+ memcpy(PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
@@ -336,7 +338,7 @@ TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
- CommitTsCtl->shared->page_buffer[slotno] +
+ PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
@@ -518,7 +520,7 @@ CommitTsShmemBuffers(void)
Size
CommitTsShmemSize(void)
{
- return SimpleLruShmemSize(CommitTsShmemBuffers(), 0) +
+ return SimpleLruShmemSize(CommitTsShmemBuffers()) +
sizeof(CommitTimestampShared);
}
@@ -553,7 +555,7 @@ CommitTsShmemInit(void)
Assert(commit_timestamp_buffers != 0);
CommitTsCtl->PagePrecedes = CommitTsPagePrecedes;
- SimpleLruInit(CommitTsCtl, "commit_timestamp", CommitTsShmemBuffers(), 0,
+ SimpleLruInit(CommitTsCtl, "commit_timestamp", CommitTsShmemBuffers(),
"pg_commit_ts", LWTRANCHE_COMMITTS_BUFFER,
LWTRANCHE_COMMITTS_SLRU,
SYNC_HANDLER_COMMIT_TS,
@@ -615,11 +617,17 @@ static int
ZeroCommitTsPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(CommitTsCtl, pageno);
+ page = CommitTsCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -985,12 +993,12 @@ CommitTsPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
+ return XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c
index 83b578dced..8d934c7bb3 100644
--- a/src/backend/access/transam/multixact.c
+++ b/src/backend/access/transam/multixact.c
@@ -83,6 +83,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "postmaster/autovacuum.h"
+#include "storage/bufpage.h"
#include "storage/pmsignal.h"
#include "storage/proc.h"
#include "storage/procarray.h"
@@ -105,7 +106,7 @@
*/
/* We need four bytes per offset */
-#define MULTIXACT_OFFSETS_PER_PAGE (BLCKSZ / sizeof(MultiXactOffset))
+#define MULTIXACT_OFFSETS_PER_PAGE (SizeOfPageContents / sizeof(MultiXactOffset))
#define MultiXactIdToOffsetPage(xid) \
((xid) / (MultiXactOffset) MULTIXACT_OFFSETS_PER_PAGE)
@@ -118,8 +119,8 @@
* additional flag bits for each TransactionId. To do this without getting
* into alignment issues, we store four bytes of flags, and then the
* corresponding 4 Xids. Each such 5-word (20-byte) set we call a "group", and
- * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 409 groups
- * per page. This wastes 12 bytes per page, but that's OK -- simplicity (and
+ * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 408 groups
+ * per page. This wastes 8 bytes per page, but that's OK -- simplicity (and
* performance) trumps space efficiency here.
*
* Note that the "offset" macros work with byte offset, not array indexes, so
@@ -137,7 +138,7 @@
/* size in bytes of a complete group */
#define MULTIXACT_MEMBERGROUP_SIZE \
(sizeof(TransactionId) * MULTIXACT_MEMBERS_PER_MEMBERGROUP + MULTIXACT_FLAGBYTES_PER_GROUP)
-#define MULTIXACT_MEMBERGROUPS_PER_PAGE (BLCKSZ / MULTIXACT_MEMBERGROUP_SIZE)
+#define MULTIXACT_MEMBERGROUPS_PER_PAGE (SizeOfPageContents / MULTIXACT_MEMBERGROUP_SIZE)
#define MULTIXACT_MEMBERS_PER_PAGE \
(MULTIXACT_MEMBERGROUPS_PER_PAGE * MULTIXACT_MEMBERS_PER_MEMBERGROUP)
@@ -364,7 +365,7 @@ static bool MultiXactOffsetWouldWrap(MultiXactOffset boundary,
MultiXactOffset start, uint32 distance);
static bool SetOffsetVacuumLimit(bool is_startup);
static bool find_multixact_start(MultiXactId multi, MultiXactOffset *result);
-static void WriteMZeroPageXlogRec(int64 pageno, uint8 info);
+static XLogRecPtr WriteMZeroPageXlogRec(int64 pageno, uint8 info);
static void WriteMTruncateXlogRec(Oid oldestMultiDB,
MultiXactId startTruncOff,
MultiXactId endTruncOff,
@@ -885,7 +886,7 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
* take the trouble to generalize the slru.c error reporting code.
*/
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
*offptr = offset;
@@ -934,12 +935,12 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
}
memberptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
*memberptr = members[i].xid;
flagsptr = (uint32 *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
flagsval = *flagsptr;
flagsval &= ~(((1 << MXACT_MEMBER_BITS_PER_XACT) - 1) << bshift);
@@ -1364,7 +1365,7 @@ retry:
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
@@ -1413,7 +1414,7 @@ retry:
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, tmpMXact);
}
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
nextMXOffset = *offptr;
@@ -1470,7 +1471,7 @@ retry:
}
xactptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
if (!TransactionIdIsValid(*xactptr))
{
@@ -1481,7 +1482,7 @@ retry:
flagsoff = MXOffsetToFlagsOffset(offset);
bshift = MXOffsetToFlagsBitShift(offset);
- flagsptr = (uint32 *) (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ flagsptr = (uint32 *) (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
ptr[truelength].xid = *xactptr;
ptr[truelength].status = (*flagsptr >> bshift) & MXACT_MEMBER_XACT_BITMASK;
@@ -1880,8 +1881,8 @@ MultiXactShmemSize(void)
mul_size(sizeof(MultiXactId) * 2, MaxOldestSlot))
size = SHARED_MULTIXACT_STATE_SIZE;
- size = add_size(size, SimpleLruShmemSize(multixact_offset_buffers, 0));
- size = add_size(size, SimpleLruShmemSize(multixact_member_buffers, 0));
+ size = add_size(size, SimpleLruShmemSize(multixact_offset_buffers));
+ size = add_size(size, SimpleLruShmemSize(multixact_member_buffers));
return size;
}
@@ -1897,14 +1898,14 @@ MultiXactShmemInit(void)
MultiXactMemberCtl->PagePrecedes = MultiXactMemberPagePrecedes;
SimpleLruInit(MultiXactOffsetCtl,
- "multixact_offset", multixact_offset_buffers, 0,
+ "multixact_offset", multixact_offset_buffers,
"pg_multixact/offsets", LWTRANCHE_MULTIXACTOFFSET_BUFFER,
LWTRANCHE_MULTIXACTOFFSET_SLRU,
SYNC_HANDLER_MULTIXACT_OFFSET,
false);
SlruPagePrecedesUnitTests(MultiXactOffsetCtl, MULTIXACT_OFFSETS_PER_PAGE);
SimpleLruInit(MultiXactMemberCtl,
- "multixact_member", multixact_member_buffers, 0,
+ "multixact_member", multixact_member_buffers,
"pg_multixact/members", LWTRANCHE_MULTIXACTMEMBER_BUFFER,
LWTRANCHE_MULTIXACTMEMBER_SLRU,
SYNC_HANDLER_MULTIXACT_MEMBER,
@@ -1999,11 +2000,17 @@ static int
ZeroMultiXactOffsetPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactOffsetCtl, pageno);
+ page = MultiXactOffsetCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2015,11 +2022,17 @@ static int
ZeroMultiXactMemberPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactMemberCtl, pageno);
+ page = MultiXactMemberCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2143,10 +2156,10 @@ TrimMultiXact(void)
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, nextMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
- MemSet(offptr, 0, BLCKSZ - (entryno * sizeof(MultiXactOffset)));
+ MemSet(offptr, 0, SizeOfPageContents - (entryno * sizeof(MultiXactOffset)));
MultiXactOffsetCtl->shared->page_dirty[slotno] = true;
LWLockRelease(lock);
@@ -2177,9 +2190,9 @@ TrimMultiXact(void)
memberoff = MXOffsetToMemberOffset(offset);
slotno = SimpleLruReadPage(MultiXactMemberCtl, pageno, true, offset);
xidptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
- MemSet(xidptr, 0, BLCKSZ - memberoff);
+ MemSet(xidptr, 0, SizeOfPageContents - memberoff);
/*
* Note: we don't need to zero out the flag bits in the remaining
@@ -2834,7 +2847,7 @@ find_multixact_start(MultiXactId multi, MultiXactOffset *result)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(MultiXactOffsetCtl, pageno, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
LWLockRelease(SimpleLruGetBankLock(MultiXactOffsetCtl, pageno));
@@ -3268,12 +3281,12 @@ MultiXactOffsetPrecedes(MultiXactOffset offset1, MultiXactOffset offset2)
* Write an xlog record reflecting the zeroing of either a MEMBERs or
* OFFSETs page (info shows which)
*/
-static void
+static XLogRecPtr
WriteMZeroPageXlogRec(int64 pageno, uint8 info)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_MULTIXACT_ID, info);
+ return XLogInsert(RM_MULTIXACT_ID, info);
}
/*
diff --git a/src/backend/access/transam/slru.c b/src/backend/access/transam/slru.c
index f99ec38a4a..aded4d9708 100644
--- a/src/backend/access/transam/slru.c
+++ b/src/backend/access/transam/slru.c
@@ -68,6 +68,7 @@
#include "access/xlogutils.h"
#include "miscadmin.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/shmem.h"
#include "utils/guc_hooks.h"
@@ -155,13 +156,13 @@ typedef enum
SLRU_WRITE_FAILED,
SLRU_FSYNC_FAILED,
SLRU_CLOSE_FAILED,
+ SLRU_DATA_CORRUPTED,
} SlruErrorCause;
static SlruErrorCause slru_errcause;
static int slru_errno;
-static void SimpleLruZeroLSNs(SlruCtl ctl, int slotno);
static void SimpleLruWaitIO(SlruCtl ctl, int slotno);
static void SlruInternalWritePage(SlruCtl ctl, int slotno, SlruWriteAll fdata);
static bool SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno);
@@ -181,7 +182,7 @@ static inline void SlruRecentlyUsed(SlruShared shared, int slotno);
*/
Size
-SimpleLruShmemSize(int nslots, int nlsns)
+SimpleLruShmemSize(int nslots)
{
int nbanks = nslots / SLRU_BANK_SIZE;
Size sz;
@@ -200,9 +201,6 @@ SimpleLruShmemSize(int nslots, int nlsns)
sz += MAXALIGN(nbanks * sizeof(LWLockPadded)); /* bank_locks[] */
sz += MAXALIGN(nbanks * sizeof(int)); /* bank_cur_lru_count[] */
- if (nlsns > 0)
- sz += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr)); /* group_lsn[] */
-
return BUFFERALIGN(sz) + BLCKSZ * nslots;
}
@@ -227,7 +225,6 @@ SimpleLruAutotuneBuffers(int divisor, int max)
* ctl: address of local (unshared) control structure.
* name: name of SLRU. (This is user-visible, pick with care!)
* nslots: number of page slots to use.
- * nlsns: number of LSN groups per page (set to zero if not relevant).
* ctllock: LWLock to use to control access to the shared control structure.
* subdir: PGDATA-relative subdirectory that will contain the files.
* buffer_tranche_id: tranche ID to use for the SLRU's per-buffer LWLocks.
@@ -235,7 +232,7 @@ SimpleLruAutotuneBuffers(int divisor, int max)
* sync_handler: which set of functions to use to handle sync requests
*/
void
-SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
const char *subdir, int buffer_tranche_id, int bank_tranche_id,
SyncRequestHandler sync_handler, bool long_segment_names)
{
@@ -246,7 +243,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
Assert(nslots <= SLRU_MAX_ALLOWED_BUFFERS);
shared = (SlruShared) ShmemInitStruct(name,
- SimpleLruShmemSize(nslots, nlsns),
+ SimpleLruShmemSize(nslots),
&found);
if (!IsUnderPostmaster)
@@ -260,7 +257,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
memset(shared, 0, sizeof(SlruSharedData));
shared->num_slots = nslots;
- shared->lsn_groups_per_page = nlsns;
pg_atomic_init_u64(&shared->latest_page_number, 0);
@@ -287,12 +283,6 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
shared->bank_cur_lru_count = (int *) (ptr + offset);
offset += MAXALIGN(nbanks * sizeof(int));
- if (nlsns > 0)
- {
- shared->group_lsn = (XLogRecPtr *) (ptr + offset);
- offset += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr));
- }
-
ptr += BUFFERALIGN(offset);
for (int slotno = 0; slotno < nslots; slotno++)
{
@@ -314,7 +304,7 @@ SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
}
/* Should fit to estimated shmem size */
- Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots, nlsns));
+ Assert(ptr - (char *) shared <= SimpleLruShmemSize(nslots));
}
else
{
@@ -378,11 +368,8 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
shared->page_dirty[slotno] = true;
SlruRecentlyUsed(shared, slotno);
- /* Set the buffer to zeroes */
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
-
- /* Set the LSNs for this new page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
+ /* Initialize the page. */
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
/*
* Assume this page is now the latest active page.
@@ -400,26 +387,6 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
return slotno;
}
-/*
- * Zero all the LSNs we store for this slru page.
- *
- * This should be called each time we create a new page, and each time we read
- * in a page from disk into an existing buffer. (Such an old page cannot
- * have any interesting LSNs, since we'd have flushed them before writing
- * the page in the first place.)
- *
- * This assumes that InvalidXLogRecPtr is bitwise-all-0.
- */
-static void
-SimpleLruZeroLSNs(SlruCtl ctl, int slotno)
-{
- SlruShared shared = ctl->shared;
-
- if (shared->lsn_groups_per_page > 0)
- MemSet(&shared->group_lsn[slotno * shared->lsn_groups_per_page], 0,
- shared->lsn_groups_per_page * sizeof(XLogRecPtr));
-}
-
/*
* Wait for any active I/O on a page slot to finish. (This does not
* guarantee that new I/O hasn't been started before we return, though.
@@ -546,9 +513,6 @@ SimpleLruReadPage(SlruCtl ctl, int64 pageno, bool write_ok,
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
- /* Set the LSNs for this newly read-in page to zero */
- SimpleLruZeroLSNs(ctl, slotno);
-
/* Re-acquire bank control lock and update page state */
LWLockAcquire(banklock, LW_EXCLUSIVE);
@@ -815,7 +779,7 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
ereport(LOG,
(errmsg("file \"%s\" doesn't exist, reading as zeroes",
path)));
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
return true;
}
@@ -838,6 +802,13 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
return false;
}
+ if (!PageIsVerifiedExtended(shared->page_buffer[slotno], pageno, PIV_REPORT_STAT))
+ {
+ slru_errcause = SLRU_DATA_CORRUPTED;
+ slru_errno = 0;
+ return false;
+ }
+
return true;
}
@@ -864,6 +835,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
off_t offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd = -1;
+ Page page = shared->page_buffer[slotno];
+ XLogRecPtr lsn;
/* update the stats counter of written pages */
pgstat_count_slru_page_written(shared->slru_stats_idx);
@@ -873,40 +846,18 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
* write out data before associated WAL records. This is the same action
* performed during FlushBuffer() in the main buffer manager.
*/
- if (shared->group_lsn != NULL)
+ lsn = PageGetLSN(page);
+ if (!XLogRecPtrIsInvalid(lsn))
{
/*
- * We must determine the largest async-commit LSN for the page. This
- * is a bit tedious, but since this entire function is a slow path
- * anyway, it seems better to do this here than to maintain a per-page
- * LSN variable (which'd need an extra comparison in the
- * transaction-commit path).
+ * As noted above, elog(ERROR) is not acceptable here, so if
+ * XLogFlush were to fail, we must PANIC. This isn't much of a
+ * restriction because XLogFlush is just about all critical
+ * section anyway, but let's make sure.
*/
- XLogRecPtr max_lsn;
- int lsnindex;
-
- lsnindex = slotno * shared->lsn_groups_per_page;
- max_lsn = shared->group_lsn[lsnindex++];
- for (int lsnoff = 1; lsnoff < shared->lsn_groups_per_page; lsnoff++)
- {
- XLogRecPtr this_lsn = shared->group_lsn[lsnindex++];
-
- if (max_lsn < this_lsn)
- max_lsn = this_lsn;
- }
-
- if (!XLogRecPtrIsInvalid(max_lsn))
- {
- /*
- * As noted above, elog(ERROR) is not acceptable here, so if
- * XLogFlush were to fail, we must PANIC. This isn't much of a
- * restriction because XLogFlush is just about all critical
- * section anyway, but let's make sure.
- */
- START_CRIT_SECTION();
- XLogFlush(max_lsn);
- END_CRIT_SECTION();
- }
+ START_CRIT_SECTION();
+ XLogFlush(lsn);
+ END_CRIT_SECTION();
}
/*
@@ -971,6 +922,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
}
}
+ PageSetChecksumInplace(shared->page_buffer[slotno], pageno);
+
errno = 0;
pgstat_report_wait_start(WAIT_EVENT_SLRU_WRITE);
if (pg_pwrite(fd, shared->page_buffer[slotno], BLCKSZ, offset) != BLCKSZ)
@@ -1091,6 +1044,13 @@ SlruReportIOError(SlruCtl ctl, int64 pageno, TransactionId xid)
errdetail("Could not close file \"%s\": %m.",
path)));
break;
+ case SLRU_DATA_CORRUPTED:
+ ereport(ERROR,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("could not access status of transaction %u", xid),
+ errdetail("Invalid page from file \"%s\" at offset %d.",
+ path, offset)));
+ break;
default:
/* can't get here, we trust */
elog(ERROR, "unrecognized SimpleLru error cause: %d",
diff --git a/src/backend/access/transam/subtrans.c b/src/backend/access/transam/subtrans.c
index 50bb1d8cfc..5549cac3a6 100644
--- a/src/backend/access/transam/subtrans.c
+++ b/src/backend/access/transam/subtrans.c
@@ -34,6 +34,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "utils/guc_hooks.h"
+#include "storage/bufpage.h"
#include "utils/snapmgr.h"
@@ -51,7 +52,7 @@
*/
/* We need four bytes per xact */
-#define SUBTRANS_XACTS_PER_PAGE (BLCKSZ / sizeof(TransactionId))
+#define SUBTRANS_XACTS_PER_PAGE (SizeOfPageContents / sizeof(TransactionId))
/*
* Although we return an int64 the actual value can't currently exceed
@@ -97,7 +98,7 @@ SubTransSetParent(TransactionId xid, TransactionId parent)
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(SubTransCtl, pageno, true, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
/*
@@ -137,7 +138,7 @@ SubTransGetParent(TransactionId xid)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
parent = *ptr;
@@ -213,7 +214,7 @@ SUBTRANSShmemBuffers(void)
Size
SUBTRANSShmemSize(void)
{
- return SimpleLruShmemSize(SUBTRANSShmemBuffers(), 0);
+ return SimpleLruShmemSize(SUBTRANSShmemBuffers());
}
void
@@ -241,7 +242,7 @@ SUBTRANSShmemInit(void)
Assert(subtransaction_buffers != 0);
SubTransCtl->PagePrecedes = SubTransPagePrecedes;
- SimpleLruInit(SubTransCtl, "subtransaction", SUBTRANSShmemBuffers(), 0,
+ SimpleLruInit(SubTransCtl, "subtransaction", SUBTRANSShmemBuffers(),
"pg_subtrans", LWTRANCHE_SUBTRANS_BUFFER,
LWTRANCHE_SUBTRANS_SLRU, SYNC_HANDLER_NONE, false);
SlruPagePrecedesUnitTests(SubTransCtl, SUBTRANS_XACTS_PER_PAGE);
@@ -366,7 +367,6 @@ CheckPointSUBTRANS(void)
TRACE_POSTGRESQL_SUBTRANS_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that SUBTRANS has room for a newly-allocated XID.
*
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c
index d0891e3f0e..a400c54c03 100644
--- a/src/backend/commands/async.c
+++ b/src/backend/commands/async.c
@@ -140,6 +140,7 @@
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/procsignal.h"
@@ -160,7 +161,7 @@
* than that, so changes in that data structure won't affect user-visible
* restrictions.
*/
-#define NOTIFY_PAYLOAD_MAX_LENGTH (BLCKSZ - NAMEDATALEN - 128)
+#define NOTIFY_PAYLOAD_MAX_LENGTH (SizeOfPageContents - NAMEDATALEN - 128)
/*
* Struct representing an entry in the global notify queue
@@ -309,7 +310,7 @@ static SlruCtlData NotifyCtlData;
#define NotifyCtl (&NotifyCtlData)
#define QUEUE_PAGESIZE BLCKSZ
-
+#define QUEUE_PAGE_CAPACITY (QUEUE_PAGESIZE - MAXALIGN(SizeOfPageHeaderData))
#define QUEUE_FULL_WARN_INTERVAL 5000 /* warn at most once every 5s */
/*
@@ -490,7 +491,7 @@ AsyncShmemSize(void)
size = mul_size(MaxBackends, sizeof(QueueBackendStatus));
size = add_size(size, offsetof(AsyncQueueControl, backend));
- size = add_size(size, SimpleLruShmemSize(notify_buffers, 0));
+ size = add_size(size, SimpleLruShmemSize(notify_buffers));
return size;
}
@@ -535,7 +536,7 @@ AsyncShmemInit(void)
* names are used in order to avoid wraparound.
*/
NotifyCtl->PagePrecedes = asyncQueuePagePrecedes;
- SimpleLruInit(NotifyCtl, "notify", notify_buffers, 0,
+ SimpleLruInit(NotifyCtl, "notify", notify_buffers,
"pg_notify", LWTRANCHE_NOTIFY_BUFFER, LWTRANCHE_NOTIFY_SLRU,
SYNC_HANDLER_NONE, true);
@@ -1295,14 +1296,14 @@ asyncQueueAdvance(volatile QueuePosition *position, int entryLength)
* written or read.
*/
offset += entryLength;
- Assert(offset <= QUEUE_PAGESIZE);
+ Assert(offset <= QUEUE_PAGE_CAPACITY);
/*
* In a second step check if another entry can possibly be written to the
* page. If so, stay here, we have reached the next position. If not, then
* we need to move on to the next page.
*/
- if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGESIZE)
+ if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGE_CAPACITY)
{
pageno++;
offset = 0;
@@ -1405,7 +1406,7 @@ asyncQueueAddEntries(ListCell *nextNotify)
offset = QUEUE_POS_OFFSET(queue_head);
/* Check whether the entry really fits on the current page */
- if (offset + qe.length <= QUEUE_PAGESIZE)
+ if (offset + qe.length <= QUEUE_PAGE_CAPACITY)
{
/* OK, so advance nextNotify past this item */
nextNotify = lnext(pendingNotifies->events, nextNotify);
@@ -1417,14 +1418,14 @@ asyncQueueAddEntries(ListCell *nextNotify)
* only check dboid and since it won't match any reader's database
* OID, they will ignore this entry and move on.
*/
- qe.length = QUEUE_PAGESIZE - offset;
+ qe.length = QUEUE_PAGE_CAPACITY - offset;
qe.dboid = InvalidOid;
qe.data[0] = '\0'; /* empty channel */
qe.data[1] = '\0'; /* empty payload */
}
/* Now copy qe into the shared buffer page */
- memcpy(NotifyCtl->shared->page_buffer[slotno] + offset,
+ memcpy(PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + offset,
&qe,
qe.length);
@@ -1955,10 +1956,10 @@ asyncQueueReadAllNotifications(void)
else
{
/* fetch all the rest of the page */
- copysize = QUEUE_PAGESIZE - curoffset;
+ copysize = QUEUE_PAGE_CAPACITY - curoffset;
}
- memcpy(page_buffer.buf + curoffset,
- NotifyCtl->shared->page_buffer[slotno] + curoffset,
+ memcpy(PageGetContents(page_buffer.buf) + curoffset,
+ PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + curoffset,
copysize);
/* Release lock that we got from SimpleLruReadPage_ReadOnly() */
LWLockRelease(SimpleLruGetBankLock(NotifyCtl, curpage));
@@ -2029,7 +2030,7 @@ asyncQueueProcessPageEntries(volatile QueuePosition *current,
if (QUEUE_POS_EQUAL(thisentry, stop))
break;
- qe = (AsyncQueueEntry *) (page_buffer + QUEUE_POS_OFFSET(thisentry));
+ qe = (AsyncQueueEntry *) (PageGetContents(page_buffer) + QUEUE_POS_OFFSET(thisentry));
/*
* Advance *current over this message, possibly to the next page. As
diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index 3f378c0099..c26ae8ddd2 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -207,6 +207,7 @@
#include "miscadmin.h"
#include "pgstat.h"
#include "port/pg_lfind.h"
+#include "storage/bufpage.h"
#include "storage/predicate.h"
#include "storage/predicate_internals.h"
#include "storage/proc.h"
@@ -326,8 +327,8 @@ static SlruCtlData SerialSlruCtlData;
#define SerialSlruCtl (&SerialSlruCtlData)
#define SERIAL_PAGESIZE BLCKSZ
-#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
-#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE / SERIAL_ENTRYSIZE)
+#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
+#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE - MAXALIGN(SizeOfPageHeaderData) / SERIAL_ENTRYSIZE)
/*
* Set maximum pages based on the number needed to track all transactions.
@@ -337,7 +338,7 @@ static SlruCtlData SerialSlruCtlData;
#define SerialNextPage(page) (((page) >= SERIAL_MAX_PAGE) ? 0 : (page) + 1)
#define SerialValue(slotno, xid) (*((SerCommitSeqNo *) \
- (SerialSlruCtl->shared->page_buffer[slotno] + \
+ (PageGetContents(SerialSlruCtl->shared->page_buffer[slotno]) + \
((((uint32) (xid)) % SERIAL_ENTRIESPERPAGE) * SERIAL_ENTRYSIZE))))
#define SerialPage(xid) (((uint32) (xid)) / SERIAL_ENTRIESPERPAGE)
@@ -789,10 +790,13 @@ SerialPagePrecedesLogicallyUnitTests(void)
* requires burning ~2B XIDs in single-user mode, a negligible
* possibility. Moreover, if it does happen, the consequence would be
* mild, namely a new transaction failing in SimpleLruReadPage().
+ *
+ * NOTE: After adding the page header, the defect affects two pages.
+ * We now assert correct treatment of its second to prior page.
*/
headPage = oldestPage;
targetPage = newestPage;
- Assert(SerialPagePrecedesLogically(headPage, targetPage - 1));
+ Assert(SerialPagePrecedesLogically(headPage, targetPage - 2));
#if 0
Assert(SerialPagePrecedesLogically(headPage, targetPage));
#endif
@@ -812,7 +816,7 @@ SerialInit(void)
*/
SerialSlruCtl->PagePrecedes = SerialPagePrecedesLogically;
SimpleLruInit(SerialSlruCtl, "serializable",
- serializable_buffers, 0, "pg_serial",
+ serializable_buffers, "pg_serial",
LWTRANCHE_SERIAL_BUFFER, LWTRANCHE_SERIAL_SLRU,
SYNC_HANDLER_NONE, false);
#ifdef USE_ASSERT_CHECKING
@@ -1377,7 +1381,7 @@ PredicateLockShmemSize(void)
/* Shared memory structures for SLRU tracking of old committed xids. */
size = add_size(size, sizeof(SerialControlData));
- size = add_size(size, SimpleLruShmemSize(serializable_buffers, 0));
+ size = add_size(size, SimpleLruShmemSize(serializable_buffers));
return size;
}
diff --git a/src/backend/storage/page/bufpage.c b/src/backend/storage/page/bufpage.c
index be6f1f62d2..e8193d7f56 100644
--- a/src/backend/storage/page/bufpage.c
+++ b/src/backend/storage/page/bufpage.c
@@ -59,6 +59,31 @@ PageInit(Page page, Size pageSize, Size specialSize)
/* p->pd_prune_xid = InvalidTransactionId; done by above MemSet */
}
+/*
+ * PageInitSLRU
+ * Initializes the contents of an SLRU page.
+ * Note that we don't calculate an initial checksum here; that's not done
+ * until it's time to write.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
/*
* PageIsVerifiedExtended
diff --git a/src/bin/pg_checksums/pg_checksums.c b/src/bin/pg_checksums/pg_checksums.c
index 9e6fd435f6..06f14f1d2d 100644
--- a/src/bin/pg_checksums/pg_checksums.c
+++ b/src/bin/pg_checksums/pg_checksums.c
@@ -16,6 +16,7 @@
#include <dirent.h>
#include <limits.h>
+#include <stdbool.h>
#include <time.h>
#include <sys/stat.h>
#include <unistd.h>
@@ -593,12 +594,20 @@ main(int argc, char *argv[])
{
total_size = scan_directory(DataDir, "global", true);
total_size += scan_directory(DataDir, "base", true);
+ total_size += scan_directory(DataDir, "pg_commit_ts", true);
+ total_size += scan_directory(DataDir, "pg_multixact", true);
+ total_size += scan_directory(DataDir, "pg_serial", true);
total_size += scan_directory(DataDir, "pg_tblspc", true);
+ total_size += scan_directory(DataDir, "pg_xact", true);
}
(void) scan_directory(DataDir, "global", false);
(void) scan_directory(DataDir, "base", false);
+ (void) scan_directory(DataDir, "pg_commit_ts", false);
+ (void) scan_directory(DataDir, "pg_multixact", false);
+ (void) scan_directory(DataDir, "pg_serial", false);
(void) scan_directory(DataDir, "pg_tblspc", false);
+ (void) scan_directory(DataDir, "pg_xact", false);
if (showprogress)
progress_report(true);
diff --git a/src/bin/pg_resetwal/t/001_basic.pl b/src/bin/pg_resetwal/t/001_basic.pl
index 9829e48106..7b9e034e19 100644
--- a/src/bin/pg_resetwal/t/001_basic.pl
+++ b/src/bin/pg_resetwal/t/001_basic.pl
@@ -206,7 +206,7 @@ push @cmd,
sprintf("%d,%d", hex($files[0]) == 0 ? 3 : hex($files[0]), hex($files[-1]));
@files = get_slru_files('pg_multixact/offsets');
-$mult = 32 * $blcksz / 4;
+$mult = 32 * ($blcksz - 24) / 4;
# -m argument is "new,old"
push @cmd, '-m',
sprintf("%d,%d",
@@ -214,11 +214,11 @@ push @cmd, '-m',
hex($files[0]) == 0 ? 1 : hex($files[0] * $mult));
@files = get_slru_files('pg_multixact/members');
-$mult = 32 * int($blcksz / 20) * 4;
+$mult = 32 * int(($blcksz - 24) / 20) * 4;
push @cmd, '-O', (hex($files[-1]) + 1) * $mult;
@files = get_slru_files('pg_xact');
-$mult = 32 * $blcksz * 4;
+$mult = 32 * ($blcksz - 24) * 4;
push @cmd,
'-u', (hex($files[0]) == 0 ? 3 : hex($files[0]) * $mult),
'-x', ((hex($files[-1]) + 1) * $mult);
diff --git a/src/bin/pg_upgrade/file.c b/src/bin/pg_upgrade/file.c
index beba376f2e..6978410f17 100644
--- a/src/bin/pg_upgrade/file.c
+++ b/src/bin/pg_upgrade/file.c
@@ -9,6 +9,7 @@
#include "postgres_fe.h"
+#include <dirent.h>
#include <sys/stat.h>
#include <limits.h>
#include <fcntl.h>
@@ -453,3 +454,174 @@ check_hard_link(void)
unlink(new_link_file);
}
+
+
+/*
+ * Copy SLRU_PAGES_PER_SEGMENT from access/slru.h to avoid including it.
+ */
+#define SLRU_PAGES_PER_SEGMENT 32
+
+#define SEGMENT_SIZE (BLCKSZ * SLRU_PAGES_PER_SEGMENT)
+
+/*
+ * Copy PageInitSLRU from storage/bufpage.c to avoid linking to the backend.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
+
+/*
+ * Filter function for scandir(3) to select only segment files.
+ */
+static int
+segment_file_filter(const struct dirent *dirent)
+{
+ return strspn(dirent->d_name, "0123456789ABCDEF") == strlen(dirent->d_name);
+}
+
+static void
+upgrade_file(const char *src_dir, const char *src_file, const char *dst_dir)
+{
+ char src[MAXPGPATH];
+ char dst[MAXPGPATH];
+
+ int seg_name_len;
+ int src_segno;
+ int64 src_pageno;
+ int dst_segno;
+ int64 dst_pageno;
+ int dst_offset;
+
+ int src_fd;
+ int dst_fd;
+
+ char *src_buf;
+ ssize_t src_len;
+ ssize_t src_buf_offset;
+ PGAlignedBlock dst_block;
+ Page page = dst_block.data;
+ int len_to_copy;
+
+ seg_name_len = strlen(src_file);
+ src_segno = (int) strtol(src_file, NULL, 16);
+ src_pageno = src_segno * SLRU_PAGES_PER_SEGMENT;
+
+ dst_pageno = src_pageno * BLCKSZ / SizeOfPageContents;
+ dst_offset = src_pageno * BLCKSZ - dst_pageno * SizeOfPageContents;
+ dst_segno = dst_pageno / SLRU_PAGES_PER_SEGMENT;
+
+ snprintf(src, sizeof(src), "%s/%s", src_dir, src_file);
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+
+ src_buf = pg_malloc(SEGMENT_SIZE);
+ if ((src_fd = open(src, O_RDONLY | PG_BINARY, 0)) == -1)
+ pg_fatal("could not open file \"%s\": %s", src, strerror(errno));
+ if ((src_len = read(src_fd, src_buf, SEGMENT_SIZE)) == -1)
+ pg_fatal("could not read file \"%s\": %s", src, strerror(errno));
+
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+
+ /*
+ * Read the destination page at dst_pageno into the buffer. The page may contain
+ * data from the previous source segment. Initialize the page if the page is new.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+ if (read(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not read file \"%s\": %s", dst, strerror(errno));
+ if (PageIsNew(page))
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Rewind the file position, so the first write will overwrite the page.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+
+ src_buf_offset = 0;
+ while (src_buf_offset < src_len)
+ {
+ len_to_copy = Min(src_len - src_buf_offset, SizeOfPageContents - dst_offset);
+ memcpy(PageGetContents(page) + dst_offset, src_buf + src_buf_offset, len_to_copy);
+ src_buf_offset += len_to_copy;
+
+ if (new_cluster.controldata.data_checksum_version > 0)
+ ((PageHeader) page)->pd_checksum = pg_checksum_page(page, dst_pageno);
+ if (write(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not write file \"%s\": %s", dst, strerror(errno));
+
+ dst_pageno++;
+ dst_offset = 0;
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Switch segments if we reached the end of the current segment.
+ */
+ if (dst_pageno % SLRU_PAGES_PER_SEGMENT == 0)
+ {
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ dst_segno++;
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+ }
+ }
+
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ pg_free(src_buf);
+ close(src_fd);
+}
+
+void
+upgrade_xact_cache(const char *src_subdir, const char *dst_subdir)
+{
+ char src_dir[MAXPGPATH];
+ char dst_dir[MAXPGPATH];
+
+ DIR *src_dirp;
+ struct dirent *src_dirent;
+
+ snprintf(src_dir, sizeof(src_dir), "%s/%s", old_cluster.pgdata, src_subdir);
+ snprintf(dst_dir, sizeof(dst_dir), "%s/%s", new_cluster.pgdata, dst_subdir);
+
+ if ((src_dirp = opendir(src_dir)) == NULL)
+ pg_fatal("could not open directory \"%s\": %s", src_dir, strerror(errno));
+
+ while (errno = 0, (src_dirent = readdir(src_dirp)) != NULL)
+ {
+ if (segment_file_filter(src_dirent))
+ upgrade_file(src_dir, src_dirent->d_name, dst_dir);
+ }
+
+ if (closedir(src_dirp) != 0)
+ pg_fatal("could not close directory \"%s\": %s", src_dir, strerror(errno));
+}
diff --git a/src/bin/pg_upgrade/pg_upgrade.c b/src/bin/pg_upgrade/pg_upgrade.c
index 10c94a6c1f..ed4a0f5c43 100644
--- a/src/bin/pg_upgrade/pg_upgrade.c
+++ b/src/bin/pg_upgrade/pg_upgrade.c
@@ -660,14 +660,23 @@ copy_subdir_files(const char *old_subdir, const char *new_subdir)
static void
copy_xact_xlog_xid(void)
{
+ bool slru_header_changed = false;
+
/*
* Copy old commit logs to new data dir. pg_clog has been renamed to
* pg_xact in post-10 clusters.
*/
- copy_subdir_files(GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact",
- GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact");
+ char *xact_old_dir = GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+ char *xact_new_dir = GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+
+ if (new_cluster.controldata.cat_ver >= SLRU_PAGE_HEADER_CAT_VER &&
+ old_cluster.controldata.cat_ver < SLRU_PAGE_HEADER_CAT_VER)
+ slru_header_changed = true;
+
+ if (slru_header_changed)
+ upgrade_xact_cache(xact_old_dir, xact_new_dir);
+ else
+ copy_subdir_files(xact_old_dir, xact_new_dir);
prep_status("Setting oldest XID for new cluster");
exec_prog(UTILITY_LOG_FILE, NULL, true, true,
@@ -702,7 +711,8 @@ copy_xact_xlog_xid(void)
* server doesn't attempt to read multis older than the cutoff value.
*/
if (old_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
- new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
+ !slru_header_changed)
{
copy_subdir_files("pg_multixact/offsets", "pg_multixact/offsets");
copy_subdir_files("pg_multixact/members", "pg_multixact/members");
@@ -722,7 +732,8 @@ copy_xact_xlog_xid(void)
new_cluster.pgdata);
check_ok();
}
- else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER ||
+ slru_header_changed)
{
/*
* Remove offsets/0000 file created by initdb that no longer matches
diff --git a/src/bin/pg_upgrade/pg_upgrade.h b/src/bin/pg_upgrade/pg_upgrade.h
index 857d715049..dc81020b62 100644
--- a/src/bin/pg_upgrade/pg_upgrade.h
+++ b/src/bin/pg_upgrade/pg_upgrade.h
@@ -115,6 +115,11 @@ extern char *output_files[];
*/
#define MULTIXACT_FORMATCHANGE_CAT_VER 201301231
+/*
+ * A page header was added to each SLRU page in 17.0.
+ */
+#define SLRU_PAGE_HEADER_CAT_VER 202403061
+
/*
* large object chunk size added to pg_controldata,
* commit 5f93c37805e7485488480916b4585e098d3cc883
@@ -412,6 +417,7 @@ void rewriteVisibilityMap(const char *fromfile, const char *tofile,
void check_file_clone(void);
void check_copy_file_range(void);
void check_hard_link(void);
+void upgrade_xact_cache(const char *src_subdir, const char *dst_subdir);
/* fopen_priv() is no longer different from fopen() */
#define fopen_priv(path, mode) fopen(path, mode)
diff --git a/src/include/access/slru.h b/src/include/access/slru.h
index 8a8d191873..5d5bb13cb0 100644
--- a/src/include/access/slru.h
+++ b/src/include/access/slru.h
@@ -96,17 +96,6 @@ typedef struct SlruSharedData
*/
int *bank_cur_lru_count;
- /*
- * Optional array of WAL flush LSNs associated with entries in the SLRU
- * pages. If not zero/NULL, we must flush WAL before writing pages (true
- * for pg_xact, false for everything else). group_lsn[] has
- * lsn_groups_per_page entries per buffer slot, each containing the
- * highest LSN known for a contiguous group of SLRU entries on that slot's
- * page.
- */
- XLogRecPtr *group_lsn;
- int lsn_groups_per_page;
-
/*
* latest_page_number is the page number of the current end of the log;
* this is not critical data, since we use it only to avoid swapping out
@@ -184,9 +173,9 @@ SimpleLruGetBankLock(SlruCtl ctl, int64 pageno)
return &(ctl->shared->bank_locks[bankno].lock);
}
-extern Size SimpleLruShmemSize(int nslots, int nlsns);
+extern Size SimpleLruShmemSize(int nslots);
extern int SimpleLruAutotuneBuffers(int divisor, int max);
-extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
+extern void SimpleLruInit(SlruCtl ctl, const char *name, int nslots,
const char *subdir, int buffer_tranche_id,
int bank_tranche_id, SyncRequestHandler sync_handler,
bool long_segment_names);
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index bedc2a0d72..854fb01281 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
*/
/* yyyymmddN */
-#define CATALOG_VERSION_NO 202403052
+#define CATALOG_VERSION_NO 202403061
#endif
diff --git a/src/include/storage/bufpage.h b/src/include/storage/bufpage.h
index d0df02d39c..2dc83451a6 100644
--- a/src/include/storage/bufpage.h
+++ b/src/include/storage/bufpage.h
@@ -201,6 +201,7 @@ typedef PageHeaderData *PageHeader;
* handling pages.
*/
#define PG_PAGE_LAYOUT_VERSION 4
+#define PG_SLRU_PAGE_LAYOUT_VERSION 1
#define PG_DATA_CHECKSUM_VERSION 1
/* ----------------------------------------------------------------
@@ -257,6 +258,11 @@ PageGetContents(Page page)
return (char *) page + MAXALIGN(SizeOfPageHeaderData);
}
+/*
+ * Space available for storing page contents.
+ */
+#define SizeOfPageContents (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
+
/* ----------------
* functions to access page size info
* ----------------
@@ -486,6 +492,7 @@ StaticAssertDecl(BLCKSZ == ((BLCKSZ / sizeof(size_t)) * sizeof(size_t)),
"BLCKSZ has to be a multiple of sizeof(size_t)");
extern void PageInit(Page page, Size pageSize, Size specialSize);
+extern void PageInitSLRU(Page page, Size pageSize, Size specialSize);
extern bool PageIsVerifiedExtended(Page page, BlockNumber blkno, int flags);
extern OffsetNumber PageAddItemExtended(Page page, Item item, Size size,
OffsetNumber offsetNumber, int flags);
diff --git a/src/test/modules/test_slru/test_slru.c b/src/test/modules/test_slru/test_slru.c
index 068a21f125..026aa79912 100644
--- a/src/test/modules/test_slru/test_slru.c
+++ b/src/test/modules/test_slru/test_slru.c
@@ -17,6 +17,7 @@
#include "access/slru.h"
#include "access/transam.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/shmem.h"
@@ -72,8 +73,8 @@ test_slru_page_write(PG_FUNCTION_ARGS)
TestSlruCtl->shared->page_status[slotno] = SLRU_PAGE_VALID;
/* write given data to the page, up to the limit of the page */
- strncpy(TestSlruCtl->shared->page_buffer[slotno], data,
- BLCKSZ - 1);
+ strncpy(PageGetContents(TestSlruCtl->shared->page_buffer[slotno]), data,
+ SizeOfPageContents - 1);
SimpleLruWritePage(TestSlruCtl, slotno);
LWLockRelease(lock);
@@ -101,7 +102,7 @@ test_slru_page_read(PG_FUNCTION_ARGS)
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(TestSlruCtl, pageno,
write_ok, InvalidTransactionId);
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(lock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -120,7 +121,7 @@ test_slru_page_readonly(PG_FUNCTION_ARGS)
pageno,
InvalidTransactionId);
Assert(LWLockHeldByMe(lock));
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(lock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -201,7 +202,7 @@ test_slru_shmem_request(void)
prev_shmem_request_hook();
/* reserve shared memory for the test SLRU */
- RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS, 0));
+ RequestAddinShmemSpace(SimpleLruShmemSize(NUM_TEST_BUFFERS));
}
static bool
@@ -240,7 +241,7 @@ test_slru_shmem_startup(void)
TestSlruCtl->PagePrecedes = test_slru_page_precedes_logically;
SimpleLruInit(TestSlruCtl, "TestSLRU",
- NUM_TEST_BUFFERS, 0, slru_dir_name,
+ NUM_TEST_BUFFERS, slru_dir_name,
test_buffer_tranche_id, test_tranche_id, SYNC_HANDLER_NONE,
long_segment_names);
}
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-03-08 21:22 ` Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Jeff Davis @ 2024-03-08 21:22 UTC (permalink / raw)
To: Li, Yong <[email protected]>; Aleksander Alekseev <[email protected]>; +Cc: PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
On Wed, 2024-03-06 at 12:01 +0000, Li, Yong wrote:
> Rebase the patch against the latest HEAD.
The upgrade logic could use more comments explaining what's going on
and why. As I understand it, it's a one-time conversion that needs to
happen between 16 and 17. Is that right?
Was the way CLOG is upgraded already decided in some earlier
discussion?
Given that the CLOG is append-only and gets truncated occasionally, I
wonder whether we can just have some marker that xids before some
number are the old CLOG, and xids beyond that number are in the new
CLOG. I'm not necessarily suggesting that; just an idea.
Regards,
Jeff Davis
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
@ 2024-03-11 10:01 ` Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2024-03-11 10:01 UTC (permalink / raw)
To: Jeff Davis <[email protected]>; +Cc: Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
> On Mar 9, 2024, at 05:22, Jeff Davis <[email protected]> wrote:
>
> External Email
>
> On Wed, 2024-03-06 at 12:01 +0000, Li, Yong wrote:
>> Rebase the patch against the latest HEAD.
>
> The upgrade logic could use more comments explaining what's going on
> and why. As I understand it, it's a one-time conversion that needs to
> happen between 16 and 17. Is that right?
>
> Regards,
> Jeff Davis
>
> In the new code we effectively store only one LSN per page, which I
> understand is strictly worse. Maybe the idea of doing away with these
> LSN groups should be reconsidered ... unless I completely misunderstand
> the whole thing.
>
> --
> Álvaro Herrera PostgreSQL Developer —
Thanks for the comments on LSN groups and pg_upgrade.
I have updated the patch to address both comments:
- The clog LSN group has been brought back.
Now the page LSN on each clog page is used for honoring the write-ahead rule
and it is always the highest LSN of all the LSN groups on the page.
The LSN groups are used by TransactionIdGetStatus() as before.
- New comments have been added to pg_upgrade to mention the SLRU
page header change as the reason for upgrading clog files.
Regards,
Yong
Attachments:
[application/octet-stream] slru_page_header_v6.patch (44.8K, ../../[email protected]/2-slru_page_header_v6.patch)
download | inline diff:
src/backend/access/transam/clog.c | 45 ++++++---
src/backend/access/transam/commit_ts.c | 22 ++--
src/backend/access/transam/multixact.c | 55 ++++++----
src/backend/access/transam/slru.c | 68 ++++++-------
src/backend/access/transam/subtrans.c | 8 +-
src/backend/commands/async.c | 23 +++--
src/backend/storage/lmgr/predicate.c | 12 ++-
src/backend/storage/page/bufpage.c | 25 +++++
src/bin/pg_checksums/pg_checksums.c | 9 ++
src/bin/pg_resetwal/t/001_basic.pl | 6 +-
src/bin/pg_upgrade/file.c | 178 +++++++++++++++++++++++++++++++++
src/bin/pg_upgrade/pg_upgrade.c | 28 ++++--
src/bin/pg_upgrade/pg_upgrade.h | 6 ++
src/include/catalog/catversion.h | 2 +-
src/include/storage/bufpage.h | 7 ++
src/test/modules/test_slru/test_slru.c | 9 +-
16 files changed, 395 insertions(+), 108 deletions(-)
diff --git a/src/backend/access/transam/clog.c b/src/backend/access/transam/clog.c
index 44c253246b..86348e242b 100644
--- a/src/backend/access/transam/clog.c
+++ b/src/backend/access/transam/clog.c
@@ -42,6 +42,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/proc.h"
#include "storage/sync.h"
#include "utils/guc_hooks.h"
@@ -61,7 +62,7 @@
/* We need two bits per xact, so four xacts fit in a byte */
#define CLOG_BITS_PER_XACT 2
#define CLOG_XACTS_PER_BYTE 4
-#define CLOG_XACTS_PER_PAGE (BLCKSZ * CLOG_XACTS_PER_BYTE)
+#define CLOG_XACTS_PER_PAGE (SizeOfPageContents * CLOG_XACTS_PER_BYTE)
#define CLOG_XACT_BITMASK ((1 << CLOG_BITS_PER_XACT) - 1)
/*
@@ -90,7 +91,13 @@ TransactionIdToPage(TransactionId xid)
/* We store the latest async LSN for each group of transactions */
#define CLOG_XACTS_PER_LSN_GROUP 32 /* keep this a power of 2 */
-#define CLOG_LSNS_PER_PAGE (CLOG_XACTS_PER_PAGE / CLOG_XACTS_PER_LSN_GROUP)
+
+/*
+ * Use BLCKSZ instead of SizeOfPageContents so that CLOG_LSNS_PER_PAGE is
+ * a power of 2. Using BLCKSZ wastes the last 4 LSN groups per page, but
+ * this is acceptable given that each page has 1,024 LSN groups.
+ */
+#define CLOG_LSNS_PER_PAGE ((BLCKSZ * CLOG_XACTS_PER_BYTE) / CLOG_XACTS_PER_LSN_GROUP)
#define GetLSNIndex(slotno, xid) ((slotno) * CLOG_LSNS_PER_PAGE + \
((xid) % (TransactionId) CLOG_XACTS_PER_PAGE) / CLOG_XACTS_PER_LSN_GROUP)
@@ -112,7 +119,7 @@ static SlruCtlData XactCtlData;
static int ZeroCLOGPage(int64 pageno, bool writeXlog);
static bool CLOGPagePrecedes(int64 page1, int64 page2);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXact,
Oid oldestXactDb);
static void TransactionIdSetPageStatus(TransactionId xid, int nsubxids,
@@ -665,13 +672,15 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
char *byteptr;
char byteval;
char curval;
+ Page page;
Assert(XactCtl->shared->page_number[slotno] == TransactionIdToPage(xid));
Assert(LWLockHeldByMeInMode(SimpleLruGetBankLock(XactCtl,
XactCtl->shared->page_number[slotno]),
LW_EXCLUSIVE));
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
curval = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
/*
@@ -700,7 +709,7 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*byteptr = byteval;
/*
- * Update the group LSN if the transaction completion LSN is higher.
+ * Update the page & group LSN if the transaction completion LSN is higher.
*
* Note: lsn will be invalid when supplied during InRecovery processing,
* so we don't need to do anything special to avoid LSN updates during
@@ -709,10 +718,13 @@ TransactionIdSetStatusBit(TransactionId xid, XidStatus status, XLogRecPtr lsn, i
*/
if (!XLogRecPtrIsInvalid(lsn))
{
- int lsnindex = GetLSNIndex(slotno, xid);
+ int lsnindex = GetLSNIndex(slotno, xid);
if (XactCtl->shared->group_lsn[lsnindex] < lsn)
XactCtl->shared->group_lsn[lsnindex] = lsn;
+
+ if (PageGetLSN(page) < lsn)
+ PageSetLSN(page, lsn);
}
}
@@ -739,13 +751,15 @@ TransactionIdGetStatus(TransactionId xid, XLogRecPtr *lsn)
int bshift = TransactionIdToBIndex(xid) * CLOG_BITS_PER_XACT;
int slotno;
int lsnindex;
+ Page page;
char *byteptr;
XidStatus status;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(XactCtl, pageno, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ page = XactCtl->shared->page_buffer[slotno];
+ byteptr = PageGetContents(page) + byteno;
status = (*byteptr >> bshift) & CLOG_XACT_BITMASK;
@@ -860,11 +874,17 @@ static int
ZeroCLOGPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(XactCtl, pageno);
+ page = XactCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -917,12 +937,12 @@ TrimCLOG(void)
char *byteptr;
slotno = SimpleLruReadPage(XactCtl, pageno, false, xid);
- byteptr = XactCtl->shared->page_buffer[slotno] + byteno;
+ byteptr = PageGetContents(XactCtl->shared->page_buffer[slotno]) + byteno;
/* Zero so-far-unused positions in the current byte */
*byteptr &= (1 << bshift) - 1;
/* Zero the rest of the page */
- MemSet(byteptr + 1, 0, BLCKSZ - byteno - 1);
+ MemSet(byteptr + 1, 0, SizeOfPageContents - byteno - 1);
XactCtl->shared->page_dirty[slotno] = true;
}
@@ -946,7 +966,6 @@ CheckPointCLOG(void)
TRACE_POSTGRESQL_CLOG_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that CLOG has room for a newly-allocated XID.
*
@@ -1070,12 +1089,12 @@ CLOGPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
+ return XLogInsert(RM_CLOG_ID, CLOG_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/commit_ts.c b/src/backend/access/transam/commit_ts.c
index f221494687..99f9fc5076 100644
--- a/src/backend/access/transam/commit_ts.c
+++ b/src/backend/access/transam/commit_ts.c
@@ -29,6 +29,7 @@
#include "access/xlogutils.h"
#include "funcapi.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/shmem.h"
#include "utils/fmgrprotos.h"
#include "utils/guc_hooks.h"
@@ -61,7 +62,7 @@ typedef struct CommitTimestampEntry
sizeof(RepOriginId))
#define COMMIT_TS_XACTS_PER_PAGE \
- (BLCKSZ / SizeOfCommitTimestampEntry)
+ (SizeOfPageContents / SizeOfCommitTimestampEntry)
/*
@@ -118,7 +119,7 @@ static int ZeroCommitTsPage(int64 pageno, bool writeXlog);
static bool CommitTsPagePrecedes(int64 page1, int64 page2);
static void ActivateCommitTs(void);
static void DeactivateCommitTs(void);
-static void WriteZeroPageXlogRec(int64 pageno);
+static XLogRecPtr WriteZeroPageXlogRec(int64 pageno);
static void WriteTruncateXlogRec(int64 pageno, TransactionId oldestXid);
/*
@@ -253,11 +254,12 @@ TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
+ Assert(xid == slotno * COMMIT_TS_XACTS_PER_PAGE + entryno);
entry.time = ts;
entry.nodeid = nodeid;
- memcpy(CommitTsCtl->shared->page_buffer[slotno] +
+ memcpy(PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
@@ -336,7 +338,7 @@ TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
- CommitTsCtl->shared->page_buffer[slotno] +
+ PageGetContents(CommitTsCtl->shared->page_buffer[slotno]) +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
@@ -615,11 +617,17 @@ static int
ZeroCommitTsPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(CommitTsCtl, pageno);
+ page = CommitTsCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteZeroPageXlogRec(pageno);
+ {
+ lsn = WriteZeroPageXlogRec(pageno);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -985,12 +993,12 @@ CommitTsPagePrecedes(int64 page1, int64 page2)
/*
* Write a ZEROPAGE xlog record
*/
-static void
+static XLogRecPtr
WriteZeroPageXlogRec(int64 pageno)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
+ return XLogInsert(RM_COMMIT_TS_ID, COMMIT_TS_ZEROPAGE);
}
/*
diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c
index 83b578dced..b6f5428327 100644
--- a/src/backend/access/transam/multixact.c
+++ b/src/backend/access/transam/multixact.c
@@ -83,6 +83,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "postmaster/autovacuum.h"
+#include "storage/bufpage.h"
#include "storage/pmsignal.h"
#include "storage/proc.h"
#include "storage/procarray.h"
@@ -105,7 +106,7 @@
*/
/* We need four bytes per offset */
-#define MULTIXACT_OFFSETS_PER_PAGE (BLCKSZ / sizeof(MultiXactOffset))
+#define MULTIXACT_OFFSETS_PER_PAGE (SizeOfPageContents / sizeof(MultiXactOffset))
#define MultiXactIdToOffsetPage(xid) \
((xid) / (MultiXactOffset) MULTIXACT_OFFSETS_PER_PAGE)
@@ -118,8 +119,8 @@
* additional flag bits for each TransactionId. To do this without getting
* into alignment issues, we store four bytes of flags, and then the
* corresponding 4 Xids. Each such 5-word (20-byte) set we call a "group", and
- * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 409 groups
- * per page. This wastes 12 bytes per page, but that's OK -- simplicity (and
+ * are stored as a whole in pages. Thus, with 8kB BLCKSZ, we keep 408 groups
+ * per page. This wastes 8 bytes per page, but that's OK -- simplicity (and
* performance) trumps space efficiency here.
*
* Note that the "offset" macros work with byte offset, not array indexes, so
@@ -137,7 +138,7 @@
/* size in bytes of a complete group */
#define MULTIXACT_MEMBERGROUP_SIZE \
(sizeof(TransactionId) * MULTIXACT_MEMBERS_PER_MEMBERGROUP + MULTIXACT_FLAGBYTES_PER_GROUP)
-#define MULTIXACT_MEMBERGROUPS_PER_PAGE (BLCKSZ / MULTIXACT_MEMBERGROUP_SIZE)
+#define MULTIXACT_MEMBERGROUPS_PER_PAGE (SizeOfPageContents / MULTIXACT_MEMBERGROUP_SIZE)
#define MULTIXACT_MEMBERS_PER_PAGE \
(MULTIXACT_MEMBERGROUPS_PER_PAGE * MULTIXACT_MEMBERS_PER_MEMBERGROUP)
@@ -364,7 +365,7 @@ static bool MultiXactOffsetWouldWrap(MultiXactOffset boundary,
MultiXactOffset start, uint32 distance);
static bool SetOffsetVacuumLimit(bool is_startup);
static bool find_multixact_start(MultiXactId multi, MultiXactOffset *result);
-static void WriteMZeroPageXlogRec(int64 pageno, uint8 info);
+static XLogRecPtr WriteMZeroPageXlogRec(int64 pageno, uint8 info);
static void WriteMTruncateXlogRec(Oid oldestMultiDB,
MultiXactId startTruncOff,
MultiXactId endTruncOff,
@@ -885,7 +886,7 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
* take the trouble to generalize the slru.c error reporting code.
*/
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
*offptr = offset;
@@ -934,12 +935,12 @@ RecordNewMultiXact(MultiXactId multi, MultiXactOffset offset,
}
memberptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
*memberptr = members[i].xid;
flagsptr = (uint32 *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
flagsval = *flagsptr;
flagsval &= ~(((1 << MXACT_MEMBER_BITS_PER_XACT) - 1) << bshift);
@@ -1364,7 +1365,7 @@ retry:
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
@@ -1413,7 +1414,7 @@ retry:
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, tmpMXact);
}
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
nextMXOffset = *offptr;
@@ -1470,7 +1471,7 @@ retry:
}
xactptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
if (!TransactionIdIsValid(*xactptr))
{
@@ -1481,7 +1482,7 @@ retry:
flagsoff = MXOffsetToFlagsOffset(offset);
bshift = MXOffsetToFlagsBitShift(offset);
- flagsptr = (uint32 *) (MultiXactMemberCtl->shared->page_buffer[slotno] + flagsoff);
+ flagsptr = (uint32 *) (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + flagsoff);
ptr[truelength].xid = *xactptr;
ptr[truelength].status = (*flagsptr >> bshift) & MXACT_MEMBER_XACT_BITMASK;
@@ -1999,11 +2000,17 @@ static int
ZeroMultiXactOffsetPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactOffsetCtl, pageno);
+ page = MultiXactOffsetCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_OFF_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2015,11 +2022,17 @@ static int
ZeroMultiXactMemberPage(int64 pageno, bool writeXlog)
{
int slotno;
+ Page page;
+ XLogRecPtr lsn = 0;
slotno = SimpleLruZeroPage(MultiXactMemberCtl, pageno);
+ page = MultiXactMemberCtl->shared->page_buffer[slotno];
if (writeXlog)
- WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ {
+ lsn = WriteMZeroPageXlogRec(pageno, XLOG_MULTIXACT_ZERO_MEM_PAGE);
+ PageSetLSN(page, lsn);
+ }
return slotno;
}
@@ -2143,10 +2156,10 @@ TrimMultiXact(void)
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(MultiXactOffsetCtl, pageno, true, nextMXact);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
- MemSet(offptr, 0, BLCKSZ - (entryno * sizeof(MultiXactOffset)));
+ MemSet(offptr, 0, SizeOfPageContents - (entryno * sizeof(MultiXactOffset)));
MultiXactOffsetCtl->shared->page_dirty[slotno] = true;
LWLockRelease(lock);
@@ -2177,9 +2190,9 @@ TrimMultiXact(void)
memberoff = MXOffsetToMemberOffset(offset);
slotno = SimpleLruReadPage(MultiXactMemberCtl, pageno, true, offset);
xidptr = (TransactionId *)
- (MultiXactMemberCtl->shared->page_buffer[slotno] + memberoff);
+ (PageGetContents(MultiXactMemberCtl->shared->page_buffer[slotno]) + memberoff);
- MemSet(xidptr, 0, BLCKSZ - memberoff);
+ MemSet(xidptr, 0, SizeOfPageContents - memberoff);
/*
* Note: we don't need to zero out the flag bits in the remaining
@@ -2834,7 +2847,7 @@ find_multixact_start(MultiXactId multi, MultiXactOffset *result)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(MultiXactOffsetCtl, pageno, multi);
- offptr = (MultiXactOffset *) MultiXactOffsetCtl->shared->page_buffer[slotno];
+ offptr = (MultiXactOffset *) PageGetContents(MultiXactOffsetCtl->shared->page_buffer[slotno]);
offptr += entryno;
offset = *offptr;
LWLockRelease(SimpleLruGetBankLock(MultiXactOffsetCtl, pageno));
@@ -3268,12 +3281,12 @@ MultiXactOffsetPrecedes(MultiXactOffset offset1, MultiXactOffset offset2)
* Write an xlog record reflecting the zeroing of either a MEMBERs or
* OFFSETs page (info shows which)
*/
-static void
+static XLogRecPtr
WriteMZeroPageXlogRec(int64 pageno, uint8 info)
{
XLogBeginInsert();
XLogRegisterData((char *) (&pageno), sizeof(pageno));
- (void) XLogInsert(RM_MULTIXACT_ID, info);
+ return XLogInsert(RM_MULTIXACT_ID, info);
}
/*
diff --git a/src/backend/access/transam/slru.c b/src/backend/access/transam/slru.c
index f99ec38a4a..189b776cef 100644
--- a/src/backend/access/transam/slru.c
+++ b/src/backend/access/transam/slru.c
@@ -68,6 +68,7 @@
#include "access/xlogutils.h"
#include "miscadmin.h"
#include "pgstat.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/shmem.h"
#include "utils/guc_hooks.h"
@@ -155,6 +156,7 @@ typedef enum
SLRU_WRITE_FAILED,
SLRU_FSYNC_FAILED,
SLRU_CLOSE_FAILED,
+ SLRU_DATA_CORRUPTED,
} SlruErrorCause;
static SlruErrorCause slru_errcause;
@@ -378,8 +380,8 @@ SimpleLruZeroPage(SlruCtl ctl, int64 pageno)
shared->page_dirty[slotno] = true;
SlruRecentlyUsed(shared, slotno);
- /* Set the buffer to zeroes */
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ /* Initialize the page. */
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
/* Set the LSNs for this new page to zero */
SimpleLruZeroLSNs(ctl, slotno);
@@ -815,7 +817,7 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
ereport(LOG,
(errmsg("file \"%s\" doesn't exist, reading as zeroes",
path)));
- MemSet(shared->page_buffer[slotno], 0, BLCKSZ);
+ PageInitSLRU(shared->page_buffer[slotno], BLCKSZ, 0);
return true;
}
@@ -838,6 +840,13 @@ SlruPhysicalReadPage(SlruCtl ctl, int64 pageno, int slotno)
return false;
}
+ if (!PageIsVerifiedExtended(shared->page_buffer[slotno], pageno, PIV_REPORT_STAT))
+ {
+ slru_errcause = SLRU_DATA_CORRUPTED;
+ slru_errno = 0;
+ return false;
+ }
+
return true;
}
@@ -864,6 +873,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
off_t offset = rpageno * BLCKSZ;
char path[MAXPGPATH];
int fd = -1;
+ Page page = shared->page_buffer[slotno];
+ XLogRecPtr lsn;
/* update the stats counter of written pages */
pgstat_count_slru_page_written(shared->slru_stats_idx);
@@ -872,41 +883,21 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
* Honor the write-WAL-before-data rule, if appropriate, so that we do not
* write out data before associated WAL records. This is the same action
* performed during FlushBuffer() in the main buffer manager.
+ *
+ * The largest async-commit LSN for the page is maintained through page LSN.
*/
- if (shared->group_lsn != NULL)
+ lsn = PageGetLSN(page);
+ if (!XLogRecPtrIsInvalid(lsn))
{
/*
- * We must determine the largest async-commit LSN for the page. This
- * is a bit tedious, but since this entire function is a slow path
- * anyway, it seems better to do this here than to maintain a per-page
- * LSN variable (which'd need an extra comparison in the
- * transaction-commit path).
+ * As noted above, elog(ERROR) is not acceptable here, so if
+ * XLogFlush were to fail, we must PANIC. This isn't much of a
+ * restriction because XLogFlush is just about all critical
+ * section anyway, but let's make sure.
*/
- XLogRecPtr max_lsn;
- int lsnindex;
-
- lsnindex = slotno * shared->lsn_groups_per_page;
- max_lsn = shared->group_lsn[lsnindex++];
- for (int lsnoff = 1; lsnoff < shared->lsn_groups_per_page; lsnoff++)
- {
- XLogRecPtr this_lsn = shared->group_lsn[lsnindex++];
-
- if (max_lsn < this_lsn)
- max_lsn = this_lsn;
- }
-
- if (!XLogRecPtrIsInvalid(max_lsn))
- {
- /*
- * As noted above, elog(ERROR) is not acceptable here, so if
- * XLogFlush were to fail, we must PANIC. This isn't much of a
- * restriction because XLogFlush is just about all critical
- * section anyway, but let's make sure.
- */
- START_CRIT_SECTION();
- XLogFlush(max_lsn);
- END_CRIT_SECTION();
- }
+ START_CRIT_SECTION();
+ XLogFlush(lsn);
+ END_CRIT_SECTION();
}
/*
@@ -971,6 +962,8 @@ SlruPhysicalWritePage(SlruCtl ctl, int64 pageno, int slotno, SlruWriteAll fdata)
}
}
+ PageSetChecksumInplace(shared->page_buffer[slotno], pageno);
+
errno = 0;
pgstat_report_wait_start(WAIT_EVENT_SLRU_WRITE);
if (pg_pwrite(fd, shared->page_buffer[slotno], BLCKSZ, offset) != BLCKSZ)
@@ -1091,6 +1084,13 @@ SlruReportIOError(SlruCtl ctl, int64 pageno, TransactionId xid)
errdetail("Could not close file \"%s\": %m.",
path)));
break;
+ case SLRU_DATA_CORRUPTED:
+ ereport(ERROR,
+ (errcode(ERRCODE_DATA_CORRUPTED),
+ errmsg("could not access status of transaction %u", xid),
+ errdetail("Invalid page from file \"%s\" at offset %d.",
+ path, offset)));
+ break;
default:
/* can't get here, we trust */
elog(ERROR, "unrecognized SimpleLru error cause: %d",
diff --git a/src/backend/access/transam/subtrans.c b/src/backend/access/transam/subtrans.c
index 50bb1d8cfc..869ae7a25d 100644
--- a/src/backend/access/transam/subtrans.c
+++ b/src/backend/access/transam/subtrans.c
@@ -34,6 +34,7 @@
#include "miscadmin.h"
#include "pg_trace.h"
#include "utils/guc_hooks.h"
+#include "storage/bufpage.h"
#include "utils/snapmgr.h"
@@ -51,7 +52,7 @@
*/
/* We need four bytes per xact */
-#define SUBTRANS_XACTS_PER_PAGE (BLCKSZ / sizeof(TransactionId))
+#define SUBTRANS_XACTS_PER_PAGE (SizeOfPageContents / sizeof(TransactionId))
/*
* Although we return an int64 the actual value can't currently exceed
@@ -97,7 +98,7 @@ SubTransSetParent(TransactionId xid, TransactionId parent)
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(SubTransCtl, pageno, true, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
/*
@@ -137,7 +138,7 @@ SubTransGetParent(TransactionId xid)
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
- ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
+ ptr = (TransactionId *) PageGetContents(SubTransCtl->shared->page_buffer[slotno]);
ptr += entryno;
parent = *ptr;
@@ -366,7 +367,6 @@ CheckPointSUBTRANS(void)
TRACE_POSTGRESQL_SUBTRANS_CHECKPOINT_DONE(true);
}
-
/*
* Make sure that SUBTRANS has room for a newly-allocated XID.
*
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c
index d0891e3f0e..a4cb773f73 100644
--- a/src/backend/commands/async.c
+++ b/src/backend/commands/async.c
@@ -140,6 +140,7 @@
#include "libpq/libpq.h"
#include "libpq/pqformat.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/ipc.h"
#include "storage/lmgr.h"
#include "storage/procsignal.h"
@@ -160,7 +161,7 @@
* than that, so changes in that data structure won't affect user-visible
* restrictions.
*/
-#define NOTIFY_PAYLOAD_MAX_LENGTH (BLCKSZ - NAMEDATALEN - 128)
+#define NOTIFY_PAYLOAD_MAX_LENGTH (SizeOfPageContents - NAMEDATALEN - 128)
/*
* Struct representing an entry in the global notify queue
@@ -309,7 +310,7 @@ static SlruCtlData NotifyCtlData;
#define NotifyCtl (&NotifyCtlData)
#define QUEUE_PAGESIZE BLCKSZ
-
+#define QUEUE_PAGE_CAPACITY (QUEUE_PAGESIZE - MAXALIGN(SizeOfPageHeaderData))
#define QUEUE_FULL_WARN_INTERVAL 5000 /* warn at most once every 5s */
/*
@@ -1295,14 +1296,14 @@ asyncQueueAdvance(volatile QueuePosition *position, int entryLength)
* written or read.
*/
offset += entryLength;
- Assert(offset <= QUEUE_PAGESIZE);
+ Assert(offset <= QUEUE_PAGE_CAPACITY);
/*
* In a second step check if another entry can possibly be written to the
* page. If so, stay here, we have reached the next position. If not, then
* we need to move on to the next page.
*/
- if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGESIZE)
+ if (offset + QUEUEALIGN(AsyncQueueEntryEmptySize) > QUEUE_PAGE_CAPACITY)
{
pageno++;
offset = 0;
@@ -1405,7 +1406,7 @@ asyncQueueAddEntries(ListCell *nextNotify)
offset = QUEUE_POS_OFFSET(queue_head);
/* Check whether the entry really fits on the current page */
- if (offset + qe.length <= QUEUE_PAGESIZE)
+ if (offset + qe.length <= QUEUE_PAGE_CAPACITY)
{
/* OK, so advance nextNotify past this item */
nextNotify = lnext(pendingNotifies->events, nextNotify);
@@ -1417,14 +1418,14 @@ asyncQueueAddEntries(ListCell *nextNotify)
* only check dboid and since it won't match any reader's database
* OID, they will ignore this entry and move on.
*/
- qe.length = QUEUE_PAGESIZE - offset;
+ qe.length = QUEUE_PAGE_CAPACITY - offset;
qe.dboid = InvalidOid;
qe.data[0] = '\0'; /* empty channel */
qe.data[1] = '\0'; /* empty payload */
}
/* Now copy qe into the shared buffer page */
- memcpy(NotifyCtl->shared->page_buffer[slotno] + offset,
+ memcpy(PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + offset,
&qe,
qe.length);
@@ -1955,10 +1956,10 @@ asyncQueueReadAllNotifications(void)
else
{
/* fetch all the rest of the page */
- copysize = QUEUE_PAGESIZE - curoffset;
+ copysize = QUEUE_PAGE_CAPACITY - curoffset;
}
- memcpy(page_buffer.buf + curoffset,
- NotifyCtl->shared->page_buffer[slotno] + curoffset,
+ memcpy(PageGetContents(page_buffer.buf) + curoffset,
+ PageGetContents(NotifyCtl->shared->page_buffer[slotno]) + curoffset,
copysize);
/* Release lock that we got from SimpleLruReadPage_ReadOnly() */
LWLockRelease(SimpleLruGetBankLock(NotifyCtl, curpage));
@@ -2029,7 +2030,7 @@ asyncQueueProcessPageEntries(volatile QueuePosition *current,
if (QUEUE_POS_EQUAL(thisentry, stop))
break;
- qe = (AsyncQueueEntry *) (page_buffer + QUEUE_POS_OFFSET(thisentry));
+ qe = (AsyncQueueEntry *) (PageGetContents(page_buffer) + QUEUE_POS_OFFSET(thisentry));
/*
* Advance *current over this message, possibly to the next page. As
diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index 3f378c0099..1cc664eee3 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -207,6 +207,7 @@
#include "miscadmin.h"
#include "pgstat.h"
#include "port/pg_lfind.h"
+#include "storage/bufpage.h"
#include "storage/predicate.h"
#include "storage/predicate_internals.h"
#include "storage/proc.h"
@@ -326,8 +327,8 @@ static SlruCtlData SerialSlruCtlData;
#define SerialSlruCtl (&SerialSlruCtlData)
#define SERIAL_PAGESIZE BLCKSZ
-#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
-#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE / SERIAL_ENTRYSIZE)
+#define SERIAL_ENTRYSIZE sizeof(SerCommitSeqNo)
+#define SERIAL_ENTRIESPERPAGE (SERIAL_PAGESIZE - MAXALIGN(SizeOfPageHeaderData) / SERIAL_ENTRYSIZE)
/*
* Set maximum pages based on the number needed to track all transactions.
@@ -337,7 +338,7 @@ static SlruCtlData SerialSlruCtlData;
#define SerialNextPage(page) (((page) >= SERIAL_MAX_PAGE) ? 0 : (page) + 1)
#define SerialValue(slotno, xid) (*((SerCommitSeqNo *) \
- (SerialSlruCtl->shared->page_buffer[slotno] + \
+ (PageGetContents(SerialSlruCtl->shared->page_buffer[slotno]) + \
((((uint32) (xid)) % SERIAL_ENTRIESPERPAGE) * SERIAL_ENTRYSIZE))))
#define SerialPage(xid) (((uint32) (xid)) / SERIAL_ENTRIESPERPAGE)
@@ -789,10 +790,13 @@ SerialPagePrecedesLogicallyUnitTests(void)
* requires burning ~2B XIDs in single-user mode, a negligible
* possibility. Moreover, if it does happen, the consequence would be
* mild, namely a new transaction failing in SimpleLruReadPage().
+ *
+ * NOTE: After adding the page header, the defect affects two pages.
+ * We now assert correct treatment of its second to prior page.
*/
headPage = oldestPage;
targetPage = newestPage;
- Assert(SerialPagePrecedesLogically(headPage, targetPage - 1));
+ Assert(SerialPagePrecedesLogically(headPage, targetPage - 2));
#if 0
Assert(SerialPagePrecedesLogically(headPage, targetPage));
#endif
diff --git a/src/backend/storage/page/bufpage.c b/src/backend/storage/page/bufpage.c
index be6f1f62d2..e8193d7f56 100644
--- a/src/backend/storage/page/bufpage.c
+++ b/src/backend/storage/page/bufpage.c
@@ -59,6 +59,31 @@ PageInit(Page page, Size pageSize, Size specialSize)
/* p->pd_prune_xid = InvalidTransactionId; done by above MemSet */
}
+/*
+ * PageInitSLRU
+ * Initializes the contents of an SLRU page.
+ * Note that we don't calculate an initial checksum here; that's not done
+ * until it's time to write.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
/*
* PageIsVerifiedExtended
diff --git a/src/bin/pg_checksums/pg_checksums.c b/src/bin/pg_checksums/pg_checksums.c
index 9e6fd435f6..06f14f1d2d 100644
--- a/src/bin/pg_checksums/pg_checksums.c
+++ b/src/bin/pg_checksums/pg_checksums.c
@@ -16,6 +16,7 @@
#include <dirent.h>
#include <limits.h>
+#include <stdbool.h>
#include <time.h>
#include <sys/stat.h>
#include <unistd.h>
@@ -593,12 +594,20 @@ main(int argc, char *argv[])
{
total_size = scan_directory(DataDir, "global", true);
total_size += scan_directory(DataDir, "base", true);
+ total_size += scan_directory(DataDir, "pg_commit_ts", true);
+ total_size += scan_directory(DataDir, "pg_multixact", true);
+ total_size += scan_directory(DataDir, "pg_serial", true);
total_size += scan_directory(DataDir, "pg_tblspc", true);
+ total_size += scan_directory(DataDir, "pg_xact", true);
}
(void) scan_directory(DataDir, "global", false);
(void) scan_directory(DataDir, "base", false);
+ (void) scan_directory(DataDir, "pg_commit_ts", false);
+ (void) scan_directory(DataDir, "pg_multixact", false);
+ (void) scan_directory(DataDir, "pg_serial", false);
(void) scan_directory(DataDir, "pg_tblspc", false);
+ (void) scan_directory(DataDir, "pg_xact", false);
if (showprogress)
progress_report(true);
diff --git a/src/bin/pg_resetwal/t/001_basic.pl b/src/bin/pg_resetwal/t/001_basic.pl
index 9829e48106..7b9e034e19 100644
--- a/src/bin/pg_resetwal/t/001_basic.pl
+++ b/src/bin/pg_resetwal/t/001_basic.pl
@@ -206,7 +206,7 @@ push @cmd,
sprintf("%d,%d", hex($files[0]) == 0 ? 3 : hex($files[0]), hex($files[-1]));
@files = get_slru_files('pg_multixact/offsets');
-$mult = 32 * $blcksz / 4;
+$mult = 32 * ($blcksz - 24) / 4;
# -m argument is "new,old"
push @cmd, '-m',
sprintf("%d,%d",
@@ -214,11 +214,11 @@ push @cmd, '-m',
hex($files[0]) == 0 ? 1 : hex($files[0] * $mult));
@files = get_slru_files('pg_multixact/members');
-$mult = 32 * int($blcksz / 20) * 4;
+$mult = 32 * int(($blcksz - 24) / 20) * 4;
push @cmd, '-O', (hex($files[-1]) + 1) * $mult;
@files = get_slru_files('pg_xact');
-$mult = 32 * $blcksz * 4;
+$mult = 32 * ($blcksz - 24) * 4;
push @cmd,
'-u', (hex($files[0]) == 0 ? 3 : hex($files[0]) * $mult),
'-x', ((hex($files[-1]) + 1) * $mult);
diff --git a/src/bin/pg_upgrade/file.c b/src/bin/pg_upgrade/file.c
index beba376f2e..663525f364 100644
--- a/src/bin/pg_upgrade/file.c
+++ b/src/bin/pg_upgrade/file.c
@@ -9,6 +9,7 @@
#include "postgres_fe.h"
+#include <dirent.h>
#include <sys/stat.h>
#include <limits.h>
#include <fcntl.h>
@@ -453,3 +454,180 @@ check_hard_link(void)
unlink(new_link_file);
}
+
+
+/*
+ * Copy SLRU_PAGES_PER_SEGMENT from access/slru.h to avoid including it.
+ */
+#define SLRU_PAGES_PER_SEGMENT 32
+
+#define SEGMENT_SIZE (BLCKSZ * SLRU_PAGES_PER_SEGMENT)
+
+/*
+ * Copy PageInitSLRU from storage/bufpage.c to avoid linking to the backend.
+ */
+void
+PageInitSLRU(Page page, Size pageSize, Size specialSize)
+{
+ PageHeader p = (PageHeader) page;
+
+ specialSize = MAXALIGN(specialSize);
+
+ Assert(pageSize == BLCKSZ);
+ Assert(pageSize > specialSize + SizeOfPageHeaderData);
+
+ /* Make sure all fields of page are zero, as well as unused space */
+ MemSet(p, 0, pageSize);
+
+ p->pd_flags = 0;
+ p->pd_lower = SizeOfPageHeaderData;
+ p->pd_upper = pageSize - specialSize;
+ p->pd_special = pageSize - specialSize;
+ PageSetPageSizeAndVersion(page, pageSize, PG_SLRU_PAGE_LAYOUT_VERSION);
+}
+
+/*
+ * Filter function for scandir(3) to select only segment files.
+ */
+static int
+segment_file_filter(const struct dirent *dirent)
+{
+ return strspn(dirent->d_name, "0123456789ABCDEF") == strlen(dirent->d_name);
+}
+
+/*
+ * Upgrade a single clog segment to add a page header on each page.
+ */
+static void
+upgrade_file(const char *src_dir, const char *src_file, const char *dst_dir)
+{
+ char src[MAXPGPATH];
+ char dst[MAXPGPATH];
+
+ int seg_name_len;
+ int src_segno;
+ int64 src_pageno;
+ int dst_segno;
+ int64 dst_pageno;
+ int dst_offset;
+
+ int src_fd;
+ int dst_fd;
+
+ char *src_buf;
+ ssize_t src_len;
+ ssize_t src_buf_offset;
+ PGAlignedBlock dst_block;
+ Page page = dst_block.data;
+ int len_to_copy;
+
+ seg_name_len = strlen(src_file);
+ src_segno = (int) strtol(src_file, NULL, 16);
+ src_pageno = src_segno * SLRU_PAGES_PER_SEGMENT;
+
+ dst_pageno = src_pageno * BLCKSZ / SizeOfPageContents;
+ dst_offset = src_pageno * BLCKSZ - dst_pageno * SizeOfPageContents;
+ dst_segno = dst_pageno / SLRU_PAGES_PER_SEGMENT;
+
+ snprintf(src, sizeof(src), "%s/%s", src_dir, src_file);
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+
+ src_buf = pg_malloc(SEGMENT_SIZE);
+ if ((src_fd = open(src, O_RDONLY | PG_BINARY, 0)) == -1)
+ pg_fatal("could not open file \"%s\": %s", src, strerror(errno));
+ if ((src_len = read(src_fd, src_buf, SEGMENT_SIZE)) == -1)
+ pg_fatal("could not read file \"%s\": %s", src, strerror(errno));
+
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+
+ /*
+ * Read the destination page at dst_pageno into the buffer. The page may contain
+ * data from the previous source segment. Initialize the page if the page is new.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+ if (read(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not read file \"%s\": %s", dst, strerror(errno));
+ if (PageIsNew(page))
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Rewind the file position, so the first write will overwrite the page.
+ */
+ if (lseek(dst_fd, (dst_pageno % SLRU_PAGES_PER_SEGMENT) * BLCKSZ, SEEK_SET) == -1)
+ pg_fatal("could not seek in file \"%s\": %s", dst, strerror(errno));
+
+ src_buf_offset = 0;
+ while (src_buf_offset < src_len)
+ {
+ len_to_copy = Min(src_len - src_buf_offset, SizeOfPageContents - dst_offset);
+ memcpy(PageGetContents(page) + dst_offset, src_buf + src_buf_offset, len_to_copy);
+ src_buf_offset += len_to_copy;
+
+ if (new_cluster.controldata.data_checksum_version > 0)
+ ((PageHeader) page)->pd_checksum = pg_checksum_page(page, dst_pageno);
+ if (write(dst_fd, page, BLCKSZ) == -1)
+ pg_fatal("could not write file \"%s\": %s", dst, strerror(errno));
+
+ dst_pageno++;
+ dst_offset = 0;
+ PageInitSLRU(page, BLCKSZ, 0);
+
+ /*
+ * Switch segments if we reached the end of the current segment.
+ */
+ if (dst_pageno % SLRU_PAGES_PER_SEGMENT == 0)
+ {
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ dst_segno++;
+ snprintf(dst, sizeof(dst), "%s/%0*X", dst_dir, seg_name_len, dst_segno);
+ if ((dst_fd = open(dst, O_RDWR | O_CREAT | PG_BINARY, S_IRUSR | S_IWUSR)) == -1)
+ pg_fatal("could not open file \"%s\": %s", dst, strerror(errno));
+ if (ftruncate(dst_fd, SEGMENT_SIZE) == -1)
+ pg_fatal("could not truncate file \"%s\": %s", dst, strerror(errno));
+ }
+ }
+
+ if (fsync(dst_fd) == -1)
+ pg_fatal("could not fsync file \"%s\": %s", dst, strerror(errno));
+ if (close(dst_fd) == -1)
+ pg_fatal("could not close file \"%s\": %s", dst, strerror(errno));
+
+ pg_free(src_buf);
+ close(src_fd);
+}
+
+/*
+ * Upgrade the clog files to add a page header to each SLRU page.
+ */
+void
+upgrade_xact_cache(const char *src_subdir, const char *dst_subdir)
+{
+ char src_dir[MAXPGPATH];
+ char dst_dir[MAXPGPATH];
+
+ DIR *src_dirp;
+ struct dirent *src_dirent;
+
+ snprintf(src_dir, sizeof(src_dir), "%s/%s", old_cluster.pgdata, src_subdir);
+ snprintf(dst_dir, sizeof(dst_dir), "%s/%s", new_cluster.pgdata, dst_subdir);
+
+ if ((src_dirp = opendir(src_dir)) == NULL)
+ pg_fatal("could not open directory \"%s\": %s", src_dir, strerror(errno));
+
+ while (errno = 0, (src_dirent = readdir(src_dirp)) != NULL)
+ {
+ if (segment_file_filter(src_dirent))
+ upgrade_file(src_dir, src_dirent->d_name, dst_dir);
+ }
+
+ if (closedir(src_dirp) != 0)
+ pg_fatal("could not close directory \"%s\": %s", src_dir, strerror(errno));
+}
diff --git a/src/bin/pg_upgrade/pg_upgrade.c b/src/bin/pg_upgrade/pg_upgrade.c
index bb261353bd..7472c44f8d 100644
--- a/src/bin/pg_upgrade/pg_upgrade.c
+++ b/src/bin/pg_upgrade/pg_upgrade.c
@@ -674,14 +674,28 @@ copy_subdir_files(const char *old_subdir, const char *new_subdir)
static void
copy_xact_xlog_xid(void)
{
+ bool slru_header_changed = false;
+
/*
* Copy old commit logs to new data dir. pg_clog has been renamed to
* pg_xact in post-10 clusters.
*/
- copy_subdir_files(GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact",
- GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ?
- "pg_clog" : "pg_xact");
+ char *xact_old_dir = GET_MAJOR_VERSION(old_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+ char *xact_new_dir = GET_MAJOR_VERSION(new_cluster.major_version) <= 906 ? "pg_clog" : "pg_xact";
+
+ /*
+ * In post-17 clusters, a page header is added to each SLRU page.
+ * Perform a one-time conversion of the clog files if the old
+ * cluster and the new cluster use different SLRU formats.
+ */
+ if (new_cluster.controldata.cat_ver >= SLRU_PAGE_HEADER_CAT_VER &&
+ old_cluster.controldata.cat_ver < SLRU_PAGE_HEADER_CAT_VER)
+ slru_header_changed = true;
+
+ if (slru_header_changed)
+ upgrade_xact_cache(xact_old_dir, xact_new_dir);
+ else
+ copy_subdir_files(xact_old_dir, xact_new_dir);
prep_status("Setting oldest XID for new cluster");
exec_prog(UTILITY_LOG_FILE, NULL, true, true,
@@ -716,7 +730,8 @@ copy_xact_xlog_xid(void)
* server doesn't attempt to read multis older than the cutoff value.
*/
if (old_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
- new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER &&
+ !slru_header_changed)
{
copy_subdir_files("pg_multixact/offsets", "pg_multixact/offsets");
copy_subdir_files("pg_multixact/members", "pg_multixact/members");
@@ -736,7 +751,8 @@ copy_xact_xlog_xid(void)
new_cluster.pgdata);
check_ok();
}
- else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER)
+ else if (new_cluster.controldata.cat_ver >= MULTIXACT_FORMATCHANGE_CAT_VER ||
+ slru_header_changed)
{
/*
* Remove offsets/0000 file created by initdb that no longer matches
diff --git a/src/bin/pg_upgrade/pg_upgrade.h b/src/bin/pg_upgrade/pg_upgrade.h
index c0bfb002d2..989f428e97 100644
--- a/src/bin/pg_upgrade/pg_upgrade.h
+++ b/src/bin/pg_upgrade/pg_upgrade.h
@@ -115,6 +115,11 @@ extern char *output_files[];
*/
#define MULTIXACT_FORMATCHANGE_CAT_VER 201301231
+/*
+ * A page header was added to each SLRU page in 17.0.
+ */
+#define SLRU_PAGE_HEADER_CAT_VER 202403111
+
/*
* large object chunk size added to pg_controldata,
* commit 5f93c37805e7485488480916b4585e098d3cc883
@@ -412,6 +417,7 @@ void rewriteVisibilityMap(const char *fromfile, const char *tofile,
void check_file_clone(void);
void check_copy_file_range(void);
void check_hard_link(void);
+void upgrade_xact_cache(const char *src_subdir, const char *dst_subdir);
/* fopen_priv() is no longer different from fopen() */
#define fopen_priv(path, mode) fopen(path, mode)
diff --git a/src/include/catalog/catversion.h b/src/include/catalog/catversion.h
index f085221155..0c8349f50e 100644
--- a/src/include/catalog/catversion.h
+++ b/src/include/catalog/catversion.h
@@ -57,6 +57,6 @@
*/
/* yyyymmddN */
-#define CATALOG_VERSION_NO 202403091
+#define CATALOG_VERSION_NO 202403111
#endif
diff --git a/src/include/storage/bufpage.h b/src/include/storage/bufpage.h
index d0df02d39c..2dc83451a6 100644
--- a/src/include/storage/bufpage.h
+++ b/src/include/storage/bufpage.h
@@ -201,6 +201,7 @@ typedef PageHeaderData *PageHeader;
* handling pages.
*/
#define PG_PAGE_LAYOUT_VERSION 4
+#define PG_SLRU_PAGE_LAYOUT_VERSION 1
#define PG_DATA_CHECKSUM_VERSION 1
/* ----------------------------------------------------------------
@@ -257,6 +258,11 @@ PageGetContents(Page page)
return (char *) page + MAXALIGN(SizeOfPageHeaderData);
}
+/*
+ * Space available for storing page contents.
+ */
+#define SizeOfPageContents (BLCKSZ - MAXALIGN(SizeOfPageHeaderData))
+
/* ----------------
* functions to access page size info
* ----------------
@@ -486,6 +492,7 @@ StaticAssertDecl(BLCKSZ == ((BLCKSZ / sizeof(size_t)) * sizeof(size_t)),
"BLCKSZ has to be a multiple of sizeof(size_t)");
extern void PageInit(Page page, Size pageSize, Size specialSize);
+extern void PageInitSLRU(Page page, Size pageSize, Size specialSize);
extern bool PageIsVerifiedExtended(Page page, BlockNumber blkno, int flags);
extern OffsetNumber PageAddItemExtended(Page page, Item item, Size size,
OffsetNumber offsetNumber, int flags);
diff --git a/src/test/modules/test_slru/test_slru.c b/src/test/modules/test_slru/test_slru.c
index 068a21f125..06cf7656f7 100644
--- a/src/test/modules/test_slru/test_slru.c
+++ b/src/test/modules/test_slru/test_slru.c
@@ -17,6 +17,7 @@
#include "access/slru.h"
#include "access/transam.h"
#include "miscadmin.h"
+#include "storage/bufpage.h"
#include "storage/fd.h"
#include "storage/ipc.h"
#include "storage/shmem.h"
@@ -72,8 +73,8 @@ test_slru_page_write(PG_FUNCTION_ARGS)
TestSlruCtl->shared->page_status[slotno] = SLRU_PAGE_VALID;
/* write given data to the page, up to the limit of the page */
- strncpy(TestSlruCtl->shared->page_buffer[slotno], data,
- BLCKSZ - 1);
+ strncpy(PageGetContents(TestSlruCtl->shared->page_buffer[slotno]), data,
+ SizeOfPageContents - 1);
SimpleLruWritePage(TestSlruCtl, slotno);
LWLockRelease(lock);
@@ -101,7 +102,7 @@ test_slru_page_read(PG_FUNCTION_ARGS)
LWLockAcquire(lock, LW_EXCLUSIVE);
slotno = SimpleLruReadPage(TestSlruCtl, pageno,
write_ok, InvalidTransactionId);
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(lock);
PG_RETURN_TEXT_P(cstring_to_text(data));
@@ -120,7 +121,7 @@ test_slru_page_readonly(PG_FUNCTION_ARGS)
pageno,
InvalidTransactionId);
Assert(LWLockHeldByMe(lock));
- data = (char *) TestSlruCtl->shared->page_buffer[slotno];
+ data = (char *) PageGetContents(TestSlruCtl->shared->page_buffer[slotno]);
LWLockRelease(lock);
PG_RETURN_TEXT_P(cstring_to_text(data));
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-03-14 22:27 ` Jeff Davis <[email protected]>
2024-03-19 06:48 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Jeff Davis @ 2024-03-14 22:27 UTC (permalink / raw)
To: Li, Yong <[email protected]>; +Cc: Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
On Mon, 2024-03-11 at 10:01 +0000, Li, Yong wrote:
> - The clog LSN group has been brought back.
> Now the page LSN on each clog page is used for honoring the write-
> ahead rule
> and it is always the highest LSN of all the LSN groups on the page.
> The LSN groups are used by TransactionIdGetStatus() as before.
I like where this is going.
Álvaro, do you still see a problem with this approach?
> - New comments have been added to pg_upgrade to mention the SLRU
> page header change as the reason for upgrading clog files.
That seems reasonable, but were any alternatives discussed? Do we have
consensus that this is the right thing to do?
And if we use this approach, is there extra validation or testing that
can be done?
Regards,
Jeff Davis
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
@ 2024-03-19 06:48 ` Li, Yong <[email protected]>
2024-06-10 07:19 ` Re: Proposal to add page headers to SLRU pages Bertrand Drouvot <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2024-03-19 06:48 UTC (permalink / raw)
To: Jeff Davis <[email protected]>; +Cc: Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Debnath, Shawn <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
>> - New comments have been added to pg_upgrade to mention the SLRU
>> page header change as the reason for upgrading clog files.
>
> That seems reasonable, but were any alternatives discussed? Do we have
> consensus that this is the right thing to do?
In general, there are two approaches. Either we convert the existing clog files,
or we don’t. The patch chooses to convert.
If we don’t, then the clog file code must be able to handle both formats. For,
XIDs in the range where the clog is written in the old format, segment and offset
computation must be done in one way, and for XIDs in a different range, it must
be computed in a different way. To avoid changing the format in the middle of a
page, which must not happen, the new format must start from a clean page,
possibly in a clean new segment. If the database is extremely small and has only
a few transactions on the first page of clog, then we must either convert the whole
page (effectively the whole clog file), or we must skip the rest of the XIDs on the
page and ask the database to start from XIDs on the second page on restart.
Also, we need to consider where to store the cut-off XID and when to remove it.
All these details feel very complex and error prone to me. Performing a one-time
conversion is the most efficient and straightforward approach to me.
>
> And if we use this approach, is there extra validation or testing that
> can be done?
>
> Regards,
> Jeff Davis
Unfortunately, the test requires a setup of two different versions of PG. I am not
aware of an existing test infrastructure which can run automated tests using two
PGs. I did manually verify the output of pg_upgrade.
Regards,
Yong
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-19 06:48 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-06-10 07:19 ` Bertrand Drouvot <[email protected]>
2024-06-10 08:01 ` Re: Proposal to add page headers to SLRU pages Michael Paquier <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Bertrand Drouvot @ 2024-06-10 07:19 UTC (permalink / raw)
To: Li, Yong <[email protected]>; +Cc: Jeff Davis <[email protected]>; Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
Hi,
On Tue, Mar 19, 2024 at 06:48:33AM +0000, Li, Yong wrote:
>
> Unfortunately, the test requires a setup of two different versions of PG. I am not
> aware of an existing test infrastructure which can run automated tests using two
> PGs. I did manually verify the output of pg_upgrade.
I think there is something in t/002_pg_upgrade.pl (see src/bin/pg_upgrade/TESTING),
that could be used to run automated tests using an old and a current version.
Regards,
--
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-19 06:48 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-06-10 07:19 ` Re: Proposal to add page headers to SLRU pages Bertrand Drouvot <[email protected]>
@ 2024-06-10 08:01 ` Michael Paquier <[email protected]>
2024-06-13 08:41 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Michael Paquier @ 2024-06-10 08:01 UTC (permalink / raw)
To: Bertrand Drouvot <[email protected]>; +Cc: Li, Yong <[email protected]>; Jeff Davis <[email protected]>; Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
On Mon, Jun 10, 2024 at 07:19:56AM +0000, Bertrand Drouvot wrote:
> On Tue, Mar 19, 2024 at 06:48:33AM +0000, Li, Yong wrote:
>> Unfortunately, the test requires a setup of two different versions of PG. I am not
>> aware of an existing test infrastructure which can run automated tests using two
>> PGs. I did manually verify the output of pg_upgrade.
>
> I think there is something in t/002_pg_upgrade.pl (see src/bin/pg_upgrade/TESTING),
> that could be used to run automated tests using an old and a current version.
Cluster.pm relies on install_path for stuff, where it is possible to
create tests with multiple nodes pointing to different installation
paths. This allows mixing nodes with different build options, or just
different major versions like pg_upgrade's perl tests.
--
Michael
Attachments:
[application/pgp-signature] signature.asc (833B, ../../[email protected]/2-signature.asc)
download
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-19 06:48 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-06-10 07:19 ` Re: Proposal to add page headers to SLRU pages Bertrand Drouvot <[email protected]>
2024-06-10 08:01 ` Re: Proposal to add page headers to SLRU pages Michael Paquier <[email protected]>
@ 2024-06-13 08:41 ` Li, Yong <[email protected]>
2024-11-09 17:32 ` Re: Proposal to add page headers to SLRU pages Rishu Bagga <[email protected]>
0 siblings, 1 reply; 29+ messages in thread
From: Li, Yong @ 2024-06-13 08:41 UTC (permalink / raw)
To: Michael Paquier <[email protected]>; +Cc: Bertrand Drouvot <[email protected]>; Jeff Davis <[email protected]>; Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
> On Jun 10, 2024, at 16:01, Michael Paquier <[email protected]> wrote:
>
> External Email
>
> From: Michael Paquier <[email protected]>
> Subject: Re: Proposal to add page headers to SLRU pages
> Date: June 10, 2024 at 16:01:50 GMT+8
> To: Bertrand Drouvot <[email protected]>
> Cc: "Li, Yong" <[email protected]>, Jeff Davis <[email protected]>, Aleksander Alekseev <[email protected]>, PostgreSQL Hackers <[email protected]>, "Bagga, Rishu" <[email protected]>, Robert Haas <[email protected]>, Andrey Borodin <[email protected]>, "Shyrabokau, Anton" <[email protected]>
>
>
> On Mon, Jun 10, 2024 at 07:19:56AM +0000, Bertrand Drouvot wrote:
>> On Tue, Mar 19, 2024 at 06:48:33AM +0000, Li, Yong wrote:
>>> Unfortunately, the test requires a setup of two different versions of PG. I am not
>>> aware of an existing test infrastructure which can run automated tests using two
>>> PGs. I did manually verify the output of pg_upgrade.
>>
>> I think there is something in t/002_pg_upgrade.pl (see src/bin/pg_upgrade/TESTING),
>> that could be used to run automated tests using an old and a current version.
>
> Cluster.pm relies on install_path for stuff, where it is possible to
> create tests with multiple nodes pointing to different installation
> paths. This allows mixing nodes with different build options, or just
> different major versions like pg_upgrade's perl tests.
> —
> Michael
>
>
Thanks for pointing this out. Here is what I have tried:
1. Manually build and install PostgreSQL from the latest source code.
2. Following the instructions from src/bin/pg_upgrade to manually dump the regression database.
3. Apply the patch to the latest code, and build from the source.
4. Run “make check” by following the instructions from src/bin/pg_upgrade and setting up the olddump and oldinstall to point to the “old” installation used in step 2.
All tests pass.
Yong
^ permalink raw reply [nested|flat] 29+ messages in thread
* Re: Proposal to add page headers to SLRU pages
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-19 06:48 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-06-10 07:19 ` Re: Proposal to add page headers to SLRU pages Bertrand Drouvot <[email protected]>
2024-06-10 08:01 ` Re: Proposal to add page headers to SLRU pages Michael Paquier <[email protected]>
2024-06-13 08:41 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
@ 2024-11-09 17:32 ` Rishu Bagga <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Rishu Bagga @ 2024-11-09 17:32 UTC (permalink / raw)
To: Li, Yong <[email protected]>; +Cc: Michael Paquier <[email protected]>; Bertrand Drouvot <[email protected]>; Jeff Davis <[email protected]>; Aleksander Alekseev <[email protected]>; PostgreSQL Hackers <[email protected]>; Bagga, Rishu <[email protected]>; Robert Haas <[email protected]>; Andrey Borodin <[email protected]>; Shyrabokau, Anton <[email protected]>
> Thanks for pointing this out. Here is what I have tried:
> 1. Manually build and install PostgreSQL from the latest source code.
> 2. Following the instructions from src/bin/pg_upgrade to manually dump the regression
> database.
> 3. Apply the patch to the latest code, and build from the source.
> 4. Run “make check” by following the instructions from src/bin/pg_upgrade and setting up
> the olddump and oldinstall to point to the “old” installation used in step 2.
> All tests pass.
Hi all,
Following up on this. What remaining work do we need to do to get this in?
Thanks,
Rishu Bagga
On Sat, Nov 9, 2024 at 8:50 AM Li, Yong <[email protected]> wrote:
>
>
>
> > On Jun 10, 2024, at 16:01, Michael Paquier <[email protected]> wrote:
> >
> > External Email
> >
> > From: Michael Paquier <[email protected]>
> > Subject: Re: Proposal to add page headers to SLRU pages
> > Date: June 10, 2024 at 16:01:50 GMT+8
> > To: Bertrand Drouvot <[email protected]>
> > Cc: "Li, Yong" <[email protected]>, Jeff Davis <[email protected]>, Aleksander Alekseev <[email protected]>, PostgreSQL Hackers <[email protected]>, "Bagga, Rishu" <[email protected]>, Robert Haas <[email protected]>, Andrey Borodin <[email protected]>, "Shyrabokau, Anton" <[email protected]>
> >
> >
> > On Mon, Jun 10, 2024 at 07:19:56AM +0000, Bertrand Drouvot wrote:
> >> On Tue, Mar 19, 2024 at 06:48:33AM +0000, Li, Yong wrote:
> >>> Unfortunately, the test requires a setup of two different versions of PG. I am not
> >>> aware of an existing test infrastructure which can run automated tests using two
> >>> PGs. I did manually verify the output of pg_upgrade.
> >>
> >> I think there is something in t/002_pg_upgrade.pl (see src/bin/pg_upgrade/TESTING),
> >> that could be used to run automated tests using an old and a current version.
> >
> > Cluster.pm relies on install_path for stuff, where it is possible to
> > create tests with multiple nodes pointing to different installation
> > paths. This allows mixing nodes with different build options, or just
> > different major versions like pg_upgrade's perl tests.
> > —
> > Michael
> >
> >
>
> Thanks for pointing this out. Here is what I have tried:
> 1. Manually build and install PostgreSQL from the latest source code.
> 2. Following the instructions from src/bin/pg_upgrade to manually dump the regression database.
> 3. Apply the patch to the latest code, and build from the source.
> 4. Run “make check” by following the instructions from src/bin/pg_upgrade and setting up the olddump and oldinstall to point to the “old” installation used in step 2.
>
> All tests pass.
>
>
> Yong
>
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH] Add hook for plugins to acquire sample rows during ANALYZE
@ 2026-05-29 12:35 Samba Siva <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Samba Siva @ 2026-05-29 12:35 UTC (permalink / raw)
- Introduced AcquireSampleRowsFunc_hook for extensions to override row sampling.
- Updated analyze.c to utilize the hook if registered.
- Added tests to ensure ANALYZE completes without errors with the new hook.
---
doc/src/sgml/xfunc.sgml | 10 ++++++++++
src/backend/commands/analyze.c | 24 ++++++++++++++++++++----
src/include/commands/vacuum.h | 11 +++++++++++
src/test/regress/expected/analyze.out | 20 ++++++++++++++++++++
src/test/regress/parallel_schedule | 1 +
src/test/regress/sql/analyze.sql | 24 ++++++++++++++++++++++++
6 files changed, 86 insertions(+), 4 deletions(-)
create mode 100644 src/test/regress/expected/analyze.out
create mode 100644 src/test/regress/sql/analyze.sql
diff --git a/doc/src/sgml/xfunc.sgml b/doc/src/sgml/xfunc.sgml
index 1eb5abffd8..f1ef025ae2 100644
--- a/doc/src/sgml/xfunc.sgml
+++ b/doc/src/sgml/xfunc.sgml
@@ -4203,4 +4203,14 @@ supportfn(internal) returns internal
To create such conditions, the support function must implement
the <literal>SupportRequestIndexCondition</literal> request type.
</para>
+
+ <para>
+ The hook variable <varname>AcquireSampleRowsFunc_hook</varname> allows extensions
+ to override the row sampling function during <command>ANALYZE</command> for regular
+ heap relations. This is useful for extensions/systems which implement distributed
+ databases and want to sample rows from remote nodes instead of the local heap.
+ The hook function fills the provided <literal>rows[]</literal> buffer with at most
+ <literal>targrows</literal> heap tuples and sets <literal>*totalrows</literal> to the
+ estimated total live row count of the relation.
+ </para>
</sect1>
diff --git a/src/backend/commands/analyze.c b/src/backend/commands/analyze.c
index 4fffb76e55..3560acdff1 100644
--- a/src/backend/commands/analyze.c
+++ b/src/backend/commands/analyze.c
@@ -74,6 +74,8 @@ int default_statistics_target = 100;
static MemoryContext anl_context = NULL;
static BufferAccessStrategy vac_strategy;
+/* Hook for plugins to acquire sample rows for ANALYZE */
+AcquireSampleRowsFunc_hook_type AcquireSampleRowsFunc_hook = NULL;
static void do_analyze_rel(Relation onerel,
VacuumParams *params, List *va_cols,
@@ -188,8 +190,15 @@ analyze_rel(Oid relid, RangeVar *relation,
if (onerel->rd_rel->relkind == RELKIND_RELATION ||
onerel->rd_rel->relkind == RELKIND_MATVIEW)
{
- /* Regular table, so we'll use the regular row acquisition function */
- acquirefunc = acquire_sample_rows;
+ /*
+ * Regular table, so we'll use the regular row acquisition function.
+ * If a plugin has registered a hook to acquire sample rows, use it;
+ * otherwise use the default function.
+ */
+ if (AcquireSampleRowsFunc_hook)
+ acquirefunc = AcquireSampleRowsFunc_hook;
+ else
+ acquirefunc = acquire_sample_rows;
/* Also get regular table's size */
relpages = RelationGetNumberOfBlocks(onerel);
}
@@ -1467,8 +1476,15 @@ acquire_inherited_sample_rows(Relation onerel, int elevel,
if (childrel->rd_rel->relkind == RELKIND_RELATION ||
childrel->rd_rel->relkind == RELKIND_MATVIEW)
{
- /* Regular table, so use the regular row acquisition function */
- acquirefunc = acquire_sample_rows;
+ /*
+ * Regular table, so use the regular row acquisition function.
+ * If a plugin has registered a hook to acquire sample rows, use it;
+ * otherwise use the default function.
+ */
+ if (AcquireSampleRowsFunc_hook)
+ acquirefunc = AcquireSampleRowsFunc_hook;
+ else
+ acquirefunc = acquire_sample_rows;
relpages = RelationGetNumberOfBlocks(childrel);
}
else if (childrel->rd_rel->relkind == RELKIND_FOREIGN_TABLE)
diff --git a/src/include/commands/vacuum.h b/src/include/commands/vacuum.h
index bc37a80dc7..146f936861 100644
--- a/src/include/commands/vacuum.h
+++ b/src/include/commands/vacuum.h
@@ -21,6 +21,7 @@
#include "catalog/pg_class.h"
#include "catalog/pg_statistic.h"
#include "catalog/pg_type.h"
+#include "foreign/fdwapi.h"
#include "parser/parse_node.h"
#include "storage/buf.h"
#include "storage/lock.h"
@@ -113,6 +114,9 @@ typedef void (*AnalyzeAttrComputeStatsFunc) (VacAttrStatsP stats,
int samplerows,
double totalrows);
+/* Hook type for plugins to acquire sample rows for ANALYZE */
+typedef AcquireSampleRowsFunc AcquireSampleRowsFunc_hook_type;
+
typedef struct VacAttrStats
{
/*
@@ -334,6 +338,13 @@ extern PGDLLIMPORT int vacuum_cost_limit;
extern PGDLLIMPORT int64 parallel_vacuum_worker_delay_ns;
+/*
+ * Hook for plugins to override row sampling during ANALYZE.
+ * Also applies to child relations of partitioned/inherited tables.
+ * See acquire_sample_rows() in src/backend/commands/analyze.c.
+ */
+extern PGDLLIMPORT AcquireSampleRowsFunc_hook_type AcquireSampleRowsFunc_hook;
+
/* in commands/vacuum.c */
extern void ExecVacuum(ParseState *pstate, VacuumStmt *vacstmt, bool isTopLevel);
extern void vacuum(List *relations, VacuumParams *params,
diff --git a/src/test/regress/expected/analyze.out b/src/test/regress/expected/analyze.out
new file mode 100644
index 0000000000..5bb6bbc45b
--- /dev/null
+++ b/src/test/regress/expected/analyze.out
@@ -0,0 +1,20 @@
+-- Test AcquireSampleRowsFunc_hook
+-- Usually this would be tested via a C extension.
+-- Here we just confirm this does not break the existing ANALYZE code
+-- by verifying that ANALYZE completes without error.
+CREATE TABLE employees (
+ id SERIAL PRIMARY KEY,
+ name TEXT,
+ department TEXT,
+ salary NUMERIC
+);
+INSERT INTO employees
+ SELECT
+ i,
+ 'Employee ' || i,
+ 'Department ' || (i % 5),
+ (i % 100) * 1000 + 50000
+ FROM generate_series(1, 1000) i;
+-- Should complete without error
+ANALYZE employees;
+DROP TABLE employees;
diff --git a/src/test/regress/parallel_schedule b/src/test/regress/parallel_schedule
index e1e0c54019..71978ef858 100644
--- a/src/test/regress/parallel_schedule
+++ b/src/test/regress/parallel_schedule
@@ -94,6 +94,7 @@ test: vacuum_parallel
# Run this alone, because concurrent DROP TABLE would make non-superuser
# "ANALYZE;" fail with "relation with OID $n does not exist".
test: maintain_every
+test: analyze
# no relation related tests can be put in this group
test: publication subscription
diff --git a/src/test/regress/sql/analyze.sql b/src/test/regress/sql/analyze.sql
new file mode 100644
index 0000000000..e0bc84e9f9
--- /dev/null
+++ b/src/test/regress/sql/analyze.sql
@@ -0,0 +1,24 @@
+-- Test AcquireSampleRowsFunc_hook
+-- Usually this would be tested via a C extension.
+-- Here we just confirm this does not break the existing ANALYZE code
+-- by verifying that ANALYZE completes without error.
+
+CREATE TABLE employees (
+ id SERIAL PRIMARY KEY,
+ name TEXT,
+ department TEXT,
+ salary NUMERIC
+);
+
+INSERT INTO employees
+ SELECT
+ i,
+ 'Employee ' || i,
+ 'Department ' || (i % 5),
+ (i % 100) * 1000 + 50000
+ FROM generate_series(1, 1000) i;
+
+-- Should complete without error
+ANALYZE employees;
+
+DROP TABLE employees;
--
2.50.1 (Apple Git-155)
------=_Part_741589_445982825.1782438414800--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH] Add hook for plugins to acquire sample rows during ANALYZE
@ 2026-05-29 12:35 Samba Siva <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Samba Siva @ 2026-05-29 12:35 UTC (permalink / raw)
- Introduced AcquireSampleRowsFunc_hook for extensions to override row sampling.
- Updated analyze.c to utilize the hook if registered.
- Added tests to ensure ANALYZE completes without errors with the new hook.
---
doc/src/sgml/xfunc.sgml | 10 ++++++++++
src/backend/commands/analyze.c | 24 ++++++++++++++++++++----
src/include/commands/vacuum.h | 11 +++++++++++
src/test/regress/expected/analyze.out | 20 ++++++++++++++++++++
src/test/regress/parallel_schedule | 1 +
src/test/regress/sql/analyze.sql | 24 ++++++++++++++++++++++++
6 files changed, 86 insertions(+), 4 deletions(-)
create mode 100644 src/test/regress/expected/analyze.out
create mode 100644 src/test/regress/sql/analyze.sql
diff --git a/doc/src/sgml/xfunc.sgml b/doc/src/sgml/xfunc.sgml
index 1eb5abffd8..f1ef025ae2 100644
--- a/doc/src/sgml/xfunc.sgml
+++ b/doc/src/sgml/xfunc.sgml
@@ -4203,4 +4203,14 @@ supportfn(internal) returns internal
To create such conditions, the support function must implement
the <literal>SupportRequestIndexCondition</literal> request type.
</para>
+
+ <para>
+ The hook variable <varname>AcquireSampleRowsFunc_hook</varname> allows extensions
+ to override the row sampling function during <command>ANALYZE</command> for regular
+ heap relations. This is useful for extensions/systems which implement distributed
+ databases and want to sample rows from remote nodes instead of the local heap.
+ The hook function fills the provided <literal>rows[]</literal> buffer with at most
+ <literal>targrows</literal> heap tuples and sets <literal>*totalrows</literal> to the
+ estimated total live row count of the relation.
+ </para>
</sect1>
diff --git a/src/backend/commands/analyze.c b/src/backend/commands/analyze.c
index 4fffb76e55..3560acdff1 100644
--- a/src/backend/commands/analyze.c
+++ b/src/backend/commands/analyze.c
@@ -74,6 +74,8 @@ int default_statistics_target = 100;
static MemoryContext anl_context = NULL;
static BufferAccessStrategy vac_strategy;
+/* Hook for plugins to acquire sample rows for ANALYZE */
+AcquireSampleRowsFunc_hook_type AcquireSampleRowsFunc_hook = NULL;
static void do_analyze_rel(Relation onerel,
VacuumParams *params, List *va_cols,
@@ -188,8 +190,15 @@ analyze_rel(Oid relid, RangeVar *relation,
if (onerel->rd_rel->relkind == RELKIND_RELATION ||
onerel->rd_rel->relkind == RELKIND_MATVIEW)
{
- /* Regular table, so we'll use the regular row acquisition function */
- acquirefunc = acquire_sample_rows;
+ /*
+ * Regular table, so we'll use the regular row acquisition function.
+ * If a plugin has registered a hook to acquire sample rows, use it;
+ * otherwise use the default function.
+ */
+ if (AcquireSampleRowsFunc_hook)
+ acquirefunc = AcquireSampleRowsFunc_hook;
+ else
+ acquirefunc = acquire_sample_rows;
/* Also get regular table's size */
relpages = RelationGetNumberOfBlocks(onerel);
}
@@ -1467,8 +1476,15 @@ acquire_inherited_sample_rows(Relation onerel, int elevel,
if (childrel->rd_rel->relkind == RELKIND_RELATION ||
childrel->rd_rel->relkind == RELKIND_MATVIEW)
{
- /* Regular table, so use the regular row acquisition function */
- acquirefunc = acquire_sample_rows;
+ /*
+ * Regular table, so use the regular row acquisition function.
+ * If a plugin has registered a hook to acquire sample rows, use it;
+ * otherwise use the default function.
+ */
+ if (AcquireSampleRowsFunc_hook)
+ acquirefunc = AcquireSampleRowsFunc_hook;
+ else
+ acquirefunc = acquire_sample_rows;
relpages = RelationGetNumberOfBlocks(childrel);
}
else if (childrel->rd_rel->relkind == RELKIND_FOREIGN_TABLE)
diff --git a/src/include/commands/vacuum.h b/src/include/commands/vacuum.h
index bc37a80dc7..146f936861 100644
--- a/src/include/commands/vacuum.h
+++ b/src/include/commands/vacuum.h
@@ -21,6 +21,7 @@
#include "catalog/pg_class.h"
#include "catalog/pg_statistic.h"
#include "catalog/pg_type.h"
+#include "foreign/fdwapi.h"
#include "parser/parse_node.h"
#include "storage/buf.h"
#include "storage/lock.h"
@@ -113,6 +114,9 @@ typedef void (*AnalyzeAttrComputeStatsFunc) (VacAttrStatsP stats,
int samplerows,
double totalrows);
+/* Hook type for plugins to acquire sample rows for ANALYZE */
+typedef AcquireSampleRowsFunc AcquireSampleRowsFunc_hook_type;
+
typedef struct VacAttrStats
{
/*
@@ -334,6 +338,13 @@ extern PGDLLIMPORT int vacuum_cost_limit;
extern PGDLLIMPORT int64 parallel_vacuum_worker_delay_ns;
+/*
+ * Hook for plugins to override row sampling during ANALYZE.
+ * Also applies to child relations of partitioned/inherited tables.
+ * See acquire_sample_rows() in src/backend/commands/analyze.c.
+ */
+extern PGDLLIMPORT AcquireSampleRowsFunc_hook_type AcquireSampleRowsFunc_hook;
+
/* in commands/vacuum.c */
extern void ExecVacuum(ParseState *pstate, VacuumStmt *vacstmt, bool isTopLevel);
extern void vacuum(List *relations, VacuumParams *params,
diff --git a/src/test/regress/expected/analyze.out b/src/test/regress/expected/analyze.out
new file mode 100644
index 0000000000..5bb6bbc45b
--- /dev/null
+++ b/src/test/regress/expected/analyze.out
@@ -0,0 +1,20 @@
+-- Test AcquireSampleRowsFunc_hook
+-- Usually this would be tested via a C extension.
+-- Here we just confirm this does not break the existing ANALYZE code
+-- by verifying that ANALYZE completes without error.
+CREATE TABLE employees (
+ id SERIAL PRIMARY KEY,
+ name TEXT,
+ department TEXT,
+ salary NUMERIC
+);
+INSERT INTO employees
+ SELECT
+ i,
+ 'Employee ' || i,
+ 'Department ' || (i % 5),
+ (i % 100) * 1000 + 50000
+ FROM generate_series(1, 1000) i;
+-- Should complete without error
+ANALYZE employees;
+DROP TABLE employees;
diff --git a/src/test/regress/parallel_schedule b/src/test/regress/parallel_schedule
index e1e0c54019..71978ef858 100644
--- a/src/test/regress/parallel_schedule
+++ b/src/test/regress/parallel_schedule
@@ -94,6 +94,7 @@ test: vacuum_parallel
# Run this alone, because concurrent DROP TABLE would make non-superuser
# "ANALYZE;" fail with "relation with OID $n does not exist".
test: maintain_every
+test: analyze
# no relation related tests can be put in this group
test: publication subscription
diff --git a/src/test/regress/sql/analyze.sql b/src/test/regress/sql/analyze.sql
new file mode 100644
index 0000000000..e0bc84e9f9
--- /dev/null
+++ b/src/test/regress/sql/analyze.sql
@@ -0,0 +1,24 @@
+-- Test AcquireSampleRowsFunc_hook
+-- Usually this would be tested via a C extension.
+-- Here we just confirm this does not break the existing ANALYZE code
+-- by verifying that ANALYZE completes without error.
+
+CREATE TABLE employees (
+ id SERIAL PRIMARY KEY,
+ name TEXT,
+ department TEXT,
+ salary NUMERIC
+);
+
+INSERT INTO employees
+ SELECT
+ i,
+ 'Employee ' || i,
+ 'Department ' || (i % 5),
+ (i % 100) * 1000 + 50000
+ FROM generate_series(1, 1000) i;
+
+-- Should complete without error
+ANALYZE employees;
+
+DROP TABLE employees;
--
2.50.1 (Apple Git-155)
------=_Part_741589_445982825.1782438414800--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH] Add hook for plugins to acquire sample rows during ANALYZE
@ 2026-05-29 12:35 Samba Siva <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Samba Siva @ 2026-05-29 12:35 UTC (permalink / raw)
- Introduced AcquireSampleRowsFunc_hook for extensions to override row sampling.
- Updated analyze.c to utilize the hook if registered.
- Added tests to ensure ANALYZE completes without errors with the new hook.
---
doc/src/sgml/xfunc.sgml | 10 ++++++++++
src/backend/commands/analyze.c | 24 ++++++++++++++++++++----
src/include/commands/vacuum.h | 11 +++++++++++
src/test/regress/expected/analyze.out | 20 ++++++++++++++++++++
src/test/regress/parallel_schedule | 1 +
src/test/regress/sql/analyze.sql | 24 ++++++++++++++++++++++++
6 files changed, 86 insertions(+), 4 deletions(-)
create mode 100644 src/test/regress/expected/analyze.out
create mode 100644 src/test/regress/sql/analyze.sql
diff --git a/doc/src/sgml/xfunc.sgml b/doc/src/sgml/xfunc.sgml
index 1eb5abffd8..f1ef025ae2 100644
--- a/doc/src/sgml/xfunc.sgml
+++ b/doc/src/sgml/xfunc.sgml
@@ -4203,4 +4203,14 @@ supportfn(internal) returns internal
To create such conditions, the support function must implement
the <literal>SupportRequestIndexCondition</literal> request type.
</para>
+
+ <para>
+ The hook variable <varname>AcquireSampleRowsFunc_hook</varname> allows extensions
+ to override the row sampling function during <command>ANALYZE</command> for regular
+ heap relations. This is useful for extensions/systems which implement distributed
+ databases and want to sample rows from remote nodes instead of the local heap.
+ The hook function fills the provided <literal>rows[]</literal> buffer with at most
+ <literal>targrows</literal> heap tuples and sets <literal>*totalrows</literal> to the
+ estimated total live row count of the relation.
+ </para>
</sect1>
diff --git a/src/backend/commands/analyze.c b/src/backend/commands/analyze.c
index 4fffb76e55..3560acdff1 100644
--- a/src/backend/commands/analyze.c
+++ b/src/backend/commands/analyze.c
@@ -74,6 +74,8 @@ int default_statistics_target = 100;
static MemoryContext anl_context = NULL;
static BufferAccessStrategy vac_strategy;
+/* Hook for plugins to acquire sample rows for ANALYZE */
+AcquireSampleRowsFunc_hook_type AcquireSampleRowsFunc_hook = NULL;
static void do_analyze_rel(Relation onerel,
VacuumParams *params, List *va_cols,
@@ -188,8 +190,15 @@ analyze_rel(Oid relid, RangeVar *relation,
if (onerel->rd_rel->relkind == RELKIND_RELATION ||
onerel->rd_rel->relkind == RELKIND_MATVIEW)
{
- /* Regular table, so we'll use the regular row acquisition function */
- acquirefunc = acquire_sample_rows;
+ /*
+ * Regular table, so we'll use the regular row acquisition function.
+ * If a plugin has registered a hook to acquire sample rows, use it;
+ * otherwise use the default function.
+ */
+ if (AcquireSampleRowsFunc_hook)
+ acquirefunc = AcquireSampleRowsFunc_hook;
+ else
+ acquirefunc = acquire_sample_rows;
/* Also get regular table's size */
relpages = RelationGetNumberOfBlocks(onerel);
}
@@ -1467,8 +1476,15 @@ acquire_inherited_sample_rows(Relation onerel, int elevel,
if (childrel->rd_rel->relkind == RELKIND_RELATION ||
childrel->rd_rel->relkind == RELKIND_MATVIEW)
{
- /* Regular table, so use the regular row acquisition function */
- acquirefunc = acquire_sample_rows;
+ /*
+ * Regular table, so use the regular row acquisition function.
+ * If a plugin has registered a hook to acquire sample rows, use it;
+ * otherwise use the default function.
+ */
+ if (AcquireSampleRowsFunc_hook)
+ acquirefunc = AcquireSampleRowsFunc_hook;
+ else
+ acquirefunc = acquire_sample_rows;
relpages = RelationGetNumberOfBlocks(childrel);
}
else if (childrel->rd_rel->relkind == RELKIND_FOREIGN_TABLE)
diff --git a/src/include/commands/vacuum.h b/src/include/commands/vacuum.h
index bc37a80dc7..146f936861 100644
--- a/src/include/commands/vacuum.h
+++ b/src/include/commands/vacuum.h
@@ -21,6 +21,7 @@
#include "catalog/pg_class.h"
#include "catalog/pg_statistic.h"
#include "catalog/pg_type.h"
+#include "foreign/fdwapi.h"
#include "parser/parse_node.h"
#include "storage/buf.h"
#include "storage/lock.h"
@@ -113,6 +114,9 @@ typedef void (*AnalyzeAttrComputeStatsFunc) (VacAttrStatsP stats,
int samplerows,
double totalrows);
+/* Hook type for plugins to acquire sample rows for ANALYZE */
+typedef AcquireSampleRowsFunc AcquireSampleRowsFunc_hook_type;
+
typedef struct VacAttrStats
{
/*
@@ -334,6 +338,13 @@ extern PGDLLIMPORT int vacuum_cost_limit;
extern PGDLLIMPORT int64 parallel_vacuum_worker_delay_ns;
+/*
+ * Hook for plugins to override row sampling during ANALYZE.
+ * Also applies to child relations of partitioned/inherited tables.
+ * See acquire_sample_rows() in src/backend/commands/analyze.c.
+ */
+extern PGDLLIMPORT AcquireSampleRowsFunc_hook_type AcquireSampleRowsFunc_hook;
+
/* in commands/vacuum.c */
extern void ExecVacuum(ParseState *pstate, VacuumStmt *vacstmt, bool isTopLevel);
extern void vacuum(List *relations, VacuumParams *params,
diff --git a/src/test/regress/expected/analyze.out b/src/test/regress/expected/analyze.out
new file mode 100644
index 0000000000..5bb6bbc45b
--- /dev/null
+++ b/src/test/regress/expected/analyze.out
@@ -0,0 +1,20 @@
+-- Test AcquireSampleRowsFunc_hook
+-- Usually this would be tested via a C extension.
+-- Here we just confirm this does not break the existing ANALYZE code
+-- by verifying that ANALYZE completes without error.
+CREATE TABLE employees (
+ id SERIAL PRIMARY KEY,
+ name TEXT,
+ department TEXT,
+ salary NUMERIC
+);
+INSERT INTO employees
+ SELECT
+ i,
+ 'Employee ' || i,
+ 'Department ' || (i % 5),
+ (i % 100) * 1000 + 50000
+ FROM generate_series(1, 1000) i;
+-- Should complete without error
+ANALYZE employees;
+DROP TABLE employees;
diff --git a/src/test/regress/parallel_schedule b/src/test/regress/parallel_schedule
index e1e0c54019..71978ef858 100644
--- a/src/test/regress/parallel_schedule
+++ b/src/test/regress/parallel_schedule
@@ -94,6 +94,7 @@ test: vacuum_parallel
# Run this alone, because concurrent DROP TABLE would make non-superuser
# "ANALYZE;" fail with "relation with OID $n does not exist".
test: maintain_every
+test: analyze
# no relation related tests can be put in this group
test: publication subscription
diff --git a/src/test/regress/sql/analyze.sql b/src/test/regress/sql/analyze.sql
new file mode 100644
index 0000000000..e0bc84e9f9
--- /dev/null
+++ b/src/test/regress/sql/analyze.sql
@@ -0,0 +1,24 @@
+-- Test AcquireSampleRowsFunc_hook
+-- Usually this would be tested via a C extension.
+-- Here we just confirm this does not break the existing ANALYZE code
+-- by verifying that ANALYZE completes without error.
+
+CREATE TABLE employees (
+ id SERIAL PRIMARY KEY,
+ name TEXT,
+ department TEXT,
+ salary NUMERIC
+);
+
+INSERT INTO employees
+ SELECT
+ i,
+ 'Employee ' || i,
+ 'Department ' || (i % 5),
+ (i % 100) * 1000 + 50000
+ FROM generate_series(1, 1000) i;
+
+-- Should complete without error
+ANALYZE employees;
+
+DROP TABLE employees;
--
2.50.1 (Apple Git-155)
------=_Part_741589_445982825.1782438414800--
^ permalink raw reply [nested|flat] 29+ messages in thread
* [PATCH v1] Mark object class descriptions for translation
@ 2026-07-03 06:06 Kyotaro Horiguchi <[email protected]>
0 siblings, 0 replies; 29+ messages in thread
From: Kyotaro Horiguchi @ 2026-07-03 06:06 UTC (permalink / raw)
ObjectPropertyType.class_descr was originally intended for internal
error messages, but is now also used in a translated error message.
Mark the object class descriptions for translation and translate the
return value of get_object_class_descr().
---
src/backend/catalog/objectaddress.c | 88 ++++++++++++++---------------
src/backend/catalog/pg_depend.c | 2 +-
2 files changed, 45 insertions(+), 45 deletions(-)
diff --git a/src/backend/catalog/objectaddress.c b/src/backend/catalog/objectaddress.c
index af0e4703616..12dc76b712a 100644
--- a/src/backend/catalog/objectaddress.c
+++ b/src/backend/catalog/objectaddress.c
@@ -100,8 +100,8 @@
*/
typedef struct
{
- const char *class_descr; /* string describing the catalog, for internal
- * error messages */
+ const char *class_descr; /* string describing the catalog, for error
+ * messages */
Oid class_oid; /* oid of catalog */
Oid oid_index_oid; /* oid of index on system oid column */
SysCacheIdentifier oid_catcache_id; /* id of catcache on system oid column */
@@ -124,7 +124,7 @@ typedef struct
static const ObjectPropertyType ObjectProperty[] =
{
{
- "access method",
+ gettext_noop("access method"),
AccessMethodRelationId,
AmOidIndexId,
AMOID,
@@ -138,7 +138,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "access method operator",
+ gettext_noop("access method operator"),
AccessMethodOperatorRelationId,
AccessMethodOperatorOidIndexId,
SYSCACHEID_INVALID,
@@ -152,7 +152,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "access method procedure",
+ gettext_noop("access method procedure"),
AccessMethodProcedureRelationId,
AccessMethodProcedureOidIndexId,
SYSCACHEID_INVALID,
@@ -166,7 +166,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "cast",
+ gettext_noop("cast"),
CastRelationId,
CastOidIndexId,
SYSCACHEID_INVALID,
@@ -180,7 +180,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "collation",
+ gettext_noop("collation"),
CollationRelationId,
CollationOidIndexId,
COLLOID,
@@ -194,7 +194,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "constraint",
+ gettext_noop("constraint"),
ConstraintRelationId,
ConstraintOidIndexId,
CONSTROID,
@@ -208,7 +208,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "conversion",
+ gettext_noop("conversion"),
ConversionRelationId,
ConversionOidIndexId,
CONVOID,
@@ -222,7 +222,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "database",
+ gettext_noop("database"),
DatabaseRelationId,
DatabaseOidIndexId,
DATABASEOID,
@@ -236,7 +236,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "default ACL",
+ gettext_noop("default ACL"),
DefaultAclRelationId,
DefaultAclOidIndexId,
SYSCACHEID_INVALID,
@@ -250,7 +250,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "extension",
+ gettext_noop("extension"),
ExtensionRelationId,
ExtensionOidIndexId,
SYSCACHEID_INVALID,
@@ -264,7 +264,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "foreign-data wrapper",
+ gettext_noop("foreign-data wrapper"),
ForeignDataWrapperRelationId,
ForeignDataWrapperOidIndexId,
FOREIGNDATAWRAPPEROID,
@@ -278,7 +278,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "foreign server",
+ gettext_noop("foreign server"),
ForeignServerRelationId,
ForeignServerOidIndexId,
FOREIGNSERVEROID,
@@ -292,7 +292,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "function",
+ gettext_noop("function"),
ProcedureRelationId,
ProcedureOidIndexId,
PROCOID,
@@ -306,7 +306,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "language",
+ gettext_noop("language"),
LanguageRelationId,
LanguageOidIndexId,
LANGOID,
@@ -320,7 +320,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "large object metadata",
+ gettext_noop("large object metadata"),
LargeObjectMetadataRelationId,
LargeObjectMetadataOidIndexId,
SYSCACHEID_INVALID,
@@ -334,7 +334,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "operator class",
+ gettext_noop("operator class"),
OperatorClassRelationId,
OpclassOidIndexId,
CLAOID,
@@ -348,7 +348,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "operator",
+ gettext_noop("operator"),
OperatorRelationId,
OperatorOidIndexId,
OPEROID,
@@ -362,7 +362,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "operator family",
+ gettext_noop("operator family"),
OperatorFamilyRelationId,
OpfamilyOidIndexId,
OPFAMILYOID,
@@ -376,7 +376,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "property graph element",
+ gettext_noop("property graph element"),
PropgraphElementRelationId,
PropgraphElementObjectIndexId,
PROPGRAPHELOID,
@@ -390,7 +390,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "property graph element label",
+ gettext_noop("property graph element label"),
PropgraphElementLabelRelationId,
PropgraphElementLabelObjectIndexId,
-1,
@@ -404,7 +404,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "property graph label",
+ gettext_noop("property graph label"),
PropgraphLabelRelationId,
PropgraphLabelObjectIndexId,
PROPGRAPHLABELOID,
@@ -418,7 +418,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "property graph label property",
+ gettext_noop("property graph label property"),
PropgraphLabelPropertyRelationId,
PropgraphLabelPropertyObjectIndexId,
-1,
@@ -432,7 +432,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "property graph property",
+ gettext_noop("property graph property"),
PropgraphPropertyRelationId,
PropgraphPropertyObjectIndexId,
-1,
@@ -446,7 +446,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "role",
+ gettext_noop("role"),
AuthIdRelationId,
AuthIdOidIndexId,
AUTHOID,
@@ -460,7 +460,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "role membership",
+ gettext_noop("role membership"),
AuthMemRelationId,
AuthMemOidIndexId,
SYSCACHEID_INVALID,
@@ -474,7 +474,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "rule",
+ gettext_noop("rule"),
RewriteRelationId,
RewriteOidIndexId,
SYSCACHEID_INVALID,
@@ -488,7 +488,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "schema",
+ gettext_noop("schema"),
NamespaceRelationId,
NamespaceOidIndexId,
NAMESPACEOID,
@@ -502,7 +502,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "relation",
+ gettext_noop("relation"),
RelationRelationId,
ClassOidIndexId,
RELOID,
@@ -516,7 +516,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "tablespace",
+ gettext_noop("tablespace"),
TableSpaceRelationId,
TablespaceOidIndexId,
TABLESPACEOID,
@@ -530,7 +530,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "transform",
+ gettext_noop("transform"),
TransformRelationId,
TransformOidIndexId,
TRFOID,
@@ -544,7 +544,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "trigger",
+ gettext_noop("trigger"),
TriggerRelationId,
TriggerOidIndexId,
SYSCACHEID_INVALID,
@@ -558,7 +558,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "policy",
+ gettext_noop("policy"),
PolicyRelationId,
PolicyOidIndexId,
SYSCACHEID_INVALID,
@@ -572,7 +572,7 @@ static const ObjectPropertyType ObjectProperty[] =
false
},
{
- "event trigger",
+ gettext_noop("event trigger"),
EventTriggerRelationId,
EventTriggerOidIndexId,
EVENTTRIGGEROID,
@@ -586,7 +586,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "text search configuration",
+ gettext_noop("text search configuration"),
TSConfigRelationId,
TSConfigOidIndexId,
TSCONFIGOID,
@@ -600,7 +600,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "text search dictionary",
+ gettext_noop("text search dictionary"),
TSDictionaryRelationId,
TSDictionaryOidIndexId,
TSDICTOID,
@@ -614,7 +614,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "text search parser",
+ gettext_noop("text search parser"),
TSParserRelationId,
TSParserOidIndexId,
TSPARSEROID,
@@ -628,7 +628,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "text search template",
+ gettext_noop("text search template"),
TSTemplateRelationId,
TSTemplateOidIndexId,
TSTEMPLATEOID,
@@ -642,7 +642,7 @@ static const ObjectPropertyType ObjectProperty[] =
true,
},
{
- "type",
+ gettext_noop("type"),
TypeRelationId,
TypeOidIndexId,
TYPEOID,
@@ -656,7 +656,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "publication",
+ gettext_noop("publication"),
PublicationRelationId,
PublicationObjectIndexId,
PUBLICATIONOID,
@@ -670,7 +670,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "subscription",
+ gettext_noop("subscription"),
SubscriptionRelationId,
SubscriptionObjectIndexId,
SUBSCRIPTIONOID,
@@ -684,7 +684,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "extended statistics",
+ gettext_noop("extended statistics"),
StatisticExtRelationId,
StatisticExtOidIndexId,
STATEXTOID,
@@ -698,7 +698,7 @@ static const ObjectPropertyType ObjectProperty[] =
true
},
{
- "user mapping",
+ gettext_noop("user mapping"),
UserMappingRelationId,
UserMappingOidIndexId,
USERMAPPINGOID,
diff --git a/src/backend/catalog/pg_depend.c b/src/backend/catalog/pg_depend.c
index 9a7a401aced..0d28d34412d 100644
--- a/src/backend/catalog/pg_depend.c
+++ b/src/backend/catalog/pg_depend.c
@@ -810,7 +810,7 @@ dependencyLockAndCheckObject(Oid classId, Oid objectId)
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("referenced %s was concurrently dropped",
- get_object_class_descr(classId))));
+ _(get_object_class_descr(classId)))));
systable_endscan(scan);
table_close(rel, AccessShareLock);
--
2.52.0
----Next_Part(Fri_Jul__3_15_37_42_2026_665)----
^ permalink raw reply [nested|flat] 29+ messages in thread
end of thread, other threads:[~2026-07-03 06:06 UTC | newest]
Thread overview: 29+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2019-02-20 11:35 [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely Etsuro Fujita <[email protected]>
2019-03-06 11:37 [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely Etsuro Fujita <[email protected]>
2019-03-07 09:49 [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely Etsuro Fujita <[email protected]>
2019-03-20 11:34 [PATCH 3/3] postgres_fdw: Perform UPPERREL_FINAL step remotely Etsuro Fujita <[email protected]>
2019-03-27 07:43 [PATCH 4/4] postgres_fdw: Perform UPPERREL_FINAL step remotely Etsuro Fujita <[email protected]>
2023-12-07 10:19 Re: Proposal to add page headers to SLRU pages Andrey M. Borodin <[email protected]>
2023-12-07 14:16 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2023-12-07 17:32 ` Re: Proposal to add page headers to SLRU pages Andrey Borodin <[email protected]>
2023-12-07 18:27 ` Re: Proposal to add page headers to SLRU pages Debnath, Shawn <[email protected]>
2023-12-07 19:51 ` Re: Proposal to add page headers to SLRU pages Robert Haas <[email protected]>
2023-12-08 09:35 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2023-12-19 02:23 ` Re: Proposal to add page headers to SLRU pages Bagga, Rishu <[email protected]>
2023-12-19 07:28 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-02 11:35 ` Re: Proposal to add page headers to SLRU pages Aleksander Alekseev <[email protected]>
2024-01-04 15:57 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-01-16 09:12 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-06 12:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-08 21:22 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-11 10:01 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-03-14 22:27 ` Re: Proposal to add page headers to SLRU pages Jeff Davis <[email protected]>
2024-03-19 06:48 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-06-10 07:19 ` Re: Proposal to add page headers to SLRU pages Bertrand Drouvot <[email protected]>
2024-06-10 08:01 ` Re: Proposal to add page headers to SLRU pages Michael Paquier <[email protected]>
2024-06-13 08:41 ` Re: Proposal to add page headers to SLRU pages Li, Yong <[email protected]>
2024-11-09 17:32 ` Re: Proposal to add page headers to SLRU pages Rishu Bagga <[email protected]>
2026-05-29 12:35 [PATCH] Add hook for plugins to acquire sample rows during ANALYZE Samba Siva <[email protected]>
2026-05-29 12:35 [PATCH] Add hook for plugins to acquire sample rows during ANALYZE Samba Siva <[email protected]>
2026-05-29 12:35 [PATCH] Add hook for plugins to acquire sample rows during ANALYZE Samba Siva <[email protected]>
2026-07-03 06:06 [PATCH v1] Mark object class descriptions for translation Kyotaro Horiguchi <[email protected]>
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox