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 4C6813858407; Tue, 10 Jan 2023 23:36:28 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 4C6813858407
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1673393788;
	bh=VqTgqx+4emZWq8KLwMAWjWYd7AUNsCw0kln0qt/GARY=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=GiT1YwHNFVfGanqmcrDKIfZGaOaR+RNDocZymhWlNmWMK8LFQ7fE4hgOSFm4fmp3W
	 f/n6eOOPcXwRg57tcIq50HCs0VLbaQORC39elfDjVDhDiXnx8vs2L4qPfgeY2FP7s2
	 eCsJKCN5b1RmMS0VggPzQSQq4odcWUiJE4LNowQ4=
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/108365] [9/10/11/12/13 Regression] Wrong code with
 -O0
Date: Tue, 10 Jan 2023 23:36:27 +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: 13.0
X-Bugzilla-Keywords: diagnostic, wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jakub 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: 10.5
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-108365-4-r0CfrjIIcg@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108365-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108365-4@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=3D108365

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
With r9-1730 or later, I think the problem is that something decides to nar=
row
the division from long long to int.  In long long it is well defined if b is
non-zero
as -2147483648LL / -1LL is 2147483648LL.  But when we instead narrow it to
-2147483648 / -1 is UB which triggers division by zero exception.
We don't seem to narrow:
int
foo (int x, int y)
{
  return (long long) x / (long long) y;
}
so probably we do it only if the dividend is constant or something similar;=
 if
yes, then the fix would be stop narrowing if the dividend is the minimum of=
 the
narrower type.=