Re: [PATCH net-next v11 09/23] ovpn: implement basic RX path (UDP)

From: Antonio Quartulli
Date: Fri Nov 15 2024 - 10:05:08 EST


On 12/11/2024 01:16, Sergey Ryazanov wrote:
On 29.10.2024 12:47, Antonio Quartulli wrote:
+static void ovpn_netdev_write(struct ovpn_peer *peer, struct sk_buff *skb)
+{
+    unsigned int pkt_len;
+
+    /* we can't guarantee the packet wasn't corrupted before entering the
+     * VPN, therefore we give other layers a chance to check that
+     */
+    skb->ip_summed = CHECKSUM_NONE;
+
+    /* skb hash for transport packet no longer valid after decapsulation */
+    skb_clear_hash(skb);
+
+    /* post-decrypt scrub -- prepare to inject encapsulated packet onto the
+     * interface, based on __skb_tunnel_rx() in dst.h
+     */
+    skb->dev = peer->ovpn->dev;
+    skb_set_queue_mapping(skb, 0);
+    skb_scrub_packet(skb, true);
+
+    skb_reset_network_header(skb);
+    skb_reset_transport_header(skb);
+    skb_probe_transport_header(skb);
+    skb_reset_inner_headers(skb);
+
+    memset(skb->cb, 0, sizeof(skb->cb));
+
+    /* cause packet to be "received" by the interface */
+    pkt_len = skb->len;
+    if (likely(gro_cells_receive(&peer->ovpn->gro_cells,
+                     skb) == NET_RX_SUCCESS))

nit: to improve readability, the packet delivery call can be composed like this:

      pkt_len = skb->len;
      res = gro_cells_receive(&peer->ovpn->gro_cells, skb);
      if (likely(res == NET_RX_SUCCESS))


hm, you don't like calls on two lines? :-)

ok, will change it.

Regards,

+        /* update RX stats with the size of decrypted packet */
+        dev_sw_netstats_rx_add(peer->ovpn->dev, pkt_len);
+}

--
Antonio Quartulli
OpenVPN Inc.