From: Tom Lane <tgl@sss.pgh.pa.us>
To: Thomas Munro <thomas.munro@gmail.com>
cc: Andres Freund <andres@anarazel.de>, Andrew Dunstan <andrew@dunslane.net>,
        Robert Haas <robertmhaas@gmail.com>,
        pgsql-hackers <pgsql-hackers@postgresql.org>
Subject: Re: Cleaning up historical portability baggage
In-reply-to: 
 <CA+hUKGK=N8W2uFPdJr07Rs4dgRtbzKaYvCaPQasX3S8=aTO5PQ@mail.gmail.com>
References: 
 <CA+hUKGJ3LHeP9w5Fgzdr4G8AnEtJ=z=p6hGDEm4qYGEUX5B6fQ@mail.gmail.com>
 <CA+Tgmoazusui5EoU_7Z=AgwC5pTAqDndTa7d-owjvQwG3zoRhg@mail.gmail.com>
 <CA+hUKGLLJUMDgCzOK5K4NE6PpU87Ka7t7RQsZoZAZ_EKfL5L+A@mail.gmail.com>
 <CA+Tgmoaw6VBC9Xzr4vOaC0kc-94jb9nzH8MY38Gqn7=wy=i_3w@mail.gmail.com>
 <20220712173352.trdeq66asbv6vf2q@awork3.anarazel.de>
 <1662559.1657648462@sss.pgh.pa.us>
 <CA+hUKG+MOzmDkr27ceG8+uwgFxbJUUv699V3VP3DCioQigrf7A@mail.gmail.com>
 <1249825.1658622186@sss.pgh.pa.us>
 <CA+hUKGLryBW4fNYoXzMG9y1e0H-QOTdSiTG46xCr1DNuuPwbeg@mail.gmail.com>
 <CA+hUKGJMcwbzRyfTocfxbPBUhERqKE42d-L22jQ+vpB7an-NjQ@mail.gmail.com>
 <20220804034343.unm4rdynedbigamv@awork3.anarazel.de>
 <3322724.1659586159@sss.pgh.pa.us>
 <CA+hUKGJ5ej33AJL19sQph7Z8SFZ6+aUSKxFTS6pvk2Y8J9Tcwg@mail.gmail.com>
 <295419.1659918447@sss.pgh.pa.us>
 <CA+hUKGK=N8W2uFPdJr07Rs4dgRtbzKaYvCaPQasX3S8=aTO5PQ@mail.gmail.com>
Comments: In-reply-to Thomas Munro <thomas.munro@gmail.com>
	message dated "Tue, 09 Aug 2022 11:35:05 +1200"
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-ID: <747619.1660002360.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Mon, 08 Aug 2022 19:46:00 -0400
Message-ID: <747620.1660002360@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/747620.1660002360%40sss.pgh.pa.us>
Precedence: bulk

Thomas Munro <thomas.munro@gmail.com> writes:
> Yeah, that's not true anymore, and QueryPerformanceCounter() is faster
> than Get=C2=ADSystem=C2=ADTime=C2=ADPrecise=C2=ADAs=C2=ADFile=C2=ADTime(=
)[1], but there doesn't
> really seem to be any point in mentioning that or gettimeofday() at
> all here.  I propose to cut it down to just:

>   * This file provides an abstraction layer to hide portability issues i=
n
> - * interval timing.  On Unix we use clock_gettime() if available, else
> - * gettimeofday().  On Windows, gettimeofday() gives a low-precision re=
sult
> - * so we must use QueryPerformanceCounter() instead.  These macros also=
 give
> - * some breathing room to use other high-precision-timing APIs.
> + * interval timing.  On Unix we use clock_gettime(), and on Windows we =
use
> + * QueryPerformanceCounter().  These macros also give some breathing ro=
om to
> + * use other high-precision-timing APIs.

WFM.

> FWIW I expect this stuff to get whacked around some more for v16[2].
> [2] https://commitfest.postgresql.org/39/3751/

Meh.  I think trying to use rdtsc is a fool's errand; you'll be fighting
CPU quirks forever.

			regards, tom lane