Date: Sat, 17 Jul 2021 07:03:01 +0200 (CEST)
From: Fabien COELHO <coelho@cri.ensmp.fr>
To: Yugo NAGATA <nagata@sraoss.co.jp>
cc: pgsql-hackers@postgresql.org
Subject: Re: pgbench: using prepared BEGIN statement in a pipeline could
 cause an error
In-Reply-To: <20210716153013.fc53b1c780b06fccc07a7f0d@sraoss.co.jp>
Message-ID: <alpine.DEB.2.22.394.2107170650260.3143659@pseudo>
References: <20210716153013.fc53b1c780b06fccc07a7f0d@sraoss.co.jp>
User-Agent: Alpine 2.22 (DEB 394 2020-01-19)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Archived-At: 
 <https://www.postgresql.org/message-id/alpine.DEB.2.22.394.2107170650260.3143659%40pseudo>
Precedence: bulk


Hello Yugo-san,

> [...] One way to avoid these errors is to send Parse messages before 
> pipeline mode starts. I attached a patch to fix to prepare commands at 
> starting of a script instead of at the first execution of the command.

> What do you think?

ISTM that moving prepare out of command execution is a good idea, so I'm 
in favor of this approach: the code is simpler and cleaner.

ISTM that a minor impact is that the preparation is not counted in the 
command performance statistics. I do not think that it is a problem, even 
if it would change detailed results under -C -r -M prepared.

Patch applies & compiles cleanly, global & local make check ok. However 
the issue is not tested. I think that the patch should add a tap test case 
for the problem being addressed.

I'd suggest to move the statement preparation call in the 
CSTATE_CHOOSE_SCRIPT case.

In comments: not yet -> needed.

-- 
Fabien.