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 420113844028; Mon, 13 May 2024 13:05:57 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 420113844028
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1715605557;
	bh=cU227iR1ajyET/0t0bf87e/AYBvHkEnvBNixk4+oYS0=;
	h=From:To:Subject:Date:From;
	b=FTKwrpN+HnKq27PcgYr1+8mTiunlUljcGp1j/Bzjr+bdQcoeCZoq0pRMR3oEIdnOC
	 3ESaEJFT2QZiXIPS6dknDbodu+gXgSHqVlVQ9b6AW1DdOvSYBWLNclDRY8YcJ83RNE
	 w7c4CIUcsli1wUiyKJVqd1r16a3DQMaqEys5BxHM=
From: "colin.king at intel dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/115069] New: 8 bit integer vector performance
 regression, x86, between gcc-14 and gcc-13 using avx2 target clones on
 skylake platform
Date: Mon, 13 May 2024 13:05:56 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: colin.king at intel 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: bug_id short_desc product version bug_status
 bug_severity priority component assigned_to reporter target_milestone
 attachments.created
Message-ID: <bug-115069-4@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=3D115069

            Bug ID: 115069
           Summary: 8 bit integer vector performance regression, x86,
                    between gcc-14 and gcc-13 using avx2 target clones on
                    skylake platform
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: colin.king at intel dot com
  Target Milestone: ---

Created attachment 58188
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=3D58188&action=3Dedit
reproducer source code

I'm seeing a ~12-14% performance regression in gcc-14 compared to gcc-13, u=
sing
gcc on Ubuntu 24.04:

Versions:
gcc version 13.2.0 (Ubuntu 13.2.0-23ubuntu4)=20
gcc version 14.0.1 20240412 (experimental) [master r14-9935-g67e1433a94f]
(Ubuntu 14-20240412-0ubuntu1)=20

cking@skylake:~$ gcc-13 reproducer-vecmath.c -O2
cking@skylake:~$ ./a.out=20
13540.16 vec8 ops per sec, duration =3D 14.77 secs

cking@skylake:~$ gcc-14 reproducer-vecmath.c -O2
cking@skylake:~$ ./a.out=20
11720.25 vec8 ops per sec, duration =3D 17.06 secs

The original issue appeared when regression testing stress-ng vecmath stres=
sor
[1]. I've managed to extract the attached reproducer from the original code
(see attached).

Salient point to focus on:

1. The issue is also dependant on the TARGET_CLONES macro being defined as
__attribute__((target_clones("mmx,avx,avx2,default")))  - the avx2 target
clones seems to be an issue in reproducing this problem, remove it for gcc-=
14
and the performance regression is reduced.

Attached are the reproducer C source and disassembled object code.=20

References: [1]
https://github.com/ColinIanKing/stress-ng/blob/master/stress-vecmath.c=