Horton

Re: [postgresql 10 beta3] unrecognized node type: 90

Amit Kapila <amit.kapila16@gmail.com>

View thread View on postgr.es
From: Amit Kapila <amit.kapila16@gmail.com>
To: Noah Misch <noah@leadboat.com>
Cc: Robert Haas <robertmhaas@gmail.com>, "Adam, Etienne (Nokia-TECH/Issy Les Moulineaux)" <etienne.adam@nokia.com>, Tom Lane <tgl@sss.pgh.pa.us>, PostgreSQL Bugs <pgsql-bugs@postgresql.org>, "Duquesne, Pierre (Nokia-TECH/Issy Les Moulineaux)" <pierre.duquesne@nokia.com>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2017-08-15T11:31:44Z
Lists: pgsql-bugs, pgsql-hackers

Attachments

On Tue, Aug 15, 2017 at 8:22 AM, Noah Misch <noah@leadboat.com> wrote:
> On Fri, Aug 11, 2017 at 11:59:14AM -0400, Tom Lane wrote:
>> "Adam, Etienne (Nokia-TECH/Issy Les Moulineaux)" <etienne.adam@nokia.com> writes:
>> > ERROR:  XX000: unrecognized node type: 90
>> > LOCATION:  ExecReScan, execAmi.c:284
>>
>> (gdb) p (NodeTag) 90
>> $1 = T_GatherMergeState
>>
>> So, apparently somebody wrote ExecReScanGatherMerge, but never bothered
>> to plug it into ExecReScan.

Attached patch fixes the issue for me.  I have locally verified that
the gather merge gets executed in rescan path.  I haven't added a test
case for the same as having gather or gather merge on the inner side
of join can be time-consuming.  However, if you or others feel that it
is important to have a test to cover this code path, then I can try to
produce one.

-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com

Commits

  1. Separate reinitialization of shared parallel-scan state from ExecReScan.

  2. Restore test case from a2b70c89ca1a5fcf6181d3c777d82e7b83d2de1b.

  3. Force rescanning of parallel-aware scan nodes below a Gather[Merge].

  4. Fix ExecReScanGatherMerge.

  5. Add missing call to ExecReScanGatherMerge.