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 692743858C27; Thu, 27 Apr 2023 12:03:32 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 692743858C27
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1682597012;
	bh=rSmM36XVC1jDZ1CfM+8uKLba2qXjtigMeHyKEkfz98Y=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=JRg5AbvdOJ1+h/1vkWgrxffqFHD3GNMULptag6WxOgYT51hvbylB/6sCSsvQvRgSN
	 /p7cAXdgLtjD8gEmalW4wPUQzaaJvzISVIbkyVvgfNANXhj+oZUwl9ZRU72D9Ts0Cf
	 5kfD7uklvSBnElGMdCvn+CAiy/DsvL979MnZelzg=
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/107131] [11/12 Regression] wrong code with -Os
 -fno-ipa-vrp -fno-tree-bit-ccp
Date: Thu, 27 Apr 2023 12:03:32 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 13.0
X-Bugzilla-Keywords: wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P2
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 11.4
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-107131-4-YgemCHeKwx@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-107131-4@http.gcc.gnu.org/bugzilla/>
References: <bug-107131-4@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=3D107131

--- Comment #12 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #11)
> (In reply to Jakub Jelinek from comment #8)
> > I think the problem happens during combine.  Parts of the vector code
> > related to:
> >   V b =3D (V) { } >=3D o;
> >   V c =3D b <=3D (V)(b >=3D (V) { 0, 0, 0, 0, 0, 0x90DF0BE3990AC871ULL =
});
> > - in particular the scalar c[5] computation, i.e.
> > unsigned long long b =3D 0 >=3D o ? -1ULL : 0ULL;
> > unsigned long long c =3D b <=3D (b >=3D 0x90DF0BE3990AC871ULL ? -1ULL :=
 0ULL) ?
> > -1ULL : 0ULL;
> > evaluates to 0 rather than -1 that it should when o is 0.
> >=20
> > This boils down to:
> > __attribute__((noipa)) unsigned long long
> > foo (unsigned char o)
> > {
> >   unsigned long long t1 =3D -(long long) (o =3D=3D 0);
> >   unsigned long long t2 =3D -(long long) (t1 > 10439075533421201520ULL);
> >   unsigned long long t3 =3D -(long long) (t1 <=3D t2);
> >   return t3;
> > }
> >=20
> > int
> > main ()
> > {
> >   if (foo (0) !=3D -1ULL)
> >     __builtin_abort ();
> >   return 0;
> > }
> > at -O2 or -Os, which was miscompiled starting with r12-303-g86403f4e6e5=
f7216
> > and got fixed with r13-3530-g0e36a9c6915c713d30.
> > No idea why this is marked as 11 regression though...
> >=20
> > So, do we need to backport PR107172 change to 12 branch?
>=20
> Yes.  Testing that now.

Not.  Not picking cleanly.=