Re: Should io_method=worker remain the default?
Thomas Munro <thomas.munro@gmail.com>
From: Thomas Munro <thomas.munro@gmail.com>
To: Jeff Davis <pgsql@j-davis.com>
Cc: Andres Freund <andres@anarazel.de>, pgsql-hackers@postgresql.org
Date: 2025-09-08T02:44:57Z
Lists: pgsql-hackers
On Mon, Sep 8, 2025 at 2:39 PM Thomas Munro <thomas.munro@gmail.com> wrote: > You might also be interested in the phenomenon discussed in the index > prefetching thread, where concurrent reads of the *same* block from > different backs currently force a wait, which can be avoided with a > patch he posted. In that thread it was happening due to index scans Sorry, bad edit: I meant a patch that Andres posted.
Commits
-
meson: Build numeric.c with -ftree-vectorize.
- 9016fa7e3bcd 19 (unreleased) landed
- 87ea6e9b66b6 18.0 landed
- 509c779293f2 16.11 landed
- 15f9eeef624f 17.7 landed
-
meson: build checksums with extra optimization flags.
- 2de24ca6ca51 16.11 landed
- e25453a3631f 17.7 landed
- a7024398b80a 18.0 landed
- 9af672bcb245 19 (unreleased) landed