[dpdk-dev] [RFC PATCH 02/11] net/virtio: implement 1.1 guest Tx

[Jens Freimann <jfreiman@redhat.com>]

From: Yuanhan Liu <yuanhan.liu@linux.intel.com>

build only so far

Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
---
 drivers/net/virtio/Makefile          |   1 +
 drivers/net/virtio/virtio_ethdev.c   |  24 ++++--
 drivers/net/virtio/virtio_ethdev.h   |   3 +
 drivers/net/virtio/virtio_rxtx.c     |   3 +
 drivers/net/virtio/virtio_rxtx_1.1.c | 159 +++++++++++++++++++++++++++++++++++
 5 files changed, 183 insertions(+), 7 deletions(-)
 create mode 100644 drivers/net/virtio/virtio_rxtx_1.1.c

diff --git a/drivers/net/virtio/Makefile b/drivers/net/virtio/Makefile
index b21b878..4c4ff42 100644
--- a/drivers/net/virtio/Makefile
+++ b/drivers/net/virtio/Makefile
@@ -49,6 +49,7 @@ LIBABIVER := 1
 SRCS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtqueue.c
 SRCS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtio_pci.c
 SRCS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtio_rxtx.c
+SRCS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtio_rxtx_1.1.c
 SRCS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtio_ethdev.c
 SRCS-$(CONFIG_RTE_LIBRTE_VIRTIO_PMD) += virtio_rxtx_simple.c
 
diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c
index a5db503..553c27a 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -334,12 +334,12 @@ struct rte_virtio_xstats_name_off {
 		vq->vq_desc_tail_idx = (uint16_t)(vq->vq_nentries - 1);
 
 		vring_desc_init(vr->desc, size);
-	}
 
-	/*
-	 * Disable device(host) interrupting guest
-	 */
-	virtqueue_disable_intr(vq);
+		/*
+		 * Disable device(host) interrupting guest
+		 */
+		virtqueue_disable_intr(vq);
+	}
 }
 
 static int
@@ -625,7 +625,8 @@ struct rte_virtio_xstats_name_off {
 	}
 
 	vtpci_reset(hw);
-	virtio_dev_free_mbufs(dev);
+	if (!vtpci_version_1_1(hw))
+		virtio_dev_free_mbufs(dev);
 	virtio_free_queues(hw);
 }
 
@@ -1534,7 +1535,6 @@ static int virtio_dev_xstats_get_names(struct rte_eth_dev *dev,
 	RTE_BUILD_BUG_ON(RTE_PKTMBUF_HEADROOM < sizeof(struct virtio_net_hdr_mrg_rxbuf));
 
 	eth_dev->dev_ops = &virtio_eth_dev_ops;
-	eth_dev->tx_pkt_burst = &virtio_xmit_pkts;
 
 	if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
 		if (!hw->virtio_user_dev) {
@@ -1578,6 +1578,12 @@ static int virtio_dev_xstats_get_names(struct rte_eth_dev *dev,
 	if (ret < 0)
 		return ret;
 
+	/* FIXME: as second process? */
+	if (vtpci_version_1_1(hw))
+		eth_dev->tx_pkt_burst = &virtio_xmit_pkts_1_1;
+	else
+		eth_dev->tx_pkt_burst = &virtio_xmit_pkts;
+
 	/* Setup interrupt callback  */
 	if (eth_dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
 		rte_intr_callback_register(eth_dev->intr_handle,
@@ -1749,6 +1755,10 @@ static int eth_virtio_pci_remove(struct rte_pci_device *pci_dev)
 		}
 	}
 
+	/*no rx support for virtio 1.1 yet*/
+	if (vtpci_version_1_1(hw))
+		return 0;
+
 	/*Notify the backend
 	 *Otherwise the tap backend might already stop its queue due to fullness.
 	 *vhost backend will have no chance to be waked up
diff --git a/drivers/net/virtio/virtio_ethdev.h b/drivers/net/virtio/virtio_ethdev.h
index c3413c6..deed34e 100644
--- a/drivers/net/virtio/virtio_ethdev.h
+++ b/drivers/net/virtio/virtio_ethdev.h
@@ -69,6 +69,7 @@
 	 1u << VIRTIO_NET_F_MTU	| \
 	 1u << VIRTIO_RING_F_INDIRECT_DESC |    \
 	 1ULL << VIRTIO_F_VERSION_1       |	\
+	 1ULL << VIRTIO_F_VERSION_1_1     |	\
 	 1ULL << VIRTIO_F_IOMMU_PLATFORM)
 
 #define VIRTIO_PMD_SUPPORTED_GUEST_FEATURES	\
@@ -104,6 +105,8 @@ uint16_t virtio_recv_mergeable_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
 
 uint16_t virtio_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
 		uint16_t nb_pkts);
+uint16_t virtio_xmit_pkts_1_1(void *tx_queue, struct rte_mbuf **tx_pkts,
+		uint16_t nb_pkts);
 
 uint16_t virtio_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 		uint16_t nb_pkts);
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index fbc96df..e697192 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -427,6 +427,9 @@
 
 	PMD_INIT_FUNC_TRACE();
 
+	if (vtpci_version_1_1(hw))
+		return 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);
diff --git a/drivers/net/virtio/virtio_rxtx_1.1.c b/drivers/net/virtio/virtio_rxtx_1.1.c
new file mode 100644
index 0000000..e47a346
--- /dev/null
+++ b/drivers/net/virtio/virtio_rxtx_1.1.c
@@ -0,0 +1,159 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2010-2016 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+
+#include <rte_cycles.h>
+#include <rte_memory.h>
+#include <rte_memzone.h>
+#include <rte_branch_prediction.h>
+#include <rte_mempool.h>
+#include <rte_malloc.h>
+#include <rte_mbuf.h>
+#include <rte_ether.h>
+#include <rte_ethdev.h>
+#include <rte_prefetch.h>
+#include <rte_string_fns.h>
+#include <rte_errno.h>
+#include <rte_byteorder.h>
+#include <rte_cpuflags.h>
+#include <rte_net.h>
+#include <rte_ip.h>
+#include <rte_udp.h>
+#include <rte_tcp.h>
+
+#include "virtio_logs.h"
+#include "virtio_ethdev.h"
+#include "virtio_pci.h"
+#include "virtqueue.h"
+#include "virtio_rxtx.h"
+
+/* Cleanup from completed transmits. */
+static void
+virtio_xmit_cleanup(struct virtqueue *vq)
+{
+	uint16_t idx;
+	uint16_t size = vq->vq_nentries;
+	struct vring_desc_1_1 *desc = vq->vq_ring.desc_1_1;
+
+	idx = vq->vq_used_cons_idx & (size - 1);
+	while ((desc[idx].flags & DESC_HW) == 0) {
+		struct vq_desc_extra *dxp;
+
+		dxp = &vq->vq_descx[idx];
+		if (dxp->cookie != NULL) {
+			rte_pktmbuf_free(dxp->cookie);
+			dxp->cookie = NULL;
+		}
+
+		idx = (++vq->vq_used_cons_idx) & (size - 1);
+		vq->vq_free_cnt++;
+
+		if (vq->vq_free_cnt >= size)
+			break;
+	}
+}
+
+static inline void
+virtio_xmit(struct virtnet_tx *txvq, struct rte_mbuf *mbuf)
+{
+	struct virtio_tx_region *txr = txvq->virtio_net_hdr_mz->addr;
+	struct virtqueue *vq = txvq->vq;
+	struct vring_desc_1_1 *desc = vq->vq_ring.desc_1_1;
+	uint16_t idx;
+	uint16_t head_idx = (vq->vq_avail_idx++) & (vq->vq_nentries - 1);
+
+	idx = head_idx;
+	vq->vq_free_cnt -= mbuf->nb_segs + 1;
+	vq->vq_descx[idx].cookie = mbuf;
+
+	desc[idx].addr  = txvq->virtio_net_hdr_mem +
+			  RTE_PTR_DIFF(&txr[idx].tx_hdr, txr);
+	desc[idx].len   = vq->hw->vtnet_hdr_size;
+	desc[idx].flags = VRING_DESC_F_NEXT;
+
+	do {
+		idx = (vq->vq_avail_idx++) & (vq->vq_nentries - 1);
+		desc[idx].addr  = VIRTIO_MBUF_DATA_DMA_ADDR(mbuf, vq);
+		desc[idx].len   = mbuf->data_len;
+		desc[idx].flags = DESC_HW | VRING_DESC_F_NEXT;
+	} while ((mbuf = mbuf->next) != NULL);
+
+	desc[idx].flags &= ~VRING_DESC_F_NEXT;
+
+	/*
+	 * update the head last, so that when the host saw such flag
+	 * is set, it means all others in the same chain is also set
+	 */
+	rte_smp_wmb();
+	desc[head_idx].flags |= DESC_HW;
+}
+
+uint16_t
+virtio_xmit_pkts_1_1(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
+{
+	struct virtnet_tx *txvq = tx_queue;
+	struct virtqueue *vq = txvq->vq;
+	uint16_t i;
+
+	if (unlikely(nb_pkts < 1))
+		return nb_pkts;
+
+	PMD_TX_LOG(DEBUG, "%d packets to xmit", nb_pkts);
+
+	for (i = 0; i < nb_pkts; i++) {
+		struct rte_mbuf *txm = tx_pkts[i];
+
+		if (unlikely(txm->nb_segs + 1 > vq->vq_free_cnt)) {
+			virtio_xmit_cleanup(vq);
+
+			if (unlikely(txm->nb_segs + 1 > vq->vq_free_cnt)) {
+				PMD_TX_LOG(ERR,
+					   "No free tx descriptors to transmit");
+				break;
+			}
+		}
+
+		virtio_xmit(txvq, txm);
+		txvq->stats.bytes += txm->pkt_len;
+	}
+
+	txvq->stats.packets += i;
+	txvq->stats.errors  += nb_pkts - i;
+
+	return i;
+}
-- 
1.8.3.1