#include <sys/mman.h>
#include <unistd.h>
#include <sys/eventfd.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
#include "vhost.h"
#include "virtio_user_dev.h"
#include "../virtio_ethdev.h"
+static int
+virtio_user_create_queue(struct virtio_user_dev *dev, uint32_t queue_sel)
+{
+ /* Of all per virtqueue MSGs, make sure VHOST_SET_VRING_CALL come
+ * firstly because vhost depends on this msg to allocate virtqueue
+ * pair.
+ */
+ struct vhost_vring_file file;
+
+ file.index = queue_sel;
+ file.fd = dev->callfds[queue_sel];
+ dev->ops->send_request(dev, VHOST_USER_SET_VRING_CALL, &file);
+
+ return 0;
+}
+
static int
virtio_user_kick_queue(struct virtio_user_dev *dev, uint32_t queue_sel)
{
- int callfd, kickfd;
struct vhost_vring_file file;
struct vhost_vring_state state;
struct vring *vring = &dev->vrings[queue_sel];
.flags = 0, /* disable log */
};
- /* May use invalid flag, but some backend leverages kickfd and callfd as
- * criteria to judge if dev is alive. so finally we use real event_fd.
- */
- callfd = eventfd(0, EFD_CLOEXEC | EFD_NONBLOCK);
- if (callfd < 0) {
- PMD_DRV_LOG(ERR, "callfd error, %s\n", strerror(errno));
- return -1;
- }
- kickfd = eventfd(0, EFD_CLOEXEC | EFD_NONBLOCK);
- if (kickfd < 0) {
- close(callfd);
- PMD_DRV_LOG(ERR, "kickfd error, %s\n", strerror(errno));
- return -1;
- }
-
- /* Of all per virtqueue MSGs, make sure VHOST_SET_VRING_CALL come
- * firstly because vhost depends on this msg to allocate virtqueue
- * pair.
- */
- file.index = queue_sel;
- file.fd = callfd;
- vhost_user_sock(dev->vhostfd, VHOST_USER_SET_VRING_CALL, &file);
- dev->callfds[queue_sel] = callfd;
-
state.index = queue_sel;
state.num = vring->num;
- vhost_user_sock(dev->vhostfd, VHOST_USER_SET_VRING_NUM, &state);
+ dev->ops->send_request(dev, VHOST_USER_SET_VRING_NUM, &state);
+ state.index = queue_sel;
state.num = 0; /* no reservation */
- vhost_user_sock(dev->vhostfd, VHOST_USER_SET_VRING_BASE, &state);
+ dev->ops->send_request(dev, VHOST_USER_SET_VRING_BASE, &state);
- vhost_user_sock(dev->vhostfd, VHOST_USER_SET_VRING_ADDR, &addr);
+ dev->ops->send_request(dev, VHOST_USER_SET_VRING_ADDR, &addr);
/* Of all per virtqueue MSGs, make sure VHOST_USER_SET_VRING_KICK comes
* lastly because vhost depends on this msg to judge if
* virtio is ready.
*/
- file.fd = kickfd;
- vhost_user_sock(dev->vhostfd, VHOST_USER_SET_VRING_KICK, &file);
- dev->kickfds[queue_sel] = kickfd;
+ file.index = queue_sel;
+ file.fd = dev->kickfds[queue_sel];
+ dev->ops->send_request(dev, VHOST_USER_SET_VRING_KICK, &file);
return 0;
}
-int
-virtio_user_start_device(struct virtio_user_dev *dev)
+static int
+virtio_user_queue_setup(struct virtio_user_dev *dev,
+ int (*fn)(struct virtio_user_dev *, uint32_t))
{
- uint64_t features;
uint32_t i, queue_sel;
- int ret;
-
- /* construct memory region inside each implementation */
- ret = vhost_user_sock(dev->vhostfd, VHOST_USER_SET_MEM_TABLE, NULL);
- if (ret < 0)
- goto error;
for (i = 0; i < dev->max_queue_pairs; ++i) {
queue_sel = 2 * i + VTNET_SQ_RQ_QUEUE_IDX;
- if (virtio_user_kick_queue(dev, queue_sel) < 0) {
- PMD_DRV_LOG(INFO, "kick rx vq fails: %u", i);
- goto error;
+ if (fn(dev, queue_sel) < 0) {
+ PMD_DRV_LOG(INFO, "setup rx vq fails: %u", i);
+ return -1;
}
}
for (i = 0; i < dev->max_queue_pairs; ++i) {
queue_sel = 2 * i + VTNET_SQ_TQ_QUEUE_IDX;
- if (virtio_user_kick_queue(dev, queue_sel) < 0) {
- PMD_DRV_LOG(INFO, "kick tx vq fails: %u", i);
- goto error;
+ if (fn(dev, queue_sel) < 0) {
+ PMD_DRV_LOG(INFO, "setup tx vq fails: %u", i);
+ return -1;
}
}
- /* After setup all virtqueues, we need to set_features so that these
- * features can be set into each virtqueue in vhost side. And before
- * that, make sure VHOST_USER_F_PROTOCOL_FEATURES is added if mq is
- * enabled, and VIRTIO_NET_F_MAC is stripped.
- */
+ return 0;
+}
+
+int
+virtio_user_start_device(struct virtio_user_dev *dev)
+{
+ uint64_t features;
+ int ret;
+
+ /* Step 0: tell vhost to create queues */
+ if (virtio_user_queue_setup(dev, virtio_user_create_queue) < 0)
+ goto error;
+
+ /* Step 1: set features */
features = dev->features;
- if (dev->max_queue_pairs > 1)
- features |= VHOST_USER_MQ;
+ /* Strip VIRTIO_NET_F_MAC, as MAC address is handled in vdev init */
features &= ~(1ull << VIRTIO_NET_F_MAC);
- ret = vhost_user_sock(dev->vhostfd, VHOST_USER_SET_FEATURES, &features);
+ /* Strip VIRTIO_NET_F_CTRL_VQ, as devices do not really need to know */
+ features &= ~(1ull << VIRTIO_NET_F_CTRL_VQ);
+ features &= ~(1ull << VIRTIO_NET_F_STATUS);
+ ret = dev->ops->send_request(dev, VHOST_USER_SET_FEATURES, &features);
if (ret < 0)
goto error;
PMD_DRV_LOG(INFO, "set features: %" PRIx64, features);
+ /* Step 2: share memory regions */
+ ret = dev->ops->send_request(dev, VHOST_USER_SET_MEM_TABLE, NULL);
+ if (ret < 0)
+ goto error;
+
+ /* Step 3: kick queues */
+ if (virtio_user_queue_setup(dev, virtio_user_kick_queue) < 0)
+ goto error;
+
+ /* Step 4: enable queues
+ * we enable the 1st queue pair by default.
+ */
+ dev->ops->enable_qp(dev, 0, 1);
+
return 0;
error:
/* TODO: free resource here or caller to check */
int virtio_user_stop_device(struct virtio_user_dev *dev)
{
- return vhost_user_sock(dev->vhostfd, VHOST_USER_RESET_OWNER, NULL);
+ uint32_t i;
+
+ for (i = 0; i < dev->max_queue_pairs; ++i)
+ dev->ops->enable_qp(dev, i, 0);
+
+ return 0;
}
static inline void
}
}
+int
+is_vhost_user_by_type(const char *path)
+{
+ struct stat sb;
+
+ if (stat(path, &sb) == -1)
+ return 0;
+
+ return S_ISSOCK(sb.st_mode);
+}
+
+static int
+virtio_user_dev_init_notify(struct virtio_user_dev *dev)
+{
+ uint32_t i, j;
+ int callfd;
+ int kickfd;
+
+ for (i = 0; i < VIRTIO_MAX_VIRTQUEUES; ++i) {
+ if (i >= dev->max_queue_pairs * 2) {
+ dev->kickfds[i] = -1;
+ dev->callfds[i] = -1;
+ continue;
+ }
+
+ /* May use invalid flag, but some backend uses kickfd and
+ * callfd as criteria to judge if dev is alive. so finally we
+ * use real event_fd.
+ */
+ callfd = eventfd(0, EFD_CLOEXEC | EFD_NONBLOCK);
+ if (callfd < 0) {
+ PMD_DRV_LOG(ERR, "callfd error, %s", strerror(errno));
+ break;
+ }
+ kickfd = eventfd(0, EFD_CLOEXEC | EFD_NONBLOCK);
+ if (kickfd < 0) {
+ PMD_DRV_LOG(ERR, "kickfd error, %s", strerror(errno));
+ break;
+ }
+ dev->callfds[i] = callfd;
+ dev->kickfds[i] = kickfd;
+ }
+
+ if (i < VIRTIO_MAX_VIRTQUEUES) {
+ for (j = 0; j <= i; ++j) {
+ close(dev->callfds[j]);
+ close(dev->kickfds[j]);
+ }
+
+ return -1;
+ }
+
+ return 0;
+}
+
+static int
+virtio_user_fill_intr_handle(struct virtio_user_dev *dev)
+{
+ uint32_t i;
+ struct rte_eth_dev *eth_dev = &rte_eth_devices[dev->port_id];
+
+ if (!eth_dev->intr_handle) {
+ eth_dev->intr_handle = malloc(sizeof(*eth_dev->intr_handle));
+ if (!eth_dev->intr_handle) {
+ PMD_DRV_LOG(ERR, "fail to allocate intr_handle");
+ return -1;
+ }
+ memset(eth_dev->intr_handle, 0, sizeof(*eth_dev->intr_handle));
+ }
+
+ for (i = 0; i < dev->max_queue_pairs; ++i)
+ eth_dev->intr_handle->efds[i] = dev->callfds[i];
+ eth_dev->intr_handle->nb_efd = dev->max_queue_pairs;
+ eth_dev->intr_handle->max_intr = dev->max_queue_pairs + 1;
+ eth_dev->intr_handle->type = RTE_INTR_HANDLE_VDEV;
+ if (dev->vhostfd >= 0)
+ eth_dev->intr_handle->fd = dev->vhostfd;
+
+ return 0;
+}
+
+static int
+virtio_user_dev_setup(struct virtio_user_dev *dev)
+{
+ uint32_t q;
+
+ dev->vhostfd = -1;
+ dev->vhostfds = NULL;
+ dev->tapfds = NULL;
+
+ if (is_vhost_user_by_type(dev->path)) {
+ dev->ops = &ops_user;
+ } else {
+ dev->ops = &ops_kernel;
+
+ dev->vhostfds = malloc(dev->max_queue_pairs * sizeof(int));
+ dev->tapfds = malloc(dev->max_queue_pairs * sizeof(int));
+ if (!dev->vhostfds || !dev->tapfds) {
+ PMD_INIT_LOG(ERR, "Failed to malloc");
+ return -1;
+ }
+
+ for (q = 0; q < dev->max_queue_pairs; ++q) {
+ dev->vhostfds[q] = -1;
+ dev->tapfds[q] = -1;
+ }
+ }
+
+ if (dev->ops->setup(dev) < 0)
+ return -1;
+
+ if (virtio_user_dev_init_notify(dev) < 0)
+ return -1;
+
+ if (virtio_user_fill_intr_handle(dev) < 0)
+ return -1;
+
+ return 0;
+}
+
+/* Use below macro to filter features from vhost backend */
+#define VIRTIO_USER_SUPPORTED_FEATURES \
+ (1ULL << VIRTIO_NET_F_MAC | \
+ 1ULL << VIRTIO_NET_F_STATUS | \
+ 1ULL << VIRTIO_NET_F_MQ | \
+ 1ULL << VIRTIO_NET_F_CTRL_MAC_ADDR | \
+ 1ULL << VIRTIO_NET_F_CTRL_VQ | \
+ 1ULL << VIRTIO_NET_F_CTRL_RX | \
+ 1ULL << VIRTIO_NET_F_CTRL_VLAN | \
+ 1ULL << VIRTIO_NET_F_CSUM | \
+ 1ULL << VIRTIO_NET_F_HOST_TSO4 | \
+ 1ULL << VIRTIO_NET_F_HOST_TSO6 | \
+ 1ULL << VIRTIO_NET_F_MRG_RXBUF | \
+ 1ULL << VIRTIO_RING_F_INDIRECT_DESC | \
+ 1ULL << VIRTIO_NET_F_GUEST_CSUM | \
+ 1ULL << VIRTIO_NET_F_GUEST_TSO4 | \
+ 1ULL << VIRTIO_NET_F_GUEST_TSO6 | \
+ 1ULL << VIRTIO_F_VERSION_1)
+
int
virtio_user_dev_init(struct virtio_user_dev *dev, char *path, int queues,
- int cq, int queue_size, const char *mac)
+ int cq, int queue_size, const char *mac, char **ifname)
{
- strncpy(dev->path, path, PATH_MAX);
+ snprintf(dev->path, PATH_MAX, "%s", path);
dev->max_queue_pairs = queues;
dev->queue_pairs = 1; /* mq disabled by default */
dev->queue_size = queue_size;
dev->mac_specified = 0;
parse_mac(dev, mac);
- dev->vhostfd = -1;
- dev->vhostfd = vhost_user_setup(dev->path);
- if (dev->vhostfd < 0) {
+ if (*ifname) {
+ dev->ifname = *ifname;
+ *ifname = NULL;
+ }
+
+ if (virtio_user_dev_setup(dev) < 0) {
PMD_INIT_LOG(ERR, "backend set up fails");
return -1;
}
- if (vhost_user_sock(dev->vhostfd, VHOST_USER_SET_OWNER, NULL) < 0) {
+ if (dev->ops->send_request(dev, VHOST_USER_SET_OWNER, NULL) < 0) {
PMD_INIT_LOG(ERR, "set_owner fails: %s", strerror(errno));
return -1;
}
- if (vhost_user_sock(dev->vhostfd, VHOST_USER_GET_FEATURES,
- &dev->features) < 0) {
+ if (dev->ops->send_request(dev, VHOST_USER_GET_FEATURES,
+ &dev->device_features) < 0) {
PMD_INIT_LOG(ERR, "get_features failed: %s", strerror(errno));
return -1;
}
if (dev->mac_specified)
- dev->features |= (1ull << VIRTIO_NET_F_MAC);
+ dev->device_features |= (1ull << VIRTIO_NET_F_MAC);
- if (!cq) {
- dev->features &= ~(1ull << VIRTIO_NET_F_CTRL_VQ);
- /* Also disable features depends on VIRTIO_NET_F_CTRL_VQ */
- dev->features &= ~(1ull << VIRTIO_NET_F_CTRL_RX);
- dev->features &= ~(1ull << VIRTIO_NET_F_CTRL_VLAN);
- dev->features &= ~(1ull << VIRTIO_NET_F_GUEST_ANNOUNCE);
- dev->features &= ~(1ull << VIRTIO_NET_F_MQ);
- dev->features &= ~(1ull << VIRTIO_NET_F_CTRL_MAC_ADDR);
- } else {
- /* vhost user backend does not need to know ctrl-q, so
- * actually we need add this bit into features. However,
- * DPDK vhost-user does send features with this bit, so we
- * check it instead of OR it for now.
+ if (cq) {
+ /* device does not really need to know anything about CQ,
+ * so if necessary, we just claim to support CQ
*/
- if (!(dev->features & (1ull << VIRTIO_NET_F_CTRL_VQ)))
- PMD_INIT_LOG(INFO, "vhost does not support ctrl-q");
+ dev->device_features |= (1ull << VIRTIO_NET_F_CTRL_VQ);
+ } else {
+ dev->device_features &= ~(1ull << VIRTIO_NET_F_CTRL_VQ);
+ /* Also disable features depends on VIRTIO_NET_F_CTRL_VQ */
+ dev->device_features &= ~(1ull << VIRTIO_NET_F_CTRL_RX);
+ dev->device_features &= ~(1ull << VIRTIO_NET_F_CTRL_VLAN);
+ dev->device_features &= ~(1ull << VIRTIO_NET_F_GUEST_ANNOUNCE);
+ dev->device_features &= ~(1ull << VIRTIO_NET_F_MQ);
+ dev->device_features &= ~(1ull << VIRTIO_NET_F_CTRL_MAC_ADDR);
}
- if (dev->max_queue_pairs > 1) {
- if (!(dev->features & VHOST_USER_MQ)) {
- PMD_INIT_LOG(ERR, "MQ not supported by the backend");
- return -1;
- }
- }
+ /* The backend will not report this feature, we add it explicitly */
+ if (is_vhost_user_by_type(dev->path))
+ dev->device_features |= (1ull << VIRTIO_NET_F_STATUS);
+
+ dev->device_features &= VIRTIO_USER_SUPPORTED_FEATURES;
return 0;
}
{
uint32_t i;
+ virtio_user_stop_device(dev);
+
for (i = 0; i < dev->max_queue_pairs * 2; ++i) {
close(dev->callfds[i]);
close(dev->kickfds[i]);
}
close(dev->vhostfd);
+
+ if (dev->vhostfds) {
+ for (i = 0; i < dev->max_queue_pairs; ++i)
+ close(dev->vhostfds[i]);
+ free(dev->vhostfds);
+ free(dev->tapfds);
+ }
+
+ free(dev->ifname);
}
static uint8_t
}
for (i = 0; i < q_pairs; ++i)
- ret |= vhost_user_enable_queue_pair(dev->vhostfd, i, 1);
+ ret |= dev->ops->enable_qp(dev, i, 1);
for (i = q_pairs; i < dev->max_queue_pairs; ++i)
- ret |= vhost_user_enable_queue_pair(dev->vhostfd, i, 0);
+ ret |= dev->ops->enable_qp(dev, i, 0);
dev->queue_pairs = q_pairs;