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 A6E27385701D; Sun, 15 Aug 2021 10:51:16 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org A6E27385701D
From: "pinskia at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/82730] extra store/reload of an XMM for every byte
 extracted
Date: Sun, 15 Aug 2021 10:51:16 +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: 8.0
X-Bugzilla-Keywords: missed-optimization, ssemmx
X-Bugzilla-Severity: enhancement
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: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: bug_status bug_severity everconfirmed
 cf_reconfirmed_on
Message-ID: <bug-82730-4-j52QYxdxGk@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-82730-4@http.gcc.gnu.org/bugzilla/>
References: <bug-82730-4@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: Sun, 15 Aug 2021 10:51:16 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D82730

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
           Severity|normal                      |enhancement
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2021-08-15

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Note the gimple level looks good:

  _20 =3D BIT_FIELD_REF <in_18(D), 8, 120>;
  _1 =3D (int) _20;
  _21 =3D BIT_FIELD_REF <in_18(D), 8, 112>;
  _2 =3D (int) _21;
  _22 =3D BIT_FIELD_REF <in_18(D), 8, 104>;
  _3 =3D (int) _22;
  _23 =3D BIT_FIELD_REF <in_18(D), 8, 96>;
  _4 =3D (int) _23;
  _24 =3D BIT_FIELD_REF <in_18(D), 8, 88>;
  _5 =3D (int) _24;
  _25 =3D BIT_FIELD_REF <in_18(D), 8, 80>;
  _6 =3D (int) _25;
  _26 =3D BIT_FIELD_REF <in_18(D), 8, 72>;
  _7 =3D (int) _26;
  _27 =3D BIT_FIELD_REF <in_18(D), 8, 64>;
  _8 =3D (int) _27;
  _28 =3D BIT_FIELD_REF <in_18(D), 8, 56>;
  _9 =3D (int) _28;
  _29 =3D BIT_FIELD_REF <in_18(D), 8, 48>;
  _10 =3D (int) _29;
  _30 =3D BIT_FIELD_REF <in_18(D), 8, 40>;
  _11 =3D (int) _30;
  _31 =3D BIT_FIELD_REF <in_18(D), 8, 32>;
  _12 =3D (int) _31;
  _32 =3D BIT_FIELD_REF <in_18(D), 8, 24>;
  _13 =3D (int) _32;
  _33 =3D BIT_FIELD_REF <in_18(D), 8, 16>;
  _14 =3D (int) _33;
  _34 =3D BIT_FIELD_REF <in_18(D), 8, 8>;
  _15 =3D (int) _34;
  _35 =3D BIT_FIELD_REF <in_18(D), 8, 0>;
  _16 =3D (int) _35;
----- CUT ----

It is the way extractions are done for bytes is not good.
Note MSVC is the only one which does extractions in a register only and not=
 do
a store to the stack.=