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 13AF23858C2C; Tue,  4 Jan 2022 21:44:08 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 13AF23858C2C
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/103905] [12 Regression] Miscompiled i386-expand.c with
 -march=bdver1 and -O3 since r12-1789-g836328b2c99f5b8d
Date: Tue, 04 Jan 2022 21:44:07 +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: 12.0
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: 12.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-103905-4-yKxAnLG9ST@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-103905-4@http.gcc.gnu.org/bugzilla/>
References: <bug-103905-4@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, 04 Jan 2022 21:44:08 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D103905

--- Comment #6 from Uro=C5=A1 Bizjak <ubizjak at gmail dot com> ---
@Jakub: It looks the problem is in expand_vec_perm_pshufb, where permutation
vector is recalculated for partial vectors:

  if (vmode =3D=3D V4QImode
      || vmode =3D=3D V8QImode)
    {
      rtx m128 =3D GEN_INT (-128);

      /* Remap elements from the second operand, as we have to
         account for inactive top elements from the first operand.  */
      if (!d->one_operand_p)
        {
          int sz =3D GET_MODE_SIZE (vmode);

          for (i =3D 0; i < nelt; ++i)
            {
              int ival =3D INTVAL (rperm[i]);
              if (ival >=3D sz)
                ival +=3D 16-sz;
              rperm[i] =3D GEN_INT (ival);
            }
        }

      /* V4QI/V8QI is emulated with V16QI instruction, fill inactive
         elements in the top positions with zeros.  */
      for (i =3D nelt; i < 16; ++i)
        rperm[i] =3D m128;

      vpmode =3D V16QImode;
    }

I must admit I only eyeballed the generated code, so perhaps there lies the
dragon.=