Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757512AbcCCTG2 (ORCPT ); Thu, 3 Mar 2016 14:06:28 -0500 Received: from torg.zytor.com ([198.137.202.12]:50974 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1757012AbcCCTG1 (ORCPT ); Thu, 3 Mar 2016 14:06:27 -0500 User-Agent: K-9 Mail for Android In-Reply-To: <20160303203414-mutt-send-email-mst@redhat.com> References: <20160303152739.GA16303@gmail.com> <20160303153453.GR6356@twins.programming.kicks-ass.net> <20160303203414-mutt-send-email-mst@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Subject: Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD? From: "H. Peter Anvin" Date: Thu, 03 Mar 2016 11:05:43 -0800 To: "Michael S. Tsirkin" , Peter Zijlstra CC: Ingo Molnar , Dexuan Cui , "linux-x86_64@vger.kernel.org" , Thomas Gleixner , Ingo Molnar , David Howells , "Paul E. McKenney" , "linux-kernel@vger.kernel.org" Message-ID: <66C0F0F8-5D2C-47DB-8C7A-EF8A15F263DB@zytor.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2764 Lines: 74 On March 3, 2016 10:35:50 AM PST, "Michael S. Tsirkin" wrote: >On Thu, Mar 03, 2016 at 04:34:53PM +0100, Peter Zijlstra wrote: >> On Thu, Mar 03, 2016 at 04:27:39PM +0100, Ingo Molnar wrote: >> > >> > * Dexuan Cui wrote: >> > >> > > Hi, >> > > My understanding about arch/x86/include/asm/barrier.h is: >obviously Linux >> > > more likes {L,S,M}FENCE -- Locked ADD is only used in x86_32 >platforms that >> > > don't support XMM2. >> > > >> > > However, it looks people say Locked Add is much faster than the >FENCE >> > > instructions, even on modern Intel CPUs like Haswell, e.g., >please see >> > > the three sources: >> > > >> > > " 11.5.1 Locked Instructions as Memory Barriers >> > > Optimization >> > > Use locked instructions to implement Store/Store and Store/Load >barriers. >> > > " >> > > http://support.amd.com/TechDocs/47414_15h_sw_opt_guide.pdf >> > > >> > > "lock addl %(rsp), 0 is a better solution for StoreLoad barrier >": >> > > http://shipilev.net/blog/2014/on-the-fence-with-dependencies/ >> > > >> > > "...locked instruction are more efficient barriers...": >> > > >http://www.pvk.ca/Blog/2014/10/19/performance-optimisation-~-writing-an-essay/ >> > > >> > > I also found that FreeBSD prefers Locked Add. >> > > >> > > So, I'm curious why Linux prefers MFENCE. >> > > I guess I may be missing something. >> > > >> > > I tried to google the question, but didn't find an answer. >> > >> > It's being worked on, see this thread on lkml from a few weeks ago: >> > >> > C Jan 13 Michael S. Tsir | [PATCH v3 0/4] x86: faster >mb()+documentation tweaks >> > C Jan 13 Michael S. Tsir | ├─>[PATCH v3 1/4] x86: add cc >clobber for addl >> > C Jan 13 Michael S. Tsir | ├─>[PATCH v3 2/4] x86: drop a >comment left over from X86_OOSTORE >> > C Jan 13 Michael S. Tsir | ├─>[PATCH v3 3/4] x86: tweak the >comment about use of wmb for IO >> > C Jan 13 Michael S. Tsir | ├─>[PATCH v3 4/4] x86: drop mfence >in favor of lock+addl >> > >> > The 4th patch changes MFENCE to a LOCK ADDL locked instruction. >> >> Lots of additional chatter here: >> >> lkml.kernel.org/r/20160112150032-mutt-send-email-mst@redhat.com >> >> And some useful bits here: >> >> lkml.kernel.org/r/56957D54.5000602@zytor.com >> >> latest version here: >> >> lkml.kernel.org/r/1453921746-16178-1-git-send-email-mst@redhat.comZ > >It's ready as far as I am concerned. >Basically we are just waiting for ack from hpa. And I'm still discussing this with the hardware people. It seems we can do this for *most* things, but not all; the question is where exactly we need to do something different. -- Sent from my Android device with K-9 Mail. Please excuse brevity and formatting.