Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Vladimir Oltean <vladimir.oltean@nxp.com>
To:     Florian Fainelli <f.fainelli@gmail.com>
CC:     "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
        Andrew Lunn <andrew@lunn.ch>,
        Vivien Didelot <vivien.didelot@gmail.com>,
        "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        open list <linux-kernel@vger.kernel.org>,
        "olteanv@gmail.com" <olteanv@gmail.com>,
        "nikolay@nvidia.com" <nikolay@nvidia.com>
Subject: Re: [PATCH net-next 1/2] net: dsa: untag the bridge pvid from rx skbs
Thread-Topic: [PATCH net-next 1/2] net: dsa: untag the bridge pvid from rx
 skbs
Thread-Index: AQHWkVdQ8QBtwDGDZECVVPbKJwPzZ6l14IgA
Date:   Wed, 23 Sep 2020 08:14:51 +0000
Message-ID: <20200923081450.2ghr6am4vjci6cd4@skbuf>
References: <20200923031155.2832348-1-f.fainelli@gmail.com>
 <20200923031155.2832348-2-f.fainelli@gmail.com>
In-Reply-To: <20200923031155.2832348-2-f.fainelli@gmail.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <514A92A2362A2C42A8A648BC225AAA14@eurprd04.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 3b609197-1c7c-422c-d483-08d85f98bf0f
X-MS-Exchange-CrossTenant-originalarrivaltime: 23 Sep 2020 08:14:51.7229
 (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: xCNeR52+3qGbrL/Zn6h5BIKYZSxEsLCAxi2CFREX//tILwpOWPfafoCfe7QhJbmSC8zU7Ianweo7NpxNtVXhIQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: VE1PR04MB6701
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, Sep 22, 2020 at 08:11:54PM -0700, Florian Fainelli wrote:
> diff --git a/include/net/dsa.h b/include/net/dsa.h
> index d16057c5987a..b539241a7533 100644
> --- a/include/net/dsa.h
> +++ b/include/net/dsa.h
> @@ -301,6 +301,14 @@ struct dsa_switch {
>  	 */
>  	bool			configure_vlan_while_not_filtering;
> =20
> +	/* If the switch driver always programs the CPU port as egress tagged
> +	 * despite the VLAN configuration indicating otherwise, then setting
> +	 * @untag_bridge_pvid will force the DSA receive path to pop the bridge=
's
> +	 * default_pvid VLAN tagged frames to offer a consistent behavior
> +	 * between a vlan_filtering=3D0 and vlan_filtering=3D1 bridge device.
> +	 */
> +	bool			untag_bridge_pvid;
> +
>  	/* In case vlan_filtering_is_global is set, the VLAN awareness state
>  	 * should be retrieved from here and not from the per-port settings.
>  	 */
> diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
> index 5c18c0214aac..dec4ab59b7c4 100644
> --- a/net/dsa/dsa.c
> +++ b/net/dsa/dsa.c
> @@ -225,6 +225,15 @@ static int dsa_switch_rcv(struct sk_buff *skb, struc=
t net_device *dev,
>  	skb->pkt_type =3D PACKET_HOST;
>  	skb->protocol =3D eth_type_trans(skb, skb->dev);
> =20
> +	if (unlikely(cpu_dp->ds->untag_bridge_pvid)) {
> +		nskb =3D dsa_untag_bridge_pvid(skb);
> +		if (!nskb) {
> +			kfree_skb(skb);
> +			return 0;
> +		}
> +		skb =3D nskb;
> +	}
> +
>  	s =3D this_cpu_ptr(p->stats64);
>  	u64_stats_update_begin(&s->syncp);
>  	s->rx_packets++;

I was thinking a lot simpler. Maybe you could just tail-call
dsa_untag_bridge_pvid(skb) at the end of your .rcv function instead of
putting it in the common receive path. I specifically wrote it to look
at hdr->h_vlan_proto instead of skb->protocol, so it wouldn't depend on
eth_type_trans().

Something like:

diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c
index 9c6c30649d13..118d253af5a7 100644
--- a/net/dsa/tag_brcm.c
+++ b/net/dsa/tag_brcm.c
@@ -192,7 +192,7 @@ static struct sk_buff *brcm_tag_rcv(struct sk_buff *skb=
, struct net_device *dev,
 		nskb->data - ETH_HLEN - BRCM_TAG_LEN,
 		2 * ETH_ALEN);
=20
-	return nskb;
+	return dsa_untag_bridge_pvid(nskb);
 }
=20
 static const struct dsa_device_ops brcm_netdev_ops =3D {
@@ -220,8 +220,14 @@ static struct sk_buff *brcm_tag_rcv_prepend(struct sk_=
buff *skb,
 					    struct net_device *dev,
 					    struct packet_type *pt)
 {
+	struct sk_buff *nskb;
+
 	/* tag is prepended to the packet */
-	return brcm_tag_rcv_ll(skb, dev, pt, ETH_HLEN);
+	nskb =3D brcm_tag_rcv_ll(skb, dev, pt, ETH_HLEN);
+	if (!nskb)
+		return nskb;
+
+	return dsa_untag_bridge_pvid(nskb);
 }
=20
 static const struct dsa_device_ops brcm_prepend_netdev_ops =3D {

Thanks,
-Vladimir=