Horton

Re: Reducing memory consumed by RestrictInfo list translations in partitionwise join planning

Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>

View thread
From: Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>
To: Alena Rybakina <lena.ribackina@yandex.ru>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>, Richard Guo <guofenglinux@gmail.com>, David Rowley <dgrowleyml@gmail.com>
Date: 2023-11-24T10:54:28Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Add assertion to verify derived clause has constant RHS

  2. Make derived clause lookup in EquivalenceClass more efficient

  3. Doc: improve documentation for jsonpath behavior.

  4. Work around implementation restriction in adjust_appendrel_attrs.

On Fri, Nov 24, 2023 at 3:56 PM Alena Rybakina <lena.ribackina@yandex.ru> wrote:
>
> On 24.11.2023 13:20, Alena Rybakina wrote:
>
> Hi! Thank you for your work on the subject, I think it's a really useful feature.
>
> I've reviewed your patch and I have a few questions.
>
> First of all, have you thought about creating a gun parameter to display memory scheduling information? I agree that this is an important feature, but I think only for debugging.

Not a GUC parameter but I have a proposal to use EXPLAIN for the same. [1]

>
> Secondly, I noticed that for the child_rinfo_hash key you use a counter (int) and can it lead to collisions? Why didn't you generate a hash from childRestrictInfo for this? For example, something like how it is formed here [0].

Not usually. But that's the only "key" we have to access a set of
sematically same RestrictInfos. Relids is another key to access the
exact RestrictInfo. A child RestrictInfo can not be used since there
will many child RestrictInfos. Similar parent RestrictInfo can not be
used since there will be multiple forms of the same RestrictInfo.

[1] https://commitfest.postgresql.org/45/4492/

-- 
Best Wishes,
Ashutosh Bapat