From mboxrd@z Thu Jan  1 00:00:00 1970
From: Stephen Hemminger <stephen@networkplumber.org>
Subject: Re: [PATCH] mbuf: fix atomic refcnt update
 synchronization
Date: Fri, 2 Sep 2016 09:12:36 -0700
Message-ID: <20160902091236.57250d1a@xeon-e3>
References: <1472793906-5699-1-git-send-email-slayercat.subscription@gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: dev@dpdk.org, =?UTF-8?B?5p2O5p6X5ZOy?= <lilinzhe@ijinshan.com>
To: lilinzhe <slayercat.subscription@gmail.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-pf0-f181.google.com (mail-pf0-f181.google.com
 [209.85.192.181]) by dpdk.org (Postfix) with ESMTP id 5F8292BB3
 for <dev@dpdk.org>; Fri,  2 Sep 2016 18:12:27 +0200 (CEST)
Received: by mail-pf0-f181.google.com with SMTP id x72so43712098pfd.2
 for <dev@dpdk.org>; Fri, 02 Sep 2016 09:12:27 -0700 (PDT)
In-Reply-To: <1472793906-5699-1-git-send-email-slayercat.subscription@gmail.com>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://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: <http://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 Fri,  2 Sep 2016 13:25:06 +0800
lilinzhe <slayercat.subscription@gmail.com> wrote:

> From: =E6=9D=8E=E6=9E=97=E5=93=B2 <lilinzhe@ijinshan.com>
>=20
> chagne atomic ref update to always call atomic_add
>=20
> when mbuf is allocated by cpu1 and freed by cpu2. cpu1 cache may not be u=
pdated by such a set operation.
> causes refcnt reads incorrect values.

What architecture are you dealing with? On X86 memory is cache coherent.

Doing atomic operation all the time on each mbuf free would significantly
slow down performance.