From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <bruce.richardson@intel.com>
Received: from mga09.intel.com (mga09.intel.com [134.134.136.24])
 by dpdk.org (Postfix) with ESMTP id C886737B8
 for <dev@dpdk.org>; Fri,  6 May 2016 16:45:54 +0200 (CEST)
Received: from orsmga001.jf.intel.com ([10.7.209.18])
 by orsmga102.jf.intel.com with ESMTP; 06 May 2016 07:45:27 -0700
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.24,587,1455004800"; d="scan'208";a="947653504"
Received: from bricha3-mobl3.ger.corp.intel.com ([10.237.220.75])
 by orsmga001.jf.intel.com with SMTP; 06 May 2016 07:45:25 -0700
Received: by  (sSMTP sendmail emulation); Fri, 06 May 2016 15:45:24 +0025
Date: Fri, 6 May 2016 15:45:24 +0100
From: Bruce Richardson <bruce.richardson@intel.com>
To: Rasesh Mody <rasesh.mody@qlogic.com>
Cc: dev@dpdk.org, Dept-EngDPDKDev@qlogic.com,
 Harish Patil <harish.patil@qlogic.com>
Message-ID: <20160506144524.GB8880@bricha3-MOBL3>
References: <1462340302-25683-1-git-send-email-rasesh.mody@qlogic.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1462340302-25683-1-git-send-email-rasesh.mody@qlogic.com>
Organization: Intel Shannon Ltd.
User-Agent: Mutt/1.5.23 (2014-03-12)
Subject: Re: [dpdk-dev] [PATCH v4 1/3] bnx2x: refactor Tx routine for single
 doorbell mechanism
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: Fri, 06 May 2016 14:45:55 -0000

On Tue, May 03, 2016 at 10:38:20PM -0700, Rasesh Mody wrote:
> Change the Tx routine logic to ring the doorbell once per burst and not
> on every Tx packet. This driver-level optimization is necessary to achieve
> line rates for larger frame sizes (1k or more).
> 
> Fixes: 540a211084a7 ("bnx2x: driver core")
> 
> Signed-off-by: Rasesh Mody <rasesh.mody@qlogic.com>
> Signed-off-by: Harish Patil <harish.patil@qlogic.com>

Hi Rasesh, Harish,

there is quite a bit of churn to the tx_encap routine in this patch, but the
reason for that function change is not really explained well here - why it's
changed from a burst function to a single mbuf function. This is confusing when
the overall objective of the patch seems to be to do the opposite - have things
more burst oriented.

It might be better to have this patch split into two patches - one changing the
tx_encap function and then a second one to restructure the xmit function.

Regards,
/Bruce