Re: remaining sql/json patches - Mailing list pgsql-hackers

From Andres Freund
Subject Re: remaining sql/json patches
Date
Msg-id 20231129174235.rfddutj3r4ozdum4@awork3.anarazel.de
Whole thread Raw
In response to Re: remaining sql/json patches  (Andrew Dunstan <andrew@dunslane.net>)
Responses Re: remaining sql/json patches
List pgsql-hackers
Hi,

On 2023-11-29 07:37:53 -0500, Andrew Dunstan wrote:
> On 2023-11-28 Tu 21:10, Andres Freund wrote:
> > Hi,
> >
> > On 2023-11-28 20:58:41 -0500, Andrew Dunstan wrote:
> > > On 2023-11-28 Tu 19:32, Tom Lane wrote:
> > > > Andrew Dunstan <andrew@dunslane.net> writes:
> > > > So I'm now a bit baffled.  Can you provide more color on what
> > > > your test setup is?
> > >
> > > *sigh* yes, you're right. I inadvertently used a setup that used meson for
> > > building REL16_STABLE and HEAD. When I switch it to autoconf I get results
> > > that are similar to the earlier branches:
> > >
> > >
> > > ==== REL_16_STABLE ====
> > > Time: 3401.625 ms (00:03.402)
> > > ==== HEAD ====
> > > Time: 3419.088 ms (00:03.419)
> > >
> > >
> > > It's not clear to me why that should be. I didn't have assertions enabled
> > > anywhere. It's the same version of bison, same compiler throughout. Maybe
> > > meson sets a higher level of optimization? It shouldn't really matter, ISTM.
> > Is it possible that you have CFLAGS set in your environment? For reasons that
> > I find very debatable, configure.ac only adds -O2 when CFLAGS is not set:
> >
> > # C[XX]FLAGS are selected so:
> > # If the user specifies something in the environment, that is used.
> > # else:  If the template file set something, that is used.
> > # else:  If coverage was enabled, don't set anything.
> > # else:  If the compiler is GCC, then we use -O2.
> > # else:  If the compiler is something else, then we use -O, unless debugging.
> >
> > if test "$ac_env_CFLAGS_set" = set; then
> >    CFLAGS=$ac_env_CFLAGS_value
> > elif test "${CFLAGS+set}" = set; then
> >    : # (keep what template set)
> > elif test "$enable_coverage" = yes; then
> >    : # no optimization by default
> > elif test "$GCC" = yes; then
> >    CFLAGS="-O2"
> > else
> >    # if the user selected debug mode, don't use -O
> >    if test "$enable_debug" != yes; then
> >      CFLAGS="-O"
> >    fi
> > fi
> >
> > So if you have CFLAGS set in the environment, we'll not add -O2 to the
> > compilation flags.
> >
> > I'd check what the actual flags are when building a some .o.
> >
>
> I do have a CFLAGS setting, but for meson I used '-Ddebug=true' and no
> buildtype  or optimization setting. However, I see that in meson.build we're
> defaulting to "buildtype=debugoptimized" as opposed to the standard meson
> "buildtype=debug", so I guess that accounts for it.
>
> Still getting used to this stuff.

What I meant was whether you set CFLAGS for the *autoconf* build, because that
will result in an unoptimized build unless you explicitly add -O2 (or whatnot)
to the flags.  Doing benchmarking without compiler optimizations is pretty
pointless.

Greetings,

Andres Freund



pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: Whose Cirrus CI credits are used when making a PR to the GitHub mirror?
Next
From: "Tristan Partin"
Date:
Subject: Re: psql not responding to SIGINT upon db reconnection