From: | Peter Geoghegan <pg(at)heroku(dot)com> |
---|---|
To: | Kevin Grittner <kgrittn(at)gmail(dot)com> |
Cc: | Andres Freund <andres(at)anarazel(dot)de>, Robert Haas <robertmhaas(at)gmail(dot)com>, Andreas Seltenreich <seltenreich(at)gmx(dot)de>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: [PATCH] Improve spinlock inline assembly for x86. |
Date: | 2016-01-18 22:53:07 |
Message-ID: | CAM3SWZQiha9kA9tc-tuLRBbHHJUy90Sks6dKAivbbWtQd3aMkw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Mon, Jan 18, 2016 at 2:39 PM, Kevin Grittner <kgrittn(at)gmail(dot)com> wrote:
>> but I don't think that Andreas' patch is necessarily a
>> performance patch. There can be value in removing superfluous
>> code; doing so sometimes clarifies intent and understanding.
>
> Well, that's why I said I would be satisfied with a neutral
> benchmark result -- when there is a tie, the shorter, simpler code
> should generally win. I'm not really sure what there was in what I
> said to argue about; since that I've just been trying figure that
> out. If we all agree, let's let it drop.
If we don't want to apply this, then I think that a sensible
compromise would be to add a code comment that says that we don't
believe the LOCK prefix matters.
--
Peter Geoghegan
From | Date | Subject | |
---|---|---|---|
Next Message | Kevin Grittner | 2016-01-18 22:56:22 | Re: [PATCH] Improve spinlock inline assembly for x86. |
Previous Message | Andres Freund | 2016-01-18 22:50:47 | Re: [PATCH] Improve spinlock inline assembly for x86. |