Horton

Re: PG 15 (and to a smaller degree 14) regression due to ExprEvalStep size

Andrew Dunstan <andrew@dunslane.net>

View thread View on postgr.es
From: Andrew Dunstan <andrew@dunslane.net>
To: Andres Freund <andres@anarazel.de>
Cc: Michael Paquier <michael@paquier.xyz>, Tom Lane <tgl@sss.pgh.pa.us>, David Rowley <dgrowleyml@gmail.com>, Peter Geoghegan <pg@bowt.ie>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Nikita Glukhov <n.gluhov@postgrespro.ru>
Date: 2022-07-05T19:04:05Z
Lists: pgsql-hackers 
On 2022-07-05 Tu 14:36, Andres Freund wrote:
> Hi,
>
> On 2022-06-24 10:29:06 -0400, Andrew Dunstan wrote:
>> On 2022-06-23 Th 21:51, Andres Freund wrote:
>>> On 2022-06-23 16:38:12 +0900, Michael Paquier wrote:
>>>> On Tue, Jun 21, 2022 at 05:41:07PM -0400, Andrew Dunstan wrote:
>>>>> Yes, but I don't guarantee to have a fix in time for Beta2.
>>>> IMHO, it would be nice to get something done for beta2.  Now the
>>>> thread is rather fresh and I guess that more performance study is
>>>> required even for 0004, so..
>>> I don't think there's a whole lot of performance study needed for 0004 - the
>>> current state is obviously wrong.
>>>
>>> I think Andrew's beta 2 comment was more about my other architectural
>>> complains around the json expression eval stuff.
>>
>> Right. That's being worked on but it's not going to be a mechanical fix.
> Any updates here?


Not yet. A colleague and I are working on it. I'll post a status this
week if we can't post a fix.


>
> I'd mentioned the significant space use due to all JsonCoercionsState for all
> the types. Another related aspect is that this code is just weird - the same
> struct name (JsonCoercionsState), nested in each other?
>
>     struct JsonCoercionsState
>     {
>         struct JsonCoercionState
>         {
>             JsonCoercion *coercion; /* coercion expression */
>             ExprState  *estate; /* coercion expression state */
>         }           null,
>                     string,
>         numeric    ,
>                     boolean,
>                     date,
>                     time,
>                     timetz,
>                     timestamp,
>                     timestamptz,
>                     composite;
>     }           coercions;      /* states for coercion from SQL/JSON item
>                                  * types directly to the output type */
>
> Also note the weird numeric indentation that pgindent does...


Yeah, we'll try to fix that.


>
>
>> The attached very small patch applies on top of your 0002 and deals with
>> the FmgrInfo complaint.
> Now that the FmgrInfo is part of a separately allocated struct, that doesn't
> seem necessary anymore.


Right, but you complained that we should do it the same way as it's done
elsewhere, so I thought I'd do that anyway.


cheers


andrew


--
Andrew Dunstan
EDB: https://www.enterprisedb.com

Commits

  1. JSON_TABLE: Add support for NESTED paths and columns

  2. Add basic JSON_TABLE() functionality

  3. Add SQL/JSON query functions

  4. Add soft error handling to some expression nodes

  5. Adjust populate_record_field() to handle errors softly

  6. Refactor code used by jsonpath executor to fetch variables

  7. Add more SQL/JSON constructor functions

  8. SQL/JSON: support the IS JSON predicate

  9. SQL/JSON: add standard JSON constructor functions

  10. Add static assertion ensuring sizeof(ExprEvalStep) <= 64 bytes

  11. Remove size increase in ExprEvalStep caused by hashed saops

  12. pgstat: reduce timer overhead by leaving timer running.

  13. expression eval: Fix EEOP_JSON_CONSTRUCTOR and EEOP_JSONEXPR size.

  14. SQL/JSON query functions

  15. Speedup ScalarArrayOpExpr evaluation