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 02FDF3858D20; Wed, 15 Nov 2023 14:42:45 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 02FDF3858D20
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1700059366;
	bh=/w7pK5Z74lBeRG5l6M27JgBDzBm8JPCp1y7VtrsYSb4=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=tF7cyUh2by9WKfbceSTaDcqNkAjqc51IRVjmDdFEhFYXZXyXi54QbMrQYlCN6u9Yi
	 bGvABQuCXZcO8reWahysbHyF2jbzXYNkkq1uS46aTFI9qiZB76L4/O6T7P4/JPcfpJ
	 x0QQacrbMfy5DWcrOtRa9cPQq9FRVAB0xa1dUm6I=
From: "rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/109088] GCC does not always vectorize
 conditional reduction
Date: Wed, 15 Nov 2023 14:42:45 +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: missed-optimization
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: rguenther at suse dot de
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: 
Message-ID: <bug-109088-4-IORyqQQql9@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-109088-4@http.gcc.gnu.org/bugzilla/>
References: <bug-109088-4@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=3D109088

--- Comment #20 from rguenther at suse dot de <rguenther at suse dot de> ---
On Wed, 15 Nov 2023, juzhe.zhong at rivai dot ai wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D109088
>=20
> --- Comment #19 from JuzheZhong <juzhe.zhong at rivai dot ai> ---
> I have added:
>=20
> +      if (!ANY_INTEGRAL_TYPE_P (TREE_TYPE (PHI_RESULT (phi)))

For TYPE_OVERFLOW_UNDEFINED you have to convert the ops to unsigned
to avoid spurious undefined overflow.

> +         && !(FLOAT_TYPE_P (TREE_TYPE (PHI_RESULT (phi)))
> +              && !HONOR_SIGNED_ZEROS (TREE_TYPE (PHI_RESULT (phi)))
> +              && !HONOR_SIGN_DEPENDENT_ROUNDING (TREE_TYPE (PHI_RESULT (=
phi)))
> +              && !HONOR_NANS (TREE_TYPE (PHI_RESULT (phi)))))

You should check flag_associative_math which covers one half
and !flag_trapping_math which covers spurious FP exceptions.

> +       return false;
>=20
> for floating-point. I failed to see which situation will cause FP excepti=
ons ?

Ops with NaN cause INVALID, but there's also INEXACT which can be
set differently after re-association.=