Horton

Thread

Commits

  1. Use MAX_PARALLEL_WORKER_LIMIT for max_parallel_maintenance_workers

  1. Use MAX_PARALLEL_WORKER_LIMIT consistently in guc_tables.c

    Matthias van de Meent <boekewurm@gmail.com> — 2024-10-09T12:34:55Z

    Hi,

Whilst doing some digging in parallel code, I noticed that
max_parallel_maintenance_workers is registered as guc with a manual
value of 1024, while max_parallel_workers_per_gather uses
MAX_PARALLEL_WORKER_LIMIT (also 1024). After some archeology, the
discrepancy seems to have existed ever since
max_parallel_maintenance_workers was originally introduced, as the
patch development that introduced the GUC that eventually got
committed predates the use of MAX_PARALLEL_WORKER_LIMIT in guc.c (now
guc_tables.c), and the change to the definition of sibling GUCs of
max_parallel_workers and max_parallel_workers_per_gather wasn't
noticed during that development.

PFA a trivial one-line patch that makes that a bit more consistent.

Kind regards,

Matthias van de Meent

  2. Re: Use MAX_PARALLEL_WORKER_LIMIT consistently in guc_tables.c

    Zhang Mingli <zmlpostgres@gmail.com> — 2024-10-09T14:28:05Z

    Hi,


Zhang Mingli
www.hashdata.xyz
On Oct 9, 2024 at 20:35 +0800, Matthias van de Meent <boekewurm@gmail.com>, wrote:
> Hi,
>
> Whilst doing some digging in parallel code, I noticed that
> max_parallel_maintenance_workers is registered as guc with a manual
> value of 1024, while max_parallel_workers_per_gather uses
> MAX_PARALLEL_WORKER_LIMIT (also 1024). After some archeology, the
> discrepancy seems to have existed ever since
> max_parallel_maintenance_workers was originally introduced, as the
> patch development that introduced the GUC that eventually got
> committed predates the use of MAX_PARALLEL_WORKER_LIMIT in guc.c (now
> guc_tables.c), and the change to the definition of sibling GUCs of
> max_parallel_workers and max_parallel_workers_per_gather wasn't
> noticed during that development.
>
> PFA a trivial one-line patch that makes that a bit more consistent.
LGTM.

  3. Re: Use MAX_PARALLEL_WORKER_LIMIT consistently in guc_tables.c

    Michael Paquier <michael@paquier.xyz> — 2024-10-09T23:51:40Z

    On Wed, Oct 09, 2024 at 10:28:05PM +0800, Zhang Mingli wrote:
> On Oct 9, 2024 at 20:35 +0800, Matthias van de Meent <boekewurm@gmail.com>, wrote:
>> PFA a trivial one-line patch that makes that a bit more consistent.
>
> LGTM.

Makes sense to me as well.  Thanks for the patch.
--
Michael