On Mon, 15 Apr 2024 at 10:33, Andres Freund <and...@anarazel.de> wrote:
> - The new bump allocator has a fair amount of uncovered functionality:
>
> https://anarazel.de/postgres/cov/16-vs-HEAD-2024-04-14/src/backend/utils/mmgr/bump.c.gcov.html#L293
The attached adds a test to tuplesort to exercise BumpAllocLarge()
>
> https://anarazel.de/postgres/cov/16-vs-HEAD-2024-04-14/src/backend/utils/mmgr/bump.c.gcov.html#L613
I don't see a way to exercise those. They're meant to be "can't
happen" ERRORs. I could delete them and use BogusFree, BogusRealloc,
BogusGetChunkContext, BogusGetChunkSpace instead, but the ERROR
message would be misleading. I think it's best just to leave this.
David
diff --git a/src/test/regress/expected/tuplesort.out
b/src/test/regress/expected/tuplesort.out
index 0e8b5bf4a3..6dd97e7427 100644
--- a/src/test/regress/expected/tuplesort.out
+++ b/src/test/regress/expected/tuplesort.out
@@ -343,6 +343,19 @@ ORDER BY ctid DESC LIMIT 5;
(5 rows)
ROLLBACK;
+----
+-- test sorting of large datums VALUES
+----
+-- Ensure the order is correct and values look intact
+SELECT LEFT(a,10),b FROM
+ (VALUES(REPEAT('a', 512 * 1024),1),(REPEAT('b', 512 * 1024),2)) v(a,b)
+ORDER BY v.a DESC;
+ left | b
+------------+---
+ bbbbbbbbbb | 2
+ aaaaaaaaaa | 1
+(2 rows)
+
----
-- test forward and backward scans for in-memory and disk based tuplesort
----
diff --git a/src/test/regress/sql/tuplesort.sql
b/src/test/regress/sql/tuplesort.sql
index 658fe98dc5..8476e594e6 100644
--- a/src/test/regress/sql/tuplesort.sql
+++ b/src/test/regress/sql/tuplesort.sql
@@ -146,6 +146,15 @@ FROM abbrev_abort_uuids
ORDER BY ctid DESC LIMIT 5;
ROLLBACK;
+----
+-- test sorting of large datums VALUES
+----
+
+-- Ensure the order is correct and values look intact
+SELECT LEFT(a,10),b FROM
+ (VALUES(REPEAT('a', 512 * 1024),1),(REPEAT('b', 512 * 1024),2)) v(a,b)
+ORDER BY v.a DESC;
+
----
-- test forward and backward scans for in-memory and disk based tuplesort
----
