Message-ID: <63fd7f0d-2d49-4952-b1ec-fa877bbf0765@tantorlabs.com>
Date: Mon, 20 Jan 2025 19:14:07 +0300
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: pg_stat_statements: improve loading and saving routines for the
 dump file
To: pgsql-hackers@lists.postgresql.org
References: <192f7185defa370d083e3a237727b066@postgrespro.ru>
Content-Language: en-US
From: Ivan Kush <ivan.kush@tantorlabs.com>
Organization: Tantor Labs LLC
In-Reply-To: <192f7185defa370d083e3a237727b066@postgrespro.ru>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: <https://www.postgresql.org/message-id/63fd7f0d-2d49-4952-b1ec-fa877bbf0765%40tantorlabs.com>
Precedence: bulk

Hello, Mikhail.

1) I'd add to comment a reason, why mutex should be last.

// Mutex should be last field, as this field isn't copied to dump file

+	/* protects the counters only. Should be the very last field, as this field isn't copied to dump file
+	slock_t		mutex;
  } pgssEntry;

2) You didn't take into account the upgrade. Saved in Postgres with this 
byte and try to load in version without this byte.

On 1/20/25 16:49, m.litsarev@postgrespro.ru wrote:
> Hi!
>
> Currently in pg_stat_statements save/load routines the whole pgssEntry 
> entity data are written/read with its last field
> slock_t        mutex;
> which is actually not used then.
> This small patch fixes this issue. Hope, it will be useful.
>
> Respectfully,
>
> Mikhail Litsarev,
> Postgres Professional: https://postgrespro.com

-- 
Best wishes,
Ivan Kush
Tantor Labs LLC