Re: JIT breaks PostGIS - Mailing list pgsql-hackers

From Darafei "Komяpa" Praliaskouski
Subject Re: JIT breaks PostGIS
Date
Msg-id CAC8Q8tL4E9vVSq813PgW3bX1fotPTDX-mvEEH6rb8+YiKgeE1Q@mail.gmail.com
Whole thread Raw
In response to Re: JIT breaks PostGIS  (Andres Freund <andres@anarazel.de>)
List pgsql-hackers
Hi,

Here's somewhat minimized example.

https://gist.github.com/Komzpa/cc3762175328ff5d11de4b972352003d 

You can put this file into regress/jitbug.sql in PostGIS code tree and run after building postgis:

perl regress/run_test.pl regress/jitbug.sql --expect
perl regress/run_test.pl regress/jitbug.sql


сб, 21 июл. 2018 г. в 23:39, Andres Freund <andres@anarazel.de>:
Hi,

On 2018-07-21 22:36:44 +0200, Christoph Berg wrote:
> Re: Andres Freund 2018-07-21 <20180721202543.ri5jyfclj6kb6lag@alap3.anarazel.de>
> > Could you attempt to come up with a smaller reproducer?
>
> The original instructions in
> https://trac.osgeo.org/postgis/ticket/4125 are fairly easy to follow;
> there's also a postgresql-11-postgis-2.5{,-scripts} package (not
> mentioned in there) that exhibits the bug.

Sure, but a more minimal example (than a 1kloc regression script, wiht
possible inter statement dependencies) still makes the debugging
easier...

Greetings,

Andres Freund
--
Darafei Praliaskouski
Support me: http://patreon.com/komzpa

pgsql-hackers by date:

Previous
From: Fabien COELHO
Date:
Subject: Re: pgbench-ycsb
Next
From: Alexander Korotkov
Date:
Subject: Re: [HACKERS] Bug in to_timestamp().