From mboxrd@z Thu Jan  1 00:00:00 1970
From: "De Lara Guarch, Pablo" <pablo.de.lara.guarch@intel.com>
Subject: Re: [PATCH v2] app/testpmd: add bitrate stats option
Date: Tue, 2 May 2017 11:18:52 +0000
Message-ID: <E115CCD9D858EF4F90C690B0DCB4D89747814CE8@IRSMSX108.ger.corp.intel.com>
References: <1493211774-28249-1-git-send-email-remy.horton@intel.com>
 <1493377213-156955-1-git-send-email-pablo.de.lara.guarch@intel.com>
 <D52CE0DB.13C49C%Harish.Patil@cavium.com> <1984316.gczFh5jcb3@xps>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Cc: "dev@dpdk.org" <dev@dpdk.org>, "Wu, Jingjing" <jingjing.wu@intel.com>
To: Thomas Monjalon <thomas@monjalon.net>, "Patil, Harish"
 <Harish.Patil@cavium.com>, "Horton, Remy" <remy.horton@intel.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mga14.intel.com (mga14.intel.com [192.55.52.115])
 by dpdk.org (Postfix) with ESMTP id 9BF54374E
 for <dev@dpdk.org>; Tue,  2 May 2017 13:18:55 +0200 (CEST)
In-Reply-To: <1984316.gczFh5jcb3@xps>
Content-Language: en-US
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>



> -----Original Message-----
> From: Thomas Monjalon [mailto:thomas@monjalon.net]
> Sent: Monday, May 01, 2017 9:22 PM
> To: Patil, Harish; Horton, Remy
> Cc: dev@dpdk.org; De Lara Guarch, Pablo; Wu, Jingjing
> Subject: Re: [dpdk-dev] [PATCH v2] app/testpmd: add bitrate stats option
>=20
> 01/05/2017 22:07, Patil, Harish:
> > Hi Remy,
> > Have a small suggestion here.
> > Since testpmd uses new libraries of librte_latencystats and
> > librte_bitratestats it hurts packet processing performance.
> > Many users who use testpmd to do the initial performance benchmarks
> may
> > not be aware of such a feature is default enabled.
>=20
> Yes, the default config of testpmd must give good performance.
>=20
> > So can we disable this feature by default in the config?
> > *         CONFIG_RTE_LIBRTE_BITRATE=3Dn
> > *         CONFIG_RTE_LIBRTE_LATENCY_STATS=3Dn
> > Only those folks interested in latency/jitter measurements can recompil=
e
> > with those configs enabled.
>=20
> I disagree about compile-time options.
> It should be a run-time option of testpmd.
>=20
> Please Remy (or others),
> disable the metrics in the default configuration of testpmd,
> before the 17.05 release.
> You have few days, it is urgent.

Bitrate stats are disabled by default, in testpmd.
I assume that the code that you want to avoid is:

                for (sm_id =3D 0; sm_id < nb_fs; sm_id++)
                        (*pkt_fwd)(fsm[sm_id]);
#ifdef RTE_LIBRTE_BITRATE
                if (bitrate_enabled !=3D 0 &&
                                bitrate_lcore_id =3D=3D rte_lcore_id()) {
                        tics_current =3D rte_rdtsc();
                        if (tics_current - tics_datum >=3D tics_per_1sec) {

Unless --bitrate-stats is used, bitrate_enabled =3D 0, so all this code won=
't be run.

I can send a patch to check the latencystats_enabled flag first, following =
the approach above, here:

#ifdef RTE_LIBRTE_LATENCY_STATS
                if (latencystats_lcore_id =3D=3D rte_lcore_id())


Thanks,
Pablo