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 A906D3858432; Fri, 12 Apr 2024 07:12:53 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org A906D3858432
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1712905973;
	bh=+HIxPV6Q2YQvCItHy6UXOgAgUMlrPe0/f0KQZvqFFpU=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=iSa6kDzVLmyJeBKVmKmxbNjCxNWO1wEYc1YeIvJcDh/4uRjh1iysMh6BlvrI2Ih19
	 QXZIiRNi+lyxB4yJeP/0o5leNz70ixMV0TSmzT9iyPbDIYlUoc1j5BWbpwBTCiVFTq
	 bYX7tJpiefXqWLCmVxoRgfkyoCezjm6cjJyf3JC4=
From: "lin1.hu at intel dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/114700] Front-end optimization generates wrong code
 with -fsanitize=undefined
Date: Fri, 12 Apr 2024 07:12:52 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: unknown
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: lin1.hu at intel 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: 
Message-ID: <bug-114700-4-sidnzhgeBV@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114700-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114700-4@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=3D114700

--- Comment #12 from Hu Lin <lin1.hu at intel dot com> ---
(In reply to Hu Lin from comment #11)
> (In reply to Richard Biener from comment #9)
> > That that GCC doesn't promise that -ftrapv preserves all overflows and
> > traps, it merely guarantees that all overflows that actually happen tra=
p.=20
> > So GCC is fine to contract some expressions where the overall number of
> > overflows can only
> > decrease.
> >=20
> > That's not a bug with -ftrapv.
> >=20
> > It is considered a bug with -fsanitize=3Dundefined though.
>=20
> I think it doesn't mean that's not a bug with -ftrapv, it should preserve
> all overflow traps. Because it doesn't work, we use -fsanitize=3Dundefined
> instead of it.
>=20
> refer: Gcc's trapv is known not always to work correctly.
>=20
> The current behavior is correct for -fsanitize=3Dundefined, because the
> integer signed overflow is well-defined, so GCC can eliminate some
> variables. I just think GCC can optimize `z =3D c  - y  - c + a  + y - b`=
 to
> `z =3D a - b`. But it doesn't mean is a bug for -fsanitize=3Dundefined.

I was wrong about signed overflow, it's undefined behavior in c++20
(https://en.cppreference.com/w/cpp/language/ub). If I'm not mistaken about =
the
source=EF=BC=8C it's a bug.=