Re: Segment fault when excuting SPI function On PG with commit 41c6a5be - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Segment fault when excuting SPI function On PG with commit 41c6a5be
Date
Msg-id YQPyqYrMy/6JTAKa@paquier.xyz
Whole thread Raw
In response to Segment fault when excuting SPI function On PG with commit 41c6a5be  ("liuhuailing@fujitsu.com" <liuhuailing@fujitsu.com>)
Responses Re: Segment fault when excuting SPI function On PG with commit 41c6a5be  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Fri, Jul 30, 2021 at 08:57:35AM +0000, liuhuailing@fujitsu.com wrote:
> When I used SPI_execute_plan function on PG12 which commit 41c6a5be is used,
> Segment fault occurred.
>
> PS: If commit 41c6a5be is not used, this phenomenon will not happen.

I see nothing wrong in this commit, FWIW.

> Reproduce:
> In a background process, the following steps are executed.
> --------------------------
> StartTransactionCommand();
> SPI_connect();       
> plan = SPI_prepare(query,0,NULL); ★the query is a SELECT SQL.
> SPI_keepplan(plan);        
> SPI_finish();           
> CommitTransactionCommand();   
> StartTransactionCommand();    
> SPI_connect();
> SPI_execute_plan(plan, NULL, NULL, true, 0); ★Segment fault
> --------------------------

But I see an issue with your code.  It seems to me that you should
push a snapshot before doing SPI_prepare() and SPI_execute_plan(),
as of:
PushActiveSnapshot(GetTransactionSnapshot());
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Ranier Vilela
Date:
Subject: Re: Segment fault when excuting SPI function On PG with commit 41c6a5be
Next
From: Robert Haas
Date:
Subject: Re: needless complexity in StartupXLOG