From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757273Ab3A1RQw (ORCPT ); Mon, 28 Jan 2013 12:16:52 -0500 Received: from lennier.cc.vt.edu ([198.82.162.213]:46990 "EHLO lennier.cc.vt.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755879Ab3A1RQu (ORCPT ); Mon, 28 Jan 2013 12:16:50 -0500 X-Mailer: exmh version 2.8.0 04/21/2012 with nmh-1.4-dev To: ling.ma.program@gmail.com Cc: mingo@redhat.com, tglx@linutronix.de, hpa@zytor.com, linux-kernel@vger.kernel.org, Ma Ling Subject: Re: [PATCH] [x86]: Compiler Option Os is better on latest x86 In-Reply-To: Your message of "Fri, 25 Jan 2013 09:11:01 -0500." <1359123061-6139-1-git-send-email-ling.ma@alipay.com> From: Valdis.Kletnieks@vt.edu References: <1359123061-6139-1-git-send-email-ling.ma@alipay.com> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="==_Exmh_1359393357_2147P"; micalg=pgp-sha1; protocol="application/pgp-signature" Content-Transfer-Encoding: 7bit Date: Mon, 28 Jan 2013 12:15:57 -0500 Message-ID: <26437.1359393357@turing-police.cc.vt.edu> X-Mirapoint-Received-SPF: 198.82.161.152 auth3.smtp.vt.edu Valdis.Kletnieks@vt.edu 2 pass X-Junkmail-Status: score=10/50, host=steiner.cc.vt.edu X-Junkmail-Signature-Raw: score=unknown, refid=str=0001.0A020207.5106B251.01A4,ss=1,re=0.000,fgs=0, ip=0.0.0.0, so=2011-07-25 19:15:43, dmn=2011-05-27 18:58:46, mode=single engine X-Junkmail-IWF: false Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --==_Exmh_1359393357_2147P Content-Type: text/plain; charset=us-ascii On Fri, 25 Jan 2013 09:11:01 -0500, ling.ma.program@gmail.com said: > Based on above reasons, we compiled linux kernel 3.6.9 with O2 and Os > respectively. The results show Os improve performance netperf 4.8%, > 2.7% for volano as below Am I allowed to NAK this? What the numbers given so far *actually* show is 4.8% more instructions executed, *not* 4.8% better performance. I'm having a *very* hard time convincing myself that what we're seeing isn't simply the expected behavior of loops *not* being unrolled and similar non-optimizations done by -Os, so more instructions get executed to do the same amount of work. Rather than "run for 10 seconds and count instructions", can we "run for 50,000 syscalls and count clock time" or similar that shows an *actual* improvement? --==_Exmh_1359393357_2147P Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.13 (GNU/Linux) Comment: Exmh version 2.5 07/13/2001 iQIVAwUBUQayTQdmEQWDXROgAQLcag//fsIee+NSfNUG76PoLuLDk4GG7mMRUaIC oC/snm7QvQVZYy83dHIZr0bJPVa3H8tHCavL63pzfLwNXBC3NzNnWjHD6x/8s0ZN FWcRun4Nn2P9QlzgUcZhxxZAnE++vMz//FNMIuuatPzNJ83uQ/Gl841lNRNYlQjb ZxNhOmThU4Kfttj0ci3OXaLA0Yl+55uY/nZXsxlCq98g4RBo6rrkIFMrcr310/D6 cR7MW9W2zcqWZ3jIdhL/nmY6yaRBl/3czlw50rK4Rxhh4o9p+dE/FbbNVPV+BK0x Z163S9cxLFstQs+6+nhzoEh50uNcnRhZpnEJ73L68xELV2ndFjE3Iglp+FKjsIn4 shvRqQVFJZKPsfGY/5mRbF8v4aqGSU/QQqjezUySxKZg4YvoePqzv9iDxdLKwJFC 36meM1pju7zSDe1dneEzNdYiSTy9BSRk6ezxXq+7GI4DuZez95dQExyyHf1wu3Bz VmOEPHLbdrfE0e/r9FDK/cil8cQKRzL/4Mtkgh0ddeMWnGYEIiyB4vpmnB1G57C7 x2/xepCBM1o1x35icldAGEgpPLVLhN1U38JNpYYhtkypRLlvECmHF4gIIvSdOAEr TojezVq5AOjAlHQdTd/VAcWFWLAFYycW8o7e0FBtkA78N+fCfriqBlYBIGvUu5df CAHjoamNw7Q= =g52A -----END PGP SIGNATURE----- --==_Exmh_1359393357_2147P--