On Fri, 20 Feb 2026 18:45:21 -0800 [email protected] wrote: > + rte_rwlock_read_lock(&hv->vf_lock); > + vf_dev = hn_get_vf_dev(hv); > + if (hv->vf_ctx.vf_vsc_switched && vf_dev && > + vf_dev->data->dev_started) { > + void *sub_q = vf_dev->data->tx_queues[queue_id]; > + > + nb_tx = (*vf_dev->tx_pkt_burst) > + (sub_q, tx_pkts, nb_pkts); > rte_rwlock_read_unlock(&hv->vf_lock); > + return nb_tx; > } > + rte_rwlock_read_unlock(&hv->vf_lock);
ok, but read locks not free. It does incur an additional cache miss which may impact performance. Might be better in future to have some form of lockless access here.
