Jul 2022

5 commits
Doc: Acknowledge historically supported CPUs and OSes.
11 Jul 2022Thomas Munromaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 98% surepart of a posted patchsetlight changesfrom patchset #461 of 1 reviewer saw the final version

Commits only the historical-platform documentation patch from the two-patch submission #46. The eight-line addition to doc/src/sgml/installation.sgml differs only by replacing OSF with Tru64 UNIX and reflowing the list, as explained in #48.

  • Tom Lanesaw the final versionLLM analysis, 96% sureIn #47, he reviewed the submission and considered PlayStation 2 as another historical example; only a small OS-name adjustment followed.
Further tidy-up for old CPU architectures.
11 Jul 2022Thomas Munromaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 98% surepart of a posted patchsetlight changesfrom patchset #461 of 2 reviewers saw the final version

Commits the first patch of the two-patch submission #46, leaving its historical-platform paragraph to a separate commit. The diff matches that first patch exactly in doc/src/sgml/installation.sgml, src/backend/storage/lmgr/s_lock.c, and src/include/storage/s_lock.h.

  • Tom Lanesaw the final versiontrailerHe proposed removing the untested architectures in #41 and explicitly found the resulting patch sane by inspection in #47.
  • Robert Haassaw an earlier versiontrailerIn #43, he endorsed removing the untested-platform claims carried by #38, before the implementation deletions in #46; the commit explicitly credits his review of the idea rather than the patch.
Tidy up claimed supported CPUs and OSes.
10 Jul 2022Thomas Munromaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 100% suremodified before commitlight changesfrom patchset #382 of 2 reviewers saw the final version

Matches all code changes in #38, including VAX spinlock removal and the Alpha cleanup in contrib/pgcrypto/crypt-blowfish.c. The sole diff adjustment removes the redundant Linux parenthetical from doc/src/sgml/installation.sgml following #39.

  • Tom Lanesaw the final versiontrailerHe requested clearer CPU-variant wording in #35, then approved the revised patch with one documentation suggestion in #39.
  • Greg Starksaw the final versiontrailerIn #36, he confirmed the VAX floating-point incompatibility and explained why restoring support would require disproportionate work; subsequent changes were documentation edits.
Remove HP/Intel Itanium support.
8 Jul 2022Thomas Munromaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 99% surepart of a posted patchsetlight changesfrom patchset #242 of 2 reviewers saw the final version

Commits the Itanium-removal component of the two-patch submission #24. Compared with that component, it only adds the requested removal of an obsolete Itanium reference in src/include/port/atomics/generic-msvc.h.

  • Tom Lanesaw the final versiontrailerHe reported successful NetBSD build and regression testing of #21 in #30 and explained in #26 why generated flex references should remain.
  • Andres Freundsaw the final versiontrailerHe requested separate HP-UX and Itanium patches in #22, then approved the split version while identifying the remaining MSVC comment in #25.
Remove HP-UX port.
8 Jul 2022Thomas Munromaster16beta1REL_19_STABLE16beta1REL_18_STABLE16beta1REL_17_STABLE16beta1REL_16_STABLE16beta1
LLM analysis, 99% surepart of a posted patchsetlight changesfrom patchset #243 of 3 reviewers saw the final version

Commits the HP-UX-removal component of #24, with the Itanium component landing separately. Apart from rebase context, the only difference from that component is retaining the existing /dev/tty comment in src/common/sprompt.c following #28.

  • Tom Lanesaw the final versiontrailerHe motivated retaining PA-RISC in #15, reported successful NetBSD testing in #30, and supported preserving the /dev/tty comment in #29.
  • Andres Freundsaw the final versiontrailerHe requested preservation of the PA-RISC instruction explanation and separation of the removals in #22, then approved the HP-UX patch in #25.
  • Peter Eisentrautsaw the final versiontrailerIn #28, he explained why deleting the historical HP-UX example would make the /dev/tty comment misleading, and the committed patch retained it.

Credits across 5 commits

Author
Thomas MunroLLM5
Reviewer
Tom Lane4
Andres Freund2
Greg Stark1
Peter Eisentraut1
Robert Haas1
Committer
Thomas Munro5

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