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 99CEA3858D28; Tue, 23 Nov 2021 10:41:38 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 99CEA3858D28
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/103376] [12 Regression] wrong code at -Os and
 above on x86_64-linux-gnu since r12-5453-ga944b5dec3adb28e
Date: Tue, 23 Nov 2021 10:41:38 +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: unknown
X-Bugzilla-Keywords: 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: P1
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 12.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: cc
Message-ID: <bug-103376-4-Zof3Bwsxk7@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-103376-4@http.gcc.gnu.org/bugzilla/>
References: <bug-103376-4@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
X-BeenThere: gcc-bugs@gcc.gnu.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Gcc-bugs mailing list <gcc-bugs.gcc.gnu.org>
List-Unsubscribe: <https://gcc.gnu.org/mailman/options/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=unsubscribe>
List-Archive: <https://gcc.gnu.org/pipermail/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-request@gcc.gnu.org?subject=help>
List-Subscribe: <https://gcc.gnu.org/mailman/listinfo/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 23 Nov 2021 10:41:38 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D103376

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
So, we have:
  e_11 =3D a;
...
  a.3_5 =3D a;
  _17 =3D a.3_5 ^ e_11;
and no stores in between those 2 reads, initially there has been a bb that
could fall through and that is the reason why the two loads are still there,
but it has been changed a few passes back to end with __builtin_trap.
bswap pass changes this to:
  e_11 =3D a;
...
+  load_dst_9 =3D MEM[(long long int *)&a];
+  _17 =3D (long long int) load_dst_9;
   a.3_5 =3D a;
-  _17 =3D a.3_5 ^ e_11;
For | instead of ^ that is a correct optimization a | a is still a, but for=
 the
newly added operations a ^ a is 0 rather than a and a + a is usually differ=
ent
than a too.
So, I guess for ^ and + we need to perform extra checking.
perform_symbolic_merge is doing:
  for (i =3D 0, mask =3D MARKER_MASK; i < size; i++, mask <<=3D BITS_PER_MA=
RKER)
    {
      uint64_t masked1, masked2;

      masked1 =3D n1->n & mask;
      masked2 =3D n2->n & mask;
      if (masked1 && masked2 && masked1 !=3D masked2)
        return NULL;
    }
which is correct solely for bitwise or.  For other operations it would need=
 to
do instead if (masked1 && masked2) return NULL; instead.=