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 3111E38582A1; Tue, 18 Jul 2023 10:11:09 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 3111E38582A1
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1689675069;
	bh=X9ELCHLh7+JCuEswV/NHHLyod44D4tjrsE7l/WdbeGQ=;
	h=From:To:Subject:Date:From;
	b=IIGUShxyg301xtW1+i+CzWnPSwEI7w/BaETJMXCPPzX3q6USabE0c6CuJtc9qw64W
	 QF0QR4vaxPMw40Sw2sLxgtvvZT3uwLWow5jzUlVpG1bCO6DtQGtCtjhBFxg7Mj0T2N
	 x6jBwDUdn7PpTomQLXUdOYLn+PpfjPyLPqH8aUns=
From: "mrks2023 at proton dot me" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/110711] New: possible missed optimization for
 std::max with -march=znver2
Date: Tue, 18 Jul 2023 10:11:08 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: 13.1.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: mrks2023 at proton dot me
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 target_milestone
Message-ID: <bug-110711-4@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
List-Id: <gcc-bugs.sourceware.org>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D110711

            Bug ID: 110711
           Summary: possible missed optimization for std::max with
                    -march=3Dznver2
           Product: gcc
           Version: 13.1.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: mrks2023 at proton dot me
  Target Milestone: ---

I think I found a missed optimization involving std::max() for -march=3Dznv=
er2
(sorry if it was already reported, but I didn't find anything related in the
bug tracker).

I have two functions that compute the maximum element of an array:

- function k_std_max uses std::max() and is never vectorized
- function k_max uses conditional assignment and is vectorized, when the
optimization flags allow for it

The code (also https://godbolt.org/z/hW49nbqMY):

#include <cassert>
#include <algorithm>

double k_std_max(size_t n_els, double * a)
{
    assert(n_els > 0);
    double m =3D a[0];
    #ifdef _OPENMP
        #pragma omp simd reduction(max:m)
    #endif
    for (size_t i =3D 1; i < n_els; ++i) {
        m =3D std::max(m, a[i]);
    }

    return m;
}

double k_max(size_t n_els, double * a)
{
    assert(n_els > 0);
    double m =3D a[0];
    #ifdef _OPENMP
        #pragma omp simd reduction(max:m)
    #endif
    for (size_t i =3D 1; i < n_els; ++i) {
        m =3D m < a[i] ? a[i] : m;
    }

    return m;
}

Compiling with "-O3 -fopenmp -march=3Dznver2 -Wall -Wextra -DNDEBUG" vector=
izes
k_max:

.L19:
        vmovupd ymm3, YMMWORD PTR [rax+8]
        add     rax, 32
        vmaxpd  ymm1, ymm3, ymm1
        cmp     rax, rdx
        jne     .L19

but for k_std_max still scalar instructions are used:

.L3:
        vmovsd  xmm0, QWORD PTR [rax]
        add     rax, 8
        vmaxsd  xmm0, xmm0, xmm1
        cmp     rdx, rax
        jne     .L5

Note that I had to use -fopenmp as using only -fopenmp-simd did not vectori=
ze
k_max.

Even when I use "-Ofast" or "-Ofast -fopenmp" instead of "-O3" k_std_max is=
 not
vectorized:

.L3:
        vmaxsd  xmm0, xmm0, QWORD PTR [rax]
        add     rax, 8
        cmp     rdx, rax
        jne     .L3=