/*
* Same as __wtd_down, but sem->sleepers is not incremented when
* coming from a wakeup.
*/
void wtd_down_common(struct semaphore * sem, struct worktodo *wtd,
int do_incr)
{
int gotit;
int sleepers;
init_waitqueue_func_entry(&wtd->wait, __wtd_down_waiter);
wtd->data = sem;
spin_lock_irq(&semaphore_lock);
sem->sleepers += do_incr;
sleepers = sem->sleepers;
gotit = add_wait_queue_exclusive_cond(&sem->wait, &wtd->wait,
atomic_add_negative(sleepers - 1, &sem->count));
if (gotit)
sem->sleepers = 0;
else
sem->sleepers = 1;
spin_unlock_irq(&semaphore_lock);
if (gotit) {
wake_up(&sem->wait);
wtd_queue(wtd);
}
}
/* Init poll structure */
void vhost_poll_init(struct vhost_poll *poll, work_func_t func,
unsigned long mask)
{
INIT_WORK(&poll->work, func);
init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
init_poll_funcptr(&poll->table, vhost_poll_func);
poll->mask = mask;
}
/* Init poll structure */
void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
unsigned long mask, struct vhost_dev *dev)
{
init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup);
init_poll_funcptr(&poll->table, vhost_poll_func);
poll->mask = mask;
poll->dev = dev;
vhost_work_init(&poll->work, fn);
}
void __pollwait(struct file * filp, wait_queue_head_t * wait_address, poll_table *p)
{
struct poll_table_page *table;
#ifdef CONFIG_EPOLL
/* If there is a qproc set (in this case, that implies it's an
* eventpoll poll_table), we may be casting the poll_table from
* something else so make sure we don't dereference any other
* poll_table fields in this case. */
if (p->qproc) {
p->qproc(filp, wait_address, p);
return;
}
#endif /* CONFIG_EPOLL */
table = p->table;
if (!table || POLL_TABLE_FULL(table)) {
struct poll_table_page *new_table;
new_table = (struct poll_table_page *) __get_free_page(GFP_KERNEL);
if (!new_table) {
p->error = -ENOMEM;
__set_current_state(TASK_RUNNING);
return;
}
new_table->size = PAGE_SIZE;
new_table->entry = new_table->entries;
new_table->next = table;
p->table = new_table;
table = new_table;
}
/* Add a new entry */
{
struct poll_table_entry * entry = table->entry;
table->entry = entry+1;
get_file(filp);
entry->filp = filp;
entry->wait_address = wait_address;
entry->p = p;
if (p->iocb) /* async poll */
init_waitqueue_func_entry(&entry->wait, async_poll_waiter);
else
init_waitqueue_entry(&entry->wait, current);
add_wait_queue(wait_address,&entry->wait);
}
}
static void skb_async_read_worker(void *_data)
{
struct skb_async_info *info = _data;
struct sock *sk = info->sk;
struct sk_buff *skb;
int error;
/* Caller is allowed not to check sk->err before skb_recv_datagram() */
error = sock_error(sk);
if (error)
goto no_packet;
init_waitqueue_func_entry(&info->wtd.wait, skb_async_read_waiter);
/* Attempted to dequeue and process any skbs that already arrived.
* Note that add_wait_queue_cond is used to check against a race
* where an skb is added to the queue after we checked but before
* the callback is added to the wait queue.
*/
do {
skb = skb_dequeue(&sk->receive_queue);
if (skb) {
info->finish(sk, info->cb, info->len, skb);
kfree(info);
return;
}
} while ( add_wait_queue_cond( sk->sleep, &info->wtd.wait,
(!(error = sock_error(sk)) &&
skb_queue_empty(&sk->receive_queue)) )
&& !error);
if (!error)
return;
no_packet:
info->cb.fn(info->cb.data, info->cb.vec, error);
kfree(info);
return;
}
/*
* 该函数在调用f_op->poll()时会被调用.
* 也就是epoll主动poll某个fd时, 用来将epitem与指定的fd关联起来的.
* 关联的办法就是使用等待队列(waitqueue)
*/
static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead,
poll_table *pt)
{
struct epitem *epi = ep_item_from_epqueue(pt);
struct eppoll_entry *pwq;
if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, GFP_KERNEL))) {
/* 初始化等待队列, 指定ep_poll_callback为唤醒时的回调函数,
* 当我们监听的fd发生状态改变时, 也就是队列头被唤醒时,
* 指定的回调函数将会被调用. */
init_waitqueue_func_entry(&pwq->wait, ep_poll_callback);
pwq->whead = whead;
pwq->base = epi;
/* 将刚分配的等待队列成员加入到头中, 头是由fd持有的 */
add_wait_queue(whead, &pwq->wait);
list_add_tail(&pwq->llink, &epi->pwqlist);
/* nwait记录了当前epitem加入到了多少个等待队列中,
* 我认为这个值最大也只会是1... */
epi->nwait++;
} else {
/* We have to signal that an error occurred */
epi->nwait = -1;
}
}
static int virqfd_enable(struct vfio_pci_device *vdev,
int (*handler)(struct vfio_pci_device *, void *),
void (*thread)(struct vfio_pci_device *, void *),
void *data, struct virqfd **pvirqfd, int fd)
{
struct fd irqfd;
struct eventfd_ctx *ctx;
struct virqfd *virqfd;
int ret = 0;
unsigned int events;
virqfd = kzalloc(sizeof(*virqfd), GFP_KERNEL);
if (!virqfd)
return -ENOMEM;
virqfd->pvirqfd = pvirqfd;
virqfd->vdev = vdev;
virqfd->handler = handler;
virqfd->thread = thread;
virqfd->data = data;
INIT_WORK(&virqfd->shutdown, virqfd_shutdown);
INIT_WORK(&virqfd->inject, virqfd_inject);
irqfd = fdget(fd);
if (!irqfd.file) {
ret = -EBADF;
goto err_fd;
}
ctx = eventfd_ctx_fileget(irqfd.file);
if (IS_ERR(ctx)) {
ret = PTR_ERR(ctx);
goto err_ctx;
}
virqfd->eventfd = ctx;
/*
* virqfds can be released by closing the eventfd or directly
* through ioctl. These are both done through a workqueue, so
* we update the pointer to the virqfd under lock to avoid
* pushing multiple jobs to release the same virqfd.
*/
spin_lock_irq(&vdev->irqlock);
if (*pvirqfd) {
spin_unlock_irq(&vdev->irqlock);
ret = -EBUSY;
goto err_busy;
}
*pvirqfd = virqfd;
spin_unlock_irq(&vdev->irqlock);
/*
* Install our own custom wake-up handling so we are notified via
* a callback whenever someone signals the underlying eventfd.
*/
init_waitqueue_func_entry(&virqfd->wait, virqfd_wakeup);
init_poll_funcptr(&virqfd->pt, virqfd_ptable_queue_proc);
events = irqfd.file->f_op->poll(irqfd.file, &virqfd->pt);
/*
* Check if there was an event already pending on the eventfd
* before we registered and trigger it as if we didn't miss it.
*/
if (events & POLLIN) {
if ((!handler || handler(vdev, data)) && thread)
schedule_work(&virqfd->inject);
}
/*
* Do not drop the file until the irqfd is fully initialized,
* otherwise we might race against the POLLHUP.
*/
fdput(irqfd);
return 0;
err_busy:
eventfd_ctx_put(ctx);
err_ctx:
fdput(irqfd);
err_fd:
kfree(virqfd);
return ret;
}