Re: Removing unneeded self joins

Alexander Korotkov <aekorotkov@gmail.com>

From: Alexander Korotkov <aekorotkov@gmail.com>

To: Andrei Lepikhov <lepihov@gmail.com>

Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Tom Lane <tgl@sss.pgh.pa.us>, Robert Haas <robertmhaas@gmail.com>, Richard Guo <guofenglinux@gmail.com>, Alexander Lakhin <exclusion@gmail.com>, "Gregory Stark (as CFM)" <stark.cfm@gmail.com>, Michał Kłeczek <michal@kleczek.org>, Dean Rasheed <dean.a.rasheed@gmail.com>

Date: 2025-02-09T11:41:10Z

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 →

Remove GUC_NOT_IN_SAMPLE from enable_self_join_elimination
- 717d0e8dd945 18.0 landed
Put enable_self_join_elimination into postgresql.conf.sample
- c2d329260cd8 18.0 landed
Get rid of ojrelid local variable in remove_rel_from_query()
- e167191dc146 18.0 landed
Implement Self-Join Elimination
- fc069a3a6319 18.0 cited
Revert: Remove useless self-joins
- d1d286d83c0e 17.0 landed
Replace lateral references to removed rels in subqueries
- 466979ef031a 17.0 landed
Replace relids in lateral subquery parse tree during SJE
- 489072ab7a9e 17.0 landed
Forbid SJE with result relation
- 8c441c082797 17.0 landed
Fix misuse of RelOptInfo.unique_for_rels cache by SJE
- 30b4955a4668 17.0 landed
Replace the relid in some missing fields during SJE
- a7928a57b9f0 17.0 landed
Revert 56-bit relfilenode change and follow-up commits.
- a448e49bcbe4 16.0 cited
Stabilize timetz test across DST transitions.
- 4a071afbd056 14.0 cited
Speed up finding EquivalenceClasses for a given set of rels
- 3373c7155350 13.0 cited
Fix mark-and-restore-skipping test case to not be a self-join.
- 24d08f3c0a1f 12.0 landed

Attachments

v11-0001-Remove-useless-self-joins.patch (application/octet-stream) patch v11-0001

Hi Andrei!

On Fri, Jan 31, 2025 at 6:57 AM Andrei Lepikhov <lepihov@gmail.com> wrote:
> On 1/10/25 10:13, Alexander Korotkov wrote:
> > I've got an off-list report from Alexander Lakhin.  The failing query
> > is added to the regression tests in the revised patch.  The query was
> > failing with an error "negative bitmapset member not allowed" issued
> > in adjust_relid_set().  In order to fix that I've to teach
> > adjust_relid_set() about negative newrelid, it became even more
> > similar to replace_relid().  Notable this error happens in
> > remove_leftjoinrel_from_query(). This seems to be consequence that we
> > made self-join removal and left-joins removal use common
> > infrastructure.
> I have analysed your change. It pertains to the situation where PHV
> references both the outer join and its inner join in phrels but actually
> utilises only the outer side of the join.
>
> I agree with your proposed fix and have just rewritten the query to make
> it simpler and more consistent with the regression tests (see new
> version of the patch in attachment).
>
> Additionally, I noticed that the remove_useless_joins code uses the term
> "join removal" in the join.sql tests. I wonder if it might be more
> appropriate to revert the name from "Self Join Elimination" back to
> "Self Join Removal."

Great, thank you.

Regarding adjust_relid_set() and replace_relid().  I think they are
now strictly equivalent, except for the case then old relid is given
and not found.  In this case adjust_relid_set() returns the original
relids while replace_relid() returns a copy.  The behavior of
adjust_relid_set() appears more desirable as we don't need extra
copying when no modification is done.  So, I've replaced all
replace_relid() with adjust_relid_set().

Also, I did some grammar correction to your new comment in tests.

------
Regards,
Alexander Korotkov
Supabase