examples/vhost_blk/vhost_blk.c

/* SPDX-License-Identifier: BSD-3-Clause
 * Copyright(c) 2010-2019 Intel Corporation
 */

#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
#include <pthread.h>
#include <sched.h>

#include <stdint.h>
#include <unistd.h>
#include <stdbool.h>
#include <signal.h>
#include <assert.h>
#include <semaphore.h>
#include <linux/virtio_blk.h>
#include <linux/virtio_ring.h>

#include <rte_atomic.h>
#include <rte_cycles.h>
#include <rte_log.h>
#include <rte_malloc.h>
#include <rte_vhost.h>

#include "vhost_blk.h"
#include "blk_spec.h"

#define VIRTQ_DESC_F_NEXT   1
#define VIRTQ_DESC_F_AVAIL  (1 << 7)
#define VIRTQ_DESC_F_USED   (1 << 15)

#define MAX_TASK        12

#define VHOST_BLK_FEATURES ((1ULL << VIRTIO_F_RING_PACKED) | \
                (1ULL << VIRTIO_F_VERSION_1) |\
                (1ULL << VIRTIO_F_NOTIFY_ON_EMPTY) | \
                (1ULL << VHOST_USER_F_PROTOCOL_FEATURES))
#define CTRLR_NAME      "vhost.socket"

enum CTRLR_WORKER_STATUS {
    WORKER_STATE_START = 0,
    WORKER_STATE_STOP,
};

struct vhost_blk_ctrlr *g_vhost_ctrlr;

/* Path to folder where character device will be created. Can be set by user. */
static char dev_pathname[PATH_MAX] = "";
static sem_t exit_sem;
static enum CTRLR_WORKER_STATUS worker_thread_status;

struct vhost_blk_ctrlr *
vhost_blk_ctrlr_find(const char *ctrlr_name)
{
    if (ctrlr_name == NULL)
        return NULL;

    /* currently we only support 1 socket file fd */
    return g_vhost_ctrlr;
}

static uint64_t
gpa_to_vva(struct vhost_blk_ctrlr *ctrlr, uint64_t gpa, uint64_t *len)
{
    assert(ctrlr->mem != NULL);

    return rte_vhost_va_from_guest_pa(ctrlr->mem, gpa, len);
}

static void
enqueue_task(struct vhost_blk_task *task)
{
    struct vhost_blk_queue *vq = task->vq;
    struct vring_used *used = vq->vring.used;

    rte_vhost_set_last_inflight_io_split(task->ctrlr->vid,
        vq->id, task->req_idx);

    /* Fill out the next entry in the "used" ring.  id = the
     * index of the descriptor that contained the blk request.
     * len = the total amount of data transferred for the blk
     * request. We must report the correct len, for variable
     * length blk CDBs, where we may return less data than
     * allocated by the guest VM.
     */
    used->ring[used->idx & (vq->vring.size - 1)].id = task->req_idx;
    used->ring[used->idx & (vq->vring.size - 1)].len = task->data_len;
    rte_smp_mb();
    used->idx++;
    rte_smp_mb();

    rte_vhost_clr_inflight_desc_split(task->ctrlr->vid,
        vq->id, used->idx, task->req_idx);

    /* Send an interrupt back to the guest VM so that it knows
     * a completion is ready to be processed.
     */
    rte_vhost_vring_call(task->ctrlr->vid, vq->id);
}

static void
enqueue_task_packed(struct vhost_blk_task *task)
{
    struct vhost_blk_queue *vq = task->vq;
    struct vring_packed_desc *desc;

    rte_vhost_set_last_inflight_io_packed(task->ctrlr->vid, vq->id,
                        task->inflight_idx);

    desc = &vq->vring.desc_packed[vq->last_used_idx];
    desc->id = task->buffer_id;
    desc->addr = 0;

    rte_smp_mb();
    if (vq->used_wrap_counter)
        desc->flags |= VIRTQ_DESC_F_AVAIL | VIRTQ_DESC_F_USED;
    else
        desc->flags &= ~(VIRTQ_DESC_F_AVAIL | VIRTQ_DESC_F_USED);
    rte_smp_mb();

    rte_vhost_clr_inflight_desc_packed(task->ctrlr->vid, vq->id,
                       task->inflight_idx);

    vq->last_used_idx += task->chain_num;
    if (vq->last_used_idx >= vq->vring.size) {
        vq->last_used_idx -= vq->vring.size;
        vq->used_wrap_counter = !vq->used_wrap_counter;
    }

    /* Send an interrupt back to the guest VM so that it knows
     * a completion is ready to be processed.
     */
    rte_vhost_vring_call(task->ctrlr->vid, vq->id);
}

static bool
descriptor_has_next_packed(struct vring_packed_desc *cur_desc)
{
    return !!(cur_desc->flags & VRING_DESC_F_NEXT);
}

static bool
descriptor_has_next_split(struct vring_desc *cur_desc)
{
    return !!(cur_desc->flags & VRING_DESC_F_NEXT);
}

static int
desc_payload_to_iovs(struct vhost_blk_ctrlr *ctrlr, struct iovec *iovs,
             uint32_t *iov_index, uintptr_t payload, uint64_t remaining)
{
    void *vva;
    uint64_t len;

    do {
        if (*iov_index >= VHOST_BLK_MAX_IOVS) {
            fprintf(stderr, "VHOST_BLK_MAX_IOVS reached\n");
            return -1;
        }
        len = remaining;
        vva = (void *)(uintptr_t)gpa_to_vva(ctrlr,
                 payload, &len);
        if (!vva || !len) {
            fprintf(stderr, "failed to translate desc address.\n");
            return -1;
        }

        iovs[*iov_index].iov_base = vva;
        iovs[*iov_index].iov_len = len;
        payload += len;
        remaining -= len;
        (*iov_index)++;
    } while (remaining);

    return 0;
}

static struct vring_desc *
vring_get_next_desc(struct vhost_blk_queue *vq, struct vring_desc *desc)
{
    if (descriptor_has_next_split(desc))
        return &vq->vring.desc[desc->next];

    return NULL;
}

static struct vring_packed_desc *
vring_get_next_desc_packed(struct vhost_blk_queue *vq, uint16_t *req_idx)
{
    if (descriptor_has_next_packed(&vq->vring.desc_packed[*req_idx])) {
        *req_idx = (*req_idx + 1) % vq->vring.size;
        return &vq->vring.desc_packed[*req_idx];
    }

    return NULL;
}

static struct rte_vhost_inflight_desc_packed *
vring_get_next_inflight_desc(struct vhost_blk_queue *vq,
            struct rte_vhost_inflight_desc_packed *desc)
{
    if (!!(desc->flags & VRING_DESC_F_NEXT))
        return &vq->inflight_ring.inflight_packed->desc[desc->next];

    return NULL;
}

static int
setup_iovs_from_descs_split(struct vhost_blk_ctrlr *ctrlr,
                struct vhost_blk_queue *vq, uint16_t req_idx,
                struct iovec *iovs, uint32_t *iovs_idx,
                uint32_t *payload)
{
    struct vring_desc *desc = &vq->vring.desc[req_idx];

    do {
        /* does not support indirect descriptors */
        assert((desc->flags & VRING_DESC_F_INDIRECT) == 0);

        if (*iovs_idx >= VHOST_BLK_MAX_IOVS) {
            fprintf(stderr, "Reach VHOST_BLK_MAX_IOVS\n");
            return -1;
        }

        if (desc_payload_to_iovs(ctrlr, iovs, iovs_idx,
            desc->addr, desc->len) != 0) {
            fprintf(stderr, "Failed to convert desc payload to iovs\n");
            return -1;
        }

        *payload += desc->len;

        desc = vring_get_next_desc(vq, desc);
    } while (desc != NULL);

    return 0;
}

static int
setup_iovs_from_descs_packed(struct vhost_blk_ctrlr *ctrlr,
                 struct vhost_blk_queue *vq, uint16_t req_idx,
                 struct iovec *iovs, uint32_t *iovs_idx,
                 uint32_t *payload)
{
    struct vring_packed_desc *desc = &vq->vring.desc_packed[req_idx];

    do {
        /* does not support indirect descriptors */
        assert((desc->flags & VRING_DESC_F_INDIRECT) == 0);

        if (*iovs_idx >= VHOST_BLK_MAX_IOVS) {
            fprintf(stderr, "Reach VHOST_BLK_MAX_IOVS\n");
            return -1;
        }

        if (desc_payload_to_iovs(ctrlr, iovs, iovs_idx,
            desc->addr, desc->len) != 0) {
            fprintf(stderr, "Failed to convert desc payload to iovs\n");
            return -1;
        }

        *payload += desc->len;

        desc = vring_get_next_desc_packed(vq, &req_idx);
    } while (desc != NULL);

    return 0;
}

static int
setup_iovs_from_inflight_desc(struct vhost_blk_ctrlr *ctrlr,
                  struct vhost_blk_queue *vq, uint16_t req_idx,
                  struct iovec *iovs, uint32_t *iovs_idx,
                  uint32_t *payload)
{
    struct rte_vhost_ring_inflight *inflight_vq;
    struct rte_vhost_inflight_desc_packed *desc;

    inflight_vq = &vq->inflight_ring;
    desc = &inflight_vq->inflight_packed->desc[req_idx];

    do {
        /* does not support indirect descriptors */
        assert((desc->flags & VRING_DESC_F_INDIRECT) == 0);

        if (*iovs_idx >= VHOST_BLK_MAX_IOVS) {
            fprintf(stderr, "Reach VHOST_BLK_MAX_IOVS\n");
            return -1;
        }

        if (desc_payload_to_iovs(ctrlr, iovs, iovs_idx,
            desc->addr, desc->len) != 0) {
            fprintf(stderr, "Failed to convert desc payload to iovs\n");
            return -1;
        }

        *payload += desc->len;

        desc = vring_get_next_inflight_desc(vq, desc);
    } while (desc != NULL);

    return 0;
}

static void
process_blk_task(struct vhost_blk_task *task)
{
    uint32_t payload = 0;

    if (task->vq->packed_ring) {
        struct rte_vhost_ring_inflight *inflight_ring;
        struct rte_vhost_resubmit_info *resubmit_inflight;

        inflight_ring = &task->vq->inflight_ring;
        resubmit_inflight = inflight_ring->resubmit_inflight;

        if (resubmit_inflight != NULL &&
            resubmit_inflight->resubmit_list != NULL) {
            if (setup_iovs_from_inflight_desc(task->ctrlr, task->vq,
                task->req_idx, task->iovs, &task->iovs_cnt,
                &payload)) {
                fprintf(stderr, "Failed to setup iovs\n");
                return;
            }
        } else {
            if (setup_iovs_from_descs_packed(task->ctrlr, task->vq,
                task->req_idx, task->iovs, &task->iovs_cnt,
                &payload)) {
                fprintf(stderr, "Failed to setup iovs\n");
                return;
            }
        }
    } else {
        if (setup_iovs_from_descs_split(task->ctrlr, task->vq,
            task->req_idx, task->iovs, &task->iovs_cnt, &payload)) {
            fprintf(stderr, "Failed to setup iovs\n");
            return;
        }
    }

    /* First IOV must be the req head. */
    task->req = (struct virtio_blk_outhdr *)task->iovs[0].iov_base;
    assert(sizeof(*task->req) == task->iovs[0].iov_len);

    /* Last IOV must be the status tail. */
    task->status = (uint8_t *)task->iovs[task->iovs_cnt - 1].iov_base;
    assert(sizeof(*task->status) == task->iovs[task->iovs_cnt - 1].iov_len);

    /* Transport data len */
    task->data_len = payload - task->iovs[0].iov_len -
        task->iovs[task->iovs_cnt - 1].iov_len;

    if (vhost_bdev_process_blk_commands(task->ctrlr->bdev, task))
        /* invalid response */
        *task->status = VIRTIO_BLK_S_IOERR;
    else
        /* successfully */
        *task->status = VIRTIO_BLK_S_OK;

    if (task->vq->packed_ring)
        enqueue_task_packed(task);
    else
        enqueue_task(task);
}

static void
blk_task_init(struct vhost_blk_task *task)
{
    task->iovs_cnt = 0;
    task->data_len = 0;
    task->req = NULL;
    task->status = NULL;
}

static void
submit_inflight_vq(struct vhost_blk_queue *vq)
{
    struct rte_vhost_ring_inflight *inflight_ring;
    struct rte_vhost_resubmit_info *resubmit_inflight;
    struct vhost_blk_task *task;

    inflight_ring = &vq->inflight_ring;
    resubmit_inflight = inflight_ring->resubmit_inflight;

    if (resubmit_inflight == NULL ||
        resubmit_inflight->resubmit_num == 0)
        return;

    fprintf(stdout, "Resubmit inflight num is %d\n",
        resubmit_inflight->resubmit_num);

    while (resubmit_inflight->resubmit_num-- > 0) {
        uint16_t desc_idx;

        desc_idx = resubmit_inflight->resubmit_list[
                    resubmit_inflight->resubmit_num].index;

        if (vq->packed_ring) {
            uint16_t task_idx;
            struct rte_vhost_inflight_desc_packed *desc;

            desc = inflight_ring->inflight_packed->desc;
            task_idx = desc[desc[desc_idx].last].id;
            task = &vq->tasks[task_idx];

            task->req_idx = desc_idx;
            task->chain_num = desc[desc_idx].num;
            task->buffer_id = task_idx;
            task->inflight_idx = desc_idx;

            vq->last_avail_idx += desc[desc_idx].num;
            if (vq->last_avail_idx >= vq->vring.size) {
                vq->last_avail_idx -= vq->vring.size;
                vq->avail_wrap_counter =
                    !vq->avail_wrap_counter;
            }
        } else
            /* In split ring, the desc_idx is the req_id
             * which was initialized when allocated the task pool.
             */
            task = &vq->tasks[desc_idx];

        blk_task_init(task);
        process_blk_task(task);
    }

    free(resubmit_inflight->resubmit_list);
    resubmit_inflight->resubmit_list = NULL;
}

/* Use the buffer_id as the task_idx */
static uint16_t
vhost_blk_vq_get_desc_chain_buffer_id(struct vhost_blk_queue *vq,
                      uint16_t *req_head, uint16_t *num)
{
    struct vring_packed_desc *desc = &vq->vring.desc_packed[
                        vq->last_avail_idx];

    *req_head = vq->last_avail_idx;
    *num = 1;

    while (descriptor_has_next_packed(desc)) {
        vq->last_avail_idx = (vq->last_avail_idx + 1) % vq->vring.size;
        desc = &vq->vring.desc_packed[vq->last_avail_idx];
        *num += 1;
    }

    /* Point to next desc */
    vq->last_avail_idx = (vq->last_avail_idx + 1) % vq->vring.size;
    if (vq->last_avail_idx < *req_head)
        vq->avail_wrap_counter = !vq->avail_wrap_counter;

    return desc->id;
}

static uint16_t
vq_get_desc_idx(struct vhost_blk_queue *vq)
{
    uint16_t desc_idx;
    uint16_t last_avail_idx;

    last_avail_idx = vq->last_avail_idx & (vq->vring.size - 1);
    desc_idx = vq->vring.avail->ring[last_avail_idx];
    vq->last_avail_idx++;

    return desc_idx;
}

static int
vhost_blk_vq_is_avail(struct vhost_blk_queue *vq)
{
    if (vq->packed_ring) {
        uint16_t flags = vq->vring.desc_packed[
                    vq->last_avail_idx].flags;
        bool avail_wrap_counter = vq->avail_wrap_counter;

        return (!!(flags & VIRTQ_DESC_F_AVAIL) == avail_wrap_counter &&
            !!(flags & VIRTQ_DESC_F_USED) != avail_wrap_counter);
    } else {
        if (vq->vring.avail->idx != vq->last_avail_idx)
            return 1;

        return 0;
    }
}

static void
process_vq(struct vhost_blk_queue *vq)
{
    struct vhost_blk_task *task;

    if (vq->packed_ring) {
        while (vhost_blk_vq_is_avail(vq)) {
            uint16_t task_idx, req_idx, last_idx, chain_num;

            task_idx = vhost_blk_vq_get_desc_chain_buffer_id(vq,
                    &req_idx, &chain_num);
            task = &vq->tasks[task_idx];

            blk_task_init(task);
            task->req_idx = req_idx;
            task->chain_num = chain_num;
            task->buffer_id = task_idx;
            last_idx = (req_idx + chain_num - 1) % vq->vring.size;

            rte_vhost_set_inflight_desc_packed(task->ctrlr->vid,
                               vq->id,
                               task->req_idx,
                               last_idx,
                               &task->inflight_idx);

            process_blk_task(task);
        }
    } else {
        while (vhost_blk_vq_is_avail(vq)) {
            uint16_t desc_idx;

            desc_idx = vq_get_desc_idx(vq);
            task = &vq->tasks[desc_idx];

            blk_task_init(task);
            rte_vhost_set_inflight_desc_split(task->ctrlr->vid,
                              vq->id,
                              task->req_idx);
            process_blk_task(task);
        }
    }
}

static void *
ctrlr_worker(void *arg)
{
    struct vhost_blk_ctrlr *ctrlr = (struct vhost_blk_ctrlr *)arg;
    cpu_set_t cpuset;
    pthread_t thread;
    int i;

    fprintf(stdout, "Ctrlr Worker Thread start\n");

    if (ctrlr == NULL || ctrlr->bdev == NULL) {
        fprintf(stderr,
            "%s: Error, invalid argument passed to worker thread\n",
            __func__);
        exit(0);
    }

    thread = pthread_self();
    CPU_ZERO(&cpuset);
    CPU_SET(0, &cpuset);
    pthread_setaffinity_np(thread, sizeof(cpu_set_t), &cpuset);

    for (i = 0; i < NUM_OF_BLK_QUEUES; i++)
        submit_inflight_vq(&ctrlr->queues[i]);

    while (worker_thread_status != WORKER_STATE_STOP)
        for (i = 0; i < NUM_OF_BLK_QUEUES; i++)
            process_vq(&ctrlr->queues[i]);

    fprintf(stdout, "Ctrlr Worker Thread Exiting\n");
    sem_post(&exit_sem);
    return NULL;
}

static int
alloc_task_pool(struct vhost_blk_ctrlr *ctrlr)
{
    struct vhost_blk_queue *vq;
    int i, j;

    for (i = 0; i < NUM_OF_BLK_QUEUES; i++) {
        vq = &ctrlr->queues[i];

        vq->tasks = rte_zmalloc(NULL,
            sizeof(struct vhost_blk_task) * vq->vring.size, 0);
        if (!vq->tasks) {
            fprintf(stderr, "Failed to allocate task memory\n");
            return -1;
        }

        for (j = 0; j < vq->vring.size; j++) {
            vq->tasks[j].req_idx = j;
            vq->tasks[j].ctrlr = ctrlr;
            vq->tasks[j].vq = vq;
        }
    }

    return 0;
}

static void
free_task_pool(struct vhost_blk_ctrlr *ctrlr)
{
    int i;

    for (i = 0; i < NUM_OF_BLK_QUEUES; i++)
        rte_free(ctrlr->queues[i].tasks);
}

static int
new_device(int vid)
{
    struct vhost_blk_ctrlr *ctrlr;
    struct vhost_blk_queue *vq;
    char path[PATH_MAX];
    uint64_t features;
    pthread_t tid;
    int i, ret;
    bool packed_ring;

    ret = rte_vhost_get_ifname(vid, path, PATH_MAX);
    if (ret) {
        fprintf(stderr, "Failed to get the socket path\n");
        return -1;
    }

    ctrlr = vhost_blk_ctrlr_find(path);
    if (!ctrlr) {
        fprintf(stderr, "Failed to find controller\n");
        return -1;
    }

    if (ctrlr->started)
        return 0;

    ctrlr->vid = vid;
    ret = rte_vhost_get_negotiated_features(vid, &features);
    if (ret) {
        fprintf(stderr, "Failed to get the negotiated features\n");
        return -1;
    }
    packed_ring = !!(features & (1ULL << VIRTIO_F_RING_PACKED));

    /* Disable Notifications and init last idx */
    for (i = 0; i < NUM_OF_BLK_QUEUES; i++) {
        vq = &ctrlr->queues[i];
        vq->id = i;

        assert(rte_vhost_get_vhost_vring(ctrlr->vid, i,
                         &vq->vring) == 0);
        assert(rte_vhost_get_vring_base(ctrlr->vid, i,
                           &vq->last_avail_idx,
                           &vq->last_used_idx) == 0);
        assert(rte_vhost_get_vhost_ring_inflight(ctrlr->vid, i,
                        &vq->inflight_ring) == 0);

        if (packed_ring) {
            /* for the reconnection */
            assert(rte_vhost_get_vring_base_from_inflight(
                ctrlr->vid, i,
                &vq->last_avail_idx,
                &vq->last_used_idx) == 0);

            vq->avail_wrap_counter = vq->last_avail_idx &
                (1 << 15);
            vq->last_avail_idx = vq->last_avail_idx &
                0x7fff;
            vq->used_wrap_counter = vq->last_used_idx &
                (1 << 15);
            vq->last_used_idx = vq->last_used_idx &
                0x7fff;
        }

        vq->packed_ring = packed_ring;
        rte_vhost_enable_guest_notification(vid, i, 0);
    }

    assert(rte_vhost_get_mem_table(vid, &ctrlr->mem) == 0);
    assert(ctrlr->mem != NULL);
    assert(alloc_task_pool(ctrlr) == 0);

    /* start polling vring */
    worker_thread_status = WORKER_STATE_START;
    fprintf(stdout, "New Device %s, Device ID %d\n", path, vid);
    if (pthread_create(&tid, NULL, &ctrlr_worker, ctrlr) < 0) {
        fprintf(stderr, "Worker Thread Started Failed\n");
        return -1;
    }

    /* device has been started */
    ctrlr->started = 1;
    pthread_detach(tid);
    return 0;
}

static void
destroy_device(int vid)
{
    char path[PATH_MAX];
    struct vhost_blk_ctrlr *ctrlr;
    struct vhost_blk_queue *vq;
    int i, ret;

    ret = rte_vhost_get_ifname(vid, path, PATH_MAX);
    if (ret) {
        fprintf(stderr, "Destroy Ctrlr Failed\n");
        return;
    }

    fprintf(stdout, "Destroy %s Device ID %d\n", path, vid);
    ctrlr = vhost_blk_ctrlr_find(path);
    if (!ctrlr) {
        fprintf(stderr, "Destroy Ctrlr Failed\n");
        return;
    }

    if (!ctrlr->started)
        return;

    worker_thread_status = WORKER_STATE_STOP;
    sem_wait(&exit_sem);

    for (i = 0; i < NUM_OF_BLK_QUEUES; i++) {
        vq = &ctrlr->queues[i];
        if (vq->packed_ring) {
            vq->last_avail_idx |= (vq->avail_wrap_counter <<
                15);
            vq->last_used_idx |= (vq->used_wrap_counter <<
                15);
        }

        rte_vhost_set_vring_base(ctrlr->vid, i,
                     vq->last_avail_idx,
                     vq->last_used_idx);
    }

    free_task_pool(ctrlr);
    free(ctrlr->mem);

    ctrlr->started = 0;
}

static int
new_connection(int vid)
{
    /* extend the proper features for block device */
    vhost_session_install_rte_compat_hooks(vid);

    return 0;
}

struct vhost_device_ops vhost_blk_device_ops = {
    .new_device =  new_device,
    .destroy_device = destroy_device,
    .new_connection = new_connection,
};

static struct vhost_block_dev *
vhost_blk_bdev_construct(const char *bdev_name,
    const char *bdev_serial, uint32_t blk_size, uint64_t blk_cnt,
    bool wce_enable)
{
    struct vhost_block_dev *bdev;

    bdev = rte_zmalloc(NULL, sizeof(*bdev), RTE_CACHE_LINE_SIZE);
    if (!bdev)
        return NULL;

    snprintf(bdev->name, sizeof(bdev->name), "%s", bdev_name);
    snprintf(bdev->product_name, sizeof(bdev->product_name), "%s",
         bdev_serial);
    bdev->blocklen = blk_size;
    bdev->blockcnt = blk_cnt;
    bdev->write_cache = wce_enable;

    fprintf(stdout, "Blocklen=%d, blockcnt=%"PRIx64"\n", bdev->blocklen,
        bdev->blockcnt);

    /* use memory as disk storage space */
    bdev->data = rte_zmalloc(NULL, blk_cnt * blk_size, 0);
    if (!bdev->data) {
        fprintf(stderr, "No enough reserved huge memory for disk\n");
        free(bdev);
        return NULL;
    }

    return bdev;
}

static struct vhost_blk_ctrlr *
vhost_blk_ctrlr_construct(const char *ctrlr_name)
{
    int ret;
    struct vhost_blk_ctrlr *ctrlr;
    char *path;
    char cwd[PATH_MAX];

    /* always use current directory */
    path = getcwd(cwd, PATH_MAX);
    if (!path) {
        fprintf(stderr, "Cannot get current working directory\n");
        return NULL;
    }
    snprintf(dev_pathname, sizeof(dev_pathname), "%s/%s", path, ctrlr_name);

    unlink(dev_pathname);

    if (rte_vhost_driver_register(dev_pathname, 0) != 0) {
        fprintf(stderr, "Socket %s already exists\n", dev_pathname);
        return NULL;
    }

    ret = rte_vhost_driver_set_features(dev_pathname, VHOST_BLK_FEATURES);
    if (ret != 0) {
        fprintf(stderr, "Set vhost driver features failed\n");
        rte_vhost_driver_unregister(dev_pathname);
        return NULL;
    }

    /* set vhost user protocol features */
    vhost_dev_install_rte_compat_hooks(dev_pathname);

    ctrlr = rte_zmalloc(NULL, sizeof(*ctrlr), RTE_CACHE_LINE_SIZE);
    if (!ctrlr) {
        rte_vhost_driver_unregister(dev_pathname);
        return NULL;
    }

    /* hardcoded block device information with 128MiB */
    ctrlr->bdev = vhost_blk_bdev_construct("malloc0", "vhost_blk_malloc0",
                        4096, 32768, 0);
    if (!ctrlr->bdev) {
        rte_free(ctrlr);
        rte_vhost_driver_unregister(dev_pathname);
        return NULL;
    }

    rte_vhost_driver_callback_register(dev_pathname,
                       &vhost_blk_device_ops);

    return ctrlr;
}

static void
vhost_blk_ctrlr_destroy(struct vhost_blk_ctrlr *ctrlr)
{
    if (ctrlr->bdev != NULL) {
        if (ctrlr->bdev->data != NULL)
            rte_free(ctrlr->bdev->data);

        rte_free(ctrlr->bdev);
    }
    rte_free(ctrlr);

    rte_vhost_driver_unregister(dev_pathname);
}

static void
signal_handler(__rte_unused int signum)
{
    struct vhost_blk_ctrlr *ctrlr;

    ctrlr = vhost_blk_ctrlr_find(dev_pathname);
    if (ctrlr == NULL)
        return;

    if (ctrlr->started)
        destroy_device(ctrlr->vid);

    vhost_blk_ctrlr_destroy(ctrlr);
    exit(0);
}

int main(int argc, char *argv[])
{
    int ret;

    /* init EAL */
    ret = rte_eal_init(argc, argv);
    if (ret < 0)
        rte_exit(EXIT_FAILURE, "Error with EAL initialization\n");

    g_vhost_ctrlr = vhost_blk_ctrlr_construct(CTRLR_NAME);
    if (g_vhost_ctrlr == NULL) {
        fprintf(stderr, "Construct vhost blk controller failed\n");
        return 0;
    }

    if (sem_init(&exit_sem, 0, 0) < 0) {
        fprintf(stderr, "Error init exit_sem\n");
        return -1;
    }

    signal(SIGINT, signal_handler);

    ret = rte_vhost_driver_start(dev_pathname);
    if (ret < 0) {
        fprintf(stderr, "Failed to start vhost driver.\n");
        return -1;
    }

    /* loop for exit the application */
    while (1)
        sleep(1);

    /* clean up the EAL */
    rte_eal_cleanup();

    return 0;
}