Re: Shared hash table allocations

From: Heikki Linnakangas <hlinnaka(at)iki(dot)fi>
To: Ashutosh Bapat <ashutosh(dot)bapat(dot)oss(at)gmail(dot)com>
Cc: Tomas Vondra <tomas(at)vondra(dot)me>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>, Robert Haas <robertmhaas(at)gmail(dot)com>, Rahila Syed <rahilasyed90(at)gmail(dot)com>, Matthias van de Meent <boekewurm+postgres(at)gmail(dot)com>
Subject: Re: Shared hash table allocations
Date: 2026-04-03 17:32:07
Message-ID: d581882f-5c34-4f8d-ac01-4f7cbd851149@iki.fi
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 03/04/2026 16:03, Ashutosh Bapat wrote:
> On Thu, Apr 2, 2026 at 10:15 PM Heikki Linnakangas <hlinnaka(at)iki(dot)fi> wrote:
>>> I think the release notes should "nudge" all the
>>> users who use non-default max_locks_per_transaction to increase it if
>>> they see "out of memory" errors. I don't think it should provide a
>>> blanket advise to double their locks
>>
>> How about:
>>
>> "If you had previously set max_locks_per_transaction, you might need to
>> set it to a higher value in v19 to avoid "out of shared memory" errors.
>> If you are unsure what to set it to and don't mind the increased memory
>> usage, you can double the value to ensure that you can acquire at least
>> as many locks as before"
>
> The wiggle room is 100KB fixed + 10% of other two structures, so value
> by which it should be increased is partly fixed and partly a multiple
> of current value. "double the value" is simplest advice we can give.
> +1.

Ok, committed these patches to remove the safety margins, make LOCK and
PROCLOCK fixed-size, and change the default to
max_locks_per_transaction=128. I will do one final self-review of the
remaining earlier patches from this thread next; I believe they're ready
to be committed too.

Thanks for the review!

- Heikki

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Daniel Gustafsson 2026-04-03 17:56:46 Re: Changing the state of data checksums in a running cluster
Previous Message Andres Freund 2026-04-03 17:30:01 Re: AIO / read stream heuristics adjustments for index prefetching