Horton
Menu

Re: Asynchronous MergeAppend

Alexander Korotkov <aekorotkov@gmail.com>

View thread
From: Alexander Korotkov <aekorotkov@gmail.com>
To: Alexander Pyhalov <a.pyhalov@postgrespro.ru>
Cc: Pgsql Hackers <pgsql-hackers@postgresql.org>, Matheus Alcantara <matheusssilv97@gmail.com>
Date: 2026-03-30T01:20:41Z
Lists: pgsql-hackers

Attachments

Hi!

On Wed, Mar 18, 2026 at 8:17 AM Alexander Pyhalov
<a.pyhalov@postgrespro.ru> wrote:
> Alexander Pyhalov писал(а) 2026-03-02 09:44:
> > Alexander Pyhalov писал(а) 2026-02-12 10:08:
> >> Alexander Pyhalov писал(а) 2025-12-30 22:04:
> >>> Matheus Alcantara писал(а) 2025-12-30 18:04:
> >>>> On Tue Dec 30, 2025 at 10:15 AM -03, Alexander Pyhalov wrote:
> >>>>> Looks good. What do you think about
> >>>>> classify_matching_subplans_common()?
> >>>>> Should it stay where it is or should we hide it to
> >>>>> src/include/executor/execAppend.h ?
> >>>>>
> >>>> Yeah, sounds better to me to move classify_matching_subplans_common
> >>>> to
> >>>> execAppend.h since it very specific for the MergeAppend and Append
> >>>> execution. I've declared the function as static inline on
> >>>> execAppend.h
> >>>> but I'm not sure if it's the best approach.
> >>>>
> >>>> I've also wrote a proper commit message for this new version.
> >>>
> >>> Looks good to me.
> >>
> >> Hi.
> >> Rebased patches over fresh master.
> >
> > Hi.
> > Rebased patches over fresh master.
>
> Hi.
> Rebased patches over fresh master.

Thank you for your work on this subject.
I have revised the patchset.  I think it would be better if common
infrastructure goes first.  Otherwise we commit async merge append and
immediately revise it.  I also did some minor improvements.

------
Regards,
Alexander Korotkov
Supabase

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Handle interrupts while waiting on Append's async subplans