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 9CC863860C3F; Thu, 27 May 2021 14:03:54 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 9CC863860C3F
From: "jl_gccbugs at conductive dot de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug ipa/100801] New: Aggressive loop optimizations cause incorrect
 warning
Date: Thu, 27 May 2021 14:03:54 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: ipa
X-Bugzilla-Version: 11.1.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jl_gccbugs at conductive dot de
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: bug_id short_desc product version bug_status
 bug_severity priority component assigned_to reporter cc target_milestone
Message-ID: <bug-100801-4@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, 27 May 2021 14:03:54 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D100801

            Bug ID: 100801
           Summary: Aggressive loop optimizations cause incorrect warning
           Product: gcc
           Version: 11.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: ipa
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jl_gccbugs at conductive dot de
                CC: marxin at gcc dot gnu.org
  Target Milestone: ---

The following warning is triggered

> $ gcc-11 -c constproploopopt.c -O2 -Wall -mavx -g
> constproploopopt.c: In function =E2=80=98test=E2=80=99:
> constproploopopt.c:22:18: warning: iteration 4611686018427387903 invokes =
undefined behavior [-Waggressive-loop-optimizations]
>    22 |     dest[i] =3D src[i];
>       |                  ^
> constproploopopt.c:21:12: note: within this loop
>    21 |   for (; i < count; ++i) {  // handle residual elements
>       |          ~~^~~~~~~

by this (minimal) code:

> #include <stdint.h>
> #include <stdio.h>
> #if defined(_MSC_VER)
> #include <intrin.h>
> #else
> #include <x86intrin.h>
> #endif
>=20
> void copy_32_unaligned(uint32_t* dest, const uint32_t* src, size_t count)=
 {
>   // invariant/nop
>   __m128i shufmask =3D  _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, =
5, 4, 3, 2, 1, 0);
>=20
>   size_t i;
>   for (i =3D 0; i + 4 <=3D count; i +=3D 4) {
>     __m128i input =3D _mm_loadu_si128((const __m128i*)(&src[i]));
>     __m128i output =3D input;
>     // no warning without the shuffle:
>     output =3D _mm_shuffle_epi8(input, shufmask);
>     _mm_storeu_si128((__m128i*)(&dest[i]), output);
>   }
>   for (; i < count; ++i) {  // handle residual elements
>     dest[i] =3D src[i];
>   }
> }
>=20
> void test(uint32_t* buf1, uint32_t* buf2) {
>     copy_32_unaligned(buf2, buf1,
>                       // multiples of 4 and greater or equal then 12 trig=
ger it:
>                       12);
> }

>>From objdump output I believe the generated code is correct though. The war=
ning
seems to be incorrect in this context, especially since the "residual" loop
should be skipped for count=3Dn*4 anyways.=