Re: Memory consumed by child SpecialJoinInfo in partitionwise join planning

On Wed, 4 Oct 2023 at 04:02, Ashutosh Bapat
<ashutosh.bapat.oss@gmail.com> wrote:
>
> On Fri, Sep 29, 2023 at 8:36 AM Amit Langote <amitlangote09@gmail.com> wrote:
> > IOW, something
> > like the following would have sufficed:
> >
> > @@ -1735,6 +1735,10 @@ build_child_join_sjinfo(PlannerInfo *root,
> > SpecialJoinInfo *parent_sjinfo,
> >  /*
> >   * free_child_sjinfo_members
> >   *     Free memory consumed by members of a child SpecialJoinInfo.
> > + *
> > + * Only members that are translated copies of their counterpart in the parent
> > + * SpecialJoinInfo are freed here.  However, members that could be referenced
> > + * elsewhere are not freed.
> >   */
> >  static void
> >  free_child_sjinfo_members(SpecialJoinInfo *child_sjinfo)
> > @@ -1755,10 +1759,7 @@ free_child_sjinfo_members(SpecialJoinInfo *child_sjinfo)
> >     bms_free(child_sjinfo->syn_lefthand);
> >     bms_free(child_sjinfo->syn_righthand);
> >
> > -   /*
> > -    * But the list of operator OIDs and the list of expressions may be
> > -    * referenced somewhere else. Do not free those.
> > -    */
> > +   /* semi_rhs_exprs may be referenced, so don't free. */
> >  }
>
> Works for me. PFA patchset with these changes. I have still left the
> changes addressing your comments as a separate patch for easier
> review.

CFBot shows that the patch does not apply anymore as in [1]:

=== Applying patches on top of PostgreSQL commit ID
55627ba2d334ce98e1f5916354c46472d414bda6 ===
=== applying patch
./0001-Report-memory-used-for-planning-a-query-in--20231003.patch
...
Hunk #1 succeeded at 89 (offset -3 lines).
patching file src/test/regress/expected/explain.out
Hunk #5 FAILED at 285.
Hunk #6 succeeded at 540 (offset 4 lines).
1 out of 6 hunks FAILED -- saving rejects to file
src/test/regress/expected/explain.out.rej

Please post an updated version for the same.

[1] - http://cfbot.cputube.org/patch_46_4500.log

Regards,
Vignesh