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 83A613858D28; Tue, 20 Jun 2023 11:55:59 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 83A613858D28
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1687262159;
	bh=ybI8QuQYSaOR0tVpTmR/RtInSd5RHnZx5PJqQQCzstU=;
	h=From:To:Subject:Date:From;
	b=B6hDayrVg89SRAkJS20ZIHm/5EvDl3Um597mrbPBNYXKOoARSrTQGhyi6P3qDr8EQ
	 AeTmd73jB9pWAE4t/D4xLOK9qAlvi6cXwJI1eo9jbePZ199GIwGRP0+GfsXJvIHz/g
	 HZTyspLpzaeCJFHzzGw0FBcZiC91/fBuWwf/QWFk=
From: "lorien.lopez at unizar dot es" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug libgomp/110319] New: Performance slowdown using a pointer to
 perform a reduction vs. using a normal variable
Date: Tue, 20 Jun 2023 11:55:59 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: libgomp
X-Bugzilla-Version: 11.3.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: lorien.lopez at unizar dot es
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-110319-4@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=3D110319

            Bug ID: 110319
           Summary: Performance slowdown using a pointer to perform a
                    reduction vs. using a normal variable
           Product: gcc
           Version: 11.3.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libgomp
          Assignee: unassigned at gcc dot gnu.org
          Reporter: lorien.lopez at unizar dot es
                CC: jakub at gcc dot gnu.org
  Target Milestone: ---

When performing an OpenMP sum reduction into a regular variable, GCC uses a
"lock cmpxchg" instruction. In contrast, when the reduction is performed in=
to a
pointer, it uses an OpenMP atomic region. The second version is several tim=
es
slower in an Intel Skylake CPU.=20

The original report can be found in Stack Overflow:
https://stackoverflow.com/questions/76480632/performance-slowdown-using-a-p=
ointer-to-perform-a-reduction-vs-using-a-normal-v=