From mboxrd@z Thu Jan  1 00:00:00 1970
From: Thomas Monjalon <thomas@monjalon.net>
Subject: Re: [PATCH] config: make AVX and AVX512 configurable
Date: Thu, 27 Apr 2017 11:20:22 +0200
Message-ID: <1558565.2fZJ5xpf6Y@xps>
References: <1493310859-49106-1-git-send-email-zhihong.wang@intel.com>
 <2616350.yKybQRU0rL@xps>
 <8F6C2BD409508844A0EFC19955BE09415126F7FE@SHSMSX103.ccr.corp.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7Bit
Cc: "Richardson, Bruce" <bruce.richardson@intel.com>, dev@dpdk.org,
 yuanhan.liu@linux.intel.com
To: "Wang, Zhihong" <zhihong.wang@intel.com>
Return-path: <dev-bounces@dpdk.org>
Received: from out4-smtp.messagingengine.com (out4-smtp.messagingengine.com
 [66.111.4.28]) by dpdk.org (Postfix) with ESMTP id 5F1322C4B
 for <dev@dpdk.org>; Thu, 27 Apr 2017 11:20:31 +0200 (CEST)
In-Reply-To: <8F6C2BD409508844A0EFC19955BE09415126F7FE@SHSMSX103.ccr.corp.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>

27/04/2017 11:18, Wang, Zhihong:
> From: Thomas Monjalon [mailto:thomas@monjalon.net]
> > 27/04/2017 18:34, Zhihong Wang:
> > > Making AVX and AVX512 configurable is useful for performance and power
> > > testing.
> > >
> > > The similar kernel patch at https://patchwork.kernel.org/patch/9618883/.
> > [...]
> > > +#
> > > +# Recognize/ignore the AVX/AVX512 CPU flags for performance/power
> > testing
> > > +#
> > > +CONFIG_RTE_ENABLE_AVX=y
> > > +CONFIG_RTE_ENABLE_AVX512=n
> > 
> > It is disabling AVX512 in default configuration.
> > Please explain this behaviour change.
> 
> Though AVX512 rte_memcpy has been in DPDK for quite a while it's still
> unproven in hardware with rich use cases. Mark it as experimental for
> now, user can enable it for their own testing.
> 
> Will enable it with enough field tests and possible optimization.
> 
> Should I add the explanation in commit log, or comments in the source,
> or both?

Yes please, add the explanation in the commit log and experimental comment
in the config.