[PATCH] Implement dynamic predicate lock ratio limits - Mailing list pgsql-hackers

From Arkady Skvorcov
Subject [PATCH] Implement dynamic predicate lock ratio limits
Date
Msg-id CAChax7z=RyVL=4RCHX3OnHF9n++fRk-2gQPOdebk2W7iPpZTEQ@mail.gmail.com
Whole thread Raw
List pgsql-hackers
Hi,

This patch series implements the TODO in predicate.c that suggests adding 
ratio-based predicate lock limits.
(Resending with patch attachment, cause previous send was missing the patch file)
The series is split into two logical commits:
1. Core implementation of the GUC variables and predicate lock logic
2. Comprehensive regression tests

Patch overview:
- Adds max_predicate_locks_per_relation_ratio and max_predicate_locks_per_page_ratio GUCs
- When ratios are > 0, calculates lock limits based on relation size estimates
- Maintains full backward compatibility (default 0.0 = use existing fixed limits)
- Includes complete regression test coverage

This addresses the TODO comment in MaxPredicateChildLocks() for more
intelligent lock allocation in mixed workloads.
Looking forward to your review! ArkadySkv
Attachment

pgsql-hackers by date:

Previous
From: vignesh C
Date:
Subject: Re: Logical Replication of sequences
Next
From: Robert Haas
Date:
Subject: Re: apply_scanjoin_target_to_paths and partitionwise join