| From: | David Rowley <dgrowleyml(at)gmail(dot)com> |
|---|---|
| To: | nospam-pg-abuse(at)bloodgate(dot)com |
| Cc: | PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: Speed up Hash Join by teaching ExprState about hashing |
| Date: | 2024-08-19 06:41:58 |
| Message-ID: | CAApHDvpk6yH-t72H7qtpnd_vaT_0OJPdWbunFK6Jp-m8PCxkkA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Thanks for having a look.
On Sat, 17 Aug 2024 at 23:21, Tels <nospam-pg-abuse(at)bloodgate(dot)com> wrote:
> Is it nec. to rotate existing_hash here before checking for isnull?
> Because in case of isnull, isn't the result of the rotate thrown away?
Yeah, I think that it's worthwhile moving that to after the isnull
check so as not to waste the effort. Unfortunately doing the same in
the JIT code meant copying and pasting the code that emits the
rotation code.
The attached v5 patch includes this change.
David
| Attachment | Content-Type | Size |
|---|---|---|
| v5-0001-Speed-up-Hash-Join-by-making-ExprStates-support-h.patch | application/octet-stream | 40.8 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Heikki Linnakangas | 2024-08-19 06:49:41 | Re: Refactoring postmaster's code to cleanup after child exit |
| Previous Message | Amit Kapila | 2024-08-19 06:39:43 | Re: Conflict detection and logging in logical replication |