Apr 2023

2 commits
Document BaseBackupSync and BaseBackupWrite wait events.
12 Apr 2023Robert Haasmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1REL_15_STABLE15.3
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #2311 of 1 reviewer saw the final version

The monitoring.sgml additions exactly match the first component of #231; the shell error-check patch is not included.

  • Thomas Munrosaw the final versionLLM analysis, 95% sureThe commit credits Thomas’s missing-wait-event documentation report, represented in the thread by #229.
Support long distance matching for zstd compression
6 Apr 2023Tomas Vondramaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1Also in: zstd compression for pg_dump (part of a posted patchset)

Mar 2022

12 commits
Fix possible NULL-pointer-deference in backup_compression.c.
30 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #2132 of 2 reviewers saw the final version

The commit keeps the vend/kwend guard from #213, drops its deliberately uncommittable example keyword and expands the comment; it does not use the alternative condition in #215.

  • Tom Lanesaw the final versionLLM analysis, 99% sureTom reviewed the fix and requested an explanatory comment in #214 and #218.
  • Justin Pryzbysaw the final versionLLM analysis, 99% sureJustin reproduced the valueless-option crash with long mode and discussed the fix in #215 and #217.
Allow parallel zstd compression when taking a base backup.
30 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 99% surepart of a posted patchsetlight changesfrom patchset #2150 of 2 reviewers saw the final version

#215 republishes Robert’s final parallel-ZSTD patch from #212; the commit takes that component with the run_log prerequisite already split out.

  • Andres Freundsaw an earlier versionLLM analysis, 97% sureAndres examined asynchronous worker lifetime and cancellation risks in #193 and #198, before parameter handling was revised.
  • Dagfinn Ilmari Mannsåkersaw an earlier versionLLM analysis, 95% sureIlmari reviewed test-output handling in #201, before later client/server option-handling revisions.
Make PostgreSQL::Test::Cluster::run_log() return a useful value.
29 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 99% surepart of a posted patchsetlight changesfrom patchset #215

The three-line run_log return-value hunk is present in Robert’s parallel patch republished in #215 and was committed separately; #220 came afterward.

Fix a few goofs in new backup compression code.
28 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #2061 of 1 reviewer saw the final version

The first component of #206 exactly republishes Robert’s #203 fix for compression defaults, ZSTD error checks and failed-context exit handling.

  • Dipesh Panditsaw the final versionLLM analysis, 100% sureDipesh approved the compression-default and error-handling fixes in #204.
Replace BASE_BACKUP COMPRESSION_LEVEL option with COMPRESSION_DETAIL.
23 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 99% suremodified before commitlight changesfrom patchset #1902 of 3 reviewers saw the final version

Compared with #190, the commit corrects SGML option-tag boundaries and omits typedef-list additions; common compression-detail parsing and validation match.

  • Justin Pryzbysaw the final versionLLM analysis, 99% sureJustin found parser and validation failures and the final typo in #176, #180 and #187, after which only light changes followed.
  • Dagfinn Ilmari Mannsåkersaw the final versionLLM analysis, 100% sureIlmari corrected the compression-option documentation in #189, incorporated in #190.
  • Tom Lanesaw an earlier versionLLM analysis, 98% sureTom reviewed integer parsing and empty-input detection in #179 and #182, before further parser and validation changes.
Add 'basebackup_to_shell' contrib module.
15 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #1421 of 1 reviewer saw the final version

The basebackup_to_shell component of #142 matches exactly, including command substitution, role checks and documentation.

  • Abhijit Menon-Sensaw the final versionLLM analysis, 99% sureAbhijit reviewed the extensible-target API and shell example in #126; the remaining changes were documentation and rebasing.
Allow extensions to add new backup targets.
15 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #1421 of 1 reviewer saw the final version

The target-registration component of #142 matches; the commit additionally includes an accidental src/backend/replication/Makefile.orig file.

  • Abhijit Menon-Sensaw the final versionLLM analysis, 99% sureAbhijit reviewed the extensible-target API and shell example in #126; the remaining changes were documentation and rebasing.
Change HAVE_LIBLZ4 and HAVE_LIBZSTD tests to USE_LZ4 and USE_ZSTD.
15 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1561 of 1 reviewer saw the final version

The USE_LZ4/USE_ZSTD replacements exactly match #156.

  • Jeevan Ladhesaw the final versionLLM analysis, 100% sureJeevan confirmed in #157 that the patch covered the required feature-macro replacements.
pg_basebackup: Clean up some bogus file extension tests.
11 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

The pg_basebackup.c filename-suffix cleanup comes from Justin’s inline patch in #162, with no attached patchset.

  • Robert Haassaw the final versioncommittercommitted the patch
pg_basebackup: Avoid unclean failure with server-compression and -D -.
11 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1611 of 1 reviewer saw the final version

The rejection of compressed manifest injection into standard-output backups exactly matches #161.

  • Justin Pryzbysaw the final versionLLM analysis, 100% sureJustin reported the failure in #160 and endorsed the proposed rejection in #162.
Fix LZ4 tests for remaining buffer space.
8 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1551 of 1 reviewer saw the final version

Both client and server LZ4 buffer-space comparisons exactly match #155.

  • Robert Haassaw the final versioncommittercommitted the patch
Add support for zstd base backup compression.
8 Mar 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1541 of 4 reviewers saw the final version

Applying both components of #154 to the commit’s parent produces byte-identical files, including Robert’s buffer-bound and cosmetic edits.

  • Tushar Ahujasaw an earlier versionLLM analysis, 99% sureTushar reported a large-backup failure in #93 and retested ZSTD in #143, before client compression and decompression were added.
  • Justin Pryzbysaw an earlier versionLLM analysis, 99% sureJustin identified configuration, return-path and assertion problems in #144, corrected before the later client-side additions.
  • Alvaro Herrerasaw an earlier versionLLM analysis, 90% sureÁlvaro proposed clearer compression-help syntax in #147, adopted in #149 before subsequent buffer-logic changes.
  • Robert Haassaw the final versioncommittercommitted the patch

Jan - Feb 2022

10 commits
pg_basebackup: Allow client-side LZ4 (de)compression.
11 Feb 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% suremodified before commitmoderate changesfrom patchset #1292 of 3 reviewers saw the final version

The client LZ4 streamers come from #129; the commit adds documentation, bare lz4 option parsing and correct uncompressed streamed-WAL handling.

  • Jeevan Ladhesaw an earlier versionLLM analysis, 99% sureJeevan reviewed buffer handling and redundant state in #128; later committer changes also corrected option parsing and WAL handling.
  • Tushar Ahujasaw the final versionLLM analysis, 97% sure751b8d23b7 explicitly credits Tushar’s review and testing, although no client-LZ4-specific precommit review is visible here.
  • Robert Haassaw the final versioncommittercommitted the patch
Add suport for server-side LZ4 base backup compression.
11 Feb 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% suremodified before commitmoderate changesfrom patchset #1301 of 3 reviewers saw the final version

Compared with #130, the commit removes an erroneous compresslevel assertion before runtime validation and reformats a comment in basebackup_lz4.c.

  • Tushar Ahujasaw an earlier versionLLM analysis, 97% sureTushar tested server-side LZ4 options in #87, before compression levels and additional validation were added.
  • Dipesh Panditsaw an earlier versionLLM analysis, 90% sureDipesh tested his client implementation over the rebased server patch in #131, before the final assertion correction.
  • Robert Haassaw the final versioncommittercommitted the patch
Remove superfluous variable.
28 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #1191 of 1 reviewer saw the final version

The first component of #119 exactly removes the unused have_zlib test variable; the accompanying LZ4 feature was committed later.

  • Robert Haassaw the final versioncommittercommitted the patch
pg_basebackup: Cleaner handling when compression is multiply specified.
28 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1141 of 1 reviewer saw the final version

The compression-option overwrite fix exactly matches #114.

  • Tushar Ahujasaw the final versionLLM analysis, 100% sureTushar reported repeated-option inconsistencies in #113 and tested the fix in #117.
Allow server-side compression to be used with -Fp.
28 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 99% suremodified before commitmoderate changesfrom patchset #1151 of 1 reviewer saw the final version

The gzip decompressor and tests follow Robert’s consolidation #115, with Dipesh’s guard-removal delta #116 and a test skip-count correction folded in.

  • Robert Haassaw the final versioncommittercommitted the patch
pg_basebackup: Fix a couple of recently-introduced bugs.
27 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1091 of 1 reviewer saw the final version

The default compression-level and incompatible -R/target checks exactly match #109.

  • Tushar Ahujasaw the final versionLLM analysis, 100% sureTushar reported both failures in #104 and confirmed the correction in #111.
Tidy up a few cosmetic issues related to pg_basebackup.
25 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #1033 of 3 reviewers saw the final version

The help and SGML corrections exactly match #103.

  • Shinoda, Noriyoshi (PN Japan FSIP)saw the final versionLLM analysis, 97% sureShinoda identified and patched the missing client/server help choices in #101; Dagfinn refined the notation and documentation.
  • Tushar Ahujasaw the final versionLLM analysis, 90% sureTushar independently reported the missing compression-help information in #104 before this documentation fix landed.
  • Robert Haassaw the final versioncommittercommitted the patch
Server-side gzip compression.
24 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surelanded as postedfrom patchset #972 of 4 reviewers saw the final version

The server-side gzip implementation and tests from #97 match modulo the announced trivial rebase.

  • Tushar Ahujasaw an earlier versionLLM analysis, 97% sureTushar tested server targets and compression and reported option problems in #24 and #87, before later feature and validation changes.
  • Suraj Kharagesaw the final versionLLM analysis, 96% sure0ad8032910 explicitly credits Suraj’s review or testing, although no gzip-specific review is visible here.
  • Dipesh Panditsaw the final versionLLM analysis, 96% sure0ad8032910 explicitly credits Dipesh’s review or testing beyond his separately authored decompressor.
  • Mark Dilgersaw an earlier versionLLM analysis, 96% sureMark reviewed protocol, options and test coverage in #26, before later target and compression revisions.
Support base backup targets.
20 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #942 of 4 reviewers saw the final version

The first component of #94 supplies server and blackhole targets; only three lines of blackhole documentation wording differ.

  • Tushar Ahujasaw an earlier versionLLM analysis, 97% sureTushar tested server targets and compression and reported option problems in #24 and #87, before later feature and validation changes.
  • Suraj Kharagesaw the final versionLLM analysis, 96% sure3500ccc39b explicitly credits Suraj’s review or testing, although no target-specific review is visible here.
  • Dipesh Panditsaw the final versionLLM analysis, 96% sure3500ccc39b explicitly credits Dipesh’s review or testing, although no target-specific review is visible here.
  • Mark Dilgersaw an earlier versionLLM analysis, 96% sureMark reviewed protocol, options and test coverage in #26, before later target revisions.
Modify pg_basebackup to use a new COPY subprotocol for base backups.
18 Jan 2022Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #773 of 5 reviewers saw the final version

The single-COPY subprotocol component of #77 matches exactly, including message typing, archive names and explicit progress.

  • Jeevan Ladhesaw an earlier versionLLM analysis, 94% sureJeevan exercised the protocol through target testing and identified missing tablespace-map handling in #55, before protocol documentation and refinements.
  • Tushar Ahujasaw the final versionLLM analysis, 94% sureTushar tested the main series in #87, after the final protocol patch #77.
  • Suraj Kharagesaw the final versionLLM analysis, 96% surecc333f3233 explicitly credits Suraj’s review or testing, although no protocol-specific review is visible here.
  • Dipesh Panditsaw the final versionLLM analysis, 96% surecc333f3233 explicitly credits Dipesh’s review or testing, although no protocol-specific review is visible here.
  • Mark Dilgersaw an earlier versionLLM analysis, 98% sureMark reviewed COPY protocol details and parsing in #26, before later protocol refinements.

Dec 2021

1 commit
Document that tar archives are now properly terminated.
1 Dec 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #77

The protocol.sgml tar-termination documentation component exactly matches #77.

Nov 2021

5 commits
Fix thinko in bbsink_throttle_manifest_contents.
15 Nov 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surewritten from the discussion, no matching patch1 of 1 reviewer saw the final version

Dmitry diagnosed the wrong sink argument in bbsink_throttle_manifest_contents in #75; the one-line commit follows that discussion without an attachment.

  • Dmitry Dolgovsaw the final versionLLM analysis, 99% sureDmitry reported and diagnosed the manifest forwarding error in #75.
Have the server properly terminate tar archives.
9 Nov 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #720 of 1 reviewer saw the final version

The second component of #72 exactly adds server tar terminators and limits client compatibility padding to older servers.

  • Tom Lanesaw an earlier versionLLM analysis, 90% sureTom diagnosed the AIX tar-termination failure in #71, prompting the new client and server fixes in #72.
Minimal fix for unterminated tar archive problem.
8 Nov 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #720 of 1 reviewer saw the final version

The first component of #72 exactly adds a client tar-terminator streamer for archives that bypass parsing.

  • Tom Lanesaw an earlier versionLLM analysis, 90% sureTom diagnosed the AIX tar-termination failure in #71, prompting the new client and server fixes in #72.
Introduce 'bbstreamer' abstraction to modularize pg_basebackup.
5 Nov 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #636 of 9 reviewers saw the final version

The bbstreamer component in #63 matches the committed client code; only typedef-list additions are omitted.

  • Andres Freundsaw an earlier versionLLM analysis, 97% sureAndres reviewed client/server protocol and archive handling in #16, before subsequent architectural revisions.
  • Sumanta Mukherjeesaw the final versionLLM analysis, 96% sure23a1c6578c explicitly credits Sumanta, although his visible early reviews predate the client bbstreamer component.
  • Dilip Kumarsaw the final versionLLM analysis, 96% sure23a1c6578c explicitly credits Dilip, although no client-component-specific review is visible here.
  • Suraj Kharagesaw the final versionLLM analysis, 96% sure23a1c6578c explicitly credits Suraj, although his visible early testing predates the client component.
  • Dipesh Panditsaw the final versionLLM analysis, 96% sure23a1c6578c explicitly credits Dipesh, although his visible early testing predates the client component.
  • Tushar Ahujasaw an earlier versionLLM analysis, 94% sureTushar tested refactored backup paths and reported failures in #20 and #24, before later client revisions.
  • Mark Dilgersaw an earlier versionLLM analysis, 98% sureMark reviewed tar parsing and recovery-injector abstractions in #26 and #30, before later client error-handling changes.
  • Sergei Kornilovsaw the final versionLLM analysis, 96% sure23a1c6578c explicitly credits Sergei, although his visible review concerns other series components.
  • Jeevan Ladhesaw the final versionLLM analysis, 96% sure23a1c6578c explicitly credits Jeevan, although no client-component-specific review is visible here.
Introduce 'bbsink' abstraction to modularize base backup code.
5 Nov 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #631 of 8 reviewers saw the final version

The bbsink component of #63 matches with WAL timeline rebase adjustments and omitted typedef-list additions.

  • Andres Freundsaw an earlier versionLLM analysis, 98% sureAndres reviewed callback costs, progress semantics and protocol design in #3 and #16, before subsequent architectural revisions.
  • Sumanta Mukherjeesaw an earlier versionLLM analysis, 97% sureSumanta examined buffering and performance methodology in #5 and #10, before the sink lifecycle was revised.
  • Dilip Kumarsaw an earlier versionLLM analysis, 98% sureDilip reviewed sink forwarding and abstraction boundaries in #6 and #23, before buffer and cleanup changes.
  • Suraj Kharagesaw an earlier versionLLM analysis, 99% sureSuraj tested backup performance across buffer sizes in #9 and #12, before later refactoring.
  • Dipesh Panditsaw an earlier versionLLM analysis, 99% sureDipesh reported repeated backup performance measurements in #13, before subsequent sink changes.
  • Tushar Ahujasaw an earlier versionLLM analysis, 97% sureTushar tested refactored backup paths and reported failures in #20 and #24, before later buffer and cleanup revisions.
  • Mark Dilgersaw an earlier versionLLM analysis, 98% sureMark reviewed sink comments and abstraction design in #26, before cleanup callbacks were introduced.
  • Jeevan Ladhesaw the final versionLLM analysis, 98% sureJeevan exercised sink sizing and cleanup, proposing size_t in #43 and diagnosing the final-series gzip cleanup failure in #64.

Oct 2021

3 commits
Refactor basebackup.c's _tarWriteDir() function.
12 Oct 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #56

The convert_link_to_directory refactoring component of #56 exactly matches the basebackup.c change.

Flexible options for CREATE_REPLICATION_SLOT.
5 Oct 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1Also in: extensible options syntax for replication parser? (part of a posted patchset)
Flexible options for BASE_BACKUP.
5 Oct 2021Robert Haasmaster15beta1REL_19_STABLE15beta1REL_18_STABLE15beta1REL_17_STABLE15beta1REL_16_STABLE15beta1REL_15_STABLE15beta1Also in: extensible options syntax for replication parser? (part of a posted patchset)

Credits across 33 commits

Author
Robert HaasLLM23
Jeevan LadheLLM4
Dipesh PanditLLM2
Justin Pryzby1 LLM2
Dagfinn Ilmari MannsåkerLLM1
Reviewer
Robert Haascommitter8
Jacob Champion1
Tomas Vondra1
Committer
Robert Haas32
Tomas Vondra1

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