From: | Michael Paquier <michael(at)paquier(dot)xyz> |
---|---|
To: | Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com> |
Cc: | leiyanliang(at)highgo(dot)com, pgsql-bugs(at)lists(dot)postgresql(dot)org |
Subject: | Re: BUG #16972: parameter parallel_leader_participation's category problem |
Date: | 2021-04-21 10:44:58 |
Message-ID: | YIACKgzEThp6J81I@paquier.xyz |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
On Wed, Apr 21, 2021 at 09:15:23AM +0530, Bharath Rupireddy wrote:
> If we arrange only the "Asynchronous Behaviour" subsection in
> alphabetical order, I think the order may not be maintained in case of
> new GUCs that may get added there. Because all the other subsections
> are unordered and there's no note of maintaining the order as such.
> And, it looks like the relevant GUCs are grouped for better
> readability. For instance, all "parallelism", "io_concurrency", "jit_"
> related GUCs are together. Developers tend to add the new GUCs in
> relevant areas.
That's up to the committers adding them to be careful, but I of course
agree that the context is important. IMV, we can do a slightly better
organization in "Asynchronous Behaviour". First, backend_flush_after
is independent, and could just be first.
parallel_leader_participation can also be moved after
max_parallel_workers without impacting the readability nor impacting
the set of parallel-ish parameters grouped together.
--
Michael
From | Date | Subject | |
---|---|---|---|
Next Message | Bharath Rupireddy | 2021-04-21 11:37:37 | Re: BUG #16972: parameter parallel_leader_participation's category problem |
Previous Message | Thomas Munro | 2021-04-21 09:40:11 | Re: postgres has no spinlock support on riscv rv64imafdc |