Re: optimize lookups in snapshot [sub]xip arrays - Mailing list pgsql-hackers

From Masahiko Sawada
Subject Re: optimize lookups in snapshot [sub]xip arrays
Date
Msg-id CAD21AoA_MEAENt9yaS5-rQ+_4yW4+GTYN16Jy0ntY+o8o7TtKg@mail.gmail.com
Whole thread Raw
In response to Re: optimize lookups in snapshot [sub]xip arrays  (Nathan Bossart <nathandbossart@gmail.com>)
Responses Re: optimize lookups in snapshot [sub]xip arrays
List pgsql-hackers
On Tue, Aug 9, 2022 at 7:33 AM Nathan Bossart <nathandbossart@gmail.com> wrote:
>
> On Mon, Aug 08, 2022 at 01:46:48PM +0700, John Naylor wrote:
> > Okay, I think it's basically in good shape. Since it should be a bit
> > faster than a couple versions ago, would you be up for retesting with
> > the original test having 8 to 512 writers?
>
> Sure thing.  The results are similar.  As before, the improvements are most
> visible when the arrays are large.
>
>         writers  head  patch
>         8        672   680
>         16       639   664
>         32       701   689
>         64       705   703
>         128      628   653
>         256      576   627
>         512      530   584
>         768      450   536
>         1024     350   494
>
> > And also add the const
> > markers we discussed upthread?
>
> Oops, sorry about that.  This is done in v9.
>
> > Aside from that, I plan to commit this
> > week unless there is further bikeshedding.
>
> Great, thanks.

The patch looks good to me. One minor point is:

+ * IDENTIFICATION
+ *   src/port/pg_lfind.h

The path doesn't match to the actual file path, src/include/port/pg_lfind.h.

Regards,


--
Masahiko Sawada
EDB:  https://www.enterprisedb.com/



pgsql-hackers by date:

Previous
From: Masahiko Sawada
Date:
Subject: Fix unmatched file identifications
Next
From: John Naylor
Date:
Subject: Re: Fix unmatched file identifications