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 44C3E3858D37; Wed, 28 Jun 2023 09:22:41 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 44C3E3858D37
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1687944161;
	bh=x3A0DS92NhHR/ZYh0TOwqcaW/K+9iN9dCWlc6Uwvsgo=;
	h=From:To:Subject:Date:From;
	b=pFs2zaJlHqzxh/qQDHBBurC1J+Vt5gvEbDBVYHmoF10/9/Gl3bY7Acy5Gf4gxnncK
	 5Jp3WE94NCwftV13ivK1S10/JFkDHke+Hz6fF41NgW0Kvz9/UxjrFTwstqkwYd0Vvm
	 yPaw1UIEyr/hwPhqXXsNN6M0vUkCjGu9ZxWrMlXQ=
From: "hliu at amperecomputing dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/110449] New: Vect: use a small step to
 calculate the loop induction if the loop is unrolled during loop
 vectorization
Date: Wed, 28 Jun 2023 09:22:40 +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-110449-4@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=3D110449

            Bug ID: 110449
           Summary: Vect: use a small step to calculate the loop induction
                    if the loop is unrolled during loop vectorization
           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: ---

This is inspired by clang. Compile the follwing case with "-mcpu=3Dneoverse=
-n2
-O3":

void foo(int *arr, int val, int step) {
  for (int i =3D 0; i < 1024; i++) {
    arr[i] =3D val;
    val +=3D step;
  }
}

It will be unrolled by 2 during vectorization. GCC generates code:
        fmov    s29, w2                 # step
        shl     v27.2s, v29.2s, 3       # 8*step
        shl     v28.2s, v29.2s, 2       # 4*step
        ...
.L2:
        mov     v30.16b, v31.16b
        add     v31.4s, v31.4s, v27.4s  # +=3D 8*step
        add     v29.4s, v30.4s, v28.4s  # +=3D 4*step
        stp     q30, q29, [x0]
        add     x0, x0, 32
        cmp     x1, x0
        bne     .L2

The v27 (i.e. "8*step") is actually not necessary. We can use v29 + v28 (i.=
e.
"+ 4*step") and generate simpler code:
        fmov    s29, w2                 # step
        shl     v28.2s, v29.2s, 2       # 4*step
        ...
.L2:
        add     v29.4s, v30.4s, v28.4s  # +=3D 4*step
        stp     q30, q29, [x0]
        add     x0, x0, 32
        add     v30.4s, v29.4s, v28.4s  # +=3D 4*step
        cmp     x1, x0
        bne     .L2

This has two benefits:
(1) Save 1 vector register and one "mov" instructon
(2) For floating point, the result value of small step should be closer to =
the
original scalar result value than large step. I.e. "A + 4*step + ... + 4*st=
ep"
should be closer to "A + step + ... + step" than "A + 8*step + ... 8*step".

Do you think if this is reasonable?=20

I have a simple patch to enhance the tree-vect-loop.cc
"vectorizable_induction()" to achieve this. Will send out the patch for code
review later.=