From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <jianbo.liu@linaro.org>
Received: from mail-yw0-f178.google.com (mail-yw0-f178.google.com
 [209.85.161.178]) by dpdk.org (Postfix) with ESMTP id E6E28275D
 for <dev@dpdk.org>; Wed,  3 Aug 2016 11:33:04 +0200 (CEST)
Received: by mail-yw0-f178.google.com with SMTP id z8so222486903ywa.1
 for <dev@dpdk.org>; Wed, 03 Aug 2016 02:33:04 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google;
 h=mime-version:in-reply-to:references:from:date:message-id:subject:to
 :cc; bh=HP1E90bCuquZZMNcXHmZuEniSdJLW7RkrPrR7i6IKXk=;
 b=hq/PtwTSKFkW1+TVNVB6HEmup5Kq2FGKC0B5pIyjV7O3VVSsxIqmaP6SchWX9RWW/L
 Zp8UD4S34kWZPCWgvcn30N1KgLtkEPe5/wdESio/35T0dESWm+h/NhfcwlgXS08bFaNe
 WXQ2Xn6Ht3FThhgHqNUqwKzgUDsWryLjqq+io=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
 d=1e100.net; s=20130820;
 h=x-gm-message-state:mime-version:in-reply-to:references:from:date
 :message-id:subject:to:cc;
 bh=HP1E90bCuquZZMNcXHmZuEniSdJLW7RkrPrR7i6IKXk=;
 b=CMj3lj1+myhFZPVr1SfX2eTJymW0cUj8tUZ8jULIuPfivTqvn+L8p5Lw0D8DhX/kSb
 xqh1G4ok6AwAVgP98AF5D4a7V9knyxf8Z/5K+UXgx/HHb6cDikYuaEPuO2njsPr6ucpx
 P1wu07DkqOdKHe28d9dprMS2mE5y/gTX+6Q+3wvsEGvqUibfHVMmWKjY1g7qV9yST/ot
 L8B+dK5lnbErVAExxco4TAjNRDyxcD9jCnOuQn5Z8V30G9mbo0FlUZRIp9z8EB4QiGIR
 UDUDIZApx3TX6Z5U5yC36WeGp7SC/8gDKbG6XhTzhhyxEQSDyzKN3atKXnD3KEl4aeH3
 +hDA==
X-Gm-Message-State: AEkoouu133r0ZvtkccHyj3xVDW4iqVpqB98Ioj7NPC0roBxILnO+VPPkqPWfeAsW50U27MdoGxnmGm11TaB+GkJE
X-Received: by 10.129.91.84 with SMTP id p81mr52981856ywb.25.1470216784232;
 Wed, 03 Aug 2016 02:33:04 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.37.3.132 with HTTP; Wed, 3 Aug 2016 02:33:03 -0700 (PDT)
In-Reply-To: <3008268.1MMmpoWvzz@xps13>
References: <1470121107-7622-1-git-send-email-jianbo.liu@linaro.org>
 <2DBBFF226F7CF64BAFCA79B681719D9502046341@shsmsx102.ccr.corp.intel.com>
 <CAP4Qi39C0vNSbkOaVapiSGBbt=em7kef9XLP-W3UjDjKMWoTCg@mail.gmail.com>
 <3008268.1MMmpoWvzz@xps13>
From: Jianbo Liu <jianbo.liu@linaro.org>
Date: Wed, 3 Aug 2016 17:33:03 +0800
Message-ID: <CAP4Qi3-y3vJh7Z6jpwOEevZ=H0qxERf6Jo6bjnSWs2z1agbJHA@mail.gmail.com>
To: Thomas Monjalon <thomas.monjalon@6wind.com>
Cc: dev@dpdk.org, "Yao, Lei A" <lei.a.yao@intel.com>, 
 "Zhang, Helin" <helin.zhang@intel.com>, "Wu, Jingjing" <jingjing.wu@intel.com>
Content-Type: text/plain; charset=UTF-8
Subject: Re: [dpdk-dev] [PATCH] i40e: enable i40e pmd on ARM platform
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
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>
X-List-Received-Date: Wed, 03 Aug 2016 09:33:05 -0000

Hi Thomas,

On 3 August 2016 at 15:58, Thomas Monjalon <thomas.monjalon@6wind.com> wrote:
> 2016-08-03 14:02, Jianbo Liu:
>> I think we can place a compiling condition before rte_rmb() to avoid
>> performance decrease on x86.
>> For example:  #if defined(RTE_ARCH_ARM) || defined(RTE_ARCH_ARM64)
>
> Please could you explain why a memory barrier would be needed on ARM but

The reason is that ARM is weealy ordered processor, and data access
will be executed out of order to improve performance.
In this case, we have to read 2 times, 8 descriptors each. The read
statuses could be wrong if no memory barrier.
I also got the outdated status for some descriptors in my testing.

> not on x86? What about other architectures?

I think Konstantin gave me a good solution, by using rte_smp_rmb :)

Jianbo