Re: Shared detoast Datum proposal - Mailing list pgsql-hackers

From Tomas Vondra
Subject Re: Shared detoast Datum proposal
Date
Msg-id d2e3ef5c-ac09-4b10-a99f-4c964b490f8c@enterprisedb.com
Whole thread Raw
In response to Re: Shared detoast Datum proposal  (Andy Fan <zhihuifan1213@163.com>)
Responses Re: Shared detoast Datum proposal
List pgsql-hackers
On 3/3/24 07:10, Andy Fan wrote:
> 
> Hi,
> 
> Here is a updated version, the main changes are:
> 
> 1. an shared_detoast_datum.org file which shows the latest desgin and
> pending items during discussion. 
> 2. I removed the slot->pre_detoast_attrs totally.
> 3. handle some pg_detoast_datum_slice use case.
> 4. Some implementation improvement.
> 

I only very briefly skimmed the patch, and I guess most of my earlier
comments still apply. But I'm a bit surprised the patch needs to pass a
MemoryContext to so many places as a function argument - that seems to
go against how we work with memory contexts. Doubly so because it seems
to only ever pass CurrentMemoryContext anyway. So what's that about?

regards

-- 
Tomas Vondra
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company



pgsql-hackers by date:

Previous
From: Tomas Vondra
Date:
Subject: Re: Shared detoast Datum proposal
Next
From: Tom Lane
Date:
Subject: Re: pgsql: Improve performance of subsystems on top of SLRU