pgsql: tests: Rename conflicting role names

2025-12-11 Thread Michael Paquier
tests: Rename conflicting role names These cause problems when running installcheck-world USE_MODULE_DB=1 with -j. Note: This patch has been originally applied to v16 and newer branches as of 6a20b04f0408. Sami Imseih has reported this issue as being possible to reach for other branches still su

pgsql: tests: Rename conflicting role names

2025-12-11 Thread Michael Paquier
tests: Rename conflicting role names These cause problems when running installcheck-world USE_MODULE_DB=1 with -j. Note: This patch has been originally applied to v16 and newer branches as of 6a20b04f0408. Sami Imseih has reported this issue as being possible to reach for other branches still su

pgsql: Fix some comments.

2025-12-11 Thread Nathan Bossart
Fix some comments. Like commit 123661427b, these were discovered while reviewing Aleksander Alekseev's proposed changes to pgindent. Branch -- master Details --- https://git.postgresql.org/pg/commitdiff/b4cbc106a6ced69a0a13a980e1b4b212bb127458 Modified Files -- src/backend/l

pgsql: Fix infer_arbiter_index for partitioned tables

2025-12-11 Thread Álvaro Herrera
Fix infer_arbiter_index for partitioned tables The fix for concurrent index operations in bc32a12e0db2 started considering indexes that are not yet marked indisvalid as arbiters for INSERT ON CONFLICT. For partitioned tables, this leads to including indexes that may not exist in partitions, causi

pgsql: Fix comment on how temp files and subtransactions are handled

2025-12-11 Thread Heikki Linnakangas
Fix comment on how temp files and subtransactions are handled The comment was accurate a long time ago, but not any more. I failed to update the comment in commit ab3148b712. Branch -- master Details --- https://git.postgresql.org/pg/commitdiff/b65f1ad9b12767dbd45d9588ce8ed2e593dbddbf M

pgsql: Add runtime checks for bogus multixact offsets

2025-12-11 Thread Heikki Linnakangas
Add runtime checks for bogus multixact offsets It's not far-fetched that we'd try to read a multixid with an invalid offset in case of bugs or corruption. Or if you call pg_get_multixact_members() after a crash that left behind invalid but unused multixids. Better to get a somewhat descriptive err

pgsql: Make consistently available in frontend and backend

2025-12-11 Thread Peter Eisentraut
Make consistently available in frontend and backend Previously, c.h made only available in frontends (#ifdef FRONTEND), which was probably reasonable, because the only thing it would give you is assert(), which you generally shouldn't use in the backend. But with C11, also makes available stat