Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Ioana Ciornei <ioana.ciornei@nxp.com>
To:     Andrew Lunn <andrew@lunn.ch>
CC:     "davem@davemloft.net" <davem@davemloft.net>,
        "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
        "f.fainelli@gmail.com" <f.fainelli@gmail.com>,
        Ioana Ciocoi Radulescu <ruxandra.radulescu@nxp.com>
Subject: Re: [PATCH] dpaa2-ethsw: move the DPAA2 Ethernet Switch driver out of
 staging
Thread-Topic: [PATCH] dpaa2-ethsw: move the DPAA2 Ethernet Switch driver out
 of staging
Thread-Index: AQHVTsse6Qsm2zNpN0akGPBvGQTj7Q==
Date:   Tue, 13 Aug 2019 07:21:41 +0000
Message-ID: <VI1PR0402MB28005DB52A30278BA8CC16D0E0D20@VI1PR0402MB2800.eurprd04.prod.outlook.com>
References: <1565366213-20063-1-git-send-email-ioana.ciornei@nxp.com>
 <20190809190459.GW27917@lunn.ch>
 <VI1PR0402MB2800FF2E5C4DE24B25E7D843E0D10@VI1PR0402MB2800.eurprd04.prod.outlook.com>
 <20190812135746.GL14290@lunn.ch>
Accept-Language: en-US
Content-Language: en-US
received-spf: None (protection.outlook.com: nxp.com does not designate
 permitted sender hosts)
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 5248ee45-1ea3-453e-56d7-08d71fbee352
X-MS-Exchange-CrossTenant-originalarrivaltime: 13 Aug 2019 07:21:41.2676
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 686ea1d3-bc2b-4c6f-a92c-d99c5c301635
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: 0r3lIv1gjmca4qx3QTc4KtWyjFs7MIaqYB1qadbXZN7ml7QcUc6InmfObTy/TKIYWXEDeZ3fQr6+QAV4IRC7DA==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: VI1PR0402MB3933
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 8/12/19 4:57 PM, Andrew Lunn wrote:=0A=
>> In the DPAA2 architecture MACs are not the only entities that can be=0A=
>> connected to a switch port.=0A=
>> Below is an exemple of a 4 port DPAA2 switch which is configured to=0A=
>> interconnect 2 DPNIs (network interfaces) and 2 DPMACs.=0A=
>>=0A=
>>=0A=
>>    [ethA]     [ethB]     [ethC]     [ethD]     [ethE]     [ethF]=0A=
>>       :          :          :          :          :          :=0A=
>>       :          :          :          :          :          :=0A=
>> [eth drv]  [eth drv]  [                ethsw drv              ]=0A=
>>       :          :          :          :          :          :        ke=
rnel=0A=
>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=0A=
>>       :          :          :          :          :          :=0A=
>> hardware=0A=
>>    [DPNI]      [DPNI]     [=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D DPSW =
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D]=0A=
>>       |          |          |          |          |          |=0A=
>>       |           ----------           |       [DPMAC]    [DPMAC]=0A=
>>        -------------------------------            |          |=0A=
>>                                                   |          |=0A=
>>                                                 [PHY]      [PHY]=0A=
>>=0A=
>> You can see it as a hardware-accelerated software bridge where=0A=
>> forwarding rules are managed from the host software partition.=0A=
> =0A=
> Hi Ioana=0A=
=0A=
Hi Andrew,=0A=
=0A=
> =0A=
> What are the use cases for this?=0A=
> =0A=
> Configuration is rather unintuitive. To bridge etha and ethb you need=0A=
> to=0A=
> =0A=
> ip link add name br0 type bridge=0A=
> ip link set ethc master br0=0A=
> ip link set ethd master br0=0A=
> =0A=
> And once you make ethc and ethd actually send/receive frames, etha and=0A=
> ethc become equivalent.=0A=
> =0A=
> If this was a PCI device, i could imagine passing etha into a VM as a=0A=
> PCI VF. But i don't think it is PCI?=0A=
=0A=
Indeed it's not PCI but we can pass etha to a VM. That's the main use =0A=
case of having DPNIs connected to a switch object.=0A=
=0A=
Our direct assignment solution for DPAA2 is not upstream yet (the case =0A=
with many of our drivers :) ) but the main idea is exposing to the VM a =0A=
fsl-mc bus on which it can find any DPAA2 objects needed to configure a =0A=
network interface (using firmware calls).=0A=
=0A=
=0A=
> =0A=
> I'm not sure moving etha into a different name space makes much sense=0A=
> either. My guess would be, a veth pair with one end connected to the=0A=
> software bridge would be more efficient than DMAing the packet out and=0A=
> then back in again. >=0A=
>       Thanks=0A=
> 	Andrew=0A=
>=0A=
=0A=
=0A=
That's really an interesting test to be made since we can make a veth =0A=
pair from DPAA2 objects just connecting two DPNIs back to back. I'll =0A=
make some performance tests and compare against the software veth.=0A=
=0A=
=0A=
Thanks,=0A=
Ioana=0A=
=0A=
=0A=