/*
* Wait for a DCCP event.
*
* Note that we don't need to lock the socket, as the upper poll layers
* take care of normal races (between the test and the event) and we don't
* go look at any of the socket buffers directly.
*/
static unsigned int dccp_poll(struct file *file, struct socket *sock,
poll_table *wait)
{
unsigned int mask;
struct sock *sk = sock->sk;
poll_wait(file, sk->sk_sleep, wait);
if (sk->sk_state == DCCP_LISTEN)
return inet_csk_listen_poll(sk);
/* Socket is not locked. We are protected from async events
by poll logic and correct handling of state changes
made by another threads is impossible in any case.
*/
mask = 0;
if (sk->sk_err)
mask = POLLERR;
if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == DCCP_CLOSED)
mask |= POLLHUP;
if (sk->sk_shutdown & RCV_SHUTDOWN)
mask |= POLLIN | POLLRDNORM;
/* Connected? */
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_RESPOND)) {
if (atomic_read(&sk->sk_rmem_alloc) > 0)
mask |= POLLIN | POLLRDNORM;
if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) {
mask |= POLLOUT | POLLWRNORM;
} else { /* send SIGIO later */
set_bit(SOCK_ASYNC_NOSPACE,
&sk->sk_socket->flags);
set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
/* Race breaker. If space is freed after
* wspace test but before the flags are set,
* IO signal will be lost.
*/
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk))
mask |= POLLOUT | POLLWRNORM;
}
}
}
return mask;
}
unsigned int dccp_poll(struct file *file, struct socket *sock,
poll_table *wait)
{
unsigned int mask;
struct sock *sk = sock->sk;
sock_poll_wait(file, sk->sk_sleep, wait);
if (sk->sk_state == DCCP_LISTEN)
return inet_csk_listen_poll(sk);
mask = 0;
if (sk->sk_err)
mask = POLLERR;
if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == DCCP_CLOSED)
mask |= POLLHUP;
if (sk->sk_shutdown & RCV_SHUTDOWN)
mask |= POLLIN | POLLRDNORM | POLLRDHUP;
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_RESPOND)) {
if (atomic_read(&sk->sk_rmem_alloc) > 0)
mask |= POLLIN | POLLRDNORM;
if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) {
mask |= POLLOUT | POLLWRNORM;
} else {
set_bit(SOCK_ASYNC_NOSPACE,
&sk->sk_socket->flags);
set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk))
mask |= POLLOUT | POLLWRNORM;
}
}
}
return mask;
}
/**
* sk_stream_write_space - stream socket write_space callback.
* @sk: socket
*
* FIXME: write proper description
*/
void sk_stream_write_space(struct sock *sk)
{
struct socket *sock = sk->sk_socket;
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk) && sock) {
clear_bit(SOCK_NOSPACE, &sock->flags);
if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
wake_up_interruptible(sk->sk_sleep);
if (sock->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN))
sock_wake_async(sock, 2, POLL_OUT);
}
}
static void
ksocknal_write_space(struct sock *sk)
{
struct ksock_conn *conn;
int wspace;
int min_wpace;
/* interleave correctly with closing sockets... */
LASSERT(!in_irq());
read_lock(&ksocknal_data.ksnd_global_lock);
conn = sk->sk_user_data;
wspace = sk_stream_wspace(sk);
min_wpace = sk_stream_min_wspace(sk);
CDEBUG(D_NET, "sk %p wspace %d low water %d conn %p%s%s%s\n",
sk, wspace, min_wpace, conn,
!conn ? "" : (conn->ksnc_tx_ready ?
" ready" : " blocked"),
!conn ? "" : (conn->ksnc_tx_scheduled ?
" scheduled" : " idle"),
!conn ? "" : (list_empty(&conn->ksnc_tx_queue) ?
" empty" : " queued"));
if (!conn) { /* raced with ksocknal_terminate_conn */
LASSERT(sk->sk_write_space != &ksocknal_write_space);
sk->sk_write_space(sk);
read_unlock(&ksocknal_data.ksnd_global_lock);
return;
}
if (wspace >= min_wpace) { /* got enough space */
ksocknal_write_callback(conn);
/*
* Clear SOCK_NOSPACE _after_ ksocknal_write_callback so the
* ENOMEM check in ksocknal_transmit is race-free (think about
* it).
*/
clear_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
}
read_unlock(&ksocknal_data.ksnd_global_lock);
}
void sk_stream_write_space(struct sock *sk)
{
struct socket *sock = sk->sk_socket;
struct socket_wq *wq;
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk) && sock) {
clear_bit(SOCK_NOSPACE, &sock->flags);
rcu_read_lock();
wq = rcu_dereference(sk->sk_wq);
if (wq_has_sleeper(wq))
wake_up_interruptible_poll(&wq->wait, POLLOUT |
POLLWRNORM | POLLWRBAND);
if (wq && wq->fasync_list && !(sk->sk_shutdown & SEND_SHUTDOWN))
sock_wake_async(sock, SOCK_WAKE_SPACE, POLL_OUT);
rcu_read_unlock();
}
}
/**
* xs_tcp_write_space - callback invoked when socket buffer space
* becomes available
* @sk: socket whose state has changed
*
* Called when more output buffer space is available for this socket.
* We try not to wake our writers until they can make "significant"
* progress, otherwise we'll waste resources thrashing kernel_sendmsg
* with a bunch of small requests.
*/
static void xs_tcp_write_space(struct sock *sk)
{
read_lock(&sk->sk_callback_lock);
/* from net/core/stream.c:sk_stream_write_space */
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) {
struct socket *sock;
struct rpc_xprt *xprt;
if (unlikely(!(sock = sk->sk_socket)))
goto out;
if (unlikely(!(xprt = xprt_from_sock(sk))))
goto out;
if (unlikely(!test_and_clear_bit(SOCK_NOSPACE, &sock->flags)))
goto out;
xprt_write_space(xprt);
}
out:
read_unlock(&sk->sk_callback_lock);
}
