From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <konstantin.ananyev@intel.com>
Received: from mga01.intel.com (mga01.intel.com [192.55.52.88])
by dpdk.org (Postfix) with ESMTP id 081F2C31A
for <dev@dpdk.org>; Fri, 5 Feb 2016 16:15:58 +0100 (CET)
Received: from orsmga003.jf.intel.com ([10.7.209.27])
by fmsmga101.fm.intel.com with ESMTP; 05 Feb 2016 07:15:57 -0800
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.22,401,1449561600"; d="scan'208";a="741162458"
Received: from irsmsx104.ger.corp.intel.com ([163.33.3.159])
by orsmga003.jf.intel.com with ESMTP; 05 Feb 2016 07:15:57 -0800
Received: from irsmsx105.ger.corp.intel.com ([169.254.7.237]) by
IRSMSX104.ger.corp.intel.com ([163.33.3.159]) with mapi id 14.03.0248.002;
Fri, 5 Feb 2016 15:15:15 +0000
From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: "Pattan, Reshma" <reshma.pattan@intel.com>, Thomas Monjalon
<thomas.monjalon@6wind.com>
Thread-Topic: [dpdk-dev] [PATCH 1/3] examples/distributor: fix build for
non-x86 arch
Thread-Index: AQHRYCVYDGUmOEp2uEqs3il22R8SI58djaMAgAAAhdA=
Date: Fri, 5 Feb 2016 15:15:14 +0000
Message-ID: <2601191342CEEE43887BDE71AB97725836B0465D@irsmsx105.ger.corp.intel.com>
References: <1454684049-27195-1-git-send-email-thomas.monjalon@6wind.com>
<1454684049-27195-2-git-send-email-thomas.monjalon@6wind.com>
<3AEA2BF9852C6F48A459DA490692831FFBEB4C@IRSMSX109.ger.corp.intel.com>
In-Reply-To: <3AEA2BF9852C6F48A459DA490692831FFBEB4C@IRSMSX109.ger.corp.intel.com>
Accept-Language: en-IE, en-US
Content-Language: en-US
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
x-originating-ip: [163.33.239.181]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 1/3] examples/distributor: fix build for
non-x86 arch
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, 05 Feb 2016 15:15:59 -0000
> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Pattan, Reshma
> Sent: Friday, February 05, 2016 3:08 PM
> To: Thomas Monjalon
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/3] examples/distributor: fix build for n=
on-x86 arch
>=20
> Hi Thomas,
>=20
> > -----Original Message-----
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Thomas Monjalon
> > Sent: Friday, February 5, 2016 2:54 PM
> > To: Richardson, Bruce; Horton, Remy
> > Cc: dev@dpdk.org
> > Subject: [dpdk-dev] [PATCH 1/3] examples/distributor: fix build for non=
-x86 arch
> >
> > _mm_prefetch is defined only in x86 compilers.
> > rte_prefetch functions are defined in EAL for each arch, and must be pr=
eferred
> > over compiler intrinsics.
> >
> > Fixes: 07db4a975094 ("examples/distributor: new sample app")
> >
> > Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
> > ---
> > examples/distributor/main.c | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/examples/distributor/main.c b/examples/distributor/main.c =
index
> > 4e74f8f..87344ac 100644
> > --- a/examples/distributor/main.c
> > +++ b/examples/distributor/main.c
> > @@ -335,13 +335,13 @@ lcore_tx(struct rte_ring *in_r)
> >
> > /* for traffic we receive, queue it up for transmit */
> > uint16_t i;
> > - _mm_prefetch((void *)bufs[0], 0);
> > - _mm_prefetch((void *)bufs[1], 0);
> > - _mm_prefetch((void *)bufs[2], 0);
> > + rte_prefetch0((void *)bufs[0]);
> > + rte_prefetch0((void *)bufs[1]);
> > + rte_prefetch0((void *)bufs[2]);
>=20
> Some time back Jerin Jacob has sent patch for replacing the _mm_prefetch =
with rte_prefetch_non_temporal. This is FYI.
Yep:
http://dpdk.org/dev/patchwork/patch/9369/
http://dpdk.org/dev/patchwork/patch/9370/
And on IA _mm_prefetch(p, 0) !=3D rte_prefetch0(p).
I think, as in Jacob implementation, it should be PREFETCHNTA.
Konstantin
>=20
> Thanks,
> Reshma
>=20
> > for (i =3D 0; i < nb_rx; i++) {
> > struct output_buffer *outbuf;
> > uint8_t outp;
> > - _mm_prefetch((void *)bufs[i + 3], 0);
> > + rte_prefetch0((void *)bufs[i + 3]);
> > /*
> > * workers should update in_port to hold the
> > * output port value
> > --
> > 2.7.0