#include <stdint.h>
#include <stdlib.h>
#ifdef RTE_LIBRTE_VHOST_NUMA
+#include <numa.h>
#include <numaif.h>
#endif
return 0;
}
+#define VHOST_LOG_PAGE 4096
+
+/*
+ * Atomically set a bit in memory.
+ */
+static __rte_always_inline void
+vhost_set_bit(unsigned int nr, volatile uint8_t *addr)
+{
+#if defined(RTE_TOOLCHAIN_GCC) && (GCC_VERSION < 70100)
+ /*
+ * __sync_ built-ins are deprecated, but __atomic_ ones
+ * are sub-optimized in older GCC versions.
+ */
+ __sync_fetch_and_or_1(addr, (1U << nr));
+#else
+ __atomic_fetch_or(addr, (1U << nr), __ATOMIC_RELAXED);
+#endif
+}
+
+static __rte_always_inline void
+vhost_log_page(uint8_t *log_base, uint64_t page)
+{
+ vhost_set_bit(page % 8, &log_base[page / 8]);
+}
+
+void
+__vhost_log_write(struct virtio_net *dev, uint64_t addr, uint64_t len)
+{
+ uint64_t page;
+
+ if (unlikely(!dev->log_base || !len))
+ return;
+
+ if (unlikely(dev->log_size <= ((addr + len - 1) / VHOST_LOG_PAGE / 8)))
+ return;
+
+ /* To make sure guest memory updates are committed before logging */
+ rte_smp_wmb();
+
+ page = addr / VHOST_LOG_PAGE;
+ while (page * VHOST_LOG_PAGE < addr + len) {
+ vhost_log_page((uint8_t *)(uintptr_t)dev->log_base, page);
+ page += 1;
+ }
+}
+
+void
+__vhost_log_cache_sync(struct virtio_net *dev, struct vhost_virtqueue *vq)
+{
+ unsigned long *log_base;
+ int i;
+
+ if (unlikely(!dev->log_base))
+ return;
+
+ rte_smp_wmb();
+
+ log_base = (unsigned long *)(uintptr_t)dev->log_base;
+
+ for (i = 0; i < vq->log_cache_nb_elem; i++) {
+ struct log_cache_entry *elem = vq->log_cache + i;
+
+#if defined(RTE_TOOLCHAIN_GCC) && (GCC_VERSION < 70100)
+ /*
+ * '__sync' builtins are deprecated, but '__atomic' ones
+ * are sub-optimized in older GCC versions.
+ */
+ __sync_fetch_and_or(log_base + elem->offset, elem->val);
+#else
+ __atomic_fetch_or(log_base + elem->offset, elem->val,
+ __ATOMIC_RELAXED);
+#endif
+ }
+
+ rte_smp_wmb();
+
+ vq->log_cache_nb_elem = 0;
+}
+
+static __rte_always_inline void
+vhost_log_cache_page(struct virtio_net *dev, struct vhost_virtqueue *vq,
+ uint64_t page)
+{
+ uint32_t bit_nr = page % (sizeof(unsigned long) << 3);
+ uint32_t offset = page / (sizeof(unsigned long) << 3);
+ int i;
+
+ for (i = 0; i < vq->log_cache_nb_elem; i++) {
+ struct log_cache_entry *elem = vq->log_cache + i;
+
+ if (elem->offset == offset) {
+ elem->val |= (1UL << bit_nr);
+ return;
+ }
+ }
+
+ if (unlikely(i >= VHOST_LOG_CACHE_NR)) {
+ /*
+ * No more room for a new log cache entry,
+ * so write the dirty log map directly.
+ */
+ rte_smp_wmb();
+ vhost_log_page((uint8_t *)(uintptr_t)dev->log_base, page);
+
+ return;
+ }
+
+ vq->log_cache[i].offset = offset;
+ vq->log_cache[i].val = (1UL << bit_nr);
+ vq->log_cache_nb_elem++;
+}
+
+void
+__vhost_log_cache_write(struct virtio_net *dev, struct vhost_virtqueue *vq,
+ uint64_t addr, uint64_t len)
+{
+ uint64_t page;
+
+ if (unlikely(!dev->log_base || !len))
+ return;
+
+ if (unlikely(dev->log_size <= ((addr + len - 1) / VHOST_LOG_PAGE / 8)))
+ return;
+
+ page = addr / VHOST_LOG_PAGE;
+ while (page * VHOST_LOG_PAGE < addr + len) {
+ vhost_log_cache_page(dev, vq, page);
+ page += 1;
+ }
+}
+
+void *
+vhost_alloc_copy_ind_table(struct virtio_net *dev, struct vhost_virtqueue *vq,
+ uint64_t desc_addr, uint64_t desc_len)
+{
+ void *idesc;
+ uint64_t src, dst;
+ uint64_t len, remain = desc_len;
+
+ idesc = rte_malloc(__func__, desc_len, 0);
+ if (unlikely(!idesc))
+ return NULL;
+
+ dst = (uint64_t)(uintptr_t)idesc;
+
+ while (remain) {
+ len = remain;
+ src = vhost_iova_to_vva(dev, vq, desc_addr, &len,
+ VHOST_ACCESS_RO);
+ if (unlikely(!src || !len)) {
+ rte_free(idesc);
+ return NULL;
+ }
+
+ rte_memcpy((void *)(uintptr_t)dst, (void *)(uintptr_t)src, len);
+
+ remain -= len;
+ dst += len;
+ desc_addr += len;
+ }
+
+ return idesc;
+}
+
void
cleanup_vq(struct vhost_virtqueue *vq, int destroy)
{
{
if (!(dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
- goto out;
+ return -1;
if (vq_is_packed(dev)) {
if (vring_translate_packed(dev, vq) < 0)
if (vring_translate_split(dev, vq) < 0)
return -1;
}
-out:
vq->access_ok = 1;
return 0;
vq->desc = NULL;
vq->avail = NULL;
vq->used = NULL;
+ vq->log_guest_addr = 0;
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_wr_unlock(vq);
dev->flags = VIRTIO_DEV_BUILTIN_VIRTIO_NET;
dev->slave_req_fd = -1;
dev->vdpa_dev_id = -1;
+ dev->postcopy_ufd = -1;
rte_spinlock_init(&dev->slave_req_lock);
return i;
dev->vdpa_dev_id = did;
}
-void
-vhost_detach_vdpa_device(int vid)
-{
- struct virtio_net *dev = get_device(vid);
-
- if (dev == NULL)
- return;
-
- vhost_user_host_notifier_ctrl(vid, false);
-
- dev->vdpa_dev_id = -1;
-}
-
void
vhost_set_ifname(int vid, const char *if_name, unsigned int if_len)
{
{
struct virtio_net *dev = get_device(vid);
- if (!dev)
+ if (dev == NULL || mtu == NULL)
return -ENODEV;
if (!(dev->flags & VIRTIO_DEV_READY))
int numa_node;
int ret;
- if (dev == NULL)
+ if (dev == NULL || numa_available() != 0)
return -1;
ret = get_mempolicy(&numa_node, NULL, 0, dev,
{
struct virtio_net *dev = get_device(vid);
- if (dev == NULL)
+ if (dev == NULL || buf == NULL)
return -1;
len = RTE_MIN(len, sizeof(dev->ifname));
struct virtio_net *dev;
dev = get_device(vid);
- if (!dev)
+ if (dev == NULL || features == NULL)
return -1;
*features = dev->features;
size_t size;
dev = get_device(vid);
- if (!dev)
+ if (dev == NULL || mem == NULL)
return -1;
size = dev->mem->nregions * sizeof(struct rte_vhost_mem_region);
struct vhost_virtqueue *vq;
dev = get_device(vid);
- if (!dev)
+ if (dev == NULL || vring == NULL)
return -1;
if (vring_idx >= VHOST_MAX_VRING)
{
struct virtio_net *dev;
struct vhost_virtqueue *vq;
+ uint16_t ret = 0;
dev = get_device(vid);
if (!dev)
return 0;
vq = dev->virtqueue[queue_id];
- if (!vq->enabled)
- return 0;
- return *(volatile uint16_t *)&vq->avail->idx - vq->last_used_idx;
+ rte_spinlock_lock(&vq->access_lock);
+
+ if (unlikely(!vq->enabled || vq->avail == NULL))
+ goto out;
+
+ ret = *(volatile uint16_t *)&vq->avail->idx - vq->last_used_idx;
+
+out:
+ rte_spinlock_unlock(&vq->access_lock);
+ return ret;
}
-static inline void
-vhost_enable_notify_split(struct vhost_virtqueue *vq, int enable)
+static inline int
+vhost_enable_notify_split(struct virtio_net *dev,
+ struct vhost_virtqueue *vq, int enable)
{
- if (enable)
- vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
- else
- vq->used->flags |= VRING_USED_F_NO_NOTIFY;
+ if (vq->used == NULL)
+ return -1;
+
+ if (!(dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX))) {
+ if (enable)
+ vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
+ else
+ vq->used->flags |= VRING_USED_F_NO_NOTIFY;
+ } else {
+ if (enable)
+ vhost_avail_event(vq) = vq->last_avail_idx;
+ }
+ return 0;
}
-static inline void
+static inline int
vhost_enable_notify_packed(struct virtio_net *dev,
struct vhost_virtqueue *vq, int enable)
{
uint16_t flags;
- if (!enable)
+ if (vq->device_event == NULL)
+ return -1;
+
+ if (!enable) {
vq->device_event->flags = VRING_EVENT_F_DISABLE;
+ return 0;
+ }
flags = VRING_EVENT_F_ENABLE;
if (dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX)) {
rte_smp_wmb();
vq->device_event->flags = flags;
+ return 0;
}
int
{
struct virtio_net *dev = get_device(vid);
struct vhost_virtqueue *vq;
+ int ret;
if (!dev)
return -1;
vq = dev->virtqueue[queue_id];
+ rte_spinlock_lock(&vq->access_lock);
+
if (vq_is_packed(dev))
- vhost_enable_notify_packed(dev, vq, enable);
+ ret = vhost_enable_notify_packed(dev, vq, enable);
else
- vhost_enable_notify_split(vq, enable);
+ ret = vhost_enable_notify_split(dev, vq, enable);
- return 0;
+ rte_spinlock_unlock(&vq->access_lock);
+
+ return ret;
}
void
{
struct virtio_net *dev;
struct vhost_virtqueue *vq;
+ uint32_t ret = 0;
dev = get_device(vid);
if (dev == NULL)
if (vq == NULL)
return 0;
+ rte_spinlock_lock(&vq->access_lock);
+
if (unlikely(vq->enabled == 0 || vq->avail == NULL))
- return 0;
+ goto out;
+
+ ret = *((volatile uint16_t *)&vq->avail->idx) - vq->last_avail_idx;
- return *((volatile uint16_t *)&vq->avail->idx) - vq->last_avail_idx;
+out:
+ rte_spinlock_unlock(&vq->access_lock);
+ return ret;
}
int rte_vhost_get_vdpa_device_id(int vid)
{
struct virtio_net *dev = get_device(vid);
- if (!dev)
- return -1;
-
- if (unlikely(!(dev->flags & VIRTIO_DEV_BUILTIN_VIRTIO_NET))) {
- RTE_LOG(ERR, VHOST_DATA,
- "(%d) %s: built-in vhost net backend is disabled.\n",
- dev->vid, __func__);
+ if (dev == NULL || log_base == NULL || log_size == NULL)
return -1;
- }
*log_base = dev->log_base;
*log_size = dev->log_size;
{
struct virtio_net *dev = get_device(vid);
- if (!dev)
+ if (dev == NULL || last_avail_idx == NULL || last_used_idx == NULL)
return -1;
- if (unlikely(!(dev->flags & VIRTIO_DEV_BUILTIN_VIRTIO_NET))) {
- RTE_LOG(ERR, VHOST_DATA,
- "(%d) %s: built-in vhost net backend is disabled.\n",
- dev->vid, __func__);
- return -1;
- }
-
*last_avail_idx = dev->virtqueue[queue_id]->last_avail_idx;
*last_used_idx = dev->virtqueue[queue_id]->last_used_idx;
if (!dev)
return -1;
- if (unlikely(!(dev->flags & VIRTIO_DEV_BUILTIN_VIRTIO_NET))) {
- RTE_LOG(ERR, VHOST_DATA,
- "(%d) %s: built-in vhost net backend is disabled.\n",
- dev->vid, __func__);
- return -1;
- }
-
dev->virtqueue[queue_id]->last_avail_idx = last_avail_idx;
dev->virtqueue[queue_id]->last_used_idx = last_used_idx;
return 0;
}
+
+int rte_vhost_extern_callback_register(int vid,
+ struct rte_vhost_user_extern_ops const * const ops, void *ctx)
+{
+ struct virtio_net *dev = get_device(vid);
+
+ if (dev == NULL || ops == NULL)
+ return -1;
+
+ dev->extern_ops = *ops;
+ dev->extern_data = ctx;
+ return 0;
+}