Mar 2023

1 commit
Ignore BRIN indexes when checking for HOT updates
20 Mar 2023Tomas Vondramaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1Also in: [PATCH] Don't block HOT update by BRIN index (not this thread)
LLM analysis, 100% suremodified before commitlight changesfrom patchset #161 of 2 reviewers saw the final version

Both attachments in #16 apply to the commit's parent, leaving only assertion cleanup in src/backend/catalog/indexing.c, comments in src/backend/utils/cache/relcache.c, and equivalent enum declarations in src/include/access/tableam.h, consistent with #17 and #18. This March 2023 submission supersedes the November 2021 source #491786 claimed by thread #44477, whose attribution I disagree with because this thread contains the latest matching implementation.

  • Tomas Vondrasaw the final versiontrailerTomas questioned tuple-header signaling in #11, reviewed predicate handling in #13, supplied final cleanup in #16, and committed the result as confirmed in #18.
  • Alvaro Herrerano review foundtrailerAlvaro is credited in 19d8e2308b, but no review activity by him appears in this thread.

Jun 2022

1 commit
Revert changes in HOT handling of BRIN indexes
16 Jun 2022Tomas Vondramaster15beta2REL_19_STABLE15beta2REL_18_STABLE15beta2REL_17_STABLE15beta2REL_16_STABLE15beta2REL_15_STABLE15beta2revert
LLM analysis, 100% surelanded as postedfrom patchset #73 of 3 reviewers saw the final version

The committed diff matches #7 after excluding patch metadata and hunk positions. Tomas confirms pushing the revert in #9.

  • Andres Freundsaw the final versionLLM analysis, 86% sureBefore a revert attachment existed, Andres recommended reverting and retrying for PostgreSQL 16 in #5, the approach Tomas adopted in #7.
  • Robert Haassaw the final versionLLM analysis, 88% sureRobert endorsed reverting and explained why allowing HOT still requires updating BRIN summaries in #6, supporting the approach adopted in #7.
  • Michael Paquiersaw the final versionLLM analysis, 100% sureMichael explicitly approved removing the IndexAmRoutine field despite the beta-period ABI change in #8.

Credits across 2 commits

Author
Matthias van de MeentLLM1
Tomas VondraLLM1
Reviewer
Alvaro Herrera1
Tomas Vondra1
Reported by
Tomas Vondra1
Committer
Tomas Vondra2

N LLM means N of the credits come from LLM analysis, not from commit trailers.