From: | Andrew Gierth <andrew(at)tao11(dot)riddles(dot)org(dot)uk> |
---|---|
To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Vik Fearing <vik(dot)fearing(at)2ndquadrant(dot)com>, Stephen Frost <sfrost(at)snowman(dot)net>, Andres Freund <andres(at)anarazel(dot)de>, gajus(at)gajus(dot)com, "pgsql-hackers\(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: IMMUTABLE and PARALLEL SAFE function markings |
Date: | 2018-11-27 04:20:24 |
Message-ID: | 87y39fnosv.fsf@news-spur.riddles.org.uk |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
>>>>> "Robert" == Robert Haas <robertmhaas(at)gmail(dot)com> writes:
>> I'm a bit more concerned by the fact that inlining the function
>> isn't affecting the parallel safety of the query - the fact that
>> parallel safety is being checked prior to inlining means that if
>> inlining *introduces* a parallel hazard, it will go unnoticed?
Robert> If a function is marked parallel-safe but internally calls
Robert> parallel-restricted or parallel-unsafe functions, it wasn't
Robert> really parallel-safe in the first place. So I think that if
Robert> inlining introduces a parallel hazard, the user has mislabeled
Robert> some functions and any resulting injury is self-inflicted.
But the combination of inlining and polymorphism, in particular, makes
it impossible for the function author to know this. Take the OP's
example; it is parallel safe if and only if the selected type's equal
function is parallel safe - how is the author supposed to know? What if
the type is one installed later?
--
Andrew (irc:RhodiumToad)
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2018-11-27 04:21:42 | Re: IMMUTABLE and PARALLEL SAFE function markings |
Previous Message | Robert Haas | 2018-11-27 04:04:35 | Re: Remove Deprecated Exclusive Backup Mode |