On Wed, Oct 12, 2022 at 10:50 PM Peter Eisentraut
<peter.eisentraut@enterprisedb.com> wrote:
> On 06.10.22 15:29, Amit Langote wrote:
> > I tried in the attached 0004. ModifyTable gets a new member
> > extraUpdatedColsBitmaps, which is List of Bitmapset "nodes".
> >
> > Actually, List of Bitmapsets turned out to be something that doesn't
> > just-work with our Node infrastructure, which I found out thanks to
> > -DWRITE_READ_PARSE_PLAN_TREES. So, I had to go ahead and add
> > first-class support for copy/equal/write/read support for Bitmapsets,
> > such that writeNode() can write appropriately labeled versions of them
> > and nodeRead() can read them as Bitmapsets. That's done in 0003. I
> > didn't actually go ahead and make*all* Bitmapsets in the plan trees
> > to be Nodes, but maybe 0003 can be expanded to do that. We won't need
> > to make gen_node_support.pl emit *_BITMAPSET_FIELD() blurbs then; can
> > just use *_NODE_FIELD().
>
> Seeing that on 64-bit platforms we have a 4-byte padding gap in the
> Bitmapset struct, sticking a node tag in there seems pretty sensible.
> So turning Bitmapset into a proper Node and then making the other
> adjustments you describe makes sense to me.
>
> Making a new thread about this might be best.
>
> (I can't currently comment on the rest of the patch set. So I don't
> know if you'll really end up needing lists of bitmapsets. But from here
> it looks like turning bitmapsets into nodes might be a worthwhile change
> just by itself.)
Ok, thanks. I'll start a new thread about it.
--
Thanks, Amit Langote
EDB: http://www.enterprisedb.com
