Re: Accidentally parallel unsafe functions

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Andreas Karlsson <andreas(at)proxel(dot)se>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Accidentally parallel unsafe functions
Date: 2016-05-02 14:42:00
Message-ID: CA+TgmoZvCqwEYXBsiJcYvjri7=i_9OwmsHW7PhcVoCms3mF6mw@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, Apr 29, 2016 at 6:06 PM, Andreas Karlsson <andreas(at)proxel(dot)se> wrote:
> I am currently looking into adding the correct parallel options to all
> functions in the extensions and I noticed that some built-in functions seems
> to have been marked as unsafe by accident. The main culprit is
> system_views.sql which redefines these functions and removes the parallel
> safe flag.
>
> I think this counts as a 9.6 bug unlike my work on adding the flags to all
> extensions which is for 9.7.
>
> I have attached a patch which marks them and all conversion functions as
> parallel safe. I also added the flag to ts_debug() when I was already
> editing system_views.sql, feel free to ignore that one if you like.
>
> Affected functions:
>
> - json_populate_record()
> - json_populate_recordset()
> - jsonb_insert()
> - jsonb_set()
> - make_interval()
> - parse_ident()
> - Loads of conversion functions

Hmm. The new pg_start_backup() is not parallel-safe. It's
parallel-restricted, because it relies on backend-private state. I'll
go fix that.

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

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Ants Aasma 2016-05-02 15:15:40 Re: Re: [COMMITTERS] pgsql: Avoid extra locks in GetSnapshotData if old_snapshot_threshold <
Previous Message Tom Lane 2016-05-02 14:32:44 Re: 9.6 and fsync=off