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 4439F389803F; Wed, 18 Nov 2020 13:39:57 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 4439F389803F
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/97832] AoSoA complex caxpy-like loops:
 AVX2+FMA -Ofast 7 times slower than -O3
Date: Wed, 18 Nov 2020 13:39:57 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 10.2.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: ASSIGNED
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: rguenth at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-97832-4-XwEo1EmBDp@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-97832-4@http.gcc.gnu.org/bugzilla/>
References: <bug-97832-4@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
X-BeenThere: gcc-bugs@gcc.gnu.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Gcc-bugs mailing list <gcc-bugs.gcc.gnu.org>
List-Unsubscribe: <https://gcc.gnu.org/mailman/options/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=unsubscribe>
List-Archive: <https://gcc.gnu.org/pipermail/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-request@gcc.gnu.org?subject=help>
List-Subscribe: <https://gcc.gnu.org/mailman/listinfo/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=subscribe>
X-List-Received-Date: Wed, 18 Nov 2020 13:39:57 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D97832

--- Comment #9 from Richard Biener <rguenth at gcc dot gnu.org> ---
There's then also a permute optimization left on the plate:

t.c:16:3: note:   node 0x3a19590 (max_nunits=3D4, refcnt=3D2)
t.c:16:3: note:         stmt 0 _153 =3D f11_im_76 * x1_im_142;
t.c:16:3: note:         stmt 1 _213 =3D f11_re_72 * x1_re_202;
t.c:16:3: note:         stmt 2 _275 =3D f11_re_72 * x1_re_264;
t.c:16:3: note:         stmt 3 _337 =3D f11_re_72 * x1_re_326;
t.c:16:3: note:         stmt 4 _155 =3D f11_im_76 * x1_re_140;
t.c:16:3: note:         stmt 5 _217 =3D f11_im_76 * x1_re_202;
t.c:16:3: note:         stmt 6 _279 =3D f11_im_76 * x1_re_264;
t.c:16:3: note:         stmt 7 _341 =3D f11_im_76 * x1_re_326;
t.c:16:3: note:         children 0x3a19600 0x3a19670
t.c:16:3: note:   node (external) 0x3a19600 (max_nunits=3D1, refcnt=3D1)
t.c:16:3: note:         { f11_im_76, f11_re_72, f11_re_72, f11_re_72,
f11_im_76, f11_im_76, f11_im_76, f11_im_76 }
t.c:16:3: note:   node 0x3a19670 (max_nunits=3D4, refcnt=3D1)
t.c:16:3: note:         stmt 0 x1_im_142 =3D *_141;
t.c:16:3: note:         stmt 1 x1_re_202 =3D *_201;
t.c:16:3: note:         stmt 2 x1_re_264 =3D *_263;
t.c:16:3: note:         stmt 3 x1_re_326 =3D *_325;
t.c:16:3: note:         stmt 4 x1_re_140 =3D *_139;
t.c:16:3: note:         stmt 5 x1_re_202 =3D *_201;
t.c:16:3: note:         stmt 6 x1_re_264 =3D *_263;
t.c:16:3: note:         stmt 7 x1_re_326 =3D *_325;
t.c:16:3: note:         load permutation { 4 1 2 3 0 1 2 3 }

which we currently do not handle (there's a FIXME as to permute externals,
currently we only handle splats as transparent for permutes).=