| From: | Peter Geoghegan <pg(at)bowt(dot)ie> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Klaudie Willis <Klaudie(dot)Willis(at)protonmail(dot)com>, PostgreSQL mailing lists <pgsql-bugs(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: When Update balloons memory |
| Date: | 2022-01-11 21:15:34 |
| Message-ID: | CAH2-WznKE7uFnXRGw+U0DpXcmho0_s6Otd9dqGTcOnsvqJMZyA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs pgsql-general |
On Tue, Jan 11, 2022 at 11:59 AM Peter Geoghegan <pg(at)bowt(dot)ie> wrote:
> On Tue, Jan 11, 2022 at 11:54 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > Pass NULL for that, per 6867f963e:
> >
> > https://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=6867f963e#patch2
>
> Will look into that.
Took a look. Not sure that using pull_varnos would represent an
improvement, though. Do you feel strongly about it?
The current approach has the advantage of allowing
index_expression_changed_walker to directly apply an
FirstLowInvalidHeapAttributeNumber offset to each Var's varattno. We
can't just use bms_overlap to check for overlap between the bms
returned by pull_varnos and our updatedColumns bms. It would be
possible to transform the bms returned by pull_varnos to compensate,
of course, but that seems rather awkward
--
Peter Geoghegan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Peter Geoghegan | 2022-01-11 22:19:38 | Re: BUG #17362: Error "could not find block containing chunk" when using index with icu collation on CentOS 7 |
| Previous Message | Peter Geoghegan | 2022-01-11 19:59:12 | Re: When Update balloons memory |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kyotaro Horiguchi | 2022-01-12 01:18:05 | Re: [Ext:] Re: Stream Replication not working |
| Previous Message | Peter Geoghegan | 2022-01-11 19:59:12 | Re: When Update balloons memory |