From mboxrd@z Thu Jan  1 00:00:00 1970
From: Stephen Hemminger <stephen@networkplumber.org>
Subject: Re: [PATCH v3 2/2] eal/x86: Use lock-prefixed
 instructions to reduce cost of rte_smp_mb()
Date: Tue, 30 Jan 2018 11:33:16 -0800
Message-ID: <20180130113316.146088f5@xeon-e3>
References: <1516028680-2342-3-git-send-email-konstantin.ananyev@intel.com>
 <1516028971-3409-1-git-send-email-konstantin.ananyev@intel.com>
 <3552780.vzqTqTtnLr@xps>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Cc: Konstantin Ananyev <konstantin.ananyev@intel.com>, dev@dpdk.org
To: Thomas Monjalon <thomas@monjalon.net>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-pf0-f195.google.com (mail-pf0-f195.google.com
 [209.85.192.195]) by dpdk.org (Postfix) with ESMTP id 0E2DE1B6FC
 for <dev@dpdk.org>; Tue, 30 Jan 2018 20:33:20 +0100 (CET)
Received: by mail-pf0-f195.google.com with SMTP id b25so10055858pfd.9
 for <dev@dpdk.org>; Tue, 30 Jan 2018 11:33:19 -0800 (PST)
In-Reply-To: <3552780.vzqTqTtnLr@xps>
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Mon, 29 Jan 2018 16:47:45 +0100
Thomas Monjalon <thomas@monjalon.net> wrote:

> 15/01/2018 16:09, Konstantin Ananyev:
> > On x86 it  is possible to use lock-prefixed instructions to get
> > the similar effect as mfence.
> > As pointed by Java guys, on most modern HW that gives a better
> > performance than using mfence:
> > https://shipilev.net/blog/2014/on-the-fence-with-dependencies/
> > That patch adopts that technique for rte_smp_mb() implementation.
> > On BDW 2.2 mb_autotest on single lcore reports 2X cycle reduction,
> > i.e. from ~110 to ~55 cycles per operation.
> > 
> > Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> > Acked-by: Bruce Richardson <bruce.richardson@intel.com>  
> 
> Applied, thanks

Does this change since lfence is one of the ways to block Spectre variant.