Horton

Re: [PATCH] Push limit to sort through a subquery

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Amit Kapila <amit.kapila16@gmail.com>
Cc: Robert Haas <robertmhaas@gmail.com>, Douglas Doole <dougdoole@gmail.com>, Konstantin Knizhnik <k.knizhnik@postgrespro.ru>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2017-08-25T14:58:41Z
Lists: pgsql-hackers 
Amit Kapila <amit.kapila16@gmail.com> writes:
> On Fri, Aug 25, 2017 at 7:35 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> it would be stupid to put a filter on that node rather than its
>> children, but I see this in both nodeGather.c and nodeGatherMerge.c:
>>
>> gatherstate->ps.qual =
>> ExecInitQual(node->plan.qual, (PlanState *) gatherstate);
>> 
>> It doesn't look like the qual is actually used anywhere in either node
>> type.  Am I right in thinking this is dead code?

> I also think so.  I think this was required in some initial versions
> of gather node patch where we were thinking of having a single node
> (instead of what we have now that Gather node and beneath there will
> be partial scan node) to perform parallel scans.

Thanks for confirming.  I'll replace that with something like
Assert(!node->plan.qual).  I have some other code-review-ish
fixes to do in nodeGatherMerge, too.

			regards, tom lane

Commits

  1. Propagate sort instrumentation from workers back to leader.

  2. Push tuple limits through Gather and Gather Merge.

  3. Code review for pushing LIMIT through subqueries.

  4. Make new regression test case parallel-safe, and improve its output.

  5. Push limit through subqueries to underlying sort, where possible.

  6. Move targetlist SRF handling from expression evaluation to new executor node.