#include <errno.h>
#include <unistd.h>
-#include <rte_ethdev.h>
+#include <rte_ethdev_driver.h>
#include <rte_ethdev_pci.h>
#include <rte_memcpy.h>
#include <rte_string_fns.h>
#include <rte_pci.h>
#include <rte_bus_pci.h>
#include <rte_ether.h>
+#include <rte_ip.h>
+#include <rte_arp.h>
#include <rte_common.h>
#include <rte_errno.h>
#include <rte_cpuflags.h>
#include <rte_memory.h>
#include <rte_eal.h>
#include <rte_dev.h>
+#include <rte_cycles.h>
#include "virtio_ethdev.h"
#include "virtio_pci.h"
uint8_t stat_idx,
uint8_t is_rx);
+int virtio_logtype_init;
+int virtio_logtype_driver;
+
+static void virtio_notify_peers(struct rte_eth_dev *dev);
+static void virtio_ack_link_announce(struct rte_eth_dev *dev);
+
/*
* The set of PCI devices this driver supports
*/
PMD_INIT_LOG(ERR, "Control queue is not supported.");
return -1;
}
+
+ rte_spinlock_lock(&cvq->lock);
vq = cvq->vq;
head = vq->vq_desc_head_idx;
"vq->hw->cvq = %p vq = %p",
vq->vq_desc_head_idx, status, vq->hw->cvq, vq);
- if ((vq->vq_free_cnt < ((uint32_t)pkt_num + 2)) || (pkt_num < 1))
+ if (vq->vq_free_cnt < pkt_num + 2 || pkt_num < 1) {
+ rte_spinlock_unlock(&cvq->lock);
return -1;
+ }
memcpy(cvq->virtio_net_hdr_mz->addr, ctrl,
sizeof(struct virtio_pmd_ctrl));
result = cvq->virtio_net_hdr_mz->addr;
+ rte_spinlock_unlock(&cvq->lock);
return result->status;
}
/* do nothing */
}
-static int
-virtio_get_queue_type(struct virtio_hw *hw, uint16_t vtpci_queue_idx)
-{
- if (vtpci_queue_idx == hw->max_queue_pairs * 2)
- return VTNET_CQ;
- else if (vtpci_queue_idx % 2 == 0)
- return VTNET_RQ;
- else
- return VTNET_TQ;
-}
-
static uint16_t
virtio_get_nr_vq(struct virtio_hw *hw)
{
return 0;
}
+int
+virtio_dev_pause(struct rte_eth_dev *dev)
+{
+ struct virtio_hw *hw = dev->data->dev_private;
+
+ rte_spinlock_lock(&hw->state_lock);
+
+ if (hw->started == 0) {
+ /* Device is just stopped. */
+ rte_spinlock_unlock(&hw->state_lock);
+ return -1;
+ }
+ hw->started = 0;
+ /*
+ * Prevent the worker threads from touching queues to avoid contention,
+ * 1 ms should be enough for the ongoing Tx function to finish.
+ */
+ rte_delay_ms(1);
+ return 0;
+}
+
/*
- * Process Virtio Config changed interrupt and call the callback
- * if link state changed.
+ * Recover hw state to let the worker threads continue.
+ */
+void
+virtio_dev_resume(struct rte_eth_dev *dev)
+{
+ struct virtio_hw *hw = dev->data->dev_private;
+
+ hw->started = 1;
+ rte_spinlock_unlock(&hw->state_lock);
+}
+
+/*
+ * Should be called only after device is paused.
+ */
+int
+virtio_inject_pkts(struct rte_eth_dev *dev, struct rte_mbuf **tx_pkts,
+ int nb_pkts)
+{
+ struct virtio_hw *hw = dev->data->dev_private;
+ struct virtnet_tx *txvq = dev->data->tx_queues[0];
+ int ret;
+
+ hw->inject_pkts = tx_pkts;
+ ret = dev->tx_pkt_burst(txvq, tx_pkts, nb_pkts);
+ hw->inject_pkts = NULL;
+
+ return ret;
+}
+
+static void
+virtio_notify_peers(struct rte_eth_dev *dev)
+{
+ struct virtio_hw *hw = dev->data->dev_private;
+ struct virtnet_rx *rxvq = dev->data->rx_queues[0];
+ struct rte_mbuf *rarp_mbuf;
+
+ rarp_mbuf = rte_net_make_rarp_packet(rxvq->mpool,
+ (struct ether_addr *)hw->mac_addr);
+ if (rarp_mbuf == NULL) {
+ PMD_DRV_LOG(ERR, "failed to make RARP packet.");
+ return;
+ }
+
+ /* If virtio port just stopped, no need to send RARP */
+ if (virtio_dev_pause(dev) < 0) {
+ rte_pktmbuf_free(rarp_mbuf);
+ return;
+ }
+
+ virtio_inject_pkts(dev, &rarp_mbuf, 1);
+ virtio_dev_resume(dev);
+}
+
+static void
+virtio_ack_link_announce(struct rte_eth_dev *dev)
+{
+ struct virtio_hw *hw = dev->data->dev_private;
+ struct virtio_pmd_ctrl ctrl;
+
+ ctrl.hdr.class = VIRTIO_NET_CTRL_ANNOUNCE;
+ ctrl.hdr.cmd = VIRTIO_NET_CTRL_ANNOUNCE_ACK;
+
+ virtio_send_command(hw->cvq, &ctrl, NULL, 0);
+}
+
+/*
+ * Process virtio config changed interrupt. Call the callback
+ * if link state changed, generate gratuitous RARP packet if
+ * the status indicates an ANNOUNCE.
*/
void
virtio_interrupt_handler(void *param)
if (virtio_dev_link_update(dev, 0) == 0)
_rte_eth_dev_callback_process(dev,
RTE_ETH_EVENT_INTR_LSC,
- NULL, NULL);
+ NULL);
}
+ if (isr & VIRTIO_NET_S_ANNOUNCE) {
+ virtio_notify_peers(dev);
+ virtio_ack_link_announce(dev);
+ }
}
/* set rx and tx handlers according to what is supported */
/* Reset the device although not necessary at startup */
vtpci_reset(hw);
+ if (hw->vqs) {
+ virtio_dev_free_mbufs(eth_dev);
+ virtio_free_queues(hw);
+ }
+
/* Tell the host we've noticed this device. */
vtpci_set_status(hw, VIRTIO_CONFIG_STATUS_ACK);
} else {
PMD_INIT_LOG(DEBUG, "config->max_virtqueue_pairs=1");
hw->max_queue_pairs = 1;
+ hw->max_mtu = VIRTIO_MAX_RX_PKTLEN - ETHER_HDR_LEN -
+ VLAN_TAG_LEN - hw->vtnet_hdr_size;
}
ret = virtio_alloc_queues(eth_dev);
if (rxmode->enable_lro &&
(!vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_TSO4) ||
- !vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_TSO4))) {
+ !vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_TSO6))) {
PMD_DRV_LOG(ERR,
"Large Receive Offload not available on this host");
return -ENOTSUP;
return -EBUSY;
}
+ rte_spinlock_init(&hw->state_lock);
+
hw->use_simple_rx = 1;
hw->use_simple_tx = 1;
-#if defined RTE_ARCH_ARM64 || defined CONFIG_RTE_ARCH_ARM
+#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;
static void virtio_dev_free_mbufs(struct rte_eth_dev *dev)
{
+ struct virtio_hw *hw = dev->data->dev_private;
+ uint16_t nr_vq = virtio_get_nr_vq(hw);
+ const char *type __rte_unused;
+ unsigned int i, mbuf_num = 0;
+ struct virtqueue *vq;
struct rte_mbuf *buf;
- int i, mbuf_num = 0;
-
- for (i = 0; i < dev->data->nb_rx_queues; i++) {
- struct virtnet_rx *rxvq = dev->data->rx_queues[i];
-
- PMD_INIT_LOG(DEBUG,
- "Before freeing rxq[%d] used and unused buf", i);
- VIRTQUEUE_DUMP(rxvq->vq);
-
- PMD_INIT_LOG(DEBUG, "rx_queues[%d]=%p", i, rxvq);
- while ((buf = virtqueue_detatch_unused(rxvq->vq)) != NULL) {
- rte_pktmbuf_free(buf);
- mbuf_num++;
- }
+ int queue_type;
- PMD_INIT_LOG(DEBUG, "free %d mbufs", mbuf_num);
- PMD_INIT_LOG(DEBUG,
- "After freeing rxq[%d] used and unused buf", i);
- VIRTQUEUE_DUMP(rxvq->vq);
- }
+ for (i = 0; i < nr_vq; i++) {
+ vq = hw->vqs[i];
+ if (!vq)
+ continue;
- for (i = 0; i < dev->data->nb_tx_queues; i++) {
- struct virtnet_tx *txvq = dev->data->tx_queues[i];
+ queue_type = virtio_get_queue_type(hw, i);
+ if (queue_type == VTNET_RQ)
+ type = "rxq";
+ else if (queue_type == VTNET_TQ)
+ type = "txq";
+ else
+ continue;
PMD_INIT_LOG(DEBUG,
- "Before freeing txq[%d] used and unused bufs",
- i);
- VIRTQUEUE_DUMP(txvq->vq);
+ "Before freeing %s[%d] used and unused buf",
+ type, i);
+ VIRTQUEUE_DUMP(vq);
- mbuf_num = 0;
- while ((buf = virtqueue_detatch_unused(txvq->vq)) != NULL) {
+ while ((buf = virtqueue_detach_unused(vq)) != NULL) {
rte_pktmbuf_free(buf);
mbuf_num++;
}
- PMD_INIT_LOG(DEBUG, "free %d mbufs", mbuf_num);
PMD_INIT_LOG(DEBUG,
- "After freeing txq[%d] used and unused buf", i);
- VIRTQUEUE_DUMP(txvq->vq);
+ "After freeing %s[%d] used and unused buf",
+ type, i);
+ VIRTQUEUE_DUMP(vq);
}
+
+ PMD_INIT_LOG(DEBUG, "%d mbufs freed", mbuf_num);
}
/*
PMD_INIT_LOG(DEBUG, "stop");
+ rte_spinlock_lock(&hw->state_lock);
if (intr_conf->lsc || intr_conf->rxq)
virtio_intr_disable(dev);
hw->started = 0;
memset(&link, 0, sizeof(link));
virtio_dev_atomic_write_link_status(dev, &link);
+ rte_spinlock_unlock(&hw->state_lock);
}
static int
RTE_PMD_EXPORT_NAME(net_virtio, __COUNTER__);
RTE_PMD_REGISTER_PCI_TABLE(net_virtio, pci_id_virtio_map);
RTE_PMD_REGISTER_KMOD_DEP(net_virtio, "* igb_uio | uio_pci_generic | vfio-pci");
+
+RTE_INIT(virtio_init_log);
+static void
+virtio_init_log(void)
+{
+ virtio_logtype_init = rte_log_register("pmd.virtio.init");
+ if (virtio_logtype_init >= 0)
+ rte_log_set_level(virtio_logtype_init, RTE_LOG_NOTICE);
+ virtio_logtype_driver = rte_log_register("pmd.virtio.driver");
+ if (virtio_logtype_driver >= 0)
+ rte_log_set_level(virtio_logtype_driver, RTE_LOG_NOTICE);
+}