Horton

Re: Using Valgrind to detect faulty buffer accesses (no pin or buffer content lock held)

Peter Geoghegan <pg@bowt.ie>

View thread View on postgr.es
From: Peter Geoghegan <pg@bowt.ie>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Anastasia Lubennikova <a.lubennikova@postgrespro.ru>, Daniel Gustafsson <daniel@yesql.se>, Andres Freund <andres@anarazel.de>, Alvaro Herrera <alvherre@2ndquadrant.com>, Masahiko Sawada <masahiko.sawada@2ndquadrant.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-07-19T02:45:42Z
Lists: pgsql-hackers

Attachments

On Sat, Jul 18, 2020 at 7:36 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I wonder whether skink's failure today is due to this change:
>
> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=skink&dt=2020-07-18%2018%3A01%3A10

That seems extremely likely. I think that I need to do something like
what you see in the attached.

Anyway, I'll take care of it tomorrow. Sorry for missing it before my commit.

-- 
Peter Geoghegan

Commits

  1. Add nbtree Valgrind buffer lock checks.

  2. Add Valgrind buffer access instrumentation.

  3. Fix bug in nbtree VACUUM "skip full scan" feature.

  4. Fix another minor page deletion buffer lock issue.

  5. Fix minor nbtree page deletion buffer lock issue.

  6. Skip full index scan during cleanup of B-tree indexes when possible