Horton

Re: inefficient loop in StandbyReleaseLockList()

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
Cc: bossartn@amazon.com, Kyotaro Horiguchi <horikyota.ntt@gmail.com>, michael@paquier.xyz, sulamul@gmail.com, pgsql-hackers@postgresql.org
Date: 2021-11-06T22:32:54Z
Lists: pgsql-hackers 
Andres Freund <andres@anarazel.de> writes:
> On 2021-11-06 14:06:12 -0400, Tom Lane wrote:
>> + * Note that this takes time proportional to the length of the list,
>> + * since the remaining entries must be moved.
>> */
>> List *
>> list_delete_first(List *list)

> Perhaps we could point to list_delete_last()? But it's an improvement without
> that too.

Good point.  The note at list_delete_last that it's O(1) isn't really
on point --- instead, the text for list_delete_first should be like

+ * Note that this takes time proportional to the length of the list,
+ * since the remaining entries must be moved.  Consider reversing the
+ * list order so that you can use list_delete_last() instead.  However,
+ * if that causes you to replace lappend() with lcons(), you haven't
+ * improved matters.

			regards, tom lane

Commits

  1. Doc: add some notes about performance of the List functions.

  2. Avoid O(N^2) behavior in SyncPostCheckpoint().

  3. Avoid some other O(N^2) hazards in list manipulation.

  4. Avoid O(N^2) behavior when the standby process releases many locks.