Re: [PoC] Improve dead tuple storage for lazy vacuum

On Fri, Dec 2, 2022 at 11:42 PM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
wrote:
>
> > On Mon, Nov 14, 2022 at 7:59 PM John Naylor <
john(dot)naylor(at)enterprisedb(dot)com> wrote:
> > >
> > > - Optimize node128 insert.
> >
> > I've attached a rough start at this. The basic idea is borrowed from
our bitmapset nodes, so we can iterate over and operate on word-sized (32-
or 64-bit) types at a time, rather than bytes.
>
> Thanks! I think this is a good idea.
>
> > To make this easier, I've moved some of the lower-level macros and
types from bitmapset.h/.c to pg_bitutils.h. That's probably going to need a
separate email thread to resolve the coding style clash this causes, so
that can be put off for later.

I started a separate thread [1], and 0002 comes from feedback on that.
There is a FIXME about using WORDNUM and BITNUM, at least with that
spelling. I'm putting that off to ease rebasing the rest as v13 -- getting
some CI testing with 0002 seems like a good idea. There are no other
changes yet. Next, I will take a look at templating local vs. shared
memory. I might try basing that on the styles of both v12 and v8, and see
which one works best with templating.

[1]
https://www.postgresql.org/message-id/CAFBsxsFW2JjTo58jtDB%2B3sZhxMx3t-3evew8%3DAcr%2BGGhC%2BkFaA%40mail.gmail.com

--
John Naylor
EDB: http://www.enterprisedb.com