Thread

Re: Add pg_buffercache_mark_dirty[_all] functions to the pg_buffercache

Nazir Bilal Yavuz <byavuz81@gmail.com> — 2025-11-28T08:27:11Z

Hi,

On Fri, 28 Nov 2025 at 03:36, Michael Paquier <michael@paquier.xyz> wrote:
>
> On Thu, Nov 27, 2025 at 10:59:47AM +0300, Nazir Bilal Yavuz wrote:
> > I agree with you, the patches make more sense this way.
> >
> > The patches are split into two in v10. There are no changes from v9,
> > except that one extra blank line was removed [1].
>
> Note that there were a couple of things incorrect in the docs.  I have
> done a sweep to improve the wording in the comments and the docs
> themselves, then applied the result.

Thanks for doing this!

> Testing the valid case for the "_all" function flavor could be costly
> for installcheck so I am feeling a bit reserved on its cost.  We are
> doing it for the evict case as well, so I have kept it at the end to
> keep the coverage.  If it proves to be an issue or if there is a
> concern with this part, I would be OK to remove it.

You are right, I did not think of this aspect.

-- 
Regards,
Nazir Bilal Yavuz
Microsoft