Sep 2025

2 commits
Add regression expected-files for older OpenSSL in FIPS mode.
16 Sep 2025Tom Lanemaster19beta1REL_19_STABLE19beta1REL_18_STABLE18.0REL_17_STABLE17.7
LLM analysis, 100% surewritten from the discussion, no matching patch

Extends the older-OpenSSL FIPS support reopened in #45 by adding contrib/pgcrypto/expected/hmac-md5_2.out and md5_2.out for the 'No such hash algorithm' errors observed by the buildfarm. Neither file nor its error wording appears in any attached patchset, and the commit explicitly cites that discussion.

Add regression expected-files for older OpenSSL in FIPS mode.
16 Sep 2025Tom Lanemaster19beta1REL_19_STABLE19beta1REL_18_STABLE18.0REL_17_STABLE17.7
LLM analysis, 100% surewritten from the discussion, no matching patch

Adds src/test/regress/expected/md5_2.out and password_2.out for the older 'disabled for FIPS' wording proposed in #45, with no corresponding attached patchset. Git ancestry identifies this candidate as the development-branch commit.

Nov 2023

5 commits
Allow tests to pass in OpenSSL FIPS mode (rest)
17 Nov 2023Peter Eisentrautmaster17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #401 of 2 reviewers saw the final version

Takes the alternative-output patch and Tom Lane's password-test delta from #40; applying those attachments reproduces all six committed files exactly. The remaining pgcrypto, TAP, and BRIN portions of the submission are outside this commit.

  • Tom Lanesaw the final versiontrailerTom tested the earlier series in #32 and supplied the final CREATE ROLE/ALTER ROLE separation with his review in #40.
  • Daniel Gustafssonsaw an earlier versiontrailerDaniel reviewed the alternative-output approach in #30, before #40 changed password-test execution so roles survive password-setting failures.
Allow tests to pass in OpenSSL FIPS mode (TAP tests)
17 Nov 2023Peter Eisentrautmaster17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #402 of 2 reviewers saw the final version

Commits only the TAP component of #40, which explicitly resends Peter's patch unchanged. Differences are Perl formatting, explanatory FIPS comments, and SCRAM capitalization in src/test/authentication/t/001_password.pl and src/test/ssl/t/002_scram.pl.

  • Tom Lanesaw the final versiontrailerTom reported successful testing apart from 3DES in #32 and endorsed the unchanged TAP patch in #40.
  • Daniel Gustafssonsaw the final versiontrailerDaniel requested the explanatory MD5-probe comment and SCRAM capitalization in #30, and those light changes appear in the commit.
pgcrypto: Allow tests to pass in OpenSSL FIPS mode
17 Nov 2023Peter Eisentrautmaster17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #402 of 2 reviewers saw the final version

The five committed pgcrypto alternative expected files exactly match #40, including Tom Lane's 3DES addition. The PGP test extraction bundled with that submission landed separately in 3af0d17ace.

  • Tom Lanesaw the final versiontrailerTom found the missing 3DES behavior in #32 and supplied the final expected file and PGP test reorganization in #40.
  • Daniel Gustafssonsaw the final versiontrailerDaniel reviewed the MD5 expected outputs in #30 and endorsed an alternative results file for disabled ciphers in #39, with only test-file organization changing afterward.
pgcrypto: Split off pgp-encrypt-md5 test
16 Nov 2023Peter Eisentrautmaster17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #402 of 2 reviewers saw the final version

Extracts precisely the six-file pgp-encrypt-md5 test separation introduced by Tom Lane in #40, including Makefile and meson.build registration. The accompanying FIPS expected files were left for a separate commit.

  • Tom Lanesaw the final versiontrailerTom reviewed Peter's pgcrypto patch and supplied the specific test separation in #40.
  • Peter Eisentrautsaw the final versioncommittercommitted the patch
citext: Allow tests to pass in OpenSSL FIPS mode
7 Nov 2023Peter Eisentrautmaster17beta1REL_19_STABLE17beta1REL_18_STABLE17beta1REL_17_STABLE17beta1
LLM analysis, 100% surepart of a posted patchsetlight changesfrom patchset #292 of 2 reviewers saw the final version

Exactly matches the citext component of #29, deleting its MD5 test from citext.sql and both expected files. The other four patches in that submission are not included.

  • Daniel Gustafssonsaw the final versionLLM analysis, 94% sureDaniel reported reading the series without finding problems beyond minor comments on other components in #30, and the citext patch landed unchanged.
  • Tom Lanesaw the final versionLLM analysis, 99% sureTom reported check-world and manually run suites passing apart from the unrelated 3DES failure in #32.

Mar 2023

2 commits
Remove incidental md5() function uses from main regression tests
13 Mar 2023Peter Eisentrautmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 99% suremodified before commitlight changesfrom patchset #131 of 1 reviewer saw the final version

Uses Tom Lane's revision of Peter's core regression patch in #13, renaming notmd5 to fipshash and simplifying helper bodies to SQL RETURN expressions. Additional rowsecurity expected-output updates accommodate intervening tests.

  • Tom Lanesaw the final versiontrailerTom reviewed and tested the core changes on a FIPS-enabled Fedora system and supplied the helper-based revision in #13, followed only by naming, syntax, and expected-output adjustments.
Improve/correct comments
9 Mar 2023Peter Eisentrautmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 88% surepart of a posted patchsetlight changesfrom patchset #191 of 1 reviewer saw the final version

Extracts the cryptohash initialization-comment correction exposed by #19 and extends it to update and finalization comments as agreed in #23 and #24. None of FAKE_FIPS_MODE's functional changes landed in this commit.

  • Daniel Gustafssonsaw the final versiontrailerDaniel identified the independently valid comment correction in #20 and endorsed matching cryptohash_openssl.c in #24.

Oct 2022

1 commit
Put tests of md5() function into separate test file
13 Oct 2022Peter Eisentrautmaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 100% surelanded as postedfrom patchset #21 of 1 reviewer saw the final version

The five-file MD5 test extraction matches #2 without substantive changes, including md5.sql, md5.out, strings test removals, and parallel_schedule.

  • Michael Paquiersaw the final versiontrailerMichael explicitly said the extracted test slice looked fine in #3, and it landed unchanged.

Credits across 10 commits

Author
Peter EisentrautLLM7
Tom Lane1 LLM3
Reviewer
Tom Lane5
Daniel Gustafsson4
Michael Paquier1
Peter Eisentrautcommitter1
Committer
Peter Eisentraut8
Tom Lane2

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