| From: | Junwang Zhao <zhjwpku(at)gmail(dot)com> |
|---|---|
| To: | Ashutosh Bapat <ashutosh(dot)bapat(dot)oss(at)gmail(dot)com> |
| Cc: | Kirill Reshke <reshkekirill(at)gmail(dot)com>, Peter Eisentraut <peter(at)eisentraut(dot)org>, assam258(at)gmail(dot)com, Amit Langote <amitlangote09(at)gmail(dot)com>, Vik Fearing <vik(at)postgresfriends(dot)org>, Ajay Pal <ajay(dot)pal(dot)k(at)gmail(dot)com>, Imran Zaheer <imran(dot)zhir(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: SQL Property Graph Queries (SQL/PGQ) |
| Date: | 2026-03-17 04:38:13 |
| Message-ID: | CAEG8a3KvXUSPqktROyzgf_z4khKY+aybLN0ULRokR9_7GRtwLw@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi Ashutosh,
On Mon, Mar 16, 2026 at 11:55 PM Ashutosh Bapat
<ashutosh(dot)bapat(dot)oss(at)gmail(dot)com> wrote:
>
> Hi Peter,
>
> > On Mon, 16 Mar 2026 at 17:43, Peter Eisentraut <peter(at)eisentraut(dot)org> wrote:
> > >
> > > On 11.03.26 08:34, Ashutosh Bapat wrote:
> > > > There are two new patches 0004 and 0005 in the attached patchset.
> > >
> > > I have committed this, including the 0004 patch.
>
> Thanks a lot.
>
> > > Let's consider the
> > > 0005 patch separately.
>
> Will share the rebased patch soon. This thread may see discussion
> about the commit itself. Should I start a new thread for 0005 or use
> this one? New one seems better to me with a new CF entry.
>
> > >
> > > The buildfarm shows some instability in the pg_upgrade test, because
> > > labels are printed by pg_get_propgraphdef() in implementation-dependent
> > > order. Attached is a quick patch to sort the labels before printing.
> > > Check please.
>
> The patch looks fine to me. While reviewing it, I noticed that the
> function has an extra loop to count the number of variables. I don't
> think it's needed. The count can be obtained from the list length. In
> the attached patch, I have removed that loop. Am I missing something?
>
> 0001 is your patch
> 0002 removes the loop + some cosmetic changes
>
> Hi Kirill,
>
> > Do we need to keep relation lock until end of function
> > (table_close(pglrel, AccessShareLock);)?
>
> I think you are right. Fixed in the attached.
>
> > I'm not sure if list_sort is
> > interruptible.
>
> I don't think it matters here. It will be very rare, if not
> impossible, to have so many labels as to let the sorting run for
> milliseconds together. The foreach loop afterwards is also not
> interruptible. Any reason you think it should be interruptible?
>
> --
> Best Wishes,
> Ashutosh Bapat
Do you think it's necessary to add a list_free_deep(label_list) at
the end? make_propgraphdef_labels itself it not on critical path
so explicit free non-used memory doesn't impact performance.
--
Regards
Junwang Zhao
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Amit Kapila | 2026-03-17 04:45:23 | Re: Report bytes and transactions actually sent downtream |
| Previous Message | Haibo Yan | 2026-03-17 04:21:22 | Re: pg_buffercache: Add per-relation summary stats |