From mboxrd@z Thu Jan  1 00:00:00 1970
From: Thomas Monjalon <thomas.monjalon@6wind.com>
Subject: Re: [dpdk-stable] [PATCH 1/2] net/virtio: fix
	performance regression due to TSO enabling
Date: Wed, 11 Jan 2017 15:51:22 +0100
Message-ID: <1610499.AMUobBPor6@xps13>
References: <1484108832-19907-1-git-send-email-yuanhan.liu@linux.intel.com>
 <1484108832-19907-2-git-send-email-yuanhan.liu@linux.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7Bit
Cc: dev@dpdk.org, Tan Jianfeng <jianfeng.tan@intel.com>,
 Wang Zhihong <zhihong.wang@intel.com>, Olivier Matz <olivier.matz@6wind.com>,
 Maxime Coquelin <maxime.coquelin@redhat.com>,
 "Michael S. Tsirkin" <mst@redhat.com>
To: Yuanhan Liu <yuanhan.liu@linux.intel.com>,
 Jianbo Liu <jianbo.liu@linaro.org>, Jerin Jacob <jerin.jacob@cavium.com>,
 Jan Viktorin <viktorin@rehivetech.com>, Chao Zhu <chaozhu@linux.vnet.ibm.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mail-wm0-f41.google.com (mail-wm0-f41.google.com [74.125.82.41])
 by dpdk.org (Postfix) with ESMTP id 618F2F614
 for <dev@dpdk.org>; Wed, 11 Jan 2017 15:51:18 +0100 (CET)
Received: by mail-wm0-f41.google.com with SMTP id f73so27312974wmf.1
 for <dev@dpdk.org>; Wed, 11 Jan 2017 06:51:18 -0800 (PST)
In-Reply-To: <1484108832-19907-2-git-send-email-yuanhan.liu@linux.intel.com>
List-Id: DPDK patches and discussions <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>

2017-01-11 12:27, Yuanhan Liu:
> The fact that virtio net header is initiated to zero in PMD driver
> init stage means that these costly writes are unnecessary and could
> be avoided:
> 
>     if (hdr->csum_start != 0)
>         hdr->csum_start = 0;
> 
> And that's what the macro ASSIGN_UNLESS_EQUAL does. With this, the
> performance drop introduced by TSO enabling is recovered: it could
> be up to 20% in micro benchmarking.

This patch is adding a condition to assignments.
We need a benchmark on other architectures like ARM. Please anyone?


[...]
> +/* avoid write operation when necessary, to lessen cache issues */
> +#define ASSIGN_UNLESS_EQUAL(var, val) do {	\
> +	if ((var) != (val))			\
> +		(var) = (val);			\
> +} while (0)