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 5E8B6385BF81; Thu, 11 Jun 2020 08:11:36 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 5E8B6385BF81
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
 s=default; t=1591863096;
 bh=duTQQa9/jWq1JsOy0rjYFDPDp6VpUEhNB3iKiGTyfqg=;
 h=From:To:Subject:Date:In-Reply-To:References:From;
 b=YR8/UFCez0heFcNvAaIgM1SJXroU3REBVVZQsTYCe9VAcgPcv6mypL/HMkYIMLv2R
 lUNOrDxBe2P5DireGGepX3TdKQO7CMZivlmpfOJliFob1+4NvGxdzfZYthPJj/WF1K
 resfTdmmPrXSCwv36d9z0gPSVD3vs3TOOTJyce7I=
From: "crazylht at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/95488] Suboptimal multiplication codegen for v16qi
Date: Thu, 11 Jun 2020 08:11:36 +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: 11.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: crazylht 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-95488-4-HyCrNqzWnh@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-95488-4@http.gcc.gnu.org/bugzilla/>
References: <bug-95488-4@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: <http://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: <http://gcc.gnu.org/mailman/listinfo/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=subscribe>
X-List-Received-Date: Thu, 11 Jun 2020 08:11:36 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D95488

--- Comment #5 from Hongtao.liu <crazylht at gmail dot com> ---
Microbenchmark
----
cat test.c

#include <stdio.h>
#include <stdlib.h>
#include <x86intrin.h>

typedef char  v16qi  __attribute__ ((vector_size (16)));
extern v16qi interleave_mul (v16qi, v16qi);
extern v16qi extend_mul (v16qi, v16qi);

#define LOOP 30000000


int
main ()
{
  int i;
  unsigned long long start, end;
  unsigned long long diff;
  unsigned int aux;
  v16qi *p0;
  v16qi *p1;
  v16qi x, y;

  p0 =3D (v16qi *) malloc (LOOP *  sizeof (*p0));
  p1 =3D (v16qi *) malloc (LOOP *  sizeof (*p1));
  for (i =3D 0; i < LOOP; i++)
    for (int j =3D 0; j !=3D 16; j++)
    {
      p0[i][j] =3D 1 + i + j;
      p1[i][j] =3D 1 + i * i + j * j;
    }

#if 1
  start =3D __rdtscp (&aux);
  for (i =3D 0; i < LOOP; i+=3D16)
    y =3D interleave_mul (p0[i], p1[i]);
  end =3D __rdtscp (&aux);
  diff =3D end - start;

  printf ("interleave_mul : %lld\n", diff);

#endif

#if 1
  start =3D __rdtscp (&aux);
  for (i =3D 0; i < LOOP; i+=3D16)
    x =3D extend_mul (p0[i], p1[i]);
  end =3D __rdtscp (&aux);
  diff =3D end - start;

  printf ("extend_mul :    %lld\n", diff);
#endif

  free (p0);
  free (p1);

  return 0;
}
---
show a little bit improvement:

interleave_mul : 104180000
extend_mul :    103922083=