On 1 Jul 2026, at 15:58, Eli Britstein wrote:
> A netdev_dpdk object can be an ethernet dpdk or a vhost-user port.
> The dev_type field was used to determine that.
> Instead, introduce netdev_dpdk_is_vhost() and remove the dev_type field.
>
> Signed-off-by: Eli Britstein <[email protected]>
Hi Eli,
Thanks for cleaning this up based on an earlier comment.
However, I found one issue; see below.
In addition, should we maybe change the call to be the
opposite, i.e. is_eth()? If we add any more non-eth devices
in the future, would it be more likely that they also have
the same limitations? Hence we end up with multiple
device-specific APIs rather than one general? David/Kevin,
any insights or thoughts on this?
Cheers,
Eelco
> ---
> lib/netdev-dpdk.c | 37 ++++++++++++++++---------------------
> 1 file changed, 16 insertions(+), 21 deletions(-)
>
> diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
> index e375275de..9f8c72bc8 100644
> --- a/lib/netdev-dpdk.c
> +++ b/lib/netdev-dpdk.c
> @@ -210,11 +210,6 @@ struct netdev_dpdk_sw_stats {
> uint64_t tx_invalid_hwol_drops;
> };
>
> -enum dpdk_dev_type {
> - DPDK_DEV_ETH = 0,
> - DPDK_DEV_VHOST = 1,
> -};
> -
> /* Quality of Service */
>
> /* An instance of a QoS configuration. Always associated with a particular
> @@ -467,7 +462,6 @@ struct netdev_dpdk {
> int socket_id;
> int buf_size;
> int max_packet_len;
> - enum dpdk_dev_type type;
> enum netdev_flags flags;
> int link_reset_cnt;
> union {
> @@ -616,6 +610,12 @@ is_dpdk_class(const struct netdev_class *class)
> || class->destruct == netdev_dpdk_vhost_destruct;
> }
>
> +static bool
> +netdev_dpdk_is_vhost(const struct netdev *netdev)
> +{
> + return netdev->netdev_class->destruct == netdev_dpdk_vhost_destruct;
> +}
> +
> /* DPDK NIC drivers allocate RX buffers at a particular granularity,
> typically
> * aligned at 1k or less. If a declared mbuf size is not a multiple of this
> * value, insufficient buffers are allocated to accomodate the packet in its
> @@ -1046,7 +1046,7 @@ dpdk_watchdog(void *dummy OVS_UNUSED)
> ovs_mutex_lock(&dpdk_mutex);
> LIST_FOR_EACH (dev, list_node, &dpdk_list) {
> ovs_mutex_lock(&dev->mutex);
> - if (dev->type == DPDK_DEV_ETH) {
> + if (!netdev_dpdk_is_vhost(&dev->up)) {
> check_link_status(dev);
> }
> ovs_mutex_unlock(&dev->mutex);
> @@ -1506,8 +1506,7 @@ netdev_dpdk_alloc_txq(unsigned int n_txqs)
> }
>
> static int
> -common_construct(struct netdev *netdev, dpdk_port_t port_no,
> - enum dpdk_dev_type type, int socket_id)
> +common_construct(struct netdev *netdev, dpdk_port_t port_no, int socket_id)
> OVS_REQUIRES(dpdk_mutex)
> {
> struct netdev_dpdk *dev = netdev_dpdk_cast(netdev);
> @@ -1522,7 +1521,6 @@ common_construct(struct netdev *netdev, dpdk_port_t
> port_no,
> dev->socket_id = socket_id < 0 ? SOCKET0 : socket_id;
> dev->requested_socket_id = dev->socket_id;
> dev->port_id = port_no;
> - dev->type = type;
> dev->flags = 0;
> dev->requested_mtu = RTE_ETHER_MTU;
> dev->max_packet_len = MTU_TO_FRAME_LEN(dev->mtu);
> @@ -1572,7 +1570,7 @@ common_construct(struct netdev *netdev, dpdk_port_t
> port_no,
> dev->rte_xstats_ids_size = 0;
>
> dev->sw_stats = xzalloc(sizeof *dev->sw_stats);
> - dev->sw_stats->tx_retries = (dev->type == DPDK_DEV_VHOST) ? 0 :
> UINT64_MAX;
> + dev->sw_stats->tx_retries = netdev_dpdk_is_vhost(netdev) ? 0 :
> UINT64_MAX;
>
> return 0;
> }
> @@ -1599,8 +1597,7 @@ vhost_common_construct(struct netdev *netdev)
>
> dev->vhost_max_queue_pairs = VHOST_MAX_QUEUE_PAIRS_DEF;
>
> - return common_construct(netdev, DPDK_ETH_PORT_ID_INVALID,
> - DPDK_DEV_VHOST, socket_id);
> + return common_construct(netdev, DPDK_ETH_PORT_ID_INVALID, socket_id);
> }
>
> static int
> @@ -1707,8 +1704,7 @@ netdev_dpdk_construct(struct netdev *netdev)
> int err;
>
> ovs_mutex_lock(&dpdk_mutex);
> - err = common_construct(netdev, DPDK_ETH_PORT_ID_INVALID,
> - DPDK_DEV_ETH, SOCKET0);
> + err = common_construct(netdev, DPDK_ETH_PORT_ID_INVALID, SOCKET0);
> ovs_mutex_unlock(&dpdk_mutex);
> return err;
> }
> @@ -2273,7 +2269,7 @@ dpdk_set_rx_steer_config(struct netdev *netdev, struct
> netdev_dpdk *dev,
> netdev_get_name(netdev), arg);
> }
>
> - if (flags && dev->type != DPDK_DEV_ETH) {
> + if (flags && !netdev_dpdk_is_vhost(netdev)) {
This is wrong, it should be;
if (flags && netdev_dpdk_is_vhost(netdev)) {
> VLOG_WARN("%s: options:rx-steering "
> "is only supported on ethernet ports",
> netdev_get_name(netdev));
> @@ -3443,7 +3439,7 @@ netdev_dpdk_set_etheraddr__(struct netdev_dpdk *dev,
> const struct eth_addr mac)
> {
> int err = 0;
>
> - if (dev->type == DPDK_DEV_ETH) {
> + if (!netdev_dpdk_is_vhost(&dev->up)) {
> struct rte_ether_addr ea;
>
> memcpy(ea.addr_bytes, mac.ea, ETH_ADDR_LEN);
> @@ -4363,8 +4359,7 @@ netdev_dpdk_update_flags__(struct netdev_dpdk *dev,
> return 0;
> }
>
> - if (dev->type == DPDK_DEV_ETH) {
> -
> + if (!netdev_dpdk_is_vhost(&dev->up)) {
> if ((dev->flags ^ *old_flagsp) & NETDEV_UP) {
> int err;
>
> @@ -4390,7 +4385,7 @@ netdev_dpdk_update_flags__(struct netdev_dpdk *dev,
>
> netdev_change_seq_changed(&dev->up);
> } else {
> - /* If DPDK_DEV_VHOST device's NETDEV_UP flag was changed and vhost is
> + /* If a vhost device's NETDEV_UP flag was changed and vhost is
> * running then change netdev's change_seq to trigger link state
> * update. */
>
> @@ -6558,7 +6553,7 @@ netdev_dpdk_flow_api_supported(struct netdev *netdev,
> bool check_only)
>
> dev = netdev_dpdk_cast(netdev);
> ovs_mutex_lock(&dev->mutex);
> - if (dev->type == DPDK_DEV_ETH) {
> + if (!netdev_dpdk_is_vhost(netdev)) {
> if (dev->requested_rx_steer_flags && !check_only) {
> VLOG_WARN("%s: rx-steering is mutually exclusive with
> hw-offload,"
> " falling back to default rss mode",
> --
> 2.43.0
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
