From mboxrd@z Thu Jan  1 00:00:00 1970
From: ludovic.courtes@inria.fr (Ludovic =?utf-8?Q?Court=C3=A8s?=)
Subject: Re: Optionally using more advanced CPU features
Date: Mon, 04 Sep 2017 14:38:20 +0200
Message-ID: <87o9qq4p1f.fsf@inria.fr>
References: <87inhhw1ms.fsf@elephly.net> <87a82s9cw3.fsf@gnu.org>
	<87wp5ufkqs.fsf@albion.it.manchester.ac.uk> <87k21nerqa.fsf@inria.fr>
	<87inh2zog7.fsf@albion.it.manchester.ac.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Return-path: <guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org>
Received: from eggs.gnu.org ([2001:4830:134:3::10]:57728)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <ludovic.courtes@inria.fr>) id 1doqe4-0005wU-Se
	for guix-devel@gnu.org; Mon, 04 Sep 2017 08:38:33 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <ludovic.courtes@inria.fr>) id 1doqdz-0005lP-DC
	for guix-devel@gnu.org; Mon, 04 Sep 2017 08:38:28 -0400
In-Reply-To: <87inh2zog7.fsf@albion.it.manchester.ac.uk> (Dave Love's message
	of "Fri, 01 Sep 2017 11:46:16 +0100")
List-Id: "Development of GNU Guix and the GNU System distribution."
	<guix-devel.gnu.org>
List-Unsubscribe: <https://lists.gnu.org/mailman/options/guix-devel>,
	<mailto:guix-devel-request@gnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/archive/html/guix-devel/>
List-Post: <mailto:guix-devel@gnu.org>
List-Help: <mailto:guix-devel-request@gnu.org?subject=help>
List-Subscribe: <https://lists.gnu.org/mailman/listinfo/guix-devel>,
	<mailto:guix-devel-request@gnu.org?subject=subscribe>
Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org
Sender: "Guix-devel" <guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org>
To: Dave Love <fx@gnu.org>
Cc: guix-devel <guix-devel@gnu.org>

Hello,

Dave Love <fx@gnu.org> skribis:

> Ludovic Court=C3=A8s <ludovic.courtes@inria.fr> writes:

[...]

>> But that sounds similar to IFUNC in that application code would need to
>> actually use hwcap info to select the right implementation at load time,
>> right?
>
> As far as I know, it's a loader feature.  See "Hardware capabilities" in
> ld.so(1).

Indeed.

I=E2=80=99ve looked at the newish libmvec along with the =E2=80=9CVector AB=
I=E2=80=9D in the
toolchain and it=E2=80=99s really the kind of thing we=E2=80=99re looking f=
or.

> Well, you could embark on some sort of (GCC-specific?) re-write, but it
> would be better to work on <https://github.com/flame/blis/issues/129>.
> I don't think there's anywhere you can just attach GCC attributes, and
> certainly no magic will happen for currently-unsupported architectures.

Agreed, adjusting BLIS to do some load-time configuration seems like the
right thing.  Thanks for the reference, we=E2=80=99ll see how it goes.

Ludo=E2=80=99.