| From: | shihao zhong <zhong950419(at)gmail(dot)com> |
|---|---|
| To: | Nikolay Samokhvalov <nik(at)postgres(dot)ai> |
| Cc: | Roman Khapov <rkhapov(at)yandex-team(dot)ru>, Japin Li <japinli(at)hotmail(dot)com>, Kirill Reshke <reshkekirill(at)gmail(dot)com>, Paul A Jungwirth <pj(at)illuminatedcomputing(dot)com>, Andrey Borodin <x4mmm(at)yandex-team(dot)ru>, Miłosz Bieniek <bieniek(dot)milosz(at)proton(dot)me>, "pgsql-hackers(at)lists(dot)postgresql(dot)org" <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: amcheck: support for GiST |
| Date: | 2026-09-25 15:54:42 |
| Message-ID: | CAGRkXqR7ApwgQFdv6uvA4WBd4sL0k_G4Aer9Z4gVxpHa28RK=A@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
Rebased, v2026-09-25 attached. 0001 and 0002 are v2026-01-12-v2. 0003
is new and has my review fixes.
0001 keeps the VARSIZE_ANY() fix from da1eff08a5b.
In 0003, a heapallindexed check on 2M rows now uses about 1MB instead of
300MB. A link of InvalidBlockNumber is P_NEW, so the check used to
extend the index, and a root page marked as split crashed on a NULL
pointer. Both are now reported as corruption. There is also a TAP test
that corrupts the index file.
Is a committer interested in picking this up?
Regards,
Shihao
| Attachment | Content-Type | Size |
|---|---|---|
| v2026-09-25-0001-Move-normalize-tuple-logic-from-nbtcheck.patch | application/octet-stream | 10.0 KB |
| v2026-09-25-0003-amcheck-Review-fixes-for-gist_index_chec.patch | application/octet-stream | 25.2 KB |
| v2026-09-25-0002-Add-gist_index_check-function-to-verify-.patch | application/octet-stream | 33.6 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Zsolt Parragi | 2026-09-25 16:00:29 | Re: ON CONFLICT DO SELECT returns rows hidden by a view |
| Previous Message | Viktor Holmberg | 2026-09-25 15:41:47 | Re: ON CONFLICT DO SELECT returns rows hidden by a view |