Re: Fix performance of generic atomics

On 2017-09-06 15:56, Tom Lane wrote:
> Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
>> On 5 September 2017 at 21:23, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>>> Moreover, it matters which primitive you're testing, on which
>>> platform,
>>> with which compiler, because we have a couple of layers of atomic ops
>>> implementations.
>
>> If there is no gain on 2-socket, at least there is no loss either.
>
> The point I'm trying to make is that if tweaking generic.h improves
> performance then it's an indicator of missed cases in the less-generic
> atomics code, and the latter is where our attention should be focused.
> I think basically all of the improvement Sokolov got was from upgrading
> the coverage of generic-gcc.h.
>
> regards, tom lane

Not exactly. I've checked, that new version of generic
pg_atomic_fetch_or_u32
loop also gives improvement. Without that check I'd not suggest to fix
generic atomic functions. Of course, gcc intrinsic gives more gain.

--
Sokolov Yura aka funny_falcon
Postgres Professional: https://postgrespro.ru
The Russian Postgres Company