Horton

Re: pgstat include expansion

Álvaro Herrera <alvherre@kurilemu.de>

View thread View on postgr.es
From: Alvaro Herrera <alvherre@kurilemu.de>
To: Andres Freund <andres@anarazel.de>
Cc: Heikki Linnakangas <hlinnaka@iki.fi>, pgsql-hackers@postgresql.org, Amit Kapila <amit.kapila16@gmail.com>, Michael Paquier <michael@paquier.xyz>
Date: 2026-02-27T10:37:31Z
Lists: pgsql-hackers

Attachments

Hi,

On 2026-Feb-26, Andres Freund wrote:

> FWIW, regardless of what we choose here (i.e. whether and where to
> "unnecessarily" include wait_event.h), I'd be in favor of fixing all the
> fallout in our source tree, just for cleanliness' sake. It's also presumably
> the part of the patch that would be the most pain to keep updated.

Oh, that makes sense.

> I'm not sure that including in latch.h really would an improvement though -
> that's included in proc.h and libpq.h, which both are fairly widely included
> [1].

I have committed fixes for both of those, so that including an extra
file in latch.h is not as bad.

Here's one more patch.  I noticed that the call actually uses
PG_WAIT_EXTENSION, which is in storage/wait_classes.h.  Including that
seems a bit narrow-minded even if it's just for backwards compatibility,
so I decided to do wait_event_types.h instead.  This is still a bit
better than the whole wait_event.h IMO.  And also now it's IWYU: export
rather than IWYU: keep, which seems sensible.

-- 
Álvaro Herrera               48°01'N 7°57'E  —  https://www.EnterpriseDB.com/
"El miedo atento y previsor es la madre de la seguridad" (E. Burke)

Commits

  1. Don't include wait_event.h in pgstat.h

  2. Don't include proc.h in shm_mq.h

  3. Don't include latch.h in libpq/libpq.h

  4. Reduce includes in pgstat.h

  5. Avoid including utils/timestamp.h in conflict.h.

  6. Avoid including worker_internal.h in pgstat.h.

  7. Add seq_sync_error_count to subscription statistics.

  8. Don't include execnodes.h in replication/conflict.h

  9. Collect statistics about conflicts in logical replication.