Re: SI-read predicate locks on materialized views

Hello Micheal-san,

On Thu, 13 Oct 2022 17:02:06 +0900
Michael Paquier <michael@paquier.xyz> wrote:

> On Fri, Sep 30, 2022 at 10:12:13AM +0900, Yugo NAGATA wrote:
> > Thank you for comment. Do you think it can be marked as Ready for Commiter?
> 
> Matviews have been discarded from needing predicate locks since
> 3bf3ab8 and their introduction, where there was no concurrent flavor
> of refresh yet.  Shouldn't this patch have at least an isolation test
> to show the difference in terms of read-write conflicts with some
> serializable transactions and REFRESH CONCURRENTLY?

Thank you for your review. I agree that an isolation test is required.
The attached patch contains the test using the scenario as explained in
the previous post.

Regards,
Yugo Nagata

-- 
Yugo NAGATA <nagata@sraoss.co.jp>