netdev-dpdk: Stop relying on vhost-user Tx flags.

vhost-user legacy behavior has been to mark mbuf with Tx offload flags based on what the virtio-net header contained. OVS current implementation was relying on this behavior though this clashed in some corner cases, like when a guest sends a tunneled packet requesting outer udp checksum offload. Signed-off-by: David Marchand <[email protected]>
david-marchand · Feb 13, 2025 · 437ae00 · 437ae00
1 parent 1f64f79
commit 437ae00
Showing 1 changed file with 5 additions and 0 deletions.
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
@@ -6387,6 +6387,11 @@ netdev_dpdk_vhost_client_reconfigure(struct netdev *netdev)
             vhost_flags |= RTE_VHOST_USER_POSTCOPY_SUPPORT;
         }
 
+        /* Prefer that the vhost library provides Rx ol_flags, rather than let
+         * it mark the packet with Tx ol_flags (which may collide with OVS Tx
+         * offloads decisions). */
+        vhost_flags |= RTE_VHOST_USER_NET_COMPLIANT_OL_FLAGS;
+
         /* Enable External Buffers if TCP Segmentation Offload is enabled. */
         if (enable_tso) {
             vhost_flags |= RTE_VHOST_USER_EXTBUF_SUPPORT;