From: Ashish Jain Date: Fri, 14 Sep 2018 11:48:09 +0000 (+0530) Subject: crypto/dpaa2_sec: support atomic queues X-Git-Url: http://git.droids-corp.org/?a=commitdiff_plain;h=a77db24643b7dc5ee053924f834256d4400bc709;p=dpdk.git crypto/dpaa2_sec: support atomic queues Queues attached to event crypto device can be parallel, atomic and ordered. This patch add support for atomic queues processing for dpaa2_sec queues. Signed-off-by: Ashish Jain Acked-by: Akhil Goyal --- diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c index e79c243c0e..24e7ebc7a6 100644 --- a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c +++ b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c @@ -1210,6 +1210,7 @@ dpaa2_sec_enqueue_burst(void *qp, struct rte_crypto_op **ops, struct dpaa2_sec_qp *dpaa2_qp = (struct dpaa2_sec_qp *)qp; struct qbman_swp *swp; uint16_t num_tx = 0; + uint32_t flags[MAX_TX_RING_SLOTS] = {0}; /*todo - need to support multiple buffer pools */ uint16_t bpid; struct rte_mempool *mb_pool; @@ -1241,6 +1242,15 @@ dpaa2_sec_enqueue_burst(void *qp, struct rte_crypto_op **ops, dpaa2_eqcr_size : nb_ops; for (loop = 0; loop < frames_to_send; loop++) { + if ((*ops)->sym->m_src->seqn) { + uint8_t dqrr_index = (*ops)->sym->m_src->seqn - 1; + + flags[loop] = QBMAN_ENQUEUE_FLAG_DCA | dqrr_index; + DPAA2_PER_LCORE_DQRR_SIZE--; + DPAA2_PER_LCORE_DQRR_HELD &= ~(1 << dqrr_index); + (*ops)->sym->m_src->seqn = DPAA2_INVALID_MBUF_SEQN; + } + /*Clear the unused FD fields before sending*/ memset(&fd_arr[loop], 0, sizeof(struct qbman_fd)); mb_pool = (*ops)->sym->m_src->pool; @@ -1257,7 +1267,7 @@ dpaa2_sec_enqueue_burst(void *qp, struct rte_crypto_op **ops, while (loop < frames_to_send) { loop += qbman_swp_enqueue_multiple(swp, &eqdesc, &fd_arr[loop], - NULL, + &flags[loop], frames_to_send - loop); } @@ -1282,6 +1292,9 @@ sec_simple_fd_to_mbuf(const struct qbman_fd *fd, __rte_unused uint8_t id) DPAA2_IOVA_TO_VADDR(DPAA2_GET_FD_ADDR(fd)), rte_dpaa2_bpid_info[DPAA2_GET_FD_BPID(fd)].meta_data_size); + diff = len - mbuf->pkt_len; + mbuf->pkt_len += diff; + mbuf->data_len += diff; op = (struct rte_crypto_op *)(size_t)mbuf->buf_iova; mbuf->buf_iova = op->sym->aead.digest.phys_addr; op->sym->aead.digest.phys_addr = 0L; @@ -1292,9 +1305,6 @@ sec_simple_fd_to_mbuf(const struct qbman_fd *fd, __rte_unused uint8_t id) mbuf->data_off += SEC_FLC_DHR_OUTBOUND; else mbuf->data_off += SEC_FLC_DHR_INBOUND; - diff = len - mbuf->pkt_len; - mbuf->pkt_len += diff; - mbuf->data_len += diff; return op; } @@ -2880,6 +2890,38 @@ dpaa2_sec_process_parallel_event(struct qbman_swp *swp, qbman_swp_dqrr_consume(swp, dq); } +static void +dpaa2_sec_process_atomic_event(struct qbman_swp *swp __attribute__((unused)), + const struct qbman_fd *fd, + const struct qbman_result *dq, + struct dpaa2_queue *rxq, + struct rte_event *ev) +{ + uint8_t dqrr_index; + struct rte_crypto_op *crypto_op = (struct rte_crypto_op *)ev->event_ptr; + /* Prefetching mbuf */ + rte_prefetch0((void *)(size_t)(DPAA2_GET_FD_ADDR(fd)- + rte_dpaa2_bpid_info[DPAA2_GET_FD_BPID(fd)].meta_data_size)); + + /* Prefetching ipsec crypto_op stored in priv data of mbuf */ + rte_prefetch0((void *)(size_t)(DPAA2_GET_FD_ADDR(fd)-64)); + + ev->flow_id = rxq->ev.flow_id; + ev->sub_event_type = rxq->ev.sub_event_type; + ev->event_type = RTE_EVENT_TYPE_CRYPTODEV; + ev->op = RTE_EVENT_OP_NEW; + ev->sched_type = rxq->ev.sched_type; + ev->queue_id = rxq->ev.queue_id; + ev->priority = rxq->ev.priority; + + ev->event_ptr = sec_fd_to_mbuf(fd, ((struct rte_cryptodev *) + (rxq->dev))->driver_id); + dqrr_index = qbman_get_dqrr_idx(dq); + crypto_op->sym->m_src->seqn = dqrr_index + 1; + DPAA2_PER_LCORE_DQRR_SIZE++; + DPAA2_PER_LCORE_DQRR_HELD |= 1 << dqrr_index; + DPAA2_PER_LCORE_DQRR_MBUF(dqrr_index) = crypto_op->sym->m_src; +} int dpaa2_sec_eventq_attach(const struct rte_cryptodev *dev, @@ -2895,6 +2937,8 @@ dpaa2_sec_eventq_attach(const struct rte_cryptodev *dev, if (event->sched_type == RTE_SCHED_TYPE_PARALLEL) qp->rx_vq.cb = dpaa2_sec_process_parallel_event; + else if (event->sched_type == RTE_SCHED_TYPE_ATOMIC) + qp->rx_vq.cb = dpaa2_sec_process_atomic_event; else return -EINVAL; @@ -2906,7 +2950,10 @@ dpaa2_sec_eventq_attach(const struct rte_cryptodev *dev, cfg.options |= DPSECI_QUEUE_OPT_USER_CTX; cfg.user_ctx = (size_t)(qp); - + if (event->sched_type == RTE_SCHED_TYPE_ATOMIC) { + cfg.options |= DPSECI_QUEUE_OPT_ORDER_PRESERVATION; + cfg.order_preservation_en = 1; + } ret = dpseci_set_rx_queue(dpseci, CMD_PRI_LOW, priv->token, qp_id, &cfg); if (ret) {