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 F3F9F3857C5B; Tue, 18 Jan 2022 08:36:11 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org F3F9F3857C5B
From: "crazylht at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/103771] [12 Regression] Missed vectorization under
 -mavx512f -mavx512vl after r12-5489
Date: Tue, 18 Jan 2022 08:36:11 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 12.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: crazylht at gmail dot com
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: 12.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-103771-4-fzRRoiOU9W@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-103771-4@http.gcc.gnu.org/bugzilla/>
References: <bug-103771-4@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: Tue, 18 Jan 2022 08:36:12 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D103771

--- Comment #22 from Hongtao.liu <crazylht at gmail dot com> ---
(In reply to Hongtao.liu from comment #21)
> (In reply to Hongtao.liu from comment #20)
> > (In reply to Richard Biener from comment #19)
> > > Ah, so the issue is missing -mavx512bw which means we end up with a A=
VX2
> > > style
> > > mask for V32QImode.  With -mavx512bw the code vectorizes fine.
> >=20
> > Vectorization code is worse than before, now we need to pack vectorized=
 mask
> > which takes extra 3 instructions.
>=20
> Current ifcvt convert
>=20
> ---------dump of .ch_vect-------
>   if (x.1_14 > 255)
>     goto <bb 4>; [50.00%]
>   else
>     goto <bb 5>; [50.00%]
>=20
>   <bb 4> [local count: 477815112]:
>   _17 =3D -_5;
>   _18 =3D _17 >> 31;
>   iftmp.0_19 =3D (unsigned char) _18;
>   goto <bb 6>; [100.00%]
>=20
>   <bb 5> [local count: 477815112]:
>   iftmp.0_20 =3D (unsigned char) _5;
>=20
>   <bb 6> [local count: 955630225]:
>   # iftmp.0_21 =3D PHI <iftmp.0_19(4), iftmp.0_20(5)>
> -------dump end---------
>=20
>=20
> to=20
> ---- dump of .ifcvt---------
>   _41 =3D -x.1_14;
>   _17 =3D (int) _41;
>   _18 =3D _17 >> 31;
>   iftmp.0_19 =3D (unsigned char) _18; -- vec_pack_trunc
>   iftmp.0_20 =3D (unsigned char) _5; -- vec_pack_trunc
>   iftmp.0_21 =3D x.1_14 > 255 ? iftmp.0_19 : iftmp.0_20; -- vec_pack_trunc
>   *_6 =3D iftmp.0_21;
>   x_16 =3D x_24 + 1;
> -----dump end----------
>=20
>=20
> if ifcvt output things like
> ------------optimal .ifcvt------
>   _41 =3D -x.1_14;
>   _17 =3D (int) _41;
>   _18 =3D _17 >> 31;
>   iftmp.0_21 =3D x.1_14 > 255 ? _18 : _5;
>   iftmp.0_22 =3D (unsigned char) iftmp.0_21; --- vec_pack_trunc
>   *_6 =3D iftmp.0_22;
>   x_16 =3D x_24 + 1;
> ------------end------------
>=20
> we can save operations for packing mask(3 vec_pack_trunc vs 1
> vec_pack_trunc?).

Or maybe a gimple simplification for it?=