/* This is where we get a request from a caller to read something */
static BlockDriverAIOCB *tar_aio_readv(BlockDriverState *bs,
        int64_t sector_num, QEMUIOVector *qiov, int nb_sectors,
        BlockDriverCompletionFunc *cb, void *opaque)
{
    BDRVTarState *s = bs->opaque;
    SparseCache *sparse;
    int64_t sec_file = sector_num + s->file_sec;
    int64_t start = sector_num * SECTOR_SIZE;
    int64_t end = start + (nb_sectors * SECTOR_SIZE);
    int i;
    TarAIOCB *acb;

    for (i = 0; i < s->sparse_num; i++) {
        sparse = &s->sparse[i];
        if (sparse->start > end) {
            /* We expect the cache to be start increasing */
            break;
        } else if ((sparse->start < start) && (sparse->end <= start)) {
            /* sparse before our offset */
            sec_file -= (sparse->end - sparse->start) / SECTOR_SIZE;
        } else if ((sparse->start <= start) && (sparse->end >= end)) {
            /* all our sectors are sparse */
            char *buf = g_malloc0(nb_sectors * SECTOR_SIZE);

            acb = qemu_aio_get(&tar_aiocb_info, bs, cb, opaque);
            qemu_iovec_from_buf(qiov, 0, buf, nb_sectors * SECTOR_SIZE);
            g_free(buf);
            acb->bh = qemu_bh_new(tar_sparse_cb, acb);
            qemu_bh_schedule(acb->bh);

            return &acb->common;
        } else if (((sparse->start >= start) && (sparse->start < end)) ||
                   ((sparse->end >= start) && (sparse->end < end))) {
            /* we're semi-sparse (worst case) */
            /* let's go synchronous and read all sectors individually */
            char *buf = g_malloc(nb_sectors * SECTOR_SIZE);
            uint64_t offs;

            for (offs = 0; offs < (nb_sectors * SECTOR_SIZE);
                 offs += SECTOR_SIZE) {
                bdrv_pread(bs, (sector_num * SECTOR_SIZE) + offs,
                           buf + offs, SECTOR_SIZE);
            }

            qemu_iovec_from_buf(qiov, 0, buf, nb_sectors * SECTOR_SIZE);
            acb = qemu_aio_get(&tar_aiocb_info, bs, cb, opaque);
            acb->bh = qemu_bh_new(tar_sparse_cb, acb);
            qemu_bh_schedule(acb->bh);

            return &acb->common;
        }
    }

    return bdrv_aio_readv(s->hd, sec_file, qiov, nb_sectors,
                          cb, opaque);
}

static void tpm_tis_realizefn(DeviceState *dev, Error **errp)
{
    TPMState *s = TPM(dev);
    TPMTISEmuState *tis = &s->s.tis;

    s->be_driver = qemu_find_tpm(s->backend);
    if (!s->be_driver) {
        error_setg(errp, "tpm_tis: backend driver with id %s could not be "
                   "found", s->backend);
        return;
    }

    s->be_driver->fe_model = TPM_MODEL_TPM_TIS;

    if (s->be_driver->ops->init(s->be_driver, s, tpm_tis_receive_cb)) {
        error_setg(errp, "tpm_tis: backend driver with id %s could not be "
                   "initialized", s->backend);
        return;
    }

    if (tis->irq_num > 15) {
        error_setg(errp, "tpm_tis: IRQ %d for TPM TIS is outside valid range "
                   "of 0 to 15.\n", tis->irq_num);
        return;
    }

    tis->bh = qemu_bh_new(tpm_tis_receive_bh, s);

    isa_init_irq(&s->busdev, &tis->irq, tis->irq_num);
}

static void *sh_timer_init(uint32_t freq, int feat, qemu_irq irq)
{
    sh_timer_state *s;
    QEMUBH *bh;

    s = (sh_timer_state *)qemu_mallocz(sizeof(sh_timer_state));
    s->freq = freq;
    s->feat = feat;
    s->tcor = 0xffffffff;
    s->tcnt = 0xffffffff;
    s->tcpr = 0xdeadbeef;
    s->tcr = 0;
    s->enabled = 0;
    s->irq = irq;

    bh = qemu_bh_new(sh_timer_tick, s);
    s->timer = ptimer_init(bh);

    sh_timer_write(s, OFFSET_TCOR >> 2, s->tcor);
    sh_timer_write(s, OFFSET_TCNT >> 2, s->tcnt);
    sh_timer_write(s, OFFSET_TCPR >> 2, s->tcpr);
    sh_timer_write(s, OFFSET_TCR  >> 2, s->tcpr);
    /* ??? Save/restore.  */
    return s;
}

static int xilinx_timer_init(SysBusDevice *dev)
{
    struct timerblock *t = FROM_SYSBUS(typeof (*t), dev);
    unsigned int i;

    /* All timers share a single irq line.  */
    sysbus_init_irq(dev, &t->irq);

    /* Init all the ptimers.  */
    t->timers = g_malloc0(sizeof t->timers[0] * num_timers(t));
    for (i = 0; i < num_timers(t); i++) {
        struct xlx_timer *xt = &t->timers[i];

        xt->parent = t;
        xt->nr = i;
        xt->bh = qemu_bh_new(timer_hit, xt);
        xt->ptimer = ptimer_init(xt->bh);
        ptimer_set_freq(xt->ptimer, t->freq_hz);
    }

    memory_region_init_io(&t->mmio, &timer_ops, t, "xlnx.xps-timer",
                          R_MAX * 4 * num_timers(t));
    sysbus_init_mmio(dev, &t->mmio);
    return 0;
}

static int grlib_gptimer_init(SysBusDevice *dev)
{
    GPTimerUnit  *unit = FROM_SYSBUS(typeof(*unit), dev);
    unsigned int  i;

    assert(unit->nr_timers > 0);
    assert(unit->nr_timers <= GPTIMER_MAX_TIMERS);

    unit->timers = g_malloc0(sizeof unit->timers[0] * unit->nr_timers);

    for (i = 0; i < unit->nr_timers; i++) {
        GPTimer *timer = &unit->timers[i];

        timer->unit   = unit;
        timer->bh     = qemu_bh_new(grlib_gptimer_hit, timer);
        timer->ptimer = ptimer_init(timer->bh);
        timer->id     = i;

        /* One IRQ line for each timer */
        sysbus_init_irq(dev, &timer->irq);

        ptimer_set_freq(timer->ptimer, unit->freq_hz);
    }

    memory_region_init_io(&unit->iomem, &grlib_gptimer_ops, unit, "gptimer",
                          UNIT_REG_SIZE + GPTIMER_REG_SIZE * unit->nr_timers);

    sysbus_init_mmio(dev, &unit->iomem);
    return 0;
}

static void
iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
                        void *command_data, void *opaque)
{
    struct IscsiTask *iTask = opaque;
    struct scsi_task *task = command_data;

    iTask->complete = 1;
    iTask->status = status;
    iTask->do_retry = 0;
    iTask->task = task;

    if (iTask->retries-- > 0 && status == SCSI_STATUS_CHECK_CONDITION
        && task->sense.key == SCSI_SENSE_UNIT_ATTENTION) {
        iTask->do_retry = 1;
        goto out;
    }

    if (status != SCSI_STATUS_GOOD) {
        error_report("iSCSI: Failure. %s", iscsi_get_error(iscsi));
    }

out:
    if (iTask->co) {
        iTask->bh = qemu_bh_new(iscsi_co_generic_bh_cb, iTask);
        qemu_bh_schedule(iTask->bh);
    }
}

int qemu_init_main_loop(Error **errp)
{
    int ret;
    GSource *src;
    Error *local_error = NULL;

    init_clocks(qemu_timer_notify_cb);

    ret = qemu_signal_init(errp);
    if (ret) {
        return ret;
    }

    qemu_aio_context = aio_context_new(&local_error);
    if (!qemu_aio_context) {
        error_propagate(errp, local_error);
        return -EMFILE;
    }
    qemu_notify_bh = qemu_bh_new(notify_event_cb, NULL);
    gpollfds = g_array_new(FALSE, FALSE, sizeof(GPollFD));
    src = aio_get_g_source(qemu_aio_context);
    g_source_set_name(src, "aio-context");
    g_source_attach(src, NULL);
    g_source_unref(src);
    src = iohandler_get_g_source();
    g_source_set_name(src, "io-handler");
    g_source_attach(src, NULL);
    g_source_unref(src);
    return 0;
}

static BlockDriverAIOCB *raw_aio_write(BlockDriverState *bs,
        int64_t sector_num, const uint8_t *buf, int nb_sectors,
        BlockDriverCompletionFunc *cb, void *opaque)
{
    RawAIOCB *acb;

    /*
     * If O_DIRECT is used and the buffer is not aligned fall back
     * to synchronous IO.
     */
    BDRVRawState *s = bs->opaque;

    if (unlikely(s->aligned_buf != NULL && ((uintptr_t) buf % 512))) {
        QEMUBH *bh;
        acb = qemu_aio_get(bs, cb, opaque);
        acb->ret = raw_pwrite(bs, 512 * sector_num, buf, 512 * nb_sectors);
        bh = qemu_bh_new(raw_aio_em_cb, acb);
        qemu_bh_schedule(bh);
        return &acb->common;
    }

    acb = raw_aio_setup(bs, sector_num, (uint8_t*)buf, nb_sectors, cb, opaque);
    if (!acb)
        return NULL;
    if (qemu_paio_write(&acb->aiocb) < 0) {
        raw_aio_remove(acb);
        return NULL;
    }
    return &acb->common;
}

static void continue_after_map_failure(void *opaque)
{
    DMAAIOCB *dbs = (DMAAIOCB *)opaque;

    dbs->bh = qemu_bh_new(reschedule_dma, dbs);
    qemu_bh_schedule(dbs->bh);
}

static void
iscsi_schedule_bh(IscsiAIOCB *acb)
{
    if (acb->bh) {
        return;
    }
    acb->bh = qemu_bh_new(iscsi_bh_cb, acb);
    qemu_bh_schedule(acb->bh);
}

/*
 * This is the callback function for rbd_aio_read and _write
 *
 * Note: this function is being called from a non qemu thread so
 * we need to be careful about what we do here. Generally we only
 * schedule a BH, and do the rest of the io completion handling
 * from rbd_finish_bh() which runs in a qemu context.
 */
static void rbd_finish_aiocb(rbd_completion_t c, RADOSCB *rcb)
{
    RBDAIOCB *acb = rcb->acb;

    rcb->ret = rbd_aio_get_return_value(c);
    rbd_aio_release(c);

    acb->bh = qemu_bh_new(rbd_finish_bh, rcb);
    qemu_bh_schedule(acb->bh);
}

void failover_request_active(Error **errp)
{
   if (failover_set_state(FAILOVER_STATUS_NONE,
        FAILOVER_STATUS_REQUIRE) != FAILOVER_STATUS_NONE) {
        error_setg(errp, "COLO failover is already actived");
        return;
    }
    failover_bh = qemu_bh_new(colo_failover_bh, NULL);
    qemu_bh_schedule(failover_bh);
}

static int etraxfs_timer_init(SysBusDevice *dev)
{
    struct etrax_timer *t = FROM_SYSBUS(typeof (*t), dev);

    t->bh_t0 = qemu_bh_new(timer0_hit, t);
    t->bh_t1 = qemu_bh_new(timer1_hit, t);
    t->bh_wd = qemu_bh_new(watchdog_hit, t);
    t->ptimer_t0 = ptimer_init(t->bh_t0);
    t->ptimer_t1 = ptimer_init(t->bh_t1);
    t->ptimer_wd = ptimer_init(t->bh_wd);

    sysbus_init_irq(dev, &t->irq);
    sysbus_init_irq(dev, &t->nmi);

    memory_region_init_io(&t->mmio, &timer_ops, t, "etraxfs-timer", 0x5c);
    sysbus_init_mmio(dev, &t->mmio);
    qemu_register_reset(etraxfs_timer_reset, t);
    return 0;
}

static void virtio_blk_handle_output(VirtIODevice *vdev, VirtQueue *vq)
{
    VirtIOBlock *s = to_virtio_blk(vdev);
    VirtIOBlockReq *req;
    MultiReqBuffer mrb = {
        .num_writes = 0,
    };

    while ((req = virtio_blk_get_request(s))) {
        virtio_blk_handle_request(req, &mrb);
    }

    virtio_submit_multiwrite(s->bs, &mrb);

    /*
     * FIXME: Want to check for completions before returning to guest mode,
     * so cached reads and writes are reported as quickly as possible. But
     * that should be done in the generic block layer.
     */
}

static void virtio_blk_dma_restart_bh(void *opaque)
{
    VirtIOBlock *s = opaque;
    VirtIOBlockReq *req = s->rq;
    MultiReqBuffer mrb = {
        .num_writes = 0,
    };

    qemu_bh_delete(s->bh);
    s->bh = NULL;

    s->rq = NULL;

    while (req) {
        virtio_blk_handle_request(req, &mrb);
        req = req->next;
    }

    virtio_submit_multiwrite(s->bs, &mrb);
}

static void virtio_blk_dma_restart_cb(void *opaque, int running,
                                      RunState state)
{
    VirtIOBlock *s = opaque;

    if (!running)
        return;

    if (!s->bh) {
        s->bh = qemu_bh_new(virtio_blk_dma_restart_bh, s);
        qemu_bh_schedule(s->bh);
    }
}

static void blk_alloc(struct XenDevice *xendev)
{
    struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev);

    QLIST_INIT(&blkdev->inflight);
    QLIST_INIT(&blkdev->finished);
    QLIST_INIT(&blkdev->freelist);
    blkdev->bh = qemu_bh_new(blk_bh, blkdev);
    if (xen_mode != XEN_EMULATE)
        batch_maps = 1;
}

static int marin_timer_init(SysBusDevice *dev)
{
    MarinTimerState *s = MARIN_TIMER(dev);

    sysbus_init_irq(dev, &s->irq);

    s->bh = qemu_bh_new(marin_timer_tick, s);
    s->ptimer = ptimer_init(s->bh);
    ptimer_set_freq(s->ptimer, 50 * 1000 * 1000);

    return 0;
}

static int milkymist_sysctl_init(SysBusDevice *dev)
{
    MilkymistSysctlState *s = FROM_SYSBUS(typeof(*s), dev);

    sysbus_init_irq(dev, &s->gpio_irq);
    sysbus_init_irq(dev, &s->timer0_irq);
    sysbus_init_irq(dev, &s->timer1_irq);

    s->bh0 = qemu_bh_new(timer0_hit, s);
    s->bh1 = qemu_bh_new(timer1_hit, s);
    s->ptimer0 = ptimer_init(s->bh0);
    s->ptimer1 = ptimer_init(s->bh1);
    ptimer_set_freq(s->ptimer0, s->freq_hz);
    ptimer_set_freq(s->ptimer1, s->freq_hz);

    memory_region_init_io(&s->regs_region, &sysctl_mmio_ops, s,
                          "milkymist-sysctl", R_MAX * 4);
    sysbus_init_mmio(dev, &s->regs_region);

    return 0;
}

static int
iscsi_schedule_bh(QEMUBHFunc *cb, IscsiAIOCB *acb)
{
    acb->bh = qemu_bh_new(cb, acb);
    if (!acb->bh) {
        error_report("oom: could not create iscsi bh");
        return -EIO;
    }

    qemu_bh_schedule(acb->bh);
    return 0;
}

static void etsec_realize(DeviceState *dev, Error **errp)
{
    eTSEC        *etsec = ETSEC_COMMON(dev);

    etsec->nic = qemu_new_nic(&net_etsec_info, &etsec->conf,
                              object_get_typename(OBJECT(dev)), dev->id, etsec);
    qemu_format_nic_info_str(qemu_get_queue(etsec->nic), etsec->conf.macaddr.a);


    etsec->bh     = qemu_bh_new(etsec_timer_hit, etsec);
    etsec->ptimer = ptimer_init(etsec->bh, PTIMER_POLICY_DEFAULT);
    ptimer_set_freq(etsec->ptimer, 100);
}

void block_job_defer_to_main_loop(BlockJob *job,
                                  BlockJobDeferToMainLoopFn *fn,
                                  void *opaque)
{
    BlockJobDeferToMainLoopData *data = g_malloc(sizeof(*data));
    data->job = job;
    data->bh = qemu_bh_new(block_job_defer_to_main_loop_bh, data);
    data->aio_context = bdrv_get_aio_context(job->bs);
    data->fn = fn;
    data->opaque = opaque;

    qemu_bh_schedule(data->bh);
}