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 163D23858D38; Mon, 30 Jan 2023 22:09:58 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 163D23858D38
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1675116598;
	bh=CHkJxAIffE/PMdsnlsyLmOJ6KlC25W5dOqLzV8mQYI0=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=BAl2RUcbwWfQ/3pZvGS4+6P6lbtFziLowR0YvQ87NwxGbUnRDwRFeDMnxnFQ/f1FS
	 7DuuktLysBDByZUTiIk0XUEsOTZnJ8gZhzfBCRclHSFb6Rff5NauD4fW6YuDIzISdx
	 rgXQeDq7VmMwoUTJEPk509OoyfYxhkG/0iOHO++g=
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/108608] [12/13 Regression] AArch64 vectorizer
 ICE in vect_transform_reduction and __builtin_fmax
Date: Mon, 30 Jan 2023 22:09:57 +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-Version: 13.0
X-Bugzilla-Keywords: ice-on-valid-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jakub at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P2
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 12.3
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-108608-4-jiLOVh1gMG@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108608-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108608-4@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=3D108608

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
The original code in kokkos had roughly:
  Real max_rel_err =3D -1;
  const Real rel_tol =3D std::numeric_limits<Real>::epsilon();


  for (int n =3D 0; n < N; n++) {
    for (int k =3D 0; k < N; k++) {
      for (int j =3D 0; j < N; j++) {
        for (int i =3D 1; i < N - 1; i++) {
          const Real analytic =3D 2.0 * (i + 1) * pow((j + 2) * (k + 3) * (=
n +
4), 2.0);
          const Real err =3D host_du(n, k, j, i - 1) - analytic;

          max_rel_err =3D fmax(fabs(err / analytic), max_rel_err);
        }
      }
    }
  }

  return max_rel_err < rel_tol;
where Real is double, and host_du is some hard to decipher C++ object.=