Re: [PATCH v20 12/25] net/pcap: support VLAN strip and insert offloads

[Stephen Hemminger]

On Mon, 16 Mar 2026 14:04:59 +0000
Bruce Richardson <bruce.richard...@intel.com> wrote:

> Are we sure this is the correct way to do this? To behave like a normal NIC
> the Tx VLAN tag should be inserted in the Tx function, rather than tx
> prepare should it not? There is no guarantee apps will call tx_prepare
> before Tx.
> 
> /Bruce

I wasn't sure, but copied what virtio PMD was doing.
If application is using VLAN's on virtio it would need to call prepare.

uint16_t
virtio_xmit_pkts_prepare(void *tx_queue __rte_unused, struct rte_mbuf **tx_pkts,
                        uint16_t nb_pkts)
{
        uint16_t nb_tx;
        int error;

        for (nb_tx = 0; nb_tx < nb_pkts; nb_tx++) {
                struct rte_mbuf *m = tx_pkts[nb_tx];

...
                /* Do VLAN tag insertion */
                if (unlikely(m->ol_flags & RTE_MBUF_F_TX_VLAN)) {
                        error = rte_vlan_insert(&m);
                        /* rte_vlan_insert() may change pointer
                         * even in the case of failure
                         */
                        tx_pkts[nb_tx] = m;

                        if (unlikely(error)) {
                                rte_errno = -error;
                                break;
                        }
                }