Thread

Re: Why isn't PG using an index-only scan?

Andrei Lepikhov <lepihov@gmail.com> — 2025-09-19T07:50:41Z

On 19/9/2025 03:05, David Rowley wrote:
> On Thu, 18 Sept 2025 at 23:55, Andrei Lepikhov <lepihov@gmail.com> wrote:
>> Perhaps we should start working on introducing this type of callback/ hook?
> There's certainly places where you could add a hook that would just
> add an unacceptable overhead that we couldn't stomach. I expect
> ExecProcNode would be one of those places.
That seems a bit odd. It is not a significant issue to instruct the 
compiler to trust the 'else' branch of the "if (PlanState::hook != 
NULL)" condition and reduce overhead to zero. However, if the hook is 
set, it indicates that the user (module or extension) genuinely needs it 
to be called for this specific node.
> I do agree that trying to come up with something for this is worthy of
> some time and effort. Helping people with performance issues when they
> can't even get EXPLAIN ANALYZE to finish is quite tricky.That's all I need; thank you. Starting a PostgreSQL project without 
positive feedback from a committer usually results in wasted time, 
unless you are purely conducting research ;).

-- regards, Andrei Lepikhov