Re: [PATCH net-next 1/2] net: airoha: Fix TSO support for header cloned skbs

[Mateusz Polchlopek <mateusz.polchlopek@intel.com>]

On 2/12/2025 7:51 AM, Lorenzo Bianconi wrote:
> For GSO packets, skb_cow_head() will reallocate the skb for TSO header
> cloned skbs in airoha_dev_xmit(). For this reason, sinfo pointer can be
> no more valid. Fix the issue relying on skb_shinfo() macro directly in
> airoha_dev_xmit().
> This is not a user visible issue since we can't currently enable TSO for
> DSA user ports since we are missing to initialize net_device
> vlan_features field.
> 
> Fixes: 23020f049327 ("net: airoha: Introduce ethernet support for EN7581 SoC")
> Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
> ---
>   drivers/net/ethernet/mediatek/airoha_eth.c | 10 +++++-----
>   1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/net/ethernet/mediatek/airoha_eth.c b/drivers/net/ethernet/mediatek/airoha_eth.c
> index 09f448f291240257c5748725848ede231c502fbd..aa5f220ddbcf9ca5bee1173114294cb3aec701c9 100644
> --- a/drivers/net/ethernet/mediatek/airoha_eth.c
> +++ b/drivers/net/ethernet/mediatek/airoha_eth.c
> @@ -2556,11 +2556,10 @@ static u16 airoha_dev_select_queue(struct net_device *dev, struct sk_buff *skb,
>   static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb,
>   				   struct net_device *dev)
>   {
> -	struct skb_shared_info *sinfo = skb_shinfo(skb);
>   	struct airoha_gdm_port *port = netdev_priv(dev);
> +	u32 nr_frags = 1 + skb_shinfo(skb)->nr_frags;
>   	u32 msg0, msg1, len = skb_headlen(skb);
>   	struct airoha_qdma *qdma = port->qdma;
> -	u32 nr_frags = 1 + sinfo->nr_frags;
>   	struct netdev_queue *txq;
>   	struct airoha_queue *q;
>   	void *data = skb->data;
> @@ -2583,8 +2582,9 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb,
>   		if (skb_cow_head(skb, 0))
>   			goto error;
>   
> -		if (sinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)) {
> -			__be16 csum = cpu_to_be16(sinfo->gso_size);
> +		if (skb_shinfo(skb)->gso_type & (SKB_GSO_TCPV4 |
> +						 SKB_GSO_TCPV6)) {
> +			__be16 csum = cpu_to_be16(skb_shinfo(skb)->gso_size);
>   
>   			tcp_hdr(skb)->check = (__force __sum16)csum;
>   			msg0 |= FIELD_PREP(QDMA_ETH_TXMSG_TSO_MASK, 1);
> @@ -2613,7 +2613,7 @@ static netdev_tx_t airoha_dev_xmit(struct sk_buff *skb,
>   	for (i = 0; i < nr_frags; i++) {
>   		struct airoha_qdma_desc *desc = &q->desc[index];
>   		struct airoha_queue_entry *e = &q->entry[index];
> -		skb_frag_t *frag = &sinfo->frags[i];
> +		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
>   		dma_addr_t addr;
>   		u32 val;
>   
> 

Thanks for this change:
Reviewed-by: Mateusz Polchlopek <mateusz.polchlopek@intel.com>