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 035B5385840C; Mon,  7 Aug 2023 09:05:23 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 035B5385840C
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1691399124;
	bh=QTet5fjyHWq9r+Tr7hdgOyO8Bg//jXPqwsC0yn/tQXE=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=dFI2QjbveEEnIDZYlFOPkYmmU279mDUkwE9fvqpvGQMTQMFRJ9W8arA6i6mtK2P1+
	 wA8FPi1ek4AINwbvUFeCXSLUh/Z0iA0FunnMwY8TfEP5jZ76d+p/0uzQGqMJR8xrBn
	 e3ngDTdzStT5SRoktttNtDik69hM113dPfiXdU98=
From: "pinskia at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/110361] [13/14 Regression] Missed Dead Code
 Elimination when using __builtin_unreachable since r13-2020-g16b013c9d9b
Date: Mon, 07 Aug 2023 09:05:21 +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: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: pinskia 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: 13.3
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: cf_reconfirmed_on everconfirmed bug_status
Message-ID: <bug-110361-4-yW0RJzOnf0@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-110361-4@http.gcc.gnu.org/bugzilla/>
References: <bug-110361-4@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=3D110361

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2023-08-07
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Confirmed.

What is interesting is we have:
```
  <bb 7> [local count: 525923045]:
  g =3D &p;
  f.0_18 =3D f;
  if (f.0_18 =3D=3D 0)
    goto <bb 8>; [50.00%]
  else
    goto <bb 13>; [50.00%]

  <bb 9> [local count: 225485782]:
  g =3D &p;
  f.0_36 =3D f;
  if (f.0_36 =3D=3D 0)
    goto <bb 8>; [50.00%]
  else
    goto <bb 13>; [50.00%]

```
Those bb's are exactly the same.

Their pred bbs are:
```
  <bb 3> [local count: 751619280]:
  if (e.4_19 =3D=3D &a)
    goto <bb 9>; [30.00%]
  else
    goto <bb 4>; [70.00%]

  <bb 4> [local count: 526133498]:
  if (e.4_19 =3D=3D &d)
    goto <bb 7>; [99.96%]
  else
    goto <bb 5>; [0.04%]
```

Their succ branch (that matters) is:
```
  <bb 13> [local count: 375914866]:
  c =3D &m;
  m =3D{v} {CLOBBER(eol)};
  _22 =3D e.4_19 =3D=3D &d;
  _1 =3D e.4_19 =3D=3D &a;
  _23 =3D _1 | _22;
  if (_23 !=3D 0)
    goto <bb 12>; [25.16%]
  else
    goto <bb 10>; [74.84%]
```

So coming into bb13, we know that `(e.4_19 =3D=3D &d) | (e.4_19 =3D=3D &a)`=
 is true.
Removing the branch to bb 10 which was the call to foo:
  <bb 10> [local count: 281323581]:
  foo ();=