Re: PGPROC alignment (was Re: pgsql: Separate RecoveryConflictReasons from procsignals)
Heikki Linnakangas <hlinnaka@iki.fi>
From: Heikki Linnakangas <hlinnaka@iki.fi>
To: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>
Cc: Andres Freund <andres@anarazel.de>,
"pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2026-02-20T21:01:33Z
Lists: pgsql-hackers
On 11/02/2026 06:40, Bertrand Drouvot wrote:
> On Tue, Feb 10, 2026 at 10:53:58PM +0200, Heikki Linnakangas wrote:
> 0003:
>
> 1/ There is one missing change in PrintLockQueue() ("links" is still used, and
> that should be replaced by "waitLink").
>
> 2/ change the comment on top of ProcWakeup?
>
> "
> /*
> * ProcWakeup -- wake up a process by setting its latch.
> *
> * Also remove the process from the wait queue and set its links invalid.
> "
>
> s/links/waitLink/?
Fixed those and pushed this "Split PGPROC 'links' field into two, for
clarity" patch. Thanks for the review!
- Heikki
Commits
-
Use C11 alignas in typedef definitions
- 5c2a8d272b9a 19 (unreleased) landed
-
Align PGPROC to cache line boundary
- 412f78c66eed 19 (unreleased) landed
-
Rearrange fields in PGPROC, for clarity
- 2e0853176f8f 19 (unreleased) landed
-
Split PGPROC 'links' field into two, for clarity
- 36bbcd5be3ff 19 (unreleased) landed
-
Remove useless store to local variable
- ab32a9e21d37 19 (unreleased) landed
-
Separate RecoveryConflictReasons from procsignals
- 17f51ea81875 19 (unreleased) cited