Horton

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

Georgios <gkokolatos@protonmail.com>

View thread View on postgr.es
From: Georgios Kokolatos <gkokolatos@protonmail.com>
To: pgsql-hackers@lists.postgresql.org
Cc: Peter Geoghegan <pg@bowt.ie>
Date: 2020-07-06T08:34:56Z
Lists: pgsql-hackers 
As a general overview, the series of patches in the mail thread do match their description. The addition of the stricter, explicit use of instrumentation does improve the design as the distinction of the use cases requiring a pin or a lock is made more clear. The added commentary is descriptive and appears grammatically correct, at least to a non native speaker.

Unfortunately though, the two bug fixes do not seem to apply.

Also, there is a small issue regarding the process, not the content of the patches. In CF app there is a latest attachment  (v3-0002-Add-nbtree-Valgrind-buffer-lock-checks.patch)  which does not appear in the mail thread.  Before changing the status, I will kindly ask for the complete latest series that applies in the mail thread.

The new status of this patch is: Waiting on Author

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