From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Axel Noltemeier" <axel(dot)noltemeier(at)gmx(dot)de> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: BUG #3067: Unnecessary lock blocks reindex |
Date: | 2007-02-25 18:15:14 |
Message-ID: | 9279.1172427314@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
"Axel Noltemeier" <axel(dot)noltemeier(at)gmx(dot)de> writes:
> -- This statement generates an unnecessary(?) lock
> -- on factory_pkey. That index is not used in the plan.
> EXPLAIN ANALYZE SELECT *
> FROM factory f
> Where f.id IN (select m.factory_id from machine m);
That behavior was changed here:
http://archives.postgresql.org/pgsql-committers/2006-07/msg00356.php
I'm disinclined to consider reverting that change...
It's possible that we could modify the planner to release locks at the
end of planning on indexes it chooses not to use, but 99.9% of the time
it'd be a waste of cycles to do that.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Eugene | 2007-02-26 11:54:33 | BUG #3079: pg_dump doesn't dump datascheme with uppercase letters in |
Previous Message | Alvaro Herrera | 2007-02-25 17:40:23 | Re: BUG #3064: In Stored Procedures (pgplgql |