Re: BUG #16122: segfault pg_detoast_datum (datum=0x0) at fmgr.c:1833 numrange query

On Tue, Nov 19, 2019 at 12:04:28PM +0900, Michael Paquier wrote:
>On Mon, Nov 18, 2019 at 06:22:11PM -0800, Adam Scott wrote:
>> Maybe a clue to the issue, why does a numrange need to be detoasted since
>> that is for large strings?
>
>I looked at the code a bit but could not really figure out what could
>go wrong. Then I switched my strategy and bisected for the first
>commit leading to the failure, and this is pointing me to the
>integration of stdbool.h for my environment (Debian SID with kernel at
>5.3):
>commit: 9a95a77d9d5d3003d2d67121f2731b6e5fc37336
>author: Peter Eisentraut <peter_e(at)gmx(dot)net>
>date: Thu, 22 Mar 2018 20:42:25 -0400
>Use stdbool.h if suitable
>
>Using the standard bool type provided by C allows some recent compilers
>and debuggers to give better diagnostics. Also, some extension code and
>third-party headers are increasingly pulling in stdbool.h, so it's
>probably saner if everyone uses the same definition.
>
>I have also checked that manually and I am getting a match. And the
>thing is that this commit did not make it through 10...

Not sure I understand. How could this be the root cause when it's not
present in the 10.x branch (which is what Adam is using)?

regards

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services