| From: | Joe Conway <mail(at)joeconway(dot)com> |
|---|---|
| To: | Craig Ringer <craig(dot)ringer(at)2ndquadrant(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Mario Becroft <mb(at)true(dot)group>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>, Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, Ilya Shkuratov <motr(dot)ilya(at)ya(dot)ru>, "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>, Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>, Gavin Flower <GavinFlower(at)archidevsys(dot)co(dot)nz>, Andrew Dunstan <andrew(dot)dunstan(at)2ndquadrant(dot)com>, Serge Rielau <serge(at)rielau(dot)com> |
| Subject: | Re: CTE inlining |
| Date: | 2017-05-05 00:17:05 |
| Message-ID: | 4f98ca61-798b-8751-b2a7-7b0af2bc237b@joeconway.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 05/04/2017 05:03 PM, Craig Ringer wrote:
> On 5 May 2017 02:52, "Tom Lane" wrote:
> I haven't been keeping close tabs either, but surely we still have
> to have
> the optimization fence in (at least) all these cases:
>
> * CTE contains INSERT/UPDATE/DELETE
> * CTE contains SELECT FOR UPDATE/SHARE (else the set of rows that get
> locked might change)
> * CTE contains volatile functions
>
> I'm willing to write off cases where, eg, a function should have been
> marked volatile and was not. That's user error and there are plenty
> of hazards of that kind already. But if the optimizer has reason
> to know that discarding the fence might change any query side-effects,
> it mustn't.
>
> I think everyone is in total agreement there.
That's great, but if so, why do we need any change in syntax at all?
Joe
--
Crunchy Data - http://crunchydata.com
PostgreSQL Support for Secure Enterprises
Consulting, Training, & Open Source Development
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2017-05-05 00:28:48 | Re: PG 10 release notes |
| Previous Message | Craig Ringer | 2017-05-05 00:16:42 | Re: CTE inlining |