RE: [PATHC] Fix minor memory leak in pg_basebackup - Mailing list pgsql-hackers

From Zhang, Jie
Subject RE: [PATHC] Fix minor memory leak in pg_basebackup
Date
Msg-id 343665364037409dafcda21e96c6d9f4@G08CNEXMBPEKD06.g08.fujitsu.local
Whole thread Raw
In response to Re: [PATHC] Fix minor memory leak in pg_basebackup  (Michael Paquier <michael@paquier.xyz>)
Responses Re: [PATHC] Fix minor memory leak in pg_basebackup  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
Hi  Michael

so much the better!

-----Original Message-----
From: Michael Paquier [mailto:michael@paquier.xyz] 
Sent: Thursday, April 16, 2020 2:31 PM
To: Zhang, Jie/张 杰 <zhangjie2@cn.fujitsu.com>
Cc: pgsql-hackers@lists.postgresql.org
Subject: Re: [PATHC] Fix minor memory leak in pg_basebackup

On Wed, Apr 15, 2020 at 10:06:52AM +0000, Zhang, Jie wrote:
> In some cases , PGresult is not cleared.
> 
> File: src\bin\pg_basebackup\streamutil.c
> 
> bool
> RetrieveWalSegSize(PGconn *conn)
> {
>     PGresult   *res;

RetrieveWalSegSize() gets called only once at the beginning of pg_basebackup and pg_receivewal, so that's not an issue
thathas major effects, still that's an issue.  The first one PQclear() is needed where you say.  Now for the second
one,I would just move it once the code is done with the query result, aka after calling PQgetvalue().
 
What do you think?  Please see the attached.
--
Michael



pgsql-hackers by date:

Previous
From: Richard Guo
Date:
Subject: Re: sqlsmith crash incremental sort
Next
From: Daniel Gustafsson
Date:
Subject: Making openssl_tls_init_hook OpenSSL specific