| Message ID | 20210303152903.11172-1-maciej.fijalkowski@intel.com |
|---|---|
| State | New |
| Headers | show |
| Series | [bpf] veth: store queue_mapping independently of XDP prog presence | expand |
On 2021/03/04 0:29, Maciej Fijalkowski wrote: > Currently, veth_xmit() would call the skb_record_rx_queue() only when > there is XDP program loaded on peer interface in native mode. > > If peer has XDP prog in generic mode, then netif_receive_generic_xdp() > has a call to netif_get_rxqueue(skb), so for multi-queue veth it will > not be possible to grab a correct rxq. > > To fix that, store queue_mapping independently of XDP prog presence on > peer interface. > > Fixes: 638264dc9022 ("veth: Support per queue XDP ring") > Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> I did like this in order to keep the default behavior for non-xdp case, but generic XDP should behave the same as native XDP, so Acked-by: Toshiaki Makita <toshiaki.makita1@gmail.com>
Hello: This patch was applied to bpf/bpf.git (refs/heads/master): On Wed, 3 Mar 2021 16:29:03 +0100 you wrote: > Currently, veth_xmit() would call the skb_record_rx_queue() only when > there is XDP program loaded on peer interface in native mode. > > If peer has XDP prog in generic mode, then netif_receive_generic_xdp() > has a call to netif_get_rxqueue(skb), so for multi-queue veth it will > not be possible to grab a correct rxq. > > [...] Here is the summary with links: - [bpf] veth: store queue_mapping independently of XDP prog presence https://git.kernel.org/bpf/bpf/c/edbea9220251 You are awesome, thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html
diff --git a/drivers/net/veth.c b/drivers/net/veth.c index aa1a66ad2ce5..34e49c75db42 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -302,8 +302,7 @@ static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev) if (rxq < rcv->real_num_rx_queues) { rq = &rcv_priv->rq[rxq]; rcv_xdp = rcu_access_pointer(rq->xdp_prog); - if (rcv_xdp) - skb_record_rx_queue(skb, rxq); + skb_record_rx_queue(skb, rxq); } skb_tx_timestamp(skb);
Currently, veth_xmit() would call the skb_record_rx_queue() only when there is XDP program loaded on peer interface in native mode. If peer has XDP prog in generic mode, then netif_receive_generic_xdp() has a call to netif_get_rxqueue(skb), so for multi-queue veth it will not be possible to grab a correct rxq. To fix that, store queue_mapping independently of XDP prog presence on peer interface. Fixes: 638264dc9022 ("veth: Support per queue XDP ring") Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> --- drivers/net/veth.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)