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 1D90C3858402; Wed, 13 Sep 2023 09:46:19 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 1D90C3858402
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1694598379;
	bh=DtFeOJZwE21rKOYr1t6aFp6Jl8EIT+E/XCh6kEJPiO4=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=ys/iJOExLGlKVzle+yetue2bUjD66yyvt0J1wVJHM/mJlu2W/7Z4wJ4gT3NLLYcFl
	 bJr7bCeix5S5GgqrTP1JQWRegQDWNDctomokxLatVmZzl3If+gZR+ig1M7Yu/8ygVc
	 bEX95p+urCDxcOfpDbQK1hRfdnhxzNHEx/QhdkKU=
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c/111401] Middle-end: Missed optimization of
 MASK_LEN_FOLD_LEFT_PLUS
Date: Wed, 13 Sep 2023 09:46:18 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: NEW
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: everconfirmed bug_status cf_reconfirmed_on
Message-ID: <bug-111401-4-yp53vUBZ5U@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-111401-4@http.gcc.gnu.org/bugzilla/>
References: <bug-111401-4@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=3D111401

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2023-09-13

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
The vectorizer sees if-converted code like

  <bb 3> [local count: 955630224]:
  # init_20 =3D PHI <_36(8), init_12(D)(18)>
  # i_22 =3D PHI <i_18(8), 0(18)>
  _1 =3D (long unsigned int) i_22;
  _2 =3D _1 * 4;
  _3 =3D cond_15(D) + _2;
  _4 =3D *_3;
  _23 =3D _4 !=3D 0;
  _6 =3D _1 * 8;
  _38 =3D _37 + _6;
  _7 =3D (double *) _38;
  _8 =3D .MASK_LOAD (_7, 64B, _23);
  _ifc__35 =3D _23 ? _8 : 0.0;
  _36 =3D init_20 + _ifc__35;
  i_18 =3D i_22 + 1;
  if (n_13(D) > i_18)

so what it produces matches up here.  There's the possibility to
modify the if-conversion handling to use a COND_ADD instead of
the COND_EXPR plus ADD, I think that would be the best thing here.
See tree-if-conv.cc:is_cond_scalar_reduction/convert_scalar_cond_reduction

I think this is also wrong code when signed zeros are involved.=