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 AF975385AC38; Thu, 11 Nov 2021 20:07:41 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org AF975385AC38
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/103193] gcc for x86_64: wrong code generation: ucomiss
 instead of comiss
Date: Thu, 11 Nov 2021 20:07:41 +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: 11.2.0
X-Bugzilla-Keywords: wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ubizjak at gmail dot com
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-103193-4-JyXzAOANpu@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-103193-4@http.gcc.gnu.org/bugzilla/>
References: <bug-103193-4@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: Thu, 11 Nov 2021 20:07:41 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D103193

--- Comment #2 from Uro=C5=A1 Bizjak <ubizjak at gmail dot com> ---
Here is compilable testcase:

_Bool a (void)
{
#pragma STDC FENV_ACCESS ON
  float f;
  _Bool b;
  f =3D __builtin_nan ("");
  b =3D f >=3D f; // ucomiss (wrong), comiss (correct)
  return b;
}

Tree optimizers (_.optimized dump) already give us:

  <bb 2> :
  f_1 =3D  Nan;
  b_2 =3D f_1 =3D=3D f_1;
  _3 =3D b_2;


However, f_1 =3D=3D f_1 expands with UNSPEC_NOTRAP RTX which gives us ucomi=
ss.=