Horton

Re: Removing unneeded self joins

Alexander Korotkov <aekorotkov@gmail.com>

View thread
From: Alexander Korotkov <aekorotkov@gmail.com>
To: Andrei Lepikhov <lepihov@gmail.com>
Cc: jian he <jian.universality@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>, Robert Haas <robertmhaas@gmail.com>, Andrei Lepikhov <a.lepikhov@postgrespro.ru>, Richard Guo <guofenglinux@gmail.com>, Alexander Lakhin <exclusion@gmail.com>, "Gregory Stark (as CFM)" <stark.cfm@gmail.com>, Michał Kłeczek <michal@kleczek.org>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2024-07-12T10:30:35Z
Lists: pgsql-hackers

Commits

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

  1. Remove GUC_NOT_IN_SAMPLE from enable_self_join_elimination

  2. Put enable_self_join_elimination into postgresql.conf.sample

  3. Get rid of ojrelid local variable in remove_rel_from_query()

  4. Implement Self-Join Elimination

  5. Revert: Remove useless self-joins

  6. Replace lateral references to removed rels in subqueries

  7. Replace relids in lateral subquery parse tree during SJE

  8. Forbid SJE with result relation

  9. Fix misuse of RelOptInfo.unique_for_rels cache by SJE

  10. Replace the relid in some missing fields during SJE

  11. Revert 56-bit relfilenode change and follow-up commits.

  12. Stabilize timetz test across DST transitions.

  13. Speed up finding EquivalenceClasses for a given set of rels

  14. Fix mark-and-restore-skipping test case to not be a self-join.

Hi, Andrei!

On Fri, Jul 12, 2024 at 6:05 AM Andrei Lepikhov <lepihov@gmail.com> wrote:
>
> On 7/11/24 14:43, jian he wrote:
> > On Tue, Jul 9, 2024 at 2:06 PM Andrei Lepikhov <lepihov@gmail.com> wrote:
> >>
> >> On 7/2/24 07:25, jian he wrote:
> >>> to make sure it's correct, I have added a lot of tests,
> >>> Some of this may be contrived, maybe some of the tests are redundant.
> >> Thanks for your job!
> >> I passed through the patches and have some notes:
> >> 1. Patch 0001 has not been applied anymore since the previous week's
> >> changes in the core. Also, there is one place with trailing whitespace.
> >
> > thanks.
> > because the previous thread mentioned the EPQ problem.
> > in remove_useless_self_joins, i make it can only process CMD_SELECT query.
> I would like to oppose here: IMO, it is just a mishap which we made
> because of a long history of patch transformations. There we lost the
> case where RowMark exists for only one of candidate relations.
> Also, after review I think we don't need so many new tests. Specifically
> for DML we already have one:
>
> EXPLAIN (COSTS OFF)
> UPDATE sj sq SET b = 1 FROM sj as sz WHERE sq.a = sz.a;
>
> And we should just add something to elaborate it a bit.
> See the patch in attachment containing my proposal to improve v4-0001
> main SJE patch. I think it resolved the issue with EPQ assertion as well
> as problems with returning value.

I tried this.  I applied 0001 from [1] and 0002 from [2].  Then I
tried the concurrent test case [3].  It still fails with assert for
me.  But assert and related stuff is the least problem.  The big
problem, as described in [3], is semantical change in query.  When EPQ
is applied, we fetch the latest tuple of the target relation
regardless snapshot.  But for the self-joined relation we should still
use the snapshot-satisfying tuple.  I don't see even attempt to
address this in your patch.  And as I pointed before, this appears
quite complex.

Links.
1. https://www.postgresql.org/message-id/96250a42-20e3-40f0-9d45-f53ae852f8ed%40gmail.com
2. https://www.postgresql.org/message-id/5b49501c-9cb3-4c5d-9d56-49704ff08143%40gmail.com
3. https://www.postgresql.org/message-id/CAPpHfduM6X82ExT0r9UzFLJ12wOYPvRw5vT2Htq0gAPBgHhKeQ%40mail.gmail.com

------
Regards,
Alexander Korotkov
Supabase