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 466A03833022; Tue, 27 Apr 2021 07:08:33 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 466A03833022
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/100267] gcc -O2 for avx512 instrincts generates
 extra warnings and less optimizations
Date: Tue, 27 Apr 2021 07:08:33 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: 10.2.0
X-Bugzilla-Keywords: diagnostic, missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth 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 keywords cf_gcctarget everconfirmed
 cf_reconfirmed_on
Message-ID: <bug-100267-4-wlQKvSSzUO@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-100267-4@http.gcc.gnu.org/bugzilla/>
References: <bug-100267-4@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, 27 Apr 2021 07:08:33 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D100267

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
           Keywords|                            |diagnostic,
                   |                            |missed-optimization
             Target|                            |x86_64-*-* i?86-*-*
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2021-04-27

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
With GCC 9 SRA decomposes pdata and substitutes uninitialized SSA names mar=
ked
with no-warning while later we no longer perform this optimization.  It's n=
ot
exactly clear what _mm256_mask_expand_epi64 does and why you need to feed it
an uninitialized first argument.

The middle-end unfortunately only sees

  _41 =3D MEM[(__m256i * {ref-all})&pdata];
  _42 =3D __builtin_ia32_expanddi256_mask (_29, _41, 15);

where it doesn't know that __builtin_ia32_expanddi256_mask doesn't actually
use the uninitialized value _41 (does it?)

You could use

  __m256i pdatau =3D pdatau;

  pdata[0] =3D _mm256_mask_expand_epi64(pdatau, m[0], nd[0]);
  pdata[1] =3D _mm256_mask_expand_epi64(pdatau, m[1], nd[1]);

to get an uninitialized __m256i and not warn about the missed initializatio=
n.
(but I suspect the data might be initialized in other uses)=