From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugzilla@gcc.gnu.org>
Received: by sourceware.org (Postfix, from userid 48)
	id A37463858C39; Mon, 12 Jun 2023 08:06:37 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org A37463858C39
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1686557197;
	bh=8pIp9V+UF6ercyq68RiWbQ5sPFj1lmL4UrEJybbj40c=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=oIwY3BGlz9JnQ6wFpQA6DT39jVK1Vu9U7XDqyWEhEnE1xTeYwLmdH7qP8C+xmtxT7
	 t85AcW7Xq8sKNbs+8c8HClmQ3GOgNMPfJM0smLWyi+AkD06ukxS094OYFQdL8rVcLL
	 S3zFaIQpf7vhvVNIRg5lFSSi8/tcY3532kBXQYOg=
From: "rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/108410] x264 averaging loop not optimized well for
 avx512
Date: Mon, 12 Jun 2023 08:06:37 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: 13.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenther at suse dot de
X-Bugzilla-Status: NEW
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-108410-4-ajTp5zPsuU@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108410-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108410-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
List-Id: <gcc-bugs.sourceware.org>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108410

--- Comment #7 from rguenther at suse dot de <rguenther at suse dot de> ---
On Mon, 12 Jun 2023, crazylht at gmail dot com wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108410
>=20
> --- Comment #6 from Hongtao.liu <crazylht at gmail dot com> ---
>=20
> > and the key thing to optimize is
> >=20
> >   ivtmp_78 =3D ivtmp_77 + 4294967232; // -64
> >   _79 =3D MIN_EXPR <ivtmp_78, 255>;
> >   _80 =3D (unsigned char) _79;
> >   _81 =3D {_80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, =
_80,
> > _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _=
80,
> > _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _=
80,
> > _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _80, _=
80,
> > _80, _80, _80, _80, _80, _80};
> >=20
> > that is we want to broadcast a saturated (to vector element precision) =
value.
>=20
> Yes, backend needs to support vec_pack_ssat_m, vec_pack_usat_m.

Can x86 do this?  We'd want to apply this to a scalar, so move ivtmp
to xmm, apply pack_usat or as you say below, the non-existing us_trunc
and then broadcast.=