Re: BUG #13640: Delete isn't using available Index Only Scan

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: finzelj(at)gmail(dot)com
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: BUG #13640: Delete isn't using available Index Only Scan
Date: 2015-09-25 22:16:50
Message-ID: 24260.1443219410@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

finzelj(at)gmail(dot)com writes:
> Here is the actual delete query SQL. This doesn't but I am convinced should
> use the same Index Only Scan as above, and I wonder if it's a bug?

No, unfortunately not: the DELETE requires that each table scan return
CTID (for possible EvalPlanQual rechecks), and index-only scans can't
return columns that aren't in the index.

Although come to think of it, we necessarily get the TID from the index
AM, so in principle it should be possible for an index-only scan to
provide that column. But that's an unimplemented feature, not a bug.

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message it 2015-09-25 23:41:00 BUG #13641: PostgreSQL (pg_ctl.exe) conflicts with GPClient.exe
Previous Message Tom Lane 2015-09-25 20:44:07 Re: BUG #13638: Exception texts from plperl has bad encoding