Re: Use extended statistics to estimate (Var op Var) clauses
От | Zhihong Yu |
---|---|
Тема | Re: Use extended statistics to estimate (Var op Var) clauses |
Дата | |
Msg-id | CALNJ-vQEPk6j2heB1_RicdfnrmQyLuaWbH_o5ZXjEWMR0nHo0Q@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Use extended statistics to estimate (Var op Var) clauses (Tomas Vondra <tomas.vondra@enterprisedb.com>) |
Список | pgsql-hackers |
On Sun, Dec 12, 2021 at 6:04 PM Tomas Vondra <tomas.vondra@enterprisedb.com> wrote:
On 8/31/21 00:14, Zhihong Yu wrote:
> Hi,
> For patch 0002,
>
> + s1 = statext_clauselist_selectivity(root, clauses,
> varRelid,
> + jointype,
> sjinfo, rel,
> +
> &estimatedclauses, false);
> +
> + estimated = (bms_num_members(estimatedclauses) == 1);
>
> I took a look at clauselist_apply_dependencies() (called by
> statext_clauselist_selectivity) where estimatedclauses is modified.
> Since the caller would not use the returned Selectivity if number of
> elements in estimatedclauses is greater than 1, I wonder
> if a parameter can be added to clauselist_apply_dependencies() which
> indicates early return if the second element is added to estimatedclauses.
>
Hmmm, I'm not sure I understand your point. Are you suggesting there's a
bug in not updating the bitmap, or would this be an optimization? Can
you give an example of a query affected by this?
Hi,
My previous comment was from 3 months ago - let me see if I can come up with an example.
Cheers
regards
--
Tomas Vondra
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
В списке pgsql-hackers по дате отправления: