[PATCH net] net: dsa: mt7530: don't change PVC_EG_TAG when CPU port becomes VLAN-aware

[Vladimir Oltean]

Hi Arınç,

On Mon, Feb 06, 2023 at 07:41:06PM +0300, Arınç ÜNAL wrote:
> Finally I got time. It's been a seismically active day where I'm from.

My deepest condolences to those who experienced tragedies after today's
earthquakes. A lot of people in neighboring countries are horrified
thinking when this will happen to them. Hopefully you aren't living in
Gaziantep or nearby cities.

> # ping 192.168.2.2
> PING 192.168.2.2
> [   39.508013] mtk_soc_eth 1b100000.ethernet eth1: dsa_switch_rcv: there is no metadata dst attached to skb 0xc2dfecc0
> 
> # ping 192.168.2.2
> PING 192.168.2.2
> [   22.674182] mtk_soc_eth 1b100000.ethernet eth1: mtk_poll_rx: received skb 0xc2d67840 without VLAN/DSA tag present

Thank you so much for testing. Would you mind cleaning everything up and
testing with this patch instead (formatted on top of net-next)?
Even if you need to adapt to your tree, hopefully you get the idea from
the commit message.

>From 218025fd0c33a06865e4202c5170bfc17e26cc75 Mon Sep 17 00:00:00 2001
From: Vladimir Oltean <vladimir.oltean at nxp.com>
Date: Mon, 6 Feb 2023 19:03:53 +0200
Subject: [PATCH] net: ethernet: mtk_eth_soc: fix DSA TX tag hwaccel for switch
 port 0
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Arınç reports that on his MT7621AT Unielec U7621-06 board and MT7623NI
Bananapi BPI-R2, packets received by the CPU over mt7530 switch port 0
(of which this driver acts as the DSA master) are not processed
correctly by software. More precisely, they arrive without a DSA tag
(in packet or in the hwaccel area - skb_metadata_dst()), so DSA cannot
demux them towards the switch's interface for port 0. Traffic from other
ports receives a skb_metadata_dst() with the correct port and is demuxed
properly.

Looking at mtk_poll_rx(), it becomes apparent that this driver uses the
skb vlan hwaccel area:

	union {
		u32		vlan_all;
		struct {
			__be16	vlan_proto;
			__u16	vlan_tci;
		};
	};

as a temporary storage for the VLAN hwaccel tag, or the DSA hwaccel tag.
If this is a DSA master it's a DSA hwaccel tag, and finally clears up
the skb VLAN hwaccel header.

I'm guessing that the problem is the (mis)use of API.
skb_vlan_tag_present() looks like this:

 #define skb_vlan_tag_present(__skb)	(!!(__skb)->vlan_all)

So if both vlan_proto and vlan_tci are zeroes, skb_vlan_tag_present()
returns precisely false. I don't know for sure what is the format of the
DSA hwaccel tag, but I surely know that lowermost 3 bits of vlan_proto
are 0 when receiving from port 0:

	unsigned int port = vlan_proto & GENMASK(2, 0);

If the RX descriptor has no other bits set to non-zero values in
RX_DMA_VTAG, then the call to __vlan_hwaccel_put_tag() will not, in
fact, make the subsequent skb_vlan_tag_present() return true, because
it's implemented like this:

static inline void __vlan_hwaccel_put_tag(struct sk_buff *skb,
					  __be16 vlan_proto, u16 vlan_tci)
{
	skb->vlan_proto = vlan_proto;
	skb->vlan_tci = vlan_tci;
}

What we need to do to fix this problem (assuming this is the problem) is
to stop using skb->vlan_all as temporary storage for driver affairs, and
just create some local variables that serve the same purpose, but
hopefully better. Instead of calling skb_vlan_tag_present(), let's look
at a boolean has_hwaccel_tag which we set to true when the RX DMA
descriptors have something. Disambiguate based on netdev_uses_dsa()
whether this is a VLAN or DSA hwaccel tag, and only call
__vlan_hwaccel_put_tag() if we're certain it's a VLAN tag.

Link: https://lore.kernel.org/netdev/704f3a72-fc9e-714a-db54-272e17612637@arinc9.com/
Fixes: 2d7605a72906 ("net: ethernet: mtk_eth_soc: enable hardware DSA untagging")
Reported-by: Arınç ÜNAL <arinc.unal at arinc9.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean at nxp.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 24 ++++++++++++---------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index f1cb1efc94cf..64b575fbe317 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -1921,7 +1921,9 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 
 	while (done < budget) {
 		unsigned int pktlen, *rxdcsum;
+		bool has_hwaccel_tag = false;
 		struct net_device *netdev;
+		u16 vlan_proto, vlan_tci;
 		dma_addr_t dma_addr;
 		u32 hash, reason;
 		int mac = 0;
@@ -2061,27 +2063,29 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 
 		if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX) {
 			if (MTK_HAS_CAPS(eth->soc->caps, MTK_NETSYS_V2)) {
-				if (trxd.rxd3 & RX_DMA_VTAG_V2)
-					__vlan_hwaccel_put_tag(skb,
-						htons(RX_DMA_VPID(trxd.rxd4)),
-						RX_DMA_VID(trxd.rxd4));
+				if (trxd.rxd3 & RX_DMA_VTAG_V2) {
+					vlan_proto = RX_DMA_VPID(trxd.rxd4);
+					vlan_tci = RX_DMA_VID(trxd.rxd4);
+					has_hwaccel_tag = true;
+				}
 			} else if (trxd.rxd2 & RX_DMA_VTAG) {
-				__vlan_hwaccel_put_tag(skb, htons(RX_DMA_VPID(trxd.rxd3)),
-						       RX_DMA_VID(trxd.rxd3));
+				vlan_proto = RX_DMA_VPID(trxd.rxd3);
+				vlan_tci = RX_DMA_VID(trxd.rxd3);
+				has_hwaccel_tag = true;
 			}
 		}
 
 		/* When using VLAN untagging in combination with DSA, the
 		 * hardware treats the MTK special tag as a VLAN and untags it.
 		 */
-		if (skb_vlan_tag_present(skb) && netdev_uses_dsa(netdev)) {
-			unsigned int port = ntohs(skb->vlan_proto) & GENMASK(2, 0);
+		if (has_hwaccel_tag && netdev_uses_dsa(netdev)) {
+			unsigned int port = vlan_proto & GENMASK(2, 0);
 
 			if (port < ARRAY_SIZE(eth->dsa_meta) &&
 			    eth->dsa_meta[port])
 				skb_dst_set_noref(skb, &eth->dsa_meta[port]->dst);
-
-			__vlan_hwaccel_clear_tag(skb);
+		} else if (has_hwaccel_tag) {
+			__vlan_hwaccel_put_tag(skb, htons(vlan_proto), vlan_tci);
 		}
 
 		skb_record_rx_queue(skb, 0);
-- 
2.34.1