Re: PostgreSQL crashes with SIGSEGV

Peter Geoghegan <pg@bowt.ie>

From: Peter Geoghegan <pg@bowt.ie>

To: Tom Lane <tgl@sss.pgh.pa.us>

Cc: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>, Aleksandr Parfenov <a.parfenov@postgrespro.ru>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>

Date: 2018-03-28T17:33:54Z

Lists: pgsql-bugs, pgsql-hackers

On Wed, Mar 28, 2018 at 10:30 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> It looks good to me.  The only real objection would be if someone came
> up with a test case proving that there's a significant performance
> degradation from the extra copies.  But given that these are back
> branches, it would take a pretty steep penalty for me to want to take
> the risk of refactoring to avoid that.
>
> I've pushed it with some cosmetic adjustments.

Thank you, Tom.

-- 
Peter Geoghegan

Commits

Fix actual and potential double-frees around tuplesort usage.
- c2d4eb1b1fa2 11.0 landed
- e4ff711582b1 9.5.13 landed
- c98f218fbf5a 10.4 landed
- 90decdba3786 9.6.9 landed
Allow the built-in ordered-set aggregates to share transition state.
- be0ebb65f512 11.0 cited
Avoid integer overflow while sifting-up a heap in tuplesort.c.
- 512f67c8d02c 10.0 cited
Allow avoiding tuple copy within tuplesort_gettupleslot().
- fa117ee40330 10.0 cited
Remove should_free arguments to tuplesort routines.
- 3856cf9607f4 10.0 cited
Fix use-after-free around DISTINCT transition function calls.
- a5f0bd77a2fa 9.6.1 cited
- d8589946ddd5 10.0 cited
Reuse abbreviated keys in ordered [set] aggregates.
- f1f5ec1efafe 9.6.0 cited