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 4043A3858D32; Wed, 28 Jun 2023 06:53:26 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 4043A3858D32
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1687935206;
	bh=mF6PPw1wqqtTl/4w0tQWqJPH9rNSsWBtl4tbe8Nc/34=;
	h=From:To:Subject:Date:From;
	b=gVfr3eP9MDUloTDgpssFGFk44Aec/IPsDzY+53YfbwcF0oaWSQjoNgXS1peJOSI3j
	 Lgr1HllRaqjpPOtxV6r1Lb1o1AYXj+NKuHdNU3Kon3iqx6uJWhyOcTlJWiQkzP/xbL
	 Sx6/Pv2jYxfPnayMHWh21lpq1/BQaXC7Bh44sVac=
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug testsuite/110445] New: [14 Regression] FAIL:
 gcc.dg/vect/slp-46.c with AVX2
Date: Wed, 28 Jun 2023 06:53:25 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: testsuite
X-Bugzilla-Version: 14.0
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-110445-4@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=3D110445

            Bug ID: 110445
           Summary: [14 Regression] FAIL: gcc.dg/vect/slp-46.c with AVX2
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: testsuite
          Assignee: unassigned at gcc dot gnu.org
          Reporter: rguenth at gcc dot gnu.org
  Target Milestone: ---

With AVX2 we fail to SLP

double x[1024], y[1024];

void __attribute__((noipa)) foo()
{
  for (int i =3D 0; i < 512; ++i)
    {
      x[2*i] =3D y[i];
      x[2*i+1] =3D y[i];
    }
}

because we hit the following:

/space/rguenther/src/gcc11queue/gcc/testsuite/gcc.dg/vect/slp-46.c:10:21: n=
ote:
  =3D=3D> examining statement: _2 =3D y[i_12];
/space/rguenther/src/gcc11queue/gcc/testsuite/gcc.dg/vect/slp-46.c:10:21:
missed:   peeling for gaps insufficient for access
/space/rguenther/src/gcc11queue/gcc/testsuite/gcc.dg/vect/slp-46.c:12:17:
missed:   not vectorized: relevant stmt not supported: _2 =3D y[i_12];
/space/rguenther/src/gcc11queue/gcc/testsuite/gcc.dg/vect/slp-46.c:10:21: n=
ote:
  removing SLP instance operations starting from: x[_1] =3D _2;
/space/rguenther/src/gcc11queue/gcc/testsuite/gcc.dg/vect/slp-46.c:10:21:
missed:  unsupported SLP instances
/space/rguenther/src/gcc11queue/gcc/testsuite/gcc.dg/vect/slp-46.c:10:21: n=
ote:
 re-trying with SLP disabled

the issue is that in the last vector iteration with VF=3D2 we are accessing
{ i, i+1, i+2, i+3 }, if we're peeling at least a single scalar iteration
we still access possibly one too much elements.

The simplest solution would be to access { i, i+1 } only which I think
we already can do.  The other solution is to peel N scalar
iterations or apply masking to not access elements in the gap if the ISA
supports that.=