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 13E743857702; Wed, 25 Oct 2023 08:14:44 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 13E743857702
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1698221684;
	bh=8G0OcqzR18vELG3O9vRXcH1mrNfuZhNakj7/6yKfbsY=;
	h=From:To:Subject:Date:From;
	b=Pzy5B8D4xbmc8WsdghpEIfUtaitHNEUthZhnN3XVMVk51hHooK1ESy2E27htugynA
	 ddMGkTYlmZiOr3oFVPBqKQ6rbV6qyieNIGDOF58r53mb8/SM+ulmbiHpJpDggQa6EA
	 6qqBi/fWLGixFTc/sN3Vhj3wm/jD4lGEWrB/BZhY=
From: "crazylht at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/111972] New: [14 regression] missed vectorzation for
 bool a = j != 1; j = (long int)a;
Date: Wed, 25 Oct 2023 08:14:43 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: crazylht at gmail dot com
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-111972-4@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=3D111972

            Bug ID: 111972
           Summary: [14 regression] missed vectorzation for bool a =3D j !=
=3D
                    1; j =3D (long int)a;
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: crazylht at gmail dot com
  Target Milestone: ---

cat test.c

double
foo() {
  long n3 =3D 3450000, xtra =3D 7270;
  long i,ix;
  long j;
  double Check;

  /* Section 3, Conditional jumps */
  j =3D 0;
  {
    for (ix=3D0; ix<xtra; ix++)
      {
        for(i=3D0; i<n3; i++)
          {
            if(j=3D=3D1)       j =3D 2;
            else           j =3D 3;
            if(j>2)        j =3D 0;
            else           j =3D 1;
            if(j<1)        j =3D 1;
            else           j =3D 0;
          }
      }
  }
  Check =3D Check + (double)j;
  return Check;
}

The different between gcc 13 dump and gcc14 dump is
GCC13 we have

  <bb 3> [local count: 1063004411]:
  # i_16 =3D PHI <i_13(8), 0(5)>
  # j_18 =3D PHI <_7(8), j_21(5)>
  # ivtmp_15 =3D PHI <ivtmp_6(8), 3450000(5)>
  _7 =3D j_18 ^ 1;
  i_13 =3D i_16 + 1;
  ivtmp_6 =3D ivtmp_15 - 1;
  if (ivtmp_6 !=3D 0)
    goto <bb 8>; [99.00%]
  else
    goto <bb 4>; [1.00%]

GCC14 we have

  <bb 3> [local count: 1063004410]:
  # i_17 =3D PHI <i_13(8), 0(5)>
  # j_19 =3D PHI <_14(8), j_22(5)>
  # ivtmp_16 =3D PHI <ivtmp_15(8), 3450000(5)>
  _9 =3D j_19 !=3D 1;
  _14 =3D (long int) _9;
  i_13 =3D i_17 + 1;
  ivtmp_15 =3D ivtmp_16 - 1;
  if (ivtmp_15 !=3D 0)
    goto <bb 8>; [98.99%]
  else
    goto <bb 4>; [1.01%]

Vectorizer can handle=20

  _7 =3D j_18 ^ 1;=20

but not

  _9 =3D j_19 !=3D 1;
  _14 =3D (long int) _9;


../test.C:11:18: note:   vect_is_simple_use: operand j_19 !=3D 1, type of d=
ef:
internal
../test.C:11:18: note:   mark relevant 2, live 0: _9 =3D j_19 !=3D 1;
../test.C:11:18: note:   worklist: examine stmt: _9 =3D j_19 !=3D 1;
../test.C:11:18: note:   vect_is_simple_use: operand j_19 =3D PHI <_14(8),
j_22(5)>, type of def: unknown
../test.C:11:18: missed:   Unsupported pattern.
../test.C:15:6: missed:   not vectorized: unsupported use in stmt.
../test.C:11:18: missed:  unexpected pattern.


The difference comes from phiopt2.=