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 9925D3858291; Mon, 29 Jan 2024 22:01:12 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 9925D3858291
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1706565672;
	bh=jD7A3lyrt+5xeoUv9Hfu6Cf9oMDpDK+ARk/+2c/y0ak=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=rqO0CgXV79bJX1tqO9u2m1Xo53BCeQITJ3iQglf4yTTD8S3ES8KeHP84v44JWJgnp
	 35irPexGtRMKPcuUnNCgBmxSLRJcT3naloP59a9cQBmSegyx6DTteTD2viEPCECyDz
	 Fo295QJ1YJ6PCxUEttnyW2sg+7+3fK5h8jdZ7W3U=
From: "pinskia at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/113664] False positive warnings with
 -fno-strict-overflow (-Warray-bounds, -Wstringop-overflow)
Date: Mon, 29 Jan 2024 22:01:12 +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: 14.0
X-Bugzilla-Keywords: diagnostic
X-Bugzilla-Severity: normal
X-Bugzilla-Who: pinskia at gcc dot gnu.org
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-113664-4-qz1zKaz3tn@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-113664-4@http.gcc.gnu.org/bugzilla/>
References: <bug-113664-4@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=3D113664

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
-fno-strict-overflow turns on -fwrapv-pointer which allows pointers to wrap
which means if s was non-null, then `s+1` can be still a null pointer ...

And then we go and prop null into dot and s is equal to null at that point.=
 and
then we still generate the code for `*s++ =3D '.';` in
```
    if (s =3D=3D dot) {
      *s++ =3D '.';
    }
```

But it is `*NULL =3D '.';` due to that.

The warning is very sensitive to the ability to optimization away null poin=
ter
checks in this case. Really `fno-strict-overflow` is normally to workaround
some "undefinedness" in the code and the code should be improved to be fixe=
d.
Using -fwrapv instead also helps because now only signed integer overflows =
and
not also pointers ...=