Re: tweak to a few index tests to hits ambuildempty() routine.

From: a(dot)kozhemyakin(at)postgrespro(dot)ru
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: tweak to a few index tests to hits ambuildempty() routine.
Date: 2022-09-12 08:47:09
Message-ID: b00c8ee096ee46cd25c183125562a1a7@postgrespro.ru
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

The commit 4fb5c794e5 eliminates the ginbulkdelete() test coverage
provided by the commit 4c51a2d1e4 two years ago.
With the following Assert added:
@@ -571,7 +571,7 @@ ginbulkdelete(IndexVacuumInfo *info,
IndexBulkDeleteResult *stats,
Buffer buffer;
BlockNumber rootOfPostingTree[BLCKSZ / (sizeof(IndexTupleData) +
sizeof(ItemId))];
uint32 nRoot;
-
+ Assert(0);
gvs.tmpCxt = AllocSetContextCreate(CurrentMemoryContext,

"Gin vacuum temporary context",

ALLOCSET_DEFAULT_SIZES);
I have check-world passed successfully.

Amul Sul писал 2021-11-29 12:04:
> Hi,
>
> Attached patch is doing small changes to brin, gin & gist index tests
> to use an unlogged table without changing the original intention of
> those tests and that is able to hit ambuildempty() routing which is
> otherwise not reachable by the current tests.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Alvaro Herrera 2022-09-12 09:09:13 Re: pg15b4: FailedAssertion("TransactionIdIsValid(xmax)
Previous Message Julien Rouhaud 2022-09-12 08:40:47 Re: pg_stat_statements locking