2 questions about volatile attribute of pg_proc.
| От | Andy Fan |
|---|---|
| Тема | 2 questions about volatile attribute of pg_proc. |
| Дата | |
| Msg-id | CAKU4AWp09TdpQzGo859MxvcYfiMvrMMjP2-R2EM=4JJELSFvgA@mail.gmail.com обсуждение исходный текст |
| Ответы |
Re: 2 questions about volatile attribute of pg_proc.
Re: 2 questions about volatile attribute of pg_proc. |
| Список | pgsql-hackers |
Hi:
--
We know volatile is very harmful for optimizers and it is the default
value (and safest value) if the user doesn't provide that. Asking user
to set the value is not a good experience, is it possible to auto-generate
the value for it rather than use the volatile directly for user defined function. I
think it should be possible, we just need to scan the PlpgSQL_stmt to see if there
is a volatile function?
The second question "It is v for “volatile” functions, whose results might change at any time.
(Use v also for functions with side-effects, so that calls to them cannot get optimized away.)"
(Use v also for functions with side-effects, so that calls to them cannot get optimized away.)"
I think they are different semantics. One of the results is volatile functions can't be removed
by remove_unused_subquery_output even if it doesn't have side effects. for example:
select b from (select an_expensive_random(), b from t); Is it by design on purpose?
Best Regards
Andy Fan (https://www.aliyun.com/)
В списке pgsql-hackers по дате отправления: