Horton

Re: BUG #18483: Segmentation fault in tests modules

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: a.kozhemyakin@postgrespro.ru, pgsql-bugs@lists.postgresql.org
Cc: Masahiko Sawada <sawada.mshk@gmail.com>, johncnaylorls@gmail.com
Date: 2024-06-03T22:40:11Z
Lists: pgsql-bugs 
On Tue, May 28, 2024 at 03:15:06AM +0000, PG Bug reporting form wrote:

Thanks for the report!  My apologies for the late reply but most
people have been busy with pgconf.dev last week.

> second case
> echo "CREATE EXTENSION test_tidstore;
> SELECT do_set_block_offsets(1, ARRAY[0]::int2[]);" >
> src/test/modules/test_tidstore/sql/test_tidstore.sql
> make check -C src/test/modules/test_tidstore
> 
> Core was generated by `postgres: postgres contrib_regression [local] SELECT 
>                         '.
> Program terminated with signal SIGSEGV, Segmentation fault.
> #0  TidStoreLockExclusive (ts=0x0) at tidstore.c:300
> 300             if (TidStoreIsShared(ts))
> (gdb) bt
> #0  TidStoreLockExclusive (ts=0x0) at tidstore.c:300
> #1  0x00007f84c48aa8a1 in do_set_block_offsets (fcinfo=<optimized out>) at

Yes, something is wrong here.  I have added an open item to track
that.

John and Sawada-san (added in CC), could you look at that?  The item
has been assigned to both of you.
--
Michael

Commits

  1. Make regress function make_tuple_indirect() able to handle plain attributes

  2. Fix segmentation fault in test_tidstore.

  3. Tighten test_predtest's input checks, and improve error messages.