From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugs-return-384856-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 29128 invoked by alias); 29 Feb 2012 12:32:57 -0000
Received: (qmail 29115 invoked by uid 22791); 29 Feb 2012 12:32:56 -0000
X-SWARE-Spam-Status: No, hits=-2.8 required=5.0	tests=ALL_TRUSTED,AWL,BAYES_00,TW_TM
X-Spam-Check-By: sourceware.org
Received: from localhost (HELO gcc.gnu.org) (127.0.0.1)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 29 Feb 2012 12:32:43 +0000
From: "evstupac at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/52252] An opportunity for x86 gcc vectorizer (gain up to 3 times)
Date: Wed, 29 Feb 2012 12:34:00 -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-Keywords: missed-optimization
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: evstupac at gmail dot com
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Changed-Fields:
Message-ID: <bug-52252-4-WHD8cucYzc@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-52252-4@http.gcc.gnu.org/bugzilla/>
References: <bug-52252-4@http.gcc.gnu.org/bugzilla/>
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
X-SW-Source: 2012-02/txt/msg02790.txt.bz2

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D52252

--- Comment #2 from Stupachenko Evgeny <evstupac at gmail dot com> 2012-02-=
29 12:32:20 UTC ---
The difference of 2 dumps from

Arm: gcc -O3 -mfpu=3Dneon test.c -S -ftree-vectorizer-verbose=3D12
X86: gcc -O3 -m32 -msse3 test.c -S -ftree-vectorizer-verbose=3D12

Starts at:

For Arm (can use vec_load_lanes):

6: =3D=3D=3D vect_make_slp_decision =3D=3D=3D=20
6: =3D=3D=3D vect_detect_hybrid_slp =3D=3D=3D
6: =3D=3D=3D vect_analyze_loop_operations =3D=3D=3D
6: examining phi: in_35 =3D PHI <in_22(7), in_5(D)(4)>

=E2=80=A6=E2=80=A6

6: can use vec_load_lanes<CI><V16QI>=20
6: vect_model_load_cost: unaligned supported by hardware.=20
6: vect_model_load_cost: inside_cost =3D 2, outside_cost =3D 0 .

For x86 (no array mode for V16QI[3]):

6: =3D=3D=3D vect_make_slp_decision =3D=3D=3D=20
6: =3D=3D=3D vect_detect_hybrid_slp =3D=3D=3D=20
6: =3D=3D=3D vect_analyze_loop_operations =3D=3D=3D=20
6: examining phi: in_35 =3D PHI <in_22(7), in_5(D)(4)>=20

.=E2=80=A6=E2=80=A6

6: no array mode for V16QI[3]=20
6: the size of the group of strided accesses is not a power of 2=20
6: not vectorized: relevant stmt not supported: r_8 =3D *in_35;=20

As I mentioned before, there is an ability for x86 to handle this (Arm can
shuffle than loads, x86 can use pshufb).