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 5C2343971835; Fri,  4 Dec 2020 09:48:28 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 5C2343971835
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/98137] New: Could use SLP to vectorize if
 split_constant_offset were smarter
Date: Fri, 04 Dec 2020 09:48:28 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 11.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: UNCONFIRMED
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: bug_id short_desc product version bug_status
 bug_severity priority component assigned_to reporter target_milestone
Message-ID: <bug-98137-4@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: Fri, 04 Dec 2020 09:48:28 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D98137

            Bug ID: 98137
           Summary: Could use SLP to vectorize if split_constant_offset
                    were smarter
           Product: gcc
           Version: 11.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: rguenth at gcc dot gnu.org
  Target Milestone: ---

void
gemm_m10_n9_k17_ldA20_ldB20_ldC10_beta0_alignedA1_alignedC1_pfsigonly(const
double* __restrict__ A, const double* __restrict__ B, double* __restrict__ =
C,
const double* A_prefetch, const double* B_prefetch, const double* C_prefetc=
h) {

  unsigned int l_m =3D 0;
  unsigned int l_n =3D 0;
  unsigned int l_k =3D 0;

  for ( l_n =3D 0; l_n < 9; l_n++ ) {
    for ( l_m =3D 0; l_m < 10; l_m++ ) { C[(l_n*10)+l_m] =3D 0.0; }

    for ( l_k =3D 0; l_k < 17; l_k++ ) {
      for ( l_m =3D 0; l_m < 10; l_m++ ) {
        C[(l_n*10)+l_m] +=3D A[(l_k*20)+l_m] * B[(l_n*20)+l_k];
      }
    }
  }
}

is nicely vectorized with BB SLP when you make l_{m,n,k} signed but when
unsigned as above then split_constant_offset gives up when seeing

C + ((unsigned long)(_286 + 1) * 8)

but we even have nice range-info:

  # RANGE [0, 80] NONZERO 126
  _286 =3D l_n_189 * 10;
  # RANGE [0, 80] NONZERO 126
  _288 =3D (long unsigned int) _286;
  # RANGE [0, 640] NONZERO 1008
  _289 =3D _288 * 8;
  # PT =3D null { D.2428 } (nonlocal, restrict)
  _290 =3D C_37(D) + _289;

^^ C + ((unsigned long)(_286) * 8)

  # RANGE [1, 81] NONZERO 127
  _296 =3D _286 + 1;
  # RANGE [1, 81] NONZERO 127
  _297 =3D (long unsigned int) _296;
  # RANGE [8, 648] NONZERO 1016
  _298 =3D _297 * 8;
  # PT =3D { D.2428 } (nonlocal, restrict)
  _299 =3D C_37(D) + _298;

^^ C + ((unsigned long)(_286 + 1) * 8

giving up means DR group analysis doesn't relate them and we do not consider
SLP vectorization.=