Horton

Re: [sqlsmith] Planner crash on foreign table join

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andrew Gierth <andrew@tao11.riddles.org.uk>, Andreas Seltenreich <seltenreich@gmx.de>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>, Simon Riggs <simon@2ndquadrant.com>
Date: 2017-04-08T21:01:06Z
Lists: pgsql-hackers 
On Sat, Apr 8, 2017 at 3:57 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> This makes me wonder whether we were being penny-wise and pound-foolish
> by not making Bitmapsets be a kind of Node, so that there could be IsA
> assertions in the bitmapset.c routines, as there are for Lists.  Most
> Bitmapsets in a typical backend probably have only one payload word
> (ie highest member < 32), so right now they occupy 8 bytes.  Adding
> a nodetag would kick them up to the next palloc category, 16 bytes,
> which is probably why I didn't do it like that to begin with.
> Still, that decision is looking unduly byte-miserly in 2017.

I think it's pretty dubious to change this, honestly.  Just because it
would have caught this one bug doesn't make it an especially valuable
thing in general.  Bytes are still not free.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Commits

  1. Improve castNode notation by introducing list-extraction-specific variants.

  2. Clean up bugs in clause_selectivity() cleanup.

  3. Add newly-symlinked files to "make clean" target.

  4. Reset API of clause_selectivity()