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 A8E733857C56; Wed, 13 Mar 2024 10:20:00 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org A8E733857C56
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1710325200;
	bh=lHz99dKxgokpIYjs+hqn0uYUaUkZYm049eGeOHpFUdI=;
	h=From:To:Subject:Date:From;
	b=YfH/cxcvGQYxDRte+g/GzrlxCXJ7eWbpHMF3A1fKrSB39bUIv+ARGUYiOZyNzMH5X
	 5DtshPY9jX8HoW/+MaUhuyDY3/6f7W5WuqXH/eU0dLMqDg6MLfTIWEBXbi4kpXDMSf
	 IeyBaEbrPUbH9K9fNrJVJfEsifDTooqfVkIMlXGU=
From: "hliu at amperecomputing dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/114322] New: [14 Regression] SCEV analysis
 failed for bases like A[(i+x)*stride] since r14-9193-ga0b1798042d033
Date: Wed, 13 Mar 2024 10:19:59 +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: 14.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: hliu at amperecomputing 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-114322-4@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=3D114322

            Bug ID: 114322
           Summary: [14 Regression] SCEV analysis failed for bases like
                    A[(i+x)*stride] since r14-9193-ga0b1798042d033
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: hliu at amperecomputing dot com
  Target Milestone: ---

Compile the following case with: gcc simp.c -Ofast -mcpu=3Dneoverse-n1 -S \
         -fdump-tree-ifcvt -fdump-tree-vect-details-scev

int
foo (short *A, int x, int stride)
{
  int sum =3D 0;

  if (stride > 1)
    {
      #pragma GCC unroll 1
      for (int i =3D 0; i < 1024; ++i)
        sum +=3D A[(i + x) * stride];
    }

  return sum;
}

The gimple in the loop is:

  <bb 3>:
  # sum_19 =3D PHI <sum_15(6), 0(5)>
  # i_20 =3D PHI <i_16(6), 0(5)>
  # ivtmp_37 =3D PHI <ivtmp_36(6), 1024(5)>
  _1 =3D x_12(D) + i_20;
  _2 =3D _1 * stride_11(D);
  _3 =3D (long unsigned int) _2;
  _4 =3D _3 * 2;
  _5 =3D A_13(D) + _4;
  _6 =3D *_5;
  _7 =3D (int) _6;
  sum_15 =3D _7 + sum_19;


Before the commit (i.e., from pr114074 bug fix), it can be vectorized:

Creating dr for *_5
analyze_innermost: (analyze_scalar_evolution=20
  (loop_nb =3D 1)
  (scalar =3D _5)
(get_scalar_evolution=20
  (scalar =3D _5)
  (scalar_evolution =3D {A_13(D) + (long unsigned int) (stride_11(D) * x_12=
(D)) *
2, +, (long unsigned int) stride_11(D) * 2}_1))
)
success.
(analyze_scalar_evolution=20
  (loop_nb =3D 1)
  (scalar =3D _5)
(get_scalar_evolution=20
  (scalar =3D _5)
  (scalar_evolution =3D {A_13(D) + (long unsigned int) (stride_11(D) * x_12=
(D)) *
2, +, (long unsigned int) stride_11(D) * 2}_1))
)
(instantiate_scev=20
  (instantiate_below =3D 5 -> 3)
  (evolution_loop =3D 1)
  (chrec =3D {A_13(D) + (long unsigned int) (stride_11(D) * x_12(D)) * 2, +,
(long unsigned int) stride_11(D) * 2}_1)
  (res =3D {A_13(D) + (long unsigned int) (stride_11(D) * x_12(D)) * 2, +, =
(long
unsigned int) stride_11(D) * 2}_1))
        base_address: A_13(D) + (sizetype) (stride_11(D) * x_12(D)) * 2
        offset from base address: 0
        constant offset from base address: 0
        step: (ssizetype) ((long unsigned int) stride_11(D) * 2)
        base alignment: 2
        base misalignment: 0
        offset alignment: 128
        step alignment: 2
        base_object: *A_13(D) + (sizetype) (stride_11(D) * x_12(D)) * 2
        Access function 0: {0B, +, (long unsigned int) stride_11(D) * 2}_1


After the commit, loop vectorized failed due to SCEV failure with *_5:

Creating dr for *_5
analyze_innermost: (analyze_scalar_evolution=20
  (loop_nb =3D 1)
  (scalar =3D _5)
(get_scalar_evolution=20
  (scalar =3D _5)
  (scalar_evolution =3D _5))
)
(analyze_scalar_evolution=20
  (loop_nb =3D 1)
  (scalar =3D _5)
(get_scalar_evolution=20
  (scalar =3D _5)
  (scalar_evolution =3D _5))
)
simp.c:11:10: missed:  failed: evolution of base is not affine.
..
  (res =3D scev_not_known))


To my understanding, '(i + x) * stride' is signed integer calculation, in w=
hich
overflow is undefined behavior and the case should be vectorized.=