| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Konstantin Knizhnik <knizhnik(at)garret(dot)ru> |
| Cc: | Peter Geoghegan <pg(at)bowt(dot)ie>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: Orphan page in _bt_split |
| Date: | 2025-09-25 06:45:21 |
| Message-ID: | aNTlAbS0By8kTcKJ@paquier.xyz |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Sep 25, 2025 at 09:41:00AM +0300, Konstantin Knizhnik wrote:
> Sorry, I have attached wrong patch.
Thanks, I was confused for a couple of minutes.
+ /*
+ * Now we are in critical section and it is not needed to maintain temporary
+ * copy of right page in the local memory. We can copy it to the destination buffer.
+ * Unlike leftpage, rightpage and ropaque are still needed to complete update
+ * of the page, so we need to correctly reinitialize them.
+ */
+ rightpage = BufferGetPage(rbuf);
+ memcpy(rightpage, rightpage_buf.data, BLCKSZ);
+ ropaque = BTPageGetOpaque(rightpage);
Hmm. This looks kind of explicit enough to document the purpose.
The wording could be simplified a bit more. I'll take it from there.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | shveta malik | 2025-09-25 06:53:40 | Re: Logical Replication of sequences |
| Previous Message | Michael Paquier | 2025-09-25 06:42:33 | Re: Get rid of pgstat_count_backend_io_op*() functions |