Re: pgv18: simple table scan take more time than pgv14 - Mailing list pgsql-hackers

From Devrim Gündüz
Subject Re: pgv18: simple table scan take more time than pgv14
Date
Msg-id 7c45639130261bbfdb20084e78380ed1f7283006.camel@gunduz.org
Whole thread Raw
List pgsql-hackers
Hi,

On Fri, 2025-06-20 at 09:46 +0800, James Pang wrote:
> same OS RHEL8, install pgv14.11 and pgv18.beta1_3, both installation
> by RPM from pgdg, and use similar postgresql.conf.

<snip>

> 18.beta1
> postgres=# \timing on
> Timing is on.
> postgres=# select * from tt where b ~~ 'a%';
>  a | b
> ---+---
> (0 rows)
>
> Time: 78.532 ms
> postgres=# select * from tt where b ~~ 'a%'
> postgres-# ;
>  a | b
> ---+---
> (0 rows)
>
> Time: 83.516 ms
> postgres=# select * from tt where b ~~ 'a%';
>  a | b
> ---+---
> (0 rows)

It's likely because v18 beta RPMs are built with  --enable-debug and 
--enable-cassert . See the note on yum.postgresql.org:

https://yum.postgresql.org/rpmchart/#pg18

Regards,



--
Devrim Gündüz
Open Source Solution Architect, PostgreSQL Major Contributor
BlueSky: @devrim.gunduz.org , @gunduz.org

Attachment

pgsql-hackers by date:

Previous
From: jian he
Date:
Subject: Re: pg18: Virtual generated columns are not (yet) safe when superuser selects from them
Next
From: Merlin Moncure
Date:
Subject: Re: [PATCH] Proposal to Enable/Disable Index using ALTER INDEX