Re: Optimize SnapBuildPurgeOlderTxn: use in-place compaction instead of temporary array

On Tue, 21 Oct 2025 at 04:31, Michael Paquier <michael(at)paquier(dot)xyz> wrote:
>
> On Sat, Oct 18, 2025 at 01:59:40PM +0500, Kirill Reshke wrote:
> > Indeed, these changes look correct.
> > I wonder why b89e151054a0 did this place this way, hope we do not miss
> > anything here.
>
> Perhaps a lack of time back in 2014? It feels like an item where we
> would need to research a bit some of the past threads, and see if this
> has been discussed, or if there were other potential alternatives
> discussed. This is not saying that what you are doing in this
> proposal is actually bad, but it's a bit hard to say what an
> "algorithm" should look like in this specific code path with XID
> manipulations. Perhaps since 2014, we may have other places in the
> tree that share similar characteristics as what's done here.
>
> So it feels like this needs a bit more historical investigation first,
> rather than saying that your proposal is the best choice on the table.
> --
> Michael

Sure

Commit b89e151054a0 comes from [0]
Comment of SnapBuildPurgeCommittedTxn tracks to [1] (it was in form
"XXX: Neater algorithm?")

Between these two messages, it was not disucccesseed...

I will also study other related threads like [2], but i don't think
they will give more insight here.

[0] https://www.postgresql.org/message-id/20140303162652.GB16654%40awork2.anarazel.de
[1] https://www.postgresql.org/message-id/20140115002223.GA17204%40awork2.anarazel.de
[2] https://www.postgresql.org/message-id/flat/20130914204913.GA4071%40awork2.anarazel.de#:~:text=20130914204913.GA4071%40awork2.anarazel.de

--
Best regards,
Kirill Reshke