model-writtenrun 6331 - may be wrong, links go to the source messages. Commit-by-commit LLM analysis is in the Commits dialog.
- Tom Lane opened the discussion because debug_invalidate_system_caches_always was cumbersome to type, proposing a shorter name that retained the debug_ prefix and dropped unnecessary words #1.
- Noah Misch argued that “always” misrepresented the multiple levels and “clobber” poorly described invalidation, while Justin Pryzby and Kyotaro Horiguchi supported removing “always”; Amul Sul and Bharath Rupireddy initially preferred debug_clobber_caches #2 #3 #4 #5 #6.
- Tom favored Noah's debug_flush_caches suggestion, but Bharath questioned its implication of writing to disk, prompting Andrew Dunstan to suggest “discard”; Tom also confirmed that initdb's --clobber-cache option should follow the chosen terminology #8 #9 #10 #11.
- Peter Eisentraut distinguished cache invalidation from the memory clobbering enabled by CLOBBER_FREED_MEMORY, and Robert Haas favored debug_discard_caches for clarity; Tom accepted that choice, with approval from Alvaro Herrera and Noah Misch #12 #14 #15 #16 #17.
- Tom posted the sole patch, renaming the GUC, initdb option, and enabling macro while updating documentation and tests; the thread ended with that proposal and his observation that PostgreSQL 14 would also need release-note adjustments #18.
- The following day, d68a003912 landed the attachment from #18 unchanged, completing the proposed development-branch rename.