Re: Confine vacuum skip logic to lazy_scan_skip

On Thu, Feb 13, 2025 at 8:30 PM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com> wrote:
>
> On Thu, Feb 13, 2025 at 4:55 PM Melanie Plageman
> <melanieplageman(at)gmail(dot)com> wrote:
> >
> > We don't want to hang onto that pin for a
> > long time. But I can't move them to the bottom of the loop after we
> > release the buffer because some of the code paths don't make it that
> > far. I don't see a good way other than how I did it or special-casing
> > block 0. What do you think?
>
> How about adding 'vacrel->scanned_pages > 0' to the if statement?
> Which seems not odd to me.

Cool. I've done this in attached v19.

> Looking at the 0002 patch, it seems you reverted the change to the
> following comment:
>
> /*
> * Vacuum the Free Space Map to make newly-freed space visible on
> - * upper-level FSM pages. Note we have not yet processed blkno.
> + * upper-level FSM pages. Note we have not yet processed blkno+1.
> */
>
> I feel that the previous change I saw in v17 is clearer:

I've reverted to the old comment. Thanks

> The rest looks good to me.

Cool! I'll plan to push this tomorrow barring any objections.

- Melanie