From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by dpdk.org (Postfix) with ESMTP id E1F4C1B430 for ; Mon, 2 Jul 2018 17:26:03 +0200 (CEST) Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 887E881663C3; Mon, 2 Jul 2018 15:26:03 +0000 (UTC) Received: from localhost.localdomain (ovpn-112-41.ams2.redhat.com [10.36.112.41]) by smtp.corp.redhat.com (Postfix) with ESMTP id 43CD6111AF34; Mon, 2 Jul 2018 15:26:02 +0000 (UTC) From: Maxime Coquelin To: tiwei.bie@intel.com, zhihong.wang@intel.com, dev@dpdk.org Cc: yong.liu@intel.com, Maxime Coquelin Date: Mon, 2 Jul 2018 17:25:44 +0200 Message-Id: <20180702152547.31423-2-maxime.coquelin@redhat.com> In-Reply-To: <20180702152547.31423-1-maxime.coquelin@redhat.com> References: <20180702152547.31423-1-maxime.coquelin@redhat.com> X-Scanned-By: MIMEDefang 2.78 on 10.11.54.3 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.8]); Mon, 02 Jul 2018 15:26:03 +0000 (UTC) X-Greylist: inspected by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.8]); Mon, 02 Jul 2018 15:26:03 +0000 (UTC) for IP:'10.11.54.3' DOMAIN:'int-mx03.intmail.prod.int.rdu2.redhat.com' HELO:'smtp.corp.redhat.com' FROM:'maxime.coquelin@redhat.com' RCPT:'' Subject: [dpdk-dev] [PATCH v4 1/4] net/virtio: remove simple Tx path X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 02 Jul 2018 15:26:04 -0000 The simple Tx path does not comply with the Virtio specification. Now that VIRTIO_F_IN_ORDER feature is supported by the Virtio PMD, let's use this optimized path instead. Signed-off-by: Maxime Coquelin --- drivers/net/virtio/virtio_ethdev.c | 11 +----- drivers/net/virtio/virtio_pci.h | 1 - drivers/net/virtio/virtio_rxtx.c | 28 +------------- drivers/net/virtio/virtio_rxtx_simple.c | 67 --------------------------------- drivers/net/virtio/virtio_rxtx_simple.h | 49 ------------------------ drivers/net/virtio/virtio_user_ethdev.c | 1 - 6 files changed, 2 insertions(+), 155 deletions(-) diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c index 02980e3d1..73c428734 100644 --- a/drivers/net/virtio/virtio_ethdev.c +++ b/drivers/net/virtio/virtio_ethdev.c @@ -1336,11 +1336,7 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev) eth_dev->rx_pkt_burst = &virtio_recv_pkts; } - if (hw->use_simple_tx) { - PMD_INIT_LOG(INFO, "virtio: using simple Tx path on port %u", - eth_dev->data->port_id); - eth_dev->tx_pkt_burst = virtio_xmit_pkts_simple; - } else if (hw->use_inorder_tx) { + if (hw->use_inorder_tx) { PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on port %u", eth_dev->data->port_id); eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder; @@ -1881,12 +1877,9 @@ virtio_dev_configure(struct rte_eth_dev *dev) rte_spinlock_init(&hw->state_lock); hw->use_simple_rx = 1; - hw->use_simple_tx = 1; if (vtpci_with_feature(hw, VIRTIO_F_IN_ORDER)) { - /* Simple Tx not compatible with in-order ring */ hw->use_inorder_tx = 1; - hw->use_simple_tx = 0; if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) { hw->use_inorder_rx = 1; hw->use_simple_rx = 0; @@ -1898,12 +1891,10 @@ virtio_dev_configure(struct rte_eth_dev *dev) #if defined RTE_ARCH_ARM64 || defined RTE_ARCH_ARM if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON)) { hw->use_simple_rx = 0; - hw->use_simple_tx = 0; } #endif if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) { hw->use_simple_rx = 0; - hw->use_simple_tx = 0; } if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM | diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h index 77f805df6..0ec216874 100644 --- a/drivers/net/virtio/virtio_pci.h +++ b/drivers/net/virtio/virtio_pci.h @@ -238,7 +238,6 @@ struct virtio_hw { uint8_t use_msix; uint8_t modern; uint8_t use_simple_rx; - uint8_t use_simple_tx; uint8_t use_inorder_rx; uint8_t use_inorder_tx; uint16_t port_id; diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c index 6394071b8..522e5dd41 100644 --- a/drivers/net/virtio/virtio_rxtx.c +++ b/drivers/net/virtio/virtio_rxtx.c @@ -708,10 +708,6 @@ virtio_dev_tx_queue_setup(struct rte_eth_dev *dev, PMD_INIT_FUNC_TRACE(); - /* cannot use simple rxtx funcs with multisegs or offloads */ - if (dev->data->dev_conf.txmode.offloads) - hw->use_simple_tx = 0; - if (nb_desc == 0 || nb_desc > vq->vq_nentries) nb_desc = vq->vq_nentries; vq->vq_free_cnt = RTE_MIN(vq->vq_free_cnt, nb_desc); @@ -746,33 +742,11 @@ virtio_dev_tx_queue_setup_finish(struct rte_eth_dev *dev, uint8_t vtpci_queue_idx = 2 * queue_idx + VTNET_SQ_TQ_QUEUE_IDX; struct virtio_hw *hw = dev->data->dev_private; struct virtqueue *vq = hw->vqs[vtpci_queue_idx]; - uint16_t mid_idx = vq->vq_nentries >> 1; - struct virtnet_tx *txvq = &vq->txq; - uint16_t desc_idx; PMD_INIT_FUNC_TRACE(); - if (hw->use_simple_tx) { - for (desc_idx = 0; desc_idx < mid_idx; desc_idx++) { - vq->vq_ring.avail->ring[desc_idx] = - desc_idx + mid_idx; - vq->vq_ring.desc[desc_idx + mid_idx].next = - desc_idx; - vq->vq_ring.desc[desc_idx + mid_idx].addr = - txvq->virtio_net_hdr_mem + - offsetof(struct virtio_tx_region, tx_hdr); - vq->vq_ring.desc[desc_idx + mid_idx].len = - vq->hw->vtnet_hdr_size; - vq->vq_ring.desc[desc_idx + mid_idx].flags = - VRING_DESC_F_NEXT; - vq->vq_ring.desc[desc_idx].flags = 0; - } - for (desc_idx = mid_idx; desc_idx < vq->vq_nentries; - desc_idx++) - vq->vq_ring.avail->ring[desc_idx] = desc_idx; - } else if (hw->use_inorder_tx) { + if (hw->use_inorder_tx) vq->vq_ring.desc[vq->vq_nentries - 1].next = 0; - } VIRTQUEUE_DUMP(vq); diff --git a/drivers/net/virtio/virtio_rxtx_simple.c b/drivers/net/virtio/virtio_rxtx_simple.c index 515207581..31e565b4c 100644 --- a/drivers/net/virtio/virtio_rxtx_simple.c +++ b/drivers/net/virtio/virtio_rxtx_simple.c @@ -27,73 +27,6 @@ #pragma GCC diagnostic ignored "-Wcast-qual" #endif -uint16_t -virtio_xmit_pkts_simple(void *tx_queue, struct rte_mbuf **tx_pkts, - uint16_t nb_pkts) -{ - struct virtnet_tx *txvq = tx_queue; - struct virtqueue *vq = txvq->vq; - struct virtio_hw *hw = vq->hw; - uint16_t nb_used; - uint16_t desc_idx; - struct vring_desc *start_dp; - uint16_t nb_tail, nb_commit; - int i; - uint16_t desc_idx_max = (vq->vq_nentries >> 1) - 1; - uint16_t nb_tx = 0; - - if (unlikely(hw->started == 0 && tx_pkts != hw->inject_pkts)) - return nb_tx; - - nb_used = VIRTQUEUE_NUSED(vq); - rte_compiler_barrier(); - - if (nb_used >= VIRTIO_TX_FREE_THRESH) - virtio_xmit_cleanup_simple(vq); - - nb_commit = nb_pkts = RTE_MIN((vq->vq_free_cnt >> 1), nb_pkts); - desc_idx = (uint16_t)(vq->vq_avail_idx & desc_idx_max); - start_dp = vq->vq_ring.desc; - nb_tail = (uint16_t) (desc_idx_max + 1 - desc_idx); - - if (nb_commit >= nb_tail) { - for (i = 0; i < nb_tail; i++) - vq->vq_descx[desc_idx + i].cookie = tx_pkts[i]; - for (i = 0; i < nb_tail; i++) { - start_dp[desc_idx].addr = - VIRTIO_MBUF_DATA_DMA_ADDR(*tx_pkts, vq); - start_dp[desc_idx].len = (*tx_pkts)->pkt_len; - tx_pkts++; - desc_idx++; - } - nb_commit -= nb_tail; - desc_idx = 0; - } - for (i = 0; i < nb_commit; i++) - vq->vq_descx[desc_idx + i].cookie = tx_pkts[i]; - for (i = 0; i < nb_commit; i++) { - start_dp[desc_idx].addr = - VIRTIO_MBUF_DATA_DMA_ADDR(*tx_pkts, vq); - start_dp[desc_idx].len = (*tx_pkts)->pkt_len; - tx_pkts++; - desc_idx++; - } - - rte_compiler_barrier(); - - vq->vq_free_cnt -= (uint16_t)(nb_pkts << 1); - vq->vq_avail_idx += nb_pkts; - vq->vq_ring.avail->idx = vq->vq_avail_idx; - txvq->stats.packets += nb_pkts; - - if (likely(nb_pkts)) { - if (unlikely(virtqueue_kick_prepare(vq))) - virtqueue_notify(vq); - } - - return nb_pkts; -} - int __attribute__((cold)) virtio_rxq_vec_setup(struct virtnet_rx *rxq) { diff --git a/drivers/net/virtio/virtio_rxtx_simple.h b/drivers/net/virtio/virtio_rxtx_simple.h index 303904d64..dc97e4ccf 100644 --- a/drivers/net/virtio/virtio_rxtx_simple.h +++ b/drivers/net/virtio/virtio_rxtx_simple.h @@ -55,53 +55,4 @@ virtio_rxq_rearm_vec(struct virtnet_rx *rxvq) vq_update_avail_idx(vq); } -#define VIRTIO_TX_FREE_THRESH 32 -#define VIRTIO_TX_MAX_FREE_BUF_SZ 32 -#define VIRTIO_TX_FREE_NR 32 -/* TODO: vq->tx_free_cnt could mean num of free slots so we could avoid shift */ -static inline void -virtio_xmit_cleanup_simple(struct virtqueue *vq) -{ - uint16_t i, desc_idx; - uint32_t nb_free = 0; - struct rte_mbuf *m, *free[VIRTIO_TX_MAX_FREE_BUF_SZ]; - - desc_idx = (uint16_t)(vq->vq_used_cons_idx & - ((vq->vq_nentries >> 1) - 1)); - m = (struct rte_mbuf *)vq->vq_descx[desc_idx++].cookie; - m = rte_pktmbuf_prefree_seg(m); - if (likely(m != NULL)) { - free[0] = m; - nb_free = 1; - for (i = 1; i < VIRTIO_TX_FREE_NR; i++) { - m = (struct rte_mbuf *)vq->vq_descx[desc_idx++].cookie; - m = rte_pktmbuf_prefree_seg(m); - if (likely(m != NULL)) { - if (likely(m->pool == free[0]->pool)) - free[nb_free++] = m; - else { - rte_mempool_put_bulk(free[0]->pool, - (void **)free, - RTE_MIN(RTE_DIM(free), - nb_free)); - free[0] = m; - nb_free = 1; - } - } - } - rte_mempool_put_bulk(free[0]->pool, (void **)free, - RTE_MIN(RTE_DIM(free), nb_free)); - } else { - for (i = 1; i < VIRTIO_TX_FREE_NR; i++) { - m = (struct rte_mbuf *)vq->vq_descx[desc_idx++].cookie; - m = rte_pktmbuf_prefree_seg(m); - if (m != NULL) - rte_mempool_put(m->pool, m); - } - } - - vq->vq_used_cons_idx += VIRTIO_TX_FREE_NR; - vq->vq_free_cnt += (VIRTIO_TX_FREE_NR << 1); -} - #endif /* _VIRTIO_RXTX_SIMPLE_H_ */ diff --git a/drivers/net/virtio/virtio_user_ethdev.c b/drivers/net/virtio/virtio_user_ethdev.c index fcd30251f..5ee59d5e5 100644 --- a/drivers/net/virtio/virtio_user_ethdev.c +++ b/drivers/net/virtio/virtio_user_ethdev.c @@ -436,7 +436,6 @@ virtio_user_eth_dev_alloc(struct rte_vdev_device *vdev) hw->use_msix = 1; hw->modern = 0; hw->use_simple_rx = 0; - hw->use_simple_tx = 0; hw->use_inorder_rx = 0; hw->use_inorder_tx = 0; hw->virtio_user_dev = dev; -- 2.14.4